I

m

'A

j:

A

1

>^>>

• W • .^ W* 9r* w * w* W*

IjlliiWill wKTil Tl iiiiwimiiiimiMHIUiliii^ ilM

"^m

SSNgygSrSSVSS^nS.^

PRIITATE LIBRARY

— •mOIF'-W- —

1) '\j, i^piiKMif.

:'T)

"Ai

:yo. z^^

\

09

r-

5

a
a

LU

o

a

■

Ra

^

A MANUAL N^H^r

m

OF THE

ANATOMY OF INVERTEBRATED

ANIMALS.

BY

THOMAS H. HUXLEY, LL. D., F. R. S.

NEW YOEK:
D. APPLETON AND COMPANY,

1, S, AND 5 BOND STREET.

1885.

PREFACE.

The ]3resent volume on the Anatomy of Invert ebrated
Animals fulfills an nndertaking to produce a treatise on
comparative anatomy for students, into whicli I entered
two-and-twenty years ago. A considerable installment of
the work, relating- wholly to the Invertebrata, appeared in
the Medical Times and Gazette for the years 1856 and
1857, under the title of " Lectures on General Natural
History." But a variety of circumstances having con-
spired, about that time, to compel me to direct my atten-
tion more particularly to the Vertebrata, I was led to in-
terrupt the publication of the " Lectures " and to com-
plete the Vertebrate half of the proposed work first. This
appeared in 1871, as a " Manual of the Anatomy of Yerte-
brated Animals."

A period of incapacity for any serious toil prevented
me from attempting, before 1874, to grapple with the im-
mense mass of new and important information respecting
the structure, and especially the development, of Inverte-
brated animals, which the activity of a host of investiga-
tors has accumulated of late years.

That my progress has been slow will not surprise any
one who is acquainted with the growth of the literature
of animal morphology, or with the expenditure of time
involved in the attempt to verify for one's self even the
cardinal facts of that science ; but I have endeavored, in

4 PREFACE.

the last chapter, to supply the most important recent ad-
ditions to om- knowledge, respecting the groups treated of
in those which have long been printed.

^^hen I commenced this work, it was mv intention to
continue the plan adopted in the " Manual of the Anatomy
of Tertehrated Animals,'' of giving a summary account
of what appeared to me to be ascertained morphokgical
facts, without referrincr to mr sources of information. I
soon found, however, that it would be inconvenient to
carrv out this scheme consistentlv : and some of mv pao^es
are, I am afi'aid, somewhat bm'dened with notes and ref-
erences.

I am the more careful to mention this circumstance as,
had it been my pui^ose to give any adequate Bibliograj^hy,
the conspicuous absence of the titles of many important
books and memoirs might appear unaccountable and in-
deed blame worth V.

My object, in writing the book, has been to make it
useful to those who wish to become acquainted with the
broad outlines of what is at present known of the moiiDhol-
02T of the Invertehraia : though I have not avoided the
incidental mention of facts connected with their physiol-
osTT and their distribution. On the other hand, I have ab-
stained from discussing questions of aetiology, not because
I underestimate their importance, or am insensible to the
interest of the great problem of Evolution ; but because,
to my mind, the growing tendency to mix up ^etiological
speculations with morphological generalizations will, if
unchecked, throw Biology into confusion.

For the student, that which is essential is a knowledge
of the facts of morphology ; and he should recollect that
generalizations are empty formulas, unless there is some-
thing in his personal experience which gives reality and
substance to the terms of the propositions in which these
generalizations are expressed.

PREFACE. 5

The dissection of a single representative of eacli of the
principal divisions of the Inveriebrata ^\SS. give the student
a more real acquaintance with their comparative anatomy
than any amount of readiug of this, or any other book.
And I have endeavored to facilitate practical study by
supplying a somewhat full description of individual forms,
in the case of the more complicated tj-pes.

That the power of repeating a " Classification of Ani-
mals," with all the appropriate definitions, has anything
to do with genuine knowledge is one of the commonest
and most mischievous delusions of both students and their
examiners.

The real business of the learner is to gain a true and
vivid conception of the characteristics of what may be
termed the natural orders of animals. The mode of ar-
rangement, or classification, of these into larger groups is
a matter of altogether secondary importance. As such, I
have relegated this subject to a subordinate place in the
last chapter ; and I have thought it unnecessary^, either to
discuss the systems proposed by others, or to give reasons
for passing over, in silence, my own former attempts in
this direction.

Of the manifold imperfections in the execution of the
task which I have set myself, few will be more sensible
than I am ; but I trust that the book, such as it is, may
be of use to the beginner.

Those who desire to pursue the study of the Inverte-
hrata further will do well to consult the excellent treatises
of Yon Siebold,' Gegenbaur,' and Glaus;' and the elabo-

* " Lelirbuch der vergleichenden Anatomie der wirbellosen Thiere," 1818.
One of the best books on the subject ever written, and still indispensable.

2 " Grundzuge der vergleichenden Anatomie," 1870 ; and " Grundriss der
vergleichenden Anatomie," 1874.

3 " Grundzuge der Zoologie." 3tte Auflage, 1876.

6 PREFACE.

rate works of Milne-Edwards* and Bronn,*" in wliicli a
very full Bibliography will be met with. Dr. Eolleston's
valuable " Types of Animal Life," and tlie "Elementary
Instruction in Practical Biology," by myself and Dr.
Martin, will prove useful adjuncts to the appliances of the
practical worker.

* " Legons sur la Physiologie et I'Anatomie comparee de I'Homme et des
Animaux." Tomes i.-xii. (incomplete).

2 " Die Klassen und Ordnungen des Tkierreiclis." Bde. i.-vi. (incomplete).

London, June^ 1877.

CONTENTS.

T±CTS

Preface, 3

Introduction : The General Principles of Biology, . 9

Chap. I. — The Distinctive Characters of Animals, . . . .44

11. — The Protozoa, 73

III. — The Porifera and the Coelenterata, 102

lY. — The Turbellaria, the Rotifera, the Trematoda, and the

Cestoidea, 15Y

V. — The Hirudinea, the Oligoch^ta, the Polych^ta, the

Gephyrea, 189

YI. — The Arthropod a, 219

YII. — The Air-breathing Arthropoda, 320

YIII. — The Polyzoa, the Brachiopoda, and the Mollusca, . . 389

IX. — The Echinodermata, 466

X. — The Tunicata or Ascidioida, 510

XI. — The Peripatidea, the Myzostomata, the Enteropneusta,
the Ch^tognatha, the Nematoidea, the Physemaria,

THE ACANTHOCEPHALA, AND THE DiCYEMIDA, . . . 534

XII. — The Taxonomy of Invertebrated Animals, . . . .561
Index, 589

THE ANATOMY

OF

IISrVERTEBEATED AITIMALS.

INTRODUCTION.

I. — THE GENERAL PRINCIPLES OP BIOLOGY.

The biolof?ical sciences are those which deal with the
phenomena manifested by living matter; and though it is
customary and convenient to group apart such of these phe-
nomena as are termed mental, ar^d such of them as are ex-
hibited by men in society, under the heads of Psychology
and Sociology, yet it must be allowed that no natural boun-
dary separates the subject-matter of the latter sciences from
that of Biology. Psychology is inseparably linked with
Physiology ; and the phases of social life exhibited by ani-
mals other than man, which sometimes curiously foreshadow
human policy, fall strictly within the province of the biolo-
gist.

On the other hand, the biological sciences are sharply
marked off from the abiological, or those which treat of the
phenomena manifested by not-living matter, in so far as the
properties of living matter distinguish it absolutely from all
other kinds of things, and as (the present state of knowledge
furnishes us with no link between the living and the not-
living.)

These distinctive properties of living matter are —

1. Its chemical C07n2)0siti07i — containing, as it invariably
does, one or more forms of a complex compound of carbon,
hydrogen, oxygen, and nitrogen, the so-called protein (which
has never yet been obtained except as a product of living
bodies) united with a large proportion of water, and forming

10 THE ANATOMY OF INVERTEBRATED ANIMALS.

the chief constituent of a substance which, in its primary un-
modified state, is known as protoplasm.

2. Its unwersal disintegration and waste hy oxidation y
and its concomitant reintegration hy the hitussusception of
new matter.

A process of waste resulting from the decomposition of
the molecules of the protoplasm, in virtue of which they
break up into more highly-oxidated products, which cease to
form any part of the living body, is a constant concomitant
of life. There is reason to believe that carbonic acid is al-
ways one of these waste products, while the others contain
the remainder of the carbon, the nitrogen, the hydrogen, and
the other elements which may enter into the composition of
the protoplasm.

The new matter taken in to make good this constant loss
is either a ready-formed protoplasmic material, supplied by
some other living being, or it consists of the elements of
protoplasm, united together in simpler combinations, which
consequently have to be built up into protoplasm by the
agency of the living matter itself. In either case, the addi-
tion of molecules to those which already existed takes place,
not at the surface of the living mass, but by interposition
between the existing molecules of the latter. If the processes
of disintegration and of reconstruction which characterize
life balance one another, the size of the mass of living matter
remains stationary, while, if the reconstructive process is the
more rapid, the living bodj^ grotcs. But the increase of size
which constitutes growth is the result of a process of molec-
ular intussusception, and therefore differs altogether from the
process of growth by accretion, which may be observed in
crystals and is effected purely by the external addition of
new matter — so that, in the M^ell-known aphorism of Linnaeus,*
the word "grow," as applied to stones, signifies a totally dif-
ferent process from what is called "growth" in plants and
animals.

3. Its tendency to undergo cyclical changes.

In the ordinary course of Nature, all living matter proceeds
from preexisting living matter, a portion of the latter being
detached and acquiring an independent existence. The new
form takes on the characters of that from which it arose ; ex-
hibits the same power of propagating itself by means of an
offshoot ; and, sooner or later, like its predecessor, ceases to

1 " Lapides crescunt: vegetdbiUa crescunt et vivunt: animalia crescunt, vi-
vunt et sentiunt."

CHARACTERS OF LIVING MATTER. n

live, and is resolved into more higlily-oxidated compounds of
its elements.

Thus an individual living body is not only constantly
changing its substance, but its size and form are undergoing
continual modifications, the end of which is the death and
decay of that individual ; the continuation of the kind being
secured by the detachment of portions which tend to run
through the same cycle of forms as the parent. No forms of
matter which are either not living, or have not been derived
from living matter, exhibit these three properties, nor any
approach to the remarkable phenomena defined under the sec-
ond and third heads. But, in addition to these distinctive
characters, living matter has some other peculiarities, the
chief of which are the dependence of all its activities upon
moisture and upon heat, within a limited range of tempera-
ture, together with the fact that it usually possesses a certain
structure, or organization.

As has been said, a large proportion of water enters into
the composition of all living matter ; a certain amount of dry-
ing arrests vital activity, and the complete abstraction of this
water is absolutely incompatible with either actual or poten-
tial life. But many of the simpler forms of life may undergo
desiccation to such an extent as to arrest their vital manifes-
tations and convert them into the semblance of not-living
matter, and yet remain potentially alive ; that is to say, on
being duly moistened they return to life again. And this
revivification may take place after months, or even years, of
arrested life.

The properties of living matter are intimately related to
temperature. Not only does exposure to heat sufficient to
decompose protein matter destroy life, by demolishing the
molecular structure upon which life depends ; but all vital
activity, all phenomena of nutritive growth, movement, and
reproduction, are possible only between certain limits of tem-
perature. As the temperature approaches these limits the
manifestations of life vanish, though they may be recovered
by return to the normal conditions ; but, if it pass far beyond
these limits, death takes place.

This much is clear ; but it is not easy to say exactly what
the limits of temperature are, as they appear to vary in part
with the kind of living matter, and in part with the con-
ditions of moisture which obtain along with the temperature.
The conditions of life are so complex in the higher organisms,
that the experimental investigation of this question can be

12 THE ANATOMY OF INTERTEBRATED ANIMALS.

satisfactorily attempted only in the lowest and simplest
forms. It appears that, in the dry state, these are able to
bear far greater extremes both of heat and cold than in the
moist condition. Thus Pasteur found that the spores of fungi,
when dry, could be exposed without destruction to a tem-
perature of 120^-125° C. (248^-257° Fahr.), while the same
spores, when moist, were all killed by exposure to 100° C.
(212° Fahr.). On the other hand, Cagniard de la Tour found
that dry yeast might be exposed to the extremely low tem-
perature of solid carbonic acid ( — 60° Cor —76° Fahr.) with-
out baing killed. In the moist state he found that it might
be frozen and cooled to —5° C. (23° Fahr.), but that it was
killed by lower temperatures. However, it is very desirable
that these experiments should be repeated, for Cohn's careful
observations on Hxcteria show that, though they fall into a
state of torpidity, and, like yeast, lose all their powers of ex-
citing fermentation at, or near, the freezing-point of water,
they are not killed by exposure for five hours to a tempera-
ture below —10° C. (14:° Fahr.), and, for some time, sinking
to —18° C. (— 0°.4 Fahr.). Specimens of Spirillum voiutans^
which had been cooled to this extent, be2'an to move about
some little time after the ice containing them thawed. But
Cohn remarks that Euglenos.f which were frozen along with
them, were all killed and disorganized, and that the same fate
had befallen the higher Infusoria and Itotifera, -with the ex-
ception of some encysted Vorticellce, in which the rhythmical
movements of the contractile vesicle showed that life was
preserved.

Thus it would app3ar that the resistance of living matter
to cold depends greatly on the special form of that matter,
and that the limit of the Euglena^ simple organism as it is, is
much higher than that of the Bacterium.

Considerations of this kind throw some light upon the
apparently anomalous conditions under which many of the
lower plants, such as Protococcns and the Diatomacece^ and
some of th3 lower animals, such as the Radiolaria^ are ob-
served to flourish. Protococcus has been found not only on
the snows of great heights in temperate latitudes, but cover-
ing extensive areas of ice and snow in the Arctic regions,
where it must be exposed to extremely low temperatures —
in the latter case for manv months tosrether : while the Arctic
and Antarctic seas swarm with Diatomacece and Radiolaria.
It is on the Diatomace':e^ as Hooker has well shown, that all
surface-life in these regions ultimately depends ; and their enor-

RESISTANCE TO HEAT AND COLD. 13

mous multitudes prove that their rate of multiplication is ade-
quate to meet the demands made upon them, and is not seri-
ously impeded by the low temperature of the waters, never
much above the freezing-point, in which they habitually live.

The maximum limit of heat which living matter can resist
is no less variable than its minimum limit. Kiihne found
that marine Amcebce were killed when the temperature
reached 35° C. (95° Fahr.), while this was not the case with
iresh-^vaiier j^7)ioebce, which survived a heat of 5°, or even 10°,
C. higher. Actlnophrys JElchhornii was not killed until the
temperature rose to 44° or 45° C. Didymiian serpula is killed
at 35° C. ; while another Myxomycete^ ^thallum septiciun^
succumbs only at 40° C.

Colin (" Untersuchungen liber Bacterien," JBeitrdye zur
l^iologie der JPftanzen^ Heft 2, 1872) has given the results of
a series of experiments conducted with the view of ascertain-
ing the temperature at which bacteria are destroyed when
living in a fluid of definite chemical composition, and free
from all such complications as must arise from the inequalities
of physical condition when solid particles other than the j5ac-
teria coexist with them. The fluid employed contained 0.1
gramme potassium phosphate, 0.1 gr. crystallized magnesium
sulphate, 0.1 gr. tribasic calcium phosphate, and 0.2 gr. am-
monium tartrate, dissolved in 20 cubic centimetres of distilled
water. If to a certain quantity of this " normal fluid " a sn.all
proportion of water containing JBacteria was added, the mul •
tiplication of the Sacteria went on with rapidity, whether the
mouth of the containing flask was open or hermetically closed.
Hermetically-sealed flasks, containing portions of the normal
fluid infected with J^acterla^ were submerged in water heated
to various temperatures, the flask being carefully shaken, with-
out being raised out of the water, during its submergence.

The result was, that in those flasks which were thus sub-
jected, for an hour, to a heat of 60°-62° C. (140°-143° Fahr.),
the Bacteria underwent no development, and the fluid re-
mained perfectly clear. On the other hand, in similar experi-
ments in which the flasks were heated only to 40° or 50° C
{104°-122° Fahr.), the fluid became turbid, in consequence of
the multiplication of the Bacteria^ in the course of from two
to three days.

I am in the habit of demonstrating annually, that Pasteur's
solution and hay-infusion, after five minutes' boilino^ in a flask
properly stopped with cotton-wool, remain perfectly clear of
living organisms, however long they may be kept. The same

14: THE ANATOMY OF INYEKTEBRATED ANIMALS.

holds good for a solution analogous to Cohn's, but in which
all the saline ingredients are ammonia salts ; ^ and in which
Bacteria flourish luxuriantly. Prof. Tyndall's large series
of experiments give the same results for fluids of the most
diverse composition. The cases of milk and some other fluids
in which Bacteria are said to appear, after they have been
heated above the boiling-point, require renewed investigation.

Both in Kuhne's and in Cohn's experiments, w^iich last have
lately been confirmed and extended by Dr. Roberts, of Man-
chester, it was noted that long exposure lo a lower temper-
ature than that which brings about immediate destruction of
life produces the same eflect as short exposure to the latter
temperature. Thus, though all the Bacteria were killed, with
certainty, in the normal fluid, by short exposure to temper-
atures at or above 60° C. (140° Fahr.), Cohn observed that,
when a flask containing infected normal fluid was heated to
50°-52° C. (122°-125° Fahr.) for only an hour, the conse-
quent multiplication of the Bacteria was manifested much
earlier than in one which had been exposed for two hours to
the same temperature.

It appears to be very generally held that the simpler vege-
table organisms are deprived of life at temperatures as high
as 60° C. (140° Fahr.) ; but it is affirmed by competent ob-
servers that Algoi have been found living in hot springs at
much higher temperatures, nam.ely, from 168° to 208° Fahr.,
for which latter surprising fact we have the high authority of
Descloiseaux. It is no explanation of these phenomena, but
only another mode of stating them, to say that these organ-
isms have become " accustomed " to such temperatures. If
this degree of heat were absolutely incompatible with the
activity of living matter, the plants could no more resist it
than they could become " accustomed " to be being made red-
hot. (Habit may modify subsidiary, but cannot aifect funda-
mental, conditions.1

Recent investigations point to the conclusion that the im-
mediate cause of the arrest of vitality, in the first place, and
of its destruction, in the second, is the coagulation of certain
substances in the protoplasm, and that the latter contains
various coagulable matters, which solidify at diff"erent temper-
atures. And it remains to be seen how far the death of any
form of living matter, at a given temperature, depends on the

1 These were as pure as I could obtain them. It is possible the fluid may
have contained an infinitesimal proportion of fixed mineral matter.

RESISTANCE TO HEAT AND COLD. 15

destruction of its fundamental substance at that beat, and
how far death is brought about by the coagulation of merely
accessory compounds.

It may be safely said of all those living things which are
laro-e enough to enable us to trust the evidence of micro-
scopes/ that they are heterogeneous optically, and that their
different parts, and especially the surface layer, as contrasted
with the interior, differ physically and chemically ; while, in
most living things, mere heterogeneity is exchanged for a
definite structure, whereby the body is distinguished into
visibly diverse parts, which possess different powers or func-
tions. Living things which present this visible structure are
said to be organized ; and so widely does organization obtain
among living beings, that organized and living are not unfre-
quently used as if they were terms of coextensive applicabil-
ity. This, however, is not exactly accurate, if it be thereby
implied that all living things have a visible organization, as
there are numerous forms of living matter of which it cannot
properly be said that they possess either a definite visible
structure or permanently specialized organs : though doubt-
less the simplest particle of living matter must possess a
highly-complex molecular structure, which is far beyond the
reach of vision.

The broad distinctions which, as a matter of fact, exist
between every known form of living substance and every other
component of the material world, justify the separation of
the biological sciences from all others. But it must not be '
supposed that the differences between living and not-living
matter are such as to bear out the assumption that the forces j
at work in the one are different from those which are to be/ e
met with in the other. [Considered apart from the phenomena ''■"' "^
of consciousness, the plienomena of life are all dependent
upon the working of the same physical and chemical forces
as those which are active in the rest of the world. It may'
be convenient to use the terms " vitality " and " vital force " to
denote the causes of certain great groups of natural opera-

1 In considering the question of the complication of molecular stracture
which even the smallest and simplest of living beings may possess, it is "O'cll
to recollect that an organic particle totoo of an inch in diameter, in -nhich our
best microscopes may be incompetent to reveal the slightest differentiation of
parts, may be made up of 1,000,000 particles rsnhoor, of an inch in diameter,
while the molecules of matter are probably much less than too^otto of an inch in
diameter. Hence in such a body there is ample scope for any amount of com-
plexity of molecular structure.

16 THE ANATOMY OF INVERTEBRATED ANIMALS^,

I tions, as we employ the names of " electricity " and " electrical
\ force " to denote others ; but it ceases to be proper to do so, if
; such a name implies the absurd assumption that either " elec-
. tricity " or "vitality" is an entity playing the part of an effi-
i cient cause of electrical or vital phenomena. A mass of living
' protoplasm is simply a molecular machine of great complexity,
\ the total results of the working of which, or its vital phenom-
i ena, depend, on the one hand, upon its construction, and, on
I the other, upon the energy supplied to it ; and to speak of
1 " vitality " as anything but the name of a series of operations
I is as if one should talk of the " horologity " of a clocE]

Living matter, or protoplasm and the products of its meta-
morphosis, may be regarded under four aspects :

(1.) It has a certain external and internal form, the laiter
being more usually called structure ;

(2.) It occupies a certain position in space and in time ;

(3.) It is the subject of the operation of certain forces, in
virtue of which it undergoes internal changes, modifies exter-
nal objects, and is modified by them ; and —

(4.) Its form, place, and powers, are the effects of certain
causes.

In correspondence with these four aspects of its subject,
Biology is divisible into four chief subdivisions — I. Moephol-
ogt; II. DiSTEiBUTiox ; III. Physiology; IV. JStiology.

I. MOEFHOLOGY.

So far as living beings have a form and structure, they
fall within the province of Anatomy Siud Histology, the latter
being merely a name for that ultimate optical analysis of
living structure which can be carried out only by the aid of
the microscope.

And, in so far as the form and structure of any living
being- are not constant during the whole of its existence, but
undergo a series of changes from the commencement of that
existence to its end, living beings have a Development. The
history of development is an accuont of the anatomy of a liv-
ing being at the successive periods of its existence, and of the
manner in which one anatomical stage passes into the next.

Finally, the systematic statement and generalization of
the facts of Morphology, in such a manner as to arrange liv-
ing beings in groups, according to their degrees of likeness,
is Taxonomy.

HISTOLOGY. 17

The study of Anatomy and Development has brought to
light certain generalizations of wide applicability and great
importance.

1. It has been said that the great majority of living beings
present a very definite structure. Unassisted vision and or-
dinary dissection suffice to separate the body of any of the
higher animals, or plants, into fabrics of different sorts, which
always present the same general arrangement in the same
organism, but are combined in different ways in different
organisms. The discrimination of these comparatively few
fabrics, or tissues, of which organisms are composed, was the
first step toward that ultimate analysis of visible structure
which has become possible only by the recent perfection of
microscopes and of methods of preparation.

Histology, which embodies the results of this analysis,
shows that every tissue of a plant is composed of more or less
modified structural elements, each of which is termed a cell /
which cell, in its simplest condition, is merely a spheroidal
mass of protoplasm, surrounded by a coat or sac — the cell-
wall — which contains cellulose. In the various tissues, these
cells may undergo innumerable modifications of form — the
protoplasm may become differentiated into a nucleus with its
nucleolus, a primordial utricle, and a cavity filled with a wa-
tery fluid, and the cell-wall may be variously altered in com-
position or in structure, or may coalesce with others. But,
however extensive these changes may be, the fact that the
tissues are made up of morphologically distinct units — the
cells — remains patent. And, if any doubt could exist on the
subject, it would be removed by the study of development,
which proves that every plant commences its existence as a
simple cell, identical in its fundamental characters with the less
modified of those cells of which the whole body is composed.

But it is not necessary to the morphological unit of the
plant that it should be always provided with a cell-wall. Cer-
tain plants, such as Protococcus, spend longer or shorter peri-
ods of their existence in the condition of a mere spheroid of
protoplasm, devoid of any cellulose wall, while, at other times,
the protoplasmic body becomes inclosed within a cell-wall, fab-
ricated by its superficial layer.

Therefore, just as the nucleus, the primordial utricle, and
the central fluid, are no essential constituents of the morpho-
logical unit of the plant, but represent results of its meta-
morphosis, so the cell-wall is equally unessential ; and either
the term " cell " must acquire a merely technical significance

18 THE ANATOMY OF INVERTEBRATED ANIMALS.

as the equivalent of morphological unit, or some new term
must be invented to describe the latter. On the whole, it is
probably least inconvenient to modify the sense of the word
"cell."

The histological analysis of animal tissues has led to sim-
ilar results, and to difficulties of terminology of precisely the
same character. In the higher animals, however, the modifi-
cations which the cells undergo are so extensive that the fact
that the tissues are, as in plants, resolvable into an aggrega-
tion of morphological units, could never have been established
without the aid of the study of development, which proves
that the animal, no less than the plant, commences its exist-
ence as a simple cell, fundamentally identical with the less
modified cells which are found in the tissues of the adult.

Though the nucleus is very constant among animal cells,
it is not universally present ; and, among the lowest forms of
animal life, the protoplasmic mass which represents the mor-
phological unit may be, as in the lowest plants, devoid of a
nucleus. In the animal the cell- wall never has the character
of a shut sac containing cellulose ; and it is not a little diffi-
cult, in many cases, to say how much of the so-called " cell-
wall " of the animal cell answers to the " primordial utricle "
and how much to the proper " cellulose cell-wall " of the vege-
table cell. Bat it is certain that in the animal, as in the
plant, neither cell-wall nor nucleus is an essential constituent
of the cell, inasmuch as bodies which are unquestionably the
equivalents of cells — true morphological units — may be mere
masses of protoplasm, devoid alike of cell-wall and nucleus.

For the whole living world, then, it results : that the mor-
phological unit — the primary and fundamental form of life —
is merely an individual mass of protoplasm, in which no fur-
ther structure is discernible ; that independent living forms
may present but little advance on this structure ; and that all
the higher forms of life are aggregates of such morphological
units or cells variously modified.

Moreover, all that is at present known tends to the conclu-
sion that, in the complex aggregates of such units of which
ail the higher animals and plants consist, no cell has arisen
otherwise than by becoming separated from the protoplasm
of a preexisting cell ; whence the aphorism, " Omnis cellula e
celluldy

It may further be added, as a general truth applicable to
nucleated cells, that the nucleus rarely undergoes any consid-
erable modification, the structures characteristic of the tis-

DEVELOPMENT. X9

sues being formed at the expense of the more superficial pro-
toplasm of the cells ; and that, when nucleated cells divide,
the division of the nucleus, as a rule, precedes that of the
whole cell.

2. In the course of its development every cell proceeds,
from a condition in which it closely resembles every other
cell, through a series of stages of gradually-increasing diver-
gence, until it reaches that condition in which it presents the
characteristic features of the elements of a special tissue.
The development of the cell is, therefore, a gradual progress
from the general to the special state.

The like holds good of the development of the body as a
whole. However complicated one of the higher animals or
plants may be, it begins its separate existence under the
form of a nucleated cell. This, by division, becomes con-
verted into an aggregate of nucleated cells — the parts of this
aggregate, following different laws of grow^th and multiplica-
tion, give rise to the rudiments of the organs ; and the parts
of these rudiments again take on those modes of growth, mul-
tiplication, and metamorphosis, which are needful to convert
the rudiment into the perfect structure.

The development of the organism as a whole, therefore,
repeats in principle the development of the cell. It is a prog-
ress from a general to a special form, resulting from the grad-
ual differentiation of the primitively similar morphological
units of which the body is composed.

Moreover, when the stages of development of two animals
are compared, the number of these stages which are similar
to one another is, as a general rule, proportional to the close-
ness of the resemblance of the adult forms ; whence it fol-
lows that the more closely any two animals are allied in adult
structure, the later are their embrj^onic conditions distinguish-
able. And this general rule holds for plants no less than for
animals.

The broad principle, that the form in which the more com-
plex living things commence their development is always the
same, was first expressed by Harvey in his famous aphorism,
" Omne vivum ex ovo^^'^ which was intended simply as a mor-
phological generalization, and in no wise implied the rejection
of spontaneous generation, as it is commonly supposed to do.
Moreover, Harvey's study of the development of the chick led
him to promulgate that theory of "epigenesis," in which the
doctrine that development is a progress from the general to
the special is implicitly contained.

20 THE ANATOMY OF INVERTEBRATED ANIMALS.

Caspar F. Wolff furnished further, and indeed conclusive,
proof of the truth of the theory of epigenesis ; but, unfortu-
nately, the authority of Haller and the speculations of Bonnet
led science astray, and it was reserved for Von Baer to put the
nature of the process of development in its true light, and to
formulate it in his famous law.

3. Development, then, is a process of differentiation by
which the primitively similar parts of the living body become
more and more unlike one another.

This process of differentiation may be effected in several
ways:

(1.) The protoplasm of the germ may not undergo divi-
sion and conversion into a cell aggregate ; but various parts
of its outer and inner substance may be metamorphosed di-
rectly into those physically and chemically different materials
which constitute the body of the adult. This occurs in such
animals as the Infusoria, and in such plants as the unicellular
A.lgce and Fungi,

(2.) The germ may undergo division, and be converted
into an aggregate of division masses, or llastomeres, which
become cells, and give rise to the tissues by undergoing a
metamorphosis of the same kind as that to which the whole
body is subjected in the preceding case.

The body, formed in either of these ways, may, as a whole,
undergo metamorphosis by differentiation of its parts ; and
this differentiation may take place without reference to any
axis of symmetry, or it may have reference to such an axis.
In the latter case, the parts of the body which become dis-
tinguishable may correspond on the two sides of the axis (bi-
lateral symmetry), or may correspond along several lines paral-
lel with the axis (radial symmetry).

The bilateral or radial symmetry of the body may be fur-
ther complicated by its segmentation, or separation by divi-
sions transverse to the axis, into parts, each of which corre-
sponds with its predecessor or successor in the series.

In the segmented body, the segments may or may not give
rise to symmetrically or asymmetrically disposed processes,
which are appendages, using that word in its most general
sense.

And the highest degree of complication of structure, in
both animals and plants, is attained by the body when it be-
comes divided into segments provided with appendages ; when
the segments not only become very different from one another,
but some coalesce and lose their primitive distinctness ; and

DIFFERENTIATION OF STRUCTURE. 21

when the appendages and the segments into which they are
subdivided similarly become diflferentiated and coalesce.

It is in virtue of such processes that the flowers of plants,
and the heads and limbs of the Arthropoda and of the Ver-
tehrata^ among animals, attain their extraordinary diversity
and complication of structure. A flower-bud is a segmented
body or axis, with a certain number of whorls of appendages ;
and the perfect flower is the result of the gradual difi'erentia-
tion and confluence of these primitively similar segments and
their appendages. The head of an insect or of a crustacean
is, in like manner, composed of a number of segments, each
with its pair of appendages, which by diff"erentiation and con-
fluence are converted into the feelers and variously modified
oral appendages of the adult.

In some complex organisms, the process of difi"erentiation
by which they pass from the condition of aggregated embryo
cells to the adult, can be traced back to the laws of growth
of the two or more cells into which the embryo cell is divided,
each of these cells giving rise to a particular portion of the
adult organism. Thus the fertilized embryo cell in thearche-
gonium of a fern divides into four cells, one of which gives
rise to the rhizome of the young fern, another to its first root-
let, while the other two are converted into a placenta-like
mass which remains imbedded in the prothallus.

The structure of the stem of Chara depends upon the dif-
ferent properties of the cells, which are successively derived
by transverse division from the apical cell. An internodal
cell, which elongates greatly, and does not divide, is suc-
ceeded by a nodal cell, which elongates but little, and becomes
greatly subdivided ; this by another internodal cell, and so
on in regular alternation. In the same way the structure of
the stem, in all the higher plants, depends upon the laws
which govern the manner of division and of metamorphosis
of the apical cells, and of their continuation in the camhiuni
layer.

In all animals which consist of cell-aggregates, the cells
of which the embryo is at first composed arrange themselves
by the splitting, or by a process of invagination, of the blas-
toderm into two layers, the epihlast and the hypohlast^ be-
tween which a third intermediate layer, the mesoblast^ ap-
pears ; and each layer gives rise to a definite group of organs
in the adult. Thus, in the Vertehrata^ the epiblast gives rise
to the cerebro-spinal axis, and to the epidermis and its deriva-
tives ; the hypoblast, to the epithelium of the alimentary

22 THE ANATOMY OF INVERTEBRATED ANIMALS.

canal and its derivatives ; and the mesoblast, to intermediate
structures. The tendency of recent inquiry is to prove that
the several layers of the germ evolve analogous organs in in-
vertebrate animals, and to indicate the possibility of tracing
the several germ-layers back to the blastomeres of the yelk,
from the subdivision of which they proceed.

It is conceivable that all the forms of life should have pre-
sented about the same differentiation of structure, and should
have differed from one another by superficial characters, each
form passing by insensible gradations into those most like it.
In this case Taxonomy^ or the classification of morphological
facts, would have had to confine itself to the formation of a
serial arrangement, representing the serial gradation of these
forms in Nature.

It is conceivable, again, that living beings should have dif-
fered as widely in structure as they actually do, but that the
interval between any two extreme forms should have been
filled up by an unbroken series of gradations ; in which case,
again, classification could only affect the formation of series —
the strict definition of groups would be as impossible as in the
former case.

As a matter of fact, living beings differ enormously, not
only in differentiation of structure, but in the modes in which
that differentiation is brought about ; and the intervals be-
tween extreme forms are not filled up, in the existing world,
by complete series of gradations. Hence it arises that living
beings are, to a great extent, susceptible of classification into
groups, the members of each group resembling one another,
and differing from all the rest, by certain definite peculiarities.

No two living beings are exactly alike, but it is a matter
of observation that, among the endless diversities of living
things, some constantly resemble one another so closely that
it is impossible to draw any line of demarkation between them,
while they differ only in such characters as are associated
with sex. Such as thus closely resemble one another consti-
tute a niorphologianl species ; while different morphological
species are defined by constant characters which are not
merely sexual.

The comparison of these lowest groups, or morphological
species, with one another, shows that more or fewer of them
possess some character or characters in common — some feat-
ure in which they resemble one another and differ from all
other species — and the group or higher order thus formed is

MORPHOLOGICAL GROUPS. 23

a genus. The generic groups thus constituted are susceptible
of being arranged in a similar manner into groups of succes-
sively higher order, which are known as families^ orders,
classes, and the like.

The method pursued in the classification of living forms is,
in fact, exactly the same as that followed by the maker of an
index in working out the heads indexed. In an alphabetical
arrangement, the classification may be truly termed a mor-
phological one, the object being to put into close relation all
those leading words which resemble one another in the
arrangement of their letters, that is, in their form, and to keep
apart those which difi'er in structure. Headings which begin
with the same word, but difi'er otherwise, might be compared
to genera with their species ; the groups of words with the
same first two syllables, to families ; those with identical first
syllables, to orders ; and those with the same initial letter, to
classes. But there is this diff'erence between the index and
the Taxonomic arrangement of living forms, that in the for-
mer there is nothing but an arbitrary relation between the
various classes, while in the latter the classes are similarly
capable of coordination into larger and larger groups, until
all are comprehended under the common definition of living
beings.

The differences between " artificial " and " natural " clas-
sifications are diff'erences in degree, and not in kind. In each
case the classification depends upon likeness ; but in an artifi-
cial classification some prominent and easilj^-observed feature
is taken as the mark of resemblance or dissemblance ; while, in
a natural classification, the things classified are arranged ac-
cording to the totality of their morphological resemblances,
and the features which are taken as the marks of groups are
those which have been ascertained by observation to be the
indications of many likenesses or unlikenesses. And thus a
natural classification is a great deal more than a mere index.
It is a statement of the marks of similarity of organization ;
of the kinds of structure which, as a matter of experience, are
found universally associated together ; and, as such, it fur-
nishes the whole foundation for those indications by which
conclusions as to the nature of the whole of an animal are
drawn from a knowledge of some part of it.

When a paleontologist argues from the characters of a
bone or a shell to the nature of the animal to which that bone
or shell belonged, he is guided by the empirical morphologi-
cal laws established by wide observation, that such a kind of

24 THE ANATOMY OF INVERTEBRATED ANIMALS.

bone or shell is associated with such and such structural feat-
ures in the rest of the body, and no others. And it is these
empirical laws which are embodied and expressed in a natural
classification.

II. DiSTEIBUTIOX.

Living beings occupy certain portions of the surface of
the earth, inhabiting either the dry land, or the fresh or salt
waters; or being competent to maintain their existence in
either. In any given locality, it is found that these different
media are inhabited by different kinds of living beings ; and
that the same medium, at different heights in the air and at
different depths in the water, has different living inhabitants.

Moreover, t!ie living populations of localities which differ
considerably in latitude, and hence in climate, always present
considerable differences. But the converse proposition is not
true — that is to say, localities which differ in longitude, even
if they resemble one another in climate, often have very dis-
similar Faunae and Florm.

It has been discovered, by careful comparison of local fau-
nae and floras, that certain areas of the earth's surface are
inhabited by groups of animals and plants which are not found
elsewhere, and which thus characterize each of these areas.
Such areas are termed Provinces of Distribution. There is
no parit}'- between these provinces in extent, nor in the phys-
ical configuration of their boundaries ; and, in reference to
existing conditions, nothing can appear to be more arbitrary
and capricious than the distribution of living beings.

The study of distribution is not confined to the present
order of Nature ; but, by the help of geology, the naturalist is
enabled to obtain clear, though too fragmentary, evidence of the
characters of the faunae and florae of antecedent epochs. The re-
mains of organisms which are contained in the stratified rocks
prove that, in any given part of the earth's surface, the living
population of earlier epochs was different from that which now
exists in the locality ; and that, on the whole, the difference
becomes greater the farther we go back in time. The organic
remains which are found in the later Cainozoic deposits of any
district are always closely allied to those now found in the
province of distribution in which that locality is included ;
while in the older Cainozoic the resemblance is less ; and in
the Mesozoic, and the Palaeozoic strata, the fossils may be
similar to creatures at present living in some other province,
or may be altogether unlike any which now exist.

DISTRIBUTION IN TIME. 25

In any given locality, the succession of liHng forms may
appear to be interrupted by numerous breaks — the associated
species in each fossiliferous bed being quite distinct from
those above and those below them. But the tendency of all
palaeontological investigation is to show that these breaks are
only apparent, and arise from the incompleteness of the series
of remains which happens to have been preserved in any given
locality. As the area over which accurate geological investi-
gations have been carried on extends, and as the fossiliferous
rocks found in one locality fill up the gaps left in another, so
do the abrupt demarkations between the faunas and floras of
successive epochs disappear — a certain proportion of the gen-
era and even of the species of every period, great or small,
being found to be continued for a longer or shorter time into
the next succeeding period. It is evident, in fact, that the
changes in the living population of the globe which have taken
place during its history have been eflected, not by the sud-
den replacement of one set of living beings by another, but
by a process of slow and gradual introduction of new species,
accompanied by the extinction of the older forms.

It is a remarkable circumstance that, in all parts of the
globe in which fossiliferous rocks have yet been examined,
the successive terms of the series of li\4ng forms which have
thus succeeded one another are analogous. The life of the
Mesozoic epoch is everywhere characterized by the abundance
of some groups of species of which no trace is to be found in
either earlier or later formations ; and the like is true of the
Palaeozoic epoch. Hence it follows, not only that there has
been a succession of species, but that the general nature of
that succession has been the same all over the globe ; and it
is on this ground that fossils are so important to the geologist
as marks of the relative age of rocks.

The determination of the morphological relations of the
species which have thus succeeded one another, is a problem
of profound importance and difficulty, the solution of which,
however, is already clearly indicated. For, in several cases,
it is possible to show that, in the same geographical area, a
form iV, which existed during a certain geological epoch, has
been replaced by another form B, at a later period ; and that
this form B has been replaced, still later, by a third form C.
When these forms. A, B, and C, are comparcl together they
are found to be organized upon the same plan, and to be
very similar even in most of the details of their structure;
but B differs from A by a sli2:ht modification of some of its
2

26 THE ANATOMY OF INYERTEBRATED ANIMALS.

parts, which icodification is carried to a still greater extent
in C.

In ether words, A, B, and C, differ from one another in
the same fashion as the earlier and later stages of the em-
bryo of the same animals differ ; and, in successive epochs,
we have the group presenting that progressive specialization
which characterizes the development of the individual. Clear
evidence that this progressive specialization of structure has
actually occurred has as yet been obtained in only a few cases
(e. g., Equidce^ Crocodilia)^ and these are confined to the
highest and most complicated forms of life ; while it is de-
monstrable that, even as reckoned by geological time, the pro-
cess must have been exceedingly slow.

Among the lower and less complicated forms, the evidence
of progressive modification, furnished by ccmiparison of the
oldest with the latest forms, is slight, or absent ; and some
of these have certainly persisted, wdth very little change,
from extremely ancient times to the present day. It is as
important to recognize the fact that certain forms of life have
thus persisted, as it is to admit that others have undergone
progressive modification.

It has been said that the successive terms in the series of
living forms are analogous in all parts of the globe. But the
species which constitute the corresponding or homotaxic terms
in the series, in different localities, are not identical. And,
though the imperfection of our knowledge at present pre-
cludes positive assertion, there is every reason to believe that
geographical provinces have existed throughout the period
durino: w^hich oro^anic remains furnish us with evidence of the
existence of life. The wide distribution of certain Palaeozoic
forms does not militate against this view ; for the recent in-
vestigations into tlie nature of the deep-sea fauna have shown
that numerous Crustacea, £chi7iodermo.ta, and other inver-
tebrate animals, have as wide a distribution now as their ana-
logues possessed in the Silurian epoch.

III. Physiologt.

Tlius far, living beings have been regarded mereh' as
definite forms of matter, and biology has presented no con-
siderations of a different order from those whicli meet the
student of mineralogy. But living tilings are not only natural
bodies, having a definite form and mode of structure, growth,
and development. They arc machines in action ; and, under

FUNCTIONS AND ORGANS. 27

this aspect, the phenomena which they present have no par-
allel in the mineral world.

The actions of living matter are termed its functions ;
and these functions, varied as they are, may be reduced to
three categories. They are either — (1), functions which affect
the material composition of the body, and determine its mass,
which is the balance of the proces^ses of waste on the one
hand and those of assimilation on the other ; or (2), they are
functions which subserve the process of reproduction, which
is essentially the detachment of a part endowed with the pow-
er of developing into an independent whole ; or (3), they are
functions in virtue of which one part of the body is able to
exert a direct influence on another, and the body, by its parts
or as a whole, becomes a source of molar motion. The first
may be termed sustentative^ the second generative, and the
third correlative functions.

Of these three classes of functions the first two only can
be said to be invariably present in living beings, all of which
are nourished, grow, and multiply. But there are some forms
of life, such as many Fungi^ which are not known to possess
any powers of changing their form ; in which the protoplasm
exhibits no movements, and reacts upon no stimulus ; and in
which any influence which the different parts of the body ex-
ert upon one another must be transmitted indirectly from
molecule to molecule of the common mass. In most of the
lowest plants, however, and in all animals yet known, the
body either constantly or temporarily changes its form, either
with or without the application of a special stimulus, and
thereby modifies the relations of its parts to one another, and
of the whole to surrounding bodies ; while, in all the higher
animals, the different parts of the body are able to affect, and be
affected by one another, by means of a special tissue, termed
nerve. Molar motion is effected on a large scale by means of
another special tissue, muscle / and the organism is brought
into relation with surrounding bodies by means of a third
kind of special tissue — that of the sensory organs — by means
of which the forces exerted by surrounding bodies are trans-
muted into affections of nerve.

In the lo^vest forms of life, the functions which have been
enumerated are seen in their simplest forms, and they are ex-
erted indifferently, or nearly so, by all parts of the proto-
plasmic body ; and the like is true of the functions of the
body of even the highest organisms, so long as they are in
the condition of the nucleated cell, which constitutes the

28 THE ANATOMY OP INVERTEBRATED ANIMALS.

starting-point of their development. But the first process in
that development is the division of the germ into a number
of morphological units or blastomeres, which, eventually, give
rise to cells ; and, as each of these possesses the same physio-
logical functions as the germ itself, it follows that each mor-
phological unit is also a physiological unit, and the multicellu-
lar mass is strictly a compound organism, made up of a mul-
titude of physiologically independent cells. The physiologi-
cal activities manifested by the complex whole represent the
sum, or rather the resultant, of the separate and independent
physiological activities resident in each of the simpler con-
stituents of that whole.

The morphological changes which the cells undergo in
the course of the further development of the organism do
not affect their individuality ; and, notwithstanding the modi-
fication and confluence of its constituent cells, the adult or-
ganism, however complex, is still an aggregate of morphologi-
cal units. Nor is it less an aggregate of physiological units,
each of which retains its fundamental independence, though
that independence becomes restricted in various ways.

Each cell, or that element of a tissue which proceeds from
the modification of a cell, must needs retain its sustentative
functions so long as it grows or maintains a condition of
equilibrium ; but the most completely metamorphosed cells
show no trace of the generative function, and many exhibit
no correlative functions. Contrariwise, those cells of the adult
organism which are the unmetamorphosed derivatives of the
germ exhibit all the primary functions, not only nourishing
themselves and growing, but multiplying, and frequently
showing more or less marked movements.

Organs are parts of the body which perform particular
functions. In strictness, perhaps, it is not quite right to
speak of organs of sustentation or generation, each of these
functions being necessarily performed by the morphological
unit which is nourished or reproduced. What are called the
organs of these functions are the apparatuses by which cer-
tain operations, subsidiary to sustentation and generation, are
carried on.

Thus, in the case of the sustentative functions, all those
organs mny be said to contribute to these functions wliich are
concerned in brinains: nutriment within the reach of the ulti-
mate cells, or in removing waste matter from them ; while in
the case of the generative function, all those organs contribute
to the function whicii produce the cells from which germs are

MUSCLE AND NERYE. 29

given off; or help in the evacution, or fertilization, or develop-
ment, of these germs.

On the other hand, the correlative functions, so long as
they are exerted by a simple undifferentiated morphological
unit or cell, are of the simplest character, consisting of those
modifications of position which can be effected by mere
changes in the form or arrangement of the parts of the pro-
toplasm, or of those prolongations of the protoplasm which
are called pseudopodia or cilia. But, in the higher animals
and plants, the movements of the organism and of its parts
are brought about by the change of the form of certain tis-
sues, the property of which is to shorten in one direction
when exposed to certain stimuli. Such tissues are termed
contractile ; and, in their most fully developed condition,
muscular. The stimulus by which this contraction is natu-
rally brought about is a molecular change, either in the sub-
stance of the contractile tissue itself, or in some other part
of the body ; in which latter case, the motion which is set up
in that part of the body must be propagated to the contractile
tissue through the intermediate substance of the body. In
plants, there seems to be no question that parts which retain
a hardly modified cellular structure may serve as channels for
the transmission of this molecular motion ; wdiether the same
is true of animals is not certain. But, in all the more com-
plex animals, a peculiar fibrous tissue — nerve — serves as the
agent by w4iich contractile tissue is affected by changes oc-
curring elsewhere, and by which contractions thus initiated
are coordinated and brought into harmonious combination.
While the sustentative functions in the higher forms of life
are still, as in the lower, fundamentally dependent upon the
powers inherent in all the physiological units which make up
the body, the correlative functions are, in the former, deputed
to two sets of specially modified units, which constitute the
muscular and the nervous tissues.

When the different forms of life are compared together as
physiological machines, tliey are found to differ as machines
of human construction do. In the lower forms, the mechan-
ism, though perfectly well adapted to do the work for which
it is required, is rough, simple, and weak ; while, in the
higher, it is finished, complicated, and powerful. Considered
as machines, there is the same sort of difference between a
polyp and a horse as there is between a distaff and a spin-
ning-jenny. In the progress from the lower to the higher
organism, there is a gradual differentiation of organs and of

30 THE ANATOMY OF INTERTEBRATED ANIMALS.

functions. Each function is separated into many parts, wbich
are severally intrusted to distinct organs. To use the strik-
ing phrase of Milne-Edwards, in passing from low to high
organisms, there is a division of physiological labor. And
exactly the same process is observable in the development of
any of the higher organisms ; so that, physiologically as well
as morphologically, development is a progress from the gen-
eral to the special.

Thus far, the physiological activities of living matter have
been considered in themselves, and without reference to an}^-
thing that may affect them in the world outside the living
body. But living matter acts on, and is powerfully affected
by, the bodies which surround it; and the study of the in-
fluence of the " conditions of existence " thus determined
constitutes a most important part of physiolog^^

The sustentative functions, for example, can only be ex-
erted under certain conditions of temperature, pressure, and
light, in certain media, and with supplies of particular kinds
of nutritive matter ; the sufficiency of which supplies, again,
is greatly influenced by the competition of other organisms,
which, striving to satisfy the same needs, give rise to the
passive " struggle for existence." The exercise of the correl-
ative functions is influenced by similar conditions, and by the
direct conflict with other organisms, which constitutes the ac-
tive struggle for existence. And, finally, the generative func-
tions are subject to extensive modifications, dependent partly
upon what are commonly called external conditions, and part-
ly upon wholly unknown agencies.

In the lowest forms of life, the only mode of generation
at present known is the division of the bodj" into two or more
parts, each of which then grows to the size and assumes the
form of its parent, and repeats the process of multiplication.
This method of multiplication by fission is properly called
generation, because the parts which are separated are sev-
erally competent to give rise to individual organisms of the
same nature as that from which they arose.

In many of the lowest organisms the process is modified
so far that, instead of the parent dividing into two equal
parts, only a small portion of its substance is detached, as a
bud, which develops into the likeness of its parent. This
is generation by gemination. Generation by fission and by
gemmation is not confined to the simplest forms of life,
however. On the contrary, both modes of multiplication are

AGAMOGENESIS. 31

common not only among plants, but among animals of con-
siderable comp]exit3^

The multiplication of flowering plants by bulbs, tliat of
annelids by fission, and that of polyps by budding, are well-
known examples of these modes of reproduction. In all
these cases, the bud or the segment consists of a multitude
of more or less metamorphosed cells. But, in other in-
stances, a single cell detached from a mass of such undiffer-
entiated cells contained in the parental organism is the foun-
dation of the new organism, and it is hard to say whether such
a detached cell may be more fitly called a bud or a segment
— whether the process is more akin to fission or to gemma-
tion.

In all these cases the development of the new being from
the detached germ takes place without the influence of other
living matter. Common as the process is in plants and in
the lower animals, it becomes rare among the higher animals.
In these, the reproduction of the whole organism from a part,
in the way indicated above, ceases. At most we find that
the cells at the end of an amputated portion of the organism
are capable of reproducing the lost part ; in the very highest
animals, even this power vanishes in the adult ; and, in most
parts of the body, though the undifferentiated cells are
capable of multiplication, their progeny grow, not into whole
organisms like that of which they form a part, but into ele-
ments of the tissues.

Throughout almost the whole series of living beings, how-
ever, we find concurrently with the process of agamog&nesis,
or asexual generation, another method of generation, in which
the development of the germ into an organism resembling
the parent depends on an influence exerted b}' living matter
different from the germ. This is gamogenesis or sexual gen-
eration. Looking at the facts broadly, and without reference
to many exceptions in detail, it may be said that there is an
inverse relation between agamogenetic and gamogenetic re-
production. In the lowest organisms gamogenesis has not
yet been observed, while in the highest agamogenesis is ab-
sent. In many of the lower forms of life agamogenesis is the
common and predominant mode of reproduction, while gamo-
genesis is exceptional ; on the contrary, in many of the high-
er, while gamogenesis is the rule, agamogenesis takes place
exceptionally.

In its simplest condition, which is termed ^^ conjugation^'*
sexual generation consists in the coalescence of two similar

32 THE ANATOMY OF IXTERTEBRATED ANIMALS.

masses of protoplasmic matter, derived from dijGferent parts
of the same organism, or from two organisms of the same
species, and the single mass which results from the fusion
develops into a new organism.

In the majority of cases, however, there is a marked mor-
phological ditference between the two factors in the process,
and then one is called the raale, and the other the female,
element. The female element is relatively large, and under-
goes but little change of form. In all the higher plants and
animals it is a nucleated coll, to which a greater or less
amount of nutritive material, constituting a food-yelk, may
be added.

The male element, on the other hand, is relatively small.
It may be conveyed to the female element by an outgrowth
of the wall of its cell, which is short in many AlgcB and Fungi,
but becomes an immensely elongated tubular filament, in the
case of the pollen-cell of flowering plants. But, more com-
monly, the protoplasm of the male cell becomes converted
into rods or filaments, which usually are in active vibratile
movement, and sometimes are propelled by num.erous cilia.
Occasionally, however, as in manv JSTeinatoidea and Arthro-
poclci, they are devoid of mobility.

The manner in which the contents of the pollen-tube
affect the embryo cell in flowering plants is unknown, as no
perforation through, which the contents of the pollen-tube
may pass, so as actually to mix with the substance of the em-
bryo cell, has been discovered ; and there is the same diffi-
culty with respect to the conjugative processes of some of the
Cryptogennia. But in the great majority of plants, and in
all animals, there can be no doubt that the substance of the
male element actually mixes with that of the female, so
that, in all these cases, the sexual process remains one of con-
jugation ; and impregnation is the physical admixture of pro-
toplasmic matter derived from two sources, which may be
either different parts of the same organism, or different organ-
isms.

The effect of impregnation appears in all cases to be that
the impregnated protoplasm tends to divide into portions
(blcistomeres), which may remain united as a single cell-aggre-
gate, or some or all of which may become separate organ-
isms. A longer or shorter period of rest, in many cases,
intervenes between the act of impregnation and the com-
mencement of the process of division.

As a general rule, the female cell, which directly receives

GAMOGENESIS. 33

the influence of the male, is that which undergoes division
and eventual development into independent germs ; but there
are some plants, such as the FloridecE, in which this is not
the case. In these, the protoplasmic body of the trichogyne,
which unites with the spermatozooids, does not undergo
division itself, but transmits some influence to adjacent cells,
in virtue of which they become subdivided into independent
germs or spores.

There is still much obscurity respecting the reproductive
processes of the Infusoria ; but, in theVorticellidce, it would
appear that conjugation merely determines a condition of the
whole organism, which gives rise to the division of the endo-
plast or so-called nucleus, by which germs are thrown ofl";
and, if this be the case, the process would have some analogy
to what takes place in the FloridecB.

On the other hand, the process of conjugation by which
two distinct Dlporpce combine into that extraordinary double
organism, the Diplozoon paradoxum, does not directly give
rise to germs, but determines the development of the sexual
organs in each of the conjugated individuals; and the same
process takes place in a large number of the Infusoria,, if
what are supposed to be male sexual elements in them are
really such.

The process of impregnation in the Floridem is remark-
ably interesting, from its bearing upon the changes which
fecundation is known to produce upon parts of the parental
organism other than the ovum, even in the highest animals
and plants.

The nature of the influence exerted by the male element
upon the female is wholly unknown. No morphological dis-
tinction can be drawn between those cells which are capable
of reproducing the whole organism without impregnation
and those which need it, as is obvious from what happens in
insects, where eggs which ordinarily require impregnation,
exceptionally, as in many moths, or regularh^, as in the case
of the drones among bees, develop without impregnation.
Even in the higher animals, such as the fowl, the earlier
stages of division of the germ may take place without im-
pregnation.

In fact, generation may be regarded as a particular case
of cell-multiplication, and impregnation simply as one of tlie
many conditions which may determine or affect that process.
In the lowest organisms the simple protoplasmic ma^s divides,
and each part retains all the physiological properties of the

34 THE ANATOMY OF INVERTEBRATED ANIMALS.

whole, and consequently constitutes a germ whence the whole
body can be reproduced. In more advanced organisms each of
the multitude of cells into which the embryo cell is converted
at first, probably retains all, or nearly all, the physiological
capabilities of the whole, and is capable of serving as a re-
productive germ ; but, as division goes on, and many of the
cells which result from division acquire special morphological
and phj'siological properties, it seems not improbable that they,
in proportion, lose their more general characters. In propor-
tion, for example, as the tendency of a giv^en cell to become a
muscle-cell or a cartilage-cell is more marked and definite, it
is readily conceivable that its primitive capacity to reproduce
the whole organism should be reduced, though it might not be
altogether abolished. If this view is well based, the power of
reproducing the whole organism would be limited to those
cells which had acquired no special tendencies, and conse-
quently had retained all the powers of the primitive cell in
which the organism commenced its existence. The more ex-
tensively diffused such cells were, the more generalh' might
multiplication by budding or fission take place ; the more lo-
calized, the more limited would be the parts of the org-anism
in which such a process would take place. And, even w^here
such cells occurred, their development or non-development
might be connected with conditions of nutrition. It depends
en the nutriment supplied to the female larva of a bee wheth-
er it shall become a neuter or a sexually perfect female ; and
the sexual perfection of a large proportion of the internal
parasites is similarly dependent upon their food, and perhaps
on other conditions, such as the temperature of the medium
in which they live. Thus the gradual disappearance of aga-
mo2:enesis in the hig-her animals would be related witli that
increasing specialization of function which is their essential
characteristic ; and, when it ceases to occur altogether, it
may be supposed that no cells are left which retain unmodified
the powers of the primitive embryo cell. The organism is
like a society in which every one is so engrossed by his spe-
cial business that he has neither time nor inclination to marrv.
Even the female elements in the highest organisms, little
as they differ to all appearance from undifferentiated cells,
and though they are directly derived from epithelial cells
which have undergone very little modification from the condi-
tion of blastomeres, are incapable of full development unless
they are subjected to the influence of the male element, which
may, as Caspar Wolff suggested, be compared to a kind of

THE ALTERNATION OF GENERATIONS. 35

nutriment. But it is a living nutriment, in some respects
comparable to that which would be supplied to an animal
kept alive by transfusion, and its molecules transfer to the
impregnated embryo cell all the special characters of the or-
ganism to w^hich it belonged.

The tendency of the germ to reproduce the characters of
its immediate parents, combined, in the case of sexual genera-
tion, with the tendency to reproduce the characters of the
male, is the source of the singular phenomena of hereditary
transmission. No structural modification is so slight, and no
functional peculiarity is so insignificant in either parent, that
it may not make its appearance in the offspring. But the
transmission of parental peculiarities depends greatly upon
the manner in w^hich they have been acquired. Such as have
arisen naturally, and have been hereditary through many an-
tecedent generations, tend to appear in the progeny with
great force ; while artificial modifications — such, for example,
as result from mutilation — are rarely, if ever, transmitted.
Circumcision through innumerable ancestral generations does
not appear to have reduced that rite to a mere formality, as
it should have done if the abbreviated prepuce had become
hereditary in the descendants of Abraham ; while modern
lambs are born with long tails, notwithstanding the long-con-
tinued practice of cutting those of every generation short.
And it remains to be seen whether the supposed hereditary
transmission of the habit of retrieving among dogs is really
what it seems at first sight to be ; on the other side, Brown-
Sequard's case of the transmission of artificially-induced epi-
lepsy in Guinea-pigs is undoubtedly very weighty.

Although the germ always tends to reproduce, directly or
indirectly, the organism from which it is derived, the result
of its development differs somewhat from the parent. Usually
the amount of variation is insignificant ; but it may be con-
siderable, as in the so-called " sports ; " and such variations,
whether useful or useless, may be transmitted with great te-
nacity to the offspring of the subjects of them.

In many plants and animals which multiply both asexually
and sexually there is no definite relation between the aga-
mogenetic and the gamogenetic phenomena. The organism
may multiply asexually before, or after, or concurrently with,
the occurrence of sexual generation.

But in a great many of the lower organisms, both animal
and vegetable, the organism (A) which results from the im-
pregnated germ produces offspring only agamogenetically.

36 THE ANATOMY OF INVERTEBRATED ANIMALS.

It thus gives rise to a series of independent organisms (B,
B, B, . . .), which are more or less different from A, and
which sooner or later acquire generative organs. From their
impregnated germs A is reproduced. The process tlius de-
scribed is what has been termed the " alternation of genera-
tions " under its simplest form — for example, as it is exhibited
by the ScdpcB, In more complicated cases the independent
organisms which correspond with B may give rise agamo-
^eaetically to others (BJ, and these to others (B^), and so
on (e. g., Aphis). Bat, however long the series, a tinal term
appears which develops sexual organs, and reproduces A.
The " alternation of generations " is, therefore, in strictness,
an alternation of asexual with sexual generation, in which
the products of the one process differ from those of the
other.

The Hi/drozoa offer a complete series of gradations be-
tween those cases in which the term B is represented by a
free, self-nourishing organism (e. g., CyanfEO), through those
in which it is free but unable to feed itself ( (Jalycophoridce),
to those in which the sexual elements are developed in bodies
which resemble free zooids, but are never detached, and are
mere generative organs of the body on which they are devel-
oped ( (7i> rdi/loph 0 ra) .

In the last case the " individual " is the total product of
the development of the impregnated embryo, all the parts of
which remain in material continuity with one another. The
multiplication of mouths and stomachs in a Cordylophora no
more makes it an aggregation of different individuals than
the multiplication of segments and legs in a centipede con-
verts that Arthropod into a compound animal. The Cordy-
lophora is a differentiation of a whole into many parts, and
the use of any terminology which implies that it results from
the coalescence of many parts into a whole is to be depre-
cated.

In Cordylophora the generative organs are incapable of
maintaining a separate existence ; but in nearly-allied Hydro-
zoa the unquestionable homologues of these organs become
free zooids, in many cases capable of feeding and growing,
and developing the sexual elements only after they have un-
dergone considerable changes of form. Morphologically, the
swarm of 3fedusrje thus set free from a Hydrozoon are as
much organs of the latter as the multitudinous pinnules of a
Comatida, with their genital glands, are organs of the Echi-
noderm. Morphologically, therefore, the equivalent of the

CAUSES OF THE PflENOMEXA OF LIFE. 37

individual Comatula is the Hydrozoic stock plus all the 3Ie-
dusce which proceed from it.

No doubt it sounds paradoxical to speak of a million of
A.phides^ for example, as parts of one morphological individ-
ual ; but beyond the momentary shock of the paradox no
harm is done. On the other hand, if the asexual Aphides
are held to be individuals, it follows, as a logical consequence,
not only that all the polyps on a Cordylophora tree are
" feeding individuals," and all the genital sacs " generative
individuals," while the stem must be a " stump individual,"
but that the eyes and legs of a lobster are "ocular" and
" locouiotive individuals." And this conception is not only
somewhat more paradoxical than the other, but suggests a
conception of the origin of the complexity 01 animal struct-
ure which is wholly inconsistent with fact.

IV. JEtiologt.

Morphology, distribution, and physiology, investigate and
determine the facts of biology. ^Etiology has for its object
the ascertainment of the causes of these facts, and the ex-
planation of biological phenomena, by showing that they con-
stitute particular cases of general physical laws. It is hardly
needful to say that aetiology, as thus conceived, is in its in-
fancy, and that the seething controversies, to which the
attempt to found this branch of science made in the " Origin
of Species " has given rise, cannot be dealt with in this place.
At most, the general nature of the problems to be solved, and
the course of inquiry needful for their solution, may be indi-
cated.

In any investigation into the causes of the phenomena of
life, the first question which arises is. Whether we have any
knowledge, and if so, what knowledge, of the origin of living
matter ?

In the case of all conspicuous and easily-studied organ-
isms, it has been obvious, since the study of Nature began,
that living beings arise by generation from living beings of
a like kind ; but, before the latter part of the seventeenth cen-
tury, learned and unlearned alike shared the conviction that
this rule was not of universal application, and that multitudes
of the smaller and more obscure organisms were produced by
the fermentation of not-living, and especially of putrefying
dead matter, by what was then termed generatio cequlvoca
or spontanea^ and is now called ahiogenesis. Redi showed

38 THE ANATOMY OF INVERTEBRATED ANIMALS.

that the general belief was erroneous in a multitude of in-
stances ; Spallanzani added largely to the list ; while the in-
vestigations of the scientific helminthologists of the present
century have eliminated a further category of cases in which
it was possible to doubt the applicability of the rule " omne
viviun e vivo'''* to the more complex organisms which consti-
tute the present fauna and flora of the earth. Even the most
extravagant supporters of abiogenesis at the present day do
not pretend that organisms of higher rank than the lowest
Fungi and Protozoa are produced otherwise than by genera-
tion from preexisting organisms. But it is pretended that
Bacteria, Torula^, certain Fungi, and *' Monads," are de-
veloped under conditions which render it impossible that
these organisms should have proceeded directly from living
matter.

The experimental evidence adduced in favor of this prop-
osition is always of one kind, and the reasoning on which
the conclusion that abiogenesis occurs is based may be stated
in the following form :

All living matter is killed by being heated to n degrees.

The contents of a vessel, the entry of germs from without
into which is prevented, have been heated to n degrees.

Therefore, all living matter which may have existed there-
in has been killed.

But living Bacteria, etc., have appeared in these contents
subsequently to their being heated.

Therefore, they have been formed abiogenetically.

No objection can be taken to the logical form of this rea-
soning, but it is obvious that its applicability to any particu-
lar case depends entirely upon the validity, in that case, of
the first and second propositions.

Suppose a fluid to be full of Bacteria in active motion,
what evidence have we that they are killed when that fluid
is heated to n degrees ? There is but one kind of conclusive
evidence, namely, that from that time forth no living Bacteria
make their appearance in the liquid, supposing it to be prop-
erly protected from the intrusion of fresh Bacteria. The
only other evidence, that, for example, which may be fur-
nished by the cessation of the motion of the Bacteria, and
such slight changes as our microscopes permit us to observe
in their optical characters, is simply presumptive evidence of
death, and no more conclusive than the stillness and paleness
of a man in a swoon are proof that he is dead. And the
caution is the more necessary in the case of Bacteria, since

ABIOGENESIS. 39

tnany of them naturally pass a considerable part of their ex-
istence in a condition in which the)^ show no marks of life
whatever save growth and multiplication.

If indeed it could be proved that, in cases which are not
open to doubt, living matter is always and invariably killed
at precisely the same temperature, there might be some
ground for the assumption that, in those which are obscure,
death must take place under the same circumstances. But
what are the facts? It has already been pointed out that,
leaving Bacteria aside, the range of high temperatures be-
tween the lowest, at which some living things are certainly
killed, and the highest, at which others certainly live, is rather
more than 100° Fahr., that is to say, between 104° Fahr. and
208° Fahr. It makes no sort of difference to the argument
how living beings have come to be able to bear such a tem-
perature as the last mentioned ; the fact that they do so is
sufficient to prove that, under certain conditions, such a tem-
perature is not sufficient to destroy life.*

Thus it appears that there is no ground for the assumption
that all living matter is killed at some given temperature be-
tween 104° and 208° Fahr.

No experimental evidence that a liquid may be heated to
n degrees, and yet subsequently give rise to living organisms,
is of the smallest value as proof that abiogenesis has taken
place, and for two reasons : Firstly, there is no proof that
organisms of the kind in question are dead, except their per-
manent incapacity to grow and reproduce their kind ; and,
secondly, since we know that conditions may largely modify
the power of resistance of such organisms to heat, it is far
more probable that such conditions existed in the experiment
in question, than that the organisms were generated afresh
out of dead matter.

Not only is the kind of evidence adduced in favor of
abiogenesis logically insufficient to furnish proof of its occur-
rence, but it may be stated, as a well-based induction, that
the more careful the investigator, and the more complete his
mastery over the endless practical difficulties which surround
experimentation on this subject, the more certain are his ex-
periments to give a negative result ; while positive results
are no less sure to crown the efforts of the clumsy and the
careless.

1 Messrs. Dallinger and Drysdale have recently sho\%Ti ffood grounds for
belie v'ing that the germs of some Monads are not destroyed dj exposure to a
temperature of 280° Fahr. or even 300° Fahr,

40 THE ANATOMY OF INVERTEBKATED ANIMALS.

It is argued that a belief in abiogenesis is a necessary
corollary from the doctrine of Evolution. This may be true
of the occurrence of abiogenesis at some time; but if the
present day, or any recorded epoch of geological time, be in
question, the exact contrary holds good. If _all living beings
have been evolved from preexisting forms of life, it is enough^
that a single particle of living protoplasm should once have
appeared on the globe, as the result of no matter what agency.
In the eyes of a consistent evolutionist, any further indepen-
dent formation of protoplasm would be sheer waste.

The production of living matter since the time of its first
appearance, only by way of biogenesis, implies that the spe-
cific forms of the lower kinds of life have undergone but little
change in the course of geological time, and this is said to be
inconsistent with the doctrine of evolution. But, in the first
place, the fact is not inconsistent with the doctrine of evolu-
tion properly understood, that doctrine being perfectly con-
sistent with either the progression, the retrogression, or the
stationary condition, of any particular species for indefinite
periods of time ; and, secondl}^ if it were, it would be so much
the worse for the doctrine of evolution, inasmuch as it is un-
questionably true that certain, even highly-organized, forms
of life have persisted without any sensible change for very
long periods. The Terehratula psittacea of the present day,
for example, is not distinguishable from that of the Cretaceous
epoch, while the highly-organized Teleostean fish, JBeryx^ of
the Chalk, differed only in minute specific characters from
that which now lives. Is it seriously suggested that the ex-
isting lerehratulcE and ISeryces are not the lineal descendants
of their Cretaceous ancestors, but that their modern repre-
sentatives have been independently developed from primordial
germs in the interval ? But if this is too fantastic a sugges-
tion for grave consideration, why are we to believe that tbe
Glohigerince of the present day are not lineally descended
from the Cretaceous forms ? And, if their unchanged genera-
tions have succeeded one another for all the enormous time
represented by the deposition of the Chalk and that of the
Tertiary and Quaternary deposits, what difficulty is there in
supposing that they may not have persisted unchanged for a
greatly longer period ?

The fact is, that at the present moment there is not a
shadow of trustworthy direct evidence that abiogenesis does
take place, or has taken place, within the period during
which the existence of life on the globe is recorded. But it

ORIGIN OF SPECIES. 41

need hardly be pointed out that the fact does not in the,
slightest degree interfere with any conclusion that may be i^ nr,
arrived at, deductively, from other considerations that, at
some time or other, abiogenesis must have taken place.^

If the hypothesis of evolution is true, living matter must
have arisen from not-living matter ; for, by the hypothesis,
the condition of the globe was at one time such that living
matter could not have existed in it,^ life being entirely in-
compatible with the gaseous state. But, living matter once
originated, there is no necessity for another origination, since
the hypothesis postulates the unlimited, though perhaps not
indefinite, modifiability of such matter.

Of the causes which have led to flie origination of living
matter, then, it may be said that we know absolutely nothing.
But postulating the existence of living matter endowed with
that power of hereditary transmission, and with that tendency
to vary which is found in all such matter, Mr. Darwin has
shown good reasons for believing that the interaction betw^een
living matter and surrounding conditions, which results in
the survival of the fittest, is sufficient to account for the
gradual evolution of plants and animals from their simplest
to their most complicated forms, and for the known phe-
nomena of Morphology, Physiology, and Distribution.

Mr. Darw^in has further endeavored to give a physical
explanation of hereditary transmission by his hypothesis
of Pangenesis ; while he seeks for the principal, if not the
only cause of variation in the influence of changing condi-
tions.

It is on this point that the chief divergence exists among
those who accept the doctrine of evolution in its general
outlines. Three views may be taken of the causes of varia-
tion :

a. In virtue of its molecular structure, the organism may
tend to vary. This variability may either be indefinite, or
may be limited to certain directions by intrinsic conditions.
In the former case, the result of the struggle for existence
would be the survival of the fittest among an indefinite
number of varieties ; in the latter case, it would be the
survival of the fittest among a certain set of varieties, the

1 It makes no difference if we adopt Sir TV. Thomson's hvpothesis, and
suppose that the germs of living things have been transported to our globe
from some other, seeing that there is as much reason for supposing that all
stellar and planetary components of the universe are or have been gaseous, as
that the earth has passed through this stage.

42 THE ANATOMY OF INVERTEBRATED ANIMALS.

nature and number of which would be predetermined by the
molecular structure of the organism.

h. The organism may have no intrinsic tendency to vary,
but variation may be brought about by the influence of con-
ditions external to it. And in this case, also, the variability
induced may be either indefinite or defined by intrinsic limi-
tation.

c. The two former cases may be combined, and variation
may to some extent depend upon intrinsic, and to some ex-
tent upon extrinsic, conditions.

At present it can hardly be said that such evidence as
would justify the positive adoption of any one of these views
exists.

If all living beings have come into existence by the gradual
modification, through a long series of generations, of a pri-
mordial living matter, the phenomena of embryonic develop-
ment ought to be explicable as particular cases of the general
law of hereditary transmission. On this view, a tadpole is
first a fish, and then a tailed amphibian, provided with both
gills and lungs, before it becomes a frog, because tlie frog
was the last term in a series of modifications whereby some
ancient fish became a urodele amphibian; and the urodele
amphibian became an anurous amphibian. In fact, the de-
velopment of the embryo is a recapitulation of the ancestral
history of the species.

If this be so, it follows that the development of any
organism should furnish the key to its ancestral history ; and
the attempt to decipher the full pedigree of organisms from
so much of the family history as is recorded in their develop-
ment has given rise to a special branch of biological specula-
tion, termed pliylogeny.

In practice, however, the reconstruction of the pedigree of
a group from the developmental history of its existing mem-
bers is fraught with difficulties. It is highly probable that
the series of developmental stages of the individual organism
never presents more than an abbreviated and condensed sum-
mary of ancestral conditions ; while this summary is often
strangely modified by variation and adaptation to conditions ;
and it must be confessed that, in most cases, we can do little
better than guess what is genuine recapitulation of ancestral
forms, and what is the effect of comparatively late adapta-
tion.

The only perfectly safe foundation for the doctrine of evolu-
tion lies in the historical, or rather archagological, evidence

PHYLOGENY. 43

that particular organisms have arisen by the gradual modifi-
cation of their predecessors, which is furnished by fossil
remains. That evidence is daily increasing in amount and in
weight ; and it is to be hoped that the comparison of the
actual pedigree of these organisms with the phenomena of
tbeir development may furnish some criterion by which the
validity of phylogenetic conclusions, deduced from the facts
of embryology alone, may be satisfactorily tested.

* " ' t

CHAPTER I.

I. THE DISTINCTIVE CHAEACTERS OF ANIMALS.

The more complicated forms of the living things, the
general characters of which have now been discussed, appear
to be readily distinguishable into widely-separated groups,
animals, and plants. The latter have no power of locomo-
tion, and only rarely exhibit any distinct movement of their
parts when these are irritated, mechanically or otherwise.
They are devoid of any digestive cavity; and the matters
which serve as their nutriment are absorbed in the gaseous
and fluid state. Ordinary animals, on the contrary, not only
possess conspicuous locomotive activit}^, but their parts
readily alter their form or position when irritated. Their
nutriment, consisting of other animals or of plants, is taken
in the solid form into a digestive cavity.

But even without descending to the very lowest forms of
animals and plants, we meet with facts which weaken the
force of these apparently broad distinctions. Among animals,
a coral or an oyster is as incapable of locomotion as an oak;
arid a tape-worm feeds by imbibition and not by the ingestion
of solid matter. On the other hand, the Sensitive-Plant and
the Sundew exhibit movements on irritation, and the recent
observations of Mr. Darwin and others leave little doubt that
the so-called " insectivorous plants " really digest and assimi-
late the nutritive matters contained in the living animals
which they catch and destroy. All the higher animals are
dependent for the protein compounds which they contain
upon other animals or upon plants. They are unable to man-
ufacture protein out of simpler substances ; and, although
positive proof is wanting that this incapacity extends to all
animals, it may safely be assumed to exist in all those forms
of animal life which take in solid nutriment, or which live
parasitically on other animals or plants, in situations in which
they are provided with abundant supphes of protein in a
dissolved state.

THE DISTINCTIVE CHARACTERS OF ANIMALS. 45

The great majority of the higher plants, on the contrary,
are able to manufacture protein when supplied with carbonic
acid, ammoniacal salts, water, and sundry mineral phosphates
and sulphates, obtaining the carbon which they require by
the decomposition of the carbonic acid, the oxygen of which
is disengaged. One essential factor in the performance of
this remarkable chemical process is the chlorophyll which
these plants contain, and another is the sun's light.

Certain animals {^Infusoria, Coelenterata, TarhellaricL)
possess chlorophyll, but there is no evidence to show what
part it plays in their economy. Some of the higher plants
when parasitic, and a great group of the lower plants, the
Fungi (which may be parasitic or not), are, however, devoid
of chlorophyll, and are consequently totally unable to derive
the carbon which they need from carbonic acid. Nevertheless
they are sharply distinguished from animals, inasmuch as they
are still, for the most part, manufacturers of protein. Thus
such a Fungus as Penicillium is able to fabricate all the con-
stituents of its body out of ammonium tartrate, sulphate, and
phosphate, dissolved in water (see siqyra, p. 14, note) ; and
the yeast-plant flourishes and multiplies with exceeding rapid-
ity in water containing sugar, ammonium tartrate, potassium
phosphate, calcium phosphate, and magnesium sulphate.

Nevertheless, the experiments of Mayer have shown that
when peptones are substituted for the ammonium tartrate,
the nutrition of the yeast-plant is favored instead of being
impeded. So that it would seem that the yeast-plant is able
to take in protein compounds and assimilate them, as if it
were an animal ; and there can be no reasonable doubt that
many parasitic Fungi, such as the Botrytis Bassiana of the
silk-worm caterpillar, the Empusa of the house-fly, and, very
probably, the Peronospora of the potato-plant, directly as-
similate the protein substances contained in the bodies of the
plants and animals which they infest ; nor is it clear that
these Fungi are able to maintain themselves upon less fully
elaborated nutriment.

Cellulose, amyloid, and saccharine compounds were former-
ly supposed to be characteristically vegetable products ; but
cellulose is found in the tests of Ascidians; and amyloid and
saccharine matters are of very wide, if not universal, occur-
rence in animals.

And on taking a comprehensive survey of the whole ani-
mal and vegetable w^orlds, the test of locomotion breaks down
as completely as does that of nutrition. For it is the rule

46 THE ANATOMY OF INVERTEBRATED ANIMALS.

rather than the exception among the lowest plants, that at
one stage or other of their existence they should be actively
locomotive, their motor organs being usually cilla^ altogether
similar in character and function to the motor organs of the
lowest animals. Moreover, the protoplasmic substance of the
body in many of these plants exhibits rhythmically pulsating
spaces or contractile vacuoles of the same nature as those
characteristic of so many animals.

No better illustration of the impossibility of drawing any
sharply-defined distinction between animals and plants can be
found than that which is supplied by the history of what are
commonly termed '"Monads,"

The name of "Monad"* has been commonly applied to
minute free or fixed, rounded or oval bodies, provided with
one or more long cilia {flagella), and usually provided with
a nucleus and a contractile vacuole. Of such bodies, all of
which would properly come under the old group of Motiadi-
dcBy the history of a few has been completely worked out ;
and the result is that, while some (e. g., Chlamydomonas,
zoospores of Peronospora and Coleochcete) are locomotive
conditions of indubitable plants, others {Hadiolaria, Nocti-
lucci) are embryonic conditions of as indubitable animals.
Yet others (zoospores of 3Iyxomyc€tes) are embryonic forms
of organisms which appear to be as much animals as plants ;
inasmuch as in one condition they take in solid nutriment,
and in another have the special morphological, if not physio-
logical peculiarities of plants; while, lastly, in the case of
such monads as those recently so carefully studied by Messrs.
Dallinger and Drysdale, the morphological characters of which
are on the whole animal, while their mode of nutrition is un-
known, it is impossible to say whether they should be regarded
as animals or as plants.

Thus, traced down to their lowest terms, the series of
plant forms gradually lose more and more of their distinctive
vegetable features, while the series of animal forms part with
more and more of their distinctive animal characters, and the
two series converge to a common term. Thfi most character-
istic morphological peculiarity of the plant is the investment
of each of its component cells by a sac, the walls of which
contain cellulose, or some closely analogous compound ; and

- 0. F. Muller, " Historia Vermium," 1773. " Vermis inconspicuus, sim-
plicissimus, pellucidus, punctiformis."

MORPHOLOGICAL DIFFERENTIATION. 47

the most characteristic physiological peculiarity of the plant
is its power of manufacturing protein from chemical com- i
pounds of a less complex nature.

The most characteristic morphological peculiarity of the i
animal is the absence of any such cellulose investment/ The '
most characteristic physiological peculiarity of the animal is
its want of power to manufacture protein out of simpler
compounds.

The great majority of living things are at once referable
to one of the two categories thus defined ; but there are some
in which the presence of one or other characteristic mark
cannot be ascertained, and others which appear at different
periods of their existence to belong to different categories.

n. THE MOEPHOLOGICAL DIFFEEEXTIATION OF A:NIMALS.

The simplest form of animal life imaginable would be a
protoplasmic body, devoid of motility, maintaining itself by
the ingestion of such proteinaceous, fatty, amyloid, and min-
eral matters as might be brought into contact with it by ex-
ternal agencies ; and increasing by simple extension of its
mass. But no animal of this degree of simplicity is known
to exist. The very humblest animals with which we are ac-
quainted exhibit contractility, and not only increase in size,
but, as they grow, divide, and thus undergo multiplication.
In the simplest known animals — the Protozoa — the proto-
plasmic substance of the body does not become differentiated
into discrete nucleated masses or cells, which by their meta-
morphosis give rise to the different tissues of which the adult
body is composed. And, in the low^est of the Protozoa^ the
body has neither a constant form nor any further distinction
of parts than a greater density of the peripheral, as com-
pared with the central, part of the protoplasm. The first
steps in complication are the appearance of one or more
rhythmically contractile vacuoles, such as are found in some
of the lower plants ; and the segregation of part of the in-

1 No analysis of the substance composing the cysts in -which so many of the
Protozoa inclose themselves temporarily has yet been made. But it is not im-
probable that it may be analogous to cliitin /"and, if so, it is 's\-ortby of remark
that, though cbitin is a nitrogenous body, it readily yields a substance ajipar-
ently identical with cellulose when heated with the double hyposulphite of
copper and ammonia. It is possible, therefore, that the difference between
the chitinous investment of an animal and the cellulose investment of a plant
may depend upon the proportion of nitrogenous matter which is present in
each case in addition to the chitiu.

48 THE ANATOMY OF INVERTEBRATED ANIMALS.

terlor protoplasm as a rounded mass, the " endoplast " or
" nucleus." Other Protozoa advance further and acquire
permanent locomotive organs. These may be developed
only on one part of the surface of the body, which may be
modified into a special organ for their support. In some, a
pedicle of attachment is formed, and the body may acquire a
dense envelope {Infusoria), or secrete an internal skeleton of
calcareous or silicious matter {Forarninifera, Itadiolaria), or
fabricate such a skeleton by gluing together extraneous par-
ticles [Forambiifera).

A mouth and gullet, with an anal aperture, may be formed,
and the permeable soft central portion of the protoplasm may
be so limited as to give rise to a virtual alimentary tract be-
tween these two apertures. The contractile vacuole may be
developed into a complicated system of canals (Paramceci-
um)y and the endoplast may take on more and more definite-
ly the characters of a reproductive organ, that is, may be the
focus of origin of germs capable of reproducing the individ-
ual ( Vorticella), In fact, rudiments of all the chief system
of organs of the higher animals, with the exception, more or
less doubtful, of the nervous, are thus sketched out in the
Protozoa, just as the organs of the higher plants are sketched
out in Cauler]}a,

In the Metazoa, which constitute the rest of the animal
kingdom, the animal, in its earliest condition, is a protoplas-
mic mass with a nucleus — is, in short, a Protozoon. But it
never acquires the morphological complexity of its adult state
by the direct metamorphosis of the protoplasmic matter of
this nucleated body — the ovum — into the different tissues.
On the contrary, the first step in the development of all the
Metazoa is the conversion of the single nucleated body into
an a <?2:re oration of such bodies of smaller size — the Morula —
by a process of division, which usually takes place with great
regularity, the ovum dividing first into two segments, which
then subdivide, giving rise to four, eight, sixteen, etc.,
portions, which are the so-called division masses or blasto-
meres.

A similar process takes place in sundry Protozoa and gives
rise to a protozoic aggregate, which is strictly comparable to
the Morxda. But the members of the protozoic aggregate
become separate, or at any rate independent existences.
What distiniruishes the metazoic asrarreorate is that, thoucfh its
component blastomeres also retain a certain degree of physi-
ological independence, they remain united into one morpho-

MORPHOLOGICAL DIFFERENTIATION. 49

logical whole, and their several metamorphoses are so ordered
and related to one another that they constitute members of a
mutually dependent commonalty.

The Metazoa are the only animals which fall under com-
mon observation, and have therefore been known from the
earliest times. All the higher languages possess general
names equivalent to our beast, bird, reptile, fish, insect, and
worm ; and this shows the very early perception of the fact
that, notwithstanding the wonderful diversity of animal forms,
they are modeled upon comparatively few great types.

In the middle of the last century the founder of modern
Taxonomy, Linnaeus, distinguished animals into Mammalia^
Aves, Amphibia, Pisces, Insecta, and Vermes, that is to say,
he converted common-sense into science by defining and giv-
ing precision to the rough distinctions arrived at by ordinary
observation.

At the end of the century, Lamarck made a most impor-
tant advance in general morphologj'', by pointing out that
mammals, birds, reptiles, and fishes, are formed upon one type
or comm^on p)lan, the essential character of which is the pos-
session of a spinal column, interposed between a cerebro-spi-
nal and a visceral cavit}' ; and that in no other animals is the
same plan of construction to be discerned. Hence he drew a
broad distinction between the former and the latter, as the
Vertebrata and the I]st^ertebrata. But the advance of
knowledge respecting the structure of invertebrated animals,
due chiefly to Swammerdam, Trembley, Reaumur, Peyssonel,
Goeze, Roesel, Ellis, Fabricius, O. F. Miiller, Lyonet, Pallas,
and Cuvier, speedily proved that the Invertebrata are not
framed upon one fundamental plan, but upon several ; and,
in 1795, Cuvier ^ showed that, at fewest, three morphological
types, as distinct from one another as they are from that of
the vertebrated animals, are distinguishable among the I?i-
vertebrata. These he named — I. Mollusques ; II. Insectes et
Vers ; III. Zoophytes. In the " Rescue animal " (1816), those
terms are Latinized, Animalia Mollusca, Articulata, and Ma-
diata. Thus, says Cuvier : " It will be found that there ex-
ist four principal forms, four general plans, if it may thus be
expressed, on which all animals appear to have been modeled ;
and the ulterior divisions of which, under whatever title natu-
ralists may have designated them, are merely slight modifica-
tions, founded on the development or addition of certain parts.

» Tableau ^lemeutaire de I'Histoire des Animaux, An vi.

3

50 THE ANATOMY OF INVERTEBRATED ANIMALS.

These four common plans are those of the Vertehrata, the MdU
lusca, the Articulata, and the Hadiata.^''

I For extent, variety, and exactness of knowledge, Cuvier
was, beyond all comparison, the greatest anatomist who has
ever lived ; but the absence of two conditions rendered if
1 impossible that his survey of the animal kingdom should be
' exhaustive, grand and comprehensive as it was.

Up to the time of Cuvier's death in 1832, microscopic in-
vestigation was in its infancy, and hence the great majority
of the lowest forms were either unknown or little understood;
and it was only in the third decade of the present century
that Rathke, Dollinger, and Von Baer, commenced that won-
derful series of exact researches into embryology which Von
Baer organized into a special branch of morphology, develop-
ing all its most important consequences and raising it to its
proper position, as the criterion of morphological theories.

Upon embryological grounds Von Baer arrived at the
same conclusion as Cuvier, that there are four ccmmon plans
of animal structure.

In the course of the last half-century the activity of anat-
omists and embryologists has been prodigious, and it may
be reasonably doubted whether any form of animal life re-
mains to be discovered which will not be found to accord
with one or other of the common plans now known. But at
the same time this increase of knowledge has abolished the
broad lines of demarkation which formerly appeared to sepa-
rate one common plan from another.

Even the hiatus between the Vertehrata and Ihe Tnver-
tehrata is partly, if not wholly, bridged over; and though
among the Invertehrata there is no difficulty in distinguish-
ing the more completely differentiated representatives of
such types or common plans as those of the Arthropoda^ the
A7ineUda, the 3Iollusca, the Tu7iicata, the Ecliinodermata^
the Coelenterata^ and the Porifera^ yet everj- year brings
forth fresh evidence to the effect that, just as the plan of the
plant is not absolutely distinct from that of the animal, so
that of the Vertebrate has its points of community with that
of certain of the Invertebrates ; that the Arthropod, the Mol-
lusk, and the Ecliinoderm plans are united by that of the
lower worms; and that the plan of the latter is separated by
no very great differences from that of the Coelenterate and
that of the Sponge.

Whatever speculative views may be held or rejected as to
ithe origin of the diversities of animal form, the facts of anat-

ANNULOSE DIFFERENTIATION. 51

omy and development compel the morphologist to regard
the whole of the Metazoa as modifications of one actual or \
ideal primitive type, which is a sac with a double cellular 1
wall, inclosing a central cavity and open at one end. This is )
what Haeckel terms a Gastrcea. The inner wall of the sac is
the hypoblast {endoderm of the adult), the outer the epiblast
.(ectoderm). Between the two, in all but the very lowest
Metazoa, a third layer, the mesoblast {mesoderm of the adult),
makes its appearance.

In the Porifera, the terminal aperture of the gastrsea
becomes the egestive opening of the adult animal, and the
ingestive apertures are numerous secondary pore-like aper-
tures formed by the separation of adjacent cells of the ec-
toderm and endoderm. The body may become variously
branched, a fibrous or spicular endoskeleton is usually de-
veloped in the ectoderm, and no perivisceral cavity is de-
veloped. There are no appendages for locomotion or pre-
hension ; no nervous system nor sensory organs are known to
exist ; nor are there any circulatory, respiratory, renal, or
generative organs.

In the Coeletiterata, the terminal aperture of the gastraea
becomes the mouth, and, if pores perforate the body-walls,
they do not subserve the ingestion of food. There is no sep-
arate perivisceral cavity, but, in man}'-, an enterocoele or sys-
tem of cavities, continuous with, but more or less separate
from, the digestive cavity, extends through the body. Pre-
hensile appendages, tefitacula, are developed in great variety.
A chitinous exoskeleton appears in some, a calcareous or chit-
iiious endoskeleton in others. There are no circulatory, re-
spiratory, or renal organs (though it is possible that certain
cells in the PorpitCG, e. g., may have a uropoietic function);
but special genital organs make their appearance, as do a
definitely-arranged nervous system and organs of sense.

The lowest Turhellaria are on nearly the same grade of
organization as the lower Coelenterata, but the thick meso-
derm is traversed by canals which constitute a loater-vascular
system. In the adult state these canals open, on the one side,
into the interstices of the mesodermal tissues, and, on the
other, communicate with the exterior. Their analogy to the
contractile vacuoles of the Infusoria on the one hand, and to
the segmental organs of the Annelids on the other, lead me
to think that they are formed by a splitting of the mesoblast,
and that they thus represent that form of perivisceral cavity
which I have termed a schizocoele, A nervous system, cou^

52 THE ANATOMY OF IXVERTEBRATED ANIMALS.

sisting of a single or double gangliou with two principal lon-
gitudinal nerve-cords, is found in many ; and there may be
eyes and auditory sacs.

Upon this foundation a gradual complication of form is
based, brought about by —

1. The elongation of the bilaterally symmetrical body and
the formation of a chitinous exoskeleton.

2. The development of a secondary aperture near the an-
terior end of the body, which becomes the permanent mouth.

3. The division of the mesoblast into successive segments
{somites).

4. The development of two nervous ganglia in each somite.

5. The outgrowth of a pair of appendages from each so-
mite, and their segmentation.

6. The gradual specialization of the somites into cephalic,
thoracic and abdominal groups ; and that of their appendages
into sense organs, jaws, locomotive limbs, and respiratory or-
gans.

7. The conversion of the schizocoele into a spacious peri-
visceral cavity containing blood ; the reduction of the water-
vascular system, and the appearance of pseudo-haemal vessels ;
and the replacement of these, in the higher forms, by a heart,
arteries, and veins, which contain blood.

8. The conversion of the simple inner sac of the gastra^a
into a hig-hly-complex alimentary canal, with special glandu-
lar appendages, representing the liver and the kidneys.

9. A similar differentiation of the genital apparatus.

10. A gradual complication of the eye, which, in its most
perfect form, presents a series of crystal-clear conical rods,
disposed perpendicularly to the transparent corneal region
of the chitinous exoskeleton, and connected by their inner
ends with the optic nerves of the pr^-oesophageal ganglia.

By such modifications as these the plan of the simple
Turbellarian gradually passes into that of the highest Ar-
thropod.

Starting from the same point, if the mesoblast does not
become distinctly segmented i if few, probably not more
than three, pairs of ganglia are formed ; if there are no seg-
mented appendages, but the chief locomotive organ is a mus-
cular foot developed in the neural aspect of the bod}^; if, in
the place of the chitinous exoskeleton, a shell is secreted by
a specially modified part of the haemal wall termed the man-
tle ; if the schizocoele is converted into a blood-cavity, which
communicates with the exterior by an organ of Bojanus, which

THE PLAN OF THE . ECHINODERMS. 53

appears to represent the water-vascular system and the seg-
mental organs ; and if, along with these changes, the aliment-
ary, circulatory, respiratory, genital, and sensory organs take
on special characters, we arrive at the complete Molluscan
plan.

From the Turbellarian to the Tunicate, or Ascidian, the
passage is indicated, if not effected, by J^alanoglofsus, which,
in its larval state, is comparable to an Appendicular ia with-
out its caudal appendage. On the other hand, the large
pharynx of the Tunicata and the circle of tentacula around
the oral aperture, with the single ganglion, approximate them
to the Polyzoa. In the perforation of the pharynx by lateral
apertures, which communicate with the exterior, either di-
rectly or by the intermediation of an atrial cavity, the Tuni-
cata resemble only Balanoglossus and tlie Vertebrata. The
axial skeleton of the caudal appendage has no parallel except
in the vertebrate notochord. In the structure of the heart
and the regular reversal of the direction of its contractions,
the Tunicata stand alone. The general presence of a test
solidified by cellulose is a marked peculiarity, but in esti-
mating its apparent singularity the existence of cellulose as
a constituent of chitin must be remembered. Finally, the
tadpole-like larvae of many Ascidians are comparable o\\\y to
the Cercarioe of Trematodes, on the one hand, and to ver-
tebrate larval forms on the other.

Yet another apparently very distinct type is met with in
the extensive group of the Echinodermata.

In all the other Metazoa, except the Porifera and Coelen-
terata, the plan of the body is, obviously, bilaterally sym-
metrical, the halves of the body on each side of a median ver-
tical plane being similar. Any disturbance of this symmetrv,
such as is found in some Artliropoda and in many Mollusca,
arises from the predominant development of one half. But,
in a Sea-urchin or Starfish, five or more similar sets of parts
are disposed around a longitudinal axis, which has the mouth
at one end and the anus at the other ; there is a radial sym-
metry, as in a sea-anemone or a Ctenophoran. Nevertheless,
close observation shows that, as is also the case in the Actinia
or Ctenophoran, this radial symmetry is never perfect, and
that the body is really bilaterally symmetrical in relation to
a median plane which traverses the centre of length of one
of the radiating metameres.

Another marked peculiarity of the Echinoderm type is

54 THE ANATOMY OF INVERTEBRATED ANIMALS.

the general, if not universal, presence of a system of " am-
bulacral vessels" consistino- of a circular canal around the
mouth, whence canals usually arise and follow the middle line
of each of the ambulacral metameres. And, in the typical
Echinoderm, these canals give off prolongations which enter
certain diverticula of the body-wall, the pedicels or suckers.

All Echinoderms have a calcareous endoskeleton.

In the chapter allotted to these animals, it will be shown
that they are modifications of the Turbellarian type, brought
about by a singular series of changes undergone by the endo-
derm and mesoderm of the larva or Echinoposdium.

III. — THE PHYSIOLOGICAL DIFFEEENTIATI0:N^ OF AI^IMALS, AND
THE MOKPHOLOGICAL DIFFERENTIATION OF THEIR ORGANS.

Regarded as machines for doing certain kinds of work,
animals differ from one another in the extent to which this
work is subdivided. Each subordinate group of actions or
functions is allotted to a particular portion of the body, which
thus becomes the organ of those functions ; and the extent
to which this division of physiological labor is carried differs
in degree within the limits of each common plan, and is the
chief cause of the diversity in the working out of the common
plan of a group exhibited by its members. Moreover, there
are certain types which never attain the same degree of physi-
ological differentiation as others do.

Thus, some of the Protozoa attain a grade of physiological
complexity as high as that which is reached by the lower Me-
tazoa. And, notwithstanding the multiplicity of its parts, no
Echinoderm is so highly differentiated a physiological ma-
chine as is a snail.

A mill with ten pairs of millstones need not be a more
complicated machine than a mill with one pair ; but if a mill
have two pairs of millstones, one for coarse and one for fine
grinding, so arranged that the substance ground passes from
one to the other, then it is a more complicated machine — a
machine of higher order — than that with ten pairs of similar
grindstones. In other words, it is not mere multiplication of
organs which constitutes physiological differentiation ; but
the multiplication of organs for different functions in the first
place, and the degree in which they are coordinated, so as to
work to a common end, in the second place. Thus, a lobster
is a higher animal, from a physiological point of view, than a

THE TEGUMENTARY SYSTEM. 55

Cyclops^ not because it has more distinguishable organs, but
because these organs are so modified as to perform a much
greater variety of functions, while they are all coordinated
toward the maintenance of the animal, by its well -developed
nervous system and sense-organs. But it is impossible to say
that, e. g., the Arthropoda, as a whole, are physiologically
higher than the 3Iollusca, inasmuch as the simplest embodi-
ments of the common plan of the Arthropoda are less differ-
entiated physiologically than the great majoritj'^ of Mollusks.
I may now rapidly indicate the mode in which physiologi-
cal differentiation is effected in the different groups of organs
of the body among the Metazoa.

Integumentary Organs. — In the lowest JTe^a^o (7, the integ-
ument and the ectoderm are identical, but, so soon as a mes-
oderm is developed, the layer of the mesoderm which is in
contact with the octoderm becomes virtually part of the in-
tegument, and in all the higher animals is distinguished as
the dermis {enderon^^ while the ectodermal cells constitute
the epiderm^is {ecdero7i). The connective tissue and muscles
of the integument are exclusively developed in the enderon ;
while, from the epidermis, all cuticular and cellular exoskele-
tal parts, and all the integumentary glands, are developed.
The latter are always involutions of the epidermis. The hard
protective skeletons in all invertebrate Metazoa, except the
JPorifera, the Actinozoa, the Echinodermata, and the Tuni-
cata, are cuticular structures, which may be variously impreg-
nated with calcareous salts formed on the outer surface of the
epidermic cells.

In the Porifera, the calcareous or silicfous deposit takes
place within the ectoderm itself, and probably the same pro-
cess occurs, to a greater or less extent, in the Actinozoa. In
those Tunicata which possess a test, it appears to be a struct-
ure sui ge7ieris, consisting of a gelatinous basis excreted by
the ectoderm, in which cells detached from the ectoderm
divide, multiply, and give rise to a deposit of cellulose. The
test may take on the structure of cartilage or even of connec-
tive tissue. In the Vertebrata alone do we find hard exo-
skeletal parts formed by the cornification and cohesion of epi-
dermic cells.

In the Actinozoa and the Echinodermata, the hard skele-
ton is, in the main, though perhaps not whoUj^ the result of
calcification of elements of the mesoderm. In some Mollusks
portions of the mesoderm are converted into true cartilage,

56 THE ANATOMY OF INYERTEBRATED ANIMALS.

while the enderon of the integument often becomes the seat
of calcareous deposit. The endoskeleton and the dermal exo-
skeleton of the Vertehrata are cellular (cartilage, notochord)
or fibrous (connective tissue) modifications of the mesoderm,
which may become calcified (bone, dentine). Recent investi-
gations tend to show that the enamel of the teeth is derived
from the ectoderm.

The Alimentary Ai^j-ciratus. — From the simple sac of the
Hydra or aproctous Turbellarian, we pass to the tubular ali-
mentary tract of the proctuchous Turhellaria, In the Itoti-
fera and Polyzoa there is a marked distinction into buccal
cavity, pharynx, oesophagus, stomach, and intestines ; while
distinct salivary, hepatic, and renal glands, are found in the
majority of the higher invertebrates, and, not unfrequently,
glands secreting an odorous or colored fluid appear in the
region of the termination of the alimentary canal.

The oral and gastric regions are armed with cuticular
teeth in many Invertehrata / but teeth formed by the calcifi-
cation of papillary elevations of the enderon of the lining of
the mouth are confined to the Vertehrata y unless, as seems
probable, the teeth of the EcMnidea have a similar origin.

The lining membrane of the oral cavity is capable of being
everted, as a proboscis, in many Invertehrata. The margins
of the mouth may be raised into folds, armed with cuticular
plates. In the Vertehrata^ the jaws are such folds, supported
by endoskeletal cartilages, belonging to the system of the
visceral arches, or by bones developed in and around them ;
but, in the Arthropoda^ what are usually termed jaws are
modified limbs.

The Blood and Circidatory Apjoaratus. — In the Coelen-
terata, the somatic cavity, or enterocoele, is in free commu-
nication with the digestive cavity, and not unfrequently
communicates with the exterior by other apertures. The fluid
which it contains represents blood ; it is moved by the con-
tractions of the body, and generally by cilia developed on the
endodermal lining of the enterocoele. In the Turhellaria^
Trematoda, and (Jestoidea^ the lacunae of the mesoderm and
the interstitial fluid of its tissues are the only representatives
of a blood-vascular system. It is probable that these com-
municate directly with the terminal ramifications of the water-
vascular system. In the Rotifer a., a spacious perivisceral
cavity separates the mesoderm into two layers, the splanch-

THE BLOOD -SYSTEM. 57

nopleure^ which forms the enderon of the alimentary canal,
and the somatopleure^ which constitutes the enderon of the
integument. The terminations of the water- vessels open into
this cavity. In Annelids, there is a similar perivisceral cavity
communicating in the same way with the segmental organs ;
but, in most, there is, in addition, a system of canals with
contractile walls, which, in some, communicate freely with
the perivisceral cavity, but, in the majority, are shut off from
it. These canals are filled by a clear, usually non-corpuscu-
lated fluid, which may be red or green, and constitute the
pseud-hf^mal system. The fluid which occupies the perivis-
ceral cavity contains nucleated corpuscles, and has the
characters of ordinary blood. It seems probable that the
fluid of the pseud-h^mal vessels, as it contains a substance
resembling hsemoglobin, represents a sort of respiratory
blood.

In the Arthropoda^ no segmental organs or pseud-haemal
vessels are known. In the lowest forms, the perivisceral
cavity and the interstices of the tissues represent the whole
blood-svstem, and colorless blood- cells float in their fluid con-
tents. In the higher forms, a valvular heart, with arteries
and capillaries, appears, but the venous system remains more
or less lacunar. In the Mollusca, the same gradual differen-
tiation of the blood- vascular system is observable. In very-
many, if not all, the blood-cavities communicate directly with
the exterior by the " organs of Bojanus " — which resemble
very simple segmental organs, and appear to be always asso-
ciated with the renal apparatus.

In the Vertebrata^ Amphioxus has a system of blood-ves-
sels, with contractile walls, and no distinct heart. In all
the other Vertebrates there is a heart with at fewest three
chambers {sinus veuosuSy atrhwi, ventricle), arteries, capil-
laries, and veins, and a system of lymphatic vessels connected
with the veins. The lymphatic fluid consists of a colorless
plasma, with equally colorless nucleated corpuscles ; the blood-
plasma contains, in addition, red corpuscles, which are nucle-
ated in Ichthyopsida and /Scmropsida, but have no nucleus
in the 3Iamnialia. The lymphatic vessels always communi-
cate with the interstitial lacunae of the tissues, and in the
lower Vertebrates are themselves, to a great extent, irregular
sinuses. The venous system presents many large sinuses in
the lower Vertebrates; while, in the higher forms, these
sinuses are for the most part replaced by definite vessels with
muscular walls. But the *' serous cavities " remain as vast

58 THE ANATOMY OF IXVERTEB RATED ANIMALS.

lymphatic lacunas. Valves make their appearance in the lym-
phatics and in tlie veins, and the heart becomes subdivided in
such a manner as to bring about a more and more complete
separation of the systemic circulatory apparatus from that
which supplies the respiratory organs.

The Respiratory System. — In the lower Metazoa respira-
tion is effected by the general surface of the body. In the
Annelids, processes of the integument, which are sometimes
branched and usually are abundantly ciliated and supplied
with pseud-haemal vessels, give rise to hranchice. Branchiae
abundantly supplied with blood-vessels, but never ciliated,
attain a great development in the Crustacea. The access of
fresh water to them is secured by their attachment to some
of the limbs ; and, in the higher Crustaceans, one of the ap-
pendages, the second maxilla, serves as an accessory organ
of respiration. Although especially adapted for aquatic res-
piration, they are converted into air-breathing organs in the
land-crabs, being protected and kept moist in a large cham-
ber formed by the carapace.

In some moUusks (e. g., Pteropoda^^ the delicate lining
membrane of the pallial cavity serves as the respiratory
organ ; but, in most, branched or laminated processes of the
body give rise to distinct branchiae. The mantle becomes an
accessory organ of respiration, being so modified as to direct,
or to cause, the flow of currents of water over the branchi?e
contained in its cavity. In many adult urodele Amphibia
{Perennihranchiata)^ and in the embryonic condition of all
Amphibia and of many fishes, branchiae of a similar character,
abundantly supplied with blood-vessels, are attached to more
or fewer of the visceral arches.

In all these cases the branchiae are external, and are de-
veloped from the integument. In Crustaceans and Mollusks
the blood with which they are supplied is returning to the
heart ; while, in the Vertebrata mentioned, it is flowing from
the heart ; and it will be observed that the gradual per-
fectioning of the respiratory machinery consists, first, in the
outgrowth of parts of the integument specially adapted to
subserve the interchange between the gases contained in the
blood and those in the surrounding medium; secondly, in the
increase of the surface of the branchiae, so as to enable them
to do their work more rapidly; thirdly, in the development
of accessory organs, by which the flow of water over the
branchiae is rendered definite and constant, and may be in-

THE RESPIRATORY SYSTEM. 59

creased or diminished in accordance with the needs of the
economy.

It is probable that the water-vascular system and the seg-
mental organs of Turbellarians and Annelids, the cloacal
tubes of the Gephyrea and of some Holothxiridea^ the ambu-
lacral vesicles of the Echinoderms, and the large pharyngeal
cavity of the Polyzoa^ to a greater or less extent, subserve
respiration, and constitute internal respiratory organs.

In Myriapoda and Insecta, the tracliece — tubes which
open on the surface of the body and contain air, and are
curiously similar in their distribution to the water-vessels of
the worms — constitute a y^Tj complete internal aerial respira-
tory apparatus.

In Arachnida^ trachese may exist alone, or be accom-
panied by folded pulmonary sacs, or the latter may exist
alone, as in the Scorpion. In this case, these lungs are sup-
plied by blood which is returning from the heart.

In these animals, the flow of air into and out of the air-
cavities is governed by the contractions of muscles of the
body, disposed so as to alter its vertical and longitudinal
dimensions. In the higher forms, the entrance and exit of
air is regulated by valves, placed at the external openings
(stigmata) of the tracheae, and provided with muscles, by
which they can be shut.

In the Enteropneusta and the Tunicata a new form of
internal aquatic respiratory apparatus appears. The large
pharynx is perforated by lateral apertures, which place its
cavity in communication with the exterior ; and water, taken
in by the mouth, is driven through these branchial clefts and
aerates the blood which circulates in their interspaces.

The respiratory apparatus of Amphioxt/s, of all adult
fishes, and of the tadpoles of the higher anurous Amp)hihia,
in a certain stage of their existence, is of an essentially simi-
lar character. The accessory respiratory apparatus for the
maintenance and the regulation of the currents of water over
the gills is furnished by the visceral arches and their mus-
cles ; and the respiratory blood flows from the heart.

In MoUusks which live on land {Pidmog aster o]?oda), the
lining wall of the mantle cavity becomes folded and highly
vascular, and subserves the aeration of the venous blood,
which flows through it on its way to the heart. The lung is
here a modification of the integument, and might be termed
an external lung. The lungs of the air-breathing J^ertehrata,
on the contrary, are diverticula of the alimentary canal, pos-

60 THE ANATOMY OF LWERTEBRATED ANIMALS.

terior to the binderraost of the visceral arches. They receive
their blood from the hindermost aortic arch. It therefore
flows from the heart. The gradual improvement of these
lungs as respiratory machines is effected, first, by the increase
of the surface over which the venous blood brought to the
lungs is distributed; secondly, by changes in the walls of
the cavity in which the lungs are contained, by which that
cavity gradually becomes shut off from the peritoneal cham-
ber, and divided from it by a muscular partition. Concur-
rently with these modifications, a series of alterations takes
place in the accessory apparatus of respiration, whereby the
machinery of inspiration, which, in the lower Vertebrata, is a
buccal force-pump, which drives air into the lungs, in the same
way as water is driven through the branchiae, is replaced by
a thoracic suction-pump, which draws air into the lungs by
dilatation of the walls of the closed cavity in which they are
contained. Alongr with these chansres, modifications of the
heart take place, in virtue of which one-half of its total
mechanical power becomes more and more exclusively ap-
propriated to the task of driving the blood through the lungs.
The term " double circulation " applied to the course of the
blood in the highest Vertebrata is, however, a misnomer. In
the highest, as in the lowest, of these animals, the blood com-
pletes but one circle, and the respiratory organ is in the
course of the outward current.

Many animals are truly amphibious, combining aquatic
and aerial respiratory organs.

Thus, among Mollusks, Ampullarla and Onchidum com-
bine branchias with pulmonary organs ; many Teleostean fishes
have the lining membrane of the enlarged branchial chamber
vascular and competent to subserve aerial respiration. And
in the Ganoids and Teleostei the presence of an air-bladder,
which is both functionally and morphologically of the same
nature as a lung, is yqtw common. But, in the majority of
the Teleostei, the air-bladder is turned aside from its pulmo-
nary function to subserve mechanical purposes, in affecting
the specific gravity of the body. On the other hand, in the
Ganoids and Dipnoi, the whole series of modifications by
which the air-bladder passes into the lung are patent. In
such lower Amphibia as Proteus and Menobranchus, bran-
chial respiration is predominant, and the lungs are subsidi-
ary ; but, in the higher, the lungs acquire greater importance,
while the branchije diminish, and eventually disappear.

THE UROPOIETIC SYSTEM. 61

The TIropoietic System. — Uropoietic organs, distinct from
the alimentary canal, are probably represented by the water-
vascular system and segmental organs of the worms. The
*' organs of Bojanus " of MoUusks are sacs or tubes opening,
on the one side, on the exterior of the body, and, on the
other, into some part of the blood-vascular system. So far,
as Gegenbaur has shown, they resemble the segmental organs
of Annelids. In the majority of the 3Iollusca, some part of
the wall of the organ of Bojanus is in close relation with the
venous system near the heart, and the nitrogenous waste of
the body is here eliminated from the venous blood. In the
Vertebrata, the renal apparatus is constructed en the same
principle. If for simplicity's sake we reduce a mammalian
kidnev to a ureter with a sino-le uriniferous tubule, it cor-
responds with an organ of Bojanus, so far as it contains a
cavity' communicating with the exterior at one end, and hav-
ing a vascular plexus — the Malpighian body — in intimate
contact with the opposite end. In the adult mammal there is
no direct communication between the urinary duct and the
blood-vascular system. But, inasmuch as recent researches
have proved that the ureter is formed by subdivision of the
Wolffian duct, and that the Wolffian duct is primitively a di-
verticulum of the peritoneal cavity, and remains for a longer
or shorter time (permanently, in some of the lower Verte-
bratn, as 3Iyxine) in communication therewith ; and since it
has further been shown that the peritoneal cavity communi-
cates directly with the lymphatics, and therefore indirectly
with the veins ; it follows that the vertebrate kidney is an
extreme modification of an organ, the primitive type of which
is to be found in the organ of Bojanus of the Mollusk, and in
the segmental organ of the Annelid ; and, to go still lower,
in tlie water-vascular system of the Turbellarian. And this, in
its lowest form, is so similar to the more complex conditions
of the contractile vacuole of a Protozoon, that it is hardly
straining analogy too far to regard the latter as the primary
form of uropoietic as well as of internal respiratory apparatus.

The j^ervous System. — In its essential nature, a nerve is
a definite tract of living substance, through w^hich the molec-
ular changes which occur in any one part of the organism
are conveyed to and affect some other part. Thus, if, in the
simple protoplasmic body of a Protozoon, a stimulus applied
to one part of the body were more readily transmitted to
some other part, along a particular tract of the protoplasm,

62 THE ANATOMY OF IxWERTEBRATED ANIMALS.

that tract would be a virtual nerve, altliougli it might have
no optical or chemical characters which should enable us to
distinguish it from the rest of the protoplasm.

It is important to have this definition of nerve clearly
before us in considering the question whether the lowest
animals possess nerves or not. Assuredly nothing of the
kind is discernible, b}' such means of investigation as we at
present possess, in Protozoa or Porifera y but an}' one who
has attentively watched the ways of a Colpodo, or still more
of a Vorticella^ will probably hesitate to deny that they
possess some apparatus by which external agencies give
rise to localized and coordinated movements. And when we
reflect that the essential elements of the highest nervous
system — the fibrils into which the axis-fibres break up — are
filaments of the extremest tenuity, devoid of any definite
structural or other characters, and that the nervous system
of animals only becomes conspicuous by the gathering to-
gether of these filaments into nerve-fibres and nerves, it will
be obvious that there are as strong morphological, as there
are physiological, grounds for suspecting that a nervous sys-
tem may exist very low down in the animal scale, and possi-
bly even in plants.

The researches of Kleinenberg, which may be readily veri-
fied, have shown that, in the common Hydra, the inner ends
of the cells of the ectoderm are prolonged into delicate pro-
cesses, which are eventually continued into very fine longi-
tudinal filaments, forming a layer between the ectoderm and
the endoderm.

Kleinenberg terms these neuro-muscular elements, and
thinks that they represent both nerve and muscle in their
undifferentiated state. But it appears to me that while the
assumed contractility of these fibres might account for the
shortening of the body of the Polyp, they can have nothing
to do with its lengthening. As the latter movements are at
least as vigorous as the former, we are therefore obliged to
assume sufficient contractility in the general constituents of
the body to account for them. And if so, what ground is
there for supposing that this contractility can be exerted by
only one tissue when the body shortens? To my mind, it is
more probable that " Kleinenherg's fibres " are solely inter-
nuncial in function, and therefore the primary form of nerve.
The prolongations of the ectodermal cells have indeed a
strangely close resemblance to those of the cells of the olfac-
tory and other sense-organs in the Vertebrata / and it seems

THE NERVOUS SYSTEM. 63

probable that they are the channels by which impulses affect-
ing any of the cells of the ectoderm are conveyed to other
cells and excite their contraction.

The researches of Eimer * upon the nervous system of the
Ctenophora are in perfect accordance with this view. The
mesoderm is traversed in all directions by very fine fibrils,
varying in diameter from ^qIqq to tj^qT ^^ ^^ inch. These
fibrils present numerous minute varicosities, and, at intervals,
larger swellings which contain nuclei, each with a large and
strongly refracting nucleolus. These fibrils take a straight
course, branch dichotomously, and end in still finer filaments,
which also divide, but become no smaller. They terminate
partly in ganglionic cells, partly in muscular fibres, partly in
the cells of the ectoderm and endoderm. Manv of the nerve-
fibrils take a longitudinal course beneath the centre of each
series of paddles, and these are accompanied by ganglionic
cells, which become particularly abundant toward the aboral
end of each series. The eight bands meet in a central tract
at the aboral pole of the body; but Eimer doubts the nervous
nature of the cellular mass which lies beneath the lithocyst
and supports the eye-spots.

The nervous system of the Ctenophoran is, therefore, just
such as would arise in Sjdra^ if the development of a thick
mesoderm gave rise to the separation and elongation of
Kleinenberg's fibres, and if special bands of such fibres,
developed in relation with the chief organs of locomotion,
united in a central tract directly connected with the higher
sensory organs. We have here, in short, virtual, though in-
completely differentiated, brain and nerves.

AH recent investigation tends more and more completely
to establish the following conclusions : firstly, that the central
ganglia of the nervous system in all animals are derived from
the ectoderm; secondly, that all the nerves of the sensory
organs terminate in cells of the ectoderm ; thirdly, that all
motor nerves end in the substance of the muscular fibres to
which they are distributed. So that, in the highest animals,
the nervous system is essentially similar to that of the lowest;
the difference consisting, in part, in the proportional size of
the nerve-centres, and, in part, in the gathering together of
the internuncial filaments into bundles, having a definite
arrangement, which are the nerves, in the ordinary anatomical
sense of the term.

» " Zoologiscln Studien auf Capri," Leipsic, 1873.

64 THE ANATOMY OF INVERTEBRATED ANIMALS.

And as respects the ectodermal cells which constitute the
fundamental part of the organs of the special senses, it is
becoming clear that the more perfect the sensory apparatus,
the more completely do these sensigenous cells take on the
form of delicate rods or filaments. Whether we consider the
organs of the lateral line in fishes and amphibia, the gusta-
tory bulbs, the olfactory cells, the auditory cells, or the
elements of the retina, this rule holds good.

Every one of the organs of the higher senses makes its
appearance in the animal series as a part of the ectoderm,
the cells of which have undergone a slight modification. In
the case of the eye, accessory structures, consisting of vari-
ously-colored masses of pigment, which surround the visual
cells, and of a transparent refracting cuticular or cellular
structure which lies superficially to them — a rudimentary
choroid and cornea — are next added. The highest form of
compound Arthropod eye differs from this only in the differ-
entiation of the layer of sensigenous cells into the crystalline
cones and their appendages, and it has not been clearly made
out that the simple eyes of most other Invertebrata have
undergone any further change.

But in JSfautiliis the nerve-cells and choroid line the walls
of a deep cup open externally ; which, though its development
has not been traced, may be safely assumed to result from
the involution of the retinal ectoderm. It may be compared
to an arthropod compound eye become concave instead of
convex.

In the higher Cephalopoda, the margins of the ocular
pouch nnite and give rise to a true cornea, which, however,
frequently remains perforated, and a crystalline lens is de-
veloped. In the higher Vertebrata the retina is still a modi-
fied portion of the ectoderm. For, inasmuch as the anterior
cerebral vesicle is formed by involution of the epiblast, and
the optic vesicle is a diverticulum of the anterior cerebral
vesicle, it necessarily follows that the outer wall of the optic
vesicle is really part of the ectoderm, its inner face being,
morphologically, a portion of the surface of the body. The
rods and cones of the vertebrate eye, therefore, exactly corre-
spond with the crj'stalline cones, etc., of the Arthropod eye;
and the reversal of the ends which are turned toward the
light in the Vertebrata is a necessary result of the extraor-
dinary change of position which the retinal surface undergoes
in them.

In the part of the ectoderm which takes on the auditory

REPRODUCTIVE ORGANS. 65

function, two kinds of accessory organs, solid particles sus-
pended in a fluid and fine hair-like filaments, are developed
in close relation with the nerve-endings. In the Crustacea
both are combined, and an involution of the sensory region
takes place, which usually remains open throughout life, and
represents the most rudimentary form of auditory labyrinth.
■^ The Crustacean ear is the parallel of the JVaiitihcs eye. In
the Vertehrata the membranous labyrinth is similarly an in-
volution of the integument, which remains open throughout
life in many fishes, but becomes shut off and surrounded by
thick mesoblastic structures in all the higher Vertebrata.
The tympanum and the ossicula auditus are additional
accessory structures, formed at the expense of the hyoman-
dibular cleft and its boundary-walls.

The Reproductive System. — The relation of the reproduc-
tive elements to the primitive layers of the germ is as yet
uncertain. E. van Beneden has brought forward very strong
evidence to the effect that in Hydractinia the spermatozoa
are modified cells of the ectoderm, and the ova of those of the
endoderm ; but, whether it can be safely concluded that this
rule holds good for animals generally, is a question that can
only be settled by much and difficult investigation. The fact
that, in the Vertehrata, the ova and spermatozoa are products
of the epithelial lining of the peritoneal cavity, and therefore
proceed from the mesoblast, appears at first sight directly to
negative any such generalization. But it must be remem-
bered that the origin of the mesoblast itself is yet uncertain,
and that it is quite possible that one portion of that layer may
originate in the ectoderm and another in the endoderm.

There is some reason to suspect that hermaphrodism was
the primitive condition of the sexual apparatus, and that uni-
sexuality is the result of the abortion of the organs of the other
sex, in males and females respectively.

Very low down in the animal series, among the Turhella-
ria^ the accessory organs of generation acquire a great com-
plexity. In the lower Turhellaria the excretory duct is a
mere short, wide passage. But, in the higher Turhellaria and
Trematoda, the female apparatus presents a germarium, in
which the ova are developed ; vitellarian glands, which give
rise to a supplemental or food yelk ; an oviduct ; a uterus and
vagina; and a spermatheca, in which the semen is stored up.
The male apparatus presents a testis, a vas deferens, and a
penis. The function of the vitellarian gland may be taken on

66 THE ANATOMY OF INVERTEBRATED ANIMALS.

by cells of the ovary, or oviduct ; or accessory yelk-substance
may be formed within the primitive ovum itself, in the Arthro-
poda and in most Mollusca ; but the reproductive organs in
all these animals are reducible to the Turbellarian type.

In the Annelids ( OUgochceta and Polychcjeta), the ovaria
and testes often have no special ducts, and their products
make their way out of the body by canals which appear to be
modified segmental organs.

In the Cephalopoda^ again, the ovaria and testes part with
their contents by dehiscence into chambers connected with the
water-cavities, which are prolongations of the organs of Boja-
nus. And they are conveyed away from these chambers by
ducts, the oviducts or vasa deferentia, which commence by
open mouths in them.

In the Vertehrata^ the reproductive organs either dehisce
and pour their contents into the peritoneal cavity, whence
they are conveyed outward by abdominal pores {Marsipo-
branchii^ many Teleostei), or they are continued into ducts
which open behind the anus separately from the renal open-
ing in the females, but in common with it in the males (most
Teleosteans) ; or their ducts are derived from portions of the
primitive renal apparatus which, as we have seen, is a struct-
ure of the same order as the organs of Bojanus and the seg-
mental organs. The testis is usually converted into a mass
of tubuli, which eventually open directly into the ducts (e^n-
didymis, vas deferens) derived from the renal organs. The
ovary, on the other hand, becomes an aggregation of sacs —
the Graafian follicles — and the oviducts open into the perito-
neal cavity.

Develojyment. — The embryo either passes through all
stages from the morula to a condition differing from the adult
only in size, proportions, and sexual characters, or it leaves the
e^g^ in a condition more or less remote from the adult state,
and sometimes exceedingly different from it. In the latter
case, the animal is said to undergo a metamorphosis. Each of
these modes of development occurs in members of the same
group, and often in closely allied forms : as, for example, the
former in the crayfish (Astacus), and the latter in the lobster
{Ilomarus).

When metamorphosis occurs, the larva may live under
conditions totally different from those under which the adult
passes its existence, and its structure may be variously modi-
fied in relation to these conditions. Thus the larva of an

DEVELOPMENT. 67

animal wbich is fixed in the adult state may be provided with
largely-developed locomotive organs ; while that of an adult
which" feeds by suction may be provided with powerful appa-
ratus for the seizure and manducation of vegetable and ani-
mal prey.

The larva of a free adult may be parasitic, or that of a
parasitic adult free and actively locomotive. Moreover, the
whole course of development maj' take place outside the body
of the parent, or more or less extensively within it ; whence
the distinction of ovijyarous, ovoviviparous^ and vivix>arous^
animals.

Finally, when development takes place within the body of
the parent, the foetus may receive nourishment from the latter
by means of an apparatus termed a placenta, by which an
exchange between the parental and foetal blood is readily
effected. Examples of placentas are found not only in the
higher mammals, but in some Plagiostome fishes and among
the Tunicata,

In many insects and in the higher Vertebrates, the em-
bryo acquires a special protective envelope, the amnion^
which is thrown off at birth ; while, in many Vertebrates,
another foetal appendage, the allantois, subserves the respi-
ration and nutrition of the foetus.

The strange phenomena included under the head of the
"Alternation of Generations," and which result from the di-
vision, by budding or otherwise, of the embryo which leaves
the egg, into a succession of independent zooids, only the last
of which acquires sexual organs, have already been gener-
ally discussed.

IV. — THE DISTEIBUTIOK^ OF ANIMALS.

The distribution of animals has to be considered under
two points of view : first, in respect of the present condi-
tion of Nature ; and secondly, in respect of past conditions.
The first is commonly termed Geo<^raphical, the second
Geological, or JPaleo7itological, Distribution. A little con-

* As eggs capable of development are alive, this terminology is etymologi-
cally bad ; and ovovivi parous is particularly objectionable, as all animals bring
forth live eggs, or that which proceeds from them. But, as understood to ap-
ply to animals which lay esrgs, to those in which the eggs are hatched within
the interior of the body without any special fcetal nutritive apparatus, and to
those in which the young are provided with such an apparatus, it has a certdn
convenience.

68 THE ANATOMY OF INVERTEBKATED ANIMALS.

sideration, however, will show that this classification of the
facts of distribution is essentially faulty, inasmuch as many
of the phenomena included under the second head are of the
same order as those comprehended under the first. Zoological
Distiibution comprehends all the facts which relate to the
occurrence of animals upon the earth's surface throughout
the time during which animal life has existed on the globe.
Therefore it embraces :

First, Zoological Chronology^ or the duration and order of
succession of living forms in time ; and —

Secondly, Zoological Geography^ or the distribution of life
on the earth's surface at any given epoch.

What is commonly termed Geographical Distribution is
simply that distribution which obtains at the present epoch ;
but it is obvious that, at any given moment in their past his-
tory, animals must have had some sort of geographical distri-
bution ; and considerable acquaintance with the nature of that
distribution has now been obtained for all the epochs, the
nature of the living population of which has been revealed by
fossil remains. I do not propose to deal at length with either
branch of distribution in this place, but a few broad truths
which have been established may be mentioned.

Geographical Distribution at the Present Epoch. — The
fauna of the deep sea (below five hundred fathoms) has been
shown, by the investigations of Wyville Thomson and his
associates of the Challenger, to present a striking general uni-
formity (in all parts of the world hitherto explored, in corre-
spondence with the general uniformity) of conditions at such
depths.

With respect to the surface of the sea, the observations of
the same naturalists tend to establish a like uniformity of the
great types of foraminiferal life throughout the tropical and
temperate zones — with a diminution in the abundance of that
life toward the arctic and antarctic regions, where it appears
to be replaced by Hadiolaria and Diatomaceous plants.

With regard to higher organisms, the oceanic Ilydrozoa
and the Ctenophora are undoubtedly very widely spread. It
is probable that they attain their maximum development in
warm seas, though the know^n facts are insufficient for the
definite conclusion. Sagitta and Appenclicularia, with many
genera of Copepoda^ Crustacea, and Pteropoda, are of world-
wide distribution ; and it is at present doubtful whether any
well-marked provinces of the ocean can be defined by the oc-

MARINE DISTRIBUTION. 69

currence of purely pelagic animals. On the other hand, shal-
low-water marine animals fall into assemblages characteristic
of definite areas or provinces of distribution — that is to say,
though many species have a world-wide distribution, others
occur only in particular localities, and certain geog'raphical
areas are marked by the existence in them of a number of
such peculiar species. The basins of the Pacific, the Indian
Ocean, the Atlantic, the Mediterranean, and the Arctic seas,
are thus especially characterized ; and even limited areas of
these great geographical divisions, such as the Celtic, (he
Lusitanian, and the Australian, have their peculiar features.

But, though the shallow-water marine faunae thus follow
the broad features of physical geography, and though, within
each great province of distribution thus marked out, temper-
ature and other physical conditions have an obvious influence
in determining the range of species ; yet, on comparing any
two great areas together, difierences in climatal conditions
are at once seen to be inadequate to account for the diifer-
ences between the faunae of the two areas. Climate in no
way enables us to understand why the Trigonia^ the pearly
N^autilus^ the Cestracion, the eared seals, and the penguins,
are found in the Pacific and not in the Atlantic area ;^ nor
why the Cetacea of the arctic and antarctic regions should be
as difl'erent as they are. When we turn to the distribution
of land-animals, the boundaries of the provinces of distribu-
tion correspond neither with physical features nor with cli-
matic conditions. Mammals, birds, reptiles, and amphibians,
are so distributed at the present day as to mark out four great
areas or provinces of distribution of very unequal extent, in
each of which a number of characteristic types, not found
elsewhere, occur. These are : 1. The Arctogceal, including
North America, Europe, Africa, and Asia as far as Wallace's
line, or the boundary between the Indian and the Papuan
divisions of the Indian Archipelago ; 2. The Austrocohim-
bian, comprising all the American Continent south of Mexico;
3. The Australian^ from Wallace's line to Tasmania ; 4. The
Kovozelanian^ including the islands of New Zealand.'^

* Penguins are found at the Cape of Good Hope and at the Falkland Islands,
"but not in the northern parts of tlae west coast of Africa, nor of the east coast
of South America. In the Pacific they stretch north to the Papuan and Peru-
vian coasts.

2 On the classification and distribution of the AlectoromorpJia, and Hetero-
morplM : Proceedings of the Zoological Society, 1868. Sclater on the " Gop-
graphical Distribution of Birds," Ibid., vol. ii. Pucheran, " Eevue et Magasin
de Zoologie," 1865. Murray, " The Geographical Distribution of Mammals,"

70 THE ANATOMY OF INVERTEBRATED ANIMALS.

There is no doubt that provinces of distribution, closely
corresponding with these, existed at the time of the Qua-
ternary and later Tertiary rocks. In Europe, North America,
and Asia, the Arctogseal province was as distinctly charac-
terized in the Miocene, and probably in the Eocene epoch, as
it is at present. What may have been the case in Austroco-
lumbia, Australasia, and JNovozelania, we have no means of
being certain, in the absence of sufficient knowledge of the
Miocene and Eocene deposits of those regions.

Our present knowledge of the geographical distribution
which obtained in the older periods does not enable us to
speak with any confidence as to the limits of the provinces of
distribution in the past. But this much is certain, that as far
back as the epoch of the Trias — at the dawn of the Secondary
period — the Meptilia and Amphibia of Europe, India, and
South Africa, and probably North America, presented the
same kind of resemblance as the mammals and birds of the
corresponding Arctogasal fauna do now. But then there is
no information respecting the reptiles and amphibians of the
corresponding epoch in Austrocolumbia and Australia, so that
it is impossible to say whether, in Triassic times, the Arcto-
gaeal province was limited as it is now.

Outside the limits of the Arctogasal province, the mate-
rials for forming a judgment of the distribution of animals
are altogether insufficient to enable us to draw any conclu-
sion as to the existence, and still less as to the boundaries, of
definite provinces of distribution in Palaeozoic times. No
remains of land-animals have yet been discovered. The
fresh-water fauna consists of Amphibians and Fishes, and we
know nothing, or next to nothing, of these in any^art of the
world except the Arctogaeal province.

A good deal is known of the older Silurian fauna outside
the boundaries of the present Arctogaeal province, and within
those of both the Austrocolumbian and Australasian prov-
inces. With a generally similar fades, the faunae of these
regions present clear differences. And, considering that the
groups of animals which are represented are chiefly deep-sea
and pelagic forms, it is not wonderful that this similarity of
facies should exist. The investigations of the Challenger
expedition show that such forms present a like similarity of
facies at the present day.

One of the most important facts which have been estab-
lished under the head of Zoological Chronology is, that in all
parts of the world the fauna of the later part of the Tertiary

THE OLDEST KNOWN FAUNA. 71

period, in any province of distribution, was made up of forms
either identical with, or very similar to, those now living in
that area.

For example, the elephants, tigers, bears, bisons, and hip-
popotamuses of the later tertiary deposits of England are all
closely allied to members of the existing Arctogaeal fauna ;
the great armadillos, anteaters, and platyrrhine apes of the
caves of South America, are as closely related to the existing
Austrocolumbian fauna ; and the fossil kangaroos, wombats
and phalangers of the Australian tertiaries to those which
now live in the Australasian province.

No remains of elephants occur in Australia, nor kangaroos
in Austrocolumbia ; nor anteaters and armadillos in Europe
in Tertiary deposits.

But, as we go back in time from the Tertiary to the Sec-
ondary, this law no longer holds good. Most of the few ter-
restrial mammals of secondary age which have been dis-
covered belong to Australasian and not to Arctogfeal types,
and the marine fauna resembles that of the existing Pacific
more than it does that of the Atlantic area, but differs from
both in the presence of numerous wholly extinct groups. It
looks as if, in the latter part of the Cretaceous epoch, a
great change in the limits of the then existing distributional
area had taken place, and the types now characteristic of
the Arctogaeal province had invaded regions from which
they had before been shut out. And the assumption of a
process of a similar character appears to me to be the only
rational explanation of the rapid advent of types absent in
the Palaeozoic deposits known to us, in the earlier Secondary
rocks.

Yet other results of first-rate importance have come out
of the study of the chronological relations of fossil remains.
Cuvier's investigations proved that the hiatuses between
existing groups of ungulate mammals tend to be filled up by
extinct forms. Later investigations have not only confirmed
this conclusion, but have shown that, in several cases, an
existing much-modified form can be shown to have been pre-
ceded in time, in the same distributional area, by exactly
such forms as it is necessar}^ should have existed, if the much-
modified existing animal had proceeded by way of evolution
from a simpler form.

For certain groups of animals, then, there is as much and
as good evidence of their having been evolved by successive
modification of a primitive form as the nature of the case per-

72 THE AXATOMY OF INYERTEBRATED ANIMALS.

mits us to expect. But the groups in which there is evi-
dence of such modifications during geologically recorded
time, all belong to the most differentiated members of their
classes. Lower forms, coextensive in duration, exhibit no
sign of having undergone any notable modification. While
the former are mutable^ the latter diie persistent types in rela-
tion to oreolocjical time.

Leaving the debatable question of the nature of Eozoon
aside, the oldest fossiliferous rocks are the Cambrian. The
scanty fauna therein preserved consists of forms which are
neither Protozoa nor Porifera^ nor even appertain to the
lowest groups of their respective classes. There is no reason
to believe that it gives a just notion of the contemporaneous
fa ana, nor is there any valid reason for the supposition that
it represents the forms of animal life which were the first to
make their appearance on our planet.

CHAPTER IL

THE PROTOZOA.

In its feeblest manifestations, the contractility of animals
results in mere changes of the form of the body, as in the
adult Gregarinoe ; but, from the sluggish shortenings and
lengthenings of the different diameters of the body which
these creatures exhibit, all gradations are traceable, through
those animals which push out and retract broad lobular pro-
cesses, to those in which the contractile prolongations take
the form of Ions: and slender filaments. Whether thick or
filamentous, such contractile processes are called "pseudo-
podia," when their movements are slow, irregular, and in-
definite ; " cilia " or " flagella," when they are rapid and occur
rhythmically in a definite direction ; but the two kinds of or-
gans are essentially of the same nature. It will be convenient
to distinguish those Protozoa which possess pseudopodia, as
myxopods^ and those which are provided with cilia or flagella,
as mastigopocls.

The Protozoa are divisible into a lower and a higher
group. In the former — the Monera — no definite structure is
discernible in the protoplasm of the body ; in the latter — the
ExDOPLASTicA — a Certain portion of this substance (the so-
called nucleus) is distinguishable from the rest;^ and, very
common!}', one or more " contractile vacuoles " are present.
The name of contractile vacuoles is given to spaces in the pro-
toplasm, which slowly become filled with a clear, watery fluid,
and, when they have attained a certain size, are suddenly
obliterated by the coming to2:ether, on all sides, of the proto-
plasm in which they lie. This systolic and diastolic move-
ment usually occurs at a fixed point in the protoplasm, at regu-
lar intervals, or rhythmically. But the vacuole has no proper

* I adopt this distinction as a matter of temporary convenience, though
I entertain great doubt whether it will stand the test of further investigation.

74 THE ANATOMY OF INYERTEBRATED ANIMALS.

wall, nor, in most cases, is any trace of it discernible at the
end of the systole. Occasionally, the vacuole certainly com-
municates with the exterior, and there is some reason to
think that such a communication mav always exist. The
function of these organs is entirely unknown, though it is an
obvious conjecture that it may be respiratory or excretory.

The "nucleus" is a structure which is often wonderful]}''
similar to the nucleus of an histological cell ; but, as its iden-
tity with this is not fully made out, it may better be termed
" endoplast." It is, usually, a rounded or oval body imbed-
ded in the protoplasm, and but slightly different therefrom
in either its optical or chemical characters. Generally it be-
comes more deeply stained by such coloring-matters as haema-
toxylin or carmine, and resists the action of acetic acid better
than the surrounding protoplasm.

In a few Protozoa there are many endoplasts in the sub-
stance of the body, and the protoplasm shows some tendency
to become partially differentiated into cells. But where, as
in the higher Infusoria^ the body presents a definite organi-
zation, with permanently differentiated constituents, which
may be properly termed tissues, these tissues do not result
from the metamorphosis of cells, but originate from the pro-
toplasm directly by changes of its physical and chemical char-
acters.

Conjugation, followed by the development of germs, which
are set free and assume the form of the parent, has been ob-
served in several groups of tlie Protozoa, but it is not yet
quite certain how far sexual distinctions are established among
these animals.

I. — THE MOXERA.

In these lowest forms of animals the entire living body
consists of a particle of gelatinous protoplasm, in which
no nucleus, contractile vacuole, or other definite structure,
is visible ; and which, at most, presents a separation into
an outer, more clear, and denser layer, the ectosarc ; and
an inner, more granular and fluid matter, the endosarc. The
outer layer is the seat of active changes of form, whereby
it is produced into pseudopodia, which attain a certain
length, and are then retracted, or are effaced by the devel-
opment of others from adjacent parts of the body. These
pseudopodia are sometimes broad, short lobes ; at others, elon-
gated filaments. When lobate, the pseudopodia remain dis-

THE MONERA. 75

tinct from one another, their margins are clear and transpar-
ent, and the granules which they may contain plainly flow
into their interior from the more fluid central part of the
body. But, when they are filiform, they are very apt to run
into one another, and give rise to networks, the constituent
filaments of which, however, readily separate and regain their
previous form ; and, whether they do this or not, the surfaces
of these pseudopodia are often beset by minute granules,
which are in incessant motion — like those which are observ-
able on the reticulations of the protoplasm of the cells in a
Trachscantia hair.

The rayxopod thus described moves about by means of its
contractile pseudopodia, and takes the solid matters which
serve as its food into all parts of its body by their aid ; w^hile
the undigested exuvia of the food are rejected from all parts
of the body in the same indiscriminate way. It is an organ-
ism which is devoid of any visible organs except pseudopodia ;
and, so far as is known at present, it multiplies by simple di-
vision.

The Protamoeha (with lobate pseudopodia) and Protoge-
nes (with filamentous pseudopodia), of Haeckel, are Monera
of this extremaly simple character. In Myxodictyiun (Haeck-
el) the pseudopodia of a number of such Monera run togeth-
er, and give rise to a complex network, or common plasmo-
dlum.

It is open to doubt, however, whether either Protamoeha,
Protogenes, or Myxodlctyum, is anything but one stage of a
cycle of forms, which are more completely, though perhaps
not yet wholly, represented by some other very interesting
Monera^ also described by Haeckel.

Thus, the genus 'Vampyrella is a myxcpod with filanien-
tous pseudopodia, a species of which infests one of the stalked
Diatomace:^, Gomplionema, feeding upon the soft parts of the
frustules of its host, by inserting some of its pseudopodia
through the raphe of the frustule, which it envelops, and
absorbing the contained protoplasm. Having thus provided
itself with abundant nourishment, by creeping from frustule
to frustule of the Gomphonema, it thrusts aside the last
evacuated frustule from its peduncle, and, taking its place,
draws in its pseudopodia, becomes spherical, and surrounds
itself with a structureless cvst, inclosed in which it remains
perched upon the peduncle of the Gomplionema. Soon its
protoj^lasm undergoes division into four equal masses, and
each of these, becoming converted into a young Vampyrella^

76

THE ANATOMY OF INVERTEBRATED ANIMALS.

escapes from the cyst, and recommences the predatory life of
its parent. In this case the myxopod becomes encysted, and

Fig, 1. — Trotomyx.a aurardiaca (Haeckel").— c, the Ptill condition snrronnded hy a
Btructurcless cyst ; h. encysted form, the protoplasm of which is dividinc:; c, the
cy?t burstini^and pivin;jf exit to the bodies into which the protoplasm breaks np.
These are at first "monads,'' d, earh beinp: provided with a flajrelliform cilinm,
by means of which it propels itself (cT). After a time each monad retracts its
cilinm. and resumes an Amoeba like form (e> ; many of these coalesce and form
a single plasmodinm. which crows and feeds nnder the form /. The specimen
fieared contains a Peridinhim. (above"), three J)ictyocyi(t(v. (below), and two I)<th'
mi'v (Diatomaceoup plants). In the centre. (Hacckel^ " Stadicn uber Mouereu,"
1870.)

THE MOXERA. 77

then undergoes fission into bodies, each of which passes direct-
ly into the form of the parent.

In another genus [3fb'xastrum) an additional complication
is introduced ; the myxopod becomes encysted, and then di-
vides into many portions ; each of these elongates, and sur-
rounds itself with a delicate, fusiform, silicious case. Thus
inclosed, the germs are set free by the bursting of the cyst ;
and, after a while, the contents of the silicious cases emerge,
and pass at once into the myxopod state.

In other genera, not only does the myxopod become en-
cysted before it undergoes fissive multiplication, but the forms
tlius produced differ from the myxopod in being free-swim-
ming organisms, propelled by a long vibratile filament or fla-
gellum, like those flagellate Infusoria which are termed "mo-
nads." After swimming about for a while, these mastigopods
draw in their flagella, and become creeping myxopod s. This
cycle of forms is exhibited by the genus Protomonas of
Haeckel. Lastl}^, in Protorayxa (Fig. 1) (Haeckel), there is
an alternation of a mastigopod (c/) with a myxopod form (e), as
in Protomonas. But each myxopod does not usually become
encysted alone. On the contrary, a certain number of the
myxopods unite together, and become fused into an active
Plasmodium ( /"), which exhibits no trace of their primitive
separation. The plasmodium becoming quiescent and sphe-
roidal, surrounds itself with a structureless cyst (a), divides
into numerous portions (^), which are converted into flagellate
mastigopods, and these finally return to the myxopod condi-
tion (c, c7, e). The cycle of life is here singularly similar to
that presented by the Myxomycetes, which have hitherto been
usually regarded as plants.

There is no means of knowing whether the cycle of forms
presented by Protomonas and Protomyxa is complete, or
whether some term of the series is still wanting ; and, con-
sidering how low down among plants the sexual process oc-
curs, it seems quite possible that some corresponding sexual
process yet waits to be discovered among the Monera. It is
posible that the fusion of separate Myxodictya and Proto-
myx(B into a plasmodium may be a process of sexual conjuga-
tion. On the other hand, it may well be that these extremely
simple organisms have not yet reached the stage of sexual
differentiation.

The Foramixifera. — Doubtless many 3Ionera remain to
be discovered, but they will probably be minute and inconspic-

78

THE AXATOMY OF INYERTEBRATED ANIMALS.

uous organisms like the rnajority of those already described.
The I orainimfera^ on the other hand, are Monera of the
Protogenes type, which, nevertheless, play and have played an
important part in the history of the globe, by reason of their
power of fabricating skeletons or shells, which may be com-
posed of horny (chitinous?) matter, or of carbonate of lime,
secreted from the water in which they live, or may be fabri-
Lated by sticking together extraneous matter, such as par-
ticles of sand.

The first step from such an organism as Protogenes to the
Forciraimfera is seen in Lieherkuhnici of Claparede, where
the pseudopodia are given off from only a small part of the
surface of the bodv, the rest remainino- naked and flexible.

In Gromia there is a similar restriction of the area from

Fig. 2.— a Eotalia, with extended pseudopodia : with an enbrged sectional view of the

chambered skeleton (after Schulze).

which pseudopodia proceed, but the rest of the body is in-
vested by a case of a membranous substance. Let this case
become hardened bv the attachment of foreiern bodies — as
particles of sand, or fragments of shelly matter, as in the so-
called arenaceous Foraminifera — or let a deposit of calca-
reous salts take place in it, and the Gromia would be con-
verted into a Foraminifer.

The infinitely diversified characters of the skeleton of the
Foraminifera depend — firsth^, upon the structure of the skele-
tal substance itself ; and, secondly, upon the form of the pro-
toplasmic body, which last, again, is largely dependent upon
the manner in which successive buds of protoplasm are devel-
oped from the parent mass, which, to begin with, is always
simple in form and commonly globular.

The substance of the calcareous skeleton itself, whatever

THE FORAMINIFERA.

79

be its form, is either perforated or imperforate. In the Im-
perforata ( Gromidce, Lituitidm^ MilioUdcB) the pseudopodia
are protruded from only one end of the body, the rest of
which is cut off from the exterior by the skeleton. In the
Perforata the substance of the shell is traversed by more or
less delicate canals filled with the protoplasm, which thus

Fig, 3.— Diagrams of Foraminifera.—A^ monothalamian ; B, (7, polj'thalamian ; i?,
horizontal ; i'and F, vertical sections of helicoid form. In E^ ihe chambers of
each turn of the spiral overlap their predecessors and conceal them, as in the
genus Numinuiites.

reaches the surface and gives off pseudopodia all over the
body. Hence, while the hard parts of the Imperforata form
a sort of exoskeleton, those of the Perforata have rather
the nature of an endoskeleton.

The simplest skeletons are spherical or flask-shaped, and
single-chambered. But complication arises by the addition
of new chambers, which may form a linear series, or be coiled
upon one another in various ways, or be irregularly aggre-
gated. Moreover, the new chambers may overlap those al-
ready formed in different degrees, and the interspaces between
the walls of the chambers may be variously filled up by sec-
ondary deposition until such large and apparently compli-
cated bodies as the Nummulites are built up.

The Foraniinifera are almost all marine animals, living
in the sea, from the surface to great depths, sometimes free,
and sometimes attached to other bodies.

The investigations of Major Owen, confirmed and extend'
ed by the recent work of H. M. S. Challenger, have proved
that such forms as Glohigerina^ Puli^inularia^ and OrhuUna^

80 THE ANATOMY OF IXVERTEBRATED ANIMALS.

constantly occur at the surface of all temperate and tropical
seas, and, together with the HacUolaria and the diatoma-
ceous plants which accompany them, form an important in-
gredient in the food of pelagic animals, such as the Salpcp..

It is no less certain that, at all depths down to 2,400 fath-
oms or thereabouts, GlobigeriiicB i\\ all stages of growth, and
containing more or less protoplasmic matter, are found at the
bottom mixed with the cases of the surface Diatoms and the
skeletons of Itadiolaria. The proportion of Glohigerincp,
OrbuUnce, and Pulvinularice, in the deep-sea mud increases
with the depth until, at depths beyond 1,000 fathoms, the
sea-bottom is composed of a fine, chalky ooze made up of
little more than the remains of these Foraminifera and their
associated Diatoms and JRadiolaria.

It may be regarded as certain, therefore, that some of the
chalky ooze arises from the precipitation to the bottom of the
skeletons of dead Glohigerinoe^ Pulmnularia?^ and Orbulince^
and it may be that the whole has this origin. On the other
hand, it may be that a greater or smaller proportion of these
Foraminifera really live at the bottom, as their congeners
are known to do at less depths.

It has been said that the condition of the surface-waters
and sea-bottom which has just been described obtains in all
temperate and hot seas ; or, speaking roughly, for 55° on
either side of the equator. Toward the northern and south-
ern limits of this zone the Foraininifera diminish, while Fui-
diolaria remain and Diatomaccce increase in proportion, so
that, in the circumpolar areas north and south of G0° in each
hemisphere, tlie surface-organisms are chiefly such as have
silicious skeletons. In accordance with this condition of the
surface-life, the ooze covering the sea-bottom in these regions
is no longer calcareous but silicious, being composed of the
cases of Diatoms and the skeletons of Fadiolaria often
largely mixed with ice, drifted mud, stones, gravel, and bowl-
ders.

If we suppose the globe to be uniformly covered with an
ocean 1,000 fathoms deep, the solid land forming its bottom
would be out of reach of rain, waves, and other agents of
degradation, and no sedimentary deposits would be formed.
But if Foraminifera and Diatoms, following the same huvs
of distribution as at present obtain, were introduced into
this ocean, the fine rain of their silicious and calcareous hard
parts would commence, and a circumpolar cap of silicious
deposit would begin to make its appearance in the north and

PROTOZOA AS ROCK-BUILDERS. 81

in the south ; while the intermediate zone would be covered
with Globijerina ooze, containing a comparatively small pro-
portion of silicious matter. The thickness of the calcareo-
silicious and silicious beds thus formed would be limited only
by time and the depth of the ocean. These strata, once ac-
cumulated, would be liable to all those influences of percolat-
ins: moisture and subterranean heat which are known to suf-
fice to convert silicious matters into opal, or quartzite, and
calcareous matters into the various forms of limestone and
marble. And such metamorphic agencies might more or less
completely obliterate the traces of their primitive structure.

But yet other changes might be effected. At the present
day, in the Gulf of Mexico, off the Agulhas Bank and else-
where, at no great depths (100 to 300 fathoms) the Fora-
miniferal mud is undergoing a metamorphosis of another
character. The chambers of the For a mi nif era become filled
by a green silicate of iron and alumina, which penetrates into
even their finest tubuli, and takes exquisite and almost in-
destructible casts of their interior. The calcareous matter is
then dissolved away, and the casts are left, constituting a
fine dark sand, which, when crushed, leaves a greenish mark,
and is known as " green-sand."

Moreover, the researches of the Challenger have shown
that in great areas of the Atlantic and Pacific Oceans over
which the sea has a depth exceeding 2,4:00 fathoms — areas in
some cases of many thousand square miles in superficies — the
bottom is covered not by Glohigerina ooze, but by a fine red
clay, which is also a silicate of iron and alumina. In this clay
no remains of Glohigerina or other calcareous organisms are
found ; but, where these great depths gradually pass into shal-
lower water, they make their appearance in a fragmentary
condition — gradually becoming more and more perfect as the
depth diminishes to 2,4:00 fathoms or thereabouts.

Xevertheless the Glohigerinm and other Foraminifera
abound at the surface over these areas as they do elsewhere,
and their remains must be rained down upon it. Why they
disappear, and what relation the red-clay mud has to them, is
a problem not yet satisfactorily solved. It has been suffo^ested
that they are dissolved and that the red clay is merely the
insoluble residue, left after the calcareous portion of their
skeletons has disappeared. In this case the red clay, like the
Glohigerina ooze, the silicious mud, and the green-sand, will
be an indirect product of living action.

Metamorphic processes operating upon clay, however, may

82 THE AXATOMY OF INVERTEBRATED ANIMALS.

convert it into slate ; and thus, all the fundamental minerals
of which rock-masses are composed may have fcrmed part of
living organisms, though no trace of their origin may be dis-
cernible in them in their final state.

Paleontology lends niuch support to the view that what
is here suggested as a theoretically possible origin of much
of the superficial crust of the globe may have been its actual
origin.

The nummulitic limestones of the Eocene epoch cover an
enormous area of Central and Southern Europe, North Africa,
West Asia, and India. And their chief mass is made up of
the more or less metamorphosed remains of Foraminifera.

The beds of chalk which underlie the nummulitic lime-
stones, and occupy a still greater area, are essentially iden-
tical with the Glohigerina ooze, the species of Globigerina
found in it being indistinguishable from those now living.
The remains of Ji oraminifera have been detected in the lime-
stones of all epochs as far as the Silurian, and Ehrenberg dis-
covered that an old Silurian green-sand, near St. Petersburg,
is composed of casts of Foraminifera just such as are now
being formed in the Gulf of Mexico. And if the JEozoon C ana-
dense be, as it appears to be, nothing but an incrusting form
of Foraminifer, the existence of these oganisms is carried back
to an epoch far beyond that at which any other evidence of
life has yet been found. So that it is possible that, as Wy-
ville Thomson has suggested, the enormously thick " azoic "
slaty and other rocks, which constitute the Laurentian and
Cambrian formations, may be to a great extent the metamor-
phosed products of Foraminiferal life.

Hence the words of Linnaeus may be literally true :

" Petrefaota non a calce, sed calx a petrefactis. Sic lapirlcs ah animalibus,
nee vice versa. Sic rupes saxei non primsevi, sed temporis filise."

And there may be no part of the common rocks, which enter
into the earth's crust, which has not passed through a living
organism at one time or another.

II. — THE e:n^doplastica.

1. The Radiolaria. — Most species of the genus Actino-
phrys or " sun-animalculo," which is common in ponds, are
simply free-swimming mvxopods with stiffish pseudopodia,
which radiate from all sides of the globular body. The sub-
stance of the latter presents one or more " contractile spaces "

THE HADIOLARIA.

83

or "vacuoles," which, rhythmically, become distended with
water, and are then obliterated by the contraction of the sur-
rounding protoplasm. But in the Actinophrys (or more
properly Actinosphceriiim) Eichornii (Fig. 4), the central
part of the protoplasm is distinguished from the rest by con-
taining a number of endoplasts. It thus leads to the RacUola-
ria [Polycistina of Ehrenberg), the simplest forms of which

zi — :? E*

Fig. A.-AcUnosphrBrium Elchhornii (after Hertwig and Lesser, " Ueber Rhi^opo-
den," Schulze's Archiv, 1876).

I.--The entire animal; c, c, contractile vacuoles. . v. x-* • i i.

II.— Part of the periphery much magnified; a, a, a, pseudopodia with stiff axial sub-
stance; n, nuclei or endoplasts. , . J i J ^;„

ni.— A very young Actinosphoerium, with only two nuclei and two pseudopodia,
much magnified.

consist essentially of a myxopod provided with filamentous,
radiating, and often anastomosing, pseudopodia. The centre
of the body is occupied bv a capsule filled with protoplasm ;

84 THE ANATOMY OF INVERTEBRATED ANIMALS.

this sometimes contains only an oil-globule, at others cells, or
nuclei, and crystalline bodies. In the layer of protoplasm

Fig. b.—Sphcerozoum pundatum.—A, a mass of the natural size ; B, two of the oval
central sacs with the colored vesicles and spicula which lie iu the investing pro-
toplasm, magnified.

Fig. Q.—Sphcerozoum ovodimare (after Haeckel), magnified.

from which the pseudopodia proceed, cell^eform bodies of a
bright-yellow color, which have been found to contain starch,
are usually developed,^ and this layer also gives rise to a skele-
ton of a horny, or, more usually, silicious character, which

» Even after the death of the Radiolarian, these yellow cells are said bv Cien-
kowsky to thrive and multiply, and the possibility that they may be parasites
must be borne in mind.

THE RADIOLARIA. 85

may have the form of detached spicula, or of coarticulated
rods, or of networks, or of plates of silicious matter, often of
the most exquisite delicacy and beauty. Most of the Hadi-
olaria are simple, solitary, and microscopical in size ; but
some, such as CollosphcBra and Splicerozoum (Figs. 5 and 6),
are formed of aggregates of such simple forms, and float, as
visible gelatinous masses, at the surface of the sea, which is
the habitation of the great majority of the Radiolaria.

The manner of multiplication and the development of the
Madiolaria have not yet been thoroughly worked out. Cien-
kowsky, however, has observed, in Collosphmra^ that the
protoplasm contained in the central capsule breaks up into
numerous rounded masses. The several capsules which are
associated together in the compound Radiolarian then be-
come isolated, by the dissolution of the protoplasm w^iich
invested and connected them, and finally burst, giving exit
to the rounded bodies ; which, while yet within the capsules,
were observed to be in active motion. The germs (for such
they appear to be) thus set free are 0.008 mm. long, ovate,
and carry two flagelliform cilia at their narrow ends ; so that
they are " monads." Each has in its interior a crystalline rod
and a few minute oil-globules. The further development of
these mastigopods has not yet been traced ; but, if, as is
probable, they pass into young Madiolaria (which, according
to Haeckel, possess no capsule, but resemble Actinosphm-
ria)^ the Radiolaria^ as members of the Endoplastica^ would
typify Protomonas among the Monera. Neither conjugation
nor fission has been observed amono; the ordinarv Radio-
laria, but both these processes take place in Actvnosphm-
rium j and, considering the resemblance of the young Radio-
laria to Actinosphmrium^ it seems probable that conjugation
and fission will yet be discovered among them.

ActinosphcBrium has been observed to undergo multipli-
cation, by division of its central substance into a certain
number of spheroids, and every spheroid becomes inclosed in
a silicious case. After a period of rest, a young Actinosphm-
riwn emerges from each of these cysts.

The marine Radiolaria all inhabit the superficial stratum
of the sea, and must fabricate their skeletons at the expense
of the infinitesimally small proportion of silex which is dis-
solved in sea-water; but, when they die, these skeletons sink
to the bottom, and there accumulate, together with the Fora-
'ininifera, in warm and temperate regions ; and with the
cases of the diatomaceous plants, which abound at the sur-

86 THE ANATOMY OF INVERTEBRATED ANIMALS.

face, along with the JRadiolaria^ all over the globe {see p. 80).
The late investigations of Archer and others have demon-
strated the existence of a considerable number of fresh-water
Radiolaria,

Extensive masses of tertiary rock, such as that which is
found at Oran, and that w^hich occurs at Bissex Hill, in Bar-
badoes, are very largely made up of exquisitely preserved
skeletons of Madiolaria. But, though there can be little
doubt that Madiolaria abounded in the Cretaceous sea, none
are found in the chalk, their silicious skeletons having prob-
ably been dissolved and redeposited as flint.

2. The Peotoplasta. — The proper Amoehce have broad
and ovate pseudopodia, and resemble JProtamoeha (p. 75) very
closely; but they present an advance upon its structure, by
exhibiting a distinct endoplast (nucleus) and a contractile
vacuole. In Arcella, there are many such nuclei. They thus
stand in somewhat the same relation to JProtamoeha as A.cti-
nophrys does to Protogenes.

Moreover, there are Amoehce in which the power of throw-
ing out pseudopodia is confined to one region of the body ;
and others, as Arcella, in w^hich a shell is formed over the
rest of the body. In other Amoehce, as A. radiosa^ the pseu-
dopodia are few, narrow, and but little mobile. But the
Amoehce present no such diversity of skeletal development as
the Forami7iifera do. They multiply by division, and in
some cases — e. g.. A, sphoerococcus of Haeckel — become en-
cysted before they divide.

Amoehce (the " proteus animalcules " of the older writers)
are not uncommon, and sometimes are very abundant, in
fresh waters ; they also occur in damp earth and in the sea,
but there is much doubt whether many of them are to be
regarded as independent organisms, or -whether they are not
rather stages in the development of other animals or even
of plants, such as Myxomycetes. Leaving out the contractile
vacuole, the resemblance of an Amoeha in its structure, man-
ner of moving, and even of feeding, to a colorless corpuscle
of the blood of one of the higher animals is particularly note-
worthy.^

3. The Gregari:^d>^ are very closely allied to the Amoe-
hce, but, in the cycle of forms through which they pass, they
curiously resemble 3Iyxastrum. In form they are spheroidal

1 Contractile vacuoles have been observed in the colorless blood-corpus-
cles of Amphibia under certain conditions.

THE GREGARIXIDyE.

87

or elongated oval bodies, sometimes divided by constrictions
into segments. Occasionally, one end of the body is pro-
duced into a sort of rostrum, which may be armed with re-
curved horny spines.

In the ordinary GregarinoPy the body presents a denser
cortical layer (ectosarc) and a more fluid inner substance
(endosarc), in which last the endoplast (nucleus) is imbed-
ded. The presence of contractility is manifested merely by
slow changes of form, and nutrition appears to be effected by
the imbibition of the fluid nutriment, prepared by the organs
of the animals in which the Gregarince are parasitic. There
is no contractile vacuole.

The Greg ar mm have a peculiar mode of multiplication,
sometimes preceded by a process which resembles conju-
gation. A single Gregarbia (or two which have become
applied together) surrounds itself with a structureless cyst.

Fio. 7.— .4, aregarina of the earthworm (after Lieberkuhn) ; B. encvsted ; C. D,
contents divided into peeudo-navicellae ; E, F, free pseudo-navicel!», G, H, free
amcebiform contents of the latter.

The nucleus disappears, and the protoplasm breaks up (in a
manner very similar to that in which the protoplasm of a

88 THE ANATOMY OF INVERTEBRATED ANIMALS,

sporangium of Mucor divides into spores) into small bodies,
each of which acquires a spindle-shaped case, and is known
as a 'pseudo-nanicella. On the bursting of the cyst these
bodies are set free, and, when placed in favorable circum-
stances, the contained protoplasm escapes as a small active
body like a Protamoeha. M. E. van Beneden has recently dis-
covered a very large Gregarina (G. gigantea), which inhab-
its the intestine of the lobster, and his careful investigation
of its structure and development has yielded very interesting
results.

Gregarina gigantea attains a length of two-thirds of an
inch. It is long and slender, and tapers at one extremity,
while the other is obtuse, rounded, and separated by a slight
constriction from the rest of the body, which is cylindroidal.
The outer investment of the body is a thin structureless cu-
ticle ; beneath this lies a thick cortical layer (ectosarc), dis-
tinguished by its clearness and firmness from the semifluid
central substance (endosarc), which contains many strongly-
refracting granules. In the centre of the body, the ellipsoid
"nucleus," with its "nucleolus," fills up the whole cavity of
the cortical layer, and thus divides the medullary substance
into two portions. The body of this Gregarina may present
longitudinal striations, arising from elevations of the inner
surface of the cortical layer, which fit into depressions of the
medullary substance ; but these are inconstant. On the other
hand, there are transverse striations which are constant, and
which arise from a layer of what are apparently muscular
fibrilloe, developed in a peripheral part of the cortical layer,
immediately below the cuticle. The fibrillas themselves are
formed of elongated corpuscles joined end to end. A trans-
verse partition separates the ceplialic enlargement from the
body, and the layer of muscular fibres only extends into the
posterior part of the enlargement.

The embryos of Gregarina gigantea^ when they leave
their pseudo-navicellse, are minute masses of protoplasm simi-
lar to ProtamoebcB, and like them devoid of nucleus and con-
tractile vacuole. They soon cease to show any change of
form, and acquire a globular shape, the peripheral region of
the body at the same time becoming clear. Next, two long
processes bud out from this body; one is actively mobile, the
other still. The former, detaching itself, assumes the appear-
ance and exhibits the motions of a minute thread-worm,
whence M. van Beneden terms it a pseudo-filaria. The en-
largement at one end becomes apparent, the pseudo-filaria

THE INFUSORIA. 89

passes into a quiescent state, and the "nucleolus " makes its
appearance in its interior. Around tliis a clear layer is differ-
entiated, giving rise to the " nucleus," and the pseudo-fllaria
passes into the condition of the adult Gregarina gigantea.

4. The Catallacta of Haeckel, represented by the genus
Magosphoera^ are, in one stage, myxopcds with long pseudo-
podia, which, broad and lobe-like at the base, break up into
fine filaments at their ends, and may therefore be said to be
intermediate between those of JProtogenes and those of Prot-
aniceha. The myxopod is provided with a distinct endoplast
and a well-marked contractile space. When fully fed, it se-
cretes a cyst and divides into a number of masses, each of
which is converted into a conical body, with its base turned
outward and its apex inward. These conical bodies are im-
bedded in gelatinous matter, and thus cohere into a ball, from
the centre of which they radiate. Each develops cilia around
its base, and contains an endoplast and a contractile vacuole.
After the complex globe thus formed has burst its envelope,
it swims about for a while, like a Volvox. The several cilia-
ted animalcules feed by taking in solid particles through the
disk. They then separate, and, finally, retracting their cilia,
become myxopods such as those with which the series started.
MagosjyJic^ra is thus very nearly an endoplastic repetition of
the moneran Protomonas — the mastigopod being provided
with many small cilia, instead of with a couple of large fla-
gella. On the other hand, the Catallacta are closely allied
to the next group, and, I am disposed to think, might well be
included in it.

5. The Ixfusoeta. — Excluding from the miscellaneous as-
semblage of heterogeneous forms, which have passed under
this name, the Pesmidicp^ Piatomaccce^ Volvochiece, and

V^ibrionidce, which are true plants, on the one hand ; and the
comparatively highly-organized Potifera^ on the other ; there
remain three assemblages of minute organisms, which may be
conveniently comprehended under the general title of Infu-
soria. These are — («) the so-called " ]\Ionads," or Pnfusoria
Jlagellata ; (b) the Acinetae^ or Pnfusoria tentacuUfera ^ and
(c) the Pnfusoria ciliata.

(a.) The Flagellata. — These are characterized by pos-
sessing only one or two long, whip-like cilia, sometimes (when
more than one are present) situated at the same end of the
body, sometimes far apart. The body very generally exhib-
its an endoplast and a contractile vacuole. There is no per-
manently open oral aperture, but there is an oral region, into

90 THE ANATOMY OF INVERTEBRATED ANIMALS.

which the food is forced, and, passing into the endosarc, re-
mains for some time smrounded by a globule of contempo-
raneously ingested water — a so-called " food- vacuole." Prof.
H. James Clark, who has recently carefully studied the Fla-
gellata, points out that, in Bicosoeca and Codonoeca, a fixed
monadiform body is inclosed within a structureless and trans-
parent calyx. In Codosiga a similar transparent substance
rises up round the base of the flagellum, like a collar. Jn
Salimigceca the collar around the base of the flagellum is
combined w ith a calycine investment for the whole animal.
In A^ithophysa, there are two motor organs — the one a stout
and comparatively stiff flagellum, which moves by occasional
jerks, and the other a very delicate cilium, which is in con-
stant vibratory motion.

The discrepancy between the two kinds of locomotive
organs attains its maximum in Aniso7ie7na.f which presents
interesting points of resemblance to Noctiluca.

Multiplication by longitudinal fission was observed in
Codosiga and Anthophy s a, find ^roh^hly occurs in the other
genera. In Codosiga the flagellum is retracted before fission
takes place, but the body does not become encysted ; in An-
thophysa the body assumes a spheroidal form, and is sur-
rounded by a structureless cyst, before division occurs.

Conjugation has not been directly observed among most
of the I)tfusoria flagellata^ nor do any of them exhibit a
structure analogous to the endoplastule of the Ciliata,

Messrs. Dallinger and Drysdale have recently worked out
the life-history of several flagellate " Monads," which occur
in putrefying infusions of fish. They show that these Ha-
gellata not only present various modes of agamic multiplica-
tion by fission, preceded or not by encyst ment, but that they
conjugate, and that the compound body which results (the
equivalent of the zygospore in ])lants) becomes encysted.
Sooner or later, the contents of the cyst become divided
either into comparatively large or excessively minute bod-
ies, which enlarge and gradually take on the form of the
parent.

The careful investigations of these authors lead them to
conclude that, while the adult forms are destroyed at from
61°-80° C, the excessively minute sporules which have been
mentioned, and which may have a diameter of less than
i^QQQQQ of an inch, may be heated to 148° C. without the
destruction of their vitality.

In Euglena viridis (which, however, may be a plant),

THE FLAGELLATA. 91

Stein ^ has observed a division of the "nucleus" to take place,
whereby it becomes converted into separate masses, some of
which acquire an ovate or fusiform shape, surrounding them-
selves with a dense coat, while others become thin-walled
sacs, full of minute granules, each of which is provided with
a single cilium. The ultimate fate of these bodies has not
been traced.

A careful study of the singular genus N'octiluca led me,
in 1855, to assign it a place among the Infusoria^ and recent
investigations have conclusively proved that it is one of the
Flagellata.

The spheroidal body of N^octiluca oniliGrls (Fig. 8) is
about one-eightieth of an inch in diameter, and, like a peach,
presents a meridional groove, at one end of which the mouth
is situated. A long and slender, transversely striated ten-
tacle overhangs the mouth, on one side of which a hard-
toothed ridge projects. Close to one end of this is a vibratile
cilium. A funnel-shaped depression leads into a central
mass of protoplasm, connected by fine radiating bands with
a layer of the same substance which lines the cuticular enve-
lope of the body. There is no contractile vacuole, but an
oval endoplast lies in the central protoplasm. Bodies which
are ingested are lodged in vacuoles of the latter until they
are digested.

According to the observations of Cienkowsky,'^ if a JS'oc-
tiluca be injured, the body bursts and collapses, but the pro-
toplasmic and other contents, together with the tentacle, form
an irregular mass, the periphery of which eventually becomes
vacuolated, enlarges, and secretes a new investment. But
even a small portion of the protoplasm of a mutilated Nocti-
luca is able to become a perfect animal. Under some condi-
tions, the tentacle of a Noctiluca may be retracted into the
body, and, at all times of the year, spheroidal N'octilucce,
devoid of flagellum, tooth, or meridional groove, but other-
wise normal, are to be found. These last are probably to be
regarded as encj^sted forms. Multiplication may take place
in at least two ways. Fission may occur in the spheroidal
forms, as well as in those possessed of a tentacle ; it is in-
augurated by the enlargement, constriction, and division, of
the endoplast. This process takes place more especially in
the latter part of the year.

1 " Oreranismus der Infusionsthlere," ii., 5fi.

» "Ueber Noctiluca miliaris." (Schulze's " Arcliiv fiir mikroskop. Anato-
mie," 1872.)

92 THE ANATOMY OF IXYERTEBRATED ANIMALS.

Another mode of a sexual multiplication, which has a sin-
gular resemblance to the process of partial yelk division,

Fig. 8. — XoctUiica miliai^.—e, gastric vacuole ; g, radiating filaments ; /, anal

aperture (.?).

occurs only in the spheroidal JS^octihicce. The endoplast dis-
appears, and the protoplasm, accumulating on the inner side
of one region of the cuticle, divides first into two, then four,
eight, sixteen, thirty-two, or more masses ; the division of the
protoplasm being accompanied by the elevation of the cuticle
into protuberances, which, at first, corresjDond in number and
dimensions with these division masses. When the division
masses have become very numerous, each protrudes upon the
surface, and is converted into a free monadiform gernr., pro-
vided with an endoplast, a beak, and a long tentacle, which
is hardly to be distinguished from a flagelliform cilium.

The process of conjugation has been directly observed.
Two JVoctilucce, applying themselves by their oral surfaces,
adhere closely together, and a bridge of protoplasm connect-
ing the endoplasts of the two becomes apparent. The ten-
tacula are thrown off, the two bodies gradually coalesce, and
the endoplasts fuse into one. The whole process occupies
five or six hours. Spheroidal or encysted JVoctllucce may
conjugate in a similar manner. In this case, the regions
nearest the endoplasts are those which become applied to-
gether. Whether this process is of a sexual nature, or not,
is not clearly made out. Cienkowsky admits that it may

THE FLAGELLATA, 93

hasten the process of multiplication by monadiform germs
described above.

Ifoctiluca is extremely abundant in the superficial waters
of the ocean, and is one of the most usual causes of the phos-
phorescence of the sea. The light is given out by the pe-
ripheral layer of protoplasm which lines the cuticle.

The PeridlnecB (see Fig. 1, /) form another aberrant
group of the Flagellata, which lead to the Ciliata. Tiie
body is inclosed in a hard case (sometimes produced into
rays), which, at one part, presents a groove-like interruption,
laying bare the contained protoplasm, in which lies an endo-
plast, and in some cases a contractile vacuole. One or more
flagelliform cilia, and usually a wreath of short cilia, are pro-
truded from the protoplasm, and serve as locomotive organs.
The mouth is a depression, whence, in some cases, an oeso-
phageal canal is continued and terminates abruptly in tlie
semi-fluid central substance of the body, the food-particles
being lodged in vacuoles formed at its extremity, as in the
Ciliata. No other mode of multiplication than that by fission
has vet been observed in the Peridinece ; but this fission is
sometimes preceded by the inclosure of the animal in an
elongated, crescent-shaped cyst.

(b.) The TEN-XACULiFER-i. — The Acinetce (Fig. 9, D, E,
F^ G) have no oral aperture of the ordinary kind, but filiform
processes or tentacula, which are usually slender, simple, and
more or less ri2:id, radiate from the surface cf the bodv gen-
erally, or from one or more regions of that surface. At first
sight, these tentacula resemble the radiating pseudopodia of
A.oti)iophrys, but, on closer inspection, they are seen to have
a different character. Eich, in fact, is a delicate tube, pre-
senting a structureless external wall, with a semi-fluid granu-
lar axis, and usually ends in a slight enlargement or knob. It
may be slowly pushed out or retracted, or diversely bent.
But, instead of playing the part of mere prehensile organs,
these tentacles act, in addition, as suckers; the Acineta ap-
plying one or more of these organs to the body of its prey ^ —

» Stein ("Der Organismus der Infusionsthiere," i., VC) thus describes the
method by which an Acineta seizes its prey : " If an Infusorium swims within
reach of the Acineta^ the nearest tentacles are swiftly thrown toward it, and, at
the same time, often become much elongated, bent, or irregularly twisted about.
The knob-like ends of these tentacle's, which come into immediate contact
with the surface of the entangled prey, spread out into disks, and adhere iixedly
to it. When many of the tentacles' have thus attached themselves, the im-
prisoned animal is no longer able to escape, its movements become slower, and
at length cease. Those tentacles which have fixed themselves most firmly
shorten and thicken, and draw the prev nearer to the body. . . . Suddenly, as

94

THE ANATOMY OF IXVERTEBRATED ANIMALS.

usually some other species of Infusorium — when the substance
of the latter travels along the interior of the sucker into the

Fio. 9.—^, Yorticella, active ; B, C, encysted ; Z>, E, F, G, Acinetoe (after Stein).

body of the Acineta. Solid food is not ingested through these
tentacles, so that the Acinetce cannot be fed with indigo or
carmine. In the interior of the body there is an endoplast *
with one or more contractile vacuoles, and it may be either
fixed by a stalk or free.

The Acinetce multiply by several methods. One of these
is simple longitudinal fission, which appears to be rare among
them. Another method consists in the development of ciliated
embryos in the interior of the body. These embr^^os result
from a separation of a portion of the endoplast, and its con-

soon as the suc'kinor (5isk has bored throncrli the cuticula of the prey, a very-
rapid stream, indicated by the fattv particles which it carries, sets along the
axis of the tentacle, and, at its base, pours into the rxcighboring part, of the
body of the Acineta. . . . The cause of the movement is unknown. It is not,
accompanied bv any discernible movement of the walls of the tentacle."

' No endoplastule, such as exists in other Infusoria^ has been observed as
yet in the Acivdoe. Under some circumstances, tlie Acinetm draw in their
radiating processes, and surround themselves with a structureless cyst; but
this process does not appear to have any relation to either mode of multiplica-
tion.

In Acineta mystacina and Porlopliryafixa, a peculiar mode of multiplication
bv division occurs. At the free end of the body a portion becomes constricted
off, together with part of the endoplast, from the remaining stalked part. The
tentacula are drawn in, and the segment becoming elongated, develops cilia
over its whole surface and swims away.

THE INFUSORIA. 95

version into a globular or oval germ, which, in some species,
is wholly covered with vibratile cilia, while, in others, the cilia
are confined to a zone around the middle of the embryo.
The germ makes its escape by bursting through the body-wall
of its parent. After a short existence (sometimes limited to
a few minutes) in the condition of a free-swimming animal-
cule, provided with an endoplast and a contractile vacuole,
but devoid of a mouth, the characteristic knobbed radiating-
processes make their appearance, the cilia vanish, and the ani-
mal passes into the Acineta state.

The Acinetce have frequently been observed to conju-
gate, the separate individuals becoming completely fused into
one and their endoplasts coalescing into the single endoplast
of the resultant Achieta y but it is not certainly made out
whether this process has, or has not, anything to do with the
process of the development of ciliated embryos just described.

(c.) The Ciliata. — The characteristic feature of the Ciliata
is, that the outer surface of the body is provided with numer-
ous vibratile cilia, which are the organs of prehension and loco-
motion. According to the distribution of the cilia, Stein has
divided them into the Holotricha^ in which the cilia are scat-
tered over the whole body, and are of one kind ; the Hetero-
tricha, in v/hich the widely-diffused cilia are of different kinds,
some larger and some smaller ; the Ilypotricha^ in which the
cilia are confined to the under or oral side of the body; and the
JPeritricha^ in which they form a zone round the body. The
great majority of these anim.als are asymmetrical.

In the simplest and smallest Ciliata^ the body resembles
that of one of the Flagellata in being differentiated merely
into an ectosarc and endosarc, with an endoplast and a con-
tractile vacuole. In most, if not all cases, however, there
is not only an oral region, through which the ingestion of
food takes place, but an oesophageal depression leads from
this into the endosarc ; and it m.ay be doubted whether, even
in the simplest Ciliata, there is not an anal area through
which the undigested parts of the food are thrown out.

The genus Colpoda, which is very common in infusions of
hay, is a good example of this low form of ciliated Infuso-
rium. It has somewhat the form of a bean flattened on one
side, and moves actively about by means of numerous cilia,
the longest of which are situated at the interior end of the
body. At i\\Q posterior end is the contractile vacuole, while
a large endoplast lies in the middle, as Stein originally dis-
covered. Colpodm frequently become quiescent, retract their

96 THE ANATOMY OF INVERTEBRATED ANIMALS.

cilia, and surround themselves with a structureless cyst. Each
encysted Colpoda then divides into two, four, or more por-
tions, ^yhich assume the adult form and escape from the cysts
to resume an active existence.

Allman has described the encystment of a Vorticellidan,
followed by division of the nucleus into many germs, with-
out any antecedent process of conjugation ; and Everts has
observed that the progeny of an encysted Yorticella take on
the form of Trichodina grandinella. The Trichodi7ice mul-
tiply by transverse divisions, and then grow into Vb'rti-
cellce.^

Encystment, whether followed or not by division, is very
common among all the Cillata^ and a species of Amphilep-
tus has been seen to swallow — or rather envelop — a stalked
bell-animalcule (Vorticella), and then become encysted upon
the stalk of its prey, just as Vam^yi/rella becomes perched
upon the stalk of the devoured Gomphonema.

In the higher Clliata^ the protoplasm of the body becomes
directly differentiated into various structures, in the same
way as has already been seen to be the case in Gregarina
gigantea^ but to a much greater degree.

Thus, in the Peritricha, of which the bell-animalcules, or
Vortieellce (Fig. 9, A^ B, (7), are the commonest examples,
the oral region presents a depression, the vestibule (Fig. 9, a)
from which a permanent oesophageal canal leads into the soft
and semi-fluid endosarc, where it terminates abruptly ; and
immediately beneath the mouth, in the vestibule, there is an
anal region which gives exit to the refuse of digestion, but
presents an opening only when fecal matters are passing
out. Except where the ciliated circlet, or rather spiral, is
situated, the outer wall of the body gives rise to a relatively
dense cuiicula^ and not unfrequently secretes a transparent
cup or case, foreshadowing the theca of hydrozoal polyps.
Moreover, in the permanently fixed Vorticelkr, the stalk of
attachment may present a central muscular fibre (Fig. 9,/"),
by the sudden contraction of which the body is retracted,
the stalk being at the same time throwm into a spiral. In
the holotrichous Paramcecium (Fig. 10) beneath the thin su-
perficial transparent cuticle from which the cilia proceed,
there is a very distinct cortical layer, fibrillated in a direc-
tion perpendicular to the surface, and, in some species of this
or other genera, as StromhkUuni and Polykricos (Biitschli),
beset with minute rod-like bodies similarly disposed, w^hich,

Allman, " Presidential Address to the Linnsean Societv," 1875,

THE INFUSORIA.

97

under some circumstances, shoot out into long filaments,
and have been termed trlchocysts. In I*, bursariay minute

WWW

Fi». 10. — Paramecium bursaria (after Stein).— ^, the animal viewed from the dorsal
side : a, cortical layer of the body ; b, endoplast ; c, contractile space ; d d', mat-
ters taken in as food ; e, chlorophyl granules.

JB, the animal viewed from the ventral side: a, depression leading to 5, mouth ;
c, gullet ; d, euJoplast; rf', endoplastule ; e. central protoplasm. In both these
figures the arrowi? indicate the direction of the circulation.

C, ParamcRcimn dividing trausvere-ly : a a\ contractile spaces ; b b, endoplast divid-
ing ; c c\ endoplastules.

green granules of chlorophyl are dispersed tlirough this layer,
and Cohn demonstrated, in 1851, that these yield the same
reactions as the chlorophyl grains of the Algae. In Salanti-
dium^ N^yctotherus^ Spirostomum^ and many others, the cor-
tical layer is divided bv linear markino^s into bands, which
there is reason to believe are rudimentary muscular fibres.

In many ClUata, the endosarc appears to be almost fluid.
The food, which is driven into the mouth and down the oesoph-
agus by the constant action of the cilia, accumulates at the
bottom of the oesophagus ; and then, with the water which
surrounds it, is passed, at intervals, with a sort of jerk, into
the endosarc, where it lies close to the end of the oesophagus,
as a food-vacuole, for a short time. But it soon begins to
move, and, along with other such vacuoles formed before and
after it, circulates in a definite course up one side of the body
and down the other, between the cortical layer and the endo-
plast. This movement is particularly free and unrestricted in
Salantidmmfi • in Parcnnoechim, the tract through which the
food- vacuoles move is more definitely limited,* while in iVyc-

1 In Faramczcium hursaria Colin observed that the circulation was completed
in Isf to 2 minutes, which gives a rate of rotation of ^oW to fj^oo oi' an inch in
a second.

98 THE ANATOMY OF INYERTEBRATED ANIMALS.

totherus it appears to be confined to a part of the body be-
tween the end of the gullet and the anal region, which in
this Infusorium is seated at one end of the body. In fact, the
finely granular endosarc of Nyctotherus so limits the passage
of the food-vacuoles that the tract along which they pass
might properly be described as a rudimentary intestinal canal.

The oral cavity is usually ciliated : sometimes, as in ChilO'
don, it has a chitinous armature, which becomes somewhat
complicated in Ermlia {Dysteria ^) and the Didinium de-
scribed by Balbiani.

Torquatella (Lankester) has a plicated membrane around
the mouth in the place of cilia.

The contractile vacuoles attain their greatest complexity
in the Paramoecia, in which there are two — one toward each
end of the body. They are lodged in the cortical layer, and,
in diastole, a portion of their outer periphery is bounded only
by the cuticle, through which it is very probable that they
communicate with the exterior. When the systole takes
place, a number of fine canals, which radiate from each vac-
uole, are seen to become distended with clear, watery fiuid.
These canals are constant in their position, and some of
them may be traced nearly as far as the mouth ; so that the
canals and vacuoles form a permanent water-vascular system.
The endoplast is finely granular, like the substance of the
endosarc. It is frequently said to be enveloped in a distinct
membrane, but I am disposed to think that this is always a
product of reagents. Attached to one part of it there is very
generally (but not in the Vorticellce) a small oval or rounded
body, the so-called "nucleolus" or e?z^op?as^w?e. The endo-
plast is commonly said to be imbededd in the cortical layer,
but this is certainly not the case in Colpoda, Paramoecium^
Palantidium, or Kyctotherus,

The outermost, or cuticular, layer of a large portion of the
body becomes hardened and forms a sort of shell, in many of
the free Infusoria. In the free marine Pictyocystida and
Codonellida of Haeckel, the body has a bell-shaped enve-
lope, which in the Pictyocystida {see Fig. 1) is strengthened
by a siliceous skeleton like that of a Radiolarian. In both
genera the circular lip which surrounds the oral end is pro-
vided with numerous long flagelliform cilia."

Most of the Ciliata, while in full activity, multiply by di-

^ Huxley, "On Dysteria." ( Quarterly Journal of Microscopical Science^ 1857.)
3 Haeckel, " Zur Morphologic der lufusorien," 1873.

THE INFUSORIA. 99

vision ; this is generally effected by the formation of a more
or less transverse constriction, whereby the body becomes
divided into two parts, which separate, each developing those
structures which are needed for its completion. The endo-
plast, however, always elongates and divides, one portion
going along with each product of fission. Neither budding
nor longitudinal fission occurs among the free Infusora, the
appearances which have been regarded as evidence of these
processes being due to the opposite operation of conjugation.
M. Balbiani,' its discoverer, thus describes the process of conju-
gation in Paramoecium hursaria :

*' The Paramoecia assemble in great numbers either tow-
ard the bottom or on the sides of the vessel in which they
are contained. They then conjugate in pairs, their anterior
ends being closely united ; and they remain in this state for
five or six days or more. During this period the nucleus and
nucleolus become transformed into sexual organs.

" The nucleolus is changed into an oval capsule, marked
superficially by longitudinal striag. Sooner or later, it usually
becomes divided into two or four portions, which grow inde-
pendently, and form many separate capsules. About the time
of separation, each of these is found to be a capsule containing
a bundle of curved rods {baguettes), enlarged in the middle,
and thinner at the ends.

" The nucleus also becomes enlarged, and gives rise — in a
manner not clearly explained — to small spherical bodies anal-
ogous to ovules.

" It is usually about the fifth or sixth day after conjuga-
tion that the first germs appear, as little rounded bodies formed
of a membrane which is rendered visible by acetic acid, and
of grayish pale homogeneous or almost imperceptibly granu-
lar contents, in which, as yet, neither nucleus nor contractile
vacuole is distinguishable. It is only later that these organs
appear. The observations of Stein and of F. Cohn have
shown how these embryos leave the body of the mother un-
der the form of Acinetce, provided with knobbed tentacles and
true suckers, by means of which they remain for some time
adherent to her, and nourish themselves from her substance.
But their investigations have not disclosed the ultimate fate
of the vouno;.

" I have been able to follow them for a long period after

» Balbiani, " Note relative a I'Existence d'une Generation Sexuelle chez
les Infusoires." {Journal de la Physiologic^ tome i., 1858.)

100 THE ANATOMY OF IXVERTEBRATED ANIMALS.

their detachment from the maternal organism ; and I have
been able to assure myself that, after having lost their ten-
tacles, becoming clothed with vibratile cilia, and acquiring a
mouth, which makes its appearance as a longitudinal groove,
they return definitely to the parental form, developing in
their interior the green granules which are characteristic of
this Pararaoecium^ without undergoing any more extensive
metamorphosis."

In Figs. 19-22 of Plate IV., which accompanies his paper,
Balbiani figures all the stages by which the acinetiform em-
brj'o becomes a Faramoecium.

So far as the fact of conjugation, the changes in the " nu-
cleolus," and the development of filaments in it, with the
subsequent detachment, by division, of masses from the " nu-
cleus," are concerned, these statements have not been modi-
fied by M. Balbiani, while they are fully confirmed by the ob-
servations made by himself, Claparede and Lachmann, Stein,
Kolliker, and others, in Paramoecium bursaria, P. aurelia,
and other ciliated Pnfusoria.

In the closely allied Paramoeciuni aiirelia, the occurrence
of the various stages of conjugation, conversion of the " nu-
cleolus " into bundles of spermatozoa, and subsequent division
of the " nucleus," is also established by the coincident testi-
mony of Balbiani and Stein. Balbiani affirms that, in this spe-
cies, the clear globular bodies which result from the division
of the " nucleus " pass out of the body without undergoing
any further modification, and he considers them to be ovules.
Stein also admits that he has never seen acinetiform embryos
in this species.

But, as it would seem, on the strength of these negative
observations in Paramoechnn aurelia^ Balbiani, in his later
publications, asserts that the " acinetiform embryos " observed
not only in Paramoecium^ but in Styloi^ychia, Stentor, and
many other ciliated Infusoria^ are not embryos at all, but
parasitic Acinetm / and he makes this assertion without ex-
plicitly withdrawing the statement given above of his own ob-
servation of the passage of the acinetiform embryo of Para-
moecium hursaria into the parental form. Engelmann and
Stein, on the other hand, hold by Balbiani's original doctrine,
and ffive stronfj reasons for so doing". Amonof- the most for-
cible analogical arguments are those afforded by the process of
sexual reproduction observed by Stein in the peritrichous In-
fusoria.

In the Perltrlcha ( VorticelUdce^ Ophrydidce^ TrlchodidcB)

THE INFUSORIA. lOl

conjugation takes place by the complete and permanent
fusion of two individuals, which are sometimes of equal
dimensions ; though, in other cases, one is much smaller than
the other, and, while it is in course of absorption, looks like a
bud, and was formerly taken for such (Fig. 9, A, g^ h). Ihe
small individuals usually take their origin from a group of
small stalked VorticellGe^ which are produced by the repeat-
ed longitudinal division of a Vorticella of the ordinary size.
The result of the conjugative act is that the " nuclei " of the
two individuals, either before or after their coalescence,
break up into a number of segments. The segments may
remain separate, or coalesce into a single mass, called by
^tQm. i^lacenta. In the former case, some of the segments
become germ-masses, while the others reunite to form a new
"nucleus ;" in the latter, the placenta throws out a number
of germ-masses, and then assumes the form of an ordinary
" nucleus." The germ-masses give oft' portions of their sub-
stance, including part of their " nucleus," and these become
converted into ciliated embrj^os, which escajDe by a special
opening. Knobbed tentacles, like those of the Acinetcp,
have not been observed in the embryos of the Peritricha^
nor has their development been traced out.

If the bodies regarded as acinetiform embryos of the
Ciliata are really such, they may be taken to represent the
myxopod stage of the Catallacta, and the relations of the
Acinitm to the Ciliata would appear to be that they arc
modifications of a common type, differing from the Catal-
lacta in having tentacula instead of ordinary pseudopodia.
In the Acinetce^ the tentaculate stage is the more permanent,
the ciliated stage transitory ; while, in the Ciliata, the cili-
ated stage is the more permanent, and the tentaculate stage
transitory.

CHAPTER III.

THE PORIFERA A]S"D THE CCELENTEKATA.

1. The Porifera or Spoxgida. — It has been seen that,
in the Protozoa^ the germ undergoes no process of division
analogous to the " yelk division " of the higher animals, and
to the corresponding process by which the embryo cell of
every plant but the very lowest becomes converted into a
cellular embryo. Consequently, there is no blastoderm ; the
body of the adult Protozoon is not resolvable into morpho-
logical units, or cells, more or less modified ; and the aliment-
ary cavity, when it exists, has no special lining. Moreover,
the occurrence of sexual reproduction in most of the Proto-
zoa is doubtful, and there is, at present, no evidence of the
existence of male elements, in the form of filamentous sper-
matozoa, in any group but the Infusoria / and even here the
real nature of these bodies is still a matter of doubt.

In all the Metazoa^ the germ has the form of a nucleated
cell. The first step in the process of development is the
production of a blastoderm by the subdivision of that cell
and the cells of the blastoderm give rise to the histoloo-ical
elements of the adult body. With the exception of certain
parasites, and the extremely modified males of a few species,
all these animals possess a permanent alimentary cavity,
lined by a special layer of cells. Sexual reproduction always
occurs ; and, very generally, though by no means invariably,
the male element has the form of filiform spermatozoa.

The lowest term in the series of the Metazoa is un-
doubtedly represented by the Porifera or Sponges, which,
after oscillating between the vegetable and the animal king-
doms, have, in recent times, been recognized as animals by
all who have sufficiently studied their structure and the
manner in which their functions are performed.

But the place in the Animal Kingdom which is to be as-
signed to the sponges has been, and still is, a matter of de-

THE PORIFERA. 103

bale. It is certain tliat an ordinary sponge is made up of an
aggregation of corpuscles, some of which have all the charac-
ters of ji)noebce, while others are no less similar to Monads ;
and therefore, taking adult structure only into account, the
comparison of a sponge to a sort of compound Protozoon is
perfectly admissible, and, in the absence of other evidence,
would justify the location of the sponges among the Protozoa,

But, within the last few years, the development of the
sponges has been carefully investigated ; and, as in so many
other cases, a knowledge of that process necessitates a recon-
sideration of the views suggested by adult structure.

The impregnated ovum undergoes regular division ; a blas-
toderm is formed, consisting of two layers of cells — an epiblast
and a hypoblast — and the young animal has the form of a
deep cup, the w^all of which is composed of two layers, an ec-
toderm and an endoderm^ which proceed respectively from the
epiblast and hypoblast. The embryo sponge is, in fact, simi-
lar to the corresponding stage of a hj'drozcon, and is totally
unlike any known condition of a protozoon.

Beyond this early stage, however, the sponge-embryo
takes a line of its own, and its subsequent condition differs
altogether from anj^thing known among the Ccelenterata j all
of which, on the other hand, present close and intimate resem-
blances in their future development, as in their adult structure.

It is not long since the only sponge of the structure and
development of w^hich we were accurately informed was the
Spongilla jfuviatilis^ or fresh-water sponge, the subject of the
elaborate researches of Lieberkuhn and Cai'ter. But, recently,
a flood of light has been thrown upon the morphology and phys-
iology of the marine sponges, particularly of those sponges
with calcareous skeletons, which are termed Ccdcispongic^^
bv Lieberkuhn, Oscar Schmidt, and especially Haeckel. It
has become clear that Spongilla is a somewhat aberrant
form, and that the fundamental type of Poriferal organization
is to be sought among the Calcispongice. In the least com-
plicated of the calcareous sponges, the body has the form of
a cup, and is attached bv its closed extremity. The open ex-
tremity is the osculum^ and leads directly into the spacious
ventricidus^ or cavity of the cup. The comparatively thin
wall of the cup is composed of two layers, readily distinguish-
able by their structure — the outer is the ectoderm^ the in-
ner the endodertn. The ectoderm is a transparent, slightly
granular, gelatinous mass in which the nuclei are scattered, but
which, in the unaltered state, shows no trace of the primitive

104 THE iiNATOMT OF IXVERTEBRATED AmifALS.

Fio. 'il.—Ascfiffa vrimorcUalis (after Ilaecke]).

I. A mature Asceffa, Tpart of one side of the body of which i? removed: o, the exhal-
ei)t aperture ; p. inhnleut pore? in the wall of the body ; i, endoderm ; <?, ecto-
derm ; (7, ova. The triradiate spicula are seen imbedded in the ectoderm.

n. A portion of the endoderm. with two pores (p); i. endodermal cells— those round
the marjrins of the pores have their cilia directed inward ; e, ectodermal syncy-
tium: (7, ova ; z, sperm-cells.

ITT. A monadiform endodermal coll.

IV. An endodermal cell, with retracted cilium, and having the characters of an
Ammbn.

V. The ciliated emhryo of A'cetfa mirrrbUh.

VI. The same embryo in optical longitudinal section: e, epiblast : i, hypoblast ; «?,
blastocoele.

THE POiaFERA. 105

distinctness of the cells which contain these nuclei, and is
therefore termed by Haeckel a syncytium. It is elastic and
contractile, and sometimes exhibits an approach to fibrillation.

The endoderm, on the contrary, is composed of a layer
of very distinct cells, each of which contains a nucleus and
one or more contractile vacuoles, and is produced at its free
extremity into a long solitary cilium or flagellum. Around
the base of this, the transparent outer portion of the proto-
plasm of the cell is produced into an upstanding ridge like a
collar, so that each cell has a wonderful resemblance to some
forms of flagellate Infusoria. Microscopic apertures — the
pores — scattered over the outer surface of the cup, lead into
short passages which perforate the ectoderm and endoderm,
and thus place the ventriculus in communication with the ex-
terior. The workinof of the flaarella of the endodermic cells
causes the water contained in the gastric cavity to flow out
of the osculum ; to make good this outflow, minute streams
set in by the pores, which have consequently been called in-
halent^ while the osculum has been termed the exhalent aper-
ture. It is said, however, that the direction of these currents
is not invariable ; and it is certain that the pores are not
constant, but that they may be temporarily or permanently
closed, and new ones formed in other positions.

The skeleton of the calcareous sponges always consists of
a multitude of separate spicula, composed of an animal sub-
stance, more or less strongly impregnated with carbonate of
lime, which is deposited in concentric layers around a central
axis, formed by the animal basis. This skeleton is devel-
oped exclusively in the ectoderm, and is not supported by
any framework of fibrous animal matter.

The calcareous sponges are frequently, if not always,
hermaphrodite. The reproductive elements are ova and
spermatozoa. There is some reason for assuming that the
latter originate in metamorphosed cells of the endoderm, as
they are found scattered between ordinary cells of the latter.
The ova, on the other hand, occur sometimes between the
cells of the endoderm, sometimes imbedded in the syncytium
itself. But the question of the origin of the sexual elements
in these and other animals needs much further investigation.
The spermatozoa are very delicate, and have minute, rod-like
heads, with long flagella. The ova present the normal ger-
minal vesicle and spot, but exhibit active amoeboid move-
ments.

Impregnation is effected, and the first stages of develop-

106 THE ANATOMY OF INVERTEBRATED ANIMALS.

ment take place, while the ova are still imbedded in the body
of the sponge.

Metschnikoff ^ has recently described the development of
Sycoti cillatuni. The ovum, after impregnation, becomes a
morula, with a central cleavage cavity or blastocoele. But
the blastomeres of the two halves of the morula take on dif-
ferent characters — those of the one half elongating and
acquiring flagelliform cilia, while those of the opposite half
remain globular and develop no cilia. The latter now coa-
lesce into a syncytium, and develope spicula, while the layer
of ciliated cells becomes invaginated within the syncytium.
More usually, however, it appears that a gastrula is formed
by invagination of the morula, the ectoderm of which has the
structure of the endoderm of the adult, while the cells of the
endoderm, or lining membrane of the gastric cavity, are de-
void of cilia. The embryo quits the parent, propelled by the
flagelliform cilia which cover the outer surface of the ecto-
derm. After a time, it fixes itself by the closed end ; the
fiagella of the cells of the ectoderm are retracted, the cells
themselves become flattened and coalesce so completely that
their boundaries cease to be distinguishable, and the ectoderm
passes into the condition of a syncytium. At the same time,
the cells of the endoderm multiply, elongate, and take on the
form which characterizes them in the adult. In this state
the young sponge is termed an Asciila. The transition to
the final condition is effected by the development of the spic-
ula in the syncytium and the separation of some of the con-
stituent cells of the syncytium to form the inhalent pores.

In the simplest Calcis]Jongice, forming the family to
which Haeckel applies the name oi Ascones^ the wall of the
ventriculus is thin, and the pores open directly into the ven-
tricular cavity ; but in another family, the Leuco7ies, the syn-
cytium becomes greatly thickened, and the pores are conse-
quently prolonged into canals (which may be ramified and
anastomose), connecting the ventriculus with the exterior.
The endodermic cells, which in these, as in the Asco7ies, at
first form a continuous layer, are eventually restricted to the

1 " Zar Entwickelungs-cfoschichte der Kalkschwammc." {ZoiUdiriftfur
wissenschqftliche Zoologic^ Bd. xxiv. ) F. E. Schuke, so far as I follow Hae-
ckel's account of his recent onservations ('•'■ Die Gastrula und die Eifurcbung
der Thiere,'' p. 158), agrees with Metschnikoff as to t'ne first sta'^es of develop-
ment, but differs in regard to subsequent stances. Haeckel withdraws his ear-
lier account of the formation of the gastrula by delamination, or splitting of the
walls of an oval shut^?aww/a-3ac into two layers, and the subsequent opening
of the planula at one end.

THE PORIFERA. 107

canals, or even to local dilatations of these canals — the so-
called " ciliated chambers."

The same relative disproportion of the ectoderm, with the
consequent development of passages which traverse the mass
of the sponge, and are provided at intervals with ciliated
chambers, is found in the silicious sponges, in which the
spicula, if they possess any, are formed by a deposit of silex ;
and in which, as a rule, the sponge-corpuscles are supported
by a more or less complete skeleton of a tough animal sub-
stance, termed Jceratose.

Malisarca^ however, is devoid both of skeleton and spicula,
and the minute structure of the curious boring-sponges — the
Clionm — has yet to be elucidated.

Ilaliphysema and Gastrophysema, of Haeckel, appear to
be sponges which get no further than the Gastrula condi-
tion, and thus form a connecting link between the Sponges
and the Hydrozoa,

The fresh-water sponge {^SpongilUi) has been studied with
extreme care by Lieberkiihn, and the following account,
based upon the investigations of that author, is given for the
use of the student to whom Spongilla flumalis is likely to
be the most readily accessible of the sponges.

The fresh-water sponge grows on the banks of docks,
canals, rivers, and on floating timber, in the form of thick
incrusting masses, which usually have a green color, and
require a constant supply of fresh water for their healthy
maintenance. The surface presents irregular conical emi-
nences perforated at their summit like small volcanic craters,
and from these exhalent funnels, which answer to the oscnla
of the Calcispongioe^ currents of the water are continually
flowing. Careful examination of the surface of the Spongilla
between the exhalent craters, shows that it is formed by a
delicate membranous expansion, separating which from the
deeper substance of the &pongilla are a number of irregular
cavities. In some case?, these run into one great water-
chamber. The superficial chambers (or chamber) communi-
cate w4th the exterior by pores, which perforate the mem-
branous expansion, are similar to those in the outer surface of
the ventricular wall of a simple calcareous sponge, and sub-
serve the same inhalent function. On their inner face, or
floor, the superficial chambers exhibit the apertures of in-
numerable canals, which traverse the deep substance of the
Spongilla in all directions, and, sooner or later, unite into
passages which lead directly into the cavities of the exhalent

103 THE AXATOMY OF INVERTEBRATSD AXIMALS.

craters. Dilatations of the canals occur at intervals, and are
lined by the characteristic monadiform endodermic cells,
which are restricted to the walls of these ciliated chambers.
It is by the working of the cilia of these cells that currents
of water are made continually to enter by the inhalent pores
and to pass out by the exhalent craters. The whole fabric
is supported and strengthened by a skeleton, which consists,
in the first place, of bands and filaments of keratose, and,
secondly, of silicious spicula, the majority of which resemble
needles pointed at each end, and contain a fine central canal
filled with an unsilicified substance. The individuality of
these animals is so little marked that two Sjyongillm, when
brought into contact, before long fuse into one; while they may
divide spontaneously, or be separated artificially into different
portions each of which will maintain an independent existence.

A process analogous to the formation of cysts, which is so
common among the Protozoa^ takes place in the deeper sub-
stance of the body, especially in the autumn. A number of
adjacent sponge-corpuscies, losing their granular appearance,
become filled with clear, strongly refracting granules, the nu-
cleus ceasing to be visible. The sponge-corpuscles which
surround these become closBly applied together, and secrete
coats of keratose, which fuse with those of the adjacent cor-
puscles. In the interior of each a singular silicious spiculum
is formed, consisting of two toothed disks, like cogged wheels,
united by an axis. As this " amjyhidlscus " enlarges, the proto-
plasm of the corpuscle disappears, and at length nothing is left
but the envelope of keratose, with the imbedded amphidisks,
disposed perpendicularly to its surface. At one point of the
spheroidal envelope a small opening is left, and the so-called
"seed " of the SponglUa is complete. It remains throughout
the winter unchanged; but, with the return of warmth, the
sponge-corpuscles inclosed within the coat of the " seed," or
more properly cyst, slowly escape through the pore, become
perforated with inhalent and exhalent apertures and canals, and
develop the characteristic spicula of a young Spongilla.

This process of encystment, which may be regarded as a
kind of budding, akin to propagation by bulbs among plants,
has not been observed among marine sponges.

Sexual propagation takes place in the same way as in the
Calcispongice^ and the embryo passes through morula and
planula stages. But the ciliated cells which form the outer
wall of the latter, and constitute its locomotive apparatus,
seem to vanish when the embryo fixes itself, and the body of

THE PORLFERA. 109

the young Flhrospongia appears to be developed out of the
inner cells, which, in the mean while, have become spiculiger-
ous. However, the details of the mode of development of the
Fibrospongim require further elucidation.

In both the marine and fresh-water sponges the ingestion
of solid matters — such as carmine and indigo — by the mo-
nadiform endodermic cells has been seen by several observ-
ers. According to Haeckel, the solid particles, which usually
are taken in between the flagellum and the collar, may also be
ingested at other parts of the surface of the endodermic cell.
In the course of such experiments, also, granules of the pig-
ment may be found in the ectoderm, but, whether they enter
it directly or secondarily from the endoderm, is unknown.
Sponges absorb oxygen, and give oif carbonic acid with great
ra{)iditv ; and the manner in which they render the water in
which they live impure, and injurious to other organisms, sug-
gests the elimination of nitrogenous waste matter.

The syncytium may contract as a whole, and is liable to
local contractions, as when the oscula or the pores shut or
open. The contours of the cells of which it is composed are
invisible in the fresh state, and hence it appears as a mere
"sarcode" or transparent gelatinous conlraciile substance,
in which nuclei and granules are imbedded here and there.
But Lieberkiihn has shown that, when the water in which
Spong'dla lives is heated to the point at which thermic coagu-
lation of the protoplasm of the cells occurs, their boundaries
at once become defined, and the cells commonly detach them-
selves from one another. The syncytium is therefore formed
by the -close union, and not by the actual fusion, of the cells
of the body.

It is a very interesting fact that thread-cells, similar to
those which are so abundant in the Coelenterata, are said to
occur in some sponges. Eimer ^ finds these structures in
species of the Renierincie. The thread-cells are scattered
through both endoderm and ectoderm, and abound on the
free surface of the former, where it limits the canals of the
sponge, but do not occur on the outer surface of the ectoderm.
The same observer states that he found partly digested re-
mains of small crustaceans in the ventricular cavities and
passages of both silicious and calcareous sponges.

The Porlfera present three principal modifications — the
Myxospongi(j>^ the Calcisponguje^ and the Fibrospongloe — the

1 " Ncsseizellen und Saamen bei See-Schwdmincn." (ArcJtiv far Mikro-
skovische Anatomies viii., 1872.)

110 THE ANATOMY OF IXYERTEBRATED ANIMALS.

Myxospongim being altogether devoid of skeleton ; the Cal-
cispongice possessing calcareous spicula, but no fibrous kera-
tose skeleton ; and the Fihrospongim having a fibrous skele-
ton, and (usually) spicula of a silicious nature. To these it is
probable that the Clionid(J& must be added, as a fourth type,
devoid of a fibrous skeleton, but possessing silicious spicula
of a very peculiar kind, by the help of which they are able
to burrow parasitically in the shells of mollusks. Finally,
Haliphysema and Gastrophyseina appear to be even simpler
than the Myxosponglm.

The division of the Myxospongice contains only the ge-
latinous Halisarca, The Calcispo?igiCB, in addition to the two
families of Asco7tes and Leiicones^ already referred to, include
a third — the Sycones^ which are essentially composite As-
coyies. The I^ibrospongice present a great diversity of form
and structure. They may have the form of flattened or glob-
ular masses, arborescent, tree-like growths, flagellate expan-
sions, or wide or deep cups. The sponge of commerce de-
rives its value from the fact that its richly-developed fibrous
skeleton is devoid of spicula. On the other hand, in such
sponges as Hyalonema and Euplectella^ the silicious spicula
attain a marvelous development and complexit}^ of arrange-
ment. In the latter genus, they form a fibrous network with
regular polygonal meshes. These appear to be the repre-
sentatives of the Ventriculites^ which were so common in
the seas of the Cretaceous epoch.

Sponges abound in the waters of all seas, but Spongilla
is the sole fresh-water form. Clionidm existed in the Silu-
rian epoch, but the most plentiful remains of sponges have
been yielded by the chalk.

The CoELE^'TERATA. — This group of the Metazoa contains
those animals which are commonly known as Polyps, Jelly-
fishes, or Medusce^ Sea-anemones, and Corals. They exhibit
two well-marked series of modifications, termed the Hydrozoa
and the Actmozoa,

The Hydrozoa. — The fundamental element in the^ struct-
ure of this group is the Hydranth., or Polypite. ^ This is es-
sentially a sac having at one end an ingestive or oral open-
ing, which leads into a digestive cavity. The wall of the sac
is composed of two cellular membranes, the outer of which is
termed the ectoderm^ and the inner the endoderm^ the former
having the morphological value of the epidermis of the higher

THE PORIFERA.

Ill

'"''m^m^^'"'

Fig. 12. — A. Hypothetical section of a SpongiTIa: a. superficial layer; b, fnhnlent
apertures ; c, ciliated chambers ; d, an esh^lent aperture ; e, deeper substance
of the sponge. The arrows indicate the direction of t'le currents. B. Kf>^msiWSpon-
qilla with a single exhalent aperture, seen from above (after Lieberkiihn) : a. in-
nalent apertures ; c, ciliated chambers ; d, exhalent aperture, C. A ciliated
chamber. D. A free-swimming ciliated embryo.

112 THE ANATOMY OF INVERTEBnATED ANIMALS.

animals, and the latter that of the epithelium of the aliment-
ary canal/ Between these two layers, a third layer — the

Fis. 13. — Diasrramp illustrative of the mutnal relations! of the HijrTrozon :

1. Hydra. 2. Sertularian. 3. Calvcophoridati. 4. Physuphoriclan. 5. Lucernariati.

a. Ectoderm, b. Endoderm. c. The digestive and somatic cavity.

P. Tentacles. N. Nectoialyx. T. Coeno?arc. B. Hydrophyllium. C. Hydrothoca. S.

Hydranth. Q Gonophore, A. Air- Vesicle contained in F. Pneumatophore. c,

Digective and somatic cavity.
I., II., III., I v\, represent the successive stages of development of a Medusiform

gonophore.

mesoderm — which represents the structures which lie between

* " The body of every Hydrozoon is essentially a sac composed of two mem-
branes, an external and an internal, which have been conveniently denomi-
nated by the terms ectoderm and endoderm. The cavity of the sac, which will
be called the somatic cavity ^ contains a fluid, charged with nutritive matter in

THE HYDROZOA. 113

the epidermis and the epithelium in more complex animals,
may be developed, and sometimes attains a great thickness,

solution, and sometimes, if not always, with suspended solid particles, which
perform the functions of the blood in animals of higher organization, and may
he termedihe S07natic fluid. . . . Notwithstanding the extreme variety of form
exhibited by the Hydrozoa, and the multiplicity and complexity of the organs
which some of them possess, they never lose the traces of this primitive sun-
plicity of organization ; and it is but rarely that it is even disguised to any con-
siderable extent. . . . This important and obvious structural peculiarity could
hardly escape notice, and I find it to have been observed by Trembley, Baker
and Laurent, Corda and Ecker in Hydra ; by Eathke, in Coryne ; by Frey and
Leuckart, in Lucernaria ; and it is given as a character of the hydroid po-
lyps in general {Hydrce^ CorynidcB^ and ISertularidce)^ in the second edition of
Ouvier's ' Legons.' I pointed it out as the general law of structure of the hy-
droid polyps, Biphydoi and PhysopJioridcc^ in a paper ^ sent to the Linna^an So-
ciety, from Australia, in 1847, but not read before tbat body till January, 1849 ;
and I extended the generalization to the whole of the Hydrozoa^ in a ' Memoir
on the Anatomy and Affinities of the Meduscx,^ read before the Koyal Society
in June, 1849.

" Prof. Allman, in his valuable memoir ' On Cordylophora ' ('Philosophical
Transactions,' 1855), has adopted and confirmed this moi-phological law, intro-
ducing the convenient terms ' ectoderm ' and ' endoderm,' to denote the imier
and outer membranes ; and Gegenbaur (' Beitriige zur niiheren Kenntniss der
Schwimmpolypen; 1854, p. 42) has partially noticed its exemplification in
Apolemia and Bkizophysa; but it seems singularly enough to have failed to
attract the attention of other excellent German observers, to whose late im-
portation investigations I shall so often have occasion to advert. The pecu-
liarity in the structure of the body walls of the Hydrozoa^ to which 1 have just
referred, possesses a singular interest in its bearing upon the truth (for, with
due limitation, it is a great truth) that there is a certain similarity between the
adult states of the lower animals and the embryonic conditions of those of
higher organization.

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

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

" Thus there is a very real and genuine analogy between the adult Hydro-
zoon and the embryonic vertebrate animal ; but I need hardly say it \)\ no
means justifies the' assumption that the Hydrozoa are in any sense 'arrested
developments ' of higher organisms. All thiit can justly be affirmed is, that the

» "Observations upon the Anatomy of the Diphydje and the UnUy of Orgnnizi-
tion of the DipJiydse and Physophoridfp." An abstract of this essay "was published
in the '• Proceedings of the Linnsean Society " for 1849.

U4

THE ANATOMY OF INYERTEBRATED ANIMALS.

but it is a secondary and, in the lower Sydrozoa^ inconspicu-
ous production.

All the Hydrozoa are provided Avith tentactda' These
are elongated and sometimes filiform organs of prehension,
which are generally diverticula of both ectoderm and endo-
derm, but may be outgrowths of only one of them.

Thread-cells, or nematocysts, are very generally distributed
through the tissues of the (Joelenterata. In its most perfect
form, a nematocyst is an elastic, thick-walled sac, coiled up in
the interior of which is a long filament, often serrated or pro-
vided with spines. The filament is hollow, and is continuous
with the wall of the sac at its thicker or basal end, while its
other pointed end is free. Very slight pressure causes the

Fig. 14— Sacculus of a tent-icle with nematocvsta of Athorybia: J., peduncle or
stalk, and 5, involucrum of the sacculus C; A filaments; </, ectoderm; e, eado-
derm ; /, neinatocysts; 1, small nematocysts of the filaments and involucrum;
2, 3, larger nematocysts of the sac ; 4, largest nematocyats.

thread to be swiftly protruded, apparently by a process of
evagination, and the nematocyst now appears as an empty

Hydrozoon travels for a certain distance aloasr the same ?reat highway of de-
velopment as the hiofher animal, before it turns off to follow the road which
leads to its special destination."

In this passage of my work on the " Oceanic Hydrozoa " (1859), I expanded
the idea enunciated in the memoir on the Medusfe here referred to, that " the
outer and inner membranes appear to bear the same physiological relation to
one another as do the serous and mucous layers of the germ." The diagram
(Fig. 13), exhibiting the relations of the different groups of the Hydrozoa, was
published in the Medical Times and Gazeiti in June, 1856.

THE HYDROZOA. 115

sac, to one end of which a long filament, often provided with
two or three spines near its base, is attached. Many of the
Coelenterata^ and notably the Fhysalla, give rise to violent
urtication when their tentacles come in contact with, the hu-
man skin, whence it may be concluded that the nematocysts
produce a like injurious effect upon the bodies of those ani-
mals which are seized and swallowed by the Polyps and Jelly-
fishes.

As regards the existence of a nervous system in the Hy-
drozoa^ very diverse opinions have been entertained, and it
may be doubted if the problem has even yet received its final
solution, I have already discussed Kleinenberg's suggestion
that the branched prolongations of the inner ends of the cells
of the ectoderm in Hyclra^ which end in the longitudinal fibres
which lie between the ectoderm and the endoderm, may be
nerves in their earliest stage of differentiation. Haeckel de-
scribes a nervous system in Glossocodon and Carmarina. It
consists of a circular band which lies on the inner side of the
circular canal of the bell-shaped swimming-organ of these
Medusae^ and presents a ganglionic enlargement at the base of
each of the lithocysts. Of these eight ganglia, the four which
correspond to the openings of the four radial canals into the
circular canal are the larger. Each of these gives off four
branches, one of which follows the course of the radial canal
to the central polypite or manubrium ; two others go to the
adjacent tentacles, and the last to the lithocyst.^

There can be little doubt that the lithocysts, or sacs con-
taining mineral particles, which are so frequently found in the
Medusas, are of the nature of auditory organs ; while the mass-
es of pigment, -with imbedded refracting bodies, which often
occur associated with the lithocysts, are doubtless rudimentary
ej'es.

The sexual reproductive elements are ova and spermato-
zoa— the ova being very often devoid of a vitelline membrane.
The fully-formed generative elements lie between the ecto-
derm and the endoderm of that part of the body -wall in which
they make their appearance. In Hydractinia, as has already
been pointed out, the ova appear to be modified cells of the
endoderm, and spermatozoa modified cells of the ectoderm;

I Haeckel, "Beitrasre zur Naturgesclnchte der Hydromednsen." The ana-
tomical disposition of this nervous apparatus accords very well with the recent
important observations of Mr. Romanes on the " Locomotor System of Medu-
sae." (" Proceedings of the Eoyal Society," December, 1S75.)

116 THE ANATOMY OF IXTERTEBRATED ANIMALS.

but it remains to be seen how far this rule is of general appli-
cation.

Usually the region of the body in which the generative
organs are produced undergoes a special modification before
the reproductive elements make their appearance in it, giving
rise to a peculiar organ, the gonophore. In its simplest con-
dition the gonophore is a mere sac-like diverticulum, or out-
ward process of the body-wall. But, from this state, the
gonophore presents every degree of complication, until it ac-
quires the form of a bell-shaped body called from its resem-
blance to a Medusa or jelly-hsh a mediisoid.^

In its most complete form, the medusoid consists of a disk
having the form of a shallow or deep cup [nectoclyx), from the
centre of the concavity of which projects a sac termed the nin-
nuhrimn. The cavity of the sac is continued into that of
sundry symmetrically disposed canals, most commonly four in
number, which radiate from the centre of the disk to its cir-
cumference, where they open into a circular marginal canaL
A membranous fold, the veliun^ which contains muscular fibres
arranged concentrically to its free margin, is attached to
the inner circumference of the mouth of the bell, and pro-
jects, like a shelf, into its interior. Lithocysts are usually
developed on the margins of the bell, wd]ich may also give
rise to tentacles. The manubrium, opening at its free end,
may become functionally, as well as structurally, a hydranth,
and may serve to feed the medusoid when it is detached from
the hydrosomn, or body of the hydrozoon. However com-
plex its structure may be, the medusoid commences as a sim-
ple bud-like outgrowth, which thickens at its free end ; the
central part of this thickening becomes the manubrium,
"while its periphery, splitting away from the manubrium, is
converted into the disk (Fig. 13). A single prolongation of
the somatic cavity is continued into the manubrium, w^hile
several, usually four, symmetrically arranged diverticula ex-
tend into the nectocalyx and become its radiating canals.
The distal ends of these subsequently throw out lateral
branches, which unite and give rise to the circular canal.

The lithocysts are usually, but not always, free and promi-

1 From the imperfection of our knowledcre respectino^ the origin of many
of the medusifonn Hydrozoa^ it is ditilcult to employ any terminology with
strict consistency. If "medusoid" is restricted to what are known to be
gonophorcs developed by gemmation, " medusa '' may be employed, in a gen-
eral sense, as the equivalent of the somewhat inconvenient vernacular term
"jelly-fish."

THE HYDROZOA. 117

nent, and the one or many solid mineral bodies which they
contain are inclosed in special envelopes. Their structure
appears to be more complicated in the Geryonidce than in
other Medusae. (Haeckel, loc. cit.)

In some of those medusoid gonophores, the reproductive
elements are developed while the gonophore is still attached
to the hydrosoma, and then they always make their appear-
ance in the wall of the manubrium. But, in other cases, the
medusoid becomes detached before the development of the
reproductive elements, and, feeding itself, increases largely
in size before the ova or spermatozoa appear. Sooner or
later, however, the reproductive organs are developed, either
in the walls of the manubrial hydranth, or in those of the
canals of the nectocalyx of the medusoid.

In an early stage of its existence, every hydrozoon is
represented by a single hydranth, but, in the great majority
of the Hydrozoa^ new hydranths are developed from that
first formed, by a process of gemmation or of fission. In
the former case the bud is almost always an outgrowth or
diverticulum of the ectoderm and endoderm, into which a
prolongation of the cavity of the body extends. Sometimes
the hydranth formed by gemmation becomes detached from
the body ; but, in many cases, the buds developed from the
primary hydranth remain connected together by a common
stem or coenosarc^ and thus give rise to a compound body, or
k}/d)'Osoma.

In many Hydrozoa^ the ectoderm gives rise to a hard cu-
ticular coating, and in some of these (Gamjjanidarida?^ Ser-
tularidce^ Fig. 13, 2), this cuticular investment, on the hy-
dranth, takes the shape of a case or " cell " — the hydrotheca
— into which the hydranth may be more or less completely
retracted. In other Hydrozoa, protective coverings are af-
forded to the hydranths by the development of processes of
the body-wall, which become thick, variously-shaped, glassy
lainella3. These appendages are termed hydrophylUa (Fig.

Again, certain groups (the CcdycojjhoridcB and most Phy-
sojjhoridce) are provided with bell-shaped organs of propul-
sion, produced by the metamorphosis of lateral buds of the
hydrosoma. These nectocalyces ha.ve the structure of a med-
usoid, devoid of a manubrium. In others {Physophoridce),
one extremity of the hydrosoma is dilated, contains air in-
closed within a sac formed by an involution of the ectoderm,
and constitutes a float or piieumatophore ; while in yet others

118 THE ANATOMY OF INVERTEBRATED ANIMALS.

(Discophora) the aboral end of the hydranth is dilated into
a disk or umbrella^ which is susceptible of rhythmical con-
tractile movements, by which the body is propelled through
the water. Thus, notwithstanding its different mode of de-
velopment, it has a close resemblance to a medusoid. Ac-
cording to the existence or absence of these various append-
ages, and the manner in which they are disposed, the Hy-
drozoa are distinguishable into three groups — 1, the Sydro-
phora ; 2, the Discophora ; 3, the Siphonophora,

1. The Hydeophoka are, in all cases but that of Hydra^
fixed ramified hydrosomes, on which many hydranths and
gonophores are developed. The somatic cavity contained in
the hydrosoma always retains a free communication with the
gastric cavities of the hydranths. In other words, it is an
enterocoele. The tentacula are either scattered over the hy-
dranths (Cory7ie), or are arranged in one circle round the
mouth [Sertularia) ; or in two circles, one close to the mouth,
and one near the aboral end { Tubular ia). Very generally —
e. g., in all Sertularidcp, GamiJanidaridm and Tuhidaridce —
there is a hard, chitinous, cuticular skeleton (perisarc of All-
man), which frequently gives rise to hydrothec^e, into which
the hydranths can be retracted (Fig. 13, 2).

The gonophores present every variety, from simple sac-
cular diverticula of the hydrosoma to free-swimming medu-
soids. The inner margin of the bell in these medusoids is
always produced into a velum, and otolithic sacs and eye-
spots are very generally disposed at regular intervals around
the circumference of the bell. The great majority of what
were formerly termed the naked-eyed MedusiB {Gymnoph-
thalmata) are merely the free-swimming gonophores of the
Hydrophora. Thus the medusoids known as Sarsiadce are
the free gonophores of the Corynidm ; the JSougainmllem
and Lizzice of the Eudendridoe ; many Oceanidm proceed
from Tubular id(B ; Thaumaiitidce and ^quoridcB from (7am-
panularidce.

In some Hydrophora (e. g., Calycella) the margins of the
hydrotheca are prolonged into triangular processes, which
serve as an operculum.

Certain PlumularidcB are provided with prominences of
the hydrosoma surrounded by a chitinous investment, which
is open at the extremity. The inclosed soft ectoderm usual-
ly contains many thread-cells, and has the power of throw-
ing out contractile pseudopodial processes. These have been

THE HYDROPHORA.

119

termed nematophores by Mr. Busk.* In Ophiodes (Hincks)
they are tentaculiform.

It frequently happens that the gonophores are developed
upon special stalks, each of which has essentially the struct-

PiG. 15.— CVzmpawwZcria (after Gesccbaur).— -4, Hydranth : e, its peduncle; «', hy-
drotlieca ; 0, mouih; <e, tentacles; ^', dii^estive cavity, continuous with the bo-
matic cavity k, contained in the peduncle and in the creeping stem, S. B, gonan-
giura containing two medusiform zc5ids or gonophores it' / the somatic cavit;

t" is in connection with that of the creeping stem. C, Bud.

ure of a mouthless hydranth. This is termed a blastostyle.
In some blastostyles (Fig. 15), during the development of the
buds of the gonophores, the ectoderm splits into two layers —
an inner, which invests the central axis formed by the endoderm
with the contained prolongation of the somatic cavity ; and
an outer, chiefly, if not wholly, chitinous layer. Into the in-
terspace between these two, the budding gonophores project,
and may emerge from the summit of the gonaiigiwn., thus
formed, either to develop the reproductive elements, and shed
them while still attached, or to be set at liberty as free medu-
soids (Fig. 16).

Allman * has shown that, in Dicoryne conferta^ the gono-

_ 1 They are described under the name of" clavate ororans," and compared
with the tentacles oi Diphydoe in rav memoir on the "Affinities of the Medu-
sae." ("Philosophical Transactions," 1849.)

2 'I Monograph of the Gymnoblastic, or Tubularian Hydroids," 1871, p. 31.
In this beautifully illustrated and elaborate work, the student will find, not

120 THE ANITOMY OF INVERTEBRATED ANIMALS.

phore contained in a gonangiuni, somewhat like that of Ziao-
medea, is set free as a ciliated bitentaculate body, on the cen-
tral axis of which the ova and spermatozoa are developed.

Fig. 16 — Medusiform zoilid of Ca7npam«?ana (after Gegenbaur) : J, nectocalyx ; te^
tentacles ; a. lithocysts; J.', velum ; k\ manubrium, inclosing the digestive cavity ;
o, mouth ; k'\ radial canals.

In the genus Aglaophenia {PlumularidcB)^ groups of
gonangia are inclosed in a common receptacle {corhula,
Allman), formed by the development and union of lateral
processes (comparable in some respects to the hydrophyllia
of the Cahjcophoridoe) from that region of the hydrosoma
which bears the gonophores.

Some medusoids, such as Sarsia prolifera and WiUsia, the
hydroid stages of which are not at present certainly known,
but which are probably coryniform, produce medusoids simi-
lar to themselves by budding. The buds may be developed
either from the manubrium, or from the maro-inal canal of the
nectocalyx, or from the bases of the tentacula, or even from
their whole length.

In August, 1849, while in the North Pacific, off the Loui-
siade Archipelago, 1 took a species of W^illsia (Fig. 17), in
which stolons were developed at the bifurcation of each of
the four principal radiating canals of the nectocalyx. Each
stolon was terminated by a knobbed extremity containing
many nematoc3^sts ( C, g), and gave rise, on one side, to a
series of buds, of which those nearest the free end of the
stolon had acquired the form of complete medusoids. They
had four uabranched radiating canals and four tentacles ; but
it is probable that they would assume the form of the parent
stock after detachment.

only a full account of the ori?anizatlon of the grouo of which it treats, but
much inforuiation respecting the Hijdrozoa in general.

THE DISCOPHORA.

121

In striking contrast with the complexity of these repro-
ductive processes, the gonophore is represented, in Hydra^

■fi^

Fig. n.—WiUsia, sp. ; A, the mednsa, with buddins stolons. B. one of the buds
developed on a stolon ; h, radial canal of the nectocalvx: e, manubrium. C, a
stolon: g, its free end beset with nematocysts ; b, c, d, budding medusoids ;
/, medusoid nearly ready to be detached; «, its manubrium; d, its nectocalyx
h, a radial canal.

by a mere enlargement of the body-wall, situated close to the
bases of the tentacula, in the case of the testes, and nearer
the attached end of the body in that of the ovary. The ovary
develops a single ovum, which, as Kleinenberg has shown,
undergoes division and invests itself with a chitinous coat
while still attached to the body of the parent. This chiti-
nous investment is more or less spinose, and is often con-
founded with an egg-shell. It obviously answers to the
perisarc of a Tubularian, and its presence in the embryo of
the Hydra., in which no perisarc is developed by the adult,
suggests that Sydra may not represent the simplest primary
condition of a Hydrophoran, but may be a reduced modifica-
tion of a Tubularian.

2. The Discophora. — These " Medusa? " resemble the
more perfect free medusoid gonophores of the Ilydrophora^
in so far as they consist of a hydranth or polypi te attached
to the centre of a gelatinous contractile swimming disk. But
they differ from the medusoids of the Hydrophora, inasmuch
as they are developed either directly from the impregnated
ovum ; or by gemmation from a Medusa which arises in this
6

123

THE ANATOMY OF INVERTEBEATED ANIMALS.

way ; or by the transverse fission of the hydriform product
of the development of the impregnated ovum.

In some of these (e. g., Carniarina, Polyxenia, ^gmopsis^
Trachynema) ^ the disk is similar to the nectocalyx of one of
the medusoicls of the Sydropliora y and, like it, is provided
with a velum. But in the rest (Lucer)iarici^ and the Stega-
nophtJialmata) the disk is either devoid of a velum, or pos-
sesses only a rudiment of that structure, and is termed an
umbrella. The edges of the umbrella are divided into lobes
by marginal notches in which the lithocysts are lodged.
Moreover, in these, the mineral particles of the lithocysts are
numerous, and not inclosed in seperate sacs. The lithocysts
are often covered by hood-like processes of the umbrella,
whence they have been termed " covered-ej^ed " or Stega-
noplithalmata,

Lucernaria is fixed by the aboral side of its umbrella
(Fig. 13, 5), by means of a longer or shorter peduncle. The
umbrella is divided into eight lobes, at the extremities of
each of which there is a group of short tentacles. The

Fig. 18.— I. Aurelia aurita: i, the prolonged angles of the mouth ; G, genital cham-
bers : m. lithocysts. x r ^v «„,i;„f5,,^

II. Under view of a segment of the disk, to show the arrangement of the radiatmg
canals: the aperture of a genital chamber and the plaited genital niembrane
showing through its ventral wall; and a lithocyst with its protective hooa (in).

hydranth stands up in the centre of the umbrella, and its
cavity communicates with a central chamber, whence four
wide chambers pass into the lobes. These chambers are
separated by septa, the free central edges of which are beset
with slender tentacles. The reproductive organs are double

THE DISCOPHORA. 123

radiating series of thickenings of the oral wall of each cham-
ber/

All the other Dlscophora^ which are what are commonly
known as " Jelly-fish," are free, and some attain a very large
size. In the adult (Fig. 18) the umbrella is thick and divided
by small marginal notches into as many (usually eight) lobes.
At the bottom of each notch, often protected by special lob-
ules, is an oval lithocyst, supported by a cylindrical pedun-
cle, the cavity of which is in direct communication with
one of the radiating canals of the umbrella (Fig. 28, IV.).
This canal communicates with the exterior on the aboral side
of the base of the peduncle.'' The thick mesoderm of which
the great mass of the umbrella consists is composed of a ge-
latinous connective tissue, in the meshes of which is a watery
fluid, containing numerous nucleated cells which exhibit amoe-
boid movements. On the oral face there is a broad zone of
striped muscle, made up of fusiform fibres placed side by
side. In Aurelia aurita, the angles of the four-sided hy-
dranth are produced into long foliaceous lips, the margins
of which are beset with minute solid tentacula (Fig. 18).
The gastric cavity contained in the hydranths terminates, be-
neath the centre of the umbrella, in a somatic cavity which
passes into four radially-disposed, wide offshoots, or genital
sinuses, the oral walls of which constitute the roof of the gen-
ital chambers (Fig. 18, II.). From their margins the narrow
branching radial canals are given off. The peripheral ends
of these unite when they reach the margin.

Each genital chamber is a recess, surrounded by a thick
wall of the oral face of the umbrella, in the centre of which
only a small aperture is left (Fig. 18, I., G). The roof of
this cavity is the floor of the genital sinus ; it is much plaited
and folded, and the genital elements are developed in it. Its
inner or endodermal wall is beset with small tentacular fila-

* The relations of Lucernaria with the Discophora were shown in my lect-
ures, Medical Times and Gazette^ 1856, Keferstein, " Untersuehungen liber
niedere Seethiere" (1862), in his monograph on the genus, fully confirms
this view, aad Prof H. J. Clark arrived independently at the same conclusion :
'-'- Lucernaria the Coenotype of the AcalephcB^'' ("Proceedings of the Boston
Society of Natural History," 1862). The Lucernaria {CarduelU, Allman)
cijathiforrnis of Sars diifers mucn from the ordinary Lucernariae^ especially in
the position of the genital organs as lonGritudinal thickeninsrs in the walls of
the gastric cavity. See Allman, " On the Structure of Carduella cyathiformu^^
("Transactions of the Microscopical Society," viii.).

''The circular canal of the nectocalyx 'communicates with the exterior by
apertures on the summits of papillose elevations in some medusoids.

124

THE ANATOMY OF INVERTEBRATED ANIMALS.

ments (Fig. 28, III.). The ova or the spermatozoa pass out
of the apertures of the genital chambers, and the ova are re-

Fig. 19. — Cephea ocdlata (?).— The entire animal : a, the umbrella ; &, the ramifications
of the brachia ; c, the tentacles which terminate them; o. the pillars which sns-
pend the bracliiferouB disk which forms the floor of the sub-umbrellar cavity ; I,
short ciavate tentacles between the oral pores.

ceived into small pouches or folds of the lips, and there under-
go the preliminary stages of their development.

In the Mhizostomidce (as was originally suggested by Von
Baer and has been proved by L. Agassiz and A. Brandt * )
the margins of the lips of the hydranth unite, leaving only a
multitude of small apertures for the ingestion of food on the
long arms, which represent prolongations of the lips of the
hydranth (Figs. 19, 20, 21). The polystomatous condition
thus brought about, by the subdivision of a primitively sim-
ple oral cavity, is obviously quite different in its nature from
that which occurs in the JPorifera.

In most of the Mhizosto'inidce, not only do the edges of
the lips unite, but the opposite walls of the hydranth beneath
the umbrella are, as it were, pushed in, so as to form four

» "Memoires de PAcademie do St.-Petersbourg," xvi., 1870.

THE RHIZOSTOMID^.

125

chambers, the walls of which unite, become perforated, and
thus give rise to a sub-umbrellar cavity with a roof formed

Fig. %Q.—Cephea ocellata (?).— ^, part of the umbrella, viewed from below, to show
the plaited genital membrane (/) and the divided attachment of one of the pillars ;
d^ place of one of the liihocysts. B^ one of the oral pores (m) surrounded by ten-
tacula (n) ; g, one of the clavate tentacles interspersed between the oral pores. C,
one of the pedunculated lithocysts {i) in its notch {d) seen from below, with the
oval plate from which muscular fibres (h) take their origin ; e, the radiating canal
with its caecal lateral branches, g.

by the umbrella and a floor, the hrachiferous disJc^ suspended
by four pillars. In the roof the plaited genital membranes

A i

Fig. 2\.— Cephea ocellata (?).— ^, lithocyst enlarged with its hood (A;) and the aboral
pore of the canal {c) ; d, the notch of the margin of the umbrella. B, the hrachifer-
ous disk with the origins of the arms ; /, eudoderm ; o, ectoderm. C, tentaculate
lip of an oral pore enlarged ; m, oral cavity; n, nematocysts.

are developed. The floor (Fig. 21, B) gives off the subdivided
arms, the free margins of which bear the oral pores, and

126 THE ANATOMY OF INVERTEBRATED ANIMALS.

Fig. 22.— a, Biphyes appendiculata.—a, hj-dranths and hydropliyllia on the hydrosoma ;
6, proximal nectocalyx ; c, aperture of distal nectocalyx; rf, somatocyst; c, pro-
longation of the distal nectocalyx, by which it is attached to the hy<lro8oma;
/, point of attacbment of the hydrosoma in the cavity, or hydrcecium, of the proxi-
mal nectocalyx. B, the distal nectocalyx with the canal (through which the bris-
tle a is passed), which is traversed by the hydrosoma in A. C\ extremity of the
distal nectocalyx, with its muscular velum.

which are traversed by canals which unite, pass through the
pillars, and open into the central cavity of the umbrella.*

Fig 23.— >4, 5, DiphyzoOid (Sphe7ioides),\SLtGT^\ and front views. ^, DiphvzoCid of
Abyla {Cuboides). cr, e. gonoijhore or reproductive oriran ; b, hvdranth; c, phyl-
locyst or cavity of hydrophyllium, with its process {d). D, free gonophore, its
manubrium (a) containing ova.

* The species of Cephea., the anatomy of which is here given, was obtained
in the South Pacific, near the LouLsiade Archipelago, on the 11th of July, 1849.

THE SIPHONOPHORA.

127

3. The Siphonophora. — In this group the hydrosoma is
always free and flexible, the ectoderm developing no hard
chitinous exoskeleton, save in the case of the pneumatophores
of some species. In most, the hydranths are of equal size ;
but in Velella and JPorpita, the hydranth situated in the
centre of the discoidal body is very much larger than the
rest, which occupy a circumferential zone around it ; and the

Fig. 2i.~A(horybia rosacea.— A, lateral view ; B. from above; C, 2), detached hydro-
phyllia ; a, polypites ; 6, tentacles ; c, sacculi of the tentacles ; d, hydropbyilia ;
/, pneumatophoie.

principal function of which is to develop the gonophores
from their pedicles. In these two genera the tentacula are
separate from the hydranths, and form the outermost circle
of appendages.

The hydranths of the Siphonophora (Fig. 25, A) never
possess a circlet of tentacula round the mouth, which, when
expanded, is trumpet-shaperl. The endoderm of the hydranth
is ciliated, and villus-like prominences project into its cavity.

The aboral surface of the umbrella was of a brownish-srrav color, variesrated
with oval white spots ; the oral surface, liorht brown with eisrht bluish-green
lines radiating toward the lithocvsts ; the brachia, gray with brown dpts.^ The
brachia divide into two at their origin, and then subdivide into an infinity of
small branches. The general color of the smaller branches is light brown, the
small interspersed clavate tentacles being white. The long tentacles which
terminate each brachium are blue and cylindrical at their origin, but become
trigonal farther on, where they are shaded with brown and green. Is it identi-
calwith the Cephea ocellata of Peron and Lesueur? The individual figured
was a young male.

128 THE ANATOMY OF INVERTEBRATED ANIMALS.

The interior of these frequently contains vacuolar spaces
(Fig. 24, B^ C). A valvular "pylorus" separates the gastric
from the somatic cavity in the Calycophoridm. Long tenta-
cles, frequently provided with unilateral series of branches,
are developed, either one from the base of each hydranth, or,
independently of the hydranths, from the coenosarc.

In the Calycophoridm and many Fhysophoridce, complex

Tig. 25.— AfhoryMa rosacea.— A. a hydranth with villi (a). B, one of the rilli in its
elongated state, enlargeii, C, a small retracted villus, still more magnified, with
its vacuolar epaces and ciliated surface.

organs, containing a sort of battery of thread-cells, terminate
each lateral branch of a tentacle (Figs. 24 and 26). Each
consists of an elongated saccidvs, terminated by two fila-
mentous appendages, and capable of being spirally coiled up.
In this state it is invested by an involucrum, which surrounds
its base. Tlie somatic cavity is continued through the branch,
v/hich constitutes the peduncle of this organ, into the saccu-
lus and its terminal filaments. In the latter it is narrow, and
their thick walls contain numerous small spherical nemato-
cysts. In the sacculus the cavity is wider. One wall is very
thick, and multitudes of elongated nematocysts, the lateral
series of which are sometimes larger than the rest, are dis-
posed parallel with one another, and perpendicular to the
surface of the sac. Like the other organs, each of these
tentacular appendages commences as a simple diverticulum
of the ectoderm and endoderm, and passes through the stages
represented in Fig. 26.

In Physalia the tentacula may be several feet long. They
have no lateral branches, but the large nematocysts are situ-

THE SIPflONOPHORA.

129

ated in transverse reniform thickenings of the wall of the ten-
tacle, which occur at regular intervals.

FiQ. 26. — Athorybia rosacea.— The ends of the tentacnlar branches in varioas stages
of developmeut. A, lateral branch, commeucin^ as a bud from the tentacle. In
B, terminal papillaa, the rudiments of ihe filaments, are developed at the extremi-
ty of the branch ; and, in (7, the sacculus is beginning to be marked off, and thread-
cells have appeared in its walls ; in Z>, the division into involucrum and sacculus
is apparent; in £", the involucrum has inve!*ted the sacculus, the extremity of
which is straight, while the lateral processes have curled round it.

Hydrophyllia are generally present, and, like the tentacu-
la, are developed either from the pedicle of a hydranth, in
which case they inclose the hydranth with its tentacle and a
group of gonophores (GalycopJioridoe)^ or, independently of
the hydranths, from the coenosarc (many Physophoridce).

The hydrophyllia are transparent, and often present very
beautifully defined forms, so that they resemble pieces of cut
glass. They are composed chiefly of the ectoderm (and meso-
derm), but contain a prolongation of the endoderm, with a
corresponding diverticulum of the somatic cavity. They are,
in fact, developed as ccecal processes of the endoderm and
ectoderm ; but the latter, with the mesodermal layer, rapidly
predominates.

The gonophores of the Slphonophora present every varie-
ty, from a simple form, in which the medusoid remains in a
state of incomplete development, to free medusoids of the
Gymnophthalmatous type. As an example of the former

130

THE ANATOMY OF INVERTEBRATED ANIMALS.

condition the gonophores of Atkoryhia may be cited (Fig*.
27) ; of the latter, the gonophores of Physalla^ Porpita^ and
Velella.

In Athoryhia, groups of gonophores, together with pyri-
form sacs, which resemble incompletely develoj^ed hydranths
(hydrocysts — Fig. 27, -4, a), are borne upon a common stem,
and constitute a gonohlastidium, (Fig. 27, A), The groups
of male and female gonophores (Fig. 27, A^ h^ c) are borne
upon separate branches of the gonohlastidium (androphores

Fig. Vl.—Athoryhia rosacea.— A, gonoblastidium bearincr three hydrocysts, a ; gyno-
phore, b ; and two androphores, c. B, female gonophores on their common i<tem
or gynophore, showinir the inchidod ovum, a, and the radical canals, b. O, X>,
female gonophores enlarged ; a, germinal vehicle ; b, vifellua; c, radial canals of
the imperfect nectocalyx ; d, canals ol the manubrial cavity. E, male gouophore,

and gyjiojihores). Each female gonophore contains only a
single ovum, which projects into the cavity of the imperfectly

THE STPHOXOPHORA. 131

differentiated manubrium, and narrowing its cavity at differ-
ent points gives rise to the irregular canals (Fig. 27, D^ d).
In the male gonophore the nectocal3'^x is more distinct from
the manubrium, and its extremity has a rounded aperture
(Fig. 27, E).

In the Calycoplioridoe^ as in the elongated Physophoridoe,^
the development of new hydranths and their appendages,
which is constantly occurring, takes place at that end of the
hydrosoma which corresponds to the fixed extremity of one
of the Hydrophora / and, if we consider this to be the proxi-
mal end, new buds are developed on the proximal side of
those already formed. Moreover, these buds are formed on
one side only of the hydrosoma. Hence the appendages are
strictly unilateral, though they may change their position so
as eventually to appear bilateral or even whorled. In the
Calycopjhoridct^^ the saccular proximal end of the coenosarc
(Fig. 22, A^ d) is inclosed within the anterior nectocalyx, at
the posterior end of which is a chamber, the hydroechctn
(Fig. ^'i^ A, c). The second, or posterior, nectocalyx is at-
tached in such a way that its anterior end is inclosed within
the hydroecium of the anterior nectocalyx, while its contrac-
tile chamber lies on the opposite side of the axis to that on
which the anterior nectocalyx is placed (Fig. 22, A). Sets
of appendages (Fig. 22, A, a ; Fig. 23), each consisting of a
hydrophyllium, a hydranth with its tentacle, and gonophores,
which last bud out from the pedicle of the hydranth — are
developed at regular intervals on the coenosarc, and the long
chain trails behind as the animal swims with a darting mo-
tion, caused by the simultaneous rhythmical contraction of
its nectocalyces, through the water (Fig. 22).

From what has been said, it follows that the distal set of
appendages is the oldest, and, as they attain their full de-
velopment, each set becomes detached, as a free-swimming,
complex Dlphyzooid (Fig. 23). In this condition they grow
and alter their form and size so much, that they were for-
merly regarded as distinct genera of what were termed mono-
gastric Diphydce. The gonophores, with which these are
provided, in their turn become detached, increase in size,
become modified in form, and are set free as a third series
of independent zooids (Fig. 23, D), But their manubrium
does not develop a mouth and become a functional hydranth ;
on the contrary, the generative elements are developed in
its wall, and are set free by its dehiscence.

In the Physophoridce, ithe proximal end of the hydrosoma

133 THE ANATOMY OF INVERTEBRATED ANIMALS.

is provided with a pneumatophore. This is a dilatation, into
which the ectoderm is invaginated, so as to form a receptacle,
which becomes filled with air and sometimes has a terminal
opening, through which the air can be expelled (Fig. 13, 4).
Tt is sometimes small, relatively to the hydrosoma (Agalma,
Physojyhora) ; sometimes so large {Athorybia, Fig. 24 ; Phy-
salia, Forpita, Velella), that the whole hydrosoma becomes
the investment of the pyriform or discoidal air-sac ; while the
latter is sometimes converted into a sort of harl inner shell,
its cavity being subdivided by septa into numerous chambers
(Porpita, Velella).

Nectocalyces may be present or absent in the Fhysojyho-
ridce. When present, their number varies, but they are con-
fined to the region of the hydrosoma which lies nearest to the
pneumatophore.

In the great majority of the Hydrozoa, the ovum under-
goes cleavage and conversion into a morula, and subsequently
into a planula, possessing a central cavity inclosed in a double
cellular wall, the inner layer of which constitutes the hypo-
blast, and the outer the epiblast.

In most S'jdrophora the ciliated, looomodve, planula be-
com3s elongated and fixed by its aboral pole. At the oppo-
site end, the mouth appears and the embryo passes into the
gastrula stage. Tentacles next bud out round the mouth,
and to this larval condition, common to all the Sydrophora,
AUman has sriven the name of Actinida.

Generally, the embryo fixes itself by its aboral extremity
at the end of tlie planula stage ; but, in certain Tubularidce^
while the embryo is still free, a circlet of tentacles is devel-
oped close to the aboral end ; and this form of larva differs
but very slightly from that w^iich is observed in the Disco-
'pJiora.

In the genus Pelagia, for example, the tentacles are de-
veloped from the circumference of the embryo, midway be-
tween the oral and aboral poles ; but it neither fixes itself
nor elongates into the ordinary actinula-form. On the con-
trary, it remains a free-swimming organism, and, by degrees,
' that moiety of the body which lies on the aboral side of the
tentacular circlet widens and is converted into the umbrella,
the other moiety becoming the hydranth, or " stomach," of
the Medusa.

In Lucernaria^ it is probable that the larva fixes itself be-
fore or during the development of the umbrella, and passes

THE DEVELOPMENT OF THE HYDROZOA. 133

directly into the adult condition. But, in most Discopliora^
tlie embryo becomes a fixed actinula (the so-called Hydra
tuba OT jScyj^histoma, Fig. 28, 1.), multiplies agamogenetically
by budding, and gives rise to permanent colonies of Hydri-
form polyps. At certain seasons of the year, some of these
enlarge and undergo a farther agamogenetic multiplication
by fission (Fig. 28, II.). In fact, each divides transversely
into a number of eight-iobed discoidal medusoids i^'' Ephyr(B "
or '■' 3Ieduscje bijidce,''^ Fig. 28, II. and III.), and thus passes
into what has been termed the Strohlla stage. The Ephyrm^
becoming detached from one another and from the stalk of
the Strobila, are set free, and, undergoing a great increase
in size, take on the form of the adult Discophore, and acquire
reproductive organs. The base of the Strobila may develop
tentacles (Fig. 28, II.) and resume the Scyphlstoma condition.

Metschnikofi"^ has recently traced out the development of
Geryonia ( Carmarina)^ Polyxenia^^glnopsis^ and other i>?5-
cophora^ which diflfer from the foregoing in possessing a velum ;
and in these, as in the Trachynema ciliatum^ observed by
Gegenbaur,' the process appears to be of essentially the same
nature as in Pelagia, The Scyi^histoma of Aurelia, Cyanoea^
and their allies, is probably to be regarded, like the larva of
Pelagia^ as a Discophore with a rudimentary disk ; in which
case the reproduction of the Ephyra-ioiWi^ of young Disco-
phora will not be comparable to the development of medusoid
gonophores among the Hydrophora^ but will merely be a pro-
cess of multiplication, by transverse fission, of a true, though
undeveloped, Discophore.

In the Siphonoijhora^ the result of yelk division Is the
formation of a ciliated body consisting of a small-celled
ectoderm investing a solid mass of large blastomeres, which
eventually pass into the cells of the endoderm. This body
does not take the form of an actinula. On the contrary, it
appears to be the rule that buds from w^hich a hydroph3dlium,
a nectocalyx, a tentacle, or pneumatophore, or even all of
them, will be developed, take their origin antecedently to the
formation of the first polypite and of the gastric cavity.

As Metschnikoff well remarks, the m.ode of development
of the Siplionopliora is wholly inconsistent with the doctrine
that the various appendages of the hydrosoma in these ani-

^ " Studien uber die Entwickelung der Medusen und Siplionoplioren."
{Zeitschrift fur wiss. Zool., xxiv.)

a " Zur Lehre der Generationsvrechsel." 1854.

3 See especially tlie late observations of AletschnikofF, he. cit.

134 THE ANATOMY OF INVERTEBEATED ANIMALS.

li liiKij i i li^if iiniil I I i\ \

' '. ' '.\ : ' ' ■ ! ' ? ?•• • ! !: ■ :'■ ;: ; •' •

ii!\iiiiin iinUniHlnlin J

riiiill nniirin \\\\\ I
iiiliUililiihUy lii \ il

i UU \iUn i \ liu uM \ I

Fro. 28.— I. and II.— Cyanoea capillata (after Van Beneden^).

I. Two Hydrce tubce (Scyphistoma stage), exhibiting their ordinary character?, and
between them two (a, b) which are undergoing fission (Sii^obila ?tage).

n. The two StrobUm, a and ft, three day? later. In a. tentacles are developed be-
neath the lowest of the Ephyrae, from the stalk of the Strobila, which will persist
as a Hydra tuba.

III. Hair the disk of an Ephyra of Aurelia aurita, seen from the oral face. The
Bmall tentacles which lie between the mouth and the band of circular muscular
fibres are inside the somatic cavity, whence sixteen short and wide radial canala
extend to the periphery, where they are united by transverse branches. Eight
of the radial canals enter the corresponding lobes, and finally divide into three
branches: one which enters the peduncle of the lithocyst. and two lateral caeca.
Radiating bands of muscular fibres accompany these canals.

rV". Side view of one of the lithocysts with its peduncle. The arrow indicates the
direction in which the cilia of the exterior work.

' "Eecherches sur la Faune littorale de Belgique. Polypes." 1866.

THE DEVELOPMENT OF THE HYDROZOA. I35

mals represent individuals. The Hydrozoa are not properly-
compound organisms, if this phrase implies a coalescence of
separate individualities ; but they are organisms, the organs
of which tend more or less completely to become independent
existences or zooids. A medusoid, though it feeds and main-
tains itself, is, in a morphological sense, simply the detached
independent generative organ of the hydrosoma on which it
was developed ; and what is termed the " alternation of gen-
erations," in these and like cases, is the result of the dissocia-
tion of those parts of the organism on which the generative
function devolves, from the rest/

In certain Z>/5ca;9^ora belonging to the group of Trachy-
Jiemata, a method of multiplication by gemmation has been
observed, which is unknown among the other Hydrozoa. It
mav be termed entogastrio gemmation, the bud growing out
from the wall of the gastric cavit}^, into which it eventually
passes on its way outward ; while, in all other cases, gemma-
tion takes place by the formation of a diverticulum of the
whole wall of the gastro-vascular cavity which projects on to
the free surface of the body, and is detached thence (if it be-
come detached), at once, into the circumjacent water. The de-
tails of this process of entogastric gemmation have been traced
by Haeckel ^ in Gannirlna hastata, one of the Geryonidm,
As in other members of that family, a conical process of the
mesoderm, covered by the endoderm, projects from the roof
of the gastric cavity and hangs freely down into its interior.
Upon the surface of this, minute elevations of -g-Jifth of an
inch in diameter make their appearance. The cells of w^hich
these outgrowths are composed next become differentiated
into two layers — an external clear and transparent layer,
which is in contact with the cone, and invests the sides of the
elevation ; and an inner darker mass. The external layer is the
ectoderm of the young medusoid, the inner its endoderm. A
cavity, which is the commencement of the gastric cavity, ap-
pears in the endodermal mass, and opens outward on the free
side of the bud. The latter, now ^tq-*^ ^^ ^^ ^"^^^^ ^" diameter,
has assumed the form of a plano-convex disk, fixed by its flat
side to the cone, and having the oral aperture in the centre of
its convex free side. The disk next increasing in height, the

» I have seen no reason to depart from the opinions on the subject of
'Animal individuality' enunciated in my lecture published in the Annals and
Magazine of Natural History for June, 1852.

2 " Beitrage zur Naturgeschi elite der Hy drome dusen," 1865.

136 THE ANATOMY OF IXVERTEBRATED ANIMALS.

body acquires the form of a flask with a wide neck. The belly
of the flask is the commencement of the umbrella of the bud-
ding medusoid ; the neck is its gastric division. The belly of
the flask, in fact, continues to widen out until it has the form
of a flat cup, from the centre of which the relatively small
gastric neck projects, and the bud is converted into an unmis-
takable medusoid, attached to the cone by the centre of the
aboral face of its umbrella. In the mean while, the gelatinous
transparent mesoderm has appeared, and, in the umbrella, has
acquired a great relative thickness. Into this, eight prolonga-
tions of the gastric cavity extend, and give rise to the radial
canals, which become united into a circular canal at the cir-
cumference of the disk. The velum, tentacula, and lithocysts
are developed, and the bud becomes detached as a free swim-
ming medusoid. But this medusoid is very different from the
Garmarina from which it has budded. For example, it has
eight radial canals, while the Cannarina has only six ; it has
solid tentacles, while the adult Cannarina has tubular tenta-
cles ; it has no gastric cone, and has differently disposed lith-
ocysts. Haeckel, in fact, identifies it with Cunina rhodo-
dactyla, a form which had hitherto been considered to be not
only specifically and generically different from Carmarina^
but to be a member of a distinct family — that of the uS^ginidm,

What makes this process of asexual multiplication more
remarkable is, that it takes place in Carmarince which have
already attained sexual maturity, and in males as well as in
females.

There is reason to believe that a similar process of ento-
gastric proliferation occurs in several other species of uSiJgi'
nidm — ^gineta prolifera (Gegenbaur), Eury stoma ruhigi-
nosum (KoUiker), and Cunina Kollikeri (F. Miiller) ; but,
in all these cases, the medusoids which result from the gem-
mative process closely resemble the stock from which they
are produced.

As might be expected, the Hydrozoa are extremely rare
in the fossil state, and probably the last animal the discovery
of fossil remains of which could be anticipated is a jelly-fish.
Nevertheless, some impressions of Medusae, in the Solenhofen
slates, are sufficiently well preserved to allow of their deter-
mination as members of the group of Rhizostomidm^ The

* Haeckel, " Ueher zwei neiie fossile Meclnsen aus der Fainilie der BH-
zostomiden." (" Jahrbuch fiir Mineralogie," 1866.)

THE ACTINOZOA. 137

apparent absence of the remains of Hydrophora in the meso-
zoic and newer palaeozoic rocks is very remarkable. Some
singular organisms, termed Graptolites^ which abound in the
Silurian rocks, may possibly be Hydrozoa, though they
present points of resemblance with the Polyzoa. They are
simple or branched stems, sometimes slender, sometimes ex-
panded or foliaceous ; occasionally the branches are connected
at their origin by a membranous expansion. The stems are
tubular, and beset on one or both sides with minute cup-
shaped prolongations, like the thecse of a Sertularian. A solid
thickening of the skeleton may have the appearance of an
independent axis. Allman has suggested that the theciform
projections of the Graptolite stem may correspond with the
mematophores of Sertularians, and that the branches may
have been terminated by hydranths. Appendages which ap-
pear to be analogous to the gonophores of the Hydrophora
have been described in some Graptolites.^

With a very few exceptions {Hydra, CordylopJiora) the
Hydrozoa are marine animals ; and a considerable number,
like the CalycoplioridcQ and Physoplioridae, are entireh' pe-
lagic in their habits.

The Actixozoa. — The essential distinctions between the
Actinozoa and the Hydrozoa are two. In the first place, the
oral aperture of an Actinozoon leads into a sac, which, with-
out prejudice to the question of its exact function, may be
termed " gastric," and which is not, like the hydranth of the
Hydrozoon, free and projecting, but is sunk within the body.
From the walls of the latter it is separated by a cavity, the
sides of which are divided- by partitions, the tnesenteries,
which radiate from the wall of the gastric sac to that of the
body, and divide the somatic cavity into a corresponding num-
ber of interinesenteric chambers. As the gastric sac is open
at its inner end, however, its cavity is in free communication
with that of the central space which communicates with the
intermesenteric chambers ; and the central space, together
with the chambers, which are often collectively termed the
*' body cavity " or " perivisceral cavity," are, in reality, one
with the digestive cavity, and, as in the Hydrozoa, consti-
stute an enter ocode. Thus an Actinozoon might be com-
pared to a Lucernarla, or still better to a Carduella, in which
the outer face of the hydranth is united with the inner face

1 Hall, " Graptolltes of the Quebec Series of North America," I860. Nichol-
son, " Monograph of the British Graptolitidiis," 1872.

138 THE ANATOMY OF INYERTEBRATED ANIMALS.

of the umbrella ; under these circumstances the canals of the
umbrella in the Hydrozoon would answer to the intermesen-
teric chambers in the Actinozoon.

Secondly, in the Actmozoa^ the reproductive elements
are developed in the walls of the chambers or canals of the en-
terocoele, just as they so commonly are in the walls of the
gastro-vascular canals of the Hjdrozoa^ but the generative
organs thus constituted do not project outwardly, nor dis-
charge their contents directly outward. On the contrary, the
ova and spermatozoa are shed into the enterocoele, and event-
ually make their way out by the mouth. In this respect,
again, the Actinozoon is comparable to a Lucernaria modi-
fied by the union of the hydranth with the ventral face of the
umbrella ; under which circumstances the reproductive ele-
ments, which in all Hydrozoa are developed, either in the
walls of the hydranth or in those of the oral face of the um-
brella, would be precluded from making their exit by any
other route than through the gastro-vascular canals and the
mouth.

In the fundam3ntal composition of the body of an ecto-
derm and endoderm, with a more or less largely developed
mesoderm, and in the abundance of thread-ceils, the Actino-
zoa agree with the Hydrozoa.

In most of the Actinozoa, the simple polyp, into which
the embryo is converted, gives rise by budding to many
zooids which form a coherent whole, termed by Lacaze-Du-
thiers a zoanthodeme.

The C0RA.LLIGEXA. — The Actinozoa comprehend two
groups — the Goralligena and the Ctenophora — which are
widely different in appearance though fund amen tall}^ similar
in structure. In the former, the mouth is always surrounded
by one or more circlets of tentacles, which may be slender
and conical, or short, broad, and fimbriated. The mouth is
usually elongated in one direction, and, at the extremities of
the long diameter, presents folds which are continued into
the gastric cavity. The arrangement of the parts of the body
is therefore not so completely radiate as it appears to be.
The enterocoele is divided into six, eight, or more wide inter-
mesenteric chambers, which communicate with the cavities of
the tentacles, and sometimes directly with the exterior, by
apertures in the parietes of the body. The mesenteries which
separate these wide chambers are thin and membranous. Two
of them, at opposite ends of a transverse diameter of the Ac-

THE CORALLIGENA.

139

tinozo(3n, are often different from the rest. Each mesentery
ends, at its aboral extremity, in a free edge, often provided

Fig. 29.— Perpendicular section of Actinia holsatica (after Frey and Leuckart).— a,
mouth; 6, j^astric cavity; c, common cavity, iuto wliich the gastric cavity and
the intermesenteric chambers open; c?, intermesenteric chambers; e, thickened
free margin, containing thread-cells of, /, a mesentery; g^ reproductive organ ; A,
tentacle.

with a thickened and folded margin ; and these free edges
look toward the centre of an axial cavity,^ into which the gas-
tric sac and all the intermesenteric chambers open.

In the Coralligena., the outer wall of the body is not pro-
vided with bands of large paddle-like cilia. Most of them are
fixed temporarily or permanently, and many give rise by
gemmation to turf-like, or arborescent, zoanthodemes. The
great majority possess a hard skeleton, composed principally
of carbonate of lime, which may be deposited in permanently
disconnected spicula in the walls of the body ; or the spicula
may run into one another, and form solid networks, or dense
plates, of calcareous matter. When the latter is the case, the
calcareous deposit may invade the base and lateral walls of
the body of the Actinozoon, thus giving rise to a simple cup,
or theca. The skeleton thus formed, freed of its soft parts, is
a " cup-coral," and receives the name of a corallite.

In a zoanthodeme, the various polyps (anthozooids)
formed by gemmation may be distinct, or their several enter-
ocoeles may communicate ; in which last case, the common
connecting mass of the body, or ccenosarc, may be traversed
by a regular S3'stem of canals. And, when such compound

» Partially-digested substances are often found in this axial space, and it is
not improbable that it may functionally represent the stomach or the com-
mencement of the intestine in higher animals.

140 THE ANATOMY OF INVEKTEBRATED ANIMALS.

Actinozoa develop skeletons, tLe corallites may be distinct,
and connected only by a substance formed by the calcifica-
tion of the coenosarc, which is termed coenenchyma j or the
thecas may be imperfectly developed, and the septa of adja-
cent corallites run into one another. There are cases, again,
in which the calcareous deposit in the several polyps of a
compound Actinozoon, and in the superficial parts of the coe-
nenchyma, remains loose and spicular, wiiile the axial por-
tion of the coenosarc is converted into a dense chitincus cr cal-
cified mass — the so-called sclerohase.

The mesoderm contains abundantly developed muscular
fibres. The question whether the Coralligena possess a ner-
vous system and organs of sense, hardly admits of a definite
answer at present. It is only in the Actinidce that the ex-
istence of such organs has been asserted ; and the nervous
circlet of Actinia^ described by Spix, has been seen by no
later investigator, and may be safely assumed to be non-exist-
ent. Prof. P. M. Duncan, F. R. S.,^ however, has recently
described a nervous apparatus, consisting of fusiform gan-
glionic cells, united by nerve-fibres, which resemble the sym-
pathetic nerve-fibrils of the Yertehrata^ and form a plexus,
which appears to extend throughout the pedal disk, and
very probably into other parts of the body. In some of the
ActmidcB (e. g.-, Actinia mesembryantliemwni)^ brightly-col-
ored bead-like bodies are situated in the oral disk outside
the tentacles. The structure of these "chromatcphores," or
" bourses calicinales," has been carefully investigated by
Schneider and Rotteken, and by Prof. Duncan. They are
diverticula of the body wall, the surface of which is com-
posed of close-set " bacilli," beneath w^iich lies a layer of
strongly-refracting spherules, followed by another layer of
no less strongly-refracting cones. Subjacent to these, Prof.
Duncan finds ganglion cells and nervous plexuses. It would
seem, therefore, that these bodies are rudimentary eyes.

The sexes are united or distinct, and the ovum is ordina-
rily, if not always, provided with a vitelline membrane. The
impregnated ovum gives rise to a ciliated morula, which may
either be discharged or undergo further development within
the somatic cavity of the parent. The morula becomes a gas-
trula, but whether by true invagination or by delamination,
as in most of the Ilydrozoa^ is not quite clear. The gastrula
usually fixes itself by its closed end, while tentacles are de-

» " On the Nervous System of Actinia." (" Proceedings of the Eoyal Socie-
ty," October 9, 1873.)

THE DEVELOPMENT OF THE CORALLIGENA, 141

veloped from its oral end. It can hardly be doubted that the
intermesenterio chambers are diverticula of the primitive en-
terocoele ; but the exact mode of their origin needs further
elucidation.

Lacaze-Duthiers ^ has recently thrown a new light upon
the development of the Coralligena^ and particularly of the
Acthiim {Actinia, Sagartia, Bimodes). These animals are
generally hermaphrodite, testes and ovaria being usually found
in the same animal, and even in the same mesenteries ; but
it may happen that the organs of one or the other sex are, at
any given time, exclusively developed. The ova undergo the
early stages of their development within the body of the
parent. The process of yelk division was not observed, and
in the earliest condition described the embryo was an oval
planula-like body, composed of an inner colored substance
and an outer colorless layer. The outer layer (epiblast = ec-
toderm) soon becomes ciliated. An oval depression appears
at one end, and becomes the mouth ^ and gastric sac, while, at
the opposite extremity, the cilia elongate into a tuft. The
ectoderm extends into and lines the gastric sac, while the in-
terior of the colored hypoblast becomes excavated by a cav-
ity, the enterocoele, which communicates with the gastric sac.
In this condition the embryo swims about with its oral pole
directed backward.

The oral aperture changes its form and becomes elongated
in one direction, which may be termed the oral axis. The
mesenteries are paired processes of the transparent outer
layer (probably of that part which constitutes the mesoderm)
which mark off corresponding segments of the enterocoele.
The first which make their appearance are directed nearly at
right angles to the oral axis near, but not exactly in, the
centre of its length. Hence they divide the enterocoele into
two primitive chambers, a smaller (A) at one end of the oral
axis, and a larger (A') at the other. This condition may be
represented by A-4-A'; the dots indicating the position of
the primitive mesenteries, and the hyphen that of the oral
axis. It is interesting to remark that, in this state, the em-

1 " Developpement, des CoralUaircs." {ArcMves de Zoologie experim^ntale^
1872.)^

2 Kowalewsky describes the formation of a ffastrula by invagination in a spe-
cies of Actinia and in Cereanthus^ the aperture of invacfination becoming the
mouth (llofmann and Schwalbe, " Jahresbericht," Bd. II., p. 269). In other
species of Actinia and in Alcijonium^ the planuhx seems to delamiuate. Ordi-
nary yelk division occurs in some Anthozoa^ while in others {Alcyonium) the
process rather resembles that which occurs in most Arthropods.

142 THE ANATOMY OF INYERTEBRATED ANIMALS.

bryo is a bilaterally symmetrical cylindrical body, with a cen-
tral canal, the future gastric sac ; and, communicating there-
with, a bilobed enterocoele, which separates the central canal
from the body- wall. In fact, in principle, it resembles the
early condition of the embryo of a Ctenophore, a Brachiopod,
or a Sag lit a.

Another pair of mesenteric processes now makes its ap-
pearance in the larger chamber A', and cuts off two lateral
chambers, B, B, which lie between these secondary mesenteries
and the primary ones. In this state the enterocoele or somat-
ic cavity is four-chambered f A-f-r? AM. Next a third pair

of mesenteries appear in the smaller chamber (A), and divide
it into three portions, one at the end of the oral axis (A),
and two lateral (C, C). In this stage there are therefore six

A p-^-T) A' ) ; but almost immediately the number

is increased to eight, by the development of a fourth pair of
mesenteries in the chambers B, B, which thus give rise to the
chambers D, D, between the primitive mesenteries and them-
selves. The embryo remains in the eight-chambered condition

(a p/-T-yA -D AM for some time, until all the chambers and their

dividing mesenteries become equal. Then a fifth and a sixth
pair of mesenteries are formed in the chambers C, C, and D, D ;
two pairs of new chambers, E and F, are produced, and thus the

Actinia acquires twelve chambers (A p p^^T^ t) "r -^ /> ^^^

of which result from the subdivision of the smaller primary
chamber, and seven from that of the larger primary chamber.
The various chambers now acquire equal dimensions, and the
tentacles begin to bud out from each. The appearance of
the tentacles, however, is not simultaneous. That which pro-
ceeds from the chamber A' is earliest to appear, and for some
time is largest, and, at first, eight of the tentacles are larger
than the other four.

The coiled marginal ends of the mesenteries appear at
first upon the edges of the two primary mesenteries ; then
upon the edge of the fourth pair, and afterward upon those
of the other pairs.

For tb.e further chan2:es of the young Actioiia, I must
refer to the work cited. Sufficient has been said to show that
the development of the ActinicB follows a law of bilateral
symmetry, and to bring out the important fact that, in the

THE OCTOCORALLA. 143

course of its development, the finally hexamerous Antho-
zoon passes through a tetramerous and an octomerous stage.

Phenomena analogous to the " alternation of generations,"
which is so common among the Hydrozoa^ are unknown
among the great majority of the Actinozoa. But Semper ^
has recently described a process of agamogenesis in two spe-
cies of Fungice, which he ranks under this head. The Fungim
bud out from a branched stem, and then become detached
and free, as is the habit of the genus. To make the parallel
with the production of a medusoid from a hydroid polyp
complete, however, the stem should be nourished by a sexless
anthozooid of a different character from the forms of Fimgice
which are produced by gemmation. And this does not appear
to be the case.

In one division of the Coralligena — the Octocoralla —
eight enterocoele chambers are developed, and as many ten-
tacles. Moreover, these tentacles are relatively broad, flat-
tened, and serrated at the edges, or even pinnatifid. The
Actinozoun developed from the e^g wi^^y remain simple
(Haimea^ Milne-Edwards), but usually gives rise to a zoan-
thodeme.

The coenosarc of the zoanthodeme in the Octocoralla is a
substance of fleshy consistence, which is formed chiefly of a
peculiar kind of connective tissue, containing many muscular
fibres developed in the thickened mesoderm. The axial cavity
of each anthozooid is in communication with a system of
large canals. In Alcyonixtm^ a single large canal descends
from each anthozooid into the interior of the zoanthodeme,
and the eight mesenteries are continued as so man}' ridges
throughout its entire length,'' so that these tubes have been
compared to the thecal canals of the Millepores. In the red
coral of commerce ( Corallium ruhruin^ Fig. 30), the large
canals run parallel with the axial skeleton. A delicate net-
work, which traverses the rest of the substance of the coeno-
sarc, appears to be sometimes solid and sometimes to form a
system of fine canals opening into the larger ones. The
anthozooids possess numerous muscles by which their move-
ments are effected. The fibres are delicate, pale, and not
striated. Nerves have not been certainly made out.

It is in these Octocoralla that the form of skeleton which
is termed a sclerobase, which is formed by cornification or

1 " Ueber Generations-Wechsel bei Steinkorallen." Leipsic, 1872.

2 Pouchet and Myevre, " Contribution a I'Auatomie des Alcyonaires."
{Journal d' Anatomie et de la Physiologie^ 1870.)

144 THE ANATOMY OF INVERTEBRATED ANIMALS.

Fig. 30. — Corallium rubrum (after Lacaze-Duthiers ^).

I. The end of a branch with A, B, C, three anthozoSids in different desrees of ex-
pansion ; k, the mouth ; a, that part of the coenosarc which rises into a cup
around the base of each anthozoOid.

II. Portion of a branch, the coenosarc of which has been divided longitudinallj' and
partially removed; B, B\ B'\ anthozooids in section; B^ anthozoOid with ex-
panded tentacles; k, mouth ; m, gastric sac ; e, its inferior edge; j^ mesenteries.

J?', anthozoOid retracted, with the tentacles {d) drawn back into the intermesenteric
chambers; c, oriftcea of the cavities of the invaginated tentacles ; e, circum-oral
cavity ; 6, the part of the body which forms the projecting tube when the antho-
zoOid is expanded ; a, festooned edges of the cup.

B" ^ antliozoOid, showing the transverse sections of the mesenteries.

-4, A, coenosarc, with its deep longitudinal canals (/'), and superficial, irregular,
reticulated canals (A). P, the hard axis of the coral, with longitudinal grooves
fj7) answering to the longitudinal vessels.

III., IV. Free ciliated embryos.

1 " Ilistoire Naturelle du Corail," 1864.

THE ACTiNOZOA. 145

calcification of the axial connective tissue of the zoantho-
deme, occurs. It is an unattached simple rod in Pennatula
and Veretilluin^ but fixed, tree-like, branched, and even retic-
ulated, in the Gorgonim and the red coral of commerce ( Co-
rallium). In the Alci/o?iia, or " Dead-men's-fingers," of our
own shores, there is no sclerobase, nor is there any in Tuhi-
pova^ the organ-coral. But, whereas in all the other Octoco-
ralla the bodies of the polyps and the coenosarc are beset with
loose spicula of carbonate of lime, Tuhipora is provided with
solid tubiform thecse, in which, however, there are no septa.

Dimorphism has been observed by Kolliker to occur exten-
sively among the Pennatulidce. Each zoanthodeme presents
at least two different sets of zooids, some being fully devel-
oped, and provided with sexual organs, while the others have
neither tentacles nor generative organs, and exhibit some
other peculiarities.* These abortive zooids are either scat-
tered irregularly among the others (e. g., S>arcophyton^ Fere-
tilliwi)^ or may occupy a definite position (e. g., Virgularia),

In the other chief division of the Coralligena — the Hexa-
eoralla — the fundamental number of enterocoele chambers and
of tentacles is six," and the tentacles are, as a rule, rounded
and conical, or filiform.

The Actinozoon developed from the ^gg in some of the
Hexacoralla remains simple, and attains a considerable size.
Of these — the Actinidce — many are to some extent locomo-
tive, and some {Minyas) float freely by the help of their
contractile pedal region. The most remarkable form of this
group is the genus Cereanthiis, which has two circlets, each
composed of numerous tentacles, one immediately around the
oral aperture, the other at the margin of the disk. The foot
is elongated, subcorneal, and generally presents a pore at its
apex. Of the diametral folds of the oral aperture, one pair is
much longer than the other, and is produced as far as the
pedal pore. The larva is curiously like a young hydrozoon
with four tentacles, and, at one time, possesses four mesen-
teries.

The ZoanthidcB differ from the Actmidce in little more
than their multiplication by buds, which remain adherent,
either by a common connecting expansion or by stolons ; and
in the possession of a rudimentary, spicular skeleton. In the
Antlpathidm there is a sclerobasic skeleton. The proper

1 " Abhandlungen der Senkenberffischen naturforschenden Gesellscliaft,"
isd. vu., vui.

2 That is to say, in the adult, they are either six or some multiple of six.

7

146 THE ANATOMY OF INVERTEBRATED ANIMALS.

stone-corals are essentially Actinice, which become converted
into zoanthodemes by gemmation or fission, and develop a
continuous skeleton.

The skeletal parts ^ of all the Acti7iozoa, consist either of
a substance of a horny character ; or of an organic basis im-
pregnated with earthy salts (chiefly of lime and magnesia),
but which can be isolated by the action of dilute acids ; or,
finally, of calcareous salts in an almost crystalline state, form-
ing rods or corpuscles, which, when treated with acids, leave
only an inappreciable and structureless film of organic matter.
The hard parts of all the Aporosa, Perforata^ and Tcibulata
of Milne-Edwards are in the last-mentioned condition ; while,
in the Octocoralla^ except Tuhipora, and in the Antipathidc^^
andZoa^z^A/f^tf, among the Hexacoralla, the skeleton is either
horny ; or consists, at any rate, to begin with, of definitely
formed spicula, which contain an organic basis, and frequently
present a laminated structure. In the organ-coral [Tuhi2:)ora),
the skeleton has the character of that of the ordinary stone-
corals, except that it is perforated by numerous minute canals.

The skeleton appears, in all cases, to be deposited within
the mesoderm, and in the intercellular substance of that layer
of the body. Even the definitely shaped spicula of the Octo-
coralla seem not to result from the metamorphosis of cells.
In the simple aporose corals the calcification of the base and
side walls of the body gives rise to the cup or theca / from
the base the calcification extends upward in lamelhe, which
correspond with the interspaces between the mesenteries, and
gives rise to as many vertical septa^ the spaces between which
are termed loculi ; while, in the centre, either by union of the
septa or independently, a column, the columella^ grows up.
Small separate pillars between the columella and the septa are
termed p)aluli. From the sides of adjacent septa scattered
processes of calcified substance, or synapticnlw, may grow
out toward one another, as in the Fimgidm ; or the interrup-
tion of the cavities of the loculi may be more complete in
consequence of the formation of shelves stretching from sep-
tum to septum, but lying at different heights in adjacent
loculi. These are interseptal dissephnents. Finally, in the
Tahidata, horizontal plates, which stretch completely across
tlie cavity of the thcca, are formed one above the other and
constitute tahidar dissejnments.

1 See Kolliker, " Icones ITistolosricoe," 1860.

2 Lacaze-Duthiers's investifxations on Astr(xa cali/cularis i>TO\'e that the septa
begin to be fonnecl before the theca.

THE "TABULATA." 147

In the Aj^orosa the theca and septa are almost invariably
imperforate; but, in the Perforata^ they present apertures,
and, in some Madrepores, the whole skeleton is reduced to
a mere network of dense calcareous substance. When the
Hexacoralla multiply by gemmation or fission, and thus give
rise to compound massive or arborescent aggregations, each
newly-formed coral polyp develops a skeleton of its own, which
is either confluent with that of the others, or is united with
them by calcification of the connecting substance of the com-
mon body. Tliis intermediate skeletal layer is then termed
coenenchyma.

The septa in the adult Hexacoralla are often very numer-
ous and of different lengths, some approaching the centre
more closely than others do. Those of the same lengths are
members of one " cycle ; " and the cycles are numbered ac-
cording to the lengths of the septa, the longest being counted
as the first. In the young, six equal septa constitute the first
cycle. As the coral grows, another cycle of six septa arises
by the development of a new septum between each pair of
the first cycle ; and then a third cycle of twelve septa^ di-
vides the previously existing twelve interseptal chambers into
twenty-four. If we mark the septa of the first cycle A, those
of the second B, and those of the third C, then the space be-
tween any two septa (A A) of the first cycle will be thus rep-
resented when the third cycle is formed — A C B C A.

When additional septa are developed, the fourth and fol-
lowing cycles do not consist of more than twelve septa each ;
hence the septa of each new cycle appear in twelve of the
previously existing interseptal spaces, and not in all of them;
and the order of their appearance follows a definite law, which
has been worked out by Milne-Edwards and Haime. Thus,
the septa of the fourth cycle of twelve (d) bisect the inter-
septal space A C ; and those of the fifth cycle (e) the inter-
septal space B C ; the septa of the sixth cycle (f), A d and
d A ; those of thes eventh cycle (g), e B and B e ; those of the
eighth cycle (h), d C and C d; and those of the ninth cycle
(i), C e and e C.

Hence, after the formation of nine cycles, the septa added
between every pair of primary septa (A, A) will be thus ar-
ranged— A fdhCiegBgeiChdf A.^

The stone-corals ordinaril}'^ known as Mlllepores are char-

1 That the order of oceurreTice of the septa of various lensfths, at the differ-
ent stagfes of growth of a corallite, is that indicated, seems to be clear, whatever
may be the exact mode of development of the septa in each cycle.

148 THE ANATOMY OF INVERTEBRATED ANIMALS.

acterized by being traversed by numerous tubular cavities,
which open at the surface, and the deeper parts of which are
divided by numerous close-set transverse partitions, or tabular
cUsse2nments, while vertical septa are rudimentary or alto-
gether absent. These were regarded as Anthozoa, and
classed together in the division of Tabidata^ until the elder
Agassiz * published his observations on the living Millepora
alcicornis, which led him to the conclusion that the Tabulata
are Hydrozoa allied to Hydractinia, and that the extinct Mu-
gosa were probably of the same nature.

The evidence adduced by Agassiz, however, was insuffi-
cient to prove his conclusions ; and the subsequent discovery
by Verrill that another tabulate coral, Pocillopora, is a true
Hexacorallan, while Moseley ^ has proved that Heliopora
coeridea is an Octocorallan, gave further justification to those
who hesitated to accept Agassiz's views.

The recent very thorough and careful investigation of a
species of Millepora occurring at Tahiti,^ by Mr. Moseley,
although it still leaves us in ignorance of one important
point, namely, the characters of the reproductive organs, yet
permits no doubt that Millepora is a true Hydrozoon allied to
Hydractinia^ as Agassiz maintained. The surface of the
living Millepora presents short, broad hydranths, the mouth
of which is surrounded by four short tentacles. Around each
of these alimentary zooids is disposed a zone of from five to
twenty or more, much longer, mouthless zooids, over the bod-
ies of which numerous short tentacles are scattered. Each
of these zooids expands at its base into a dilatation, whence
tubular processes proceed, which ramify and anastomose, giv-
ing rise to a thin expanded hydrosoma. The calcareous mat-
ter (composed as usual of carbonate, with a small proportion
of phosphate of lime) forms a dense continuous crust upon
the ectoderm of the ramifications of the hydrosoma, that part
of it which underlies the dilatations of the zooids constituting
the septa. As the first formed hydrosomal expansion is com-
pleted, another is formed on its outer surface, and it dies.
The " thecal " canals of the coral arise from the correspond-
ence in position of the dilatations of the zooids of successive
hydrosomal layers, and the tabulte are their supporting plates.

Thus the grou|) of the Tabulata ceases to exist, and its

» " Natural History of the United States," vols. iii. and iv., 1860-'62.
2 Moseley, " The Structure and Relations of the Alcyonaruxn, Heliopora,
carulea^'' etc. (" Proceedinijs of the Roval Society," November, 1875.)
8 " Proceedin^rs of the Pvoval Societv," 1876.

THE REEF-BUILDIXG CORALS. 149

members must be grouped either with the Scxacoralla^ the
Octocoralla, or the Hydrozoa.

The Hugosa constitute a group of extinct and mainl^r
Palaeozoic stone-corals, the thecae of which are provided with
tabular dissepiments, and generally have the septa less de-
veloped than those of the ordinary stone-corals. The arrange-
ment of the parts of the adult Hugosa in fours, and the
bilateral symmetry which they sometimes exhibit, are inter-
esting peculiarities when taken in connection with the te-
tramerous and asymmetrical states of the embryonic Hexaco-
ralla. On the other hand, some of the Hugosa possess oper-
cula, which are comparable to the skeletal appendages of the
Alcyonarian Prlmnoa observed by LindstrOm, and the te-
tramerous arrangement of their parts suggests affinity with
the Octocoralla. It seems not improbable tiiat these ancient
corals represent an intercalary type between the Hexccoralla
and the Octocoralla.

All the Actbiozoa are marine animals. The Actinice,
among the Hexacoralla^ and various forms of Gctocoralla^
have an exceedingly wide distribution, while the latter are
found at very great depths.

The stone-corals, again, have a wide range, both as respects
depth and temperature, but they are most abundant in hot
seas, and many are confined to such regions. Some of these
stone-corals are solitary in habit, while others are social, grow-
ing together in great fields, and forming what are called
" coral reefs." The latter are restricted within that compara-
tively narow zone of the earth's surface which lies between
the isotherms of 60°, or, in other words, they do not extend
for more than about 30° on either side of the equator. It is
not conditions of temperature alone, however, which limit
their distribution ; for, within this zone, the reef-builders are
not found alive at a greater depth than from fifteen to twenty
fathoms, while at the equator, an average temperature of 68°
is not reached within a depth of 100 fathoms.

Not only heat, then, but light, and probably rapid and
effectual aeration, are essential conditions for the activity of
the reef-building Actinozoa. But, even within the coral zone,
the distribution of the reef-builders appears to be singularly
capricious. None are found on the west coast of Africa, very
few on the east coast of South America, none on the west
coast of North America ; while in the Indian Ocean, the Pa-
cific, and the Caribbean Sea, they cover thousands of square

150 THE ANATOMY OF IXVERTEBRATED ANIMALS.

miles. It is by no means certain, however, that any one
species of West India reef-coral is identical with any East
Indian species, and the corals of the central Paciiic differ very
considerably from those of the Indian Ocean.

Different species of corals exhibit great differences as to
the rapidity of their growth, and the depth at which they
flourish best ; and no one must be taken as evidence for anoth-
er in these respects. Certain species of Perforata {Madre-
poridce and Foritidce) appear to be at once the fastest grow-
ers, and those which delight in the shallowest waters. The
AstrceidcB among the Aporosa, and Seriatopora among the
Tahulata^ live at greater depths, and are probably slower of
increase.

Under the peculiar conditions of existence winch have
just been described, it would seem easy enough to compre-
hend, a priori^ the necessary arrangement of coral-reefs. As
the reef-building Actinozoa cannot live at greater depths than »
twenty fathoms, or thereabouts, it is clear that no reef can
be originally formed at a greater depth below the surface, and
such a depih usually implies no very great distance from land.
Furthermore, we should expect that the growth of the coral
would fill up all the space between the shore and this farthest
limit of its growth ; so that the shores of coral seas would
be fringed by a sort of flat terrace of coral, covered, at most,
by a very few feet of water ; that this terrace would extend
out until the shelving land upon which it had grown descended
to a depth of some twenty fathoms ; and that then it would
suddenly end in a steep wall, the summit and upper parts of
which would be crowned with overhann^inof ledges of livina:
coral, while its base would be hidden by a talus of dead
fragments, torn off and accumulated by the waves. Such a
"fringing reef" as this, in fact, surrounds the island of
Mauritius. The beach here does not gradually shelve down
into the depths of the sea, but passes into a flat, irregular
bank, covered by a few feet of water, and terminating at a
greater or less distance from the shore in a ridge, over which
the sea constantly breaks, and the seaward face of which
slopes at once sheer down into fifteen or twenty fathoms of
water.

The structure of a frino-ingr reef varies at different dis-
tances from the land, and at different depths in its seaward
face. The edge beaten by the surf is composed of living
masses of Porites, and of the coral-like plant, the Nulllpore ;
deeper than this is a zone of Aporosa (Astrceidce), and of

FRINGING REEFS.— ATOLLS. 151

Millepores {Seriato2^ora) ; while, deeper still, all living coral
ceases ; the lead bringing up either dead branches, or show-
ing the existence of a flat, gently-sloping floor, the true sea-
bottom, covered with fine coral sand and mud. Passing from
the edge of the reef landward, the Poritldm cease, and are
replaced by a ridge of agglomerated dead branches and sand,
coated with Nullipore ; the floor of th3 shallow basin, or
" lagoon," inclosed between the reef and the land, is formed
by a conglomerate, composed of fragments of coral cemented
by mud ; and, on this, Meandrmce and FungicB rest and
flourish, exhibiting the most gaudy coloration, and sometimes
attaining a great size. During storms, masses of coral are
hurled on to the floor of the lagoon, and there gradually in-
crease the accumulation of rocky conglomerate ; but in no
other way can a frin-jino: reef, which has once attained its
limit in depth, increase in size, unless, indeed, the talus ac-
cumulatins: at the foot of its outer wall should ever rise sufti-
ciently high to aff'ord a footing for the corals within their pre-
scribed limits of depth.

Such is the structure of a fringing reef ; but the great
majority of reefs in the Pacific are very different in their
character. Along the northeastern coasts of New Holland,
for instance, a vast aggregation of reefs lies at a distance
from the shore which varies from a hundred to ten miles ;
forming a mighty wall or barrier against the waves of the
Pacific. At a few hundred yards outside this " barrier reef "
no bottom can be obtained with a sounding-line of a thousand
fathoms ; between the reef and the mainland, on the con-
trary, the sea is hardly ever more than thirty fathoms deep.
Many of the islands of the Pacific, again, are encircled with
reefs corresponding exactly in their character with the barrier
reef ; separated, that is, by a relatively shallow channel from
the land, but facing the sea with an almost perpendicular wall
which rises from a very great depth.

Finally, in many cases, especially among the single reefs,
which taken together constitute the great Australian barrier,
there is no trace of any central island ; but a circular reef,
usually having an opening on its leeward side, stands out in
the midst of the sea. These reefs, apparently unconnected
with other land, are what are called " Atolls."

How have these barrier reefs, encircling reefs, and atolls,
been formed ? It is certain that the fabricators of these reefs
cannot live at a greater depth than in the fringing reefs.
How can they have grown up, then, from a thousand fathoms

152 THE ANATOMY OF INYERTEBRATED ANIMALS.

or more ? Why do they take so generally tbe circular form ?
What is the connection, finally, between fringing reefs and
atolls ? The only thoroughly satisfactory answer to these
questions has been given by Mr. Darwin, from whose beauti-
ful work on " Coral Reefs " I have borrowed most of the fore-
going details. Consider for a moment what would be the
effect of a slow and gradual submergence of the island of
Mauritius — a submergence, perhaps, of a few feet in a century
(at any rate, not greater than the rate of upward growth of
coral), continued for age after age. As the edge of the fring-
ing reef sank, new coral would grow up from it to the sur-
face ; and, as the most active and important of the reef- build-
ers flourish best in the very surf of the breakers, so the margin
of the reef would grow faster than its inner portion, and the
discrepancy would increase as the latter, sinking deeper and
deeper, became farther removed from the region of active
growth. Nevertheless, the sea-bottom within the reef w^ould
constantly tend to be raised by the accumulation of frag-
ments, and by the deposit of fine mud, in its sheltered and
comparatively calm waters. On the other hand, on the sea-
ward face of the reef, no possible extension could take place
by direct growth; and that by accumulation must be exceed-
ingly slow, the incessant wash of tides, waves, and currents,
tending incessantly to spread any talus over a wider and
wider area.

Thus, then, the edge of the reef unceasingly compensates
itself for the depression which it undergoes, while, inside the
reef, only a partial compensation takes place, and, outside,
hardly any at all. Continue the sinking process until its
highest peak was but a few hundred feet above the surface,
and all that would be left of Mauritius would be an island
surrounded by an encircling reef ; carry on the depression
further still, and a circular reef, or atoll, alone would remain.
But the region of the coral-reefs is, for the most part, that of
constant winds. During the whole process of growth of the
reef, therefore, one of its sides — that to windward — has been
exposed to more surf than that to leeward. Not only will
the greater quantity of debris, therefore, have been heaped
up by storms upon the windward side, but the coral-builders
themselves will here have been better fed, better aerated, and
consequently more active. Hence it is that, other things
being alike, there is a probability that the leeward side of
the reef will grow more slowly, and repair any damages less
easily, than the windward side ; and hence, again, as a result,

ANCIENT REEFS. 153

the known fact that the practicable channels of entrance into
encircling reefs or atolls are usually to leeward.

The winds and waves are singularly aided in grinding
down the corals into mud and fragments by the Scari and
Holothurim which haunt the reefs ; the former browsing
upon the living polyps, with their hard and parrot-like jaws,
and passing a fine calcareous mud in their excrements ; the
latter, more probably, swallowing only the smaller fragments
and mud, and, having extracted from them such nourishment
as they may contain, casting out a similar product. It is
curious to reflect upon the similarity of action of these worm-
like Holothurlw upon the sea-meadows of coral, to that
which the Earthworms, as Darwin has shown, exert upon our
land-meadows !

In the Palaeozoic period reefs like those which have just
been described appear to have abounded in our own latitudes ;
and there is the most striking superficial resemblance be-
tween the ancient beds of calcareous rock which record their
existence, and the masses of coral limestone, hard enough to
clink with a hammer, which are now being formed in the
Pacific, by the processes of accumulation of coral mud and
fragments, and their consolidation by percolating water.
Closer examination, however, shows an important difi'erence
in the nature of the corals which compose the two reefs. The
modern limestones are made up of Perforata, Millepores,
and Aporos'i. The ancient ones contain Millepores, but usu-
ally neither Perforata nor Aporosa — both these groups being
replaced by the Hiigosa, none of whose members (with some
doubtful exceptions) have survived the Palaeozoic period.
On the other hand, Palmocyclus and Pleurodictyon are the
only genera belonging to the Aporosa or Perforata, which
have yet been discovered in strata of greater than mesozoic
age.

The Ctexophora.' — These are freely-swimming marine
animals, which never give rise by gemmation to compound
organisms, and are always of a soft and gelatinous consist-
ence, their chief bulk being made up by the greatly -devel-
oped mesoderm. Many are oval or rounded {Beroe, Pleuro-

1 Allman (" Monograpli of the Tubularian HydrolcU," 1871, pagre 3) consid-
ers that the CUnophora are more properly arranfjed amonsj the Hydrozoa. I
confess, however, that I see no reason to depart irom the conclusion to which
I was led by the study of the structure of Pleurobrachia^ many years ago, that
the Ctenophora are peculiarly modified Actinozoa.

154 THE ANATOMY OF INYERTEBRATED ANIMALS.

hrachia^ Fig. 31), while in others the body is produced into
lobes {Callianh'a), or may even be ribbon-shaped (Cestwn) ;
but, whatever their form, they present a distinct bilateral
symmetry, similar parts being disposed upon opposite sides
of a median plane, which is traversed by the axis of the
body. The mouth is situated at one end of this axis, which
may be termed the oral pole. At the opposite, or aboral
pole, there is no median aperture, but usually, if not inva-
riably, a pair of apertures a short distance apart. The faces
of the halves of the body present four longitudinal bands of
long and strong cilia, disposed in transverse rows, like so
many paddles ; these constitute the chief organs of locomo-
tion. Each half is also often provided with a long retractile
tentacle ; and lobed processes of the body, or non-retractile
tentacula, may be developed on its oral face. The mouth
leads into a wide, but flattened, gastric sac, the aboral end of
which is perforated, and leads into a chamber termed the
infundibuluin. From the aboral face of this, a canal which
bifurcates, or two canals, lead to the aboral apertures. On
opposite sides of the infundibulum a canal is given off toward
the middle of each half of the body, which sooner or later
divides into two, and these two again subdivide, so that four
canals, which diverge and radiate toward the inner faces of
the rows of paddles, are eventually formed. Having reached
the surface, each radiating canal enters a longitudinal canal,
which underlies the row of paddles, and may give off branches,
or unite with the other longitudinal canals in a circular canal
at the aboral end of the body. In addition, two other canals,
which run parallel with each flat face of the gastric sac, open
into the infundibulum. And, w^hen retractile tentacula are
present, their cavities also communicate with the same cham-
ber.

The entire system of canals is in free communication with
the gastric cavity, and corresponds Avith the enterocoele of
an JLctinia. Indeed, an Actinia with only eight mesenter-
ies, and these exceedingly thick, whereby the intermesenteric
chambers would be reduced to canals ; with two aboral pores
instead of the one pore, which exists in Cereanthus j and
with eight bands of cilia corresponding with the reduced
intermesenteric chambers, would have all the essential pecu-
liarities of a Ctenophoran.

The question whether the Ctenopliora possess a nervous
system or not is still under debate. Between the aboral aper-
tures there is a rounded cellular body, on which there is

THE CTENOPHORA.

155

seated, in many cases, a sac containing solid particles, like
one of the lithocysts of the medusiform Hydrozoa, I see no
reason to doubt that the rounded body is a ganglion and the
sac a rudimentary auditory organ. Bands which radiate
from the ganglion to the rows of paddles may be regarded
as nerves ; though they may contain other than nervous
structures.^

The ova and spermatozoa are developed in the lateral
walls of the longitudinal canals, which correspond with the
faces of the mesenteries in the CoralUgena^ and the sexes
are usually united in the same individual.

Fig. 31.— Dia;r'am of Ple>jrdbrachia.—a. month ; 6, stomach ; r, inftindlbulam ; d,
hoilzontal canal; «, one of its branches dividini; ajjain at / into two branches
which open Into the longitudinal canals, g g, parallel with which the ciliated
area runs ; h, sac of the tentacle, 2, with one of its branches, Jc ; I, canal run-
n ins; by the side of the stomach; m, tentaculigeroas canal; n n, canals opening
at tbe'aboral apertures, o, on each side of j3, the ganglion and lithocyst.

1 Grant originally described a nervous ganglionated ring, whence longitu-
dinal cords proceeded in Cydippe {PleurobracJiia)^ but Ms observation has not
been veritied by subsequent investigators. According to Milne-Edwards, fol-
lowed by others (among whom I must include myself), the nervous system
consists of a ganglion, situated at the aboral pole of the body, "whence nerves
radiate, the most conspicuous of which are eight cords which run down the
corresponding series of paddles ; and a sensor}' organ, having the characters
of an otolithic sac, is seated upon the ganglion. Agassiz and Kolliker, on the
other hand, have denied that the appearances described (though they really
exist) are justly interpreted. And again, though the body, described as an
otolithic sac, undoubtedly exists in the position indicated in all or most of the
Cterwphora^ the question has been raised whether it is an auditory or visual
organ.

These problems have been recently reinvestigated with great care, and by
the aid of the refined methods of modern histology, by Dr. Eimer, whose de-
scription of the nervous system has already been quoted {supra, p. 63).

156 THE ANATOMY OF INYERTEBRATED ANIMALS.

The development of the Ctenophora has recently been
thoroughly investigated by Kowalewsky and by A. Agassiz
(" Memoirs of the American Academy of Arts and Sciences,"
1874).

The laid egg is contained in a spacious capsule, and con-
sists of an external thin layer of protoplasm, which, in some
cases, is contractile, investing an inner vesicular substance.
After fecundation, the vitellus thus constituted divides into
two, four, and finally eight masses ; on one face of each of
these the protoplasmic layer accumulates, and is divided oif
as a blastomere of much smaller size than that from which it
arises. By repeated division, each of these gives rise to still
smaller blastomeres, which become distinctly nucleated wdien
the}'' have reached the number of thirty-two, and form a
layer of cells, which gradually spreads round the large blas-
tomeres, and invests them in a complete blastodermic sac.
At the pole of this sac, on the face opposite to that on which
these blastoderm-cells begin to make their appearance, an
ingrowth or involution of the blastoderm takes place, which,
extending: throiio'h the middle of the laro-e velk-masses tow-
ard the opposite pole, gives rise to the alimentary canal.
This, at first, ends by a rounded blind termination ; but from
it, at} a later period, prolongations are given off which be-
come the canals of the enterocoele.

At the opposite pole, in the centre of the region corre-
spondino: with that in which the cells of the blastoderm first
make their appearance, the nervous ganglion is developed by
metamorphosis of some of these cells.

The invaginated portion of the blastoderm, which gives
rise to the alimentary canal, appears to answer to the hypo-
blast, while the rest corresponds w^th the epiblast. The
large blastomeres which become inclosed between the epi-
blast and hypoblast in the manner described seem to serve
the purpose of a food-yelk ; and the space which they origi-
nally occupied is eventually filled by a gelatinous connective
tissue, which possibly derives its origin from wandering cells
of the epiblast.

In those Ctenophora the bodies of which depart widely
from the globular form in the adult state, the young undergo
a sort of metamorphosis after they leave the Qgg, and have
acquired all the essential characters of the group to which
they belong.

As might be expected from their extreme softness and
perishable nature, no fossil Ctenophora are known.

CHAPTER IV.

THE TURBELLAEIA, THE ROTIFERA, THE TREMATODA, AND THE

CESTOIDEA.

The TtTRBELLAKiA. — The animals which constitute this
group inhabit fresh and salt water and damp localities on
land. The smallest are not larger than some of the Infusoria,
which they approach very closely in appearance, while the
largest may attain a length of many feet. Some are broad,
flattened, and discoidal, while others are extremely elongated
and relatively narrow. None are divided into distinct seg-
ments, except the genus Alaurina, in which there are four ;
and the ectoderm, which constitutes the outer surface of the
body, is everywhere beset wi1?h vibratile cilia. Rod-like
bodies, similiar to those met with in some Infusoria and in
many Annelida, are often imbedded in its substance, and in
some genera (e. g., Microstomum, Thysanozoon) true thread-
cells occur. Stiff setae project from the ectoderm in some
species.

The aperture of the mouth is sometimes situated at the
anterior end of the body, sometimes in the middle, or toward
the posterior end, of its ventral face. In many, the oral
aperture is surrounded by a flexible muscular lip, which some-
times takes on the form of a protrusible proboscis.

A definite digestive cavity can hardly be said to exist in
the lowest Tarbellaria (e. g., Convoluta) in which the endo-
dermal cells are not arranged in such a manner as to bound a
central alimentary cavity, and the food finds its way through
the interstices of an endodermal parenchyma. In the higher
forms, the alimentary cavity, which may be simple or rami-
fied, provided with an anal aperture or without one, is lined
by the endoderm, between which and the ectoderm is an in-
terspace more or less completely occupied by the connective
and muscular tissues of the mesoderm. Hence there is no
definite perivisceral cavity.

153 THE ANATOMY OF INVERTEBRATED ANIMALS.

The Turhellaria possess vessels of two kinds : 1. Water-
vessels, which open externally by one or more pores, and are
ciliated. When these vessels are present, there are usually
two chief lateral trunks, from which many branches are given
off. It is probable that the ultimate ends of these branches
open into lacunar interspaces between the elements of the
tissues of the mesoderm. 2. JPsend-hcemal vessels, which ap-
pear to form a closed system, usually consisting of one median
dorsal and two lateral trunks, which anastomose anteriorly
and posteriorly. The walls of these vessels are contractile
and not ciliated, and their contents are clear, and may be
colored. These two systems of vessels have been shown by
Schulze to coexist in Tetrastemma, The nervous system con-
sists of two ganglia placed in the anterior end of the body,
from which, in addition to other branches, a longitudinal cord
extends backward on each side of the body. In some cases,
these lateral trunks exhibit ganglionic enlargements, from
which nerves are given off; and they may become approxi-
mated on the ventral side of the body, thereby showing a
tendency to the formation of the double ganglionated chain
characteristic of higher worms. Most possess eyes, and some
have auditory sacs. The Turhellaria are both monoecious
and dioecious, and the reproductive organs vary from the
utmost simplicity of structure to considerable complexity.
In most, the embryo passes by insensible gradations into the
form of the adult, but some undergo a remarkable metamor-
phosis.

The Turhellaria are divisible into two groups. In the one,
the Aprocta, the digestive cavity is caecal, having no anal
aperture ; in the other, the Proctucha, it is provided with an
anal opening. The two groups form parallel series, in each
of which organization advances, from forms w^iich are little
more than gastrulse provided with reproductive organs, to
animals of relatively high organization. In the simplest of
the Aprocta, such as Macrostomum^ the oral opening is
devoid of any protrusible muscular proboscis, and the aliment-
ary sac is a simple straight bag. The male and female gen-
erative organs are united in the same individual, and each
consists of an aggregation of cells; w^hich, in the former case,
gradually enlarge, fill with yelk-granules, and become ova;
while, in the latter, thc}^ are converted into spermatozoa.
The generative cells are contained within a sac, which opens

1 E. Van Beneden, " Eecherclies sur la Composition ct la Signification de
rCEuf," 1S70, p. 64.

THE TURBELLARIA.

159

externally by a median pore on the oral face of the body, the
male aperture being posterior to the female. The margins of
the male aperture are produced into a curved prominence, the
penis.

Those Tarhellaria which resemble 3facrostommn in having
a straight, simple digestive cavity, are termed Ukabdocoela.
They, for the most part, possess a buccal proboscis, which is
capable of being protruded from, or retracted into a chamber

§^^^

s-^:

Fig. S2.—Opisthomum (aftp.r Schulze).— a, central nerrous system ; ramifications of
the water- vessels are seen close to it; b, mouth; c, proboscis; d, testes; e, vasa
deferentia; /, vesicula seminalis ; g, penis ; A, sexual aperture ; i, vatrina ; k, sper-
raatheca ; L germarium; m, vitellarium ; n, uterus with two ova inclosed within^
their hard shells.

formed by the walls of the circum-oral region of the body
(Fig. 32, c).

In some (e. g., Prostomum) the anterior end of the body is

160 THi: ANATOMY OF INVERTEBRATED ANIMALS.

provided with a second hollow muscular proboscidiform organ,
which may be termed the frontal proboscis.

In all the higher rhabdocoelous Turhellaria^ the female
generative apparatus becomes complicated by the presence
of a special gland, the vitellariuin (Fig. 32, m), in which an
accessory vitelline substance is formed. There is a single or
double germariuin (Fig. 32, I), having nearly the same struct-
ure as the ovary of Macrostonium^ and the ova are formed
in it in the same w^ay. When detached, how^ever, they con-
tain no vitelline granules ; but the two vitellaria, which are
long and simple or branched tubes, open into the oviduct ;
and the vitelline matter which they secrete envelops the
proper ovum, and becomes more or less fused w^ith it, as it
passes into the uterine continuation of the oviduct connected
with the outer, or vaginal, end of the uterus. There is usually
a spermatheca, or receptacle for the seminal fluid (Fig. 32, k),
and the eggs, after impregnation, are inclosed within a hard
shell (Fig. 32, oi). The testes and vasa deferentia (Fig. 32,
d, e) generally have the form of two long tubes. The penis
is often eversible and covered with spines (Fig. 32, g).

In some genera a difference is observed between the eggs
produced in summer, which have a soft vitelline membrane,
and those produced later. These so-called winter ova have
hard shells.

The water-vascular system consists of lateral trunks,
which open hy a terminal pore, or by many pores, and give
off numerous ramifications. They are not contractile, but
their inner surface is ciliated.

Many of the Ilhabdocoela multiply by transverse fission ;
and, in the genus Catenula, the incompletely separated ani-
mals produced in this way swim about in long chains.

The vitellus of the impregnated ovum undergoes complete
yelk-divison, and the embryos pass directly into the form of
the parent ; but the precise nature of the steps of the devel-
opmental process requires further investigation. How^ever,
there seems little reason to doubt that the ectoderm and en-
doderm are formed by delamination.

In the remaining Ajyrocta^ termed Dendroccela^ the diges-
tive cavity gives off many csecal, frequently branched, pro-
cesses into the mesoderm, one of which is always median and
anterior (Fig. 33) ; and the mouth is always provided with a
proboscis. Some {Procotyla) have a frontal proboscis, and
others {JBdellura) a posterior sucker. The animals commonly

THE DENDKOCCELA.

161

known as Planarice belong to this division. Some are ma-
rine, some fresh-water, and some terrestrial.

In the fresh-water forms, the female reproductive appa-
ratus has a distinct vitellarium, as in the higher Hhahdocoela,
and there is only one common genital aperture. But, in the
marine Planar ioe (Fig. 33), there is no vitellarium ; the ova-
ries and testes are numerous, and scattered through the meso-
derm, being connected with the exterior by ramifications of
the oviducts and of the vasa deferentia. A ramified gland,
which secretes a viscid albumen or envelope for the eggs,

Fig. Zi.—Polycdis (Leptoplana) IcBvigata (after Quatrcfast^p)-— o, mouth; ft, buccal
cavity; c, (Esophageal orifice ; d, stomach ; e, ramifications of gastric caeca; f,
ganglia ; g, testes ; h, vesiculiB seminales ; i, male genital canal and peuis ; k, ovi-
dacts ; /, spermathecal dilatation at their junction ; m, vulva.

opens into the vagina, and the female is distinct from the
male aperture. Planaria dioica is unisexual.

In some of the Planarice there are distinct water-vascular

162 THE ANATOMY OF INYERTEBRATED ANIMALS.

canals of the ordinary kind ; but in the land Planarians ^ two
nearly simple canals, occupied by a spongy tissue, and the
connection of which with the exterior has not been observed,
occupy the place of the water-vessels.

The fresh-water Planarke, like the Rkahdocoela^ undergo
no metamorphosis in the course of their development ; and
the like is true of some of the marine Dendroccela. Kefer-
stein ' has carefully worked out the development of Lepto-
plana [Polycells). The vitellus undergoes division first into
two and then into four equal blastomeres ; next, from one
surface of these four blastomeres, four small segments are, as
it were, pinched off. These divide rapidly, and form a blas-
toderm, which grows over the more slowly dividing large seg-
ments, and eventually incloses them. So far, the process is
very similar to that which has been described in the Cteno-
phora. But though Keferstem describes and figures the
various stages by which the globular ciliated embryo attains
the form of the adult, neither his description nor the figures
enable one to say whether the alimentary cavity arises by de-
lamination or by invagination, nor to trace the mode of origi-
nation of the buccal proboscisough, th this organ is one of
the first to make its appearance, and its aperture becomes the
future mouth.

In some of the marine Planarim, however, the embryo,
wlien it leaves the Qg^, differs very widely from the adult.
Johannes Miiller described such a larva, in which the body is
provided with eight lobes or processes, one ventral and median
in front of the mouth, three lateral, and one dorso-median.
The edges of these processes are fringed by a continuous
series of cilia, which pass from one process on to another, so
as to form a complete circlet round the body. The successive
working of the cilia forming this lobed transverse girdle of the
body produces the appearance of a rotating wheel, as in the
RotAfera. The eyes are situated on the aboral face of the
embryo, in front of the ciliated circlet, while the mouth opens
immediately behind it. As development proceeds, tlie lobes
disappear, and the body takes on the ordinary Planarian
character.

As will be seen, some of the Proctucha have larvae simi-
larly provided with a pra3-oral ciliated zone ; and larvae of

* Moseley, " On the Anatomy and Histology of the Land Planarians of Cey-
lon." (" Philosophical Transactions," 1873.)

"^ "Beitrilge zur Anatomie und Entwickelungsgeschichte ciniger See-Plana-
rien," 1868.

THE PROCTUCnA.

1G3

the same fundamental type abound among tlie polycbastous
Annelida, the Uchinodermata, and the Mollusca,

mm'

Fig. 34.—^, young Tetrastemma.—aa, central ^nglia of the nervous system; 55, cil-
iated fossae ; c, aperture through which the proboscis is protrude!; d, anterior
portion of proboscis ; e, posterior muscular part, fixed to ttie parietes at/; g, in-
testine; h, anal aperture: i, water- vessels; k, rhythmically contracting vessels.
(After Schulze.) B, anterior extremity of the everted proboscis of letrastemmay
exhibiting the principal and the reserve stilets. (After Schulze.)

The lowest Proctucha^ such as 3Iicrostomiim^ have no
frontal proboscis (whence they are termed Arhynchid)^ and
they differ very little from the lowest Rhahdocoela., save in the
fact that there is an anus, and that the sexes are distinct.
But all the other Proctucha [Rhynchoccela., or Nemerteans)
are provided with a frontal proboscis, which sometimes oc-
cupies the greater part of the length of the body (Fig. 34).
It has special retractor muscles, and its internal surface is
either merely papillose, or may possess a peculiar armature,

164 THE ANATOMY OF INVERTEBRATED ANIMALS.

consisting of a sharp cLitinous style (Fig. 34, J3). There is
no buccal proboscis, but the mouth leads into a long, straight
intestine, with short, lateral, caecal dilatations.^

The Proctucha usually present only the pseud-htemal ves-
sels, though, as has been mentioned above, Schulze found water-
vessels coexisting with them in Tetrastemma (Fig. 34).

The nervous system of the Proctucha is like that of the
Aprocta ; but, in correspondence with the often extreme elon-
gation of the body, the backwardly- prolonged cords are very
stout. Moreover, the ganglia are united by an additional
commissure over the proboscis, which thus traverses a ner-
vous ring. In some, the lateral cords approach one another
on the ventral aspect of the body, and ganglionic enlarge-
ments appear where the nerves are given otr, thus present-
ing an approximation to the double ganglionated chain of
higher forms.

In addition to eyes, almost all the Proctucha possess two
ciliated foss?e, one on each side of the head (Fig. 34, bh)^
which receive nerves from the ganglia. Occasionally two
otolithic vesicles are attached to the cerebral ganglia.

The Proctucha are almost alw^ays dioecious. The simple
reproductive glands are lodged in the intervals between the
saccular dilatations of the intestine, and the ova and sper-
matozoa usually make their way out by the dehiscence of
the integument. In some, however, the embryos are devel-
oped in the ovarian sacs, or in the cavity of the body. In
most of the Proctucha, the egg, after passing through the
morula stage, acquires an alimentary cavity, apparently by
delamination, and passes, without other metamorphosis than
the shedding of a ciliated outer investment, into the form of
the adult.

Prof. A. Agassiz'' has described a free-swimming larva,
the broad anterior end of the body of which is surrounded
by a zone of cilia, immediately behind which the mouth opens ;
while around the anal aperture, at the narrow posterior end,
is a second circlet of cilia. This larva exactly resembles
those forms of polj'chffitous Annelidan larvre which are called
Telotrocha. As in these Annelids, the region of the body
which lies between the two ciliated rings elongates and be-
comes segmented, while a pair of eyes and two short tenta-

1 For the orsranization of the Rhynchocoele Tvrhdlaria, or Nemerteans, see
Dr. C. Mcintosh's elaborate monof?raph lately published by the Ray Society.

3 " On the Young Stages of a few Annelids." (Annals of the Lyceum of
New York, 1864.)

THE PROCTUCHA.

165

cles are developed on the head in front of the prae-oral ciliated
band. But, as development advances, the segmentation be-
comes obliterated, the ciliated bands and the feelers vanish,
and the worm assumes the characters of a Nemertean/

Fig. 35.

Fig. 37.

Fig. 36.

Fig. ZS-Zl.—Pllidhim gyrans (after Leuckart and Pagenstecher).

85. YoTing PUidium : a, alimentary canal; 6, radimeut of the Nemertean.

36. Pilidium with a more advanced Nemertean.

37. Newly -freed Nemertean.

In species of the genus Lineus^ the ciliated embr^'O which
leaves the egg is speedily converted into a body like a helmet
with ear-lappets, and having a tuft of cilia in place of a plume

1 It is very probable, however, that this larva belongs to the genus Poh/qor-
dius, which appears to be an annectent form between the Turbellaria and other
groups. See Schneider, " Ueber Bau und Entwickelung von Polygordius."
C' Archiv fur Anatomie und Physiologic," 1808.)

166 THE ANATOMY OF INVERTEBRATED AMMALS.

(Fig. 35). The lappets are fringed with long cilia, and be-
tween them, where the head would fit into a helmet, is the
aperture of a mouth, which leads into a csecal pouch-like ali-
mentary cavity. This larv^a w'as named by MuUer, w^ho dis-
covered it, Pllidmin gyra7is. On each side of the ventral
face of the Pllidimn^ two involutions of the integument take
place. Aggregations of cells in relation with these, and
probably forming part of the mesoblast, appear, eventually in-
close the alimentary canal of the Pllidium, and give rise to
an elongated vermiform body, in which the characteristic feat-
ures of a Nemertean soon become discernible (Fig, 36). The
worm thus developed becomes detached (Fig. 37) and falls to
the bottom, carrying with it the alimentary canal of the Pi-
lidliDn, and leaving the ciliated integument to perish.

In this remarkable process of development the formation
of the Nemertean body may be compared, on the one hand,
to that of the segmented mesoblast in Aniielida and Arthro-
poda^ and, on the other, to that of an Echinoderm (especially
Echinus), within its larva.

Tpie Rotifera. — The " wheel-animalcules," as they were
termed by the older observers, on account of the appearance
of rotation produced, as in many Annelid larvae, by the work-
ing of the vibratile cilia with which the oral end of the body
is provided, were formerly included among the Infusoria,
However, they are true Metazoa^ as their vitellus undergoes
division into blastomeres, and the tissues of the body are pro-
duced by the metamorphosis of the cells into which the blas-
tomeres are converted. They are free or adherent, but never
absolutely fixed animals, and they do not multiply by gem-
mation or fission. The oral end of the body is usually broader
than the opposite extremity, and presents the form of a disk,
sometimes produced into tentacle-like prolongations (Fig. 39),
The edges of this trochcd disk are fringed with long cilia, but
the general surface of the body, instead of being ciliated,
as in the Turhellaria^ is formed by a dense, generally chiti-
nous, cuticular layer, which is sometimes converted into a kind
of shell and variously sculptured. Transverse constrictions,
which are slight in the anterior part of the body, but may
become more marked toward its posterior end, give rise to an
imperfect segmentation. The segments do not appear to ex-
ceed six, and the divisions are less marked in the tubicolous
than in the free Motifera. The mouth is a funnel-shaped
cavity, situated in the middle, or on one side, of the trochal

THE ROTIFERA. 167

disk. The walls of this cavitv are abundantly ciliated, and
at the bottom is a muscular pharynx, or mastax^ provided
with a peculiar armature. Sometimes, as in Stephanoceros, a
large crop-like cavity lies between the mouth and the mastax,
and the aperture of communication between this crop and
the mouth is guarded by a valve formed by two broad mem-
branous folds which project into the cavity of the crop. The
armature of the mastax generally consists of four pieces — two
lateral, the mallei^ and two central, constituting the mens.
The contraction of the muscular masses, to which the mallei
are attached, causes the free ends of the latter to work back-
ward and forward upon the incus, and crush the prey which
is taken into the mouth.*

A short oesophagus, provided with cilia or vibratile mem-
branes, leads into a digestive cavity bounded by the endo-
derm. The anterior or gastric part of this cavity is usually
dilated, and gives ofif a large caecum on each side. The pos-
terior, narrower, intestinal part usually opens externally by a
cloacal chamber ; but, in some Rotifers (e. g., Kotommatd)^
the alimentary cavity is a blind sac, devoid of intestine or
anus ; and in the males, so far as they are known, the whole
alimentary canal is aborted and represented by a solid cord.

A spacious perivisceral cavity occupies the interval be-
tween the walls of the alimentary canal and the parietes of
the body. The latter contains circular and longitudinal mus-
cular fibres, which may be smooth or striated.

Opening into the cloaca there is usually a large thin-walled
vesicle with rhythmically contractile walls ; and, in connection
with this, are two delicate water-vessels, which pass forward,
often giving off short lateral branches, and eventually break
up into numerous ramifications in the trochal disk. The
branches are open at the ends, wherel^y the cavities of the
water-vessels are in communication with the perivisceral cav-
ity on the one side, and with the surrounding water on the
other. Here and there, in the course of the main trunks and
at the ends of the branches, long cilia, which, by their con-
stant undulation, give rise to a flickering motion, are situated.

The nervous system is represented by a relatively large
single ganglion placed on one side of the body, near the tro-
chal disk. One or more eye-spots are sometimes seated on
the ganglion, and there are other organs which appear to be

1 i've^, for the various forms of this apparatus, Gosse, "On the Structure,
Functions, and Tlomolosrues of the Manducating Apparatus in the Botifera,^'*
(rhilosophical Transactions, 1855.)

168

THE ANATOMY OF IXVERTEBRATED ANIMALS.

sensory. Such are the ciliated pit and the spur-like process
(calcar) or processes, provided at the end with a tuft of setae,
which occur in many Rotifers, and are more or less closely
connected with the ganglion. In some there is a sac filled
with calcareous matter (otocyst ?) attached to the ganglion.

Fig. 2S.—n ydalina senta (after Cohn).— ^, female : a, anas ; b, contractile vesicle ;
c, water-vessels ; g. ovary ; f. <,'anii:lion. B, male : a, peuis ; 6, coutractile vesicle ;
c, testis ; /, ganglion ; g, setigerous pit.

The ovarium and the testis are simple glands which open
into the cloaca, and are always placed in distinct individuals.
All the males at present known differ from the females in be-
ing much smaller, and in their digestive canal being arrested
in its development. The males copulate with the females,
and the eggs are sometimes attached to, and carried about by,
the latter — e.g., Brachionus.

In some Rotifers, the eggs are distinguishable, as in cer-
tain Turhellaria, into summer and winter ova. The latter
are inclosed in a peculiar shell. In Lacimilaria, it appeared
to me that the winter ova were segregated portions of the
ovarium, and that they were probably developed without im-
pregnation. Cohn, on the contrary, has given reasons for be-

THE ROTIFERA. 169

lieving that the summer ova are occasionally, if not alwaj-s,
developed without fecundation, and that it is the winter ova
which are fecundated.

The egg undergoes complete yelk-division, and the em-
brj'O gradually passes into the adult form. The blastomeres
are soon of unequal sizes, and the smaller, as an epiblast, in-
vest the larger, which form the hypoblast.

Salensky's ^ recent observations on jB}'achionus urceolaris
show that a depression arises on one face of the epiblast and
that the antero-lateral parts of this depression are converted
into the trochal disk, while its median posterior part grows
out into the "foot; " and he points out the resemblance of
the embryo in its early stages to that of some Gasteropods.

An involution of the epiblast at the bottom of the depres-
sion gives rise not only to the oral chamber, but also to the
mastax ; eventually communicating with the gastro-intestinal
division, w^hich is developed out of the hypoblast. The gan-
glion is a product of the epiblast.

Some of the modifications of the general structure thus
described, which occur in the different groups of the Hotife-
ra, are of considerable interest.

Thus, in the tubicolous forms, the body is elongated and
terminated posteriorly by a discoidal surface of adhesion.
The animals (of which a number are often associated together),
fixed by this disk, inclose themselves in cases, the foundation
of which is a gelatinous secretion. The intestine is bent
upon itself {Lacinularia, Fig. 39, II.), and opens upon the
face of the body opposite to that upon which the ganglion is
placed. The peduncle of attachment is therefore a process of
the neural face of the body. In these Jiotifera the trochal
disk is sometimes produced into long ciliated tentacula,
which surround the mouth symmetrically i^Stephanoceros^
Fig. 39, v.), or its edges may be provided with two circlets of
cilia, one in front of, and the other behind, the oral aperture ;
and it may be bilobed or horseshoe-shaped, as in Melicerta^
and Lacinularia ' (Fig. 39, I., II.).

In the free Rotifers, the body may be rounded, sac-like,
and devoid of appendages, as in the genus Asplanchiia, which
has neither anus nor intestine. In Albertia and Lindia^ on
the other hand, the body is elongated and vermiform. Most
of the free JRotifera (Fig. 38) are provided with a segmented

1 ZdUchHft fTir wiss. Zoologie^ 18T2.

2 Huxley, Lacinutaria socialis. (Transactions of the Microscopical Society,

8

170 THE ANATOMY OF INYERTEBRATED ANIMALS.

and sometimes telescopically-jointed " foot," usually termi-
nated by two styles, which can be approximated or divari-

FiG. 39.— Diagrams showing the arrangement of the cilia of the trochal diek in the
jRotifera. I. Larval Lacinularia. If. Adult Lacinularia. III. Philodina. IV.
Brachionus. V. StephanocerOH. M, mouth ; G, gan^ilion ; J., anus.

cated like pincers, and serve to anchor the body. This foot
is a median process of that face of the body which is opposite
to that on which the ganglion is placed, so that it is not the
homologue of the peduncle of the tubicolous forms.

Polyarthra and Iriarthra possess long, symmetrically ar-
ranged, movably articulated setas ; and Pedalion has median
appendages proceeding from both the neural and the opposite
faces of the body, as well as lateral appendages.

In most of the free Rotifers the trochal disk is large ; it
may be bilobed or folded upon itself (Fig. 39, III.), or its sur-
face may give rise to ciliated processes (Fig. 39, IV.). In
Alhertia and Notomynata tardigrada^ however, the trochal
disk is reduced to a small ciliated lip around the oral aper-
ture ; and there is no trochal disk in Apsilus, Lindia, Ta-
phrocampa, and Balatro. Some few Rotifera are parasitic.
Thus Alhertia is an entoparasite, and Bcdatro an ectopara-
site, upon oligochaetous Annelids.

Under the name of Gasterotricha^ Metschnikoff and Cla-
parede * include the curious aquatic genera ChcBtonotus, Ich-
thydiwin^ Choetura^ Cephalidium^ Dasydltis, Turhanella^ and
Ilernidasys^ the last of which alone is marine. These animals
have been united with the Motifera, but they differ from them
in the absence of a mastax and in the disposition of the cilia,
which are restricted to the ventral surface of the body. It

* Olaparede and Metschnilcoff, "Beitriige zur Kenntniss der Entwickelungs-
geschichte der Chaetopoden," 1868.

THE TREMATODA. 171

appears probable that tliey form an annectent group between
the Rotifera and the Turbellaria, which last approach the Eo-
tlfera by such forms as Dinophilus,

The free Rotifers present marked resemblances to the
telotrochous larvae of Annelids. The young Lacinularia, for
example, has a circular prae-oral disk provided with two eye-
spots and a second circle of cilia behind the mouth, and is
wonderfully like an Annelid larva (Fig. 39, I.). The append-
ages of Triarthra and Polyarthra may be compared to the
lateral bundles of long setae of the larvae of Spio and Nerine,
and the pharyngeal armature is essentially Annelidan. On
the other hand, in the sessile tubicolous Rotifera, the trcchal
disk assumes the characters of the lophophore in the Polyzoa,
and of the tentacular circlet of the Gephyrean Phoronis,
Many years ago I drew attention to the points of resem-
blance between the Rotifera and the larvae of Echinoderms
C' Oa Zctcmularia socialis,'" I. c). Of any such close and
direct relations with the Crustacea, I see no evidence ; but
Pedalion,^ with its jointed setose appendages and curious
likeness to some J^aupUus conditions of the lower Crustacea,
suggests that connecting links in this direction may be found.*
In fact, the Rotifera, as low 3Ietazoa with nascent segmenta-
tion, naturally present resemblances to all those groups which,
in their simpler forms, converge toward the lower Metazoa.

The TREiiATODA. — These are all parasitic, either upon the
exterior (ectoparasites) or in the internal organs (endopara-
sites) of other animals. Many are microscopic, and none
attain a length of more than an inch or two. Most have a
broad and flattened form, one face being ventral and tlie
other dorsal, and the hody is never segmented.

In the adult, the ectoderm is not ciliated, but its outer-
most layer is a chitinous cuticula. In most Trematoda, one
or more suckers are developed upon the ventral surface of the
body, behind the mouth. These are sometimes armed with
chitinous spines or hooks ; and setae of the same character

1 Hudson, " On a New Rotifer." {Montlily Microscopical Journal^ 1871.)
3 The singular marine genus EcTiinoderes (Dujardin) is perhaps such a link.
These are minute worm like animals, with a rounded head, followed by a num-
ber (ten or eleven) of distinct ses^raents, the last of which is bifurcated. There
are no litnbs, but the head is provided with recurved hooks, and the body seg-
ments with paired setae. The nervous system appears to be represented by a
single ganglion, which lies in the head and presents eye-spots. The develop-
ment or Echiiioderes is unknown. (See Greef, " Arcbiv flir Naturgeschichte,"
1869.)

172 THE ANATOMY OF INVERTEBRATED ANIMALS.

ma}' be developed in other parts of the body, especially in the
region of the head.

The mouth is usually terminal, but is sometimes ventral
and sub-central ; it is ordinarily placed in the centre of a
muscular sucker, rarely [)roboscidiform. The alimentary canal
is never provided with an anus. Sometimes a simple sac, it
is often bifurcated, and occasionally branched, like that of the
dendrocoele Turbellaria, Sometimes (Amphilina, Amphipty-
ches) the alimentary canal is absent ; and, according to Van
Beneden, it becomes aborted in the adult Distoma filicolle.
The interval between the endoderm and the ectoderm is oc-
cupied by a cellular or reticulated mesoderm, in which abun-
dant muscular fibres are developed. The peripheral muscular
fibres form an external circular and an internal longitudinal
layer.

The water-vascular system is well developed, and may
consist of — (1) a contractile sac, which opens externally and
communicates with (2) longitudinal vessels with contractile
non-cihated walls, from which proceed (3) non-contractile and
ciliated branches which ramify through the bodj'', and the
ultimate ramifications of which probably end by open mouths,
as in the Rotifera.

There is no pseud-hcemal system. The nervous S3"stem has
not been discovered in all ; but, when it exists, it has the
same arrangement as in the aproctous Turbellaria. Eye-
spots have been observed, but no other sense-organs. With
rare exceptions, the Trematoda are hermaphrodite, and the
reproductive organs are constructed upon the same type as
in the rhabdocoele Turbellaria , a large vitellarium being al-
ways present. The accessory vitellus is included, in the
form of numerous pellets, along with the primitive ovum, and
is absorbed pari passu with the development of the embryo.

Asjndogaster concliioola (Fig. 40) inhabits the pericardial
cavity of the fresh-water muscle ; it is a very convenient sub-
ject for examination on account of its small size, and the ease
with which it can be rendered sufficiently transparent for the
displa}^ of the arrangement of its internal organs, by the
judicious use of the compressorium. The flat oval bod}'",
rounded posteriorly, is produced in front into a truncated
cone, on the face of which the mouth opens. The ventral
sucker is very large, and its surface is subdivided into rectan-
gular areas. There is no perivisceral cavitj'-, its place being
occupied by a mass of spongy cellular tissue. The oral cavity
leads into an oval, thick-walled, muscular pharyngeal bulb.

ASPIDOGASTER COXCHICOLA.

173

"whence an elongated pyriform sac, which constitutes the rest
of the alimentary canal, is continued. This occupies a great
part of the body, and extends nearly to its posterior end ; but
there is no anus. A contractile vacuole placed at the hinder
extremity of the body opens outward by a small pore (Fig.
41, a), and gives off two lateral contractile non-ciliated canals
(b), which pass to the anterior end of the central sucker and
there end blindly ; but before reaching this termination each
gives off a non-contractile ciliated vessel (Fig. 41, c), which,
on arriving at the pharynx, turns backward and ramifies
through the body. The cilia diminish toward the extremi-
ties of these vessels, the terminations of the corresponding
canals in the Motife^'a being, on the contrary, richly ciliated.
No nerves have as yet been found in Asjndogaster,

Fio. ^(i.—Aspidogaster concMcola.—A, arransement of the alimentary and reproduc-
tive organs ; profile of the animal in outline : a, mouth ; 6. mufecular pharynx ; c,
stomach; rf, germarium ; e, internal vas deferens;/, common vitellarian duct ;
<7, vitellarium ; ;^,one of itp ducts ; i, k, oviduct ; L uterns; m, testis ; o. vagina;
p, penis, continuous posteriorly with the external vas deferens ; B, one of the
lateral contractile vessels ; (7, ramifications of the ciliated vessels.

As in most Trematoda, the genitalia (Figs. 40 and 42)
form a large part of the viscera, and the structure of the com-
plex hermaphrodite apparatus is in some respects so peculiar
that it is needful to describe it in detail. It consists of —
1. The germarium. 2. The vitellarium. 3. The oviduct.
4. The uterus and vagina. 5. The common vestibule. 6. The
testi?. 7. The vasa deferentia, internal and external. 8. The
penis and its sac. The ovary {cT) is the anterior of two round-

174 THE ANATOMY OF INVERTEBRATED ANIMALS.

ed masses lying* in the sucker. At first sight it appears to be
oval, but it is, in fact, pyriform, the larger end being anterior,
while the posterior narrower extremity is bent backward be-

FiG. 41. — A, water- vascular system of Aspidogaster conchicola: a, terminal pore;
6, lateral contractile vessels ; 6% lateral ciliated trunks, that of the left side shaded :
d, dilatation of this trunk ; B, one of larger, and C, one of the smaller, ciliated
vessels.

neath the anterior end. Before it reaches the anterior ex-
tremity of the mass, however, it is bent sharply back again,
parallel with itself, and so passes into the oviduct (Fig. 40, ^).
The ovary is surrounded by a delicate, but strong coat, inclos-
ing a mass of transparent protoplasm. At the anterior end
of the ovary minute granules are scattered through this sub-
stance, and are occasionally surrounded by a faint, clear area
(Fig. 43, A 1). These are the rudimentary germinal spots
and vesicles of the future ova, the course of whose develop-
ment may be readily traced by working from the anterior to
the posterior extremity of the ovary. The germinal spots
become larger, and gradually assume the appearance of vesic-
ular nuclei ; while the clear area around them in like manner
becomes larger, and acquires more and more the appearance
of a cavity. While this cavity is small, it has no distinct
wall, but, as it enlarges, the contour of the wall becomes dis-
tinctly marked (Fig. 43, A 2, 3, 4). On examining the ovary
close to the commencement of the oviduct, a division of the
homogeneous protoplasmic basis or matrix of the ovary into
areas surrounding each germinal vesicle becomes obvious. On
the application of pressure, the matrix breaks up into masses
corresponding with these areas in size, which are very flexible,
but when left to themselves assume a rounded or oval form,
and have all the appearance of perfect ova, except that they
possess no vitelline membrane, and that the yelk, instead of
being granular, is clear, and comparatively small. These

ASPIDOGASTER CONCHICOLA.

175

'primary ova^ as they may be termed, become detached, and
pass into the oviduct. Here they are fecundated, and, be-
coming surrounded by a great mass of accessory yelk, and a
shell, gradually acquire the appearance of the complete ova.

The accessory yelk is the product of the vitellarium — a
large double gland consisting of a number of oval, pyriform,
or irregular granular masses placed on each side, at the junc-
tion of the sucker with the body (Fig. 40, g).

These masses appear to be quite independent of one an-
other; nor do they at first present any obvious communication
with the genitalia ; but if the oviduct, just after it becomes
free from tlie ovarium, be examined, it will be found to re-
ceive a short duct (Fig. 42,/"), filled with strongly retracting
granules of the same nature as those in the vitellarium. This
duct is enlarged posteriorly, and then divides into two ducts
filled with the same matter, which take a direction toward the
vitellarium, but can be traced no further than they contain
granules (Fig. 42). By the careful application of pressure,
however, the granules may be forced from the vitellarium,
through an anterior and posterior branch upon each side, into
these ducts.

Fig. 42. — Aspidogaster concMcnla. — Reproductive onrans on a larger scale. Letters
as in Fis:. 40. The commeucemeut of the external vas deferena'ia seen behind the
vitellarian ducts.

The oviduct (Fig. 42, %) is richly ciliated internally ; it is
at first applied to the under surface of the ovarium, and when
it becomes free it receives a canal (e), which may be traced

176 THE ANATOMY OF IXYERTEBRATED ANIMALS.

back to the testis, and which would appear to correspond
with the internal vas deferens of other Trematoda described
by Von Siebold/ This canal, however, presents no dilatation,
or internal vesicula seminalis. The oviduct next receives the
duct of the vitellarium, and then becoming much convoluted
(y^), and rapidly widening, passes into the uterus (/), a wide
tube, which runs forward, disposed in many undulating curves
(Fig. 40, ^), to terminate on the left side of the anterior part
of the body, close to the male organs. Posteriorly, the walls
of the uterus are thin ; but in its anterior, or vaginal, part
they become thick and muscular. The genital vestibule into
which the vagina opens is very small.

The testis (m) is an oval body of the same size as the
ovarium, and situated just behind it. Minute water-vessels
ramify upon it, as upon the ovarium ; and it contains a gran-
ular and cellular mass, but no spermatozoa. The external
vas deferens (Figs. 40 and 42) is a delicate duct, which
passes forward and comes into contact with the ovarium,
without, however, so far as I could observe, communicating
u'ith it or with the oviduct ; it then bends backward and up-
ward, passing between the anterior vitellarian masses into
the fore part of the body. Here it suddenly becomes about
twice as wide as before, and runs forward, as an undulating
thick tube, to the penis (Fig. 40, jo), a short and conical body,
occupying the bottom of a large pyriform sac, which opens
in common with the uterus. The spermatozoa are linear.

The development of the ova presents many very interest-
ing peculiarities (Fig. 43). Above the junction of the duct
of the vitellarium with the oviduct the contents of the latter
were pale and clear, and presented no formed particles beside
the primary ova which had just been detached from the ova-
rium (Fig, 43, C). Below the insertion of the vitellarian
duct, however, the oviduct was full of granules like those in
the vitellarium, mixed up with ova in a more advanced state.
In the smallest of these (Fig. 43, D), the shell of the ovum
had commenced, but was incomplete at one end. At the op-
posite extremity, it inclosed a mass of irregularly aggregated
vitelline granules, which covered almost one-half of a round
pale mass, not larger than one of the primary ova ; in which,
however, three nuclei (two of which were very close together,

^ The connection of this duct with the testis \\\ the Trematoda has recently
been denied by Stieda (" Miiller's Arcliiv," 1871). I had no doubt of its exist-
ence in Aspidogaster^ but I have had no opportunity of reexamining this ani-
mal since the publication of Stieda' s paper.

THE DEVELOPMENT OF ASPIDOGASTER.

177

as if they had just divided) were to be distinguished. In
more advanced ova the shell was complete, but either color-
less or of a very pale-brown hue. In some of these the pri-
mary ova contained many nuclei and were imbedded in and
surrounded by a confused mass of accessory yelk-granules ;
while in others these granules were aggregated into a num-
ber of regular spheroidal masses (Fig. 43, £).

As dev:elopment proceeds, the accessory j'elk-masses grad-
ually disappear ; the primitive ovum, now become the homo-
losrue of the blastodermic disk or vesicle in other animals, to
all appearance increasing at their expense. At the same
time, clear rounded vacuoles in various numbers appear in its
substance ; but the nuclei of the germ, though very minute,
can, with proper care, be readily detected between these. In
the final stages the shell becomes browner, the vacuoles and
granules disappear, and the substance of the embryo appears
homogeneous. Bat, if carefully examined, the minute nuclei
become visible, especially if water be allowed to act on the

¥iG. i^.—A^pidogaster concMcola.—A, section of the ovary: 1, its anterior end: 2,
germinal spot surrounded by a distinct wall ; 3. 4, a complete germinal vesicle
and spot ; C, a primary ovum ; Z), voune: state of a complete ovum ; the primary
ovum partially surrounded by yelk-granules and a shell ; B, complete ovum, with
the accessory yelk aggregated into spheroids ; E, vacuolated embryonic mass ; F,
embryo.

tissue, and, if the shell be burst, and its contents poured out,
they readily break up into small but well-marked cells, each
with its nucleus. ^ At the same time, the embryo takes on a
form not very distantly resembling that possessed by the

178 THE ANATOMY OF INVERTEBRATED ANIMALS.

adult; into which it eventually passes without any metamor-
phosis.^

Thus it appears that, in Aspidogaster, the ovarium gives
rise to primary ova, which pass down the oviduct and become
fecundated, either by the spermatozoa conveyed by the inter-
nal vas deferens, or by those received by the vagina when
copulation with another individual, or, possibly, self-impreg-
nation, occurs ; that, next, the essential part of the process of
"yelk-division" takes place, the germinal spot dividing and
subdividing, and the primary ovum becoming in this way con-
verted into the spheroidal blastoderm ; that, contemporane-
ously, the blastoderm becomes invested by the accessory yelk-
granules poured in by the vitellarian duct, and by a shell ;
that the accessory yelk arranges itself into spheroidal masses,
which probably supply the blastoderm with the means of its
constant enlargement ; and that, finally, the accessory yelk
disappears, and the blastoderm becomes converted into the
embryo.

The modifications exhibited by other Trematoda concern
the number of the suckers, of which there are usually several
in the ectoparasites, but not more than one in the endopara-
sites ; their support on a chitinous framework, or the addition
to them of spines or booklets, similar to those of Cestoidea
or Acanthocephala : the bifurcation of the intestinal canal,
and the ramification of its branches, so that the forms of the
alimentary apparatus repeat the two extremes observed in
the aproctous Turhellaria ; the existence of two nervous
ganglia with a single transverse commissure in many ; and
the occasional presence of sensory organs (ej^e-spots). The
non-contractile canals of some genera are destitute of cilia,
except at their inner terminations.

The variations of the reproductive organs are rather of
position than of structure. Dioecious Trematodes are very
rare, the most important being the formidable Silha/rzia, the
male of which is the larger and retains the female in a gynm-
cophore^ or canal, which is formed by the infolding of the
margins of the concave side of the body. Bllharzia has
neither intromittent organ nor seminal pouch, and the history
of its development has not been traced beyond the escape of

1 The substance of this account of the structure and development of A'>pido-
aaster, with the illustrative fisrures, was published in IBS') in The Medical
'Times and (razeite. M. E. Van Beneden has recently thrown much light on the
mode in which the ova of the Trematoda are formed and developed, in his
"Eechcrches surla Composition et la Signification de l'(Euf."

THE DEVELOPMENT OF THE TREMATODA.

179

a ciliated embryo from the ovum. This parasite is found in
the blood-vessels of man, chiefly in those of the urinary or-
gans, the ova escaping from the body through the ulcerated
surfaces to which the parent gives rise. In the ectoparasites,

Fig. 44.— a. B, Mnnostomnm mntabile.—A. the ciliated emhryo (a) inclosinsr the
zooid, {b.) represented free in B (after Siebold) : C, Redia, or kine's yellow worm
of Distnma pacificum, containine grerms of other Redice ; D. Redia containine
CercaricE (a): E, Cercaria; F, Distoma, which results from the metamorphosis
of the Cercaria. (After Steenstrup.)

the embryo passes into a form identical with or closely resem-
bling that of the parent while still within the egg, as in As-
pidogaster. When this happens (e. g., Distoma variegatiim,
D. tereticolle), the one end of the embryo is often provided
with spines, and it is capable of slow creeping movements.
But, in most of the endoparasites, the embryo leaves the
parent as a morula, which is usually ciliated. Thus, in Disto-
ma lanceolatum^ D. hepaticimi, and 3Io7iosto7ni(m mutabiley
the embryo which escapes from the egg has a ciliated invCwSt-
ment, which propels it rapidly through the water, and may
be provided with eyespots and water-vessels (Fig. 44, A).
On becoming attached to the animal upon which it is parasit-
ic, the embryo of 3fonost07nu7n gives exit to a larva, having
the form of a cylindrical sac with two lateral prolongations
and a tapering tail. The Jiedia, as this form is called (Fig.
44, j&, (7), has a mouth and a simple c^ecal intestine, but no
other organs. In its cavity a process of internal gemmation
takes place, giving rise to bodies resembling the parent in
shape, but destitute of reproductive organs, and furnished

180 THE ANATOMY OF INYERTEBRATED ANIMALS.

with long tails, by which they are propelled. These creatures,
called Cercarice (Fig. 44, E)^ escape by bursting through the
Hedia, and, after a free-swimraing existence, penetrate the
body of some other animal, their tails dropping oif. They
then become encysted, and, under suitable conditions, assume
the adult form, and develop reproductive organs (Fig. 44, F).
The cycle of forms through which Distoma milltare passes
has been nearly completely traced, and may be briefly stated
as follows : 1. The parent form, whose habitat is the in-
testines of water-birds, bears on its anterior extremity two
alternating circles of larger and smaller booklets, and a few
others, irregularly disposed. Rings of papillie give the cen-
tre of the body an annulated aspect. The mouth, almost
terminal, leads into the lono*, straight diofestive csecum. The
generative organs are similar to those of Asjndogaster / the
testes are, however, double, and lack the internal vas deferens.
The ova are few, eight or ten in number. 2. From each
ovum issues a ciliated larva, showing the rudiments of — 3. A
Media, but the mode of development of the latter has not
been fully traced. The perfect Media is found attached to
the body of a water-snail (Maludi?ia), the ciliated investment
having disappeared. It consists of a sac, within which is
suspended a tubular bag, containing colored masses, probably
alimentary. Anteriorly, the head is represented by a kind of
crown, in which no oesophagus exists as yet, and not far from
the posterior extremity the two lateral projections, character-
istic of Distomatous Mediae, appear. Daring the rapid growth
of the zooid, the head becomes marked off by a constriction,
and a mouth and gullet, with a pharyngeal dilatation, admit
aliment to the digestive sac. In the body cavity, external to
this sac, vesicles appear, rapidly increase, and take the form
of Cercarice ; the Media bursts, and these new zooids are
set free. 4. The Cercaria has a long tail with lateral mem-
branous expansions, by means of which it swims after the
fashion of a tadpole. The pharyngeal bulb is followed by an
oesophagus, which, opposite the ventral sucker, divides ; the
two branches ending in a csecum on either side of the <;on-
tractile vacuoles of the water-vascular sj'stem. These are
median, the terminal quadrate chamber opening into an an-
terior circular one, whence are given off the two main canals
which traverse the body longitudinally, and are then lost. 5.
After swimming about freely for a while, the Cercaria fixes
itself upon, or bores its way into, a Paludina ; the tail drop-
ping off, and the body coating itself with a structureless cyst,

THE DEVELOPMENT OF THE TREMATODA.

181

in which it remains quiescent, but undergoes some further
advances in development, the coronal hooklets making their
appearance. 6. When a Paludina, thus infested, is swal-
lowed by a w^ater-bird and digested, the cj'sts are set free in
the alimentary canal of the bird ; sexual organs appear within
the included Distoma / the body elongates and narrows an-
teriorly ; the sucker moves nearer the head, and the coronal
circlets reach their full development. The Distoma gradually
assumes the form of the parent, attaches itself by its hooklets
to the intestinal walls, and acquires complete sexual organs.*
Thus the developmental stages of Distoma militare may be
summed up, as : 1. Ciliated larva. 2. liedia. 3. Cercaria,

4. Cercaria^ tailless and encysted, or incomplete Distoma,

5. Perfect Distoma,

The stages of transition vary in different genera. Thus,
several generations of Hedioe may intervene between the

Fig. 45.— Bucephalus polymorphus of the fresh-water muscle.— J, ramified sporocyst ;
^, portion of the same more man^nified: a, outer coar, 6, inner; c, <?, germ-
masses in course of development ; C, one of the germ-masses more highly mag-
nified ; i?, Bucephalus ; a, b, suckers ; c, clear cavity ; d, caudal appendages.

third and fourth stages ; or the mature animal may appear at
the close of this stage, having undergone no Cercarian meta-
morphosis.

In Bucephalus polym^orphus^ a parasite of the fresh-
water muscle (Fig. 45), two caudal appendages, which seem
to correspond with the tail of the ordinary Cercarice^ become

1 Van Beneden, " Memoire sur les Vers Intestinaux.'

182 THE ANATOMY OF INVERTEBRATED ANIMALS.

enormously elongated. They are converted into ramified
tubes called sporocysts^ "which sometimes occupy all the inter-
spaces of the viscera of the muscle. These develop new
l^ucejyhali by internal gemmation. The Trematode condition
appears to be the genus Gasterostomum^ which inhabits fresh-
water fishes.

The Sporocysts, Redias, and Cercariae, free or encysted,
are found almost exclusively in invertebrated animals, while
the corresponding adult Trematodes are met with in the verte-
brated animals which prey upon these Invertehrata.

The singular double-bodied Diplozoon ijaradoxum has
been shown by Von Siebold to result from a sort of conjuga-
tion between two individuals of a Trematode, which, in the
separate state, has been named Diporpa. The DiporpxB^
when they leave the eg^^ are ciliated and provided with two
eye-spots, with a small ventral sucker and a dorsal papilla.
After a time the Diporpm approach, each applies its ventral
sucker to the dorsal papilla of the other, and the coadapted
parts of their bodies coalesce. They acquire fully developed
sexual organs only this after union. ^

Gyrodactylus multiplies agamically by the development
of a young Trematode within the body, as a sort of internal
bud. A second generation appears within the first, and even
a third within the second, before the young Gyrodactylus is
born.

The Cestoidea. — The Tape-worms are all endoparasites,
and, in their adult condition, infest the intestines of verte-
brated animals.

The simplest form known is Caryophyllaeus^ found in
fishes of the Carp tribe. It has a slightly elongated body,
dilated and lobed at one end, so as to resemble a clove,
whence the name of the genus. In structure it resembles a
Trematode, devoid of any trace of an alimentary canal, but
provided with the characteristic water-vascular system and
with a single set of hermaphrodite reproductive organs.

In lAqida^ the body is much elongated, and, at the head-
end, exhibits two lateral depressions. It is not divided into
segments, but there are numerous sets of sexual organs ar-

1 Zeller, " TJntersuchun^en ilber die Entwickelung des Diplozoon paradox-
um." {Zeitschrift fiirwiss. Zoologie, 1872.)

2 See the " Meraoire surles Vers tntestinaux," 1858, by M. P. J. Van Beneden,
to which I am much indebted for information respecting this and other genera
of Cestoidea which have not fallen under my own observation. Also Leuckart,
*' Die menschlichen Parasiten," 1863 ; andCobbold, *' Entozoa."

THE CESTOIDEA.

183

ranged in longitudinal series. The openings of the genital
glands are situated in the middle line of the body. These
parasites inhabit fishes and amphibians, as well as water-
birds, but they attain their sexual state only in the latter.

Fig. 46 —Diagram of the structure of a cestoid worm, with only one joint. The posi-
tion of the liooks of a Tainia and of one of the proboscides of a letrarhynchua
is inoicated. -4, head and neck; 5, segment of the body corresponding with a
proglottis: a. rosteUum; 6, rostella spines {ToBnia); c, </, </', spinose eversible
proboscis {Titrarhynchus) ; rf, sucker; e, ganglion (?); /, lateral, and g, circular
water-ves;?el ; A, ramifications of the water-vessels ; k, anastomosing trunk ; *,
contractile vacuole ; /, genital vestibule ; m, penis and vas deferens ; «, vagina ;
o, common cavity ani vesictiCa seminalis interior ; p, ovary; q, uterus ; r, vitel-
larian duct.

In the more typical Cestoidea the body is elongated, and
presents, at one end, a head provided with suckers, and very
generally with chitinous hooks, either disposed circularly
around the summit of the head, or upon proboscidiform ten-
tacles, which can be retracted into, or protruded from, the
head. Sometimes the head is produced into lobes ; and very
generally, when lobes or tentacles exist, they are four in
number, and are disposed symmetricall}^ round the head. A
short distance bej^ond the latter, the slender body widens and
becomes transversely grooved, so as to be marked out into
segments. Longitudinal water-vessels run parallel with one
another through the body, and are connected by transverse
trunks in each segment, and by a circular vessel in the head.
In Bothriocephalus latus, the principal trunks are occupied
by a spongy reticulated tissue.

In most of the tape-worms, innumerable, solid, strongly-

184

THE ANATOMY OF INVERTEBRATED ANIMALS.

refracting corpuscles are scattered through the substance of
the body (Fig. 48, A). It is probable that these are more or
less calcified connective-tissue corpuscles. Similar bodies
which occur in some Trematoda were found by Claparede to
be lodged in dilated ends of the water-vessels, but it would
appear that they are not so situated in the Gestoidea.^

The distance between these transverse grooves, and their
depth, increase toward the hinder end of the body ; and each
segment is eventually found to contain a set of male and
female organs. The genital organs are constructed upon the
same general plan as those of the Trematoda, but the uterus,
as it fills with ova, usually takes the form of a ramified sac.
At the extreme end of the body, the segments become de-
tached, and may for some time retain an independent vitality.
In this condition each segment is termed a proglottis ^ and
its uterus is full of ova.

The embryo is developed in these ova in the same way as
in the Trematoda ; and, as in the latter group, it may either
be ciliated (as in JBothriocephalus) or non-ciliated, which last
is the more usual case. The embryo is a solid morula, on one
face of which four or six chitinous hooks, disposed symmet-
rically on either side of a median line, are developed.

Fig. 47.— Diagrams illustrative of the relation hetween Tcvnia, Cysticercus, Coe,nnrus^
and Echinocoecus.—A, J5, youn? Tcenim in the Scolex stajje, the latter with an
en\?^xze'^receptactilum SGoUcis,\ntowh.\ch the head and neck are withdrawn in
C. Cusficercus ; 7), C(»nurus ; E. hypothetical condition of Echinococcm, in which
" Trpnia heads" are developed only on the inner surface of the primary cysts; S^t
Echinococcus with secondary cysts; G, embryo Taenia (after Stein).

If the egg is placed in appropriate conditions, the hooked
embryo emerges from the shell, and rapidly increases in size.

» Sommer and Landois, " Ueber den Bau der ^eschlechtsreifen Glieder
von Bothriocfiphalys latus." (Zeitschrift fur wiss. Zoolocfie. 1872). Leuckart,
however, maintains the contrary opinion, " Die menschlichen Parasiten," p.
175.

THE CESTOIDEA.

185

After a time, a cavity appears in the midst of the cells of
which the morula is composed, and a chitinous cuticula is
developed upon the outer surface of the embryo. Ramified
water-vessels make their appearance in the wall of the sphe-
roidal sac thus formed, and in some cases open by an external
pore. There is, therefore, a very close resemblance between
this cestoid embryo and the sporocyst of a Trematode.

When the saccular embryo has attained a certain size, a
thickening and invagination take place, usually at one {I'm-
Olid), sometimes at many {Coenurus, Echinococcus) points of
its wall. The invagination of tlie wall elongates inward, and
becomes a caecum, the cavity of which opens outward. At
the bottom of the interior of this caecum, and therefore on
what is morphologically its external surface, the hooks of
those species which possess them are developed, while, upon

7<i^^.-..

IF

Fig. 4S.—Echinococcifs veterinoritm.—A, "Taenia head," or Scolex : a, hooks; ft,
puckers; c, cilia in water-vessels; d, oval, stronsrly refracting particles; B, single
hooks; C, portion of the elastic cyst, a; with the inner membranous primary
cyst, b; c and e, ScoUces developing from its inner surface; d, a secondary cyst.

the side-walls, elevations arise, which become converted into
suckers. The ca3cum is next evaginated or turned inside

186 THE ANATOMY OF INYERTEBRATED ANIMALS.

out, and tlie embryo has the form of a phial, of which the
evaginated caecum forms the neck. Round its apex are the
hooks, and below these the suckers, forming a complete ces-
toid head ; while the sac answers to the body of the phial.
The original hooks of the embryo are cast off in the course
of the process.

If the eggs of the Tape-worm have passed into the aliment-
ary canal of an animal in which the worm is unable to attain
its sexual condition, the hooked embryo, as soon as it is
hatched, bores its way through the walls of the alimentary
canal, and eventually becomes lodged in the connective tissue
batween the muscles, or in the liver, or in the brain or eye.
Here it oroes throusrh the chanofes which have been described,
and, generally, the sac undergoes very great dilatation. The
rearion of the wall of the sac to which the cestoid head is at-

CD

tached becomes invaginated, and thus is inclosed within a
chamber, the parietes of which are really constituted by the
outside of its own body. In this condition, the animal is
what is termed a Cystic worm, or bladder-worm ; and when
there is only one head it is a Cystlcercus, In the genera
Goenurus and EMnococcus the cystic worm has many heads;
and, in Eshmococcus, the structure of the cystic worm is still
further complicated by its proliferation, the result of which is
the formation of many bladder-worms inclosed one within
the other, and contained in a strong laminated sac or cyst, ap-
parently of a chitinous nature, secreted by the parasite (Fig. '

In the cystic condition, the Tape-worms never acquire
sexual organs ; but, if transported into the alimentary canal
of their appropriate hosts, the heads become detached from
the cysts, and, rapidly growing, give rise to segments, which
become sexual proglottides. The Tape-worms are rarely met
with in both the cystic and cestoid conditions in the same
animal ; but the cystic form is found in some creature which
serves as prey to the animal in which the cestoid form occurs.
Thus :

Cystic Form. Cestoid Form.

Cystkercus cellulosm. Tarnia solium.

(Muscles of the Pig) (Man)

Cysticercm ? Taenia mediocanellata.

(Muscles of the Ox) (Man)

Cysticerms plnformis. Tmnia serrata.

(Liver of the Rabbit) (Dog, Fox)

THE DEVELOPMEXT OF THE CESTOIDEA. 187

Cystic Form:. Cestoid Form.

Cysticercus fasciolaris. Tcenia crassicollis,

(Liver of Rats and Mice) (Cat)

Canurus cerebi'alis. Tcania ccenurus.

(Sheep's brain) (Dog)

JEchinococcus veterinorum. Tcenia Echinococcus.

(Liver of Man and of (Dog)
domestic Ungulata)

The embryo of loenia cuciimeri7ia passes, in the body of
the Dog-louse {Trichodectes cmiis), into a Cysticercoid^ or
minute unjointed and sexless Tsenia, without any terminal
dilatation. The dog devours the louse and the Cysticercoid
becomes a Tcenia cucumerlna in his intestine. The eggs of
the Tcenia^ contained in fieces adherent to the hair of the
dog, are in turn devoured by the louse, and thus the " vicious
circle" of parasitism is maintained.

The cystic Tetraphyllidea frequent osseous fishes, their
sexual maturity being attained in the bodies of Plagiostomes.
The head is provided with four suckers or lobes, which may
be stalked and unarmed, as in Echeneibothrium^ or furnished
with hooklets as in Acanthohothrhim ; while, in Tetrarhyn-
chiis^ four proboscidiform tentacles, thickly set with booklets,
are retracted into sheaths alongside of the suckers (Fig. 46).

The Diphyllidea have two suctorial disks, two armed
rostellar prominences, and a collar of hooklets on the neck.

The migrations of the JPseudophyllidea are chiefly from
fishes and amphibians to water-birds, one genus {Bothrio-
cephalus) containing species which enter the human body, prob-
ably in the flesh of fresh-water fishes. The head has neither
suckers nor lobes, but is deeply grooved on either side. In
JBothriocephalus the genital apertures are in the middle of
each segment. The embryo is ciliated, and swims actively in
water. Recent experiments tend to show that the develop-
ment of the embryo in this genus may take place directly, or
without the intervention of a Cysticercus stage.

It is obvious that the Cestoidea are very closely related to
the Trematoda. In fact, inasmuch as some of the latter are
anenterous, and some of the former are not segmented, it is
impossible to draw anv absolute line of demarkation between
the two groups. It would appear that the Cestoidea are
either Trematodes which have undergone retrogressive met-
amorphosis and have lost the alimentary canal which they
primitively possessed, or that they are modifications of a

183 THE ANATOMY OF INVERTEBRATED AXIMALS.

Trematode type, in which the endoderm has got no further
than the spongy condition which it exhibits in Convoluta
among the Turbellaria, and in which no oral aperture has
been formed ; or, lastly, it is possible that the central cavity
of the body of the embryo Taenia simply represents a blas-
tocoele.

If the Cestoidea are essentially Trematodes, modified by
the loss of their digestive organs, some trace of the digestive
apparatus ought to be discoverable in the embryo tape-worm.
Nevertheless, nothing of the kind is discernible, unless the
cavity of the saccular embryo is an enterocoele. And if this
cavit}'' is a blastoccele, and not an enterocoele, it may become
a question whether the tape-worms are anything but gigantic
morul®, so to speak, which have never passed through the
gastrula stage.

CHAPTER V.

THE HIRTJDIXEA, THE OLIGOCH^TA, THE POLTCH^TA, THE

GEPHYKEA.

The Hirudinea. — -The Leeches are aquatic or terrestrial,
more or less distinctly segmented, vermiform animals, most
of which suck blood, though some devour their prey. The
ectoderm is a cellular layer, covered externally by a chitinous
cuticula, and, except in Malacohdella^ devoid of cilia. Very
commonly it is marked by transverse constrictions into rings,
■which are more numerous than the true somites as indicated
by the ganglia and the segmental organs ; and simple glands
may open upon its surface. One or more suckers, which
serve as organs of adhesion, are developed upon it. In some
{Acanthohdelli) bundhs of setae are present ; in others [Brayi-
chelllon) the sides of the body are produced into lobe-like
appendages ; but none have true limbs, unless the lateral ap-
pendages of Histriohdella are to be considered as such ; nor
are the anterior segments of the body so modified as to give
rise to a distinct head.

The mouth is generally situated at the anterior end of the
body ; the anus at the opposite extremity, on the dorsal side
of the terminal sucker. The buccal cavity may be armed
with several serrated chitinous plates, as in the Medicinal
Leech, where there are three such teeth. By their aid the
Leech incises the skin and gives rise to the well-known tri-
ardiate mark of a leech-bite. The buccal cavity usually
opens into a muscular, sometimes protrusible, pharynx, from
which a narrow oesophagus leads into a stomach, which is fre-
quently produced into lateral coeca. In the Medicinal Leech
(Fig. 49), for example, there are eleven pairs of such creca,
increasing in length and capacity from before backward.
From the stomach a narrow intestine leads to the anus. In

190

THE ANATOMY OF INVERTEBRATED ANIMALS.

ft.

r-^

'm

t-M

\

<N^?

3falacohdella the alimentary canal is a sim-
ple tube bent several times upon itself.
The alimentary canal is lined by the cells
of the endoderm, and the space between
them and the ectoderm is occupied by the
mesoderm, which contains abundant con-
nective and muscular elements, and is ex-
cavated by the blood-channels, which some-
times have the form of wide sinuses, but
in other cases are comparatively narrow
vessels with definite walls.

In the lower Hirudinea^ as Clepsine,
the sinuses and vessels appear to form one
continuous system of cavities containing a
fluid which must be regarded as blood. But
in the Leech a distinct pseud-haemal vascu-
lar system has attained a great degree of
definition and complexity : it consists of
(1) a median dorsal trunk ; (2) a median
ventral trunk, in which the ganglionic nerve-
chain lies ; (3, 4) two wide lateral longitu-
dinal trunks (Fig. 50). These anastomose
with one another, and give off numerous
branches, which open into a rich capillary
network, situated in the muscular layer of
the mesoderm, and on the segmental and
reproductive organs. The fluid contained
within these vessels has a red color, and
contains no corpuscles.

More or fewer of the segments of the
body are provided with what are termed
segmental organs. These are tubes which
open externally on the ventral wall of the
body, while at their other extremities they
either open into the sinuses by ciliated
mouths (Clepsine)^ or form a closed and
more or less reticulated non-ciliated coil
{Hirudo). These obviously answer to the
ciliated water-vessels of the Turhellaria
and Trematoda.

The nervous system consists of a cerebral mass in front of
the mouth, proceeding from which, on each side, is a commis-

1 " Die mensclilichen Parasiten," 1863.

THE HIRUDINEA.

191

sure connecting it with a ventral cord on which ganglia, cor-
responding in number with the somites of the body, are de-

~" *^

TrV-— =

Fig. 50.— a diasrammatic view of the arransrement of the principal vessele of the
leech {Hirudo medicinalia). after Gratiolet. The inner surface of a portion of
one-half of the body is depicted: a, a, the ventral trunk; e, e', c", the lateral
trunk and its branches ; /, /', the dorsal trunk and its branches ; g, the slender
transverse trunks which branch out at each end; A, i, the transverse ventrai
branches of the lateral trunk ; k, I, the branch to the testis (e). and the segmental
organ (d) ; m, branch from the dilatation on the testis to the parietal plexuses;
b, b, vas deferens.

veloped. In Malacohdella, these cords are lateral and wide
apart, but, in all the other Hirudinea^ they come close to-
gether behind the mouth, and occupy the middle line of the
ventral face of the body. In the Leech, according to Leuck-
art, there are originally thirty pairs of post-oral ganglia, but
the seven posterior and the three anterior pairs coalesce, so
that only twenty-three pairs are distinguishable in the adult.
Nerves are given off to the pharynx and intestines, and the
former develop special ganglia.

Simple eyes are usually present on the anterior or oral
segment, and receive nerves from the supraoesophaireal gan-
glia. In the Leech these eyes are situated in the first three
segments. Cup-shaped depressions of the integument of the
anterior segments of the body, lined by peculiar glassy cells

192 THE ANATOMY OF INVERTEBRATED ANIMALS.

and in relation with nerves which terminate in fine filaments,
have been discovered by Leydig in several of the Slrudinea,^

The elongated spindle-shaped muscle-cells of thel)ody are
abundant, and are disposed in a superficial circular and deep
longitudinal layer, while dorso-ventral bands pass from the
dorsal to the opposite body-wall.

Malacohdella and Histriohdella are dioecious, but the other
Ulrudlnea are hermaphrodite. The male organs consist of
numerous testicular sacs, situated on each side of the body,
and connected by a vas deferens, which usually opens into
a sac, terminating in an eversible penis. The spermatozoa
are often inclosed in a case or sperrnatophore. The female
organs, much smaller than the male, consist of ovaries, with
oviducts opening into a vagina. The vaginal orifice is behind
that of the penis. In the Leech the eggs are inclosed in a
sort of cocoon, formed by a viscid secretion of the integu-
ment.

The observations of Rathke and Leuckart on the develop-
ment of N^ephelis^ Clepsbie^ and Hlrudo show that, after the
division of the vitellus into a few equal-sized large blasto-
meres, small blastomeres are separated from the large ones (as
in the Ctenophora and Polycelis)^ and the rapidly-multiply-
ing small blastomeres form an investment to the slowly-divid-
ing large ones. This investment is the epiblast, and becomes
the ectoderm, while the included larger blastomeres are event-
ually converted into the cells of the endoderm. At one end
of the body the oral aperture appears, in some cases (e. g.,
Nephelis) surrounded by a raised lip, as in the embryo Pla-
narian ; and the embryo passes into the Gastrula stage. The
body now elongates, and, on the ventral face, the mesoblast
makes its appearance as a layer of cells, sometimes divided
into two longitudinal bands, separated by a median interval.
Three pairs of segmental organs, which have only a tempo-
rary existence and have been regarded as primordial kidneys,
are developed at the posterior end of the body. The meso-
blast next becomes divided transversely into the number of
somites of which the body is eventually composed, the divis-
ion first making its appearance on the ventral face of the
body. A pair of ganglia, probably derived from the epiblast,
is developed in each segment.

Thus, in the Leeches, the segmentation of the body is the
result of the segmentation of the mesoblast, which becomes

i " Arcliiv fiir Anatomie und Physiologie," 1861.

THE OLIGOCHJETA. I93

the mesoderm of the adult. And it is this seg^mentation of
the mesoblast, and consequently of the mesoderm, which con-
stitutes the most important difference between the Leech on
the one hand and the Turbellarian and Trematode on the
other.

On the other hand, in the development of a mesoblast
which undergoes division into segments, the Leeches exhibit
the fundamental character of all such seo^mented Invertebrates
as the chsetophorous A?inelkla and the Arthropoda,

The Oligoch^ta. — The earthworm (Lumhriciis) and
fresh-water worms (JVais, Tubifex^ Chcetogaster)^ which are
included under this name, are closely allied w^ith the Leeches
in the essential points of their structure and development,
much as they differ from them in habit and appearance.

They have elongated, rounded, segmented bodies, often
divided by many superficial transverse constrictions into
rings, which, as in the leeches, may be more numerous than
the proper somites. There are no limbs, but each segment
is usually provided with two or four sets of longer or shorter
chitinous seta?, which are developed and lodged in integument-
ary sacs. The outermost layer of the ectoderm is a non-cili-
ated chitinous cuticle.

The mouth is situated close to the anterior end of the
body, but a " cephalic lobe " not unfrequently projects be-
yond it on the dorsal side. The anus is at the opposite ex-
tremity of the body, and the straight alimentary tract which
connects the two and is lined by the endoderm is usually
divided into a pharyngeal, oesophageal, and gastro-intestinal
portion, the latter often being produced laterally into short
cjBca. The mesoderm presents well-developed transverse,
longitudinal, and dorso-ventral muscular fibres, as in the
Leeches. It is excavated by a spacious perivisceral cavity,
which contains a colorless corpusculated fluid, and is divided
by thin but muscular mesenteries, which stretch from the in-
testine to the parietes, and thus break up the perivisceral
cavity into partially separate chambers. In addition, there
is a system of pseud-hremal vessels, like those of the Leeches,
provided with contractile walls, and containing a red non-
corpusculated fluid. No communication has been ascertained
to exist between these vessels and the perivisceral cavity ;
but there can be little doubt that, as in the case of the
Leeches, they must be regarded as a specially differentiated
part of the general system of the perivisceral cavity.

y

194 THE AXATOMY OF INVERTEBRATED ANIMALS.

In the majority of the segments there are, as in the Si-
rudinea^ paired segmental organs ; these are ciliated and
their inner ends open into the perivisceral chamber.

The nervous system consists of prae-oral or cerebral gan-
glia, continued backward, on the ventral aspect of the body,
by commissures on each side of the oesophagus into a double
chain of closely united post-oral ganglia.

Large tubular fibres are imbedded in the neurilemma of
the ganglionic chain on its dorsal face. In the earthworm
there are three of these — one median and two lateral — ex-
tending along the whole length of the ventral end, but not
into the oesophageal commissures.^ The nature of these
structures is unknown.

These animals are hermaphrodite. The generative organs
are situated in the front part of the body, the male organs
being anterior to the female. In the aquatic Oligochceta
(N^ais, Tnbifex) the genital glands have no proper ducts, but
the segmental organs of the segments in which they are con-
tained convey the generative products outward. In the ter-
ricolous forms {Lumhricus) the vasa deferentia are continuous
with the testes, which are very large. The ovaries, on the
other hand, are minute solid bodies attached to one of the
mesenteries, and the oviducts are separate tubes with funnel-
shaped mouths, w^hich open into the cavity of the segment.

In iVa?5 and Chcjetogaster^ agamic multiplication occurs
by the development of posterior segments of the body into
zo5ids, Avhich may remain associated in chains for some time,
but eventually become detached and assume the parental
form. Schulze has observed that when a JSTais has divided
into an anterior and posterior zoOid, the last somite of the
former gradually enlarges, and becomes divided into new
somites, the anterior of which give rise to a head. A new
zouid is thus developed between the previously existing ones.
This process is repeated in what was the penultimate, but is
now the ultimate somite of the anterior zooid ; and again in
the penultimate somite when it has, in tlie same way, become
terminal.

As the Earthworm is a very accessible subject, it may be
useful to the student to be furnished with an account of some
of the chief points of its organization more in detail.

The exterior of the body of an Earthworm (Lumhricus
terrestriSf rubellus, or communis) shows a number of close-set

» Claparede, " Histologische Untersuchungen uber den Kegenwurm," 1869.

THE STRUCTURE OF THE EARTHWORM. 195

transverse grooves which divide its body into numerous nar-
row rings or segments/ The most anterior segment is small
and conical, and presents, on its under surface, a depression
which is the oral aperture. The anus is at the opposite end
of the body. Behind the mouth, the successive segments rap-
idly attain their average size ; but, in a full-grown worm, a
part of the bod}'', into which more or fewer of the segments
between the twenty-fourth and thirty-sixth inclusively (29 ?-
36, L. terrestris ; 24-29 ?, L. ruhellus ; 26-32, L. communis)
enter, is swollen, of a different color from the rest, provided
with abundant cutaneous glands, and receives the name of
cingulum or clitellum.

In the dorsal median line there is a series of small aper-
tures or pores, one for each segment except the most an-
terior, which lead into the perivisceral cavity ; while upon
the ventral surface of the anterior part of the body the eight
apertures of the organs of generation are situated. Of these,
four, situated two on each side, between the ninth and tenth,
and the tenth and eleventh segments, are the openings of the
receptacula seminis. The openings of the two oviducts are
on the fourteenth segment ; those of the two vasa deferentia
on the fifteenth. Besides these, all the segments, except
some of the most anterior, exhibit a pair of minute openings
appertaining to the segmental organs ; and they are further
perforated by the four longitudinal double rows of setas,
which project slightly beyond the surface of the integument,
and offer a certain resistance when the worm is drawn from
tail to head througrh the fino^ers.

The body is invested in a thin and transparent but dense
cuticula, perforated by excessively minute vertical canals.
Within this lies a thicker layer, consisting of a reticulated
nucleated protoplasm, the meshes of which are filled with
a transparent gelatinous substance. This layer probably rep-
resents both the dermis and epidermis^ and has been termed
the hypodermis. Internal to it lies a thick layer of circular
muscular bands, in the interstices of which pigment-granules
occur ; and, still more internallj'", is a much thicker coat of
muscular fibres, which are disposed longitudinall3\

The cavity circumscribed by this longitudinally fibrous
muscular layer is lined by a kind of connective tissue. Cor-
responding with the divisions between every pair of segments

^ The question how far all these segments represent somites may be left
open. The history of the development of the Earthworm is in favor of their
beino: true somites.

X96 THE ANATOMY OP INVERTEBRATED ANIMALS.

(except in the most anterior part of the body), this connective
tissue is continued transversely toward the axis of the bod\^,
and passes into that which forms the wall of the intestine ;
while, on the ventral side, it forms an arch over the ventral
nervous cord and the vessels which accompany it. In the
interior of each of these mesenteric septa, radiating and circu-
lar muscular fibres are abundantly developed, and the former
are connected externally with the superficial layer of trans-
verse muscles.

The perivisceral cavity is thus divided into nearly as many
short chambers as there are segments; each chamber com-
municates with the exterior, directly by the dorsal pore and
indirectly through the segmental organs, while fluid may pass
from one to the other by the supra-neural archways.

The short and curved setae project much farther into the
interior of the body than they do on to its exterior. The free
apices of each pair are situated close together, while their
inner ends diverge from one another. Each is inclosed in a
sac in which it is developed, and to which the muscles, by
which it is protruded, are attached. There are eight setae to
each somite, one pair not far from the ventral median line on
each side ; and the other pair placed in the same transverse
line, but further outward.

The mouth leads into a muscular pharynx, with a com-
paratively small internal cavity, which reaches as far back
as the seventh segment. From this a narrow oesophagus is
continued as far back as the fifteenth or sixteenth segment ;
and presents three pairs of lateral glandular diverticula, which
contain a calcareous matter,^ in the region of the twelfth and
thirteenth segments. Posteriorly, the gullet opens into a
crop, which is succeeded, about the eighteenth segment, by a
thickened and muscular gizzard.

Upon this follows the intestine, wliich has the appearance
of a simple tube; but is in reality complicated by the invo-
lution of its wall, along the dorsal median line, into a thick
fold, which projects into the interior of the intestinal cavity,
and is the so-called typhlosole. The exterior of the intestine
and the cavity of the typhlosole present a coating of yellow-
ish-brown cells.

The segmental organs are greatly convoluted tubes, situ-

1 The nature of this substance has recently been rUscnssed by M. E. Perrier,
*'!^tude sur un trenre nouveau des Loinbriciens." ("Archives de Zoologie ex-
p^rimentale," 1873.)

THE STRUCTURE OF THE EARTHWORM. 197

ated one on each side of every segment except the first, and
attached to the posterior mesenteric septum of the segment.
Each canal communicates internally, by a wide funnel-shaped
ciliated aperture, with the perivisceral cavity, while extern al-
I3' it opens by a minute pore, which is usually close to the in-
ternal pair of setie/

A colorless fluid, containing colorless corpuscles, and an-
swering to the blood of other invertebrated animals, occu-
pies the perivisceral cavity ; but, in addition to this, there is
a deep-red fluid, devoid of corpuscles, which fills a very large-
ly developed system of pseud-hicmal vessels. These consist
of longitudinal and transverse principal trunks, and of very
numerous branches which proceed from them and ramify in
all parts of the body, except the cuticle and hypodernus.

The longitudinal trunks are three : one siqyra-intesthial,
which lies along the dorsal aspect of the alimentary canal;
one sicb-i?itestinal, which corresponds with this on the ven-
tral aspect of that canal ; and one suh-neiiral, which lies be-
neath the ganglionic cord.

The supra-intestinal and sub-intestinal vessels are con-
nected in the greater number of the segments by pairs of com-
missural transverse trunks, which embrace the intestine, and
give off numerous branches to it. The supra-intestinal and
sub-neural vessels give off transverse trunks into the mesenter-
ic septa, which branch out into the muscular layers, and some
of which anastomose so as to form a second set of transverse
communications. Moreover, the sub-neural trunk and the
sub-intestinal trunk respectively send branches to each seg-
mental organ, upon which they are distributed, and, anastomo-
sing, give rise to another series of communications between
the longitudinal trunks.

In the seven most anterior segments, the longitudinal
vessels break up into a network, and there are no distinct
transverse commissural vessels. Behind these, and in the
region of the generative apparatus, the commissural vessels
are greatly dilated, and form from five to eight pairs of so-
called hearts which are attached to the anterior faces of as
many mesenteries. These contract from the dorsal toward
the ventral side.

The nervous system consists of two cerebral ganglia
lodged above the pharynx in the third segment, and united

1 Ge^enhaur, " Ueberdie sogenannten Eespiratioasorganc des Eegenwurms."
{Zeitschrift fur wiss. Zoologie^ 1852.)

198 THE ANITOMY OF INVERTEBRATED ANIMALS.

by commissural cords with the anterior ganglia of the chain,
which extends through the whole length of the body on the
ventral wall of the perivisceral cavity.

There are no eyes, nor are any other organs of special
sense known.

The Earthworm is hermaphrodite. The testes are two
pairs of large sacs, each of the anterior pair being bilobed.
The testes of opposite sides are united in a common median
reservoir, situated in the tenth and eleventh segments, from
which, on each side, ducts take their origin. The two ducts
of the testes of the same side unite into a single vas deferens,
and these two vasa deferentia open externally on the ventral
aspect of the fifteenth segment. The ovaries are two minute
solid bodies, not more than y^g- of an inch long, attached to
the posterior face of the mesenteric septum which separates
the twelfth and thirteenth segments. They therefore lie in
the cavity of the latter. The oviducts are quite distinct from
the ovaries, and open internally by wide, funnel-shaped aper-
tures, situated in the cavity of the thirteenth segment. From
these funnel-shaped ends the oviducts are continued, as
slender tubes, through the mesenteric septum which separates
the thirteenth from the fourteenth segment, and open on the
ventral face of the latter.

Four globular spermathecce, or receptacles of the sper-
matozoa, are situated, two on each side, in the tenth and
eleventh segments, and open on the ventral face between
the ninth and tenth and the tenth and eleventh seo^ments
respectively. These are filled when copulation takes place,
during which process the two worms are said to be bound
together by a tough secretion of their clitella.

The development of the Ollgochceta has recently been
carefully investigated by Kowalewsky. The eggs of the
Earthworm are laid in chitinous cocoons cr cases, which are
probably secreted by the clitella. In addition to the eggs,
the cocoons inclose an albuminous fluid, and packets of sper-
matozoa. The vitellus is invested by a membrane, and con-
tains a germinal vesicle and spot. Complete yelk-division
takes place, and eventually the blastocoele becomes reduced
to a mere cleft. The blastomeres are disposed in two layers
— one consisting of small and the other of large blastomeres.
The embryo thus formed becomes concave on the side formed
by the large blastomeres, until it assumes the form of a sac,
ciliated externally, with an opening, the future mouth, at one
end ; the cavity of the sac being the primitive alimentary

THE POLYCH^TA. 199

canal, and the layer of large blastomeres, the hypoblast. Be-
tween the two, a mesoblastic layer appears, but the exact
manner of its origin is not known. On one face of the sac-
cular embryo the mesoblast becomes divided into a series of
quadrate masses, like the protovertebras of a vertebrate em-
bryo, disposed symmetrically on each side of a median line,
which corresponds with the future ventral median line of the
body. Along this line, the epiblast becomes thickened in-
wardj and the thickening is converted into the ganglionic
chain. At the same time, each quadrate mass of the meso-
blast is excavated by the development of a cavity in its in-
terior, whereby it becomes converted into a sort of sac. The
adjacent anterior and posterior walls of successive sacs unite,
and give rise to the mesenteric septa, while their cavities
beconie the chambers of the perivisceral cavity. The seg-
mental organs commence as cellular outgrowths from the
posterior face of each septum thus formed, and only subse-
quently become excavated and communicate with the exte-
rior.

The development of the Earthvrorm, therefore, closely re-
sembles that of the Hlrudiiiea^ and more especially that of
the Medicinal Leech, in which the digestive cavity of the
embryo would seem to be formed, as in the Earthworm, by a
process which is, in a sense, invagination. It would appear
that the first-formed aperture is the mouth ; while the anus
is a secondary perforation; and the segmentation of the body
commences in the mesoblast.

In the fresh-water Oligoehceta^ Euaxes and Tuhifex^ the
vitellus also becomes divided into large and small blastomeres.
The latter extend over the larger blastomeres, and form the
epiblast (= ectoderm). A mesoblast (= mesoderm), divided
into two broad longitudinal bands, is developed, and the oral
cavity is said to be formed by invagination of the epiblast
between the anterior ends of the two bands of the mesoblast.
In this case, the mouth in these genera is a secondary forma-
tion. The innermost layer of large blastomeres becomes the
hypoblast (= endoderm).*

The PoLYcn.ETA. — Except that the Polychrpta are almost
invariably dioecious and marine, while the OVgochcEta are
monoecious, and inhabitants either of land or fresh water, it

1 Kowalewskv, " Embrjologisclie Studien." (" Momoires de 1' Acaddmie de
St. P^tersbourg," 1S61.)

200 THE ANATOMY OF INVERTEBRATED ANIMALS.

is hard to say what absolute characters separate these two
groups. The lowest forms of the Polycha^ta^ such as Capi-
tella and Polyophthahnus^ might be regarded as marine dioe-
cious KaidcB. But, in the higher Folychoeta^ each segment
of the body develops lateral processes — the parapodia^ or
rudimentary limbs, which are usually provided with abundant
strong set« ; a distinct cephalic segment, the p>rcestomii(m^
appears in front of and above the mouth, and bears eyes and
tentacles ; while those parapodia which lie in the vicinity of
the mouth may be specially modified in form and direction,
foreshadowing the jaws of the Arthropoda. Ciliated, som,e-
times plumose, processes of the dorsal walls of more or fewer
of the segments may perform the office of external hrancldoe j
and, occasionally, the dorsal surface gives rise to flat shield-
like processes, the so-called elytra.

The following detailed description of a very com.mon
species of Polynoe will give a fair conception of a pol3'ch8e-
tous Annelid, in which the highest degree of comj^lexity of
organization known in the group is attained :

Polynoe sqiiamata is an elongated vermiform animal,
about an inch long, the body of which is divided into a suc-
cession of portions, for the most part similar and equivalent
to one another, but presenting peculiar modifications at the
anterior and posterior extremities. Each such portion is
properly termed a somite j while the term "segment" may
be retained to indicate generally a portion of the body, with-
out implying its precise equivalency to one somite or to
many. Thus, then, the body of the Polynoe is composed of
a series of twenty-six "somites," terminated anteriorly by a
"segment," the prcestoynium ("Kopf-lappen," Grube), and
posteriorly by another, the pygidium^ which may or may not
represent single somites.

If one of the somites from the middle of thp body (Fig.
51, <7, Z>) be examined separatel}^ it will be found to be
transversely elongated, so as to be about three times as broad
as it is long, and to be slightly convex above and below,
presenting a deep, median, longitudinal groove inferiorly.
Laterally the somite is produced into two thick processes,
the '•'' parapodia, ^"^

Each parapodium divides at its extremity into two por-
tions, a superior and an inferior, which may be denominated
respectively the notopodium (Fig. 51, ^) and the neuropodiiim
(Jc)^ the one occupying the " ha?mal " or dorsal, the other the
"neural" or ventral aspect. The latter is, in this species

POLYNOE SQUAMATA.

201

so much the larger, that the notopodium appears like a mere
tubercle projecting from its upper surface. In other A?i?ie'
lida, however, and in the young state of Polynoe^ the notopo-

FiQ. hl.—Polynde squamata.

A. Viewed from above and enlarged : a. b, c, etc., as in Fi^. 53, 5; e, elytra ; /, space
left between the two posterior elytra ; g, sets and fimbriag oftlie elytra.

B. Po!?terior extremity, inferior view: (?, pygidial cirri; h, inferior tubercle; c,c^,
notopodial and neuropodial cirri.

O. Section of half a somite with elytron: i, notopodium; k, ncuropodium.
I). Section of half a somite with cirrus.

dium is as large as the neuropodium. Both divisions of the
parapodia are armed with peculiar stiff, hair-like appendages
{g)y composed of chitin, and developed within diverticula of
the integment, or trichophores^ in which their bases always
remain inclosed. These can be protruded and retracted by
muscles attached to their sacs, and they vary exceedingly in
form. Three distinct kinds are observable in Polynoe alone.
The notopodium and the neuropodium carry each a single,
sharp, style-like aciculum, the greater part of the length of
which is imbedded in the parapodium and its divisions, while
the point just projects at about the centre of the latter. The
neuropodial is very much longer than the notopodial aciculum.

202

THE ANATOMY OF IXVERTEBIUTED ANIMALS.

Superiorly, the notopodium carries two transverse rows of
more slender organs of a similar nature, the setoe : the proxi-
mal set are much shorter than the distal, but even the latter
do not attain a length of more than y^g of an inch (Fig. 52,

The proximal set are somewhat knife-like in shape if viewed
in profile, consisting of a comparativeh" short, straight "han-
dle," by which they are imbedded in their sacs, and of a thick,
rounded, curved blade, tapering to a fine point at its extrem-
ity. Close-set transverse ridges, finely serrated at their edges,
and inclined obliquely to the surface of the blade, traverse
its convex anterior circumference, leaving the back free. The
distal setae (Fig. 52, G) have a very similar structure, but they
are much elongated and very slender. The handle is longer;
and the blade, little curved and simply set on an angle with

f'?^3>^^

Fig. 5Z.—Po!ynoe sqnamata.

A, elvtron viewed from above. B. a tooth. C. D, nenropodial setae. F, F, part? of

the blade of the same, more highly ma;?nified. O, free extremity of a uotopodial

peta.

the handle, is produced at the end into a long and delicate
filament. The base of the blade {£J) is beset with incomplete

polynOe SQUAMATA. 203

ridges, like those of the short setse, but toward the middle
(I^) these ridges appear to encircle the blade completely, as-
suming the aspect of so many closely-imbricated concentric
scales, before finally becoming obsolete upon the extremity
of the seta.

The neuropodial aciculum needs no special notice, except
that the extremity of its trichophore projects as a sort of
papilla, less obvious in full-grown specimens, which divides
the neuropodium into an upper and a lower portion, the for-
mer containing about half as many setas as the latter. The
apertures of the trichophores are placed between lobe-like
prolongations of the neuropodium, to which the special term
of labia (Grube) may be applied. In this species they pre-
sent no remarkable peculiarity bej^ond their inequality.

The neuropodial set^e (Fig. 52, C, jD), although at first
sight very different from the notopodial setfB, are, in truth,
constructed on essentially the same plan, the blade being
short, while the handle is proportionally elongated. The
blade is subcylindrical at its base, pointed and slightly curved.
Eight or nine transverse ridges extend around about two-thirds
of the circumference of its proximal half ; the basal ridges
are narrow, and merely serrated, but toward the apex the
ridges become deeper, and the serrations pass into strong
teeth ; at the same time, one side of the ridge is elongated
into a strong point.

Attached to the under surface of the parapodium by a
somewhat enlarged base, with which it is articulated, is a
smooth, conical, very flexible filament — the neuropodial ciV'
rus (Fig. 51, c') ; it hardly reaches to the end of the neuro-
podium. Again, springing from the neural surface of the
somite, close to the parapodium, there is a small pyriform
tubercle (A), divided by longitudinal grooves into about eight
segments. This is possibly connected with the reproductive
function.

The appendage of the notopodium, or rather of the noto-
podial side of the parapodium and somite, varies according
to the particular somite which may be examined. In some
somites this appendage is a cirrus (Fig. 51, 7), c) similar to
the neuropodial cirrus, but much larger, equaling the semi-
diameter of the body in length, and presenting an enlarged
pigmented bulb of attachment to which the filament of the
cirrus, which is cylindrical for about two-thirds of its length,
and then becomes enlarged and suddenly tapers to its extrem-
ity, is articulated.

304 THE ANATOMY OF INVERTEBRATED ANIMALS.

la the other somites the notopodial appendage is a large,
thin, oval plate — the elytron (Fig. 51, O, c). It is attached
by a thick peduncle, and has its long axis directed obliquely
outward and backward. The surface of the elytron (Fig. 52,
A) is covered with an ornamentation of larger or smaller
tubercular prominences, granulated and ridged upon their
surface. A part of the inner and anterior edge of each ely-
tron overlaps or is overlapped by its fellows for a certain ex-
tent of its circumference, which is so far smooth, but in the
rest of its extent it is fringed with coarse brownish filaments
ov Jimbrice, which arise from the upper surface just within the
edge, and are obviously outgrowths of the same order as the
tubercles.

Such is the structure of one of the middle somites of
Polynoe squamata. The anterior and posterior somites, with
the exception of the first and second, present only minor dif-
ferences, as in the proportion of the setse, or in the figure of
the elytra. The first somite, which contains the mouth, is the
peristomiwn (" Mund-Ssgment " of Grube). The parapodia
of this somite are narrow and elongated (Fig. 53, J3, (7, m) ;
they are obscurely divided at their extremity into a rudimen-
tary neuropodium and notopodium, and give attachment to a
pair of large peristomial cirri (c' c) (" cirrhes tentaculaires,"
Audouin and Milne-Edwards ; " Fiihler-cirren," Grube), of
the same structure as the notopodial cirri, which stretch for-
ward by the sides of the mouth.

Tiie apex of a single small aciculum issues rather above
the point of division of the peristomial parapodium, and two
minute curved setae accompany it. These have been generally
overlooked ; ^ but they seem to demonstrate, in a very inter-
esting manner, the nature of the appendages of the peristo-
mial searment.

The second somite differs from the rest only in the great
elongation of its neuropodial cirrus, which is directed forward
and applied against the mouth.

The peristomium and the prsestomium together are ordi*
narily confounded under the common term of " head." The
latter (Fig. 53, ^, C, I) is an oval segment flattened superior-
ly, placed altogether in front of and above the mouth, pre-
senting on its post ero -lateral edges four dark spots, the eyes,
and possessing five cirriform appendages, two pairs and a

* At least, in the descriptions of the adult Poli/noe. They are particularly
mentioned, however, by Max Miiller in his valuable paper, " Ueber die Ent-
wickelung und Metamorphose der Polynoen." {MuUer^s Archiv^ 1841.)

POLYNOE SQUAMATA.

205

single median one. The latter (a), or the 2:)rce^tomial tentacle
(" antenne mediane," Milne-Edwards), is similar in structure
to an ordinary cirrus. Of the other appendages, the upper
one upon each side (supero-lateral prsestomial cirrus, " an-
tenne mitoyenne ") also resembles an ordinary cirrus (b) ; but
the lower (infero-lateral prsestomial cirrus, " antenne ex-
terne ") (b') is much larger, and is capable of extreme elon-

FiG. 5S.—Pol7jnoe sqnamafa.

A. Posterior extremity from above : c, notopodial cirrus of last somite; d, pygidial
cirri ; x, anus.

B. Anterior extremity from above : a, prsestomial tentacle ; b, superior and b'
.inferior prasstomial cirrus ; c, c', notopodial and neuropudial cirri ; e, peduncle
of first elytron ; L prsestomium ; /«, parapodium of pcrlstomium. C. Inferior
view of anierior extremity, letters as before.

gation and contraction,* while the ordinary cirri are merely
flexible. Altliough at first sight probable, yet it would ap-
pear, from Max Muller's account of the development of Poly-
noe^ that these two appendages do not, like the two peristo-
mial cirri which they essentially resemble, correspond with
the notopodial and neuropodial cirri of a single parapodium,
inasmuch as they arise from perfectly distinct portions of the
pra^stomium. It is very possible that each represents the
appendage of a somite, and in this case the praestomium
w^ould be composed of at least two somites. Whether the
prsestomial tentacle indicates another, or w'hether it is merely

* I have never o"b*;er\-ed any invagination such as is stated to occur bv
Audouin and Milne-Edwards, 1834. (" Histoire Naturelle du Littoral de la
France," p. 10.)

206 THE ANATOMY OF IXVERTEBRATED ANIMALS.

an appendage of such a nature as the labrum or the rostrnm
of a Crustacean, there is no evidence at present to show.

It is highly interesting to remark that thus, in the Poly-
noe, as in the Arthropoda^ the *'head " results from the modi-
fication of a number of somites, some of which lie in front of,
and others behind, the mouth. The movements and evident
extreme sensitiveness of the inferior praestomial cirri during
life indicate that they perform the functions, as well as occupy
the position, of antennae.

The hindermost segment of the body, or pygidimn (Fig.
51, -S, Fig. 53, A)^ is narrow, and divided at the end into two
supports for the pygidial (d) cirri which are as long as the
three last somites, and resemble the notopodial cirri in form
and structure. They extend directly backward, almost paral-
lel with one another, and with the notopodial cirri of the last
somite, which are thrown backward and downward (Fig. 53,
A^ c). It seems probable that the pygidium represents only
a siuQ-le somite.

The anus is not terminal, as in many Annelids, but is
seated in the middle of a strongly-raised papilla (Fig. 53,
A^ cc), which projects from the dorsal surface of the penulti-
mate somite ; its sides are produced into about fourteen folds.
The two last elytra have their edges excavated, so as to leave
a space over the anus (Fig. 51, A^f).

The notopodial cirri and the elytra do not coexist upon
the same somites ; and the order of arrangement of the ely-
trigerous and cirrigerous somites is very curious. The 1st or
peristomial somite is cerrigerous, and so are the 3d, 6th, 8th,
10th, 12th, 14th, 16th, 18th, 20th, 22d, 24th, 25th, and 26th ;
while the 2d, 4th, 5th, 7th, 9th, 11th, 13th, loth, 17th, 19lh,
21st, and 23d, somites bear elytra, making twelve pairs in all.

In no polychaetous Annelid is the structure of a somite
more complex than in Polynoe / and there are but very few
parts not found in Polynoe to be met with in other Annelida,
The careful study of this species, therefore, furnishes us with
an almost complete nomenclature for the external organs of
the whole group ; and it will be found that the other forms
of Annelida differ mainly in the greater or less development
and modification of the organs which have just been de-
scribed. A large proportion of the Polychata are like Poly-
noe, free and actively locomotive animals, which rarely fabri-
cate tubular habitations, and are therefore termed Errantia ^
they possess a praestomium, usually provided with eyes and
feelers, and have many parapodia, which are not confined to

THE POLYCHJ^TA. 207

the anterior region of the bod}-. They very generally have a
proboscis, provided with chitinous teeth.

The singular genus, Tomopteris, is a transparent pelagic
Annelid, with numerous parapodia, each terminated by two
lobes representing the neuropodium and notopodium, but
with setae, two of w^hich are very long, only in the cephalic
region.

The sedentary Annelids [Tuhicola) fabricate tubes, either
by gluing together particles of sand and shells, or by secret-
ing a chitinous or calcified shelly substance, in which they
remain (e. g., Protula^ Fig. 54). The praestomium is small or
wanting ; none have a proboscis ; there are no cirri ; and the
parapodia are short or rudimentary. The branchiae are devel-
oped only on the anterior somites, and the latter are often
markedly diflferent from those which constitute the posterior
part of the body.

In some (/Serpuliclce) a tentacle is enlarged and its end
secretes a shelly plate which serves as an operculum, and
shuts down over the mouth of the calcareous tube inhabited
by the animal, when it is retracted. The dilated end of the
opercular tentacle sometimes serves as a chamber in which
the young undergo their development (species of Spirorhis),

The alimentary canal of the polychaetous An7ielida rarely
presents any marked distinction between stomach and intes-
tine, and is almost always of the same length as the body, ex-
tending, without folds or convolutions, from its anterior to
its posterior extremity; but in Siphonostomum (Chloroemd)^
Pectlnaria and others, it is more or less convoluted. It is
attached by membranous bands, or more complete mesenteries,
to the walls of each somite, and very commonly presents a dila-
tation between every pair of mesenteries. In most Polychceta,
the intestine acquires in this wav merely a moniliform appear-
ance, but in Polynoe^ Aphrodite^ Sigaliony and their allies,
long cpeca are given off upon each side of the alimentary
canal, and, sometimes becoming more or less convoluted, ter-
minate at the upper part of each segment (Fig. 51, D) close
beneath, or in the branchiae, w^iere such organs exist.

The anterior portion of the alimentary canal is, in a great
number of the Polychceta^ in fact in all the typical Errantia,,
so modified as to constitute a distinct muscular pharynx, the
anterior portion of the wall of which can be everted like the
finger of a glove, from the aperture of the mouth, and the
posterior portion protruded, so as to form a proboscis. In
Polynoe squamata, the proboscis is one-fourth as long as the

208 THE ANATOMY OF INVERTEBRATED AKIMALS.

Fig. bi—Protida Thjsteri. A, the Bexual. mature animal, extracted from its calca-
reous tube : a, branchial plumes ; b, hood-like expansion of the anterior end of the
bofly : c, the mouth ; c?, the stomach ; e, the anu^ ;/, the testos ; g^ the ova. B^
a Protula in course of proliferation ; &, the branchiae of the zo5id,

bod}', and its walls are very thick and muscular. At its an-
terior extremity it is surrounded with a circle of small papil-
lae, immediately behind which are four strong, pointed and
curved horny teeth, implanted in the muscular wall (Fig. 52,

THE POLYCHiETA. 209

S). Each tooth has a little projection upon its convex edge,
which is connected by a short strong ligament with the cor-
responding projection of another tooth ; and the one pair of
teeth, thus connected, works vertically against the opposite
pair. In JSTereis, there are two powerful teeth which work
horizontally, besides minute accessory denticles. In Syllis,
the chitinous lining of the pharj'nx is produced into a circle
of sharp teeth anteriorly, and there is, in addition, a much
stronger triangular median tooth. In Glycera, which pos-
sesses a pair of teeth, the extremity of the protruded pro-
boscis is covered with very remarkable papillae. The most
complex arrangement of teeth, however, is that presented by
the EunicidiB. In Eunice^ there are altogether nine distinct
pieces : two large, flat, more or less calcified portions united
together below, and three cutting and tearing teeth on the
right side working against four on the left. As has has been
already stated, the tubicolar Annelids possess neither probos-
cis nor teeth.

No special hepatic gland appears to exist in the Annelida^
unless the intestinal ca?ca perform that function, and the
secretion of the bile is doubtless effected by the glandular
tract, which extends for a greater or less distance in the walls
of the alimentary canal. A pair of glandular CtBca, the func-
tion of which is not known, is appended to the base of the
proboscis in Nereis.

The general cavity of the body, or perivisceral cavity,
which is included between the parietes of the alimentary
canal and those of the body, is filled with a fluid which con-
tains corpuscles, which are usually, as in the Invertebrata in
general, colorless. They are red, however, in Glycera, and
in a species of Apneumea (De Quatrefages). The parapodia,
the cirri, the branchiae, and all the other important appendages
of the PolycJiceta^ contain a cavity continuous with the peri-
visceral cavity, and are therefore equally filled with the blood.
The circulation of this fluid is effected partly by the contrac-
tion of the body and its appendages, partly by the vibratile
cilia, with which a greater or less extent of the walls of the
perivisceral cavity is covered.

In a great number of the FolychfMa no part of the body
is specially adapted to perform the function of respiration,
the aeration of the blood probably taking place w^ierever
the inteofument is sufficientlv thin ; and, even when distinct
branchiae ordinarily exist, members of the same family may
be deprived of them. In Polynoe squamata, ciliated spots

210 THE ANATOMY OF INVERTEBRATED ANIMALS.

which appear to represent branchiae, may be discovered on
the dorsal side of the bases of the parapodia, at any rate, in
youno- specimens. In some species of Pohjnoe the parapodia
ffive °ise, at corresponding points, to large, richly cihated,
malleiform tubercles, in which the caeca of the ahraentary
canal terminate. In Slgalion, a filiform, cihated brancbia
depends from the upper part of the somite, beneath the ely-
tron • and, besides this, curious little ciliated palettes are
arrano-ed upon the dorsal surface of the parapodia, and upon
th- sfd-s of the anterior somites. But the best-developed
brinchiie among these Annelids are possessed by the Anqylii-
nomldx, and the JEimicld'B among the Errant la; the 'Jere-
hdUd%, and the Sirpulid:^ among the Tuhicola. In the
three former families the branchiae are ciliated branched
plumes, or tufts, attached to the dorsal surface of more or
fewer of the somites. In the last (Fig. 54) they are exclu-
sively attached to the anterior segment of the body, and
present the form of two large plumes, each consisting of a
principal stem, with many lateral branches. The stem is
supported by a kind of internal skeleton, of cartilaginous
consistence, which sends off processes into the lateral branches.

I have been unable to find any pseud-haemal vessels in
Polynoe squam'-xta, and, as Claparede ' could discover none in
the transparent P. litmdata, it is safe to assume their non-
existence. Claparede, in fact, denies them to the whole of
the Aohrodltldoe.

When it is present, the pseud-haemal system varies very
much in the arrangement of its great trunks ; but they com-
monly consist of one or two principal longitudinal dorsal and
ventral vessels, which are connected in each somite by trans-
verse branches. Where branchise exist, loops or processes of
one or other of the great trunks enter them. The dorsal and
the ventral trunks are usually rhythmically contractile, and
contractile dilatations at the bases of the branchiae (Eunice),
in portions of the lateral trunks {Are?iicola), or in those
which supply the proboscis (Eunice, Nereis), have received
the name of " hearts." The direction of the contractions is
usually such that the blood is propelled from behind forward
in the dorsal vessel, and in the opposite direction in the ven-
tral vessel ; but the course which it pursues in the lateral
trunks is probably very irregular. In Chlorcema, in which
even the smallest ramifications of the vessels are contractile, I

» "Annelides Ch^topodes du Golfe de Naples," 1868, p. 65.

THE POLYCH^TA. 211

have observed c.'ecal branches depending into the perivisceral
cavity in which the contained fluid underwent merely an alter-
nate flux and reflux. Ramified caeca of a similar kind appear
to exist in the oligochsetous genera, Euaxes and Lumbriculus.
The principal trunks give otf a great number of branches,
which ramify very minutely in some Annelids [Euyiice) and
may give rise to retia mirabilia {Nereis) ; but in many (e. g.,
Protula) there are hardly any branches and no minute capil-
lary ramifications.

In many Folychmta no segmental organs have yet been
discovered, and in others they appear to be represented by
mere openings in the parietes of the body. I have observed
short ciliated canals opening externally upon the ventral sur-
face at the bases of the parapodia in Phyllodoce viridis, and
there are indications of the existence of similar organs in
Syllis vittata. True segmental organs have, however, been
found by Ehlers and Claparede in many JPolychceta. In some
cases their walls are thick and glandular, and they probably
have a renal function. In addition, they frequently play the
part of oviducts and spermiducts. "Whether the ciliated canal
extending along the ventral surface of the intestine, which I
have described in Protula, is a structure of the same order or
not. I am not prepared to say.

The nervous system of the Polychceta usually consists of
a chain of ganglia — one pair for each somite — connected
together by "longitudinal and transverse commissural bands,
which diverge between the cerebral ganglia and the succeed-
ing pair, to allow of the passage of the oesophagus. The most
important diff'erences presented by the nervous systems of the
Polychosta result from the varying len<rth of the transverse
commissures. In Vermilia, Serpula, Sahella, these commis-
sures are very long, so that two distinct and distant series of
ganglia appear to run through the body, while, in JVepthys,
the two series of ganglia are fused into a sina^le cord enlarged
at intervals. Every transitional condition between these is
observable in Terehella^ Aonia, Glycera^ Phyllodoce^ and
Aphrodite. In most Polychrpta a very extensive series of
visceral nerves supplies the alimentary canal.

The recognizable organs of sense in the ArineUda are eyes
and auditory vesicles. The former are usually very simple,
consisting of an expansion of the extremity of the optic nerve,
imbedded in pigment, and provided occasionally, but not in-
invariably, with transparent spheroids or cones. Alciope and
Torrea have very well-developed and large eyes. The eyes

212 THE ANATOMY OF IXVERTEBRATED AXIMALS.

are usually confined to the anterior extremity of the body, and
to the prc^stomium where it exists ; but, iiQ the remarkable
genus Polyophthalmus, De Quatrefages discovered, besides

—c

Fig. 55. — A, anterior end of the nerrous system of Tclynde sq-uamcfa (aftr-r De Qna-
trefagest : o, cerebral yanfrlia ; b, oesi)phai;eal cominissures ; e. !oi,<riiuaiual com-
missures of the venrral ganglia.

B, anterior end of the nervous system of SaMla fiabellcta (after De Quatrefages) : o,
cerebral gan^rlia; b, oesophageal commissures ; c.loniritr.dinal commissures of the
venrral giinglia. Those of opposite sides are united by long transverse commie-
Eures.

the ordinary cephalic eyes, a double series of additional visual
organs, one pair being allotted to each somite. In 13ran-
chiotnma, eyes are situated at the ends of the branchial
plumes. Ehrenberg has described two caudal eyes in Ay/iphi-
cora, and De Quatrefages has shown that similarly placed
eyes exist in three other species of PolychcpAa^ two of which
are closely allied to Amphicora, while the other is an errant
form, related to Lvmhrinereis. These curious worms are said
to swim about with the caudal extremity forward.

Auditory sacs, containing m,any otoliths, have been ob-
served upon each side of the oesophageal ring in Arenicola^
and similar organs have been noticed in other Tiibicola ; but
hitherto their existence has not been certain]}'" determined in
the Kr rant la.

The crenitalia of the polychaetous Annelida are excessively
simple in their structure ; indeed, special reproductive organs
can hardly be said to exist in most, the generative products

THE DEVELOPMENT OF THE POLYCHJETA. 213

being merely developed from some part of the walls of the
perivisceral cavity, in which thev eventually freely float, mak-
ing their way out in a manner which is not quite understood
at present ; probably, however, through temporary or perma-
nent apertures at the bases of the parapodia. In many, the
segmental organs appear to serve as excretory ducts. As a
rule, the polychaetous Annelids are dioecious ; but some (e. g.,
Protula^ Fig. 54) are hermaphrodite. The ova undergo their
development within the body of the parent in some species
of EiUiice y in pouches attached to the body in Exogone / in
masses of gelatinous matter which adhere to the tubes of the
vermidom in Protula j beneath the elytra in Polynoe cir-
rata / in the cavity of the opercular tentacle in some Spii'-
orhes / while, in other cases, they appear to become, almost
immediately, free ciliated embryos.

The vitellus undergoes division, and is converted, as in
the case of the Oligochceta and Ilirudmea, into blastomeres
of two kinds. This contrast between the two components of
the embryo commences with the division of the vitellus into
two, inasmuch as the first fissure is usually so directed as to
divide the yelk into unequal portions. Both subdivide, but
the smaller much faster than the larger ; so that the former
becomes converted into very small blastomeres, which grad-
ually envelop the larger blastomeres resulting from the sub-
division of the latter. The larger included blastomeres are
destined to form the alimentary tract ; the smaller peripheral
ones, on the other hand, give rise to the ectoderm, and to the
nervous ganglia.' As in the Ollgochmta and Ulrudlnea,
again, the mesoblast forms a thick band on each side of the
median ventral line, and its transverse division originates the
segmentation of the body. But, generally, the development
of the 2:>rotosomUes, as these segments might be called, does
not occur until some time after the embryo has been hatched.
The somites increase in number by the addition of new ones
between the last and the penultimate somite.

The embryos of the Polychoeta differ from those of the Oli-
goch(Bta and Hlrudinea in being ciliated. In some cases, the
cilia form a broad zone which encircles the body, leaving at
each end an area, which is either devoid of cilia, or, as is fre-
quently the case, has a tuft of long cilia at the cephalic end.
Such larvre are termed Atrocha.

In other embryos the cilia are arranged in one or more

1 Claparede and raetschnikofF, " Beitrage zur Kenntniss der Entwickelungs-
geschiclite der Chaetopoden," 18G8.

214 THE ANATOMY OF INVERTEBRATED ANIMALS.

narrow bands, which surround the body. A very common
arrangement is one in which a band of cilia encircles the body
immediately in front of the mouth, the region in front of the
band bearing eyes, and becoming the praestomium of the adult
(e. g., Polynoe). In such embryos, there is very commonly a
second band of cilia around the anal end of the embryo, and
a tuft of cilia is attached to the centre of the praestomium.
These larvae are called Telotrocha. In other cases, one or
many bands of cilia surround the middle of the body, between
the mouth and the hinder extremity. These are Mesotrocha.

In the telotrochous larva of JPhyllodoce, a shield-shaped,
mantle-like elevation of the integument covers the dorsal
region of the body behind the prae-oral ciliated ring. In the
larvae of the Serjndidce a process of the integument grows
out behind the mouth, and surrounds the anterior part of the
body of the larva like a turned-back collar. It persists, as a
kind of hood, in the adult.

Some larvae are provided with setse of a different charac-
ter from those which are possessed by the adult, and which
are cast off as development advances.

Many Polychmta multiply by a process of zooid develop-
ment, which, in some cases, appears to be a combination of
fission with gemmation ; in others, to approach very nearly
to pure fission or pure gemmation. The result is, not infre-
quently, the formation of long chains of connected zooids.

The method of multiplication which De Quatrefages ob-
served in SylUs prollfera., is nearly siniple fission, the animal
dividing near its middle, and the posterior division acquiring
a new head.

In Myrianida., Milne-Edwards has described the occur-
rence of a sort of continuous budding between the ultimate
and penultimate segments, in which region new segments are
formed until the zooid has attained its full length.

Frey and Leuckart and Krohn have shown that Autolytus
prolifer multiplies in a somewhat similar manner ; but, in-
stead of each new zooid being formed at the expense of an
entire somite, it is developed from only a portion of one.
Finally, I found in Protula Dysteri that, when the Protida
had attained a certain lensrth, all the somites behind the six-
teenth became eventually separated as a new zooid ; but the
development of the latter is not mere fission, inasmuch as one
of the earliest steps in the process is the enlargement of the
seventeenth somite, and its conversion into the head and

AGAMOGENESIS AMONG POLYCH^TA. 215

thorax of the bud (Fig. 54, B). Sars has described a similar
mode of multiplication in his iilograna implexa^ a very close-
ly allied form.

In SylUs and in Protula, the producing and the produced
zooids alike develop generative products, but, in Autolyhis^
Krohn has shown that the primary producing zouid remains
sexless, the secondary produced zooids having a somewhat
different form, and alone giving rise to ova and spermatozoa.

In some species of the genus JVereiSy the worm, after the
development of its genital organs has taken place, takes on
the characters of what was formerly considered a distinct
genus, ITeterotiereis ; and the males and the females of the
same species of Nereis have even been regarded as different
species of Setero nereis.^

The series of forms represented by the Tiirhellaria^ the
Jlirudinea, the Oligochceta, and the Polychata^ illustrates
the manner in which a type of organization, which, in its
simplest condition, exhibits but little advance upon a mere
Gastrula, passes into one in which the body is divided into
manv segments, each provided with a pair of appendages or
rudimentary limbs.

The segmentation, or serial repetition of homologous
somites, extends to the nervous system, and, more or less, to
the vascular and reproductive organs, in the higher forms of
these "Annulose" animals; from which a further extension
of the same process of segmentation, with a fuller develop-
ment of the appendages and a more complete appropriation
of some of them to manducatory purposes, leads us to the
Arthropoda.

The Gephyrea. — These are marine vermiform animals
without distinct external segmentation or parapodial append-
ages. The ectoderm has a chitinous cuticle, and is often
provided with tubercles, hooks, or set^e, of chitin i^EcMurus^
Sternaspis). No calcareous skeleton is found in any of the
Gephyrea. The integument frequently contains numerous
simple glands, the apertures of wliich perforate the cuticle.
In one genus i^Sternaspis)^ two shield-shaped plates, fringed
with setae, are developed upon the hinder part of the ventral
surface bf the body. There are external circular and internal
longitudinal muscular fibres beneath the ectoderm. An inner

1 Ehlers, " Die Gattung Heteronereisy (" GGttlngen Nachrichten," ISCr.)

216 THE ANATOMY OF INVERTEBRATED ANIMALS.

layer of circularly disposed muscular fibres may be added.
The oral end of the body may have the form of a retractile
proboscis (I^riapiilus), or be provided with tentacular append-
ages. These may be arranged in a circle round the mouth,
and short (Slpicncidus, Fig. 56, L, T), or long (Phorojiis)^ or
there may be a single long, sometimes bifurcated and ciliated,
tentacular appendage {£onellia). Filamentous appendages,
which are probably branchiae, are given off at the hinder end
of the body in Sternaspis and JPriapulus. The endoderm is
usually ciliated throughout. The intestine is straight in most
genera, but is coiled and bent upon itself, so as to terminate
in the middle of the body, in Sipunculus (Fig. 56, I.). In
Phoronis the anus is close to the mouth. The anal aperture
is always situated upon the dorsal aspect of the body. There
is a spacious perivisceral cavity, undivided by mesenteries,
which in some cases (Priapulus^ Sipunculus) opens externally
by a terminal pore. In Echiuriis^ JBonellia, Thalassema^ a
pair of tubular, sometimes branched organs, which are ciliated
internally, and communicate by ciliated apertures with the
perivisceral cavity, open into the rectum. These appear to
represent the water-vessels of the Itotifera and the respira-
torv tubes of the Holothurioe.

A pseud-hsemal system exists in most [SijDunculus^ Sternas-
pis, PoneUla, JEJjhiurus, and Phoro7iis), and, when fully devel-
oped, consists of two longitudinal trunks — one dorsal, or su-
pra-intestinal, the other ventral, with their terminal and lateral
communications. The pseud-hremal fluid is colorless, or may
have a pale reddish tinge, in most. In Phoro7iis it is said to
contain red corpuscles. In Sipunculus, the cavities of the
tentacles communicate with a circular vessel provided with
cascal appendages ; and this circular vessel is said to open
into the pseud-htemal vessels.

The nervous system presents a collar, which surrounds the
oesophagus, and from which a simple or ganglionated cord
proceeds backward in the ventral median line, giving off lat-
eral branches. The ventral cord contains a central canal, and
the collar usually presents a cerebral ganglionic enlartrement.
Rudimentary eyes are sometimes connected with the cerebral
ganglion.

The sexes are distinct, and the reproductive elements are
developed either from the parietes of the perivisceral cavity
or in simple ccecal glands. In Sipunculus, the ova and sper-
matozoa float freely in the perivisceral cavity.

The actively locomotive embryo of Sipunculus (Fig. 56, II.)

THE GEPHYREA.

217

is surrounded by a circular band of cilia placed immediately
behind the mouth ( W] TF), and resembles a Rotifer or a meso-
trochal Annelidan larva. As developuient advances it loses

Fig. 5Q.—Sipunculus nudus (after Keferstein and Ehlers).^

I. The animal laid open longitudinally — | n. s. T, tentacles; r, the four retractor
muscles of the proboscis; r, the points at which they were*attached to the walls
of the body ; te, oesophagus ; i. intestine; a, anu:? ; X J\ loops of the intestine ;
«, y, appendages of the rectum; 2, fusiform muscle; w, ciliated groove on the
inner side of the intestine ; g, anal muscles ; s, csecal glands ; t. cseca which open
on each side of the nervous cord, and are gonerally considered to be testes ; p,
pore at the hinder end of the body; ??, nervous cord, which ends in a Inbed gan-
glionic mass, close to the mouth, and presents an enlargement, g\ at its poste-
rior end ; wi, m', m". muscles associated with the nervous cords.

II, A larval Sipunculus about rV of an inch long: 0, mouth ; ce, gnllet ; 8, csecal
gland; t. intestine with masses of fatty cells ; a, anus ; w, ciliated groove of the
intestine ; g. brain with two pairs of red eye-spots ; », nervous cord, p, pore;
t, t, so-called testes ; TF, TF, circlet of cilia.

this apparatus, and passes gradually into the adult form. In
PJioronis, the embryo is also mesotrochal, but it has two
ciliated bands, one circular, round the anus, and the other im-
mediately behind the mouth. The post-oral band of cilia is
produced into numerous tentaculiform lobes, and fringes the
free edge of a broad concave lobe of the dorsal side of the
body, vvhich arches over the mouth. In this state the embrj^o

1 "

Zoologische Beitrage," 1861.

10

218 THE ANATOMY OF INVERTEBRATED ANIMALS.

is the so-called Actinotrocha.^ An invagination of the ven-
tral integument of the larva connects itself with the middle
of the intestine, and then, becoming evaginated, pulls the in-
testine, in the form of a loop, into the ventral process thus
formed, which gives rise to the body of the Fhoronis^ while
the tentacles of the larva grow into those of the adult.
Schneider has suggested that the bell-shaped larva, with long
seta3, termed Mitraria by Miiller, is the embryo of Sternaspis.
The affinities of the Gephyrea with the Turhellaria^ with
the Annelida^ and with the JRotifera^ are unmistakable. In
fact, it may be doubted whether Sternaspis should not be
associated with the Polychoeta, and £o7iellia is in many re^
spects comparable to a colossal Rotifer. Their usually as-
sumed connection with the Echinodermata is more question-
able. The circular canal which communicates with the cavi-
ties of the tentacles in Sipunculiis has been compared to the
ambulacral system of the Echinoderms, but the manner of
its development is not yet sufficiently understood to justify
the expression of an opinion on this subject. Krohn has de-
scribed a bilobed organ on the ventral face of the gullet of the
larva of Sipunculus, which opens externally in front of the
ciliated band by a narrow ciliated duct ^ (Fig. 56, II., S). It
has a striking similarity to the " water-vessel " of the larva
of £alanoglossus, which, however, lies on the opposite side
of the body.

1" Schneider, "TJeber die Metamorpliose der Aciinotrocha Iranchiata.''^
(" Archiv fur Anatomie," 1862.)

" " Ueber die Larve des Si^vnculus nvdusy (" Archiv fiir Anatomic,"
1851.)

CHAPTER VI.

IHE AKTHROPODA.

The segmentation of the body, that is, its division into
a series of somites, each provided with a pair of lateral ap-
pendages, which is so characteristic a feature of the higher
Annelids, is exhibited in a still more marked degree by the
Arthropoda. In these animals, moreover, the appendages,
themselves are usually divided into segments, while one or
more pairs of the appendages in the neighborhood of the
mouth are modified in form and position to subserve man-
ducation. Segmental organs, at least in their Annelidan
form, are wanting in the Arthropoda, and neither in the em-
bryonic nor the adult condition do they ever possess cilia.

The process of yelk-division may be complete or incom-
plete, but no known Arthropod ovum gives rise to a vesicular
morula, nor is the alimentary cavity ordinarily formed by in-
vagination/ The precise mode of origin of the mesoblast
has yet to be worked out, but the perivisceral cavity appears
always to be developed by its splitting. In other words, it is
a schizocoele.

As with Annelids, the segmentation of the body results
from the subdivision of the mesoblast by transverse constric-
tions into 2yrotosomites ; and there is every reason to believe
that the ganglionated nervous chain arises from an involution
of the epiblast.

The neural face of the embryo is fashioned first, and its
anterior end terminates in two rounded expansions — the j^ro-
cephcdie lobes — which are converted into the sides and front
of the head. The appendages are developed as paired out-

» The recent observations of Bobretzky on the development of Oniscus and
Astacm (Hofmann and Schwalbe, " Jahresberichte," Bd. ii., 1875), however,
tend to show that the hypoblast arises by a sort of modified invasrination of
the primitive blastoderm. And in other Arthropoda there are indications of a
similar process.

220 THE ANATOMY OF INVERTEBRATED ANIMALS.

growths from the neural aspect of each somite, and, whatever
their ultimate form, they are, at first, simple bud-like pro-
cesses. Very generally, a broad median prolongation of the
sternum of the somite which lies in front of the mouth gives
rise to a labrum ; while a corresponding, but often bifid me-
dian elevation, behind the mouth, becomes a metastoma.

In many Arthropods, the haemal or tergal face of the body
grows out into lateral processes, which may either be fixed,
or more or less movable. The lateral prolongations of the
carapace in the Critstacea and the wings of J?isecta are
structures of this order.

In a number of Insects belonging to different orders of
the class, an amnionic investment is developed from the
extra-neural part of the blastoderm by a method similar to
that which gives rise to the amnion in the higher Vertebrata,

In all the higher Arthropods, a certain number of the
somites which constitute the anterior end of the body coa-
lesce and form a head, distinct from the rest of the body ;
and the appendages belonging to these confluent somites un-
dergo remarkable modifications, whereby they are converted
into organs of the higher senses and into jaws. In many
cases, the somites of the middle and posterior parts of the
body become similarly difi'erentiated into groups of poly-
somitic segments, which then receive the name of thorax and
abdomen. The somites entering into each of these groups
may remain distinct or may coalesce. The tergal expansions
of the somites of the head, or of both head and thorax, may
take the shape of a broad shield, or carapace. This may con-
stitute a continuous whole (e. g., Apus, Astacns) ; or its two
halves may be movably connected by a median hinge, like a
bivalve shell [Cypris^ Lhnnadia) ; or, finally, the tergal pro-
cesses of each side may remain distinct from one another and
freely movable on their respective somites (wings of In-
sects).

Limbs, or appendages capable of effectino;* locomotion, are
always attached either to the head or to the thorax,^ or to
both. They may be present or absent in the abdominal re-
gion. In adnlt Arachnida and Insecta^ there are no abdomi-
nal limbs, unless the accessory organs of generation, the stings
of some insects, and the peculiar appendages of the abdomen
in the Thysaniira and CoUembola, be such.

The alimentary apparatus presents very wide diversities

1 The extinct Trilohites possibly form an exception to this rule.

THE ARTHROPODA. 221

in form and structure, and in the number and nature of its
glands. The anus, which is very rarely absent, is situated in
the hindermost somite.

In like manner, the blood-vascular system varies from a
mere perivisceral cavity without any heart ( Ostracoda, Cirri-
pedicC) up to a complete, usually many-chambered heart with
well-developed arterial vessels. The venous channels, how-
ever, always have the nature of more or less definite lacunae.
The blood-corpuscles are colorless, nucleated cells.

Special respiratory organs may be absent, or they may
take one of the following forms :

1. Branc1ii<x, Externally projecting processes of the
body or limbs, supplied with venous blood, which is thus
brought into contact wdth the air dissolved in water.

2. Tracheae. Tubes w^hich traverse the body and gen-
erally open upon its exterior by apertures termed stigmata, u«x,^
and thus bring air into contact with the blood and the tissues
generally. Saccular reservoirs of air are often foim.ed by
dilatations of these tubes.

The so-called Tracheo-hranchim of some aquatic Insect
larvae are usually laterally projecting processes of more or
fewer of the thoracic or abdominal somites, containing abun-
dant tracheae, which communicate with those which traverse
the body [^Ephemeridce, Perlaridaf). They are in no sense
branchiae, but simply take the place cf stigmata. The ex-
changee of constituents between the air contained in the
tracheae of these animals and that of the surrounding medium
is effected indirectly, by diffusion through the walls of the
tracheo-branchia?, instead of directly, through the stigmata,
as in other cases.

In the aquatic larvre of many Dragon-flies (Zibelhdidc^),
the function of the tracheo-branchiae is performed by folds of
the lining membrane of the rectum, which contain abundant
tracheae. Water is drawn into, and expelled from, the cavity
of the rectum by rhythmical contractions of its walls, so as
to secure the exchanjre of g-aseous constituents between the
air which it contains and that v.'hich fills the tracheae.

3. Pulmoiiary sacs. These are met with only in some
Arachnida. They are involutions of the integument, the
walls of w^hich are folded in such a manner as to expose a
large surface to the air, which is alternately taken into, and
expelled from, their apertures. The blood is brought to these
sacs by venous channels.

The exact mode by which the separation of the nitro-

222 THE ANATOMY OF IXVERTEBRATED ANIMALS.

genous products of the waste of the tissues from the blood
is effected in Arthropods requires further elucidation. In
many, however, such products, notably uric acid, have been
found to abound in the co'rpus adiposum — a cellular mass
which lies in the walls of, and more or less fills, the peri-
visceral cavity — and in the Malpighian glands. In the latter
case, thev are conveyed out of the body by the intestine.

The nervous system consists primitively of a pair of gan-
glia for each somite, but the number of ganglia discoverable
in the adult depends on the extent to which these primitive
ganglia coalesce. There is usually, if not always, a well-
developed system of ganglionated visceral nerves, connected
with the cerebral ganglia and distributed to the gullet and
stomach.

Eyes are usually present ; and, when they exist, they are
almost always situated in the head and are connected with
the cerebral ganglia. Among the Crustacea, however, JEu-
2)hausia has eyes in some of the thoracic limbs, and in some
abdominal somites. The eyes may be simple or compound.
In the latter case there are, in correspondence with the num-
ber of parts into which the transparent corneal continuation
of the chitinous cuticula over the eye is divided, a number
of elongated bodies which lie between the outer surface of
the ganglionic expansion of the optic nerve and the inner
face of the cornea. These bodies consist of two parts: an
external transparent crystalline cone and an internal pris-
matic rod. The broad end of the cone is external, and is ap-
plied to the inner surface of the corneal facet ; its narrow
end is continuous with the outer extremity of the prismatic
rod, which, by its inner end, is connected with the ultimate
ramifications of the optic nerve. Each of these crystalline
cones and prismatic rods is separated from the rest by a pig-
mented sheath.^

Distinct auditory organs have been observed in Crus-
taceans and Insects. They are not exclusively confined to
the head. In the opossum shrimp (My sis), for example, they
are placed in the appendages of the last somite of the ab-
domen. And, in Insects, the only organs to which the audi-
tory function can be certainly assigned are situated in the
thorax or in the legs.

» Leydi^, "Das AuEje der Gliederthiere," 1864. Schulze, "Untersuch-
ungen," 1868. Mr. E. T. Newton has given a very good account of the struct-
ure of the eye of the lobster, accompanied by full references to the literature
of the subject, in the Quarterly Journal of Microscopical Science for 1875.

THE ARTHROPODA. 223

There is some reason to think that the antennae of Insects
are the seat of the olfactory function, but no certain infor-
mation of this head has been obtained. The very fine sette
to the bases of which nerves can be traced, which abound on
the antennary organs of Insecta and Crustacea, but are found
in other regions of the body, are probably partly tactile and
partly auditory organs.

As a general rule, all the muscles of the Arthropoda, even
those of the alimentary canal, are striated. Those of the
body and limbs are often attached by chitinized tendons to
the parts which they have to move. As the hard skeleton is
hollow and the muscles are inside it, it follows that the body,
or a limb, is bent toward that side of its axis which is oppo-
site to that on which a contracting muscle is situated.

Sounds are produced by many Insects ; but in most cases
they cannot be properly referred to a voice, in the sense in
which that term is applied to the sounds produced in the
higher animals, by the vibrations of the atmosphere arising
from the impact of a current of air upon the free edges of
membranes bounding the aperture of exit of the current.
The chirping and humming of Insects often arise from the
friction of their hard parts against one another, or from the
rapid vibration of their wnngs : in some instances, however,
recent investigations render it probable that they are pro-
duced by the action of expiratory currents on tense mem-
branes which bound the stigmata.

Agamogenesis is very common among some groups of the
Arthropoda, such as the Crustacea and the Insecta, but has
not yet been observed in the Myriapoda or the Arachnida,
It may be effected in one of two ways:

1. Either individuals which are, by their structure, inca-
pable of being impregnated and are therefore physiologically
sexless, though it may happen that they more or less approxi-
mate females morphologically, give rise to offspring ( Cecido-
myia larvae. Aphis) ;

2. Or individuals which are capable of being impregnated,
and are thus both morphologically and physiologically true
females, give rise to eggs which develop without impreg-
nation. (The queen-bee, so far as the production of drones is
concerned ; many Lepidoptera^.

The cases of Apus, Daphnia, and Cypris, would belong
to the latter category, if it were certain that the very same
females which, for a certain period, produce young agamo-

224: THE ANATOMY OF IXVERTEBRATED ANIMALS.

genetically, at another time undergo fecundation. Multipli-
cation by lission or external gemmation is not known to take
place in any Arthropod. Hermaphrodisrn occurs as a rule in
some few Arthropods (e. g., the CirrijKclia and Tardigradd)^
and as an abnormal "sport" in sundry Crustacea and in
many Insecta.

In absolute number of species, the ArtJiropoda far ex-
ceed all the rest of the animal kingdom put together. Thus
Gerstaecker/ while allowing 50,000 species for the latter,
estimates the number of species of Arthropoda as rather
above than below 200,000 ; by far the larger proportion of
these, probably more than 150,000, being Insects.

The Arthropoda are commonlj- divided into the Crustacea,
the Arachnida^ the Myriapoda. and the Insecta ; and though
it is impracticable to give a definition which shall absolutely
separate the first two groups, it is perhaps not worth while
to disturb an arrangement which has much practical con-
venience. But, for purely morphological purposes, it may be
instructive to regard them from another point of view.

The Arthrojjoda may, in fact, be divided into two series.
One of these consists almost whollv of air-breathing forms,
which, if they possess special respiratory organs, have either
pulmonary sacs or trachete, or both combined ; while the
other includes a corresponding predominance of water-breath-
ing animals, which, if they possess respiratory organs, have
branchiie. Tlie latter series contains the Crustacea ; the
former comprises the Arachnida, Myriapoda^ and Insecta.

In the course of the development of the higher Arthro-
poda, there is a stage in which the body begins to be seg-
mented, but the appendages are not developed. This is
followed by a stage in which appendages make their appear-
ance, but the antennary and manducatory appendages [gna-
thites) are like the other limbs: and, finally, there is a stage
in which the gnathites are completely converted into jaws.
Now, among the water-breathing Arthr o jyoda, r\o trace of
limbs has yet been certainly discovered among the Trilo-
hita ; in the JSIerostomata \Eurypterida and Xiphosura)
the gnathites are completely pediform ; while, in the Ejito-
rnostraca and Malacostraca, more or fewer of the gnathites
are so modified as to subserve manducation and no other
function.

» Bronu's "Klassen und Ordnungen des TliierreiciLS," vol. v., p. 273. 1868.

THE GROUPS OF THE ARTHROPODA. 225

In the air-breatliing series no completely apodal forms are
known. The Tardigrada and the Pentastomida appear to
have no jaws ; but the presence of oral stilets in the former,
and the position of the hooks which represent the limbs in
the latter, throw some doubt upon this point.

In the Arachnlda and the Peripatidea the gnathites are
completely pediform. But in the Myriapoda^ and still more
in the Insecta^ the gnathites lose the character of legs, and
are completely converted into manducatory organs. Thus
we arrive at the following arrangement of the Arthropoda :

Aetheopoda.

/. WitJiout Gnathites.
Teilobita. Taedigeada (?) Pextastomida (?)

//. With Pediform Gnathites.
Meeostoi>iata. Aeachnida. Peeipatidea.

///. With Maxilliforni Gnathites.
Entomosteaca. Mteiapoda.

Malacosteaca. Insecta.

Water-breathers. Air-breathers.

For the most part.

Of the four great groups, the Cnista^iea are those which
present the greatest and the most instructive variations upon
the fundamental type of structure; while the modifications
of the Insecta, Arachnlda, and JSIyriapoda, are less exten-
sive, and may be regarded as of secondary morphological im-
portance. The Crustacea will, therefore, be treated of at
some length, while the other groups will be passed over more
lightly.

THE CEUSTACEA.

The Trilobita. — These ancient Arthropods, which have
been extinct since the latter part of the PalfBozoic epoch, oc-
cur in the fossil state in great numbers, and in conditions
very favorable for their preservation ; but, up to this time, no
certain indications of the existence of appendages, nor even
of any hard, sternal body-wall, have been discovered, though

2'26 THE AXATOMY OF INVERTEBRATED AXIMALS.

a shield-shaped labrum, which lies in front of the mouth, has
been preserved in some specimens. Tlie body consists of a
cephalic shield (Fig. 57, A) ', oi a, variable number of mov-
ably-articulated thoracic somites (Fig. 57, £) ; and of a,py-
gidium^ composed of a variable number of the somites which
succeed the thorax, united together (Fig. 57, C).

Each thoracic somite presents a median portion, convex
from side to side, termed the axis or tergnin, and two flat-
tened lateral portions, the pleura. The former overlap one
another largely when the body is extended, the latter when
it is flexed, and the freedom of motion permitted by this ar-
rangement is so great that many Trilobites were able to roll
themselves up like wood-lice, and are found fossilized in that
condition. At the lateral edge of each pleuron, the cuticular
substance of which it is composed folds inward, and can be
traced on the ventral or sternal side for some distance. But
in the middle of the ventral region no indication of a sternum
is discoverable. It may, therefore, be concluded that the
sternal region of the somite was of a soft and perishable na-
ture ; and that the thoracic somite of a Trilobite resembled
one of the abdominal somites of a crab in this and in some
other respects.

The glabellum (Fig. 57, 4), or central raised ridge of the
cephalic shield, is a continuation of the thoracic axis^ the lo-
cation of its sides perhaps referring to the number of primi-
tive somites it represents. The limb, or lateral area on either
side, answers to a thoracic pleuron ; its thickened margin
(Fig. 57, 1) is produced into two longer or shorter posterior
angles {ff) ; inferiorly, the marginal band is reflected inward
for a short distance, as the subfrontal fold, the remaining
sternal area being incomplete. A median movable plate
answers to the labrum of Apiis and Limulus. On the occip-
ital or lateral margin of the limb a suture (Fig. 57, 5) com-
mences, and, passing between the eye and the glabellum,
meets that of the opposite side either in front of the latter,
or on the margin of the limb, or on the subfrontal fold, and
is connected with the labral suture by one or two sutures.
The limb is thus divided into two parts — one fixed (the fixed
gena, Fig. 57, «), attached to the glabellum ; the other sep-
arable (the movable gena, Fig. 57, h), on which the eye is
placed. The eves, are absent in some genera. In others they
occur as isolated ocelli ; or in groups, their interspaces being
occupied by the common integument ; or they may resemble
the compound eyes of other Arthropods,

THE TRILOBITA.

227

M. Barrande ^ has succeeded in tracing out the develop-
ment of some species of Trilobites. He finds that the small-

FiG. 57.— Diacrram of Dalmanites (after Pictet).— ^, head ; l,Tiiarcrinal band ; 2, mar-
ginal groove, internal to the band ; 3, occipital segment ; 4, glabellum ; 5, great
suture ; 6. eyes ; a, fixed gena ; ^>, separable gena ; g, genal angle ; B, thorax ; 7,
axis or terguni ; 8, pleuron ; C, pygidium ; 9, tergal ; 10, pleural portions of the
pygidium.

est, and therefore the youngest, forms are discoidal bodies,
without any clear evidence of segmentation. The division
into somites takes place by degrees, the number increasing
up to the adult condition. It is possible that still younger
conditions may have escaped fossilization, but the analogy
of Liinulus suggests that these small discoidal forms really
represent the condition in which the Trilobite left the e^g.

The Mekostomata.' — The only existing representative^ of
this division of the Crustacea is the genus Lirtxulus (the King
Crabs or Horseshoe Crabs), the various species of which are

^ " Systeme Silurien du centre de Boheme." tome i. Trilobites. 1852.
» H. Wood-n-ard, " A Monosrraph of the British Fossil Crustacea belonging
to the Order Merostomata," 1866.

228

THE ANATOMY OF INVERTEBRATED ANIMALS.

found in America and in the Moluccas. They are usually-
classed as a distinct order of the Crustacea^ termed ^ipho-
sura or Poecilopoda.

The body of Lhnulus (Fig. 58) is naturally divided into
three parts, which are movably articulated together. The
most anterior is a shield-shaped portion, curiously similar in
form to the head of a Trilobite. Its convex dorsal surface is
similarly divided into a median and two lateral regions; its
edges are thickened, and its posterior and external angles are
produced backward. At the anterior end of the median re-
gion two simple eyes are situated, and at its sides are two
large compound eyes. The sternal surface presents, ante-
riorly, a flattened subfrontal area^ behind which it is deeply
excavated, so that the labrum and the appendages are hidden
in a deep cavity formed by its shelving walls. The middle
division of the body of Lhnulus exhibits markings which in-
dicate that it is composed of, at fewest, six coalesced somites;
its margins are spinose, and its excavated sternal face lodges
the appendages of this region.

Fio. 58.—^, Linwlus moluccanus (dorsal view). 5. L. rotundlcauda (ventral view)
(after Milne-Edwards): a. anterior; b, mid 11? division of the body ; c, telson , d\
Bubfrontal area; e, anteunules ; f, antennae ; g, operculum; A, branchiferoua ap-
pendages.

The terminal division is a long, pointed, and laterally ser-
rated spine, which is termed the telson.

THE MEROSTOMATA. 229

The mouth is placed in the centre of the sternal surface of
the anterior division ; the anus opens on the same surface, at
the junction between the middle division and the telson. A
movable, escutcheon-shaped labrum projects backward in the
middle line, immediately behind the subfrontal area (d) ; and
on each side of it is a three-jointed appendage, the second joint
of which is prolonged in such a manner as to form with the
third a pincer or chela. The attachment of this appendage is
completely in front of the labrum, which separates it from the
mouth.

In each of the next five pairs of appendages, the basal
joint is enlarged ; and, in the anterior four, its inner edge is
beset with numerous movable spines. The attachment of the
basal joint of the foremost of these appendages (the second
of tlie whole series) is in front of the mouth ; but its pro-
longed, spinose, posterior and internal angle may be made to
project a little into the oral cavity. The basal joints of the
following three appendages are articulated at the sides of the
mouth, and the inner angle of each is provided with a spinose
process which projects into the oral cavity. The second,
third, fourth, and fifth appendages in the females are chelate ;
in the males of most species the second, and sometimes the
third, are not chelate. The large basal joint of the sixth ap-
pendage is almost devoid of spines, and bears a curved, spat-
ulate process, which is directed backward between the ante-
rior and middle divisions of the body. The fifth joint of this
limb carries four oval lamellae. The appendages which form
the seventh pair, very unlike the rest, are short, stout, and
single-jointed.

The eighth pair of appendages, again, are of a totally dif-
ferent character from those which precede them. They are
united in the middle line into a single broad plate, which
forms a sort of cover, or operculum, over the succeeding ap-
pendages, when the animal is viewed from the sternal side.
On the dorsal face of this plate are seated the two apertures
of the reproductive organs.

From the inner face of the anterior, or sternal, wall of
each half of the operculum a strong process arises, and passes
upw^ard to be attached to a corresponding process of the ter-
gal wall of the anterior division of the body. By far the
greater part of the large levator muscle of the appendage
arises from the tergal wall of the anterior division of the
body, and the nerve which supplies the limb is derived direct-
ly from the posterior part of the multiganglionate cord which

230 THE ANATOMY OF IIs^YERTEBRATED ANIMALS.

surrounds the gullet and supplies the appendages which lie
in front of the operculum.

The five pairs of appendages which remain resemble the
operculum in their general form, and have ascending process-
es, which are connected with inward prolongations of the ter-
gal wall of the middle division of the body. Their nerves are
derived from the ganglia which lie in this region of the body.

Thus there are altogether thirteen pairs of appendages,
eight of which are connected with the anterior, and five with
the middle division of the body ; and the appendages in the
region of the mouth are essentially ordinary limbs, the basal
joints of some of which are so modified as to subserve man-
ducation.

The determination of the homologies of the parts hither-
to spoken of as the anterior and middle divisions of the body,
and of their appendages, is a matter of some difficulty ; but,
on comparing the disposition of the limbs and their nervous
supply with what obtains in the higher Crustacea^ it seems
hardly doubtful that the first pair of appendages answer to
the antennules ; the second, to the antennae ; the third, to
the mandibles ; the fourth and fifth, to the maxilla? ; and the
sixth, seventh, and eighth, to the maxillipedes of Astacus or
Soinarus y and, in this case, the anterior division is a ceph-
alo-thorax. If the position of the genital openings marks
the hinder boundary of the thorax, the middle division of the
body represents an abdomen, composed cf five somites. But,
on the other hand, it may be that the genital organs open in
front of the hinder extremity of the thorax, as in female
Pod ophthalmia^ and that the five somites which form the
middle division correspond with the remaining five somites
of the thorax of a Podophthalmian. In this case, the region
which corresponds with the abdomen in the higher ciusta-
ceans is undeveloped.

The alimentary canal of Limiihis is very peculiarly ar-
ranged. The gullet passes directly forward and upward,
and gradually widens into the stomach, the walls of which
are provided with many longitudinal folds. The pylorus is
prolonged into a narrow tube which projects into the intes-
tine. The two biliary ducts on each side are far apart, and
branch out into minute tubules, which form a mass occupying
the greater part of the cavity of the body. The rectum, a
slender canal with plaited walls, and very short, opens into a
sort of cloaca situated between the telson and the sternal wall
of the abdomen.

THE MEROSTOMATA. 231

The heart, in Limidus polyphemus, is an elongated mus-
cular tube, divided into eight chambers, and having as many
pairs of lateral valvular apertures. It lies in a large peri-
cardial sinus, which, in its abdominal portion, presents on
each side five apertures, the terminations of the branchial
veins. The branchiae consist of numerous delicate semicir-
cular lamellae, attached transversely to the posterior faces of
the five post-opercular appendages, and superimposed upon
one another like the leaves of a book.

The nervous system appears, at first sight, to be very con-
centrated, its principal substance being disposed in a ring,
embracing the oesophagus ; but, on closer inspection, it is
found to consist of an anterior mass, representing the prin-
cipal part of the cerebral ganglia in most other Crustacea^
and of two ganglionic cords which proceed from the outer
and posterior angles of that mass, and extend as far as the
interval between the last and penultimate pairs of append-
ages. These cords are thick, and lie on each side of the
oesophagus, around which they converge, so as to come into
close union and almost confluence, immediately behind it.
In front of this point, however, they are connected by three
or four transverse commissures, which curve round the poste-
rior wall of the oesophagus, and become gradually shorter
from before backward.

The first of these commissures unites the tw^o cords oppo-
site the origin of the nerves to the third pair of appendages,
which I reg^ard as the homologues of the mandibles. In front
of this point, the cerebral ganglia give ofi', from their ante-
rior edges, the nerves to the ocelli, eyes, and frontal region ;
and, from their posterior and under surfaces, those to the an-
tennules. The nerves to the antennte arise from the cord
close to the outer and posterior angles of the cerebral gan-
glia, and some distance in front of those to the mandibles.
Close behind the latter arise the large nerves to the fifth and
sixth cephalo-thoracic appendages.

The nerves to the rudimentary seventh pair of append-
ages are slender, and arise rather from the under part of
the post-oesophageal ganglia ; those which suppl}" the eighth
pair of appendages, constituting the operculum, are also
slender, and seem to come off from the two longitudinal com-
missural cords, which connect the post-oesophngeal ganglia
with those which are situated in the second division of the
body, though they are, in truth, only united in one sheath
with them for a short distance, and can be readily traced to

232 THE ANATOMY OF INVERTEBRATED ANIMALS.

the post-oesophageal ganglia, internal to the nerves of the
seventh pair of appendages. The longitudinal commissures
are very long, and are inclosed in a continuation of the same
sheath ; they pass back into the second division of the body,
and there present four ganglionic enlargements, whence the
nerves of the post-opercular appendages proceed. The last
of these ganglia is much larger than the others, and appears
to consist of several confluent masses. The nerves diverge
from it in such a manner as to resemble a Cauda equina.

The reproductive organs of both sexes consist of a mass
of glandular ceeca, which ramify through the body amid the
hepatic tubules, and eventually open on papiilie situated on
the posterior face of the operculum. The males are much
smaller than the females, and present, in many species, an
external sexual distinction in the peculiarity of their second
and third appendages already referred to.

The young of Limidus acquires all its cliaracteristic
features while still within the Qgg. The interesting obser-
vations of A. Dohrn ^ have shown that, in an early stage, the
embryo is provided with the nine anterior pairs of append-
ages, and is marked out into fourteen somites by transverse
grooves upon its sternal face. The body has the form of a
thick rounded disk, divided into an anterior shield composed
of six somites, and a posterior, likewise shield-shaped region,
formed by the union of eight somites. The telson has not
made its appearance. In this condition, its resemblance,
apart from the limbs, to such a Trilobite as Trinucleiis is, as
Dohrn points out, most remarkable.

The JCiphosura were represented in the Carboniferous
epoch [Belllnurus),

Tlie Eicrypterida (Fig. 59) are extinct Crusiacea of Pa-
laeozoic (Silurian) age, which sometimes attain a very large
size and in many respects resemble Limidus^ while, in others,
they present approximations to other Crustacea,, especially
the Copepoda. An anterior, eye-bearing, shield-shaped di-
vision of the body is succeeded by a number (12 or more)
of free somites, and the body is ended by a broad, or narrow
and spine-like, telson. Five pairs, at most, of limbs, pro-

^ " Tlntcrsnchungen tiber Bau und Entwiclvclun;^ der Artliropoden." (Jena-
iscJie Zeifsclirift^ Bd. vi.) See also the observations of Lockwood and Packard,
American Naturalift, vol. iv., 1871, vol. vii., 1873, and " Memoirs of the Boston
Society of Natural History," 1872 ; with the discussion of the systematic place
of Zimulus by E. Van Beneden, Journal de Zoologie^ 1872.

THE MERCSTOMATA.

233

vided with toothed basal joints, are attached to the sternal
surface of the shield, and the mouth is covered, behind them,
by a large oval plate which appears to represent a meta-
stoma (Fig-. 59, -S, g). Some of the anterior limbs are fre-
quently chelate [Fterygotus) ; the terminal joints of the most

Fig. m.—Emimterm rsmims (after Nieszkowpki).i— J, do-sal aspect. B, ventral
aspect. Cth, t'le cepTialo-thoracic shield bearing a, the eyes, and 0, c. d, e,/, the
locomotive limhs; T, telson ; g, the metastoma; h, the sternal plates of the an-
terior free somites.

posterior pair are generally expanded and paddle-like. The
integument often presents a peculiar sculpture, simulating
minute scales. The sternal surface of one or more of the
anterior free somites is occupied by a broad plate, with a
median lobe, and two laterally-expanded side-lobes (Fig. 59,

1 " Der Enrypterm remipes^ au3 den obersilurisclien Schicliten der Insel
Oesel." 1859.

234 THE AXATOMY OF INVERTEBRATED ANIMALS.

3, A), having a remote resemblance to the operculum of
Limiilus,

The Extomostraca. — All the remaining Crustacea have
completely specialized jaws; and as many as six pairs of
appendages may be converted into gnathites.

In the Entomostraca, if the body possesses an abdomen
(reckoning as such the somites which lie behind the genital
aperture), its somites are devoid of appendages. Moreover,
the somites, counting that which bears the eyes as the first,
are more or fewer than twenty. There are never more than
three pairs of gnathites. The embryo almost always leaves
the eg^ in the condition of a Nauplius y that is, an oval
body, provided with two or three pairs of appendages, which
become converted into antennary organs and gnathites in the
adult. The division of the JEntomostraca comprises the
Copepoda^ the Epizoa^ the Branchiopoda^ the Ostracoda^
and the Pectostraca,

The Copepoda. — Tn these Entomostraca, which come
nearest to the Eurypterida^ the cephalic shield, which is dis-
coidal and not folded longitudinally, is succeeded by a certain
number of free thoracic and abdominal somites. The anten-
nules and antennae are large, and, as in the Eurypterida^ are
organs of locomotion and sometimes of prehension. The an-
terior thoracic members are converted into foot-jaws ; the
posterior serve as paddles, the limbs of each pair being often
united together in the middle line, as in Limulus. The em-
bryo leaves the ^^^ as a jSTauplhis.

The various species of the genus Cyclops^ which abound
in fresh water, afford excellent illustrations of the structure of
the Copepoda,

The minute animal (Fig. 60) is shaped something like a
split pear, the larger end corresponding with the head, and
the convex side with the dorsal surface. The anterior third
of the body is covered by a large carapace, which, at the sides,
extends downward as a free fold over the bases of the ap-
pendages, but is hardly at all free posteriorly. Anteriorly,
in the middle line, it curves forward and downward, and is
produced into a short rostrum, on each side of which a con-
siderable excavation lodges the base of the long antennule,
by tlie vigorous oar-like strokes of which the animal darts
through the water. At the anterior boundary of the head,
the double, black, median eye, which, unless very closely ex-

THE COPEPODA.

235

amined, appears single, shines through the carapace, and at
the sides of the latter, two coiled tubes with clear contents,
the so-called shell-glands^ are seen.

Four distinct and movable somites succeed the carapace,
and gradually diminish in diameter. The body then suddenly
enlarges, and becomes divided, in the female, into four seg-
ments, the last of which gives attachment to two long setose
styles, which possibly represent another somite. There is a
well-developed and prominent labrum (or conjoined epistoma
and labrum) in front of the mouth, and behind it is a bilobed
metastoma. The first pair of appendages are the long and

Fig. 60.— Cyclops. — Side-view of an adulf female carryinjj a pair of ovisacs, and ven-
tral view of the head, showing tlie labrnm. metastoma and appendages of the left
side. 1', eye ; II', antennule ; III', antenna; IV', mandible; V, "first maxiila ;
YI', second maxilla (erroneously marked YII') : a. outer ; b, inner division ; 1, 2,
3, 4, 5, thoracic limbs ; B, rostrum ; lb, labrum ; mt, metastoma.

many-jointed antenniiles, which are the chief organs of loco-
motion. These are succeeded by the short and few-jointed
antennge. The third pair of appendages, or first pair of gna-
thites, differ from the corresponding limb in Limidus in the re-
duction of the greater part of the appendage to a rudiment
terminated by setse, while the strong basal part is the princi-
pal gnathite or mandible. The second pair of gnathites are
strong and incurved ; following upon these is a third pair of

236 THE ANATOMY OF INVERTEBRATED ANIMALS.

appendages, each divided into two portions, an inner and an
outer. The latter is by far the larger, and is so constructed
that the three distal articulations can be bent back upon the
proximal ones, and opposed to the internal division, consti-
tuting a prehensile organ, the " hand " of Jurine/ Thus the
gnathites of Cyclops are a pair of mandibles followed by two
pairs of maxillae. x\t some distance behind the third pair of
gnathites the first pair of thoracic appendages is attached to
the hinder part of the cephalo-thorax. Each consists of a two-
jointed basal part (2)rotopodite), terminated by two three-
jointed divisions {exo2Dodite and endopodite). Three similar
pairs are appended to the three anterior free somites, while
a fifth rudimentary pair is connected with the next and small-
est of these somites. The suddenly-enlarged following seg-
ment of the body carries the apertures of the reproductiv^e
organs in the female, and supports the ovisacs. It is com-
monly regarded as the first abdominal somite ; but, according
to Claus, it is composed of tw^o distinct somites, which be-
come united only after the last moult.

The alimentary canal is straight and simple, and without
any distinct liver. There is no heart nor any special respira-
tory organ.

The single ovary, situated in the thorax, is provided with
two oviducts, which open on the sides of the coalesced first
and second abdominal somites. On the ventral face, between
the apertures of the oviducts, is the median aperture of a
colleterial gland which secretes the viscid matter which forms
the coat of the ovisac. Short lateral ducts connect the gland
with the extremities of the oviducts.

The male is much smaller than the female, and the tw*o
enlarged somites of the abdomen remain distinct. There ic
a single testis provided with two vasa deferentia. A special-
ly glandular portion of the latter secretes the material of
the spermatophores, or cases in which the spermatozoa are
inclosed. The antennae are thickened, and provided with a
peculiar hinge-joint, by means of which the male firmly
seizes the fourth pair of swimming-legs of the female during
copulation, and then, bending np his abdomen, deposits two
of the spermatophores on the median opening of the colle-

» That these are two divisions of the third g-nathite, and not two separate
appendaa^es, has been demonstrated by tracing; out their development. ( Claus,
"Organization und Verwandtschaft der Copepoden," Wurzburger naUirwiss.
Zeitschrift ^ 1862.) Under these circumstances 1 do uot know why they should
be termed '' maxillipedes."

THE EPIZOA. 237

terial gland, into which the spermatozoa pass on their way to
the oviducts. The gland thus plays the part of a sperma-
theca. The eggs are carried about in the ovisacs until they
are hatched.

The vitellus undergoes complete division, and a morula
results, the blastomeres of which soon become differentiated
into a superficial epiblast, surrounding a deeper-colored mass,
which gives rise to the h^^poblast and mesoblast. The whole
embryo then becomes divided by two constrictions into three
segments, and the hypoblast arises by delamination around
a central cavity, which becomes the alimentary canal. There
is a large labrum on the ventral side of the first segment in
front of the mouth. The eye appears on the tergal aspect of
the most anterior segment, as two pigment-spots which soon
coalesce into one ; and a pair of jointed setose limbs grows
out of each segment. In this N^auplius-stSitQ the young Cy-
clops leaves the Qgg,

The posterior part of the body elongates and becomes
divided into the somites of the thorax and abdomen, from
which their respective appendages bud out ; and these
changes are accompanied by exuviation of the cuticle. The
three pairs of appendages of the Nauplius are converted into
the antennules, antennae, and mandibles of the adult.

There are a few other fresh-water and many marine
genera of Copepoda, Among the latter, the Pontellidoe are
remarkable for the separation of that part of the head which
bears the antennules and the antennae, from the rest, a pecu-
liarity to which a parallel can be found only among the Sto-
tnatopoda. Corycmus has two large, more or less lateral
eyes in addition to the median eye, subchelate antennae, and
a rudimentary abdomen. The beautifully iridescent Sapphi-
rina has an extremely depressed body, short filiform an-
tennae, two eyes, and rudimentary gnathites. A short tho-
racic heart is present in some genera.

The Epizoa. — Insensibly connected by such genera as
Ergasilus and Callgus with the typical Copepods, are a great
number of very singular Crustacea^ which, from their habit
of living parasitically upon aquatic animals (whence their
vulgar name of " fish-lice "), have received the title of Eirl-
zoa, C ho ndr acanthus gibbosus^ commonly found in great
abundance on the walls of the branchial chamber of the
Fishing-frog {Lophius inscatorhts), may serve very well as

238 THE ANATOMY OF INVERTEBRATED ANIMALS.

an illustration of the most remarkable peculiarities of this
aberrant group.

The female (Fig. 61) is not more than half an inch long,
but, posteriorly, two long slender cylindrical filaments (like
the rest of the animal, of a whitish or yellowish color) are
attached to its body, which is broad and flattened, and as it
were crimped at its edges, so as to present two principal
transverse folds. The angles of the folds are elongated into
lateral processes (A, ^, /"), and similar processes (c?, e) proceed
from the middle line of the body, which by these outgrowths
and foldings becomes singularly distorted ; and the grotesque-
ness of the animal's appearance is not a little enhanced by
the bowing motion, accompanied by a flapping backward and
forward of its gouty limbs, which it executes when detached
from the integument of the Lo'phius.

The head is expanded into a sort of hood, the convex
anterior margin of which bears the antennules and antennae,
the latter being metamorphosed into the strong curved hooks
by which the Chondr acanthus is securely anchored to the
infested animal. A subquadrate labrum overhangs the mouth,
but does not inclose the mandibles and form a suctorial ap-
paratus, as it does in some Epizoa.

The mandibles and the two pairs of maxillse resemble
curved hooks or claws. Two pairs of appendages (Fig. 61,
5, c), composed each of a protopodite, terminated by an endo-
podite and exopodite and exhibiting hardly any trace of
articulation, are attached to the anterior part of the body
behind the head.

The body ends in a rounded segment, situated in the deep
notch between the hindermost marginal processes, and bear-
ing the two projecting vulvae. Above each of these is a small
triangular papillose lobe (Fig. 62, w), probably a modified ap-
pendage, to which, as we shall see, the male attaches himself,
while below them are two other rudimentary appendages
(Fig. 62, y). The alimentary canal is a straight tube running
from the mouth to the opposite extremity of the body. No
heart is discoverable, and the nervous system and organs of
sense (if any) are equally undistinguishable. The interspace
between the alimentary canal and the walls of the body is
almost wholly occupied by the ovarium, which consists^ of
four tubes, situated on each side of the intestine, and giving
off ramified caeca, in which the ova are developed. ^ Ante-
riorly, each pair of tubes opens into the oviduct of its side,
which passes down along the side of the body to terminate at

CHONDRACANTHUS GIBBOSUS.

239

the vulva. The lower part of the oviduct contains a clear
gelatinous substance, and is very similar in aspect to the ce-
ment duct of a cirripede ; this substance is secreted by the

Fig. Gl.—Chondracanthus ffibbosvs.— Female: A, lateral view. JB, ventral view, en-
larged : a, head ; b, c, appendages ; </, median dorsal process ; e, median ventral
processes ; /, i, h, lateral processes; k, terminal segment; I, male ; g, ovisacs ;
m, n, medio-doreal ovarian tubes ; p, lateral ovarian tubes ; o, oviduct. 2, 3, an-
tennules ; 4, 5, 6, antennae, gnathites.

walls of the oviduct, and forms the walls of the ovigerous sac.
The latter, as has been stated, has the form of a long cylin-
drical filament, the upper end of which is firmly held between
the prominent lips of the vulva (Fig. 62, x).

The male Chondr acanthus does not attain to a twelfth the
length of the female, and looks, at first, like a papilla upon

240 THE ANATOMY OF INVERTEBRATED ANIMALS.

her body near the vulva. On close examination, however, he
is seen to be firmly fixed by his antennary hooks to one of the
two triangular lobes described above. The hooks are doubt-
less at first attached to the lobe by muscular contraction ; but
the connection once efi'ected seems indissoluble — at least ma-
ceration in caustic soda does not cause the male to become
detached. It does not appear that more than one male is
attached to a female.

The body of the male (Fig. 62) is pyriform, and exhibits
indications of a division into six segments beside the head.

Fio. 62.— (7, Male Chondr acanthus, in situ, enlarged : x, vulvae of female ; w, tnan-
gular papillose lobes ; q, antennae of male ; r, eye-spot ; t. testis ; u, vas deferens;
V, genital aperture ; y, rudimentary appendages of the female ; g, ovisace.

The anterior extremity presents a black eye-spot imbedded
in its substance, and gives origin to a pair of rudimentary
antennules, and to the strong, hooked, prehensile antennae.
Behind and below them is a large labrum and three pairs of
hook-like gnathites. These are succeeded by two pairs of
subcylindrical appendages, which apparently represent ambu-
latory limbs.

The caudal extremity is terminated by two styles, and
there are two prominent tubercles on the ventral surface of
the penultimate somite, in which the genital apertures are
seated. The alimentary canal is a delicate, irregular tube,
having many brownish granules imbedded in its walls. A
wide oesophagus is connected with its anterior extremity ;
but the opposite end appears to be rounded, and to be united
with the ventral surface of the integument only by connec-
tive tissue. A complex muscular system, composed of striped
fibres, is visible through the integument, and the eye-spot

CHOXDRACANTHUS GIBBOSUS. 241

seems to be connected with a subjacent ganglionic mass.
The body is suificiently transparent to allow the pulsations
of a heart to be seen, but none can be discovered. The testis
is a large oval bilobed mass (t), lying like a saddle upon the
anterior part of the intestine. From this body a thick vas
deferens runs back upon each side of the intestine, and di-
lates in the penultimate and antepenultimate somites into a
thick walled pyriform sac — a sort of vesicula seminalis. The
embryo leaves the egg as a N^aujylius^ like that of Cyclops,

There are many genera of these parasites, some of which,
such as the almost completely vermiform Lernmm^ deviate
even more widely than Chondr acanthus from the ordinary
form of Crustacea^ while others, such as Ergasilus and N'oto-
delphys^ differ but little from the free Copepoda.

In Caligus, the labium and metastoma are elongated and
united into a tube in which the sharp styliform mandibles are
inclosed ; and from the prevalence of this suctorian form of
mouth in some of the best known species of parasitic Cope-
poda, they are frequently termed "suctorial" crustaceans.
Suctorial disks for attachment are developed from the coa-
lesced posterior pair of thoracic members in Achtheres ; and,
in this genus, the head, as a distinct part, becomes almost
entirely obsolete.

Argulus, the parasite so common on the Stickleback, is
worthy of notice as one of the most curious modifications of
the epizoic type.* It is extremely flattened, and is composed
of an anterior cephalo-thoracic disk, behind which lies a very
short and broad, notched abdomen. A median styliform
weapon lies in a sheath in front of the mouth, and the small
mandibles and maxillae are inclosed in a short tube formed
by the labrum and the metastoma. Six pairs of appendages
lie behind the mouth, the anterior being metamorphosed into
suckers, the next pair into strong limbs with a toothed sec-
ond joint, and the four others constituting biramous swim-
ming-feet. There are two pairs of antennary organs, and two
compound eyes. According to Leydig, the males are pro-
vided with cups on their penultimate swimming-feet ; and,
during copulation, these are filled with the seminal fluid,
which is thus transferred to the vulva of the female, and
thence to the spermatheca. The eggs are laid, and not car-
ried about in ovisacs. The larva is provided with two pairs

^ Claus (" Ueber die Entwickelunor, Orfiranization und systemntische Stellung
der Ar^uliden," 1875) has proved the close affinity of Argulus with the Cope-
poda, but proposes to regard it as the type of a special group, the Branchiura,

242 THE ANATOMY OF DsYERTEBRATED ANIMALS.

of principal swimming appendages, the future antennae and
the mandibular palps, the latter eventually entirely dis-
appearing. There is a pair of small antennules, a pair cf
strong legs in the place of the suckers, and, behind them, the
rudiments of the prehensile legs and the first pair of bira-
mous appendages, the others being rudimentary.

Kotodelphys^ which may be found very commonly in the
branchial sac of A^scidians^ closely resembles an ordinary
Copepod, except that it becomes much distorted, and that
it carries its ova in a chamber formed by the dorsum of the
carapace.

However strangely modified the adult form may be (and
it must be remembered that it is always the female which un-
dergoes the greatest amount of Change), the larvae of all
these epizoic parasites resemble those of the ordinary free
Copepoda in possessing only two (Achtheres, Tracheliastes)
or three pairs of appendages (which appertain to the anterior
region of the head) ; and they are endowed with considerable
powers of locomotion.

The Branchiopoda. — The genera Kebalia^ Apus, Bran-
cJiipus, Limnetis^ Daphnia^ and their allies, are usuall}' di-
vided into two ordiev?,^ the Phyllopoda and ihe Cladocera ^
but these pass into one another so gradually, and have so
many structural peculiarities in common, that the subdivision
of the group of Branchiopoda appears to me to be a step of
doubtful propriety. Closely resembling the lower Podoph-
thcdmia^ such as 3fysis, in some respects, these Crustaceans
are invariably distinguished from them by the possession of
a greater or less number of somites than twenty ; Kehalia,
which most nearly approximates the higher Crustacea^ hav-
ing twenty-two somites. Furthermore the thoracic and ab-
dominal appendages of the Branchiopoda are, in the majority
of cases, more or less foliaceous, resembling in many respects
the anterior maxillipede of an Astactis, and being constructed
on essentially the same plan.

Apus glacialis (Fig. 63) presents an elongated vermiform
body, terminated by two long, multiarticulate, setose styles,
and covered anteriorly by a great shield-like carapace, deeply
excavated behind. The posterior three-fifths of the carapace
are free, and merely overlap the segments of the body ; the
anterior portion, on the contrary, is united with and forms
the tergal surface of the corresponding region of the head ;
the free portion of the carapace shelves away laterally from a

THE BRAXCHIOPODA. 243

median ridge, on each side of which a curious concentric mark-
incr, indicating the position of the shell-gland (Fig. 63, B^ x),
is visible. This gland is a coiled tube with clear contents,
which, according to Glaus, opens on the base of the first pair
of thoracic appendages, immediately behind the second max-
illas. Where the free joins the fixed portion of the carapace,
the ridge is abruptly terminated by a transverse depression.
A little distance in front of this is another deeper transverse
groove, close to which, in the middle line, are the two reni-
form compound eyes, converging toward one another ante-
riorly (Fig. 63, ^, i').

The ventral surface of the anterior division of the carapace
(Fig. 63, C) presents a flattened, semilunar, subfrontal area,
as in Limulus, behind which it slopes upward on all sides
into the posterior division, thus forming a wide chamber, in
which the anterior thoracico-abdominal segments are lodged.
In the middle line, the subfrontal plate sends back a long and
wide process, movably articulated with it, and rounded at its
free end — the labrura ; above and behind which the mouth
and gnathites are situated. Behind these follow twenty-six
spinulose thoracico-abdominal segments ; the anterior twenty
of which bear the swimming-feet, while the twenty-sixth,
much larger than the others, is produced into an incurved
point posteriorly, and carries the anus and the terminal setae.

The compound eyes, as has been said, are seated upon the
upper surface of the anterior division of the carapace. On the
under surface, just above and behind the posterior boundary
of the subfrontal area, and on each side of the labrum (Fig. 63,
C, lb)y is a delicate jointed filament — the antennule (Fig. 63,
C, ii'). Behind this Zaddach found, in some specimens of
Apus cajicriformis^ a second very small filament, the rudiment
of the antenna, which in the larva is so large and important
an organ ; but I have observed nothing of the kind in A. gla-
cialls. On each side of the labrum is a large, convex, strong,
toothed mandible, and the aperture of the mouth is bounded
posteriorly by a profoundly divided plate, the metastoma.
Succeeding this are two pairs of small maxillae, the second
pair being foliaceous, and almost rudimentary. Behind these
appendages, a cervical fold marks ofi" the boundary between
the head and the thorax, and at the same time corresponds
with the commencement of the free portion of the carapace.
Whether the carapace is also to a certain extent attached to
the first thoracic somite, as Grube states,* or whether it is en-

1 ((

Bemerkungen iiber die Phyllopoden," p. 81.

244 THE ANATOMY OF INVERTEBRATED ANIMALS.

tirely cephalic, as Milne-Edwards considers, is a point upon
which I have been able to come to no very clear determina-
tion ; indeed, it is a question rather for the embryologist than
the anatomist.

Of the twenty pedigerous segments, the first eleven have
each one pair of appendages ; but, behind the eleventh, each
segment gives attachment to a gradually increasing number
of limbs, so that the twentieth carries live or six pairs. Alto-
gether twenty-eight pairs of appendages are attached to these
nine posterior thoracic segments ; these, added to the eleven
preceding, make thirty-nine appendages in all. While each
of the anterior eleven segments must be regarded as single
somites, the nature of the posterior ones is open to doubt ;
they may be single terga, the sterna and appendages of which
have multiplied; or, more probably, they each represent a
number of coalesced terga.

Each appendage consists of three divisions — an endopo-
dite, exopodite, and epipodite, supported on a protopodite
or basal division (Fig. 63, Z>, E^ F). The latter consists of
three joints — a coxopodite produced internally into a strongly
setose prominence (not represented in the figures), a basi-
podite, and an ischiopodite, the latter elongated internally
into a lanceolate process, and bearing on its outer side two
appendages, of which the proximal — the epipodite or branchia
(Fig. 63, J^, E, 7) — is pyriform and vesicular in specimens
preserved in spirit. The distal appendage, which appears to
represent the exopodite (6), is a large flat plate, provided
with long setse on its margin.

The endopodite consists of four joints, the two proximal
ones being much the longer, and, like the penultimate, giving
off internally a long process. Finally, the terminal joint is
claw-like and serrated on its concave edge.

The average form of these appendages is represented by
(jE'), taken from the middle of the series; anteriorly the limbs
become more slender and leg-like {D)\ posteriorly, on the
other hand, they are completely foliaceous, as {F) ; but the
same elements are recognizable throughout.

The eleventh pair of appendages alone depart, in any im-
portant respect, from the rest of the series, each of these
being modified so as to serve as a receptacle for the ova.
To this end the joints of the endopodite are greatly ex-
panded, and converted into a hemisjDberical bowl ; the exo-
podite, metamorphosed into another such bowl, shuts down
over the endopodite ; and into the box thus formed the

TEE BRANCHIOPODA.

245

Fig. GS.—Anux glacialis.—A, Lateral view, with the riijht half of the carapace cut
away. B, Dorsal view. C, Anterior part of the body, ventral aspect. A One of
the anterior, E, one *)i the middle, and F, one of the posterior limbs, without their
coxopodites. x, convoluted " shell-L'land " in the carapace; y, caudal filament;
lb, labrum. 1, 2, 3, 4. Endopodite. 6. Exopodite. 7. Epipodite or branchia. F,
eye ; IF, auteniiule ; IV', labrum ; Y', VF, maxillae.

246 THE ANATOMY OF INVERTEBRATED ANIMALS.

ova are conducted by means of the oviduct, which opens
into it.

On the dorsal surface of each side of the terminal seg-
ment of the body there is a tubercle produced into five spines
anteriorlv, and carrying, posteriorly, a long and delicate se-
tigerous filament (Fig. 63, jB, g).

The alimentary canal of Apus is very simple, consisting
of a vertically ascending oesophagus, which bends back into
the small stomach, situated immediatelv behind the com-
pound eyes, in the middle of the region bounded by the two
transverse furrows on the dorsum of the carapace ; from the
hinder end of the stomach the straight intestine passes back
to the anus, which is seated beneath the terminal segment.
The liver consists of caeca, which branch off from the stomach
and lie on each side of it, in the head. Zaddach describes a
pair of glands which he regards as salivary, placed above,
and opening into, the stomach itself, like the salivary glands
of the Scorpion.

The heart occupies the tergal region of the eleven ante-
rior thoracic somites, presenting as many chambers, with lat-
eral venous apertures.

The nervous system consists of a quadrate cerebral mass,
placed immediately under the compound eyes, and giving off
large nerves to them and to the remains of the single eye of
the larva, which lies in front of their anterior extremities.
Commissures pass downward and backward on either side of
the oesophagus, and connect the cerebrum with a chain of
numerous ganglia placed on the median line of the ventral
surface. It is worthy of remark, that the antennary and
antennulary nerves are given off from the commissures, far
behind the chief cerebral mass.

In the female, the ova are developed in the caecal branches
of two long tubes, situated one on each side of the bod}'', and
opening, as above described, in the eleventh pair of append-
ages. Apus usually propagates agamogenetically, and the
examination of thousands of individuals, extending over more
than thirty years, failed to reveal to Von Siebold the ex-
istence of a male form. In 1856, however, Kozubowski ^ dis-
covered a small proportion of males (16 in 160), among the
specimens taken in the neighborhood of Cracow ; and near
Rouen, in 1863, Sir John Lubbock found the largest pro-

» " Ueber den munnlichea Apus cancri/ormis." (" Archiv fiir Naturge-
scliichte," 1857.)

THE BRAXCHIOPODA. 247

portion of males to females yet known, viz., 33 in 72. On
the other hand, between 1857 and 1869, Von Siebold ex-
amined many thousands of specimens of the Bavarian A.pus
without finding a single male/

The testis is similar to the ovary in form, and its duct
opens upon the eleventh pair of appendages, as in the case of
that of the female organs. The spermatozoa are oval and
without motion.

The young Apus (cancriformis)^ when just hatched, is a
N^aupllus. The body is oval, indistinctly divided into a few
segments, and entirely destitute of appendages, except a
shorter anterior, uniramous, and a longer posterior, biramous,
pair of oar-like organs, situated at the anterior extremity, on
either side of the single median eye. The carapace is rudi-
mentary, and there are no caudal filaments. The little ani-
mal soon casts its skin, and the mandibles, which are provided
with long palps, make their appearance.'* With successive
ecdyses, the larva assumes more and more the form of the
adult, and acquires the pair of compound eyes ; the anterior
pair of appendages being converted into the antennules, the
posterior pair disappearing, or remaining as rudimentary an-
tennae, and the mandibular palps also vanishing.

Singular and highly instructive modifications are exhib-
ited by the other genera of the Branchiopoda^ such as Keha-
lia, Dranchipus {Cheiroc€phalns)^JLimnetis, and Daphnia.

In Daphnia and its allies (Fig. 64), the thoracic members
are reduced to six, five, or even four pairs, some or all of
w^hich may take the form of ordinary limbs ; the abdomen is
rudimentary ; the heart is short ; and the carapace presents
a posterior division (omostegite), obviously developed from
the anterior thoracic somites, the lateral halves of which are
deflexed so as to resemble a bivalve shell, into which the
hinder part of the body can be withdrawn. The anterior
division of the carapace (cephalostef/ite) in Daphjiia has, on
the contrary, the same structure as the corresponding part of
the carapace of Apus, but the compound eyes, represented
by a single mass, are situated at the anterior extremity of
the head, rather than on its upper surface, and the single eye
is quite distinct, and far posterior to them (Fig. 64, J5, i',
ii"). The antennules (Fig. 64, A, ii') are small, rudimentary,

» " Beitrage zur Parthenogenesis der Artbropoden,'' 1871. It appears that,
in Apys^ the impreernated ova alone give rise to males,

' According to Claus's recent investigations, this third pair of appendages
Is present from the time the young Apus leaves the egg.

248

THE ANATOMY OF INYERTEBRATED ANIMALS.

and placed at the sides of the produced frontal rostrum, but
the antennae are very large, and constitute the principal loco-

Fi6. Gi.—Daphnia. — A, Side-view; the appendages not fieured except IP, the an-
tennules; IV', the mandibles; and V, the maxillae. IIP, The place of attach-
ment of the antennae. £, Front view of bead.

cs, cephaloFtegite, or that part of the carapace which covers the head ; ms, omoste-
glte, or thoracic portion cf the carapace ; c, heart ; sf, cervical depression ; lb,
Jabrum ; P, compound eye; IP, simple e>e ; «, the ''shell-gland," which opens
behind the maxillte.

motive organs. The posterior, or second, maxiUse are obso-
lete. In I^vachie, JPolyphernus, aS'/c/ot, and other genera, sucker-
like organs of adhesion are situated on the anterior region
of the carapace. The eggs are developed in the cavity of the
carapace, and the embryos pass directly into the form of the
parent, except in JLeptodora, where they are, at first, N^auplius-
like.

Limnetis and Estheria present a Daphiiia-MVe carapace,
though more completely bivalve, combined with the numer-
ous segments of the body and the foliaceous appendages of
the typical Phyllopods (Fig. 65).

Nebalia has a large carapace, provided with a movable
rostrum, like that of Sqidlla^ and arising entirely from the
head, which is remarkable for its very slight sternal flexure.
In this genus the eyes are large and pedunculated ; there
are well-developed antennules, antennae, mandibles, and two
pairs of maxillae, the anterior of which ends in a long palp.

Branchipxis^ finally, develops no carapace either from

THE BR.VNCHIOPODA.

249

the head or the thorax, the segments of the latter being en-
tirely free, while the former is similar in shape to that of an
Insect, or Edrioplithalmous Crustacean, and carries two large
stalked eyes, two antennules (singularly modified in the
male), two antennae, a pair of mandibles, and two pairs of
maxilljie.

In Estheria and Limnetis, the males are met with in full
proportion to, and may be even more numerous than, the
females. No males are known in Llmnadia gigas^ although
thousands have been examined, while, in X. ^tanleyana^
more males than females have been found. In JBranchipus^
males are fewer than females ; in Artetma, they occur only
at rare intervals. In Daphnia^ the males are few, and appear

Fig. ^5.—Limnetis brachyurus (after Grnbe).— The upper left-hand figure is the
male, the other the female; one valve of the carapace in each case being removed.
A\ Antennules. A^, Antennae. A, Young larva. B, The same further advanced,
c, head ; o, eye ; d, carapace : &, bodv. A '. Antennae. J/, Mandible?, d', great
plate (labrum?) which covers the mouth.

only at certain seasons of the year. But notwithstanding
the rarity or absence of the males in many of these genera,
reproduction proceeds with great rapidity. The ova are capa-
ble of development without fecundation ; and isolated females

250 THE ANATOMY OF INVERTEBRATED ANIMALS.

of the genus Daphnia will thus go on producing broods for
generation after generation, without any known limit/

Under certain circumstances, however, bodies of a differ-
ent nature from these *' agamic ova," as they have been well
termed by Sir John Lubbock,^ are developed within the
ovary, the substance of which acquires an accumulation of
strongly refracting granules at one spot, and forms a dark
mass, the so-called " ephippial ovum." When fully formed,
two of these bodies pass into the dorsal chamber of the cara-
pace, the walls of which have, in the mean time, become
altered. The outer and inner layers of the integument ac-
quire a peculiar structure, a brown color, and a more firm
consistency, over a large, saddle-like area. When the next
moult takes place, these altered portions of the integument,
constituting the " ephippium," are cast off, together with the
rest of the carapace, which soon disappears, and then the
ephippium is left, as a sort of double-walled spring-box (the
spring being formed by the original dorsal junction of the
two halves of the carapace), in which the ephippial ova are
inclosed. The ephippium sinks to the bottom, and, sooner
or later, its contents give rise to young Daphnim.

Jurine's and Sir J. Lubbock's researches have proved that
the development of the ephippial ova may commence with-
out the influence of the male, and thev seem to indicate that
these ova may even be fully formed and laid without the
male influence. On the other hand, there appears, under ordi-
nary circumstances, to be a certain relation between the com-
plete development of ephippial ova and the presence of males ;
and, as yet, no ephippial ova produced by virgin females have
been directly observed to produce young. The question,
therefore, seems to stand thus, at present: the agamic ova
may certainly be produced, and give rise to embryos, without
impregnation ; the ephippial ova may certainly be produced
without impregnation ; but whether impregnation is or is not
absolutely necessary for their further development, there is,
at present, no evidence to show.

The great majority of the BrancMopoda inhabit fresh
waters. Artemia, however, delights in brine-pools. The
genus JEstheria is of Devonian age, and it seems probable

i"Ueber die Gaitun^Qn EstJieria und Limnadia.'''' ("Arcliiv fiir Natur-
peschlclite," 1854.)

2 " An Account of the Two Methods of Reproduction in DapTinia^ and of the
Structure of the Ephippium." (" Transactions of the Royal Society," 1875.)

THE OSTRACODA. 251

that the Silurian Symenocaris and its allies were related to
Apus.

The Ostracoda. — This group contains several genera of
both recent and fossil Crustacea^ for the most part of very
small size, and distinguished by their hard, often calcified,
and completely bivalve shell, provided with a distinct hinge.
The valves of this shell consist of the lateral moieties of the
carapace ; they are commonly unequal and unsymmetrical,
and present a peculiar ornamentation. The shell-gland is
very small. The Ostracoda are also remarkable for the ex-
tremely rudimental condition of their abdomen, and for the
paucity of their thoracic appendages, which, instead of being
foliaceous, are strong and subc^dindrical, like the ambulatory
legs of the higher Crustacea.

The cephalic flexure is as well marked as in the highest
Crustacea^ so that the eye, obscurely divided, and m.edian in
Cypris (Fig. 66, A), but double and lateral in Cythere (JB),
is situated in the upper part of the anterior region of the body.
The antennules and antennae, attached to their respective
somites, the sterna of which constitute the anterior boundary
of the body, are similar in form and function to ambulatory
limbs. The ducts of a peculiar gland open, according to
Zenker, at the end of the strong spine with which the an-
tenna of Cythere is provided. The labrum is conspicuous,
and the mandibles nre strong, and possess a well- developed
palp. The first maxilla is provided with a large foliaceous se-
tose appendage (epipcdite ?). The second maxilla in Cythere
is represented by the first of the three pairs of ambulatory
limbs (Fig. 66, JB, e, e, e) present in this genus. In Cypris,
which possesses a second pair of maxillae, there are only two
pairs of ambulatory limbs (Fig. 66, A, p, i., ii.). The aper-
tures of the reproductive organs, provided in the male with a
wonderfully complex, horny, copnlatory apparatus (described
with great minuteness by Zenker), are situated between the
last pair of thoracic members and the large caudal hooks.

Strong adductor muscular bundles pass from one valve of
the carapace to the other, and leave impressions discernible
from without, the form and arrangement of which furnish
valuable systematic characters.

The alimentary canal of the Ostracoda is provided ante-
riorly with an apparatus of hard parts, resembling in many re-
spects the gastric armature of the Isopoda, and gives origin
to two hepatic caeca. Cypris and Cythere have no heart;

252

THE ANATOMY OF INVERTEBRATED ANIMALS.

but, in Ci/jyridlna, Conchoecia, and Halocryptis there is, ac-
cordins: to Claus, a short saccular heart with one anterior and
two lateral apertures. The nervous system is difficult to make
out ; but, in Gythere lutea, the same observer found a large cer-
ebral ganglion in front of the mouth, whence filaments passed
to an ophthalmic ganglionic mass, and to the antennary or-
gans. A double ganglion, behind the mouth, supplies the
gnathites ; three ganglia, situated in the thorax, send fila-
ments to its appendages, and a terminal ganglion supplies the
caudal appendage and genitalia. In the female, the ovaries
lie in the valves of the carapace, and terminate in oviducts
which open by distinct apertures in front of the caudal ap-
pendage. Immediately anterior to them are the openings of

Fig. 66.—^. CypHs.—A. i. it. Antennules and AntennsB. M. i. n . m. Mandibles

and maxilhe. P. i. ir. Thoracic members ; c, caudal extremity ; 6, mandibular

palp ; o, eye. B. Maxillary appenda-je.
B. Cytfiere.—o, eye ; a, antennule ; 6, antenna ; c, mandible ; d, first maxilla ;

e, e, e, second maxilla and two thoracic members ; /, caudal extremity. (Alter

Zenker.) ^

two horny canals, called vaginae by Zenker, each of which is
continued into a long convoluted transparent tube, and event-
ually terminates in a large vesicle, the spermatheca, into
which the spermatozoa of the male are received.

In the males, the antennae, the second maxillae or some of
the thoracic limbs, are modified in such a manner as to enable
them to seize and hold the females. The testes are elongated
caeca in Cypris, globular vesicles in Cythere, and communi-
cate with a long vas deferens, which opens into the copula-
tory apparatus. In Cypris, a very singular cylindrical mu-
cous gland is connected with the vas deferens '; but perhaps
the most remarkable peculiarity about the genital apparatus
in the male consists in the size of the spermatozoa, which in
Cypris ovum are, according to Zenker, more than three times
as long as the body. They possess a spirally-wound coat, and
are totally deprived of mobility.

» " Mono^rapliie der Ootracodon." (" Archiv far Naturgesdiichtc," 1854.)

THE PECTOSTKACA. 253

The Ostracoda either attach their eggs to aquatic plants,
or carry them about between the valves of the carapace.

Glaus* hac worked out the development of Cypris^ which
passes through nine successive stages, distinguished from one
another, not merely by the shape of tlie carapace, but by the
number and form of the limbs. An ecdysis of the chitinous
cuticle of the body and carapace terminates each stage of de-
velopment. When the Cypris leaves the ^g%^ it resembles a
Naupjlixis^ in possessing a single median eye and only three
pairs of limbs (the future antennules, antennae, and mandi-
bles) ; but none of these are divided into two branches. The
body is laterally compressed and has a bivalve carapace.

The changes undergone by the marine Ostracoda after
they leave the Q^'g are much less marked.

Fossil Ostracoda abound in strata of all ages, from the
older palaeozoic formations onward ; and, so far as the char-
acters of the carapace furnish evidence, the most ancient
forms differed very little from those which nov7 exist.

The Pectostraca [Rhizocephala and Cirrijyedla) leave
the egg as a JVduplius, provided with three pairs of limb-like
appendages, of which the anterior pair are sim.ple, while the
two posterior pairs are bifurcated (Fig. 68, A). An addi-
tional pair of filiform appendages subsequently makes its ap-
pearance in front of the undivided pair of members, in most
cases ; and there is a discoidal carapace, the antero-lateral
angles of which usually become greatly produced. Subse-
quently, the carapace becomes bivalve (as in many PhyllO'
poda, and in the Cladocera and Ostracoda), and the anterior
undivided pair of limbs are converted into relatively large,
jointed appendages, provided with a sucker-like organ. The
thorax grows and usually develops six pairs of appendages.

Finally, the bivalve-shelled larva fixing itself by the
suckers of its anterior limbs, the prte-oral region of the head
becomes enlarpfed, and is converted into the base, or pe-
duncle, in ordinary Cirripedes ; while it gives off the root-
like processes which grow into the tissues of the animals on
w^hich the RhizocephaJa are parasitic. The Pectostraca are
almost all hermaphrodite, a condition which is very excep-
tional among Arthropods. They possess no heart.

The Cirripedia. — It can hardly be a matter of reproach

» " Entwickelungsgeschicbte von Cypris " (1868) ; and " Grundzuge,"
p. 487.

254 THE ANATOMY OF INVERTEBRATED ANIMALS.

to the older naturalists if they failed to discover the affinity
connecting the sedentary *' Acorn-shells " of a rocky coast
with the active Shore-crab which runs among them ; or if
they classed the Barnacles with Mollusca, instead of admit-
ting them to that place amid the Crustacea which is now
assigned to them by every naturalist of competent judgment.
Nothing, in fact, at first sight, is less suggestive of a Crusta-
cean than a Balanus^ or a Lepas / the former firmly fixed
by the base of its multivalve conical shell, the latter by its
fl[eshy and contractile peduncle ; the only sign of life in
either being the alternate protrusion and retraction, from the
valvular opening of the animal's case, of a bundle of curved
filamentous cirri, which sweep with a brushing motion through
the water, and scoop the floating nutritive matters toward
the mouth.

The valves through which the cirri make their egress are
strengthened, in both Balanus and Z,€2:)as^ by four calcified
pieces, two on each side ; those of each half being united to-
gether by an oblique suture, or by a regular articulation ;
while the two pieces of opposite sides are connected only
along one margin, either immediatelj' [J3alanus), or by means
of an intermediate piece (JLej^as).

The upper, or distal, pieces are termed the terga^ the
lower, or proximal, pieces the scuta^ the intermediate piece is
the carina. In Lepas^ there are no other hard external
pieces; but, in JBakmus, the conical shell, into which the
valves can be more or less completely retracted, is composed
of six portions or compartments. Of these, one is situated
on the same side as the opening between the valves and
another at the precisely opposite point, or on the same side
as the line of union of the valves. The latter is the homo-
logue of the intermediate piece, or carina^ in Lepas ; the
former, in Balanus^ consists of three pieces united together,
the median rostrum and the two rostro-lateral compartments.
On each side of the carina is a compartment termed carino-
lateral, and between them and the complex rostrum lies a
lateral compartment.

If the shell consisted of its eio:ht typical pieces (as it does
in the genus Octomeris), it would be found that each pre-
sented a triangular free middle portion and two lateral w^ngs.
The former is always termed the paries, but the latter re-
ceive different names, according as they overlap or are over-
lapped by others. In the former case, they are termed radii,
in the latter, alee. Thus, typically, the carinal and the ros-

THE CIRRIPEDIA. 255

tral compartments are overlapped on both sides, and their
wings are consequently both alae ; the lateral and carino-
lateral compartments are overlapped on one side, and overlap
on the other, hence they have an ala on one side, a radius on
the other ; while the rostro-lateral compartment overlaps on
both sides, and hence its wings are both radii. In Balanm^
however, the rostrum and rostro-lateral compartments being
replaced by a single compartment formed by their confluence,
this piece has radii on both sides.

Different as is the appearance of Lepas from that of
Balanus, they closely resemble one anotner in essential
structure. Thus, to commence with Lepas. On cutting away
the scutum and tergum of one side (Fig. 67, -S), the hinder
part of the body of the animal is seen within the sac of the
capitulum, formed by the valves of the shell, to which it is
attached only on the rostral side and inferiorly by a com-
paratively narrow isthmus. Immediately behind this point
the body widens, to constitute what Mr. Darwin' has termed
the prosomi^ but the thoracic segments, which succeed the
prosoma, gradually taper posteriorly. Six pairs of appendages
{a) are attached to the thorax, each limb consisting of a basal
joint (protopodite), terminated by two long multi-articulate
cirri, the representatives of the endopodite and exopodite ;
and a rudimentary abdominal segment, terminated by two
short caudal appendages, succeeds the thorax, and is pro-
duced in a long setose annulated penis (/*). Filamentous
appendages depend from some of the thoracic somites, and,
projecting from the inner wall of the sac on each side, is a
triangular process, the ovigerous frcenum (m).

The mouth is situated at the posterior part of a protuber-
ant mass, seated on the rostral face of the prosoma. This is
principally composed of a large, bullate labrum, behind which
are a pair of mandibles with large and setose palps, and two
pairs of maxillae. Anteriorly, the prosoma passes by a nar-
row isthmus into the rostral part of the peduncle, into which
it, as it were, expands ; while the posterior margins of the
peduncle become continuous with the walls of the sac.

The^ extremity of the peduncle is fixed by a peculiar
cementing substance to the body to which the Lepas ad-
heres ; but, if it be carefully detached, thei-e will be found
connected with the rostral portion of the surface a pair of
very minute, singular-looking, organs, consisting of two proxi-

» " Monograpli of the Cirripedia," 1851, 1854.

256

THE ANATOMY OF INVERTEBRATED ANIMALS.

mal joints, succeeded by an articulation which is dilated into
a sucker, and terminated by an elongated setose joint (Fig,
67, A, J3, I). These are the remains of the anterior append-
ages of the larva.

From what has been said, it follows that the fixed end of
the peducle is, in fact, the anterior extremity of the body
of the Lepas^ and that a Barnacle may be said to be a Crus-

\ tacean fixed by its head, and kicking the food into its mouth

j with its legs.

Fig. 67.—^, Diagrammatic section of Balanus ; J?, of Lepas. — a is placed in the cavity
of the sac, and lies over the labrum ; 6, prosoma ; c, carina; c, I, carino-lateral
compartment; I, lateral compartment: r, rostrum; «, scntnra; t. tergam; /, penie;
p. trnt-formed gland ; h. duct connecting this with i. k, cement-duct and glands;
/', antennae; i, peduncular or ovarian tubules; m, ovigerous fra&nam; </, anus.

The mouth of JLppas looks toward the posterior extremity
of the body, and leads into a tubular oesophagus, which
passes forward, and opens by a wide superior extremity into
the globular stomach. From this point, the alimentary canal
bends back upon itself, and gradually narrows into the in-
testine, which terminates in the anus, situated at the ex-
tremity of the abdomen, on the tergal side of the penis.
Two considerable branched cseca, probably hepatic, proceed
as diverticula from the stomach, corresponding very closely

THE CIRRIPEDIA. 257

in position with those of DapJima. No heart or other cir-
culatory organs are known to exist ; and it may be doubted
if the ovigerous frjena of Lepas exert, as they have been sup-
posed to do, a branchial function.

The nervous system consists of a pair of cerebral ganglia
situated in front of the oesophagus, and connected by long
commissures with the anterior of five pairs of thoracic gan-
glia, whence nerves are given off to the limbs. In the mid-
dle line the cerebral ganglion gives off two slender nerves,
which run parallel with one another in front of the stomach
and enlarge into two ganglia, whence they are continued to
a double mass of pigment, representing the eyes. From the
outer angles of the cerebral ganglion arise the large nerves
which proceed into the peduncle and supply the sac. These
appear to correspond with the antennary and frontal nerves
of other Crustacea ; and Mr. Darwin describes an extensive
system of splanchnic nerves.

Lepas^ like the majority of the Cirripedia^ is hermaphro-
dite. The vesicula3 seminales are readily seen in fresh speci-
mens, as white cords distended with spermatozoa, which run
from the canal of the penis, into which they open, forward,
on each side of the body, to the prosoma, where they end in
dilated extremities, which are connected with a multitude of
ramified caeca forming the proper testis.

The ovaries are ramified tubes provided with coecal dila-
tations, and lodged in the peduncle. The oviducts pass into
the body, and, according to Krohn, terminate in apertures
situated on the basal joint of the first pair of cirri.' Two
"gut-formed" glands, as they are termed by Darwin, lie,
one on each side of the stomach, and are probably accessory
glands of the reproductive organs, analogous to those which
secrete the walls of the ovisac in the Copepoda,

The mode of exit of the ova from the ovary is not cer-
tainly known, nor is the place of their impregnation ascer-
tained ; but they are eventually found cemented together by
chitin into large lamella?, which adhere to the ovigerous
frsena, and, ordinarily, at once strike the eye when the ca-
pituhim of a Cirripede is opened.

Yelk division is complete, and the embryo attains to its
earliest larval condition within the eo:g. If a series of the
fresh ovigerous lamellae be taken and pulled to pieces with

» The position of these apertures corresponds with that of the openings,
supposed to appertain to the shell-glands in Limnadia and Apus.

258

THE ANATOMY OF INVERTEBRATED ANIMALS.

needles in a watch-glass full of sea-water, one is pretty sure
to be found whence a number of active little jVauplius-\ike
animalcules are set free (Fig. 68, A), Each presents a some-

FiG. 68.—^. Larva of Balanus halanoides on leavin? the egx (after Spence-Bate").
B. Attached pupa of Lepas Aufitratis (after Darwin): /?, autennary apodemes ; i,
gut-formed gland, with cement-duct running to the antenna.

what triangular body, produced in the middle line posteriorly
and at its anterior lateral angles. The mouth is situated on
a proboscidiform projection placed nearly in the centre of
the body, and in the midst of three pairs of natatory limbs,
of which the two posterior pairs have bifid extremities. In
front of the mouth, either in this stage, or after one or two
moultings, two filaments are often developed. A single eye-
spot is situated in front of the bases of the anterior append-
ages. After moulting several times the larva assumes a
new form, passing into its second stage. The carapace is
now oval and compressed, so as more nearly to resemble that
of a Daphnia or Gypris. There are two eyes. The first
pair of swimming appendages of the Nauplius are converted
into antenniform organs, each provided with a sucker, and
the rudiments of the six pairs of cirri make their appearance
behind the mouth.^

In the third stage, the larva is, as Mr. Darwin states,
" much compressed, nearly of the shape of a Cypris or mus-
cle-shell, with the anterior end the thickest, the sternal sur-
face nearly or quite straight, and the dorsal arched. Almost
the whole of what is externally visible consists of the cara-

* Accordinfy to Claus (" Grundzage der Zoologie," 3te Auflage, p. 460), the
second pair of appendages disappears, and the third gives rise to the mandi-
bles. In this case the antennary organs represent antennules, and the limbs
of the Cirripede Nauplim correspond with those of the Copepod and Branchi-
opod NaupUus.

DEYELOPMENT OF THE CIRRIPEDIA. 259

pace ; for the thorax and limbs are hidden and inclosed by its
backward prolongation ; and, even at the anterior end of the
animal, the narrow sternal surface can be drawn up, so as to
be likewise inclosed." The larva, in this stage, is provided
with two large compound lateral eyes, while the median eye
is arrested in its development. The oral tubercle exhibits all
the gnathites of a Cirripede, but they are covered by an imper-
forate integument, so that this " locomotive pupa," as Mr.
Darwin terms it, is unable to feed. There are six pairs of
legs, and the thorax ends in an abdomen, consisting of three
somites terminated by two caudal appendages. There is no
penis. The most remarkable structures in the pupa, however,
are the "gut-formed glands," which are already well devel-
oped, and from which the cement ducts can be traced to the
disks of the antenniform organs, on the faces of which they
open. The pupa, after swimming about for a while, at length
selects its permanent resting-place, to which it adheres, at
first, only by the action of the suctorial disks. The tempo-
rary attachment, however, is speedily converted into a per-
sistent one, the cement pouring out from its excretory aper-
tures on the disks, and firmly gluing them and the anterior
end of the body down to the surface on which they rest.

Coincidently with these changes, several other important
alterations take place, during the passage of the locomotive
pupa into the fixed young Cirripede. The compound eyes are
moulted, and with them the antennary apodemes, furnished
by the integument of the deep fold which separates that part
of the body of the pupa which corresponds with the beak of a
Daphnia^ or of a Limnetis^ from the prosoma. The fold is
thus enabled to straighten itself; and, as a consequence, the
carapace of the Cirripede, instead of remaining more or less
parallel with the surface of attachment, becomes perpendicu-
lar to it. Again, in the pupa, the axis of the carapace and
that of the body are identical in direction; but, during the
last moult, the chamber of the carapace extends forward far
more on the tergal than on the sternal side, separating the
tergal part of the prosoma from the " beak," with which it
was at first continuous, and thus allowing the body of the
Cirripede to take its final position, which is nearly transverse
to the axis of the carapace.

The terga and scuta now appear as horny thickenings,
and, afterward, as calcifications in the wall of the capitulum.
The fraena and the penis make their appearance, and the
genitalia become developed in the prosoma and in the pe-

260 THE ANATOMY OF INYERTEBRATED ANIMALS.

duncle, which is produced by the gradual elongation of the
" beak " of the pupa.

With the assumption of its perfect form, the Cirripede
ceases to moult its carapace, ecdysis being hereafter confined
to the inner lining of the sac, and to the integument of the
contained body.

Such is the structure and development of a typical pedun-
culate Cirripede. In other genera, such as PoUicipes^ calca-
reous plates are developed on the peduncle, foreshadowing
the compartments of the sessile forms. The latter, of which
3alanus may be regarded as the type, differ in structure from
Lepas in no very essential particular. The peduncle, very
short and broad, instead of slender and elongated, is incased
b}' its compartments, and is sometimes fixed by a shelly basis.
The arrangement of the layers of cement is often extremely
complicated ; the scuta and terga are articulated together ;
the frgena are much larger organs, and posssibly subserve the
respiratory function ; the thoracic ganglia are concentrated
into a single mass ; and the cementing apparatus is much
more complicated.

The pedunculate and sessile Clrripedia^ taken together,
constitute by far the largest of the three great groups which
Mr. Darwin recognizes ; namely, the Thoracica, characterized
by having limbs attached to the thoracic somites, while the
abdomen is rudimentary.

The second group, the Ahdominalia, contains only one
genus, Cryptophiali(S (Fig. 69, 5, 6), which has no thoracic
limbs, but is provided with three pairs of abdominal append-
ages. The larva is very imperfect in its first and second
changes, which are undergone within the sac of the parent.

The third group, Apoda, likewise contains only one
genus, the remarkable Proteclepas (Fig. 69, 7), which is
devoid of either thoracic or abdominal limbs ; it has a vermi-
form body, and a rudimentary peduncle, represented by two
threads terminated by the characteristic antenniform organs.

In the great majority of the Cirripedia the sexual appa-
ratus is disposed as in Lepas, but Cryptophialus and Alcippe
are unisexual, the male differing very widely in form and size
from the female (Fig. 69, 3, 6).

The J^alanida?, or sessile Cirripedes, all present the nor-
mal sexual relations ; but the other division of the TJioracicOy
the Lepadldm, contains two genera, Ibla and Scalpellum,
which not only possess species having the sexes in distinct
individuals, but others presenting the unique combination of

THE CIRRIPEDIA,

261

males with hermaphrodites. Thus, Scalpellum vulgare is
hermaphrodite, possessing well-developed male and female
oro-ans. Nevertheless, on the inner side of the occludent
margin of its scutum there is a fold, over which and imbed-
ded in the spinose chitinous border of the scutum, a minute,
oval, sac-like creature is commonly found, firmly attached by

Fig. 69.— 1. Aldppe lampas ; female. 2. The same in pectional view: H, Hornv
disk of attachment ; in 1, the males are visible as dark ppecks on either side of
the upper part of the sac ; c, ovary; A, first pair of cirri ; k. I, n, three sesr-
ments of the thorax without cirri ; the other three segments, bearing the three
pairs of terminal cirri, are very short. 3. Male Alcijype : a, antennary append-
ages ; 6, vesicula seminalis ; o, eye; d, testis; k, orifice of the sac; m. pe-
nis. 4. Burrow of Alcippe in a portion of a Fusus shell. 5. Cryptophialus
minulus (female) with the outer internment removed: «, labrum : /', palpi; v,
outer maxillae; h, rudimentary maxillipede: c, wall of &ac continued above
into the rim of the aperture n. h; I, m, abdominal cirri ; k. appendages of un-
known nature. 6. Male Cryptophialus. 7. Proteolepas bivinctn ; m. month ; g, h,
peduncle and antenna; ?, A, vesicula seminalis and penis. (After Darwin.)

cement which covers the characteristic antennules of a Cir-
ripede. Within the sac is a thorax, with four pairs of rudi-
mentary appendages terminated by a short abdomen. There
is neither mouth, alimentary canal, nor gnathites, the cavity
of the body being principally occupied by a great seminal

262

THE ANATOMY OF INVERTEBRATED ANIMALS.

vesicle ; and no trace of female organs exists. This is, there-
fore, an accessory, or '* complemental " male. In Scalpellum
ornatuni the individuals are males and females, two of the
former being lodged in cavities of the scuta of one of the
latter, as in the preceding species, and in S. rutilwn. The
males have no mouth. S. rostratum has complemental males,
provided with alimentary organs attached to the interior of
the sac of the hermaphrodite, while S. JPeronii and villoswrn,
have still more perfect complemental males fixed in a like
position. In Ibla Cuiningii^ the female has a vermiform
male, provided with well-developed alimentary organs at-
tached within her sac ; but, in the only other species of this
genus, I. quadrivalvis, a similarly constructed, but here only
complemental male, is lodged in a relatively large hermaph-
rodite form.

With regard to the habits of the Cirripedia, the majority
are merely cemented to foreign bodies. Anelasina and Tubi-
cinella, however, partially bury themselves in the integuments
of the shark and whale, and thus prepare us for the com-
pletely boring habit of Cryiytophialus Lithotrya, and Alcippe,
the latter of which (Fig. 69, 1, 2, 3) burrows in dead shells
on our own coasts.

Proteolepas lives within the sac of Alejms cornuta. and

Ftg. 70. — A. N'aupUus-9'\s.^p, of Sriccnb'na purpurea : cp, cnrapace.

JB. C'yprit-ftta.<ie o^LernceodisctisporcellancB. C. Adult coirlition o^ Peltorjaster pnguri :

a. anterior end of the body ; b, aperture ; c, root-like processes. (After F.

Muller.)

appears to be truly parasitic upon it, suckinG^ the nutritive
juices from the soft prosoma of the animal which it infests.

THE MALACOSTRACA. 263

The Cirrijyedia are almost exclusively marine, only a few
species tolerating even brackish water. The Thoracica alone
have yet been found in the fossil state. The oldest known
genus, Pollicipes^ occurs in the lower oolite ; there is a single
cretaceous species of Verruca^ but the sessile Cirripedes be-
come numerous only in the tertiary epoch.

The Rhizocephala [Peltogaster^ Saccidina) are small and
parasitic ; usually upon the abdomen of other Crustacea
{Podophthalmia), The body is like a sac or disk, and devoid
of segmentation and of limbs. 1 he aperture of the sac is
funnel-shaped, and supported by a ring of chitin. The circum-
ference of the funnel gives off a number of root-like processes,
which branch out through the body of the infested animal.
The alimentary canal is obsolete, and there are no cement-
glands. They are hermaphrodite, and the j^oung, like those
of the other Pectostraca, pass through a Kauplius and a
Cypris stage. ^

The Malacosteaca. — The groups of Crustacea known as
the Podophthahnia, the Cumacea, the Edriop>hthalmia^ and
the Stomatopoda, are here included under this head.

The body consists of twenty somites (counting that which
bears the eyes as one), and, of these, six (bearing the eyes,
antennules, antennae, mandibles, and two pairs of maxilla?)
constitute the head ; eight enter into the thorax, and bear
the foot-jaws and ambulatory limbs ; and six form the abdo-
men and swimming limbs. In some few instances the num-
ber of somites is reduced, but they never exceed twenty.

The N^auplius-ioTm of the free embryo is rare, but occurs
in some cases (Penev.s). In others {Mysis) it is represented
only by a temporary condition of the embryo, during which,
however, a chitinous cuticula is formed, and subsequently
shed ; and what appear to be remains of such a transitory
record of an original Kauplius state, are seen in many Am-
phipoda and Isopoda, which nearly attain their adult form
within the eg^. In most Podophthalmia the embryo leaves
the eor^ not as a JVauplius, but as a Zocea, which has thora-
cic, but no abdominal, appendages, and in many respects re-
sembles a Copepod.

1 The term Ci/pris-stage, usually applied to that condition of the Inrvae of
the P.ctostraca in which they are provided with a bis'alve carapace, must not
be taken to imply any special affinity with the Ostracoda. On the contrary,
the larva in the C'y^m-stage is much more similar to a Copepod or Bianchi-
opod.

264 THE ANATOMY OF INVERTEBRATED ANIMALS.

The Cumacea take an intermediate position between the
Podophthahnia and the Edriophthalmia on the one hand,
and the Phyllopoda {Nehalla) on the other. They thus serve
to connect the Malacostraca with the Entomostraca.

The Pod ophthalmia. — It will be convenient to commence
the study of the 3IalacostraGa with the Podophthalmia y and
as excellent examples of this division of convenient size are
readily obtainable in the fresh-water Crayfish {Astacus fluvi-
atilis) and the Lobster {Honiarus vulgaris), and as thej fur-
nish a very intelligible guide to the general plan of structure
of the higher Arthropoda^ the organization of Astacus will
be described at length. With some unimportant modifi-
cations, what is said about it will be found to apply to the
Lobster.

The upper and anterior portion of the dense and more or
less calcified exoskeleton which covers the body of Astacus,
has the form of a large, expanded, shield-like plate, the cara-
pace, produced into a strong frontal spine between the eyes,
and bent down at the sides, so as to reach the bases of the
legs. The posterior division of the body, on the other hand,
presents a very different aspect, being divided into a series of
distinct movable somites. Tiiis is called the abdomen ; while
the anterior division, covered by the carapace, corresponds
with the head and thorax of other Arthropoda, and receives
the name of cephalo-thorax.

On turning to the ventral surface of the Crayfish, a great
number of limbs or appendages, twenty pairs in all, are seen
to be attached to the cephalo-thorax and abdomen, six pairs
belonging to the latter and fourteen pairs to the former re-
gion of the body.

The six pairs of abdominal appendages are commonly
known as the "false" or " swimming" feet ; and it will be
observed that they are attached to the six anterior segments
of the abdomen only, the seventh being unprovided with any
such organs. Of the fourteen pairs of cephalo-thoracic ap-
pendages, the five posterior are called the " ambulatory " legs,
being the organs by which the Crayfish is enabled to walk.
Strictly speaking, however, the anterior of the five pairs is
not more ambulatory than prehensile, being so modified as to
constitute the great claws, or ** chelag."

Of the six next pairs of appendages, passing from behind
forward, five are not at first sight apparent, the posterior
pair, which are applied over the mouth and cover the others,

ASTACUS FLUVIATILIS. 265

being alone visible. These, and the two pairs which lie im-
mediately under or in front of them, are called inaxillipedes^
or "foot-jaws." The next two pairs, delicate and foliaceous,
are the maxillse ; while beneath or rather in front of them
are two strong, toothed organs, the mandibles. These, the
maxillae and the maxillipedes, thus constitute six pairs of
gnathites.

The remaining three pairs of appendages occupy the sides
of the forepart of the cephalo-thoiax, in front of the moulh.
The most posterior pair, or the long feelers, are the antennae;
the next, or the short feelers, are the antennulae; while the
most anterior pair are the movable stalks, which support the
eyes upon their extremities — the " ophthalmic peduncles," or
" ophthalmites."

To arrive at an understanding of the composition of this
complex body with its multiform appendages, we must first
detach and study carefully one of the abdominal segments —
say the third. Such a segment is nearly semi-circular in ver-
tical section, the dorsal wall, or tergum, being very convex,
and where it reaches the level of the almost straight ventral
wall, or sternum, sending down a flattened lobe, which is re-
flected at its free edges into a corresponding prolongation of
the ventral wall, so that each infero-lateral angle of the seg-
ment is prolonged into a hollow process, the pleuron. Near
the outer extremities of the straight ventral portion of the
segment two rounded articular cavities, which receive the
basal joints of the appendages, are situated. A transverse
groove will be seen on the tergum, separating rather more
than the anterior third of its surface, as a smooth, convex,
lenticular facet, which is completely overlapped by the pos-
terior margin of the preceding segment, when the abdomen
is extended, and is left uncovered only in complete flexion.
This is the tergal facet. A corresponding flattened and rath-
er excavated surface upon the anterior half of the pleuron,
which is similarly overlapped by the preceding pleuron, and
is left uncovered only in complete extension, may be termed
the pleural facet. It will be observed that there is a close
correspondence between the skeleton of an abdominal somite
of a Cray-fish, and that of a thoracic somite of a Trilobite ;
except that, in the latter, the sternal region is not calcified.

The appendages of the segment (Fig. 71, K) are very sim-
ple, consisting of a cylindrical basal portion, divided into two
joints, a shorter proximal, and a longer distal, to the latter of
which two terminal many-jointed filaments are articulated.
12

266 THE ANATOilY OF IXYERTEBRATED AXIMALS.

Oh n y «

/f^ T- X

ASTACUS FLUYIATILIS. 267

Fig. 71. — Astacus fluviatUis. A. Mandible: a, ft, endopodite ; o, its terminal joints
• constitutinar the palpaa of the mandible. B. First maxilla. C. Second max-
illa. 2). First maxillipede. E. Second maxillipede. i^. Third niaxillipede. All
the preceding, except B, are left limbs. G. Ambulatory le^:. H. Appendage of
first, and /of second, abdominal somite in the male. K. Appendage of third ab-
dominal somite. L. Sixth abdominal somite, with its appendages and telson :
a, b, endopodite; c, exopodite ; d, epipodite; e, setaceous filaments attached to
coxopodite ; x, tergum of sixth abdominal somite; y, 2, the two divisions of the
telson. In G : 2, basipodite ; 3, ischiopodite ; 4, meropodite ; 5, carpopodite ;
6, propodite ; 7. dactylopodite. In A^ d marks the tendon of the adductor muscle,
and in K the joints of a 6 and c are not sufficiently numerous. M. Transverse
section of half a thoracic somite (a): 6, coxopodite : c, basipodite; rf, ischiopo-
dite ; A, branchiferous epipodite ; /, g, branchiae; e, filiform appendage. ^V. One
of the branchiferous epipodites : a, its point of attachment; 6, basal enlarge-
ment ; c, branchial filaments ; c?, terminal lobes.

The inner of these is distinguished from the outer by possess-
ing a more elongated and wider basal joint. The whole basal
division of the appendages is the protopodite ; while the in-
ternal and external terminal filaments are the endopodite
(a, b) and exopodite (c).

An abdominal segment, or somite, then, is composed of a
tergum, two pleura, and a sternum ; but it must be remem-
bered that these terms rather indicate regions than anatomi-
cal elements, the whole segment being continuously calcified,
and no sutures or other absolute demarcations separating
one portion from another. Furthermore, the somite carries
two appendages, each divided into a proximal portion or pro-
topodite, terminated by two branches, the endopodite and
exopodite.

The whole exoskeleton of the Astacus, however various
may be the appearance of its different parts, consists of so-
mites and appendages essentially similar to those which have
just been described, but which are more or less masked by
the connation, the coalescence, the abortion, or the extreme
modification of their primitive elements.

If, in the first place, we follow out these modifications in
the posterior somites, we find the fourth, fifth, and sixth
abdominal somites to be, in all essential respects, similar to
the third ; but the appendages of the sixth (Fig. 71, Z) are
singularly changed, the protopodite being represented by a
single strong, short joint, and the exopodite and endopodite
having the form of wide, oval setose plates. The exopodite
is again divided into two portions by a transverse joint. The
seventh division of the abdomen (Fig. 71, X, y, z) is the telson.
This telson bears no appendages ; dorsally it is completely
calcified, but is divided by a transverse suture into two
portions, the posterior of which is movable upon the other;
yentrally, on the contrary, it is only tlie posterior part which
is fully calcified, the middle of the anterior portion, in vrhich

268 THE ANATOMY OF INVERTEBRATED ANIMALS.

the anus is situated, being completely membranous, and the
sides only being strengthened by calcareous plates extend-
ing inward from the dorsal hard skeletal element, or sclero-
dermite.

The powerful tail-fin of the Astacus is formed by the tel-
son, combined with the two distal divisions of the sixth ab-
dominal appendages on each side. The other abdominal
appendages can have very little influence on locomotion. In
the female, however, they play an important part as the car-
riers of the eggs ; and in this sex there is nothing worthy
of special notice about the first and second abdominal somites
or their appendages, except that those of the first are rudi-
mentary. In the male the appendages of these two somites
have undergone ^ very interesting metamorphosis, whereby
they are fitted to subserve copulation. Those of the second
somite (Fig. 71, I) are enlarged, and the protopodite and
basal joint of the endopodite are much elongated ; the latter
being produced internally into a plate rolled upon itself, and
thence concave outward and forward. It is as long as the rest
of the endopodite (which, like the exopodite, is many -jointed),
and serves as a sort of sheath for the reception of the append-
age of the first abdominal somite (Fig. 71, H), which con-
sists of a single plate rolled upon itself in a similar manner,
so as to resemble a grooved style. These organs, doubtless,
help to convey the spermatophores from the male genital
apertures to the body of the female.

The compact and firm cephalo-thorax seems at first to dif-
fer widely from the flexible, many-jointed abdomen ; but the
most posterior of its somites offers an interesting transition
from the one to the other. This somite is, in fact, only united
bv membrane to that which precedes it, and is hence, to a
certain extent, movable. Its sternal portion is completely
calcified, but the epimera * are only partially calciBed.

The appendages of this somite diS'er widely from those of
the abdomen, representing (as their development shows) only
the protopodite and endopodite of the latter. Each is a long,
firm leg, composed of seven joints, the proximal one being
thicker than any of the rest, while the terminal joint is nar-
row, curved, and pointed. To these seven joints Milne-Ed-
wards has applied the following terms (Fiff. 71, G-) : The
proximal one, which articulates with the somite, is the eoxo-

^ The term epimeroTi is here employed in a more special sense than that
commonly used, to denote that part of the lateral wall of a somite which is
situated between the articulation of the appendage and the pleuron.

ASTACUS FLUYIATILIS. 269

podite (1) ; the next, small and conical, is the hcmpodite (2) ;
the third, cylindrical, short, and marked by an annular con-
striction, is the ischiopodite (3) ; next comes a long joint,
the meroi^odite (4) ; then the carpopodite (5) and pjropjodiU
(6) ; and, finally, the terminal dactylopjodite (7).^

The next four somites, proceeding anteriorly, have a sim-
ilar general character to that which has just been described,
but they cease to be movable upon one another, partly by
reason of the calcification of the interepimeral and inter-
sternal membranes, partly on account of the development of
these membranes by a folding inward, or involution, into
processes, the op)odemes, which project inward and unite
with one another in the cavitA' of the thorax. In an Astacus
w^hich has been macerated — or, better, boiled in caustic alkali
— the floor of the thoracic cavity is seen to be divided into a
number of incomplete cells, or chambers, by these apodemal
partitions, which will be observed, on careful examination,
to arise partly from the intersternal, partly from the inter-
epimeral, membrane connecting every pair of somites. The
former portion of each apodeme is the endosteriiite, the latter
the endopleurite ^ of Milne-Edwards. As a general rule, each
endosternite is distinguishable into three apophyses: the
arthrodial^ which passes outward and unites with the de-
scending division of the endopleurite to form one boundary' oi
an articular cavity for a liml3 ; the tnesopjhragmal^ which is
directed inward, uniting with its fellow, and forming an arch
over the passage left in the middle line between each pair of
endosternites — the so-called sternal canal ; lastly, the para-
phragmal division is a small process, which passes forward,
upward, and outward, and unites with the anterior division
of its own endopleurite, and with the posterior division of the
endopleurite in front of it.

The endopleurite likewise divides into three apophyses,
one descending or arthrodial, and two which pass nearly
horizontally inward : the anterior horizontal apophysis unit-
ing with its own paraphragmal apophysis, the posterior with
the paraphragmal of the antecedent endosternite. The pos-
terior horizontal apophysis, therefore, crosses the space be-
tween every pair of apoderaes diagonally, whence the ap-
pearance of a double row of longitudinal cells opening above,
on each side of the sternal canal. It will be understood,

» Probably the coxo- and basipodite together answer to tbe protopodite of
th.e abdominal appendages, the remainiag' joints representing the endopodite.

270 THE ANATOMY OF INVESTEBRATED ANIMALS.

however, that these cells are very incomplete, communicating
with one another anteriorly and posteriorly by the large
apertures left between the endosternites and endopleurites ;
and laterally, by the spaces between the endosternites,
through Avhich each series opens into the sternal canal ; while
above, they are in free communication with the thoracic
cavity. The apodemes give attachment to the muscles of
the appendages, while the chain of ganglia and the sternal
artery lie in the sternal canal.

The appendages of the penultimate resemble those of
the last thoracic somite, but the three preceding pairs differ
from them by being chelate — that is, by having the posterior
distal angle of the propodite produced so as to equal the
dactylopodite in length, and thus constitute a sort of oppos-
able finger for it (Fig. 71, G^ 6, 7). The first ambulatory or
prehensile limb, again, is remarkable for its great size and
strength, and for the ankylosis of its basipodite with the
isohiopodite.

The four anterior pairs of ambulatory limbs differ from
the last pair in possessing a long curved appendage (Fig. 71,
JST), which ascends from the coxopodite, with which it is artic-
ulated, and passes into the branchial chamber, in which it lies.
This is the epipodite ; its relation to the function of respira-
tion will be adverted to presently.

The sterna, which are wide in the three hindmost thoracic
somites, become very narrow and almost linear in the ante-
rior ones. They and their apodemes, however, remain per-
fectly recognizable.

The sternal regions of the three maxillipedary somites
have the same characters, their appendages and articular cavi-
ties becoming smaller ; while, by the contemporaneous exces-
sive narrowing^ of the interarticular resrions of the sterna,
these cavities are closely approximated.

The sternum of the next anterior somite (bearing the
second pair of maxillas), on the other hand, though very nar-
row from before backward, has a considerable width, and its
articular cavities, already much larger than those of the ante-
rior maxillipedary somites, are consequently thrown outward.
Hence results a sudden widening of the second maxillary, as
compared with the first maxillipedary somite ; and, as a con-
sequence, we find a deep fold or depression on the sides of
the body where these two somites join. This fold is directed
upward and backward on the flanks of the body, parallel with
an important impression on the carapace, the cervical groove.

ASTACUS FLUVIATILIS. 271

Not only on this ground, but because the fold really repre-
sents a true neck, or separation between the head and thorax,
it may approximately be termed the cervical fold. The
scaphognathite (Fig. 71, 6', c, c?), an important appendage of
the second maxilla, lies in this cervical fold.

The appendages of the three maxillipedary somites (Fig.
71, D^ E^ jB) are highly interesting, inasmuch as they afford
transitional forms between the ambulatory limbs and the
gnathites. Each maxillipede is composed of three divisions,
articulated with a stout protopodite. The outermost of these
divisions is a curved, elongated lamina {d ), precisely resem-
bling the epipodite of the posterior thoracic limbs in the two
hinder maxillipedes (E^ F) ; but, in the anterior (J[>), not
modified so as to serve as a branchia, and rather approaching
the scaphognathite in form.

The middle division of each maxillipede (c), answering to
the exopodite, is long, slender, many-jointed, and palpiform ;
while the inner division, or endopodite (a, ^), not only corre-
sponds with one of the ambulatory limbs, but in the posterior
maxillipede (Fig. 71, F^ very closely resembles one, and con-
tains the same number of joints. In the next maxillipede,
however (Fig. 71, E)^ the endopodite is proportionally shorter,
and in texture and form rather approaches the foliaceous en-
dopodite of the anterior maxillipede (Fig. 71, Z>), in which a
flat plate is applied to the posterior surface of the slender
exopodite. A perfect transition is thus produced between
the corresponding divisions of the second maxillipede and of
the second maxilla.

The intermaxillary apodeme, or that developed from the
connecting membrane of the two maxillary somites, is very
remarkable for its stoutness and for the great size and ex-
panded form of the mesophragmal processes, which unite into
a broad plate, whence prolongations are sent forward and
outward, in front of the tendon of the great adductor mandi-
hulcB muscle on each side. These prolongations appear to be
the calcified posterior horizontal apophyses of the mandibulo-
maxillary apodeme, which elsewhere remains membranous.

The second maxilla (Fig. 71, G) much resembles the an-
terior maxillipede, but the epipodite [d) and exopodite (c)
appear to be combined into a wide oval plate, the scapho-
gnathite^ of which mention has already been made.^ In the
first maxilla (Fig. 71, B) the epipodite and exopodite appear

1 Until the development of these appendages has been worked out, the de-
termination of the homologies of their parts must be regarded as provisional.

^72 THE ANATOMY OF INVERTEBRATED ANIMALS.

to be undeveloped, and the joints of the endopodite are com-
pletely foliaceous. The somite which supports the mandibles
is, to a great extent, membranous in its sternal region ; it is
united with the corresponding region of the first maxillary
somite, itself represented merely by a narrow, distinctly cal-
cified, band, in front of the second maxillary sternum, by mem.-
brane only. In this membranous space the elongated aper-
ture of the mouth is situated.

On each side of and behind the mouth are two little
elongated oval calcified plates, between which an oval pro-
cess, setose at its extremity, proceeds downward and for-
ward, and lies in close apposition with the posterior face of
the mandible of its side. This is one-half of what is termed
by most authors the lahiuni^ but, to avoid confusion with the
labium of Insecta^ from which it is wliolly different, it may
be called the metastoma (Fig. 72,/"). It obviously answers
to the structure so named in the Cojjej'^oda.

The mandibles fill up a large space in the sternal mem-
brane, with which their edges are continuous on each side of
the oral aperture ; externally, the sternal membrane bends
suddenly downward into the pleural ridge, continuous with
the branchiostegite of the carapace, and becomes calcified ;
while, anteriorly, it is very difficult to say where the mandi-
bular sternum terminates. In front of the mouth the sternal
membrane becomes developed into a large median lobe, con-
taining three small calcified plates on each side of the middle
line. This is the labrum (Fig. 72, e).

The mandible itself (Fig. 71, A) is thick and strong at its
inner end, Avhere it is divided by a deep excavation into an
upper and a lower portion {a, b), the edge of each being
toothed. The outer division of the mandible extends along
the whole width of tlie somite, and tapers to its extremity,
which presents an articular head, the outer condyle. At-
tached to its anterior margin is the palp (o), which represents
the terminal joints of the mandibular endopodite. The ex-
opodite and the epipodite have no representatives in this ap-
pendage. Superiorly, the outer portion of the mandible is
concave, and its posterior edge gives attachment to the cal-
cified tendon of the adductor mandibuhia (d).

In front of the labrum and mandibles is a wide, somewhat
pentagonal area, prolonged into a point in the middle line
forward, and presenting a small spine on each side ; this is
the epistoma (Fig. 72, JB, I), and it is chiefly, if not entirely,
formed by the sternum of the antennary somite. On each

ASTACUS FLUYIATILIS.

273

side of its triangular anterior extremity it presents a wide
articular cavity for the articulation of the antennae. In these
organs (Fig. 72, -S, d) the same parts can be recognized as in

Fig. 72.-4. ABtcrior extremity of the cephalo-thorax of Astaciis, with a portion of
the carapace removed. B. Vertical section of the anterior part of the cephalo-
thorax: o, rostrum ; b, ophthahnic peduncles ; c',antei:niilge ; d, antennae; e. la-
brum ; /, metastoma ; g^ oral aperture ; h. procephalic processes ; i, ophthalmic
sternum ; k^ autennulary sternum ; /, antennary sternum or epistoma.

the other appendages, viz., an imperfect basal joint, produced
into a prominent cone, perforated behind and internal to its
apex, and here called coxocerite. Next, a basicerite^ to the
outer portion of which a flattened plate, the representative
of the exopodite, and here called the scaphocerite^ is articu-
lated; wliile to its inner portion an ischiocerite is connected,
bearing a nierocerlte and carpocerite^ while the last segment,
oi procerite^ consists of a long multi-articulate filament.

The sterna of the next two somites are narrow and elon-
gated; that of the antennary somite is well calcified, but
that of the oplithalmic somite is almost entirely membranous.

The antennules (Fig. 72, 13, c) present an enlarged trigonal
basal joint, succeeded by two others. These represent the
protopodite, and carry at their extremities two many-jointed
filaments, which probably represent the exo- and endopodites.

The peduncles of the eyes (Fig. 72, 5), lastly, are com-
posed of two joints, a small proximal basiophthalmiie, and a
larger terminal podophthcdmite.

Such are the structure and arrangement of the sternal por-
tions of the several cephalo-thqracic somites, and the nature
of their appendages. On regarding the sternal region as a
whole, there are yet some very important points (the morpho-
logical value of wdiich has been fully pointed out by Milne-
Edwards) to be noticed. A longitudinal median section,
in fact, shows that, while a line drawn through the sterna of
the somites behind the mouth is nearly straight and parallel

274 THE ANATOMY OF INVERTEBRATED ANIMALS.

with the axis of the body, a similar line drawn through the
sterna of the somites, in front of the mouth, ascends as it
passes through the antennary, antennulary, and ophthalmic
sterna, and thus takes a position at right angles to the former
line (Fig. 72, B), The sterna of the somites, in front of the
mouth, are, therefore, bent up so as to look forward instead
of downward ; and it is of essential importance to bear in
mind this cephalic flexure, in considering the structure of the
head in these and other Arthropoda,

Just as the lateral regions of the abdominal somites are
produced into the pleura, so are the lateral regions of the
cephalo-thorax similarly prolonged. Thus the membranous
lateral walls of the posterior cephalo-thoracic somite are re-
flected superiorly, and bent down again to the level of the
bases of the legs, where they become continuous with a calci-
fied layer corresponding with the tergal half of the pleura,
and forming the posterior part of the carapace. In like man-
ner, the more or less calcified epimera of all the other somites
are reflected superiorly into a membrane which passes down-
ward, and the free lower edge of which is continuous with
the edges of the carapace. The carapace, therefore, corre-
sponds in position with the terga and tergal halves of the
pleura of all the somites which are thus reflected into it, and
these somites include all, without exception, from the last
thoracic to the ophthalmic. Posteriorly, the edges of the
carapace are a little prolonged beyond the last thoracic somite,
and take the form of a fold, with an under layer distinct from
the upper. Anteriorly, in the middle line, the carapace is
prolonged in a similar manner, but to a much greater extent;
it thus gives rise to the long rostrum, which overhangs the
sterna of the ophthalmic and antennulary somites. At the
sides of the antennulary and antennary somites the rostral pro-
longation of the carapace is the direct continuation outward of
the epimera of those somites, and there is nothing to be com-
pared to an apodeme ; but the sternum of the ophthalmic so-
mite, after giving off the lamella which forms the inferomedian
region of the rostrum, is prolonged on each side of the middle
line backward and outward into a free, expanded, thin, cal-
cified process, which applies itself against the carapace by its
upper surface, and by its under surface gives attachment to
the anterior gastric muscles. Corresponding processes are
developed from the carapace itself, in some Podophthalmia
(e. g., Galathea, Carcinus), for the attachment of the poste-
rior irastric muscles. From the last thoracic to the maxilli-

ASTACUS FLUVIATILIS. 275

pedary somites, the pleural, or free part of the carapace,
termed, from its function, the hranchiostegite^ or cover of the
gills, incloses a wide space, bounded internally by the epimera
of the somites. This is the branchial chamber. In front of
the maxillipedes and cervical fold, however, the chamber sud-
denly becomes narrowed by the rapid widening of the sterna
of the maxillary and mandibular somites, and by the lowering
of the point at which the reflection of their epimera into their
pleura takes place. Finally, on the antennary somite, and in
front of it, the pleuron becomes a mere fold separated by a
shallow groove, the representative of the branchial chamber,
from the epimera.

On the dorsal surface there is no indication of any divis-
ion of the carapace into terga corresponding with the sterna
of the somites, but it is marked by a well-defined, curved
groove, the posterior convexity of which extends across the
carapace, rather behind its middle, and the lateral portion of
which runs downward and forward, toward the anterior part
of the antennary sternum. This is the cervical groove ^ that
part of the carapace which lies in front of it is the cejjhaloste-
gite, while that which is behind is the otnostegite.

The omostegite, again, is divided into three portions by a
groove on each side of the middle line — the branchiocardiac
grooves. The branchiocardiac groove, and the lateral por-
tion of the cervical groove, on the dorsum of the carapace,
correspond very closely with the line at which the epimeral
is reflected into the pleural membrane, on its ventral surface.
The transverse portion of the cervical groove, on the other
hand, corresponds with the posterior boundary of the stom-
ach and the anterior extremity of the heart, and continues
inward the line of the cervical fold ; so that, in a longitudi-
nal section of an Astacus, the direction of the cervical fold,
if followed upward and backward, strikes against the inner
surface of the carapace, at a point corresponding with the
summit of the cervical groove, on its outer surface. By cut-
ting through the cervical fold, therefore, through the mem-
brane joining the second maxillary with the first maxillipe-
dary sternum, and through the carapace in the transverse
part of the cervical groove, it is possible to separate an ante-
rior portion of the cephalo-thorax, containing the whole of
the cephalostegite, and the first six somites, with their ap-
pendages, from a posterior portion, consisting of the omos-
tegite, and the last eight cephalo-thoracic somites. And, in
making this artificial separation, we should be merely carry-

276 THE ANATOMY OF INVERTEBRATED ANIMALS.

ing out a distinction between these two sets of somites,
already very clearly indicated by the cervical fold and groove.

It is for this reason that I differ from Milne-Edwards in
regarding the somite which bears the first maxillipedes as the
first of the thorax, and not as the last of the head. And the
acceptance of this natural delimitation of the head in the
higher Crustacea has the advantage of bringing its structure
into accordance with that of the same region in the Ento-
mostraca, in which it is the rule that the head possesses eyes,
antennules, antenna?, mandibles, and two pairs of maxillas.

Another mark upon the carapace is a large and rounded
convexity, occupying nearly a third of the whole width of
the posterior half of the cephalostegite. This impression is
bounded internally by a line drawn from the outer angle of
the base of the rostrum, directly backward, and externally
by a curved depression, deepening into a pit anteriorly; it
corresponds with the attachment of the base of the adduc-
tor muscle of the mandible.

The mouth of the Craj^fish is a wide aperture, situated
between the labrum in front, the metastoma behind, and the
mandibles on each side. It serves as the entrance to an
equally wide oesophagus, a short tube with plaited walls,
which takes a slightly curved direction upward and a little
backward, to open into the large stomach, which is not only
situated directly over, but extends forward in front of the
gullet. The stomach, in fact, occupies almost the whole
cavity of the body in front of the cervical suture, and is
divided by a constriction into a large anterior moiety, the
c.xrdiac division, and a small posterior, pyloric portion. The
anterior half of the cardiac division has the form of a large
membranous bag, the inner surface of which is closely set
with minute hairs ; but in the posterior half of this, and on
the whole of the pyloric division, the walls of the stomach
are strengthened by a very peculiar arrangement of uncalci-
fied and calcified plates and bars articulated together, which
are thickenings of the chitinous cuticula of the epithelium
of the alimentary canal, and constitute the gastric skeleton.
The most important part of this apparatus is that which is
developed in the posterior cardiac region.

It consists, in the first place, of a transverse, slightly
arcuated cardiac plate (Fig. 73, cr?), calcified posteriorly,
which extends across the whole width of the stomach, and
articulates at each extremity by an oblique suture with a
small curved triangular antero-lateral or pterocardiac {pt)

ASTACUS FLUVIATILIS. 277

ossicle. On each side a larcre, elongated postero-lateral or
zygocardiac ossicle (se), wider posteriorly than anteriorl}', is
connected with the lower end of the antero-lateral ossicle,
and, passing upward and backward, becomes continuous with
a transverse arcuated plate, calcified in its anterior moiety,
and situated in the roof of the anterior dilatation of the py-
loric portion; this is the pyloric ossicle (Fig. '^S,py),

These pieces, it will be observed, form a sort of six-sided
frame, the anterior and lateral angles of which are formed by
movable joints, while the posterior angles are united by the
elastic pyloric plate.

From the middle of the cardiac piece a strong calcified
iirocardiac process {ca') extends backward and downward,
and, immediately under the anterior half of the pyloric ossi-
cle, terminates in a broad, thickened extremity, which presents
in feriorly two strong rounded tuberosities, or cardiac teeth.
With this process is articulated, posteriorly, a broad p>re-
jjyloric ossicle, which passes obliquely upward and forward, in
the front wall of the anterior dilatation of the pyloric portion,
and articulates with the anterior edge of the pyloric ossicle,
thus forming a kind of elastic diagonal brace between the
urocardiac process {ca') and the pyloric ossicle. The inferior
end of this pre-pyloric ossicle is produced downward into a
strong bifid urocardiac tooth {ac). FinpJly, the inner edges
of the postero-lateral ossicles are flanged inward horizontally,
and, becoming greatly thickened ai.d ridged, form the large
lateral cardiac teeth {cc). The membrane of the stomach is
continued from the edges of the pre-pyloric to those of the
postero-lateral ossicle in such a manner as to form a kind of
pouch with elastic sides, which act, to a certain extent, as a
spring, tending to approximate the inferior face of the pre-
pyloric ossicle to the superior face of the median process of
the cardiac ossicle.

The result is, that there is a certain position of equilibrium
of the w^liole apparatus, when tlie urocardiac process and the
pre-pyloric ossicle make a small angle with one another, and
the antero-lateral ossicles form an almost unbroken transverse
curve with the cardiac. When undisturbed, the apparatus
tends to assume this position.

Two pairs of powerful muscles are attaclied to this gastric
skeleton. The anterior pair arise from the procephalic pro-
cesses, and are inserted into the roof of the stomach, some-
what in front of the cardiac ossicle; the posterior have their
origin in the carapace immediately above and behind the

278 THE ANATOMY OF INVERTEBRATED ANIMALS.

pyloric end of the stomach, and their insertion into the pylo-
ric ossicle and the wide posterior part of the postero-lateral
pieces.

Fig. 'tS.—Astacus.— Upper Figure: Longitudinal Section of Stomach.— J, anterior
gasitric muscle ; B, posterior gastric muscle ; CE", oesophagus ; P, pylorus ; ca,
cardiac ossicle ; ca'. its urocardiac process ; ac, urocardiac tooth ; py, py-
loric ossicle ; the ohlique bar, extending from the end of the cardiac to the pylo-
ric, is the pre-pyloric ossicle; pt, pterocardiac ; se, postero-lateral cardiac, with
its great tooth, cc ; I, small inferior tooth ; c, cardio-pyloric vaive ; b, infero-
medlan pyloric ridge; a, lateral pyloric ridge; d, superior pyloric ridge; up,
uro-pyloric ossicle ; oo y. line of section ; the anterior lace of the posterior segment
being shown in the lower figure.

From the attachment of these muscles it is clear that their
action must, in a general way, resemble that produced by
pulling upon the cardiac and pyloric pieces when the stomach
is removed from the body. Now, the result of doing this is
that, the cardiac and pyloric pieces being divaricated, the
pre-pyloric ossicle assumes a vertical position, and the uro-
cardiac tooth turns downward and forward. At the same
time the antero-lateral or pterocardiac pieces are pulled back*
"ward, and, owing to their oblique articulation with the car-

ASTACUS FLUVIATILIS. 279

diao piece, their inferior ends move downward, backward, and
inward, carrying with them the anterior ends of the postero-
lateral pieces, the teeth of which (lateral cardiac) come into
contact with the urooardiac and cardiac teeth with a force
proportional to that exerted in traction. On ceasing to pull,
the apparatus returns to its former position, its backward
movement being facilitated by the reaction of the elastic
pouch mentioned above, and being doubtless also assisted, in
the living state, by a pair of small cardio-pyloric muscles,
which pass, one on each side, between the cardiac and pyloric
ossicles, beneath the membrane of the stomach, the looseness
of which, in this region, where it unites the various ossicles
of the gastric mill, greatly assists the free movement of the
whole apparatus.

Nothing can be more easy than to perform the experi-
ment, and to convince one's self that these structures do really
constitute a most efficient masticatory apparatus ; and it is
surprising that Oesterlen, in his elaborate essay on the stom-
ach of Astacus, should have questioned the crushing action
of the teeth.

A great bilobed valvular process (Fig. 73, c) rises up from
the sternal region of the stomach, opposite the cardio-pyloric
constriction, and apparently prevents the food from passing
into the pyloric division until it is properly comminuted.
And, in front of this valve, the infero-lateral parietes of the
stomach are strengthened by a number of other plates and
bars ; one of which on each side bears a small tooth {infero-
lateral cardiac^ /), and is continued into a broad uncalcified
plate, lying in the hinder and lower part of the side- walls of
the stomach, and covered with hairs internally. There are,
therefore, altogether seven gastric teeth : three median, the
cardiac, and the urocardiac, and two lateral on each side, the
lateral cardiac and the infero-lateral cardiac.

In the pyloric division of the stomach the food has to
undergo a further series of comminutions and strainings. A
ridge covered with long hairs projects in the median line
above ; other hairy ridges extend inward from the sides to
meet it, and nearly close the passage laterally. These ridges
are very convex inferiorly, and their convexities abut against
the concavities of an inferior median ridge, which rises up to
meet them, and is prolonged posteriorly into a sort of valvu-
lar process, covered at its termination with long hairs, which
bar the space left between the upper parts of the lateral
ridges. The concave faces of this median process are covered

280 THE ANATOMY OF INVERTEBRATED ANIMALS,

by close-set parallel riclges, which only become free hair-like
processes at the posterior margin of the plate, each ridge
giving attachment to a regular series of minute hairs. These
are directed inward nearly parallel with the surface, which
looks at first as if it were merely ruled with close-set trans-
verse lines, connected by still finer and closer longitudinal
ones.

This apparatus constitutes the " ampoule cartilagineux "
of Milne-Edwards. Behind it there is yet another infero-
median and two lateral setose valvular prominences, which
form the last barrier between the food and the intestine.

Mr. T. J. Parker, who has recently carefully examined the
structure of the stomach of the Crayfish,^ finds that, besides
the anterior and posterior gastric and the cardio-pyloric mus-
cles, there are intrinsic fibres in the walls of tlie stomach,
some encircling the posterior pyloric region, others passing
between the hindermost accessory ossicle and the postero-
lateral and pyloric pieces ; these must tend to diminish the
cavity of the stomach, and the last-named fibres possibly
assist in mastication by bringing the lateral cardiac into con-
tact with the infero-lateral cardiac tooth. Moreover, there are
nine pairs of minor extrinsic muscles, of which two pairs pass
from the anterior wall of the stomach and gullet to the anten-
nary sternum, passing between the oesophageal commissures
and on either side of the azygos nerve of the visceral system ;
three pairs pass between the side-walls of the stomach and
oesophagus and the mandibular sterna; a sixth pair arises from
the forward processes of the intermaxillary apodeme, and is in-
serted into the oesophagus; two more pairs arise, one from
the internal thickened edge of the mandible, the other from
the intermaxillary apodeme, and are inserted into the inferior
surface of the pyloric region ; and a ninth pair arises from the
carapace just behind the posterior gastric muscles, and goes
to be inserted into the posterior pyloric dilatation. There
are also a few more inconspicuous fibres passing between the
oesophagus and the neighboring hard parts. All these, at
least when acting: togrether, must antag-onize the intrinsic
muscles, and dilate the stomach.

The pyloric portion of tlie stomach passes into the an-
terior portion of the intestine, which is smooth internally,
and presents superiorly a caecal process, the remains, accord-
ing to Rathke, of one lobe of the vitellary sac of the embrya

» Journal of Ancitomy ani P'tyshly^y^ Ojt>b3r, 18T6.

ASTACUS FLUVIATILIS. 281

This anterior portion of the intestine is, however, very-
short, and almost immediately becomes dilated into the wider
posterior division, which extends to the anus. The inner
surface of the dilatation is produced into six ridges, which are
continued into a corresponding number of series of papillge
along the rest of the intestine.

The only glandular apparatus of any kind which opens
into the alimentary canal is the liver, and the apertures of
the wide hepatic ducts are seen on each side of the pylorus.
Each duct conveys the secretion from the multitudinous
caecal tubes, which constitute the principal mass of the cor-
responding bilobed half of the liver. The two halves lie on
each side of the stomach, and, though they remain perfectly
distinct from one another, come into close contact below.

jistacKS possesses neither salivary glands nor any cascal
appendages to the intestine, such as exist in the JBrachyura
and some Macrura^ unless the short caecum just now de-
scribed is the homologue of the longer cseca of Maia and
Ilomarus.

In the spring and summer two very curious discoidal cal-
careous plates, the so-called "eyes" of the Crayfish, are
found imbedded in the walls of the dilated anterior portion
of the cardiac division of the stomach, the middle of the
lateral surface of which they occupy. These bodies com-
mence as calcareous deposits underneath the chitinous gas-
tric lining, and increase in size until the period arrives at
which the Crayfish casts its skin. They are then cast, to-
gether with this lining membrane and the gastric armature ;
and it would appear that, like the latter, they become broken
up and destroyed within the new stomach. The purpose of
these concretions is not understood; the ordinary theory,
that they are stores of calcareous matter, ready to be dis-
tributed through the young integument after ecdy^s, appear-
ing to be negatived by their small size. Oesterlen states
that they rarely weigh more than two grains, and judiciouslj^
suggests that if it be admitted that the Crayfish can derive
all the calcareous matter it requires, except two grains, from
other sources, it is hardly necessary to look on those two
grains as a special supply.

The circulatory apparatus of Astacus is well developed.
The heart (Fig. 74, C) has the shape of an irregular polygon,
and lies immediately behind the stomach and beneath the
cardiac region of the carapace, in a chamber which is com-
monly termed the " pericardium," to the walls of which it is

282

THE ANATOMY OF INVERTEBRATED ANIMALS.

attached by six ligaments, corresponding with the alge of the
heart in insects, but not, like them, muscular. Except by

Fio. 74.—Astacus, Longitudinal Section,— I. II, III. Sterna of flret, second, and third
somites ; <b, oe8opha<;us ; lb, labrum ; /, metastoma ; G, membranous part of the
stomach; c, cardiac ossicle; ;j<. pterocardiac ; mc, urocardiac ; c/, lateral cardiac;
p, cardio-pyloric valve ; pi. Inferior pyloric valvular apparatus ; m, anterior gastric
muscle ; m*, insertion of posterior gastric muscles ; pc, procephalic processes ;
A', opening of hepatic duct ; v, pyloric caecum ; i, k, intestine ; gn, testis ; gn', gn'\
vaa deferens ; C, heart ; ao, ophthalmic artery ; aa, antennary ; ah, hepatic ; as,
eternal ; ap, superior abdominal artery ; 6, cerebral ganglia ; sg, azygos visceral
nerve.

these ligaments, and by the arteries, which pass through it,
the walls of the pericardial cavity, or blood sinus (for such it

ASTACUS FLUYIATILIS. 283

really is), are wholly unconnected with the heart, which thus
is, in a manner, suspended freely in the blood.

Six apertures, two of which are superior, two inferior,
and two lateral, provided with valves which open inward,
allow the blood to enter the cavity of the heart during the
diastole, and prevent its egress, except by the arteries, dur-
ing the systole. The arterial trunks are six in number, five
being given off anteriorly, and the other from the posterior
portion of the heart.

Of the five anterior arteries, one, the ophthalmic, is single,
and situated in the middle line ; it passes forward on the
stomach to the head, where it supplies the eyes and anten-
nules. The other arteries are in pairs ; two pass on the
stomach forward and outward, giving off branches to the
carapace, and eventually supplying the antennae ; the other
two pass downward, between the anterior lobes of the geni-
talia, and divide into a multitude of branches upon the he-
patic CcEca.

The posterior trunk, or sternal artery, is the largest of
all, and presents a sort of bulbus arteriosus at its commence-
ment. It turns almost directly downward, usually on the
right side of the intestine, to the sternal canal, which it
enters, passing between the antepenultimate and penultimate
thoracic ganglia to the lower surface of the ganglionic cord ;
it gives off two abdominal branches, one superior, close to its
origin from the heart, which traverses the m.iddle of the ter-
gal region above the intestine, the other inferior, which takes
a corresponding course along its sternal region beneath the
nervous system. The arterial trunks are provided with valves
at their commencement, so arranged as to prevent the regur-
gitation of the blood. They ramify minutely, but how far a
capillary system can be said to exist, is a question requiring
further investigation. In transparent Zocece, I have plainly
observed the abrupt termination of the arterial trunks by
open mouths, through which the blood was poured into wall-
less lacuna?, and into the general cavity of the body ; nor can
there be the least doubt that a similarly lacunar condition
of the circulation exists in those lower adult Crustacea^ the
transparency of which allows of their examination with the
requisite powers of the microscope. The probability is that
a similar state of things obtains in the vascular system of all
other Crustacea, and that, after undero-oino: a srreater or less
amount of subdivision, the arterial vessels, or their capillary
continuations, cease to exist, the blood then making its way

284 THE ANATOMY OF INVERTEBRATED ANIMALS.

into lacunas between the organs, and into the general peri-
visceral cavity ; and, as in most Jlollusca, ceasing to be con-
tained in vessels with distinct walls.

The blood thus poured out eventually makes its way into
irregular sinuses or reservoirs, the chief of which, lodged in
the sternal canal, communicates by lateral channels with oth-
ers which lie above the bases of the thoracic appendages,
and from which the afferent branchial canals pass into the
stems of the branchiae, on the exterior faces of which they
ascend, giving off branches to the lateral filaments. Corre-
sponding canals return the blood from these filaments to the
efferent branchial canals, which run down the inner side of
the branchial stems, and unite above the bases of the limbs
into six trunks, which ascend beneath the epimera and open
into the sides of the pericardial sinus. The floor of this
sinus is formed by a continuous membrane, which appears to
shut it off completely from the general visceral cavity (at
least it retains air or fluid thrown into it), and, if this be
really the case, it may be said to be functionally a branchial
auricle, containing pure unmixed aerated blood.

The branchias are eighteen in number upon each side, and
are attached from the eighth to the fourteenth somites in-
clusively. Six of these branchiae are attached to the epipo-
dites of the eighth to the thirteenth somites, and differ very
considerably in appearance from the other twelve. Each
epipodite is, in fact, expanded at its upper extremity into a
broad, bilobed membrane, which is folded upon itself, so that
the two lobes are directed posteriori}-, and receive the epipo-
dite of the next limb (Fig. 71, JV). The membrane of the
lobes is obliquely plaited, so that, doubtless, they subserve
respiration to a certain extent ; but, in addition, the anterior
edge of the epipodite is beset with a number of branchial
filaments, similar to those on the other branchiae.

The latter (Fig. 71, ^,f, g) are simple plumes, consisting
of a stem, to which are attached many delicate, cylindrical
filaments. Two of these plumes are attached to the epimera
and coxo-epimeral articular membranes of the ninth, tenth,
eleventh, twelfth, and thirteenth somites. They increase in
size posteriorly. The eighth and fourteenth somites, on the
contrary, only carry one plume. A tuft of long byssus-like
filaments is attached to the coxopodite of each of the last
six thoracic appendages (Fig. 71, F, M),

The respiratory organs of the Crayfish, not being pro-
vided with cilia, require some special arrangement for the

ASTACUS FLUVIATILIS.

285

renewal of the water with which they are in contact. This
object is attained principally by the action of the scapho-
gnathite, which lies immediately behind the anterior opening
of the branchial chamber ; and, during life, is incessantly in
motion, baling out, as it were, the water w4iich has become
impure through the anterior opening, and thus compelling
the flow of fresh fluid into the branchial chamber through its
posterior and inferior opening, constituted by the space left
between the lower edge of the branchiostegite and the bases
of the limbs.

The nervous system of Astacus * is composed of thirteen
principal ganglionic masses, of which one, cerebral, lies in
the head, in front of the mouth ; six, thoracic, are situated
in the sternal canal ; and six, abdominal, lie in the median
sternal resfion of the six anterior somites of the abdomen.

The cerebral ganglia (Fig. 74, h ; Fig. 75, a) give off
nerves to the eyes and to the muscles of the ophthalmic
appendages ; to the antennules and the auditory organs
which they contain ; to the antennae and the sac of the
antennary gland ; to the carapace in front of the cervical
suture ; and, finally, they send posteriorly two long and

Fig. 75.— Visceral nerves of Astacus.— a. cerebral ganglia ; 6, commissures— that of
the right eide is cut aud turned back ; c, transverse cord uniting them behind
the oesopha^s, CE ; d, d, d, azygos nerve ; h, ganglion; i, lateral branch of
azygos, uniting with postero-lateral nerve g ,• e, antero-lat3ral nerve ; /. medio-
lateral nerve; k^ hepatic nerve ; P, pyloric; C, cardiac portion of stomach.

stout commissural cords to the anterior thoracic ganglionic
mass. These commissures are connected bv a transverse

* For tlie histology of the nervous system, see an elaborate essav by
Haeckel on the minute structure of the tissues of the Crayfish, ia the ' ArchiV
far Anatomie," 1857.

286 THE ANATOMY OF INYERTEBRATED ANIMALS.

cord immediately behind the oesophagus (Fig. 75, <^). The
size and form of the anterior thoracic ganglion would lead to
a suspicion of the complex nature which development shows
it to possess. It supplies the somites and their appendages
from the fourth to the ninth inclusively, and sends forward
delicate filaments to the oesophagus.

Posteriorly it is connected with the ganglionic mass of
the tenth somite by two commissures, and the other thoracic
ganglia are similarly brought into communication, the com-
missures of the ultimate and penultimate only being remark-
able for their brevity. The abdominal, which are much
smaller than the thoracic ganglia, are, with the exception of
the last two, united by single cords, which represent coa-
lesced double commissures. Each of these ganglia supplies
the muscles and the appendages of the somite to which it
belongs, and the posterior abdominal ganglion sends branches
into the telson.

The Cra3'fish possesses a remarkably well-developed sys-
tem of visceral or stomatogastric nerves, wliich has been the
subject of special study by Brandt, Milne-Edwards, Krohn,
and Schlemm, each of whom has described a larger or smaller
portion of the system wdth accuracy, but has omitted to men-
tion, or has denied, the existence of some other part. Each
of the great commissures (Fig. 75, 5), as it passes over the
sides of the oesophagus, becomes slightly swollen, and from
the enlargement four nerves arise ; one, external, passes
toward the mandibular muscles ; a second postero-lateral
branch (Fig. 75, g) runs upward and backward to the infero-
lateral regions of the stomach, and eventually enters into the
composition of the hepatic nerve {k) ; a third branch (f)
turns directly inward and upward, and unites upon the
oesophagus with its fellow and with an azygos nerve (<:/),
which passes up in the middle line of the anterior face of the
oesophagus and stomach, and enters a ganglion placed be-
tween the anterior gastric muscles (A), from whence a lateral
branch is given off on each side, w^hile a posterior median
branch {d) continues the direction of the azygos nerve.
Having reached the cardiac ossicle, this nerve divides into
two branches (z), each of which passes downward and out-
ward, unites with the postero-lateral nerve of its side, and
thus forms the hepatic nerve {Jc). The fourth and last, or
antero-lateral branch (e), descends at first to near the mouth,
and then, curving forward, ascends to unite on the anterior
face of the oesophagus with the anterior continuation of the

ASTACUS FLUYIATILIS. 287

azygos nerve, which passes forward and upward and enters
the cerebral mass. I am inclined to think that this part of
the azygos nerve forms a portion of a fine plexus of nervous
filaments which pass from the cerebral ganglia backward to
the lining membrane of the carapace ; but the dissection of
these fine filaments, and the demonstration of their conti-
nuity, is a matter of no ordinary difficulty.

The intestine is supplied by two nerves which aiise from
the last abdominal ganglion, and unite into a single trunk,
from which small branches are given off backward, and two
principal ones forward, which supply the greater part of the
intestine. According to Brandt, the genitalia receive branches
of the fourth, fifth, and sixth thoracic ganglia.

The only certainly known organs of sense in Astacus are
the eyes and the auditory organs. The eyes are seated at
the extremities of the ophthalmic peduncles, the integument
of the outer extremity of which becomes translucent over a
reniform space, and constitutes the corneal membrane. This
membrane is divided into a great number of minute quadri-
lateral facets, each of which corresponds with the base of a
crystalline cone.*

The upper face of the trihedral, proximal, and largest
joint of the antennule presents an oval space, covered by a
broad brush of complex hairs having their points all directed
inward. On cutting these hairs away close to their bases,
however, it is seen that they cover an aperture, wider above
than below, and about one-sixteenth of an inch long. The
hairs are attached to the outer lip of this aperture, and some
are directed so as to lie within the inner lip, but the majority
cover it. A good-sized bristle passes with great ease into
this aperture, and if the inner and outer walls of the basal
joint of the antennule be now removed, and the soft parts
carefully dissected away, the end of the bristle will be seen
to have passed into a wide delicate sac about one-twelfth of
an inch long, which is attached by a narrower neck round
the aperture, the lips of which are continuous with its walls.
The sac is filled with minute sandy particles, suspended in a
mucous, dirty-looking fluid, and when emptied of these con-
tents a band, consisting of several lines of very fine hairs,
like those which guard the mouth of the sac, but more deli-

^ Mr. E. T. Newton's careful description of the eve of the Lobster in TJie
Quarterly Journal of Microscopical Science for 1875, to which I have referred
above, may be taken as a guide to the study of the minute structure of the eye
in the Crayfish,

288 THE ANATOMY OF INVERTEBRATED ANIMALS.

cate, is seen to skirt its inner contour. The hairs, projecting
inward, come into close contact with the solid particles sus-
pended in the mucous fluid.

A nerve may be traced accompanying the antennulary
nerve to the sac, and appears to be distributed principally
alono: the setio^erous baud, so that the extremities of the
nerve fibrils come into close relation with the bases of the
hairs. Some, if not all, of the sandy particles are insoluble
in strong acetic acid, and would appear to be siliceous.^

Two glandular sacs commonly known as the green glands^
which were formerly regarded as the auditory organs, lie in
the cavity of the head. An aperture is visible on the inner
or oral side of a conical prominence, upon the inferior portion
of the coxal joint of the antenna. A bristle passed into this
aperture enters a large but very delicate and transparent
sac, filled with a clear fluid, which is usually conspicuous on
each side of the anterior end of the stomach, when the cara-
pace is carefully removed. A nerve which comes ofi" from
the cerebral mass close to the antennary nerve, passes to the
neck of this vesicle, and is distributed over its surface be-
tween the outer and inner membranes, of which it is com-
posed. Inferiorly the vesicle rests upon a large greenish, ap-
parently glandular mass, but is directly connected with the
latter only at two points, firstly by a vascular cord, which
passes to the central and usually more yellow portion of the
gland, and secondly by a short neck-like continuation of the
sac itself, which is attached over a small circular space, mid-
way between the centre and the periphery of the gland, and
opens into the circular principal duct of the gland. There is,
therefore, a free communication between the cavity of the
gland and the exterior by means of the sac, which is, in this
respect, simply a dilated duct. A section of the gland shows
it to be composed of two subtances, a central and a cortical.
The latter is composed of minute caeca, filled with a homo-
geneous gelatinous matter, containing many large nuclei ;
the former is traversed in all directions by large canals, so as
to have a spongy appearance. The casca open into the ulti-
mate ramifications of the canals, and the spongy, lung-like
texture of the central mass seems to arise merely from the
very free anastomosis of their larger branches, which event-

* See^ for a full account of the minute structure of tlie aiiditory organs in
the higher Crustacea, llenseu'a "Studien iiber das Gehororgan der Deoa-
poden,'' 1833.

ASTACUS FLUVIATILIS. 289

ually enter the circular canal which communicates with the
vesicle.

There is little in these structural features to suggest an
organ of special sensation, but much to show that the green
mass is a secreting organ, and that the vesicle acts (whatever
other purposes it may subserve) as its duct. In all proba-
biUty the green gland is an organ of the same nature as the
shell gland of the Entomostraca.

Leydig has attributed an olfactory function to certain
groups of delicate setse which occur on the joints of the outer
division of the antennule of the Crayfish.

The most remarkable part of the muscular system of the
Crayfish is the great extensor muscle of the abdomen, a com-
plex mass of fibres which is attached in part to the endo-
phragms of the thorax in front, and, behind, to the sterna of
the abdominal somites, a large part of the cavity of which it
occupies.^

The essential parts of the reproductive organs in the male
and female Astaciis are very similar to one another in form,
both ovarium and testis having the figure of a trilobed gland,
situated immediately behind the stomach, and below the
heart. Two of the lobes are applied together, and pass for-
ward ; the other lobe is directed in the middle line back-
ward. The ducts take their origin, one on each side, at the
junction of each antero-lateral with the posterior lobe.

In minute structure, however, the two organs differ wddely.
Each lobe of the testis is composed of a number of small
caeca, in which the spermatozoa are developed, and w^hich
open into a central duct. The ovarium, on the other hand, is
essentially a wide sac, produced into three large caeca, each
of which corresponds with a lobe ; and the ova are developed
in the epithelial lining of the sac. The efferent ducts, again,
have little resemblance, the oviducts being short, wide tubes
which open on the coxopodites of the antepenultimate thora-
cic appendages, while the vasa deferentia are canals as long
as the body, at first very narrow, but afterward widening,
which lie coiled up on either side of the posterior part of the
thoracic cavity, where their white contents make them very
conspicuous (Fig. 74, ^?^'). Eventuallv, they open on* the
coxopodites of the posterior thoracic appendages.

The spermatozoa, like those of many other Crustacea, are

* For details, see Suclcow, " Anatomisch-pbysioloorisclie ITntersuchunffen."
Milne-Edwards has described the muscles of the Lobster at length in the " His-
toire naturelle des Crustaces," torn. i.

13

290 THE ANATOMY OF INVERTEBRATED ANIMALS.

motionless, and have the form of cells, provided with a nu-
cleus and produced into several delicate radiating processes.
They are united in their course down the vas deferens into
cylindrical masses, w^hich, becoming invested by a fine mem-
branous coat, probably secreted by the walls of that duct,
constitute the spermatophores, which may not unfrequently
be found adhering to different parts of the body, not only of
female but of male Crayfish.

The ova are fecundated while still within the parent ; they
become surrounded, in their passage down the oviduct, by a
coat corresponding with that of the spermatophore, which is
produced into a pedicle, the extremity of which becomes at-
tached to one or other of the abdominal appendages. Great
numbers of ova, attached in this way, may be observed, dur-
ing the breeding-season, within the incubatory chamber formed
by the flexure of the abdomen upon itself; and it is in this
cavity that the embryos pass through the whole of their foetal
existence.

The development of the Crayfish has been the subject of
one of the most beautiful of the many admirable memoirs on
development, for which we are indebted to the genius and
patience of Rathke.* After fecundation a blastoderm arises
upon the surface of the yelk, and, gradually extending over
the whole yelk, becomes thickened at one part, so as to form
an oval germinal disk, with a central depression.

This disk next becomes widened and bilobed at its ante-
rior extremity, the lobes being identical with the procephalic
lobes, to be hereafter described in the embryo of 3Iysis. The
edges of the disk are raised into a fold, and within the fold a
papilla, the rudiment of the abdomen, and of the greater part
if not of the whole of the thorax, makes its appearance ; while,
anteriorly, three pairs of transverse elevations constitute the
rudiments of the antennules, the antennjB, and the mandibles.
The labrum arises as a median papilla, situated at first be-
tween the antennules. The ocular peduncles are next devel-
oped in front of the antennules as ridges, which only subse-
quently become free processes.

The thoracico-abdominal process lengthens, and the anal
aperture makes its appearance. It is to be remarked that
the anus is at first situated on the dorsal side of the extremity

1 "Ueber die Blldung und Entwickeluno^ des Flusskrebses," Bd. 29. See
also Lereboullet, "Recherclies d' Embryolopf ie comparee sur le Ueveloppement
du Brochet de la Perche et de I'Ecrevisse," 1862; and the account of Bobret-
sky's researches in Ilofmann and Schwalbe, " Jahresbericht " for ISTS (1875).

THE DEVELOPMENT OF ASTACUS. 291

of tlie abdomen, and that there is no telson. This is devel-
oped only at a much later period from the dorsum of the end
of the abdomen, and, by its outgrowth, forces the anus to
the ventral side of the body.

In the mean while, the oral aperture is developed behind
the labrum, which moves backward ; while the maxillie, max-
illipedes, and ambulatory feet appear in succession as eleva-
tions or ridges of the substance of the embryo, which are, at
first, all alike, and gradually become specialized into their
ultimate forms.

When these appendages first appear, the maxillae and first
pair of maxillipedes are attached to the embryo in front of
the thoracico-abdominal process, the second maxillipedes lie
in the angle between them, and the third maxillipedes and
following appendages are attached to the sternal surface of
the thoracico-abdominal process itself; and, as this process is
at first bent forward upon the rest of the germ, it follows that
the appendages attached to it look upward, while those at-
tached to the anterior part of the embryo look downward.
As development proceeds, however, the embryo gradually
straightens itself, more and more of the anterior part of the
thoracico-abdominal process becoming continuous in direction
with the anterior part of the embryo ; until, at length, the
whole of the cephalo-thoracic portion forms a convex surface,
parallel with the vitellary membrane, only the abdomem re-
maining bent upon the cephalo-thorax. The middle portion
of the carapace is formed by the continuous calcification of
the dorsal walls of the cephalo-thorax of the embryo. Its
pleura are developed as two distinct folds, one of which, the
rudiment of the branchiostegite, encircles the embryo poste-
riorly, and extends forward on each side as far as the mandi-
bles ; while the other, the rudiment of the rostrum, and an-
terior cephalic pleura, is developed in front of the eyes, and
extends on each side to meet the former. Rathke's clear ac-
count of this matter is in perfect accordance with what I have
observed in My sis, and shows conclusively that the carapace
is not developed from any one or two somites in particular,
but that its tergal portion corresponds with, and is formed by,
the terga of all the cephalo-thoracic somites, while the bran-
chiostegites and rostrum are developments of the lateral por-
tions of all these somites; in fact, represent their pleura,
which, like the terga, are connate and continuously calcified.

The appendages are thus, at first, similar to one another,
and each consists of a ridge which eventually takes the form

292 THE ANATOMY OF INVERTEBRATED ANIMALS.

of a plate, free at the outer end. This plate, in all the mem-
^ bers, except the ophthalmic peduncles and the mandibles,
then becomes bilobed externally, the inner lobe representing
the endopodite, while the outer is the representative of the
exopodite and epipodite. The two latter, when they are in-
dependently developed, become separated by the division of
the outer lobe. The gills arise partly as outgrowths from
the epipodites, partly as distinct processes from the parts to
which they are eventually attached. The division of the
limbs into articulations takes place from their distal toward
their proximal ends. The heart appears late, at the posterior
extremity of the cephalo-thorax, and therefore behind the
yelk-sac.

The nervous system of the post-oral portion of the ceph-
alo-thorax consists at first of eleven pairs of ganglia, cor-
responding with the mandibles, maxillae, maxillipedes, and
ambulatory legs. The six anterior post-oral ganglia of each
side soon coalesce in pairs, so as to form as many single gan-
glia ; and of these the four anterior, namely, the mandibular,
the two maxillary, and the first maxillipedary ganglia, unite
into a single mass ; the two hinder ganglia, that is to say,
those of the second maxillipedary somite, next coalesce in
the same way, and it is only subsequently that the two masses
thus formed become fused into the single anterior post-oral
ganglion of the adult. The other ganglia not only remain
separate, but become wider apart with advancing age. A
ridge on each side of the oesophagus at first represents the
cerebral ganglion and the commissural cords, the latter being
developed out of the posterior part of the ridgo, and the for-
mer from its anterior portion. The cerebral ganglia are at
first two on each side, but the posterior, whence the nerves to
the antennary organs proceed, is much larger than the other,
and would appear to represent two ganglia. The endoster-
nites arise as processes from each of the eight posterior ceph-
alo-thoracic sterna, which eventually arch over the gangli-
onic cord, and unite with one another.

The alimentary canal is produced by the gradual differen-
tiation and demarcation of the sternal part of the hypoblast,
which invests the whole yelk, from the tergal part, which be-
comes the velk-sac*

J Accorditifr to BohretsTcy (/. c.) there is no proper yelt-sac, tlie structure so
termed by Eathke beinsf the saccular hypoblast, "which is formed by invasrina-
tion of the primitive blastoderm, and encroaches upon the vitellus, until the
latter is all absorbed. The hypoblastic sac is converted into the liver and the
intestine. The stomach arises independently by invagination of the epiblast.

THE PODCPHTHALMIA. 293

After the liver, genitalia, and antennary glands are de-
veloped, the yelk-sac eventually becomes reduced to a small
c^ecal diverticulum, situated at the pyloric end of the stom-
ach. The genital ducts in both males and females are origi-
nalh'^ diverticula from the corresponding regions of the geni-
tal glands ; their external apertures and the copulatory ap-
pendages of the first abdominal somites in the male are not
developed until some time after birth.

The modifications of structure observable within the limits
of the Podophthalinia are exceedingly interesting.

Excluding, for the present, the ISquillidce, the group is
divisible on clear morphological grounds into the following
subdivisions : 1. The Brachyura ; 2. The Anomura j 3. The
Macrura ; 4. The Schizopoda.

The morphological relations of the Macrura are nearly
such as are indicated by their position in this series ; and
A-Stacus^ as a central genus of the central group, thus be-
comes a sort of natural centre for the whole of the Podoph-
thalinia^ whence we may trace a gradual series of modifica-
tions, leading on the one hand to the Schizopoda^ with their
large abdomen and small cephalo-thorax ; and on the other to
the Brachyura^ with their rudimentary abdomen and com-
paratively enorm.ous cephalo-thorax.

In all the Macrura the branchi« are numerous, and are
covered by the branchiostegites. The abdomen is large, and
is used as a locomotive organ, the appendages of its sixth
somite being well developed. The thoracic ganglia usually
form an elongated chain, and the external maxillipedes never
form broad opercular plates over the other jaws. In some of
the lower Macrura (^Peneus^ Pasiphcea) the exopodite per-
sists as an appendage at the base of the thoracic limbs ; and
in two genera, Sergestes and Acetes, the posterior thoracic
members become rudimentary, or even entirely abortive,
though the abdominal appendages remain.

In the higher Macrura^ such as Palmums^ the nervous
system exhibits a greater degree of concentration, the tho-
racic ganglia constituting an elongated oval mass ; and it is
in this genus and its allies that the head and its appendages
exhibit modifications, which prepare us for those which are
presented by the Prachyura. In this respect the Palinurus
vulgaris (Rock Lobster, Sea Crayfish, or Spiny Lobster) is
particularly worthy of attention. The rostrum is rudimentary
and represented by a mere spine, leaving the anterior cephalic

294 THE ANATOMY OF INVERTEBRATED ANIMALS.

somites uncovered. The cephalic flexure is so strong as to
throw the ophthahnic sternum, whioh is very wide, completely
to the top of the head. The basal joints of the antennae, or
coxocerites, are enormous, fixed to the surrounding parts, and
united by their anterior extremities in the middle line below.
Superiorly, they seemed to have coalesced with the antennu-
lary sternum, so as to form a projecting wedge-shaped mass,
which separates the antennules from the ophthalmic sternum,
and causes them to appear, at first, as if they were inferior
to the antennas. In this genus, the basicerite, ischiocerite,
and merocerite are much thicker and stronger than the cor-
responding joints of any of the other appendages ; and in the
closely allied Scyllarus^ the facial region of which is, on the
whole, similarly constructed, these joints become extremely
expanded and flattened, and are succeeded by no procerite.
In these genera the scaphocerite, or squame, usually attached
to the base of the antenna, is absent ; and, in Scyllarus^ there
is another approximation to Brachyuran structure in the ex-
istence of distinct orbits, formed by a lobe of the carapace,
which descends on tlie inner side of the ocular peduncle, to
mest the base of the antenna. No median septum is formed
by the rostrum, however, nor are the antennules capable of
being folded back into distinct chambers in any Macruran at
present known.

The Anomura are so completely intermediate in structure
between the Macrura and the Srachyura, that they need not
be specially noticed, except to draw attention to the singular
deviation from the ordinary habits and form of the higher
Crustaceans, presented by the Paguridm^ or Hermit Crabs,
so common on all coasts. Essentially Macruran in their or-
ganization, these Crustacea are distinguished from all true
Macrura by the uncalcified and soft condition of the integu-
ment of their unsymmetrical abdomen, the appendages of
which are for the most part abortive, those of the sixth somite
being modified so as to serve as claspers. It is by means of
these that the Hermit Crab retains firm hold of the columella
of the empty gasteropod shell into which it is his habit to
thrust his unprotected abdomen, and, covering over his re-
tracted body with the enlarged chela, which takes the place of
an operculum, resists all attempts at forcible extraction.

The internal structure of the JBrachyura is, on the whole,
similar to that of the Macrura ^ but the thoracic ganglia have
coalesced to a much greater extent than in Astacits, forming
a single rounded mass. The branchiae are few, never exceed-

THE BRACHYURA. 295

ins: nine on each side, and sometimes not more than seven.
Tiie branchiostegite fits closely down upon the bases of the
four posterior pairs of thoracic limbs, and sometimes incloses
a space which is ver\^ large in proportion to the branchiae.
This is particularly the case in the Land Crabs {Gecarcinus),
where the spacious branchial chamber is lined by a thick and
vascular membrane, which, in these almost wholly terrestrial
Crustacea, either takes on to son^e extent the respiratory
function, or serves to keep the air within the branchial cham-
ber saturated with moisture.

The abdomen in the Brachyura is comparatively small ;
its sixth somite possesses no appendages ; and the others, if
they exist at all, subserve only a sexual purpose, the two an-
terior pairs commonly forming accessory copulatory organs
in the male ; while, in the female, so many of these append-
ages as remain give attachment to the ova, which are carried
about until hatched, between the thorax and the abdomen,
which is bent up against it. The ifdm^X^ Brachyura also pos-
sess a spermatheca attached to each oviduct, which is absent
in the Macrura / and, in this sex, the abdomen is larger and
broader than in the males. In accordance with the rudimen-
tary condition of this part of the bod}^, the abdominal gan-
glia are represented only by a cord, which proceeds from the
posterior part of the great thoracic mass. It is in the con-
struction of their skeleton, however, that the Brachyura
present the most interesting deviations from the Macrura.
Thus, if we select the common Shore-crab, Carcinus mcenas
(Fig. 76), as a tj-pical example of a Brachyuran, we find that
the carapace is a wide shield, broader than long, having a
somewhat pentagonal shape, and bent sharply inward at the
sides, instead of taking an even sweep down to the base of
the legs. It is in such close contact with the four posterior
pairs of thoracic limbs as to leave no passage or aperture
such as exists in Astacus, the only inlet for the water required
for respiration being placed above the basal joints of the che-
late anterior ambulatory limbs. The edges of the carapace pass
completely in front of the basis of the limbs, and then turn sud-
denly forward, parallel with one another and with the axis of
the body, as the pterygostomial plates of Milne-Edwards, to
join the antennary sternum, which is very wide, but short from
before backward. The space included between the edges of
the pterygostomial plates and the antennary sternum is the
*' cadre buccal," or ^eris^ome/ the antennary sternum itself
receives, as in the Astacus, the specific appellation of epi-

296

THE ANATOMY OF IXVERTEBRATED ANIMALS.

Stoma / and the plate which stretches backward and supports
the labrum, within its posterior forked boundary, is the en-
dost oma.

The middle of the dorsal surface of the carapace is marked
somewhat nearer its posterior than its anterior boundary by
a short transverse depression, which is continued on each side
forward and outward, and then curves directly outward to the
edge of the carapace (Fig. 76, cs). Further than this I
cannot trace this homologue of the cervical groove of jLstacus,

Fia. 76.— Of the two nppcr fisrnres*, the left represents the dorsal surface of the cara-
pace of Carcinus mmnas : f, ros?trum; o, orbit; cs, cervical grouve; gri, epigastric
lobe; 9^2 protogaetric ; g»3^ mesogastric ; g^*, hypogastric; g'^ urogastric; c, c\ an-
terior and posterior cardiac ; h, hepatic ; 6', ^>^ b^, epibrancbial, mesobranchial
and mefabranchial lobes. The lower figure represents a ventral view of the an-
terior half of the same carapace: o, rostral septum; 6, antennary sternum ; c,
suture between these ; c?, snpraciliary lobe ; e, internal suborbitar lobe ; /, anten-
na ; g, articular cavity for the ophthalmic peduncle ; h, tlie same for the anten-
nule; o, orbit ; .s,^, subhepatic region; f/?, anterior pleural region. The riglit-
hand upper figure gives a side-view of the carapace of SfeJiorhynchvs pkalanginm,
the common " spider-crab : " o, orbit ; f^f"^, rostrum ; al, auteunule ; at, auteuua ;
ep, epistoma.

Elevations and depressions upon the surface of the carapace
in front of the cervical groove, which, as in Astacus, is com-
posed of the connate terga of the six cephalic somites, mark

THE BRACHYURA. 297

it out into certain definite regions of considerable systematic
importance. An irregular transverse depression, crossing the
carapace near the anterior margin, bounds an anterior or fa-
cial region, divided into a middle J'ro?ital lobe (/), and lateral
orbital lobes (o), from a posterior, much larger, gastro-hepatio
area, divided into small lateral he2Xitic lobes (A), and a large
complex gastric lobe (^\ ^^, etc.). Tlie latter is again sub-
divided into two ejngastric lobes (^'), two protogastric lobes
(^'^), a median onesogastric lobe (^^), two metagastrlc lobes
((/*), and two iirogastric lobes (^^), making altogether nine
subordinate divisions. The gastric lobes correspond in a gen-
eral way to the stomach ; the hepatic lobes, to a portion of
the liver. The region behind the cervical suture consists of
the connate terga of the eight thoracic somites ; it is divided
by two strong longitudinal grooves, the hranchio-cardiac
grooves, into a middle region, corresponding with the heart,
and two lateral regions, forming the roof of the branchial
chamber. A transverse depression divides the middle region
into an anterior and a posterior cardiac lobe, while the bran-
chial region is subdivided into epibranchial (b^)^ mesobran-
chial (6^), and nietabranchial (b^) lobes.

On turning to the inflected inferior portion of the cara-
pace, a sutural line or groove is seen running from the epi-
stoma, outward and backward, very nearly reaching the outer
edge of the carapace, opposite its external angle, and then
sweeping backward parallel with, and but little distant from,
its postero-lateral boundary, until it cuts its posterior edge.
The portion of the carapace internal to this sutural line is
called by Milne-Edwards the inferior branddostegite^ and is
considered by him to be composed of an anterior {ep) and
posterior epimeral piece, corresponding with the subhepatic
(sh) and subbranchial regions of the surface of the carapace
between the suture and the line of inflection. I cannot
regard these parts, however, as having any relation with the
true epimera. The suture, or rather groove, seems rather to
correspond with that which marks off the pleuron from the
rest of the somite in Astacus.

The anterior cephalic somites in Carcinus have under-
gone some singular modifications, whereby their true relations
are greatly obscured. The broad trilobed plate (Fig. 76,/)
corresponds with the elongated rostrum oi Astacus ; inferiorly
it is produced in the median line into a strong ridge or sep-
tum, the lower and posterior edge of which is convex, and
fits closely into the concavity formed by the antennulary and

298 THE ANATOMY OF INVERTEBRATED ANIMALS.

ophtbalmic sterna, as they bend back from the sternal flex-
ure. This rostral septum, therefore, abuts below and behind
on the epistoma, and constitutes a sort of partition (Fig. 76,
«), by which the cavities in which the antennules and eyes of
the two sides are lodged are completely separated from one
another. The lateral portions of the rostrum form a flattened
roof over the inner portions of these cavities, which contain
the bases of the ophthalmic peduncles and the antennules ;
but the outer angles of the rostrum are produced downward
(d), to form the supraciliary lobe. The outer half of the
lateral cavities or chambers is more excavated, and is bound-
ed by a strong pointed process, the external orbitar lobe,
which is divisible into a supraorbital and suborbital portion.
The latter passes gradually into a strong process of the sub-
hepatic region, called the internal suborbitar lobe (Fig. 76,
e) ; this turns forward and upward toward the supraciliary
lobe, which it approaches, but does not meet, the base of the
antenna being, as it were, wedged between the two.

The supraciliary, external orbitar, and internal suborbi-
tar lobes, and the antennae, thus together circumscribe a
cavity widely open in front, which is called the orbit, inas-
much as it lodges the terminal portion of the ophthalmic
peduncles, with the eyes which they support. The proximal
portions of the peduncles pass through the comparatively
narrow opening by which the inner and outer chambers com-
municate, between the antenna and the supraciliary process,
and are inserted as usual into the articular cavities on each
side of the ophthalmic sternum, which is narrow, and hardly
wider than the septum. It thus comes to pass that the eyes,
lodged in their orbits, appear to be altogether external to
the antennules, the enlarged bases of which hide the oph-
thalmic peduncles, and appear to be the sole contents of the
inner division of the subfrontal chamber ; but the true posi-
tion of the eyes is precisely the same as in Astacus, that is
to say, anterior and superior to the antennules. Another
interesting peculiarity about the facial region of the cara-
pace is that the basal joints of the antennae have coalesced
with the sternum of the antennary somite, and, consequent-
ly, that the bases of the antennas are immovable. There is
no vestige of a scaphocerite, and the aperture of the organ
which answers to the green gland of Astaciis is provided
with a peculiar movable plate, provided with a projecting
internal stem, to which delicate muscles are attached in Car-
cinus. It is this structure which has been compared to an

THE BRACHYURA. 299

auditory ossicle ; but, as in Astacus, the auditory sacs are,
in fact, lodged in the dilated basal joint of the antennule.

A cervical fold, lodging the scaphognathite, occupies the
same relative position as in Astacus, and marks off the
cephalic form of the thoracic region, on the sides of the body.
The thoracic sterna gradually increase in breadth, and the
posterior ones are marked externally by a strong median,
longitudinal depression, answering to a corresponding fold
on the inner surface. The apodemal cells are well formed,
but the sternal canal, so largely developed in the 3Iacriira, is
absent in this, as in all other JBrachynra.

The structure of the appendages is essentially the same
as in Astacus, but the third thoracic appendage, or external
maxillipede, has its ischiopodite and meropodite greatly en-
larged, so as to form a broad plate, which, with its fellow,
covers over the other organs, and hence receives the name of
the gnathostegite. The three terminal joints of the limb re-
main small, and constitute a palpiform appendage — the en-
do gnathal palp.

In some of the lower Macrura the thoracic limbs are pro-
vided with a short exopodite, and the posterior maxillipedes
become undistinguishable from the ordinary thoracic limbs.
Such forms lead us naturally to the Schizopoda^ a group the
name of which is derived from the apparent splitting of the
limbs produced by the great development of the exopcdite,
which, in these Crustacea, is as large as the endopodite. In
this group, again, a line can hardly be drawn, in many cases,
between any of the maxillipedes and the thoracic limbs, the
anterior pair only being somewhat smaller than the rest.
Hence Thysanoi^oda is admitted, hj Milne-Edwards, to have
eight pairs of thoracic limbs (" Crustaces," ii. 464). The
branchijB in the Schizopoda are frequently absent ; when
well developed, as in Thysanopoda, they are not included
under the branchiostegite, but hang down freely from the
bases of the thoracic limbs. In 3Iysis, the only represen-
tative of a branchia (if it be one in realitj^) is a process at-
tached to the first thoracic appendage. Cynthia has its
branchial appendages attached to the abdominal members.

In Thysanopoda, 3Iysis, and Cyiithia^ the general struct-
ure of the body is similar to that of the 3Iacrura, except
that, in 3fysis, the greater number of the abdominal append-
ages are rudimentary.

In Leucifer, the antennary somite is produced into a very
long and narrow peduncle, which supports the eyes, on their

300

THE ANATOMY OF IXVERTEBRATED ANIMALS.

great stalks, the antennules, and the antenna?, at its extrem-
ity, separating them from the rest of the cephalo-thorax,
which is covered by a delicate carapace, bent down at the
sides. The anterior thoracic members are rudimentary, and
the posterior pair is absent. The heart is short and rounded,
and situated, as usual, in the thorax.

It has been seen that in Astacus Jluviatilis^ as in Limit'
lus and Daphnia., the embryo slowly and gradually passes
into the form of the adult ; to which it is so similar when it
leaves the ^^^^ that the changes of the young present noth-
ing comparable to the well-known metamorphoses of Butter-
flies and Beetles.

But most Podophthalmia rather resemble the Copepoda
and the majority of the LJntotnostraca, in the fact that the
young, when they leave the Qgg^ have a totally dissimilar
form to that of the parent, and only acquire the adult con-
dition after a series of ecdyses.

The observations of Fritz Miiller ^ have shown that the

^•^v^Mj c -

Fig. n.

-Feneus.—A, ]!faupliits-BtSL<^c. B, Zomn or Copepod sta^e. C, Schlzopod-

stage. (After Miiller.)

young of a species of Prawn (Peneus) undergo a metamor-
phosis which rung parallel with that of the Copepoda. When
it leaves the egg (Fig. 77, A), the young Peneus has an

» " Far Darwin," 1864.

THE DEVELOPMENT OF THE PODOPHTHALMIA. 301

oval, unsegmented body with a single frontal eye, a large
labrum, and three pairs of natatorial appendages — it is in fact,
to all intents and purposes, a Kaiiplius. The Naiqolius-ioTm.
next develops a rounded tergal shield, or carapace ; the first
and second pairs of appendages, remaining, long, become the
antennules and the antennae ; while those of the third pair,
their bases enlarging at the expense of the rest of the append-
age, become the mandibles. Four pairs of appendages sub-
sequently appear behind the mandibles. The hinder three
pairs are bifurcated and become the two pairs of maxillae and
the first and second maxillipedes. Behind these again are
five pairs of short lamellar processes, which eventually are
converted into the rest of the thoracic appendages. The six
somites of the abdomen are long and distinct, and the last
ends in two setose processes. They are at first without ap-
pendages. In this stage (Fig. 77, ^), which answers to the
so-called Zocea-iovm of other Podophthahnia, the principal
locomotive organs are the antennae and antennules, and the
resemblance to an adult Copepod is so striking that it may
be termed the Copepod-&iQ,gQ. Next, the antennae, diminish-
ing in relation to the rest of the body, cease to be the prin-
cipal organs of locomotion, and the rapidly-elongating abdo-
men assumes that function. The stalked double eyes, which
made their appearance in the Copepod-stage, become more
fully developed. The jointed exopodite of the antenna is re-
placed by a single plate. The greatly-enlarged thoracic limbs
are provided with an endopodite and an exopodite, as in the
Schizoj^oda, the branchiae are developed from them, and the
abdominal appendages make their appearance. This may be
termed the Schizojyod-stsige (Fig. 77, C). Lastly, the me-
dian eye vanishes, the exopodite of the locomotive thoracic
limbs disappears, and the larva assumes all the characters of
the adult Peneus.

In the great majority of the Podophthcdmia the embryo
undergoes as remarkable a metamorphosis after it leaves the
Ggg- This fact was first indicated by Siebold, afterward
demonstrated by Vaughan Thompson, whose observations
have been confirmed and extended by many more recent ob-
servers, notably by Spence Bate^ and Claus.^ But the stages
of this metamorphosis differ from those observed in Peneus in

» " On the Development of Decapod Crustacea." (Philosophical Transac-
tions, 1857.)

3 " Zur Kenntniss der Malakostraccnlarven." ("Wlirzburg " Naturwissen-
Bchaftliche Zeitschrift," 1861.)

302 THE ANATOMY OF INVERTEBIJATED ANIMALS.

the apparent absence of the first or Nauplius condition. Pos-
sibly, however, this is represented by a delicate cuticular in-
vestment which the larva throws off soon after leaving the
egg. It then corresponds with the later form of the Copepod
stage of Peneus^ and is termed a Zooea, The Zocea has a
short carapace, often provided in the median frontal and dor-
sal regions with long spine-like prolongations. There is a
median simple eye between the lateral sessile faceted eyes, a
pair of antennules, a pair of antennae, a pair of mandibles,

Fio. 78.— Development of (?ar^mr/s mrenas.—^, Zocea-^tn<re. B, Megalopa-siage. C,

Fiual state. (After Cuucii.)

and tw^o pairs of maxillae ; in short, all the appendages of the
head. Of the appendages of the thorax, the first two pairs
are well developed, and terminate in an exopodite and an en-
dopodite. But behind these, which become the first and the
second pair of maxillipedes, only short rudiments of the six
remaining pairs of thoracic appendages are to be found, and
the somites of the long abdomen have no appendages at all.
Subsequently these make their appearance, the posterior tho-
racic members increase in size, the eyes become raised upon
short peduncles, and the larva resembles one of the lower
Macrura. The carapace next becomes broader, and its spines
shorter, while the ambulatory thoracic limbs take on the
characters of those of the adult, the bifurcated first and sec-
ond pairs becoming metamorphosed into the first and second
maxillipedes. The abdomen becomes relatively short and

THE DEVELOPMEXT OF MYSIS. 303

slendar, and the larva takes on the characters of one of the
Anomura, In this stage it has been named Megalopa, By
further changes in the same direction, the Anomuran con-
dition passes into that of tbe young Brachyuran. All these
modifications of form are accompanied by exuviations of the
chitinous cuticula.

The successive stages are well exemplified by the young
of the Shore-crab, Carcinus moenas (Fig. 78, A^ B^ C). The
larva, on leaving the egg^ has sessile eyes, a long pointed
rostrum, and a spine projecting from the middle of the cara-
pace ; rudimentary antennre, and two pairs of locomotive ap-
pendages— the rudinieats of the anterior maxillipedes. The
abdominal somites are without appendages, and the telson is
broad and bilobed (Fig. 78, A).

This, the Zocecc-stage^ after repeated ecdyses, assumes the
3Iiigalopa form represented in Fig. 78, B, Finally, the car-
apace becomes broader, the abdomen loses its appendages,
and is bent up under the thorax ; the peculiarities of the fa-
cial region, characteristic of the Brachyura^ are developed ;
the antennules and ambulatory members acquire their char-
acteristic proportions ; and the little Brachyuran by degrees
assumes the special peculiarities of Carcinus (Fig. 78, 6').

The development of the Opossum Shrimp {3fysisY is par-
ticularly interesting, as it appears to indicate the relations
between the two modes of development, that with and
that without metamorphosis, which obtain in the Crustacea

The ova consist of a vitelline mass, inclosed within a deli-
cate chorion. The blastoderm appears as an oval patch
upon the surface of the yelk (Fig. 79, A^ c), thickest in the
middle, and here presenting a more or less marked depression
(Fig. 79, JL, B^ c). It is sharply defined from the subjacent
yelk (5), and consists of a finely granular mass, in which mul-
titudes of nuclei, about y^o^ to ^oVo ^^ ^^ inch in diameter,
are imbedded.

The blastoderm next becomes larger at one end than at
the other, and a median sinuation gradually divides this ex-
tremity into two lobes, which will eventually form the ante-
rior parietes of the head, and may be called the procephalic
lobes?'

> Conf. E. van Beneden, " Developpemcnt des Mysis." (" Bulletin de
I'Academie de Bruxelles," 1869.)

2 It is exceedingly interestinor to remark the correspondence hetTVcen the
embryonic structure of the head of Mysis (and I may add that of other Arthro*

304

THE ANATOMY OF INVERTEBRATED ANIMALS.

The median depression becomes more decided, and, at the
end opposite the procephalic lobes, the blastoderm is produced
into a sort of papilla, directed forward. This is the rudiment
of the caudal extremity. From the anterior part of the
blastoderm there arise, on each side, two papillie, the points
of which are directed backward, and which will become the
antennules and antennge. The whole of these parts are in-
vested by a delicate cuticular membrane, w^hich gradually ex-
tends over and invests the whole yelk beneath the vitellary
membrane. At the end of the caudal papilla it forms a broad
process, produced into set^e, which sometimes appear fan-
like, sometimes so deeplj^ bifid as to resemble two styles.

The embryo has now reached what we may term its larval
stage, and, in this condition, it leaves the vitellary membrane
within which it was inclosed, and lies free in the ovigerous
pouch of the parent. At the same time, the caudal extremity
enlarges, and straightens itself out, so that no indication of
its previous inflexion against the thoracic portion of the blas-
toderm remains. The larva thus much resembles a pear
(Fig. 79, D^ JE), with four processes (2, 3), the antennules
and antennsje, which have now become much elongated, on
one surface.

The young 3fysis next grows rapidl}'" and undergoes great
changes in form : but it is a very remarkable fact, that the
primitive integument remains unaltered ; gradually enlarging,
to accommodate itself to the increased size of the foetus, in-

FiG. 79.

poda) and that of tlie head of a vertebrate embryo. The procephalic pro-
cesses resemble in a remarkable manner the trabecules cranii of the vertebrate
embryo ; and the cephalic flexure of the Crustacean or Insect has its analogue,
if not it'=( homologue, in the angle which the trabecular region of the base of
the skiill at first makes with the parachordal region iu almost all V^ertebrata.

THE DEVELOPMENT OF MYSIS.

305

Fig. "i^.— Continued.

i< J,a

Fig. 79.— The development of J/y^l*. — J., Bide view of an egg', in wiiich the blasto-
derm has just appeared. B, side view further advanced. <7. front view of embryo
at the same age, showing the procephalic lobes, here marked b. Z), larva, ventral
view. E, side view. (These two figures have been inverted by the engraver.)
J^, young pupa. G', further advanced. ^, young 3/y*^, which has left its pupa
skin. 7, anterior portion of the same, enlarged, and with the carapace thrown
back, a, vitelline membrane : 6, yelk ; c, central depression of the blastoderm ;
d, procephalic lobes ;/, larval integument; g, its caudal enlargement; A, cara-
pace. 1,2, 3,4, etc., the somites and their appendages, numbered from before
backward.

308 THE ANATOMY CF INVERTEBRATED ANIMALS.

deed, but otherwise taking no share whatever in its changes*
The young 3Iysis might, therefore, in this condition be justly
termed a pupa, for the relation of the primitive integument
to the animal which it incloses is precisely that of the pupa
skin to the imasfo of an insect.

The antennules and antennae remain intact within the
sheaths afforded by the primitive integument, but, becoming
immensely elongated and divided at their extremities, assume
more and more their proper adult conformation.

In front of the antennules, a large rounded protuberance
makes its appearance upon each procephalic lobe, and event-
ually becomes the ophthalmic peduncle. At first, the sternal
portions of the somites, corresponding with these three pairs
of appendages, occupy the same plane with one another and
the posterior sterna (Fig. 79, F, G) ; but, by degrees, they
become bent up (Fig. 79, H)^ and at length the ophthalmic
sternum occupies the upper and front part of the head (Fig.
79, I). In this way the " cephalic flexure " is produced.
The mouth is indicated behind the antennary sternum, which
projects backward in the middle line to form the labrum. On
each side of it the rudiments of the mandibles appear, and
behind these are the papillary commencements of the two
pairs of maxillae. Behind the second pair of maxillse a dis-
tinct constriction indicates the commencement of the thorax,
the appendages of which appear, at first, as tubercular eleva-
tions, all of precisely the same character, and all directed
backward parallel with one another. The abdomen is at first
very small, and the appendages of its sixth somite early ac-
quire a far larger size than the others. The telson is devel-
oped from the middle line above the anus. While all these
changes are going on, the blastoderm gradually extends over
the tergal surface of the embryo and closes it in. When the
carapace is first distinguishable it appears as a ridge arising
from the sides of the posterior thoracic somites, beginning at
the last but one, and gradually extending forward as far as
the antennary somites. The ridge increases and becomes a
fold, which overhangs the bases of the thoracic appendages
(Fig. 79, G) ; and if this fold be turned back (Fig. 79, 1), its
actual attachments may be readily demonstrated.

Having advanced thus far in its development, the foetal
Mysis^ with all its organs fully formed, though somewhat
different in appearance from those of the adult, casts its pupa-
skin and straightens its body, which, from having its pos-
terior portion bent on the anterior, as in the embryo (Fig.

THE GLASS-CRABS OR PHYLLOSOMATA.

S07

79, B)j had gradually in the pupa (Fig. 79, F^ G) assumed
the opposite curvature. Its dimensions are threefold those
of the embryo, and it exhibits vivacious movements when ex-
tracted from the pouch of the parent. It is not improbable
it may yet undergo another change of integument before ac-
quiring the full form of the adult.

Thus it appears that, in 3Iysis^ the iVai</:>//w5-stage (Fig.
79, D, £J) is passed over so rapidly that the embryo has gone
through it at a very early period, and nothing but the cuticu-
lar sheath of the body appertaining to this stage remains to
prove its existence. A step further, in the abbreviation of
the JVaupUus-stage, and there would be nothing to distin-

-cx

Fig. dO.—P/iyllosoma.—A, ventral view of ihe body, -with the limbs I'-XX' of the
left side and the bases only of XP to XIII' represented. J5, side view of the
body. C, the nervous system. D, the last cephalic and first aod second thoi-acic
liuibs.

guish the general course of the development of ]\fi/sis from
that of Astacus. On the other hand, another Schizopod,
MJifphausia, has been shown by Metschnikoff * to leave the
egg as a true N^aupUiis.

» ZeiUchrift J"ur wiss. Zool., 1871.

308 THE ANATOMY OF INVERTEBRaTED ANIMALS.

The Glass-crabs, or Phyllosomata (Fig. 80), are singular
marine pelagic Crustacea, in which the body consists almost
wholly of two large, extremely flat and transparent disks,
devoid of any segmentation. The anterior of these bears
the pedunculated eyes, the antennules and the antennae on
its anterior margin ; while the labrum, w4th the mandibles
and anterior pair of maxillae, form a small projection poste-
riorly on its ventral surface. The second pair of maxillse is
situated a little more backward and outward, and bears a
scaphognathite ; and just behind these appendages is the
fold of a cervical groove which separates the anterior disk
from the posterior. The anterior disk contains the stomach
and the liver, and in this respect, as in its appendages, cor-
responds exactly with the cephalostegite of the carapace of
an ordinary Crustacean, and its six cephalic sterna. The pos-
terior disk, on the other hand, contains the short and almost
round heart, with the intestine, and bears the eight pairs of
thoracic appendages, the anterior and posterior of which are
not uncommonly rudimentary. The abdomen is usually very
small, and situated in a notch at the posterior edge of the
thoracic disk. It is provided with six pairs of appendages.
No generative organs have been found in the Phyllosomata^
and there is reason to believe that they are merely larvae of
the Macruran genera Palinurus^ jSeyllarus, T/ienus, and their
allies.

The Cumacea. — These are very remarkable forms, allied to
the Schizopoda and JVebalia, on the one hand, and on the
other to the EdriopJithalmia and Copepoda y while they ap-
pear, in many respects, to represent persistent larvae of the
higher Crustacea.

Cuma Katlikii might, at first, be readily mistaken for a
Copepod. It possesses a comparatively small, thick carapace,
apparently produced into a rostrum anteriorly, and succeeded
by a series of twelve gradually narrowing free segments, the
appendages of which are in great part obsolete. The last of
these segments is a pointed telson ; the anterior five, belong-
ing to the thorax, bear thoracic limbs, while the eleventh, the
last true somite of the body, carries its characteristic styli-
form appendages. The appendages of the preceding abdom-
inal somites may be either absent or very small and rudimen-
tary. Dohrn has proved that this is true only of the females
among the Cumacea. The males, which were formerly re-
ferred to the genera Bodotria and Alauna^ often have well-

THE CUMACEA. 309

developed abdominal limbs, t])ouo-h they appear late. It is
interesting to find that the females, in this respect, retain
more of the larval character than the males.

On examining the apparent rostrum with care, it is found
to be divided along the middle line by a fissure which runs in
front of the eye (wiiich is here single and sessile), divides into
two branches, which run backward and outward, and termi-
nate before traversing half the length of the carapace ; they
thus cut off a median lobe, bearing the eye at its apex, from
two lateral processes. The lateral processes are simply pro-
longations of the antero-lateral regions of the posterior di-
vision of the carapace (as it were the antero-lateral angles of
the carapace of 3lysis^ excessively produced and meeting in
the middle line) ; while the middle lobe corresponds, I believe,
with the cephalostegite of the carapace in ordinary Podoxjh-
thalmia^ the insertions of the m.andibular muscles occupying
their normal position, toward its posterior boundary. The
hinder part of the carapace will therefore correspond with the
terga of the three anterior thoracic somites, the five posterior
ones being, as has been seen, free and movable.

The five anterior pairs of thoracic appendages are con-
structed much on the same plan as those of the Schizopoda /
the three posterior have no exopodite. In the female, the
sixth abdominal somite alone has appendages, but in the male
the two anterior abdominal somites are provided with styles.
Ovigerous plates are attached to the fourth, fifth, and sixth tho-
racic appendages in the female. The structure of the head is
peculiar. No ophthalmic sternum nor ophthalmic peduncles
are discernible, the single, or closely approximated two, eyes
being sessile on the median line on the superior surface of
the head. The coxopodites and basipodites of the antennules
and antennae are bent down almost at right angles with the
axis of the body, and appear to be connate, or confluent, with
their sterna. The succeeding joints are free and pass for-
ward, the antennules being much longer and stronger than
the antennae in the females, while in the males the antennre
are ver}' long; the labrum is large ; the mandibles strong and
unprovided with a palp. There is a distinct metastoraa, and
the maxillae are delicate and foliaceous. A papillose bran-
chial plate is attached to the base of the first thoracic append-
age. The surface of many parts of the body in some species
exhibits a very peculiar sculpturing, singularly like that ex-
hibited by the Exirypterida.

As in the Podophthalmia, the heart is short or mod-

310 THE ANATOMY OF INVERTEBRATED AXIMALS.

erately elongated, and situated in the posterior part of the
thorax.

Dohrn^ has shown that the development of the Cumaeea
takes place without metamorphosis. In most respects the
embrj^o resembles that of 3Iysis y but, instead of the cuticu-
lar investment of the transitory N^auplius-si-dgQ with its two
pairs of appendages, there is only a sort of cuticular sac with
a thickening in the middle line of the tergal aspect, which the
embryo bursts as it acquires a larger size. In this respect,
the resemblance of the embryonic development of the Cuma-
eea to that of the JEdriophthalmia is, as Dohrn points out,
very striking, and no doubt they form a connecting-link be-
tween the Podophthalmia and the JEdriop)1ithalmia, Having
regard to their whole organization, on the other hand, they
stand at the bottom of the Malacostracan group, and are com-
parable to a jP(gnew5-larva in the Copepod stage, the limbs and
body of which are modified in the direction of the Schizopoda^
while the fore-part of the head has remained Copepodous.

Fossil Brachyura are abundant in tertiary deposits, but
are rare in formations of earlier date. Macrura of a pecul-
iar type (Eryoii) occur in the mesozoic rocks, and perhaps
the carboniferous Gampsonyx should be referred to the Po-
dophthahnia.

The Edkiophthalmia. — These resemble the Podophthal-
mia in never possessing a greater than the typical number (20)
of somites, though, in some members of the group, the body
is composed of fewer somites, in consequence of the abortive
or rudimentary condition of the abdomen. Eyes may be
absent ; when present, they are usually simple, and are either
sessile or seated upon immovable peduncles (Jliaina). The
antennules almost disappear in the terrestrial Psopoda, while
the antenniB become rudimentary or vanish in some Atn-
phipoda. The mandibles lose their palps in the Woodlice ;
which thus, as in the presence of only one pair of well-devel-
oped antennary organs, approach Insects. Ordinarily, the
posterior seven, and, at fewest, the posterior four, thoracic
somites are perfectly distinct from, and freely movable upon,
one another. The oplithalmic and antennary somites have
coalesced with the rest of the head ; the branchiae depend
from the thoracic limbs, or are modifications of the abdomi-
nal appendages ; and the heart is elongated and many-cham-

1 " Ueber den Ban und die Entwickfdiin!? der Cnmaceen." (" Untersuchun-
gen iiber Bau und Entwickelung der Arthropoden," 1870.)

THE EDRIOPHTHALMIA.

311

bered. But the salient characters of the group will be best
understood by the study of such a genus as Amphitho'e^ the
principal details of the organization of which are represented
in Fig. 81.

The body of this animal is compressed, bent upon itself,
and divided into fifteen very distinct segments, reckoning the
head as the first and the telson as the last.

Fig. ^X.—Amphithoe.—T'hQ letters and fisrures have the same signification as in other
fi'jrnres of Crustacea, except os, oostegite; br, bianchise; C lateral view of
stomach (Z>) opened from above ; a, 6, c, different parts of tlie armature.

The head presents a rounded tergal surface ; the anterior
face is disposed perpendicularly to the axis of the body, and

312 THE ANATOMY OF INVERTEBRATED ANIMALS.

is produced anteriorly into a strong", curved, and pointed ros-
trum ; on each side it bears an aggregation of simple eyes,
and in front, immediately beneath the rostrum, this face gives
attachment to two long, many-jointed antennules. Below
these, two antennae, shorter, and fewer-jointed than the an-
tennules, are inserted, and the inferior part of the face is
completed by a large movable labrum. Behind this come
the strong, toothed palpigerous mandibles (IV), and two
pairs of more or less foliaceous maxillae. Inasmuch as the
eyes are sessile, these five pairs of appendages are all that
belong to the head proper ; but, just as in the Podophthal-
mia, certain of the anterior thoracic appendages are con-
verted into accessory gnathites, so, in A.mphithoe, the first
pair of these members are applied against the mouth, and
form a large lower lip (VII').

The "head" of AmpJiithoe^ therefore, is formed by the
coalescence of the seven anterior somites of the body, but I
believe that the tergum of the seventh (or first thoracic) so-
mite is obsolete, as in a Stomatopod, and hence that the ter-
gal surface of the head of the JEdriophthahnia corresponds
exactly with the cephalostegite (or that part of the carapace
which lies in front of the cervical groove) in Podoplithalmia.
Mr. Spence Bate has shown, in his valuable " Report on the
Elriophthalmia^'^ that in the Crustacea at present under
discussion, a strong apodeme arises on each side from the
posterior part of the sternal region of the head, and passing
inward and forward meets wilh its fellow to form an endo-
phragmal arcli, which supports the oesophagus and stomach,
and protects the nervous commissure between the first and
second sub-oesophageal ganglia, which runs under it.

The discoverer of this structure conceives that it repre-
sents the terga of the three somites immediately succeeding
the mouth ; but I cannot see that it is other than the repre-
sentative of the precisely similar mesophragm formed by the
anterior apodemes in Astacus. In fact, the correspondence
in structure between the head of an Amphithoe and the ceph-
alic portion of the cephalo-thorax of Astacus is not a little
striking. There is the same sternal flexure, the same relative
position of the stomach, and of the insertions of the mandibu-
lar muscles. The great difiference lies in the abortive condi-
tion of the ophtlialmic appendages.^

* A stron:^ endophraemal arch separates the sub-oesophageal ganprlia and com-
missures from the ijuUet in Squilla, but has different connections (Fig. 83). ^ A
very similar endophragmal arch is found in the Insect head. See the descrip-
tion of the head of Blatta {infra).

THE EDRIOPHTHALMIA. 313

The seven free somites of the thorax each give attachment
to a pair of limbs. It is characteristic of Amphithoe^ as of the
Amphipoda in general, to have the five anterior pairs of tho-
racic members directed forward. Each limb consists of an
expanded coxopodite, succeeded by the other six joints of the
typical crustacean limb.

In the male, a single vesicular lamella, the branchia, is
attached to the inner side of the coxopodite of the append-
ages of the ninth to the fourteenth somites inclusively ; but,
in the female, an additional plate, convex externally and con-
cave internally, is attached above, and internal to, the branchia
of the 9th to the 12th somite. These oostegites, as they may
be called, inclose a cavity in which the incubation of the eggs
takes place.

The abdomen consists of six somites and a very small ter-
minal telson. The appendages of the three anterior somites
are terminated by two multiarticulate setose filaments (Fig.
81, XV'), while in the three posterior the corresponding
parts are styliform, and serve as a fulcrum for the abdomen
when the animal leaps, by the sudden extension of that region
of the body.

The Edrioplithahnia are ordinarily divided into three
groups. The Amphipoda^ which resemble Amphithoe, are
characterized by their compressed form and their ordinarily
saltatory habits ; by having thoracic branchiae ; by the for-
ward direction of their four anterior locomotive limbs (2d
to 5th pairs of thoracic appendages), and by the contrast
between the three anterior and the three posterior pairs of
abdominal appendages. The common Sand-hopper is the most
familiar example of this division. The second group is that
of the LcBmodipoda^ distinguished by the rudimentary state
of the abdomen, which is reduced to a mere papilla, and by
the coalescence of the second, as well as the first, thoracic
somite with the head, so that the anterior limbs appear to be,
as it were, suspended under the neck. The strangely-formed
genera Cyamus^ the parasite of whales, and Caprella^ which
is very common upon our own coast, adhering to corallines,
sea-weeds, and starfish, belong to this group.

The Isopoda^ which constitute the third group of the
JEdriophthahnia^ are usually depressed instead of compressed,
and run or crawl instead of leaping. Many, like tlie common
Woodlouse ( Oniscus)^ possess the power of rolling them-
selves into a ball when alarmed ; some, like the last-named
genus, are terrestrial ; others, like the Asellus^ inhabit fresh

314 THE ANATOMY OF INVERTEBRATED ANIMALS.

waters, but the great majority are marine ; and among them
are many peculiarly modified parasitic forms (Fig. 82, Cymo-
thoa ; Bopyrus). The composition of the head and mouth

-^xx>

■vi'^V

Fio. 82. — Cymothoa. — The letters ar.d fisnres have the same signification as in Fig,
81, except Ab, abdominal appendages in Fig. A.

in the Isopoda is essentially the same as in the Amphipoda,
though differing considerably in details. The branchiae of
the thoracic members are absent, their functions being per-
formed by the endopodites of some of the abdominal mem-
bers, which are soft and vascular. The three anterior pairs
of thoracic members are usually directed forward — the four
posterior pairs backward. In some Isopoda the abdominal
somites, partly or wholly, coalesce Avith one another.

In all the Edriophtlialmia the alimentary canal is straight
and simple, and its anterior gastric dilatation, frequently
strongly armed, is situated in the head. The liver is repre-
sented by a variable number of straight caeca.

Occasionally there are one or two caeca which open into

THE EDRIOPHTHALMIA. 315

the posterior part of the intestine, and appear to be urinary
organs analogous to the Malpighian cagca of insects.

The respiratory organs vary greatly in structure. In most
Edriophthalmia they are simple plates or sacs, the delicacy
of the integument of which permits of the free exposure of
the blood circulating in them to the air. In the amphipod
genus Phrosina, however, the branchiae are composed of rudi-
mentary lamellce, attached to an expanded stem, and resem-
ble not a little the epipoditic branchiae of Astacus. In some
Sphceroinidm^ Duvernoy and Lereboullet found the branchial
endopodites transversely folded, so as to approach those of
the Xiphosura.

The exopodites of the abdominal members of the Isopoda
frequently cover the modified endopodites, forming opercula,
and the first pair of abdominal limbs is, in many genera, al-
tered in such a manner as to form one such large operculum
for the four pairs which succeed it. In the Idoteidce it is, on
the other hand, the sixth pair of abdominal limbs which are so
modified as to form the curious door-like opercula which cover
the gills.

In certain of the terrestrial Isopoda {Porcellio^ Arma-
dllUdium), some of the opercular plates of the branchiae,
usually the two anterior pairs, contain curiously ramified cav-
ities, which 0238n externally, and contain air. The genus
Tylos possesses respiratory organs, which present a still more
interesting approximation to those of the purely air-breath-
ing Articulata. They are thus described by Milne-Edwards :

" The abdomen presents inferiorly a deep cavity, very
similar to that of the Sphderomm^ in which the five anterior
pairs of appendages are lodged ; but this cavity, instead of
being completely open below, is imperfectly closed, in its pos-
terior half, by two series of lamellar prolongations, which
arise from the sides of the inferior faces of the third, fourth,
and fifth abdominal segments, and pass horizontally inward ;
the first pair of these plates is small, those of the third pair
are, on the other hand, very wide, and almost meet in the me-
dian line. The four anterior pairs of abdominal appendages,
lodged in this cavity, each carry a wide and short quadrilat-
eral appendage, the surface of which is raised into a transverse
series of large longitudinal elevations, and each of these eleva-
tions presents inferiorly a linear aperture leading to a respir-
atory vesicle, the parietes of which are covered with a multi-
tude of little arborescent c£eca. These vesicles when extracted
from the interior of the limb closely resemble a brush-like

316 THE ANATOMY OF INVERTEBRATED ANIMALS.

branchia, having its longitudinal canal in communication with
the atmosphere by a longitudinal stigma. The fifth pair of
abdominal members are rudimentary, while the sixth consti-
tute the door-like triangular valves covering the anus, and all
the inferior face of the last abdominal seo:ment." ^

The nervous system in the Amphipodoj consists of supra-
oesophageal or cerebral ganglia, united by commissures with
an iufra-oesophageal mass, w hence commissural cords pass un-
der the endophragm to the anterior of the thoracic ganglia, of
which there are commonly seven pairs, succeeded by five or six
pairs of abdominal ganglia. In some Isopoda ( Cymothoa,
Idotea) the abdominal ganglia are also distinct; but in others,
such as ^ga hicarinata (according to Rathke), they are
fused into a single mass placed in the anterior part of the
abdomen, presenting only traces of a division into five por-
tions. In the Cymothoadoe and terrestrial Isopoda^ again, the
abdominal ganglia appear to have completely coalesced with
the last thoracic ganglia and form a mass, whence the ab-
dominal nerves radiate. Finally, in the short-bodied Lmmo-
dipoda^ such as Cyamus, there are not more than eight pairs
of post-cesophageal ganglia, the posterior commissures of
which are so shortened that the nervous system ends in the
antepenultimate somite.

Brandt describes splanchnic ganglia like the lateral pair of
Insects in the OniscidcB. It is one of the many respects in
which the Isopoda simulate Insecta.

No other organs of sense than eyes have, as yet, been cer-
tainly demonstrated to exist in the Edriophthcdmia^ though
the fine setae which beset the antennary appendaofes have
been supposed to be organs of the olfactory sense. The eyes
vary in their structure, from the simple, more or less closely
aggregated ocelli of JLmmodipoda^ and of many Isopoda and
Amphipoda^ to the strictly compound eyes, as complex as
those of the highest Articidata, which exist in .jUga and in
Phrosina.

The female genitalia of the Edriophthahnia consist of two
simple sacs, the ducts of w^hich usually open on the ventral
surface of the antepenultimate thoracic somite, or on the bases
of the limbs of this somite. In the male, one or more caeca
on each side constitute the testis, which ordinarily opens on
the last thoracic or first abdominal somite, in connection with
one or two pairs of copulatory organs developed from the an-
terior abdominal somites.

* " HIstoire Naturello des Crust^ces," vol. iii., p. 187.

THE STOMATOPODA. 317

The eggs of the ordinary Edriophthalmia usually undergo
their development in the chamber beneath the thorax inclosed
by the oostegites of the thoracic appendages. In most cases,
the young diifer so little from the adults that no metamorpho-
sis can be said to take place. They frequentW, however,
want the last thoracic somite. The young of the parasitic
Edriophthalmia^ such as J^opyrus, Phryxus^ Cymothoa, Cy-
amus, and the Hyperinoe^ on the other hand, are widely dif-
ferent from the adults ; and not only in their metamorphosis,
but in the small proportional size and less aberrant form of
the male, Bopyriis and Fhryxus recall the parasitic Cope-
poda.

In certain Amphipods [Gaynmanis lociista and Desmo-
philus) the vitellus undergoes complete division ; while, in
closely allied forms {Gammarus Jfuviatilis and 2nd€x)^ and
still more completely in those Isopoda which have been
studied, the part of the vitellus which divides into blasto-
meres becomes more or less completely separated from the
rest immediately after fecundaticn, and the so-called partial
yelk division, take place.^

In all Edriophthalmia^ the development of which has
been examined, before any other organs appear, a cuticular
investmxnt or sac is formed, which is eventually burst and
thrown off. This appears to represent the Kauplius cuticle
of Mysis^ and, in close relation with it, are peculiar tergal
structures, such as the bifid lamellte of Asellus^ and the un-
fortunately named "micrcpyle apparatus" of other Edrioph-
thalmia.

The Edriophthalmia are not abundant in the fossil state ;
but thev may be traced back as far as the later Palaeozoic
strata {Prosop>oniscus, Amphipdtis),

The Stomatopoda. — Of the Stomatopoda of Milne-Ed-
wards, two of the three divisions, the Carido'ides and the
Eicuirasses^ have since found a place among the Schizopod-
ous Pod ophthalmia^ or among the larva? of certain Macru-
ra ; but the third, the Stomatopodes vnicuirass^s, compris-
ing Squilla, Goi\odactylus, and Coronis^ appear to me to differ
so widely and in such important structural peculiarities, not
only from the Podophthalmia proper, but from all other
Crustacea, as to require arrangement in a separate group,
for which the title of Stomatopoda may well be retained.

» E. van Bencden, " Eecherclies sur la Composition et la Signification de
I'GEuf," 1870.

318

THE ANATOMY OF INVERTEBRATED ANIMALS.

The genera named, in fact, stand alone among the CruS'
tacea^ in that the ophthalmic and antennulary somites are
complete rings, movable upon one another and the anten-

-Tir

33r

Fig. 83. — SquU'a scabricauda. — A, the entire body, with the thorax and abdomen in
lonsitudinal and vertical section. B. the head in vertical section. I-XX, somites
of which the body is composed. F-XX', their appendaj^es, the bases of most of
which are alone represented, ^i, alimentary canal ; (?, stomach ; ^n, anus; (7,
heart; br, branchia. N, erancrlia and their commissures. R^ rostrum of the cara-
pace ; j9, the penis. Pw, endopbragmal arch. The fifth thoracic appendage XI' is
figured separately.

nary somite, and that their long axis is parallel with that of
the body, so that there is no sternal flexure. Numerous pairs
of hepatic caeca open into the elongated alimentary canal.
The heart, again, is not short and broad, with at most three
pairs of apertures, and confined to the thoracic region, as in
the proper Podophthalmia / but it is greatly elongated, mul-
tilocular, and extends into the abdomen. The branchiae are
pkimes attached to the abdominal members (Fig. 83, A^ hr),
and, so far as I have been able to ascertain, the carapace is,
in all, connected exclusively with the cephalic somites. This

* Unless the freedom of the anterior segment of the head in the Pontdlidm
referred to above, when the Copepoda were under consideration, is a parallel
case.

THE STOMATOPODA. 319

is particularly well seen to be the case in Sqiiilla scahricauda
(Fig. 83), where five completely developed posterior thoracic
terga can be counted, uncovered by the short carapace, be-
neath which the tero-a of the three anterior thoracic somites
are represented by a membrane which passes forward to be
reflected into the carapace.

The free somites of the thorax, and those of the abdomen,
in this species and in the Stomatopoda generally, are so large
relatively to the carapace, that the latter is not larger in pro-
portion to the body than the tergal covering of the head in
many Edriophthalmia, with which order the Stomatopoda
present many marked affinities. Indeed, if we leave the eyes
out of consideration, the organization of the Stomatopoda is
more Edriophthalmian (and especially Amphipodan) than
Podophthalmian. The five anterior pairs of thoracic members
are turned forward, and are subchelate. The first pair are
small and slender. The second pair are the largest of all,
and have the characters of powerful prehensile limbs, the ter-
minal curved and spinose joint of w^hich shuts down into a
groove in the penultimate joint, as the blade of a pocket-knife
does into its handle. The three posterior thoracic limbs, on
the other hand, are turned outward, and terminated by an
endopodite and an exopodite.

Squilla lays its eggs in burrows in the bottom of the sea,
which the animals inhabit. The earliest condition of the free
larva is not fully known, but the young larvae have a single
ej'e, and the hinder thoracic and the abdominal appendages
are not developed.^ The larvae pass into forms which, under
the names of Alima^ Erichthys^ and Squillerichthys, were
formerly considered to be independent genera. Claus's inves-
tigations, however, have rendered it probable that the two
latter genera are simply larval stages of Gonodactylus^ and
that Alima is a larval stage of Squilla,

1 Fritz Mijller, " Fur Darwin." See also Claus, " Die Metamorphose der
Squiliiden," 1872.

CHAPTER VII.

THE AIK-BKEATHING ARTHEOPODA.

Among these Arthropoda^ no forms absolutely devoid of
limbs are at present known, though the appendages are re-
duced to two pairs of minute hooks in the vermiform parasite
Linguatula.

The Araciinida have pediform gnathites, and the least
modified forms of this group (the Arthrogastra or Scorpions
and Pseudo-scorpions) exhibit, in many respects, extraordi-
narily close resemblances to the Merostomata among the
Crustacea.

The Aetheogastea. — The anterior part of the body of
a Scorpion (Fig. 84) presents a broad, shield-like tergal plate,
resembling that of Eurypteriis in form. Two large eyes are
situated one on each side of the middle line of the shield,
while smaller eyes, which var}^ in number according to the
species, are ranged along its antero-lateral margins.

Six wide plates, representing the terga of as many so-
mites, follow the anterior shield, and are connected only by
the soft integument of the sides of the body with their
sterna. The seventh is united with its sternum (xv) poste-
riorly, while the five following terga and sterna form contin-
uous rings, which constitute the joints of the so-called " tail."
The anus is situated behind the last sternum. A movable
terminal piece, answering to the telson of a Crustacean, which
is swollen at its base, and then rapidly narrows to a curved
and pointed free end, overhangs the anus, and constitutes the
characteristic weapon of offense of the Scorpion. This sting,
in fact, contains two glands which secrete a poisonous fluid,
and their ducts convey it to the minute aperture situated at
the sharp point of the organ. On the sternal surface of
the body there are four wide and long sternal plates (xi-xiv),

THE ARTHROGASTRA.

321

which correspond with the third, fourth, fifth, and sixth, of
the free terga. Each of these bears a pair of oblique slits,

Fig. Bi.— Scorpio afer.—A, teraal, and B, sternal, view of tlie body ; At, chelicerae ;
iv', pedipalpi ; v', vi', posterior pair of cephalic appeiidaires ; vii', viii', anterior
thoracic limbs; JY, pectiues ; St, etigma ; Ctk, cephalo-thorax. (After Milne-
Edwards and Diiges.^)

which are the openings of the respiratory organs (Fig. 85, e).
The sterna of the first and second free somites (ix, x) are very
small ; that of the first carries the valves which cover the
genital aperture ; that of the second bears a pair of very
curious appendages, somewhat like combs, which are termed
the pectlnes. The nervous trunks which enter the pectines
are distributed to the numerous papillae which cover them,
and are probably tactile in function. Thus there are twelve
somites behind the eye-bearing shield, and none of these are
provided with appendages, unless the pectines be such.

The truncated anterior extremity of the body, beneath the

I "

Eegne Animal," Illustrated Edition.

322

THE AXATOMY OF IXYERTEBRATED ANIMALS.

shield, is formed by a very large setose labrum, behind and
below which, in the middle line, is the extraordinarily minute

Fig. 85.— a diagram of the body of a Scorpion, the majority of the appendages be-
ing removerl : a, the mouth ; 6, the alimentary canal ; c, the anus ; d, the heart ;
e, a pulmonary sac ; /, the positi-in of the ventral gani^lionated cord ; g, the cere-
bral canerlia;"?', the telson. VIT-XX, the seventh to the twentieth somite. IV,
V, VI, the basal joints of the pedipalpi, aud two following pairs of limbs.

aperture of the mouth (Fig. 86, 31). On each side of it is
attached a three-jointed, pincer-ended, appendage, the che-
licera. Behind these follow the pedipalpi^ large chelate limbs,
the stout basal joints (iv') of which lie on each side of the
mouth.

The following four pairs of appendages are seven-jointed
ambulatory limbs, each terminated by three claws. The ba-

THE ARTHROGASTRA.

323

sal joints of the first two (v', Yi') lie behind the mouth, the
posterior and inferior boundary of which they form, and are
directed forward. The basal joints of the last two (vn', Tin'),
on the other hand, directed inward, are firmly united together,
and are altogether excluded from the mouth.

Thus the mouth is situated between the labrum in front,
the bases of the pedipalpi and those of the first two pairs of
ambulatory limbs, at the sides and behind ; just as, in Limiilus,
the mouth lies between the labrum and the basal joints of the
third, fourth, and fifth limbs, which answer to the mandibles
and first and second maxillae of the higher Crustacea. If this
comparison is just, there is one pair of pra-oral appendages,
which exist in Limuhis, wanting in the Scorpion ; and the
difi"erence between the two maj' be represented thus :

Limulus. Antennule. Antenna. Mandible. Maxilla 1. Maxilla 2.

ScoT'pio, Chelicera. Pedipalpus. Leg 1. Leg 2.

Again, if the eye-bearing part of the head may be regarded
as a somite, then the body of the Scorpion, like that of a mala-
costracous crustacean, -will consist of twenty somites and a
telson. We may regard the six posterior somites (xv-xx) as
the homologues of those w^iich constitute the abdomen in the
crustacean ; while the eight middle somites (yii-xiy) will

j>r 2^

Fig. 86.— 5^rorp?o.— Vertical f?ection of thp cephalo-thorax: At, chelicera ; », labmm,
Jf, month; a. pharyngeal sac; iV, K', supra and infra-oesophageal gangJia ; 0,
cesophaiTue ; <?, opening of the salivary ducts ; c, intestine ; E, heart.

answer to those which enter into the thorax of the latter ;
and the head will resemble that of an Edriophthalmian with

324 THE AXATOMY OF IXVERTEBRATED ANIMALS.

one pair of antennary organs completely suppressed. Upon
this view, the eye-bearing shield is a carapace covering a
csphalo-thorax, into which the two anterior thoracic somites
only enter. These are followed by six free thoracic somites,
the four posterior of which are pulmoniferous. But no trace
of the supposed missing antennary appendage has been met
with in the embryonic condition, so that the alternative pos-
sibility that the mouth is situated one somite farther forward
in the Scorpion than in the Crustacean must be borne in mind.
It is a very interesting fact that Metschnikoff ^ has found ru-
diments of limbs on those somites of the embryo Scorpion on
which the stifj-mata are situated — a circumstance which suor-
gests the suspicion that the Scorpion is derived from some
form possessing more numerous limbs.

The minute oral aperture leads into a small pyriform lat-
erally-compressed sac (Fig. 86, a) with chitinous elastic walls.
Muscles pass from these to apodemes of the sternal wall of
the head, and doubtless act as divaricators of the wall of the
sac. As the Scorpion sucks out the juices of its prey, it is
probable that the elastic sac acts as a kind of buccal pump —
the nutritious fluid rushing in when the sides of the pump are
separated, and being squeezed into the oesophagus when the
elasticity of the walls brings them back to their first position.^

The oesophagus (Fig. 86, b) is an exceedingly narrow tube,
which springs from the tergal and posterior aspect of the sac
just mentioned, traverses the nervous ring, and then, passing
obliquely upward and backward, enlarges into a dilatation
which receives the secretion of two large salivary glands, by
a wide duct on eacli side. The alimentary canal narrows
again, and, becoming a delicate cylindrical tube which widens
posteriorly, passes straight through the body to the anus.
The numerous ducts of the liver open into the anterior part
of this region of the alimentary canal, and it receives two
delicate Malpighian tubuli.

The liver is a vast follicular gland, which occupies all the
intervals left between the other organs in the enlarged part
of the bodv, and even extends for some distance into the nar-
row posterior somites.

The eight-chambered heart (Fig. 86, j^) is a larger and
more conspicuous structure than the alimentary canal, above
which it lies, in a pericardial sinus situated in the middle

1 " Embryologie des Scorpions." {ZeiUchrift fhr wins. Zoologie, 1871.)

2 Huxley, " On the Mouth of the Scorpion." {Quarterly Journal of Micro^
scopical Science, 1860.)

THE ARTHROGASTRA.

325

line of the tergal aspect, between the eye-bearing shield and
the tail ; each chamber is wider behind and narrower in front,
and has two valvular apertures, by which blood is admitted
from the pericardial sinus at its postero-lateral angles. It
gives off small lateral arteries, and ends in front and behind
in a wide aortic trunk. Of these the anterior is larger than
the oesophagus, and both aortae give off branches which are
distributed widely through the body. A large trunk lies on
the tergal aspect of the ganglionic chain, and is united with
the anterior dorsal aorta, by a lateral aortic arch, on each side
of the body. The veins, on the other hand, are irregular pas-
sages, the blood of which is carried to two afferent pulmonary
sinuses, one for each set of respiratory organs.

These respiratory organs are four pairs of flattened sacs,
which open externally by as many stigmata, on the sterna of
the four posterior free thoracic somites (Fig. 85, xi-xiv) in

Fig. 87.—^, palmonary sac. 5, respiralory leaflets of Scorpio occitanus. (After

Blauchard.)

front of the tail. Each lies with one flat side sternal and the
other tergal, in front of its stigma, and its walls are so folded
as to divide its cavity into a multitude of subdivisions, each
of which opens into the common chamber which communi-
cates with the exterior by the stigma (Fig. 87). The organ,
in fact, somewhat resembles Vi porte-monnaie v^'ith. many pock-
ets. The blood circulates in the folds, and, after being thus
exposed to the influence of the air, is carried by efferent pul-
monary sinnses to the pericardial sinus. Expiration is effect-
ed by muscles which pass vertically between the sterna and
terga of the free somites.

The bilobed cerebral ganglion supplies nerves to the eyes
and chelicerce, and is connected by thick commissures with
the post-oesophageal ganglion, a larcre oval mass, whence
branches are given to the maxillae and following somites. A
long cord formed by two closely-applied commissures passes

O-

26 THE ANATOMY OF INVERTEBRATED ANIMALS.

to the three ganglia placed in the twelfth to the fourteenth
somites. There are four ganglia in the abdomen, two dis-
tinct cords passing from the last to its extremity. The vis-
ceral nervous system is represented by an oesophageal gan-
glion receiving roots from the cerebral ganglion, and giving
branches to the alimentary cacal.^

Two lateral ovarian tubes, connected by transverse anasto-
moses with a median tube, end in two oviducts, which open
by a fusiform vagina on the first free sternum (ix). The tu-
bular testes end in a pair of deferent ducts, on which, before
their union at the common orifice, two long and two short
cjBca are found, the former playing the part of vesiculae semi-
nales. Both male and female organs lie imbedded in the
hepatic mass in the posterior thoracic region, their ducts pass-
ing forward. Partial yelk-division takes place, and the ova
undergo development within the ovarian canals, in a manner
which is very similar to that of Astacus. Thus there is no
metamorphosis, and the young difi"er but little from the adult
in any respect but size.

The Pseudo-scorpions ( CheUfer, Obisium) resemble the
Scorpions in form and in the nature of their appendages, but
they have no aculeate telson nor poison-gland. They possess
silk-glands, which open close to the genital aperture, and their
two pairs of stigmata are connected, not with pulmonary sacs,
but with tracheal tubes. According to MetschnikoflP, the eggs
undergo complete yelk-division, and the young leave the egg
provided only with that pair of appendages, which become the
pedipalpi.

In the number of the appendages, and in the segmenta-
tion of the abdomen, Galeodes (or Solpiiga) agrees with the
Scorpions and Pseudo-scorpions. But the three somites which
bear the three hinder pairs of ambulatory limbs (yi, vii, viii,
in the Scorpion) retain their distinctness, and there is no
cephalo-thorax, in the proper sense of the word. In form and
function the pedipalps resemble the first pair of ambulatory
limbs, while the cheliceras are subch elate. The organs of res-
piration are tracheal.

The Phalangidoe (Phalangimn^ Gonyleptus) have chelate
chelicerae, but the pedipalps are filiform or limb-like, and the
articulated abdomen is relatively short and broad. They have
no silk-glands, and their respiratory organs are tracheal.

in
1843.)

' Newport, " On the Structure, etc., of the Nervous and Circulatory Systems
Myriapoda and Macrurous Araclinida." (" Philosophical Transactions,"

THE ARANEINA.

327

While tlie last-mentioned forms lead from the Arthrogastra
to the Acarina, the pulmonate FhrynidcB, or Scorpion-spiders
(ThelyphonuSy Fhrynus), are in many respects intermediate
between the Arthrogastra and the Araiieina.

-J°5s»

fjzzr.

is:/^

Fig. ^. — Mygale, C'e7nentar'ia.—A, female of the natnral size : At, chelicei'sie ; rv',
pedipalpi ; V, vi', maxillary feet; vii', viii', thoracic feet ; Cth, cephalo-thorax.
B, the last joint of the pedipalpus of the male much majmified. C, terminal joint
of the chelicera At, with the poison-gland. D, the left pulmonary sac viewed from
its dorsal aspect: Stg, stigma; Pw, pulmonary lamellae. E, the two arachnidial
mammillae of the left side— the smaller Sp 1 is situated on the base of the large one,
Sp 2. (After Diig^s, " Regne Animal.'")

The Araxetna. — The Spiders stand in somewhat the same
relation to the Scorpions as the brachyurous to the macrurous
Crustacea. That part of the body which lies behind the
cephalo-thorax and answers to the free somites of the body
of Scorpio is swollen, and presents no distinct division into
somites.

The chelicerse are subchelate, that is to sav, the distal
joint is folded down upon the next, like the blade of a pocket-
knife upon the handle. The duct of a poison-gland, lodged in
the cephalo-thorax, opens at the summit of the terminal joint.
The pedipalpi are filiform, and, in the males, their extremities
are converted into singular spring boxes, in which the sper-
matophores are received from the genital apertures and con-
veyed to the females (Fig. 88, B),

The pulmonary sacs, two or four in number, are similar to
those organs in Scorpio, and are placed in the anterior part
of the abdomen ; a tracheal system is also present, a pair of
sternal stigmata, situated either behind the pulmonary sacs,
or at the end of the abdomen, leading into two more or less
branched tubes. There is a complex pharj'ngeal apparatus,

328

THE ANATOilY 07 INVERTEBRATED ANIMALS.

probably having the same function as in Scorpio,^ The
stomach giv^es otf caecal prolongations which may extend far
into the limbs. There is usually a dilated short rectum, into
which the branched Malpighian ducts open. The nervous
system, more concentrated than that of the Arthrogastra^ is
reduced to a supra-oesophageal ganglion and a single post-
oesophageal mass, with four indentations on either side. There
are six or eight simple eyes in the anterior part of the cara-
pace. Auditory organs have not been discovered in these or
any other Arachnida.

One of the most characteristic organs of the Araneina is
the arachnidium^ or apparatus by which the fine silky threads
which constitute the web are produced. H. Meckel," who
has fully described this apparatus as it occurs in Epeira dia-
dema, states that, in the adult, more than a thousand glands,

A B

Fig. 89, k.—Mygale Blondii (after Blancharc!).—TTie sfomadi with it? caeca, and the

remainder of the alimentary canal with the liver and Malpighian tubes.
Fig. 89, B.— The heart and arterial vessels of the same.

with separate excretory ducts, secrete the viscid material,
which, when exposed to the air, hardens into silk. These

1 LyoTiet^s " Anatomie de differentes Especes d'Insectes" ("Menfi. du Mu-
seum d'Histoire Naturelle," 1829) contains an elaborate account of this appa-
ratus, as well as of the structure of the pedipalps of the male spiders.

' " Mikrowraphie einicrer Drusenapparate der niederen Thiere." (Mailer's
" Areliiv," 1846.) See also Buchholz and Landois. {Ihid.^ 1868.)

THE ACARINA. 329

glands are divisible into five different kinds (aciniform, am-
pullate, aggregate, tubuliform, and tuberous), and their ducts
ultimately enter the six prominent araclinidial mammilloe^
which, in this species, project from the hinder end of the
abdomen. The superior and inferior mammillae are three-
jointed, the middle one is two-jointed. Their terminal faces
are truncated, forming an area beset with the minute arach-
iiidial papillve by which the secretion of the glands is poured
out.

The males are smaller than the females, and their ap-
proaches to the latter are made with extreme caution, as
they run the risk of being devoured ; extending their pedi-
palps, they deposit the spermatophores in the female genital
aperture, and betake themselves to flight.

The Aranelna are oviparous, but the development of the
embryo takes place as in the Arthrogastra, and there is no
metamorphosis.*

The Acaeixa. — In the Mites and Ticks, the hinder so-
mites are, as in the Spiders, distinctly separated from one an-
other, but they are not separated by any constriction from
the anterior somites.

The bases of the chelicerte, and of the pedipalpi, coalesce
with the labrum, and give rise to a suctorial rostrum (Fig.
90).

There are usually several gastric caeca, but no distinct
liver. Saliv^ary glands occur in some, and Malpighian caeca
are occasionally found. No heart has yet been discovered.
Special respiratory organs are sometimes wanting (e. g., Sar-
cojDtes) ; when present, they are tracheal tubes, springing
brush- wise from a common trunk which opens by a stigma.
The stigmata are usually two, sometimes anterior and some-
times posterior in position. The ganglia of the nervous sys-
tem are concentrated round the gullet, as in the Spiders ; and
the reproductive aperture is situated far forward, sometimes
close to the rostrum.

The greater number of the Acarlna are parasites upon
other animals or upon plants.

Most are oviparous, but the Orihatidce are viviparous.
The course of the development of the embryo is the same as
in the Spiders. The young, when born, are frequently pro-

» Claparede, " Recherohes sur 1' Evolution des Araignees," 1862. Also
Balbiani, "Ann. des Sc. Nat.," 18T3.

330 THE ANATOMY OF INVERTEBRATED ANIMALS.

vided with only three pairs of ambulatory limbs, the fourth
pair making its appearance only after ecdysis has occurred.

Fia. 90. — Ixodes vicinus, female (after Pajfenstecher*).— a, mandibular booklets ; ft,
d, e, fourth, third, and seccmd joints of the palp ; c, booklets of sternal surface of
proboscis ; /; base of the proboscis ; g, stigma ; h, genital aperture ; i, anal
valves.

In some A.carina, a singular kind of metamorphosis
occurs.

Thus, in Atax JBonzi, Clapar^de ^ observed that, before
the limbs appear on the blastoderm, a chitinous cuticula is
separated and forms an envelope, which he terms the " sac
of the deutovum." The proper vitelline membrane bursts
into two halves, much as in Limulus, and the deutovum
emerges. In the mean while, the anterior end of the blasto-
derm becomes fashioned into two prOcephalic lobes ; while
five pairs of tubercles, answering to the rudiments of the
chelicer.ie, pedipalpi, the two posterior gnathites, and one

» " Anatomie der Milben," 18«0.

3 " Studien an Acariden." (Zdtschrift fiir wiss. Zoologie^ 1868.)

THE PYCNOGOXIDA. 331

pair of thoracic limbs of the Spiders, make their appearance
beneath the sac of the deutovura. The rudiments of the
chelicerag and pedipaipi apply themselves together, and coa-
lesce into a proboscis. Thus the first larval form is com-
pleted. It tears the pseudoval sac, emerges, and buries
itself in the branchiae of the fresh-water mussel ( JJnio)^ upon
which it is parasitic. The cuticular investment of the first
larva now becomes distended by absorption of water, and
forms a globular case, the limbs being drawn out of their
sheaths. The second larval stage completes itself within the
sac formed by this singular ecdysis. The two palpi are de-
veloped from the pedipalpal portion of the proboscis; two
horny hooks from the cheliceral portion ; and, finally, the
hinder pair of thoracic limbs is added. This second larva
gradually passes into the adult Atax.

In the Acarus {JSlyobia coarctata) of the Mouse, Claparede
observed that the deutovum stage is followed by a tritovum ;
the chitinous sac, which invests the embryo within the deuto-
vum, apparently representing the cuticle of the first larva of
Atax, In this case, it presents a parallel to the JSFaiq^lius
cuticle of 3fysis.

The Arthrogastra, the Araneiiia, and the Acarina (with
some doubtful exceptions among the latter), possess the same
number of appendages, and do not differ from one another so
much as do the different forms of the Cojyejjoda, among the
Crustacea, But the remaining groups, which are usually in-
cluded among the Arachnida, namely, the Pycnogo7iida, the
Arctisca, and the JPe?itastomida, diverge widely from the Ar-
throgastra and the Araneina, though each exhibits certain
approximations to the Acari?ia.

The Pycxogoxida. — These are marine animals, with short
bodies terminated in front by a rostrum like that of the Mites,
but with a mere tubercle in place of the posterior thoracic and
abdominal somites. The adult has four pairs of enormouslv-
elongated, many-jointed, ambulatory limbs, in front of which
are three pairs of short appendages, the anterior of which may
be chelate, while the posterior are more or less rudimentary
(Fig. 91).

The alimentary canal sends off very long caeca into the
legs. There is a short heart, but no distinct respiratory or-
gans exist. A cerebral, nervous mass is connected with a
ventral chain of four or five pairs of ganglia. Four eyes are
seated upon a dorsal tubercle above the brain. The sexes are

332 THE ANATOMY OF INVERTEBRATED ANIMALS.

distinct, and the testes and ovaria are lodged in the legs and
open upon their basal joints.

Fig, ^l.—AmmotTiea pycnogonides, female (after Quatrefage?).— a, cesopliagns ; <f,
antennae ; &, etomach witU its prolongation into the antennae and limbs e • c,
rectum.

Tlie embryo emerges from the egg as a larva provided with
a rostrum, and with three pairs of appendages, which repre-
sent the short, anterior three pairs in the adult/ The four
pairs of great limbs of the adult are produced by outgrowths
from a subsequent posterior elongation of the body.

The comparison of the embryos of the Pycnogonida with
those of the Acarma, especially such as leave the egg with
three pairs of appendages, appears to me to leave little doubt
that the rostrum of the larva Pycnogomnn is formed, as in
the Mites, by the coalesced representatives of the chelicerre
and pedipalpi. If so, the seven other pairs of limbs are, by
three pairs, in excess of the number found in any Arachnidan.

^ A. Dohm, " IJntersucbungcn liber Bau und Entwickelung der Artbro-
poden." Erstes Heft. 1870.

THE PYCNOGONIDA.

333

On the other hand, the hexapod larva of the Pycnogonida
differs from the hexapod Nauplius of the Crustacea^ inasmuch
as the three pairs of appendages of a JVauplius always repre-

PiQ. ^I.—Macrobiotus Sehultzei (x 100), — a, mouth with six oral papillte ; 6, snllet,
calcified stylets; c, salivary glauds ; d, muscular pharynx ; «, ovary:/, vesicula
Bcmiualis ; flf, testis ; 1, 2, 3, 4, limbs. (After Greeff.^)

sent antennary and mandibular appendages, and these, by the
hypothesis, are to be sought in the rostrum of the Pycnogo-
nida.

The fact to which reference has already been made, that
the embryo Scorpion has six pairs of rudimentary appendages,
attached to as many of the anterior free somites, of which one
pair only remain (as the pectines) in the adult, leads me to
suspect that the Pycnogonida may represent a much modified

1 " Untersucliungen iiber den Bau der Bllrthierclien." (" Arcliiv far mikr.
Anat.," 18G6,)

334 THE ANATOMY OF IXYERTEBRATED ANIMALS.

early Arachnidan form, from which the Arthrogastra^ Ara^
iieidea, and Acaridea, have branched off.

The ApwCTIsca, or Tardigrada, are microscopic animals,
found, in association with Hotifera^in moss and in sand, rare-
ly in water, which present many points of resemblance to the
Acarina. The body (Fig. 92) is vermiform, with four pairs
of tubercles, representing limbs, terminated by two or more
claws. The fourth pair is directed backward at the hinder
end of the body, so that if these appendages answer to the
hinder pair of limbs in the typical Arachnida, the hinder tho-
racic, and all the abdominal, somites are undeveloped. The
mouth is situated at the extremity of a rostrum provided with
two stylets, which is so like that of the Acarina, that it may
probably be regarded as formed by the coalescence of cheli-
ceral and pedipalpal tubercles. There is a muscular pharynx
leading into a wide alimentary canal, which gradually narrows
to the anus. No organs of circulation or of respiration exist.
The paired ventral ganglia, which correspond in number with
the appendages, are large. They are connected by longitu-
dinal commissures with one another, and with a prae-oesopha-
geal cerebral mass which sometimes bears two eyes. The
Arctisca are hermaphrodite, the ovarian sac and the two
testes opening together into a cloacal dilatation in which the
intestine terminates. The ova are relatively very large, and
the cuticle of the parent is cast off and incloses them when
they are laid, as a sort of ephippium. Complete yelk-division
takes place. The young have one-third the size of the adult
when they are hatched, and they undergo no metamorphosis
beyond the addition, in some cases, of one pair of limbs after
birth.'

The Pentastomtda. — A still more aberrant form is the
parasitic Linguatula^ or Pentastomum^ which is found in a
sexless condition in the lungs and liver of herbivorous mam-
mals and of reptiles, and in the sexual state in the nasal cav-
ities and maxillary antra of Carnivores. Thus, as Leuckart's
investigations have proved, PentastomuTn tcenioides, which
inhabits the latter cavities in the dog and the wolf, is the
sexual state of the P. dentlculatum^ which occurs in the liver
of hares and rabbits.'

^ Kaufraann, " EntTrickelung und systematische Stellung der Tardigraden."
{Zeit. wiss. Zooloqie^ 1851.)

8 " Bau und Entwickeluugsgeschichte der Peatastomen," 1860.

THE PENTASTOMIDA.

335

The Pentastomida are elongated vermiform animals, the
bodies of which are divided by close-set transverse coustric-

Fio. 93.—Pentastomum tcmioides.—A. Male. B. Female. C. Anterior end of the body :
a, anterior hooks ; 6, posterior hooks ; d, mouth ; c, rudimentary palpiform cr-
gans. (After Leuckart.)

tions into numerous short segments. At first sight they
appear to be entirely devoid of appendages, but, on careful
inspection, four curved hooks are found, two on each side of
the mouth, which is situated rather behind the anterior ex-
tremity of the body. Each hook is solid, and its base pro-

FiQ. 94.— Embryo of Pentastomum tcenioides.

jects into the cavity of the bod}^ and gives attachment to the
muscular bands by which it is moved.

The mouth is surrounded by a chitinous ring ; a narrow

336 THE ANATOMY OF INVERTEBRATED ANIMALS.

oesophagus leads from it into a nearly cylindrical, straight,
alimentary canal, which terminates in the anus, in the middle
line of the posterior extremity of the body. A mesentery is
attached to the whole length of the alimentary canal and
holds it in place. A nervous ring surrounds the oesophagus,
and posteriorly presents a ganglionic enlargement whence
nerves are given off to the body. The muscles are striated.
The sexes are distinct, and the males are usually much
smaller than the females.

The testicle is an elongated sac which lies on the ventral
aspect of the intestine, and is connected anteriorly with two
vasa deferentia. These terminate on the fore-part of the
ventral aspect of the body, each having a saccular dilatation
which contains a very long, coiled, chitinous penis. In the
female, the ovary is also a large sac and the oviducts come
off from its anterior end, but the genital aperture is close to
the anus.

The ova undergo their development in the ovary. The
embryos are oval, but taper to the posterior end. In the
middle line, in front, are three sharp protractile styles, of
which the middle is the longest. Two pairs of articulated
limbs are attached to the middle of the ventral aspect ; each
is terminated by a double hooked claw. The embryo of Lin-
guatulci thus resembles those of the Acarina^ on the one
hand, and those of such parasitic Crustacea as Aiichorella,
on the other.

In the case of Pentastomum tcenioides, the embryos, in-
closed in their vitelline membranes, pass out of the bodies of
the dog or wolf, along with the nasal mucus. Taken into the
body along with the food of the hare or rabbit, they emerge
from the egg, penetrate the walls of the intestine, and lodge
themselves in the liver. Here they become encysted, grow,
and go through a series of changes of form, accompanied by
repeated ecdyses, until they pass into the state known as
Pentastomum denticulatum. If the flesh of the rodent con-
taining P. denticulatum is devoured by a dog, the parasite
passes into the frontal sinuses or maxillary antra of tho
latter, gradually takes on the form of P. tcenioides, and ac-
quires sexual organs. The parasitism of the Pentastomida^
therefore, is very similar to that of the Cestoidea,

Spiders and Mites abounded in the tertiary epoch, as
their remains, preserved in amber, show. Various Arthro-
gastra occur in the mesozoic formations, while Spiders and

THE MYRIAPODA.

337

Scorpions of large size have been found in the carbonifer-
ous rocks.

The Mtkiapoda. — In these Arthropods, the body is di-
vided into many segments, the most anterior of which takes
on the characters of a distinct head ; and almost all these
segments bear articulated limbs terminated by claws. In
the Centipedes ( Chilopoda)^ the segments of the body have
broad sterna, and the bases of the limbs are far apart ; but,
in the Millipedes ( Chilognathd), the sternal region is rudi-
mentary, and the bases of the limbs are close together.

Fig. 95. — A. Sedopendra borbonica iChllopoda). B. lulus Jlavozonatus (Chilo-

gnatha).^

Moreover, in the latter Sfroup, the majority of the segments
of the body bear two pairs of limbs, and probably represent
two somites.

» " Eegne Animal." Illustrated edition.

15

338

THE ANATOMY OF INVERTEBRATED ANIMALS.

The bead is either flattened from above downward (Chi-
lopocla), or from before backward {Chilognathd), Some
species are blind, but the majority have eyes, which are gen-
erally small and not very numerous ocelli, though, in some
cases, they are large compound eyes. There is always a pair
of jointed antenna?.

The majority have the mouth constructed for biting, and
are provided with a pair of mandibles, the most important
peculiarity of which is that they are jointed, and thus depart
less from the type of the ordinary limb than do those of in-

PiG. ^^.—Scolopendra Hopei (after Newport). " ^ ,.

A. Dorsal view of the anterior part of tlie body: «, antennae ; A, cephalic eegmeiit ;
B, basilar segment.

B. Ventral view of the head • a, B, as before. ^ ^,

C. Under view of the cephalic Pegmeut, showing the antennae, a; the eyes, *; the
labrum and the mandibles, IV'. - xi -u -i

D. The second pair of gnathites V, and the first pair of appendages of the baeilar
segment VI'.

sects, while, to the same extent, they approach the gnathites
of the Peripatidea. The mandibles are more modified in the
Chilopoda (Fig. 96) than in the Chilognatha. In the latter,
tlie second pair of gnathites form a broad four-lobed plate
which plays the part of an under-lip, while, in the Chilopoda^
they are soft and jointed, and united at their bases by a bi-
lobed median process (Fig. 96, v'). In the Chilognatha the

THE MYRIAPODA. 339

four segments which follow the head are free, and their append-
ages resemble ordinary limbs. The anterior pair is turned
forward and comes into relation with the mouth, and the ter-
gum of the first somite is often enlarged ; of the other three
somites, the appendages of one appear to be always abortive.
Thus there are three segments with single pairs of legs.
The rest each bear two paii-s.

In the Ghilopoda^ on the contrary, the head is followed by
a basilar segment (Fig. 96, J5), formed, according to Newport,
by the union of four embryonic somites, and carrying three
pairs of appendages. Of these the first are limb-like, but
are turned forward beneath the mouth (Fig. 96, D, vi') ; the
second pair are the strong recurved poison-claws, and the
hindermost pair may become functional legs, resembling those
which are attached to the succeedinsf somites, but are alwavs
smaller than the others, and may be altogether aborted in the
adult. The somites of the body never bear more than one
pair of limbs.

The alimentary canal is usually straight and simple, like
that of an insect larva. There are large salivary glands, and
the intestine is provided with Malpighian tubules.

The heart extends through the greater part of the length
of the body, and is many-chambered, there being one cham-
ber for each of the somites in which it lies. Each chamber is
somewhat conical in shape, being broader behind than in front,
and admits the blood by a pair of lateral clefts, while the
blood leaves it, in part by the communication with the adja-
cent chamber, in part by lateral arterial branches. A medi-
an aortic trunk continues the heart forward, and the lateral
trunks encircle the oesophagus and unite into an artery which
lies upon the ganglionic chain. The arterial system in the
Chilopocla is, in fact, as complete as that of the Scorpions.^

The respiratory organs are tracheae, which open by stig-
mata on the lateral or ventral surface of more or fewer of the
somites. In Scutigera the stigmata are situated in the me-
dian dorsal line of the body.

The nervous system presents a ventral chain, with a pair
of ganglionic enlargements for each segment of the body, the
most anterior of which are connected by commissures, which
embrace the oesophagus, with the cerebral ganglia.

The ovary in both Chilogiiatha and Chilopoda is long,

- Newport, " On the Structure, Eelations, and Development of the Nervous
and Circulatory Systems in the Myriapoda and Macrurous Arachnida." (" Phi-
losophical Transactions," 1863.)

340 THE ANATOMY OF INVERTEBRATED ANIMALS.

single, and tubular in fcrm. It lies above the alimentary
canal in the latter, between the alimentary canal and the ner-
vous system in the former. The double vaginae open on, or
close behind, the bases of the second pair of legs in the Chi-
lognatka ; at the posterior end of the body, beneath the anus,
in the Chilopoda. Two spermathecse and colleterial glands
are very generally present.

The testes in the Chilognatha are tubular glands, which
occupy the same position as the ovary, and open in the same
region. They have lateral caeca, and are connected by trans-
verse ducts. Two copulatory organs, or penes, are developed
on the sternal face of the sixth segment which follows the
head, or are connected with the bases of the seventh pair of
legs.

In the Chilopoda there is a good deal of variation in the
structure of the testis. Thus, in Lithohius^ the testis is a
single filiform tube, connected at the hinder end with two
deferent ducts which embrace the rectum. A large caecum,
apparently a vesicula seminalis, opens into each deferent duct.
But, in most Chilopods, the testes are fusiform acini, united
by delicate ducts with a median vas deferens. Two, or four,
pairs of accessory glands are connected with the opening of
the male apparatus.

The spermatozoa are inclosed in spermatophores in Scolo-
pendra^ Cryptops^ and Geophilus.

The Chilognatha copulate. In Glomeris and Polyxenus
the genital apertures of the two sexes are brought together
during copulation ; but, in lulus^ the penes of the male are
charged with the spermatic fluid before copulation takes place,
and it is by their agency that the female is impregnated.

The Chilopoda have not been observed to copulate ; in-
deed, the female sliows a tendency to destroy the males, as
among spiders. The male Geophilus spins webs like those
of spiders across the passages which he frequents, and depos-
its a spermatophore in the centre of each.

Metschnikoff ' has recently shown that, in the Chilognatha^
the process of yelk-division is complete, and confirms the obser-
vation of Newport (" Phil. Trans.," 1841), that the sternal face
of the blastoderm becomes sharplj'' infolded down its centre,
in such a manner that the anterior and the posterior halves of

^ Favre, " Anatomie des orcranes reproductcurs des Myriapodes." (" Annales
des Sciences Naturelles," 1855.)

a " Embryologie der doppelfussigen Myriapoden {Ch'dognatha)^ {Zeit'
scliriftfur wiss. Zoologie, 1874.)

THE MYRIAPODA. 341

that face of the embryo become closely applied together.
Metschiiikoff further -points out that only two pairs of ap-
pendages are converted into gnathites, and that a chitinous
cuticula, apparently identical with w4iat Newport describes
as the " amnion " in Julus, is early thrown ofi' from the em-
bryo. In some species it develops a median tooth-like pro-
cess, which serves to burst the vitelline membrane. New-
port describes a short cord, or fu7iicifhis, which connects the
anal extremity of the embryo with the so-called " amnion."
It is not improbable that this is simply the continuation of
the first larval skin into the rectum.

The embrvo Zulus at first bursts the vitelline menbrane,
and is inclosed only in the embryonic integument. At this
period its body is divided into eight segmicnts, of which the
first represents the head. Traces of the antennse are visible
on the sides of the head, and the four following segments
exhibit papillae ; those of the second, third, and fifth segments
develop into the three pairs of functional limbs, with which
the young myriapod is at first provided.

Between the terminal segment and the seventh the bcdy
grows and becomes divided into six rudimentary new seg-
ments. The terminal segment also becom.es divided into
, two. Thus, w^hen the young escapes from the embrj^onic
integument, it consists of nine complete segments, includirg
the head, with six rudimentary segments interposed between
the penultimate and the antepenultimate — making fifteen in
all ; w^hich is the full number of segments (head -f three tho-
racic + eleven abdominal somites) possessed by an insect
larva.

There is this difference, however, between the insect and
the larval myriapod : that since, in the latter, there are only
two pairs of gnathites, which must answer to the mandibles
and first maxillae of insects, the ambulatory appendages of
the second segment miust represent the second maxillae of
insects ; and hence, though there is apparently the same
number of somites in the two cases, there must in reality be
one fewer in the myriapod. The myriapod larva, therefore,
notwithstanding its hexapod character, is essentially differ-
ent from an insect larva.

The sixth and the seventh segments develop two pairs of
legs, as do all the newly-formed segments ; and it is w^orthy
of notice that the male copulatory apparatus, inasmuch as it
is situated in the seventh (sixth postcephalic) segment in the
adult, is developed from one of the primary segments of the

342 THE ANATOMY OF INVERTEBRATED ANIMALS.

embryo, and not from the subsequently-added segments*
New segments, each giving rise to two pairs of limbs, are
developed by sixes in the germinal region between the penul-
timate segment and the hindermost of the newly-formed seg-
ments, until the full number of the adult is complete.

In all other Ghilognatha of which the development has
been traced, the young, at first, possess only three pairs of
functional legs ; and one of the four segments which follow
the head is apodal. According to Fabre, the apodal seg-
ment in Polydesmus complanatus is the second, and not, as
in lulus, the third.

In the Chilopoda the young leave the &^^ with seven
{Llthohius, Szutigera) or a greater number of pairs of ambu-
latory limbs. Scolopendra is said to be viviparous. The
early stages of development of Geophilus have been de-
scribed by MetschnikoEf.^ Complete yelk-division takes
place, and when the young leaves the Q^g it has a cylindrical
body, like that of one of the Chilogriatha, and possesses many
pairs of limbs. Newport ^ has pointed out that, in Geophi-
lus longicornis, the basilar segment is formed by the conflu-
ence of four somites, of the appendages of which only two
are ultimately developed. Thus the basilar segment of the
head of the Chilopoda appears to correspond verv closely -
with the four somites which follow the head in the Chilogna-
tha. Under these circumstances, the difference in the posi-
tion of the reproductive apertures in the two groups is ex-
ceedingly remarkable.

Fossil Myriapoda occur both in the tertiary and secondary
formations, and there seems no reason to doubt that the
Xylohius sigillarim discovered in the coal of Nova Scotia by
Lyell and Dawson is to be referred to this group.

The Ixsecta. — Notwithstanding the vast number and the
singular diversity of form of Insects, the fundamental unity
of their structure is remarkable, and, in this respect, the group
exhibits a striking contrast to the Crustacea,

The division of the body into three regions — head, thorax,
and abdomen — is usually well marked, not only by the peculiar
modifications which the cephalic and thoracic somites under-
go, but by the attachment of the three pairs of ambulatory

- Zeitschrift fur wiss. Zooloqie^ 1875.

« Mono^rapli of the class Myriapoda, order Chilopoda. (" Transactions of
the Linnjean Society," xix.)

THE INSECTA. 343

limbs exclusively to the latter. The head possesses four pairs
of appendages, that is to say, one pair of antennse and three
pairs of gnathites ; and, as a general rule, there is a pair of
compound eyes, sessile upon the sides of the head ; sometimes
simple eyes are added to them. The first pair of gnathites
are the mandibles^ which are always devoid of a palp. The
second pair are the maxillae^ which, in those insects in which
the mouth is least modified, are distinct from one another and
laterally movable ; while the third pair of gnathites are
united together in the median line, and constitute the labium
of entomologists. In front of the oral aperture is a median
plate, the labrum / while from the floor of the mouth formed
by the labium another median process, the lingua^ is usually
developed.

It is hardly open to doubt that the mandibles, the maxillae,
and the labium, answer to the mandibles and the two pairs of
maxillae of the crustacean mouth. In this case, one pair of
antennary organs found in the latter is wanting in insects, as
in other air-breathing Arthropods, and the existence of the
corresponding somite cannot be proved. But if it be sup-
posed to be present, though without any appendage, and if
the eyes be taken to represent the appendages of another
somite, the insect-head will contain six somites, the praeoral
sterna being bent up toward the tergal aspect, as in the higher
Crustacea.

The three somites which succeed the head are termed re-
spectively prothorax, mesothorax, and metathorax. A pair
of legs is normally attached to each ; and, when wings exist,
they are lateral expansions of the tergal region (correspond-
ing with the pleura of Crustacea) of the mesothorax or the
metathorax, or of both.

In the abdomen there are, at most, eleven somites, none
of which, in the adult, bear ambulatory limbs. Thus, assum-
ing the existence of six somites in the head, the normal num-
ber of somites in the body of insects will be twenty, as in the
higher Crustacea Arachnida.^

One of the commonest of insects, the Cockroach (Blatta
(Periplaneta) orientalis) is fortunately one of the oldest, least

' It is open to question whether the podlcal plates represent a somite ; and
therefore it must be recollected that the total number of somites, the existence
of which can be actually demonstrated in insects, is only seventeen, viz., four
for the head, three for the thorax, and ten for the abdomen.

344 THE AKATOMY OF INVERTEBRATED ANIMALS.

modified, and in many ways most instructive forms ; at tlie
same time, it is not too small for convenient dissection.^

In this insect the head is vertically elongated, flattened
from before backward, and connected by a distinct neck
with the prothorax. The antennae are slender, as long as, or
rather longer than, the body. Large reniform compound eyes
are situated at the sides of the head. The tergal portion of
the prothorax (iwonotum) is a wide shield, which overlaps the
head, in front, and the tergal portion of the mesothorax, or
mesonotum^ behind. The legs are strong, and increase in
length from the first pair to the last. The abdomen is flat-
tened from above downward, and bears a pair of elongated,
many-jointed, setose styles {cerci) at its hinder extremity.

The males differ very considerably from the females. They
have two pairs of wings, of which the anterior are brown, and
are of a stiff and horny texture. As they serve to cover the
posterior wings, they are termed tegmina. When closed,
the left overlaps the right, and they extend back as far as
the posterior edge of the tergum of the fifth abdominal so-
mite.

The posterior wings, on the contrary, are thin and mem-
branous ; and, in a state of rest, are folded longitudinally
upon themselves, the folded edge being internal. In this
condition they are triangular, the base of the triangle Ij'ing
close to the posterior edge of the fourth abdominal somite,
and the right a little overlapping the left. When forcibly
unfolded and made to stand out at right angles to the body,
each of these wings is seen to have a nearly straight, thick-
ened, anterior edge, while its rounded outer and posterior
edges are very thin. The wing is strengthened by radiating
thickenings, or nerviires^ united by delicate transverse ridges ;
and, when left to itself, it springs back into its folded state
with some force.

The abdomen of the male is not very broad. The sterna
of the abdominal somites are all flattened ; and, to the hind-
ermost, two minute unjointed stjdes are attached, while some
singular hooked processes are seen, on close inspection, to
protrude between the hindermost tergum and the hindermost
sternum. The abdomen of the female is verv much broader,
especially toward the middle of its length. The hindermost
sternum is convex and boat-shaped, and its posterior half is
separated along the middle line into two halves, united only

> Se^., for an excellent figure and description, Rollcston, " Forms of Animal
Life," p. 199, plate vi.

THE COCKROACH. 345

Iby a thin and flexible raembrane. Sometimes the great egg-
case, which the female carries about for some time before it
is laid, is seen protruding between the posterior terga and
sterna. The female has movable tegmina, but they are very
3mal], inasmuch as they do not extend beyond the middle of
the metathorax, and are widely separated in the middle line;
they are, in fact, mere rudiments of the anterior wings. The
posterior wings appear, at first, to be altogether wanting.
But the outer extremities of the metanotum^ or tergal portion
of the metathorax, present triangular areas, in which the in-
tegument is very thin, and exhibits markings which simulate
the nervures of the wings. There can be no doubt, in fact,
that these are undeveloped wings, and they show, in a \'^vy
instructive manner, that the wings are modifications of that
part of the insect skeleton which answers to the pleura, and
therefore to the lateral parts of the carapace, of a crustacean.

The convex dorsal wall of the head of the Cockroach (Fig.
97) is termed the epicraniuni, A median suture — the epicra-
nial suture — may be seen, especially in young Cockroaches,
traversing it from before backward, and dividing between the
eyes into two branches, one of which passes toward the artic-
ulation of each antenna. The basal joint of the antenna is
attached to a transparent flexible membrane, which occupies
an oval space, the antennary fossa, and allows of the free play
of the antenna. A little projection of the hard chitinous skel-
eton, when it bounds the inferior margin of the fossa, helps
to support the joint. On the inner side of, and above the an-
tennary fossa, there is an oval fenestra, covered only by a
thin and transparent portion of the integument, which allows
a subjacent tissue of glistening white appearance to be seen
(Fig. 97, I.,n., h). These have been regarded as rudimentary
ocelli by some entomologists ; but their structure needs care-
ful examination before this view can be adopted.

The transparent cornea of the eye, situated external to
and behind the antennary fossa, is elongated, wider above
than below, and has a concave anterior, and slightly convex
posterior, margin. The numerous facets into which the cor-
nea is divided are hexagonal in shape, and very small.

The broad flattened region of the fore-part of the head, on
the oral side of the epicranial suture, is the clypeus. It is
prolonged in front of the mouth, and w^ith the truncated edge
of this prolongation the flap-like labrum is freely articulated.
Behind the labrum are two, very stout, curved mandibles,
strongly toothed at their extremities (Fig. 97, II., mn). Each

346 THE ANATOMY OF IXVERTEBRATED ANIMALS.

mandible is articulated with the truncated edge of the lateral
part of the skeleton of the head, beneath the eyes, which is
termed the geiia^ in such a manner as to be freely movable

Fig. ^l.—Blatta orientalU.—l., TI. Side and front views of the head : a, the epicranial
suture, at the ends of the lateral branches of which are 6, the fenestras;/, the
antennas ; g, the eyes ; lb, the lahrum ; wm, the mandible; co, the cardo; st, the
Btipes ; ga, the galea ; pi, the palpus of the maxilla ; p. the palj)U8 ; q, the men-
turn and submentnm of the labium; k, the marjrins of the occipital foramen : ic,
inferior cervical sclerites ; Ic, lateral cervical sclerites; pn, pronotum. III. The
labium and the riorht maxilla, viewed from below; letters as before, except ?o,
lacinia of the maxilla; pg^ paraglossag; li^ ligula; m, mentum; 5W, submentum of
the labium.

toward and from the median line, but in no other direction.
The proximal end of the maxilla (Fig. 97, III.) is formed by
an elongated basal articulation, the cardo^ which is directed
transversely to the axis of the head, and is connected with
the inferior margin of the epicranium, or rather with a thin
skeletal band which runs round the posterior margin of the

THE COCKROACH. 347

epicranium, and is firmly united with it only od its dorsal side.
This band forms the boundary of the so-called occipital fora-
men^ by which the cavity of the head communicates with that
of the neck, the chitinous wall of the latter region being con-
tinuous with it. Articulated at right angles with the cardo
is the stipes^ or second joint of the maxilla. This is freely
movable in the lateral direction, and its outer distal angle
bears the continuation of the limb, or palpus, formed by two
short and three long joints. Two processes terminate the
stipes ; of these, the anterior and outer — the galea — is soft,
rounded, and possibly sensory in function, while the posterior
and inner — the lacinia — is a curved cutting blade with a
toothed and spinose inner edge.

The labium (Fig. 97, III.) consists of two incompletel^^-
separated median plates, the suhme7itiimhehm^, and the me7i-
tuni in front ; upon the latter follows a bilobed terminal piece,
the ligula, each lobe of which is again divided longitudinally
into two portions, which have considerable similarity to the
galea and lacinia of the maxilla. The outer is usually termed
the paraglossa.

Between the mentum and the ligula, on each outer edge
of the labium, a small piece, the palpiger, is articulated ; it
bears the three-jointed labial palpus, which is to be regarded
as the proper free termination of the second maxilla. The
resemblance between the labium and a pair of maxillae which
have coalesced, is obvious.

The submentum is not directly articulated with the cranial
skeleton, but its posterior edge is close to one of the cervical
sclerites^ or skeletal elements observable in the chitinous in-
tegument of the neck, of w^hich there are altogether seven.
One is dorsal, median, and marked by a deep longitudinal
depression. It articulates with the dorsal margin of the oc-
cipital foramen. Four are lateral, two on each side (Fig. 97,
I., Ic) ; these take an oblique course from the dorsal part of
the boundary of the occipital foramen, with a tubercle of
which the anterior piece is articulated, to the anterior edge
of the episternum of the prothorax. The inferior cervical
sclerites (Fig. 97, I., ic) are two narrow transverse plates, one
behind the other in the middle line. They appear to repre-
sent the part called gula, which, in many insects, is a large

* I use this term in the sense in which it has been employed by Milne-
Edwards, to denote any definite hardened part of the chitinous skeleton. It
is to the latter what a distinct ossification is to the skeleton of a vertebrated
animal.

348 THE ANATOMY OF INVERTEBRATED ANIMALS.

plate, confluent with the epicranium above and supporting the
submentum anteriorly. I think it is probable that these cer-
vical sclerites represent the hindermost of the cephalic somites,
while the band with which the maxillae are united, and the
genae, are all that is left of the sides and roof of the tirst max-
illary and the mandibular somites ; the epicranial expansions
being mainly formed by the upward and backward extension
of the ophthalmic and antennary sterna, which arise out of
the procephalic lobes of the embryo. In addition to these
externally-visible sclerites, there is a sort of internal skeleton
(endocranium or tentormm)^ which extends as a cruciform
partition from the inner face of the lateral walls of the cra-
nium, close to the articulation of the mandible, to the sides of
the occipital foramen. The centre of the cross is perforated
by a rounded aperture, through which the oesophageal nerve-
commissures pass. The commencement of the oesophagus
traverses the interspace between the anterior processes of the
cross ; the tendons of the great adductors of the mandible
pass through the lateral apertures ; and the backward contin-
uation of the gullet enters the thorax through the posterior
aperture, included between the tentorium and the margins of
the occipital foramen.

In each somite of the thorax a distinct median sclerite,
the sternum^ maj' be observed ; and a much larger tergal
piece, the notum. At the sides of the somite are other defi-
nitely-arranged sclerites, the anterior of which appear to an-
swer to the episternum and epimera in the Crustacea, while
the posterior, perhaps, properly belong to the attached limb.

Forked or double apodemes, the antefurca., medifurca^
and postfarca, project from the sternal wall of each somite
of the thorax into its cavity. They support the nervous cord
and give attachment to muscles.

The legs present a large basal joint, the coxa, between
which and the third, termed femur, a small articulation, the
trochanter, is interposed. Upon the femur follows an elon-
gated tibia / and this is succeeded by the tarsus, which con-
sists of six joints. Of these, the proximal joint is long and
stout, the three next are short, the fifth is elongated and slen-
der ; the sixth, very short, is terminated by two curved and
pointed claws {ungues).^

» Mr. Westwood (" Modern Classification of Insects," vol. i, p. 416) says
that the tarsi are five-jointed, and that there is a pulvillus between the ungues.
The sixth joint appears to be what Mr. Westwood terms pulvilhts, but it is a
true joint, provided with a special flexor, the slender tendon of which, how-
ever, traverses several of the joints of the tarsus.

THE COCKROACH. 349

The broad dijQferences in the structure of the abdomen of
the male and female have been already pointed out. Of the
eight terga externally visible in the female (Fig. 98), the first
is shorter than those which succeed it ; and the hindermost
(Fig. 98, 10) is escutcheon-shaped, deflexed at the sides, thin
in the middle, and notched at the end. When this tergum is
gently pulled backward, two other very narrow terga (Fig.
98, 8, 9), of which the anterior overlaps the posterior, and
which were hidden between it and the antepenultimate or
seventh tergum, become visible. The apparent eighth tergum
is therefore really the tenth. Beneath the tenth tergum are
two triangular j90C?ica^ plates (Fig. 98, 11), one on each side
of the anus. Provisionally, I take them to be the scleriies of
the eleventh abdominal somite.

The first sternum is confluent with the second, and largely
hidden by the coxce of the metathoracic limbs. The seventh
is greatly enlarged, and its posterior edge is produced into a
boat-shaped process, nearly divided into two portions by an
inward fold of the integument along the median line.

Completely hidden by the seventh sternum is a thin plate,
narrower in front than behind, where it is produced on each
side. Anteriorly, it is articulated with the sternum of the
following somite, so as to form a sort of spring-joint, which
ordinarily keeps it applied against the latter, and therefore
directed obliquely upward and a little forward. The large
aperture of the vulva (Fig. 98, v) lies in the middle of this
plate. On the sternal region behind the vulva, between it
and the anus, arises a pair of elongated processes, divided
into two portions, of which the outer is thick and soft, the
inner slender, pointed, and hard. They embrace and partly
ensheath two other processes, having somewhat the shape of
knife-blades, the anterior fixed ends of which are curved, and,
being attached to the sides of the somite to which they be-
long, are distant, while the blades meet, and are applied to-
gether in the middle line. Of these, which may be termed
gonapophyses^ the study of their development shows that the
posterior bifid pair belong to the ninth somite, while the an-
terior pair belong to the eighth.

The cerci (x) are attached to the dorso-lateral part of the
tenth somite.

In the abdomen of the male Blatta (Fig. 99) the ten terga
are readily discernible ; but the eighth and ninth are very
short, and the former overlaps the latter. The tenth tergum
is flat, and has a freely-projecting, truncated, posterior mar-

350 THE ANATOMY OF INVERTEBRATED ANIMALS.

gin. Articulated beneath its lateral edge are two multiartic-
ulate cerci (x), similar to those of the female.

Beneath the tenth tergum, and hidden by it, are the two
podical plates (11) between which the anus opens. The first
sternum is small, and may easil}^ escape notice. The second
to the sixth sterna are of nearly equal width and length. The
seventh and eighth are narrower; the ninth still narrower and
longer, about half of its length being covered by the eighth.
The covered half is different in texture from the uncovered,
being thinner and more transparent, and its anterior margin
is deeply notched. The uncovered half is strong, horny, and
dark-colored, convex below and concave above ; its free pos-
terior margin is obscurely trilobed by two lateral, shallow
notches. On each side, a slender, unjointed, setose style,
which projects backward and outward, is attached to this
sternum.

Thus, the tergal surface of the abdomen of the male essen-
tially resembles that of the female, while the sternal surface
differs in exhibiting two sterna more (namely, the eighth and
ninth) without dissection. Hence, while in the female the
opening of the recto-genital chamber lies between the tenth
tergum and the seventh sternum, in the male it lies between
the tenth tergum and the ninth sternum.

When the tenth tergum and the podical plates are removed,
a very singular apparatus, the male genital armature, comes
into view. It consists of a number of chitinous processes
having the form of plates and hooks, the exact form and dis-
position of which could be made intelligible only by numerous
figures. It may be stated generally, however, that these plates
and hooks terminate processes of the sternal region of the
tenth somite, on each side of the aperture of the vas deferens,
and therefore, though they are of the same nature as the gona-
pophyses of the female, they are not their exact homologues.

The most conspicuous division of the right gonapophysis
is a broad plate divided at the extremity into two portions,
the inner of which curves inward and ends in two or three
sharp spines, while the outer is coiled upon itself so as to
resemble a short corkscrew. The left gonapophysis is pro-
vided with a long process like a tenaculum, the incurved
extremity of which is denticulated.

The alimentary canal of the Cockroach commences by the
oral cavit}", situated between the labrum in front, the mandi-
bles and maxillae at the sides, and the labium, with the large

THE COCKROACH.

^o^

Fio. 98.

mt Vlll

xix

552

THE ANATOMY OF INVERTEBKATED ANIMALS.

Fio. 93. — ^Loni?itndinal and vertical eection of a female Cockroach (Blatta).~i to XX^
somites of the body ; 1 to 11, somites of the abdomen ; A, antenna ; lb, labrum ;
a. mouth ; 6, cesopbaguB ; c, crop ; rf, proveotricnius ; e, pyloric caeca ; /, chylific
ventricle ; g, insertion of the Malpighian caeca ; h, iutestiue ; i, rectum ; v, vulva;
/, salivary irland ; k, salivary receptacle. By an error, the duct is made to termi-
nate above instead of beneath the lingua. M, position of heart ; m, cerebral gan-
glia ; iV^, thoracic ganglia ; x, ccrci.

lingua, or h7/pophar^nx,he\i\vA. The oesophagus, beginning
as a narrow tube, passes between the anterior crura of the
tentorium, and then, leaving the head bj the occipital foramen

Fio. 99.— Longitudinal and vertical section of the abdomen of a male Cockroach
{Titafta). — 1, 2, 3, 4, etc., ter^a and sterna of the abdomen ; t, mushroom-shaped
gland ; V, aperture of the vas deferens ; A, anus.

and traversing the neck and thorax, gradually widens into
the large crop or ingluvies (Fig. 98, c), which lies in the ab-
domen. This is followed by the small thick-walled proven-
triculus (Fig. 98, c?), shaped like a pear, with its broad end
applied against the crop. The narrow end of the proventricu-
lus opens into a wide canal, the so-called chylific ventricle or

THE COCKROACH. 353

ventriculus (Fig. 98, /*), an elongated tube, the junction of
which with the intestine is marked by the insertion of the
numerous Malpighian tubes. The anterior end of the ven-
triculus is provided with seven or eight caecal diverticula of un-
equal lengths (Fig. 98, e), the pyloric cceca. The first portion
of the intestine (ileum) is narrow. The next, termed the
colon, is very wide, and somewhat sacculated. A constric-
tion marks off the region of the colon from the straight short
rectum (Fig. 98, i), which terminates in the anus, situated
at the hinder extremity of the body between the podical
plates.^

The aperture by which the mouth communicates with the
gullet is small, and situated at the superior and anterior part
of the buccal cavity. A broad projection of the posterior and
inferior wall of the buccal cavity occupies all the space between
the oesophageal opening of that cavity and the labium, and
ends in a free subcylindrical process. This is termed hypo-
p>hary)ix or lingua, but it might be well to reserve the term
lingua for the free end, and hypopjliarynx for the attached
posterior portion. The anterior surface of the hypopharynx
slopes downward and forward; its sides are supported by two
sclerites, which are narrow and rod-like above and broad be-
low, where they unite in an arch on the dorsal face, just where
the free part, or lingua, begins. On the under side of the lin-
gua are two broader sclerites, which also unite and form an
arch, which lies over the opening of the salivary duct. The
anterior surface of the lingua and hypopharynx is beset with
fine hairs.

The two salivary glands, with their receptacles, are greatly
developed in the Cockroach.^ The glands (Fig. 98, /) lie on

1 M. F. Plateau (" Eecherches sur les ^henomenes de la digestion chez les
Insectes," 1874 ; " Note sur les plienomenes de la digestion chez la Blatta
am^ricaine {Periplaneta Americana^ 1876 ; and " Eecherclies sur les pheno-
raenes de la digestion chez les Mynapodes," 1876) divides the alimentary canal
of insects and myriapods into a buccal, a median, and a terminal portion. The
buccal portion consists of the oesophagus, crop, and proventriculus — which last
he considers to be a mere strainer, and to have no masticatory function. The
middle division lies between the proventriculus and the insertion of the Mal-
pighian tubes. The terminal division extends from the latter point to the anus.
With the solitary exception of /wZi/s, the secretions of the alimentary canal are
always alkaline, and that which efi'ects the transformation of the albuminoid
elements of the food into peptones appears to be furnished by the middle di-
vision, which is lined by epithelium, devoid of any ciiticle. In carnivorous in-
sects digestion may take place in the crop by the flow of the secretion of the
middle intestine into it. The salivary fluid of Blatta rapidly efi'ects the trans-
formation of starch into sugar.

2 The salivary glands are well described by Basch, " Untersuchungen fiber
die chvlopoietische und uropoietische Svsteme der Blatta orientalis." ('' Sitzb.
Wiener Akad.," 1858.)

354 THE AXATOMY OF INVERTEBRATED ANIMALS.

each side of the oesophagus and crop, extending through the
thorax, as far as the commencement of the abdomen. Each
gland is a white mass, as much as a quarter of an inch long,
and composed of numerous acini. The ducts which arise from
these acini unite first into a single trunk on each side, and
then, beneath the sub-oesophageal ganglion, the two trunks
join to form the single short salivary duct which opens be-
neath the lingua. The ducts of the salivary glands are lined
by a transversely-ribbed chitinous membrane, so that they
greatly resemble tracheae.

The salivary receptacles (Fig. 98, k) are elongated oval
sacs, three-eighths of an inch long, each of which is situated
at the extremity of a long duct. The ducts unite in front with
one another, and with the duct of the gland, to form the short
terminal common duct. The receptacle and its ducts have a
chitinous lining similar to that of the duct of the glands, but
the spiral marking does not extend over the walls of the re-
ceptacle.

The proventriculus has a thick muscular coat, and the chi-
tinous lining which is continued into it from the ingluvies is
greatly thickened, and produced into six hard, brown, ridge-
like principal teeth. Posterior to these is a circle of six
prominent cushions covered with setae, and similar setae beset
the lining membrane of the funnel-shaped cavit\^ into which
they project. Between each pair of principal teeth are five
smaller tooth-like ridges, of which the median is the largest,
and a variable number of still finer lonijitudinal elevations lie
between them.

The proventriculus leads posteriorly into a narrow, thick-
coated canal, the tubular extremity of which projects freely
into the much wider anterior end of the chylific ventricle, and
constitutes a very efficient valve.

The short and narrow anterior division of the intestine
(ileum) is separated from the colon by a circular valve, the
surface of which is beset with small spines.

The Malpighian glands are very numerous (20-30), deli-
cate, caecal tubules, of even diameter throughout, and lined
by a small-celled epithelium inclosing a central cavity.

The communication between the colon and the rectum is
very narrow, but is not valvular. The walls of the rectum
itself are raised into six ridges, which project into its inte-
rior and are abundantly supplied with tracheae ; these are the
so-called rectal glands. Anal glands appear to be absent.

The histology of the alimentary canal has been particu-

THE COCKROACH. 355

larly studied by Basch.^ From the oral cavity to the funnel-
shaped extremity of the proventriculus, it is lined by a chiti-
nous coat continuous with the chitinous layer of the integu-
ment, and beset for the greater part of its extent with fine
setiform processes. Beneath this is the proper endoderm,
consisting of a layer of cells. Next follows a structureless
merabrana propria or basement membrane ; and this is suc-
ceeded by two layers of striped muscular fibres, the internal
disposed longitudinally and the external circularly. In the
proventriculus, the muscular layers become much thicker, and
some of those of the outer layer acquire a radial arrangement,
while the longitudinal muscles are disposed in bundles which
correspond with the six principal ridges. In the chylific ven-
tricle, the muscular layers and the basement membrane are
disposed much as before. The basement membrane presents
pits on its free surface in which rounded cells are lodged, and
is beset between these by the elongated cells of a cylinder
epithelium. The free ends of these present a thick wall,
marked by vertical striations. There is no chitinous layer.
The caeca are merely diverticula of the wall of the chylific
ventricle.

The intestine, finally, repeats tlie structure found in that
part of the alimentary canal which lies in front of the chy-
lific ventricle and is provided with a setose chitinous lining.

Basch found the secretion of the salivary glands and the
contents of the crop acid,^ and that an infusion of the sali-
vary glands, acidulated with hydrochloric acid, digested fibrin.
The contents of the chylific ventricle were neutral or alka-
line ; and an infusion of the chylific ventricle at once turned
starch into sugar. The same effect was produced by an infu-
sion of the salivary glands.

The heart (Fig. 98, h) is a slender inconspicuous tube,
which occupies the middle line of the dorsal wall of the
abdomen, and presents, at intervals, pairs of lateral apertures.
The wall of the abdomen internal to the chitinous integu-
ment is lined by a soft cellular substance {hypoderniis), the
outer layer of which represents the ectoderm or epidermis,
while the deeper part is the parietal layer of the mesoderm.
This last contains a stratum of longitudinal muscular fibres,
divided into segments or myotomes^ in correspondence with
the somites, and numerous tracheae. The heart is inclosed in

» '' Sitzungrsberlchte der Wiener Akademie," xxxili., 1858.
^ Plateau denies that the salivary secretion of Blatta is ever acid, and as-
cribes the occasional acidity of the contents of the crop to the food.

356 THE ANATOMY OF INVERTEBRATED ANIMALS.

the abdominal wall which surrounds it on all sides, leaving
only a small pericardial space.^ Beyond the slender aortic
canal in which the heart terminates anteriorly, and which
passes into the thorax and the head, no vessels appear to be
given off from the heart.

Delicate triangular sheets of muscular fibre, the alary
muscles^ are attached in pairs by their bases to the wall of
the pericardial chamber, while their apices are inserted into
the hypodermis. They occupy the interspaces left by the
principal dorsal branches of the tracheae, which form arches on
each side of the heart.

From the inner face of the abdominal wall, processes are
given off, some of which appear to hang freely into the ab-
dominal cavity, while others accompany the numerous tracheae
which pass to the alimentary canal. When the abdominal
cavity is laid open, its inner lining has a villous appearance,
and often seems to be full of free granular matter, as the pro-
cesses very readily break up into fragments. The substance
which thus fills up the interspace between the parietes of the
abdomen and the contained organs is the corjnis adiposum.
It is made up of cells often so arranged as to form a network,
and it usually has a milk-white color, which arises partly from
the air contained in the tracheae, and partly from innumerable,
strongly refracting granules contained in its component cells.

There are ten stigmata on each side of the body of Blaita^
eight in the abdomen, and two in the thorax. The latter are
situated between the prothorax and mesothorax, the meso-
thorax and the metathorax, respectively ; above the attach-
ment of the coxge and beneath the terga. The abdominal
stigmata lie in the soft integument which connects the sterna
and terga of the somites. All the stigmata are situated in
conical thickened elevations of the integument. The thoracic
stigmata are the largest, and the anterior pair have a distinctly
two-lipped aperture, the anterior lip being notched in the cen-
tre. The openings of the abdominal stigmata are more oval,
and are inclined backward. Immediately within each stigma
the tracheal trunk into which it opens is provided with a val-
vular arrangement, by which the passage can be closed or
opened.

» Cornelius (" Beitrage zur naheren Kenntniss von Periplaneta (Blatia) ori-
eidalis,^'' 1853) found that the pulsations of the heart could readily be observed
in Blattce which had recently undergone ecdysis. They were as frequent aa
eighty in the minute ; but allowance must be made for the disturbed condition
of the insects under observation.

THE COCKROACH.

357

The larffe tracliene which take their orisrin from these
stigmata immediately divide and give off dorsal and ventral
branches ; the former unite in a series of arches on each
side of the heart, while, on the ventral side, the branches
are connected by trunks which run parallel with the abdominal
ganglia. Large tracheae pass from the anterior thoracic stigma
through the neck into the head, and, in the abdomen, the vis-
cera receive an abundant supply of air-tubes.

!:AI,

r-<?

Fig. 100.— Blatta orientalis.—C, the brain with the antennary (a) and optic ih)
nerves ; c, e, ./; gr, A, etomato-gastric nerves. B, the anterior end of the gullet.
A^ the crop. D, the gizzard.

The lobes of the corpus adiposum are also plentifully
supplied with tracheas, while fine trunks enter the substance
of the ganglia and nerves and there ramify. TrachejB accom-
pany the nervures of the wings and are abundantly distrib-
uted to the muscles.

The nervous system consists of the supra-oesophageal
ganglia (Fig. 100, A), commonly termed the brain, united by

358 THE ANATOMY OF INVERTEBRATED ANIMALS.

thick and short commissures with an infra-oesophageal gan-
glionic mass, situated in the head; of three pairs of large
coalesced ganglia in the thorax, one for the prothorax, one
for the mesothorax, and one for the metathorax ; of six pairs
of closelj'-united smaller ganglia in the abdomen ; and of a
set of visceral or stomato-gastric nerves. The several pairs
of thoracic and abdominal ganglia are united by double com-
missural cords. In the males the commissures which unite
the abdominal ganglia are not straight, but are bent, as if it
were needful to make allowance for the possible elongation
of the abdomen. The supra-oesophageal ganglia give off the
nerves to the antennae from their antero-lateral angles ; while
their posterolateral angles are produced into the great optic
nerves. Above the margin of each antennary nerve there is
a small rounded tubercle which is in immediate relation with
the silvery patch which shines through the fenestra on the
inner side of the antennary fossa. Beneath this tubercle,
and on the inner side of the antennary nerve, arises the root
of the stomato-gastric system of nerves. Each root passes
forward for a short distance, then turns inward, and in the
middle line enters a heart-shaped ganglion situated on the
gullet (Fig. 100, c). From this a median cord passes back-
ward beneath the brain and enters a ganglion, which is con-
nected on each side with two others (6, e). The continuation
of the median cord passes back along the tergal wall of the
oesophagus, and where this begins to dilate into the crop
ends in a small triangular ganglion (^), whence lateral
branches are given off, which can be traced as far as the
gizzard.

The exact form and arrangement of the male organ of
generation has only recently been made out. The most con-
spicuous of these organs is a mushroom-shaped gland (Fig.
99, t) composed of a great number of short caeca attached to
the extremity of the also very short vas deferens. It is
lodged in the hinder end of the abdomen, and covers the
posterior abdominal ganglion. The contents of the casca are
viscid, granular, and usually brilliantly white. The anterior
end of the vas deferens is dilated, and the coeca are arranged
in two groups which open into each side of the dilatation.
The contents of the vas deferens are also white and viscid,
and evidently consist in great measure of the secretion of
the caeca. In the adult male, however, innumerable sperma-
tozoa with straight rod-like heads, and long flagella, are to be
found intermingled with the contents of the vas deferens and

THE COCKROACH. 359

its dilatation. On the sternal side of the musliroom-shaped
gland, between it and the last abdominal ganglion, there is
an accessory gland composed of dicliotomous monilated tubes,
lined by a columnar epithelium, all bound together by a com-
mon investment into a flattened elongated mass.

As the duct of the mushroom-shaped gland in the adult
male always contains spermatozoa, and no other organ con-
taining spermatozoa is to be found, this gland has naturally
been taken for the testis. Rajewsky,^ however, has recently
pointed out that the true testes are situated in the tergal
region of the abdomen, and that they may be found in this
region in the young and yet wingless males, though they are
much obscured by the corpus adiposum which invests them.
He traces the efl^erent duct of the testis to the glands just
mentioned. In the adult male the testes atrophy, and are
hardly to be discovered among the masses of the corpus adi-
posum. I have found the testes in the young males in the
position assigned to them by Rajewsky. They consist of
numerous oval or pyriform sacs attached by short pedicles to
a common duct.

The ovaries (Fig. 101) are two groups of eight tubes, sit-
uated on each side of the hinder half of the abdomen. The
ovarian tubes, or ovarioles, of each group communicate with
a short oviduct, which soon unites with its fellow in the mid-
dle line and opens externally by the very short and wide
vagina. The finely tapering anterior ends of the ovarioles
of each side are continued forward by delicate cellular pro-
longations. These finally unite together into one long fila-
ment, which can be traced for some distance forward among
the lobes of the corpus adiposum. It is a cellular cord, which
appears to be nothing but a process of the mesoderm. Nu-
merous nucleated cells, from some of which the ova take their
origin, while others remain as interstitial cells, which are
eventually converted into an epithelium, make up the sub-
stance of the slender anterior terminations of the ovarioles.
The ova situated behind these enlarge, and become disposed
in a single series. Further on, the epithelial cells form a thick
stratum round each egg, and possibly assist in the formation
of the large vitellus with which it is ultimately provided. As
the egg advances toward maturity, the vitellus acquires first
a finely and then a coarsely granular structure, and the ger-
minal vesicle and spot, previously conspicuous, are no longer

» Hofmann and Schwalbe, " Jahresbericlit," 1875. The original paper ia
in Russian, and 1 have not seen it.

360

THE ANATOMY OF INVERTEBRATED ANIMALS.

to be seen. Behind the junction of the oviducts with the
vagina and the last abdominal ganglion which lies upon the
latter, there is a small sac with a long neck from which a short

Fig. 101.— Blaita orientalis.—F ema]G genital organs: a, the poeterior abdominal
ganglion ; 6, the oviducts ; c, d, e, the ovarian tubes ; /, the filament by which
their extremities are united ; g, the spermatheca ; A, the coUeterial glands.

caeca! process is given off. It has a thick chitinous lining and
a muscular investment, and is the s/jermatheca. Behind it
are two large, ramified, tubular coUeterial glands^ which prob-
ably give rise to the substance of which the egg-case is
formed. Their conjoined ducts open behind the spermatheca.
The eggs are inclosed, sixteen together, in strong capsules
of a horny consistency, shaped somewhat like a cigar-case,
and presenting a longitudinal slit, the raised and serrated edges
of which are closely applied to one another. It is through
this slit that the fully-developed young make their exit. The
eggs attain one-sixth of an inch in length. Each has its own
thin but tough brownish shell, the surface of which is beauti-
fully ornamented with hexagonal patches of minute tubercles.
They are arranged parallel with one another in two opposite
series, one series occupying each half of the case. The eggs,
adapting themselves to the form of the case, are convex out-
ward and concave inward, and thus, though their ends touch,
a median space is left bet\veen the two sets. The inner con-
cave face of the egg is that on which the sternal face of the
embryo is situated. The female carries the egg-case about

THE COCKROACH. 36X

for a week or more, before depositing it. The young leave
the eggs as minute active insects, colorless, except for the
large dark eyes. Before they are hatched they acquire eyes,
antennae, gnathites, legs, and short cerci, which differ only in
detail from those of the perfect Blatta^ into which the larva
passes hy successive ecdyses. According to Cornehus (/. c,
p. 29), the Cockroach undergoes seven ecdyses : the first im-
mediately on leaving the Qg^^ the second a month later.
After the second ecdysis the insect sheds its skin only once
a year ; so that it attains its adult condition only in its fifth
summer. The chitinous cuticula splits along the median line
of the tergal aspect of the head, thorax, and abdomen, before
it is cast.

Thus the Cockroach is said to be an insect without meta-
morphosis. For although the male, in the later stages of its
growth, acquires wings, and thus does become very sensibly
metamorphosed from a merely cursorial animal to one which
has, at any rate, the capacity for flight, there is no period in
the life of this insect in which the larva passes into a resting
condition, during which it takes no food, and in the course of
which it develops its wdngs. In other words, the Cockroach
passes through no ^i/p«-state, which the insect enters as a
larva, and leaves as an imago, such as is so well known to
occur in the course of the development of Moths and Butter-
flies. The term metamorphosis^ in its technical entomologi-
cal sense, is applied only to that succession of changes of
which such a definite pupal condition forms the middle term.

It is obvious that a metamorphosis, in this sense, is a sec-
ondar}' complication superinduced upon the direct and grad-
ual process of development exhibited by such insects as the
Cockroach ; * and that the 3Ietahola, as insects having a
metamorphosis are termed, are, so far, more differentiated
than the Ametahola^ or those w^hich have no metamorphosis.
Again, in each of these divisions it is clear that the insects
which never possess wings are less differentiated, or more
embryonic, than those which are winged. And, finally, insects
with the parts of the mouth in the condition of ordinary
gnathites are less differentiated than those in which such

*

1 Sir John Lubbock has shown that the young C7ilo>}on {EpJiemera) dimidi'
atum undergoes more than twenty ecdyses, each accompanied by a sli^rht
change of form in its passage to the adult state. (" Transactions of the Lin-
nffian Society," 1863.)

16

362 THE ANATOMY OF INVERTEBRATED ANIMALS.

gnathites are changed in form and function, or become con-
fluent.

The insects which, in this view of their morphological re-

Pxo. 102.— Campodea staphylinus, one of the Thysanura (after Lubbock).*

lations, occupy the lowest position in the group, are the Col-
lemhola and Thysanura, the Mallophaga^ and the Pedicu-
llna, inasmuch as they possess no trace of wings and undergo
no metamorphosis.

The Collemhola and Thysanura undergo no metamorpho-
sis, and are always wingless. The abdomen contains six seg-
ments only in the Collemhola [Podura, S^nynthurus, Tomo-
ceros), in which group the mouth is usually provided with
mandibles and maxillae, though these, instead of being artic-
ulated with the sides of the head, are capable of being re-
tracted into its interior.** In the genus Aiioma the mouth is
suctorial.

» "Monograph on the Collemhola and Thysanura,'^'' pi. liii.
^Ibid.^ p.' 37.

THYSANURA.— PEDICULINA. 363

The Thysanura (Lepisma, Campodea^ Japyx) resemble
the young Blattce. They have ten well-marked abdominal so-
mites {^Campodea^ Fig. 102), and the gnathites conform to
the mandibulate type. The abdomen in Machetes has a pair
of elongated cylindrical appendages attached to every seg-
ment except the first ; while Campodea and Japyx have seven
pairs of such abdominal appendages.^

The Collemhola are provided with a curious tube or sucker,
which is attached to the sternum of the first abdominal so-
mite, and gives exit to a glandular process, which secretes a
viscid matter. Most of the insects belonging to this group
possess a curiously-contrived " spring and catch " attached to
the sternal region of the penultimate or antepenultimate so-
mites of the abdomen, by the help of which they execute their
vigorous leaps.

Sir John Lubbock could find no trace of tracheae in any of
the Collemhola except Smynthurus^ though they are easily
seen in many of the Thysanura. According to the same au-
thority, Lepisma has four Malpighian tubes, while Campo-
dea^ Japyx^ and many Collemhola^ have none.

The Mallopjhaga are parasites upon mammals and birds,
on the hairs and feathers of which they feed. The head and
body are depressed, the eyes simple, the gnathites of the mas-
ticatory type. The abdomen has nine or ten visible segments.

The Pedicidina^ or Lice, subsist upon the blood of the
mammals on which they are parasites. The gnathites are
converted into a piercing and sucking apparatus. The under-
side of the head presents a soft protrusible proboscis, pro-
vided externally with minute horny hooks, and traversed by
a canal which leads into the oesophagus. The proboscis in-
closes two grooved chitinous styles, which are applied to-
gether by their concave sides ; and, within the sheath thus
formed, lie two finely-pointed chitinous set^e, which can be
moved up and down in the sheath.^

The proboscis is, in all probability, formed by the union
of the labrum with the second pair of maxillae, while the two
halves of the horny sheath are the mandibles, and the seta?,
the first maxillae. The prothorax, mesothorax, and meta-

» The myriapod Scolopendrilla has similar appendages attached to each
segment along with legs. (Lubbock, I. c.)

2 Gerstfeldt, " Ueber die Mundtheile der sangenden Insecten," 1S53.

364

THE ANATOMY OF INVERTEBRATED ANIMALS.

thorax are hardly distinguishable, and the abdomen has nine

visible segments.

Fia. 103. — Terla nigra. — A. The aquatic apterous larva. B. One of the traneitional
BtageB between this aud the perfect iusect, C. (" Kegne Animal.")

The Orthoptera (Fig. 103) and the Hemiptera (Fig. 104)
are ametabolous. The majority have two pairs of similar or
more or less dissimilar wings in the adult state, and in the
apterous forms it is probable that the wings are aborted, not
typically absent. In the Orthoptera'^ (the Termites, Cock-
roaches, Grasshoppers, Crickets, Day-flies, Dragon-flies, and

Fio. lOi.— Aphis pdargonii. Apterous agamogenotic form.

Earwiirs) the mouth is constructed upon the same plan as
that of Blatta; but the Physopoda or Thysanoptera (Thrips

» The Thysanura and the Phywpoda are often united \rith the Ortloptera
in modem classifications, while the Fphemerida and Libellulidce used to be
arranged with the JWurojpUra.

THE HEMIPTERA.

365

and Its allies), small vv'inged insects which live chiefly in flow-
ers, present a modification which is transitional to the Hemip-
teran mouth (Gerstfeldt, /. c). There is a proboscis directed
backward and formed bv the union of the labrum with the
labium, which last is provided with palps, though they are
sometimes very small. The maxillae are paipigeroas, and are
united at their bases with the labium. The mandibles are
styliform setse inclosed in the proboscis.

In the Hemiptera^'' all of which suck the blood of animals
or the juices of plants (Bugs, Plant-lice, Cicadce)^ wings
may be present or absent, and the eyes are usually compound.
The visible abdominal somites mav be reduced* to six. The
gnathites are modified to form a piercing and suctorial appa-
ratus, which is similar, in many respects, to that of the Pedi-
culina. There is a usually sharp and pointed labrum, while
the mandibles and maxillae are mere tubercles, surmounted
bv long chitinous pointed styles, of which, therefore, there are
four. The labium is usually represented by a median, jointed,
fleshy, elongated body, the anterior face of which presents a
longitudinal groove in which the mandibles and maxillae are
inclosed. Neither the maxillce nor the labium are provided
with palps.

Thus, in the series of ametabolous insects there are some
with masticatory, others with suctorial, mouths. It is by no

Fio. l05.—R>/drophilus piceus.—k. Larva. B. Pupa. C. Imago. ("Regnc Animal.")

means clear that the gnathites of the suctorial mouth of the
Hemiptera are to be regarded as modifications of masticatory

» The Mallophaga and the PedicuUna are united with the Hemiptera by
some authors.

366 THE ANATOMY OF INVERTEBRATED ANIMALS.

gnathites of the type exhibited by the Orthoptera, The
absence of palps is a very significant fact, suggesting that
the Hemipteran mouth is the extreme term of a series of
modifications for the commencement of which we must go
back to the Myriapoda.

The metabolous Coleoptera, or Beetles (Fig, 105), have
masticatory mouths of the same general type as those of the
Orthoptera / with which they are closely connected through
the Earwigs. The two constituents of the labium are, how-
ever, much more completely confluent than in the Orthoptera,
There are usually two pairs of wings, the anterior pair being
converted into stiff horny elytra ; these take no part in the
act of flight, but serve as covers to the metathoracic wdngs,
which, in the state of rest, are folded up beneath them. The
number of apparent somites of the abdomen is often much
reduced. In the metabolous Neuroptera (Ant-lions, Caddis-
flies, Scorpion-flies), in some of which the insect is more or
less active during the pupa-state, the parts of the mouth are,
for the most part, very similar to those of the Orthoptera,
In two groups of Neuroptera^ however, the mouth becomes
suctorial. Thus, in the Trichoptera, or Caddis-flies, the labrum
is elongated and grooved posteriorly ; the mandibles are
aborted, the bases of all the gnathites are united, and the
labrum is a spoon-shaped body. In the Scorpion-flies [Pa-
norpind) there is, according to Gerstfeldt, a proboscis formed
in front by the elongated clypeus and labrum, and behind by
the coalesced maxillse. The mandibles are small, and the
first maxillae much elongated. The ordinary four palps are
present.

The Neuroptera have two pairs of wings of a delicate
reticulated structure. The metathoracic wings may or may
not be folded.

What appears to be a further development of this type
of mouth is found in the Lepidoptera (Butterflies and Moths).
The labrum and the mandibles abort, and the labium is rep-
resented only by a triangular plate which bears two large
palps. On the other hand, the maxilla?, the palps of which
are always very small, are often immensely elongated and
applied together by their channeled inner faces, thus consti-
tuting a sucking proboscis (Figs. 106, 107). The wings, simi-
lar in character, and covered with minute scales, are rarely
absent. Both pairs are used in flight.

In the metabolous JDiptera (Flies and Fleas, Fig. 108)

TEE DIPTERA.

367

the mouth is constructed upon the same plan as that of the
Semiptera^ so far as the conversion of the labium into an
organ of suction is concerned ; but usually the metamorpho-
sis of the gnathites is carried still further, and the maxillas
have palps. Thus, in the Fleas, which are parasitic on mam-
mals and birds, what appears to be the labrum is an elon-
gated, slender style, which lies between the two elongated
mandibles. The first maxillag are broad triangular plates,
each with a four-jointed palp. The second maxilhie (labium)
are represented by a short median lamella, which bounds the

Fig. 106.

Fia. 107.

Fig. 106.— The head, A, and parts of the mouth, B, C, of Sphinx ligustri.— a. anten-
na; b, epicranium; c, cornea; d, clypeus posterior; e, labrum;/, mandible; g,
maxilla ; h, maxillary palpus : k, labial palpua. B, base of the maxillte with the
mandibles and labrum. C, lateral view of the same. (After Newport.)

FiG.lOl.— Vane-ssaatalnnf a.— Inner or concave surface of the apical portion of the
right maxilla : a, transverse muscles ; 6, canal ; c, papilla ; d, hooks which join
the maxillae.

mouth behind, and is provided with two long palps, which
resemble knife-blades, and are imperfectly divided into four
joints. The three somites of the thorax are distinct, and the
two hinder ones have lamellar appendages, which possibly
represent wings. The abdomen has ten somites.^

In those dipterous insects which are termed I^upiparay
which are apterous, or nearly so, and parasitic upon mam-

' See L. Landois, " Anatomic des Hundeflolies." 1866.

368

THE ANATOMY OF INYERTEBRATED ANIMALS.

mals, birds, and bees, a circular wall, or short proboscis, in-
vests the other parts of the mouth. There are, first, two
lateral, protrusible, horny plates ; secondly, an anterior and a
posterior seta ; the latter stronger, and grooved longitudinally
in front. Between these is a single tine seta. Gerstfeldt
considers that the last answers to the hypopharynx ; the
second pair, to the labrum and the second maxillae ; the first
pair, to the first maxillo3 ; and that there are no mandibles.

Fig. 103.—S]/rphus ribesii.—K. Larva. B. Pupa. C. Imago. (" Regne Animal.")

The ordinnTy Dlptera, which possess one pair of functional
wings attached to the mesothorax, resemble the JSemiptera

Pig. i09.—EnstcUisfloreus.—rl, front of the head; e, labrum; /", mandible ; ff. maxilla
and palpus ; z, labium ; ?:*, c-xtremity of the labiurn separately and more masiii-
fied ; **i, inner surface of the para?lo«sse ; ***i. the rows of hairs on the inner
surface ; I, the ligula ; m, the cardo and submentura. (After Newport.)

in possessing a usually fleshy proboscis, often tumid at its
extremity, which is formed by the confluent second maxillae.

THE DIPTERA.

369

As in Hemiptera^ also, the labrum is a more or less elongated
pointed plate, and the mandibles and maxillae are usually ter-
minated by cliitinous cutting set^e (Fig. 109). But the bases
of these parts are constantly united together ; there is a pair
of maxillary palpi, and often a median, more or less styliform
structure, usually considered to be the hypopharynx. It
seems doubtful, however, whether this maj' not be formed by
the coalesced terminations of the maxillae. In the common
House-fly, the labrum, mandibles, and maxillce coalesce at

Fig. 110. — Upper flf::nre. Section of the head of Bomhu^. ft, ocellus ; c, antenna ; rf,
clypeus; e, labrum; /, mandible; g. epipharynx ; A, maxilla; ?', cardo ; j, k, I,
Bubmentum and nientum ; m, m'. labial palpus ; n. para^rlossa ; o, lingua or
median process of the ligula; w, occipital foramen; 1, 2, sclerites of the hypo-
pharynx.

Left lower flgure. Terminal portion of a maxilla.

Middle lower figure. Epipharynx and hypophamyx majmified ; 1, 2, sclerites of
the hypopharynx| 3, cut end of tl>e oesophaniis; 4. 5, sclerites In the wall of the
oesophasrus and sides of the mouth; 6, lip-like projection of the hypopharynx;
(f. epipharynx.

Right lower figure, a, quadrate sclprite connected by a triangnlar piece with c, one
of the lances of the sting; 5, duct of the poison-eland ; f, grooved median piece
in which the lances play; h, one of the lateral setose palpiform sLeath-pieces ; g,
genital aperture.

their origins to constitute the base of the proboscis, which is
mainly formed by the confluent second maxillae. Its longitu-

370 THE ANATOMY OF INVERTEBRATED ANIMALS.

dinal grooved anterior face is overhung by the elongated sty-
liform labrum. The gnathites here exhibit almost the extreme
modification of the piercing and sucking type of mouth.

Finally, the metabolous Hymenoptera., with, usually, two
pairs of reticulated scaleless wings, present a series of modifi-
cations from the essentially masticatory mouth of the Ants to
the partly masticatory and partly suctorial, or rather lapping,
mouth, such as is met with in the Bees. In the latter (Fig.
110) the labrum is small ; beneath it, a median fleshy lobe —
the epiphary7ix — overhangs the minute aperture of the mouth.
The mandibles are strong, with wide, almost spoon-shaped,
extremities. The part of the maxilla which appears to an-
swer to the lacinia in Blatta is shaped like a knife-blade,
and folds upon the stout stipes like a clasp-knife in its
handle. The short and almost rudimentary palp is attached
to the extremity of the stipes. The cardines are long and
slender, and give rise to a hinge-joint, whereby the maxillae
and labium can be folded back, like a carriage-step, under the
head. The mentum is large, the labial palps long and slen-
der ; there are two large paraglossge, and, between them, a
median, annulated, setose, cylindrical organ proceeds, which
either represents the lingua, or is an independent prolonga-
tion of the ligula. Functionally, this organ is a tongue, and
enables the bee to lap up the honey on which it feeds. The
mandibles and maxillae are employed as cutting and model-
ing implements, but appear to have little or nothing to do
with mastication, properly so called.

The gnathites and the mouth are abortive in some insects,
as the Day-flies, which take no food in the adult condition.
The development of the diff'erent divisions of the alimentary
canal varies greatly. Salivary glands are very generally
present. In many suctorial insects, the ingluvies is a sac
opening by a long duct into the gullet ; a distinct proven-
triculus, provided with chitinous ridges, may be present or
absent. The ventriculus appears to be always devoid of an
inner cuticula. It may be devoid of creca or beset with short
cseca throughout its whole extent. The number of the Mal-
pighian tubes, which are sometimes branched, varies from two
to a multitude. In many cases they have been found to con-
tain uric acid ; but no biliary matter has yet been proved to
exist in them. Anal glands are frequentlj' appended to the
termination of the rectum, and may secrete an acrid or stink-
ing fluid.

OVIPOSITORS AXD STINGS. 371

In some larvce {MyrmecoleOy Dytisciis) there is no proper
median oral aperture, but canals which open on the extremi-
ties of the mandibles lead into the oesophagus. The alimen-
tary canal has no posterior opening in the larvae of many
Symenoptera, of Myrmecoleo, and of the Pupipara, The
salivary glands secrete the silken material in which the larvas
of the Lepidoptera invest themselves ; while, in Myrmecoleo
and the Hemerohidoe^ it is the rectum which furnishes the silk.

The poison of the Hymenoptera is a fluid strongly im-
pregnated with formic acid, which is secreted by a special
gland and poured into a reservoir connected with the sling.

In many winged insects both pairs of wings are developed
and take equal shares in flight {Hymenoptera, Lepidoptera^
Neuroptera). In the Goleoptera, the anterior pair are con-
verted into horny wing-covers {elytra), and the posterior pair,
much larger than the anterior and folded up under them when
the insect is at rest, subserve flight. In the Dlptera the pos-
terior wings are represented only by short processes, the hal-
ter es. In the Strepsiptera, on the other hand, it is the ante-
rior pair of wings which abort. In all orders of winged in-
sects, individual cases of complete abortion of the wings oc-
cur either in the female alone, or in both sexes.

The posterior abdominal somites often undergo extensive
modifications ; they may be small and retracted within the
anterior somites, or they may even become more or less com-
pletely abortive. In many insects, processes of the somites
in the genital region of the females, which answer to the go-
napophyses of JBlatta, are converted into organs which assist
in the deposition of the eggs, and are termed ovipositors.
The saws of the Saw-flies and the stings of other Hymenop-
tera are to be regarded as specially modified ovipositors. The
laborious and thoughtful investigations of Lacaze-Duthiers *
led him to the conclusion that all these organs are constructed
upon the same plan ; that they are developed from that so-
mite of the abdomen which lies immediately behind the open-
ing of the vulva ; that this opening is always situated be-
tween the eighth and the ninth somite ; and is therefore
separated by three somites (the ninth, tenth, and eleventh)
from the anus.

According to Lacaze-Duthiers, in those insects which are
provided v/ith an ovipositor, saw, or sting, the ninth somite

» " Recherclies sur Tarmure grenitale femelle des lusectes." (" Annales de3
Sciences Naturelles," 1849-1853.)

372 THE ANATOMY OF INYERTEBRATED ANIMALS.

always consists of a single median tergal sclerite, to the in-
ferior angles of which are connected two small more or less
triangular pieces, each of which carries a long styliform ap-
pendage. There is a single median sclerite, which is the most
important part of the boring apparatus ; two small sclerites
are united with the lateral angles of this piece, and there are
two other elongated sclerites which constitute a valvular
sheath. Thus, according to Lacaze-Duthiers's view, in the
sting of Bombus (Fig. 94) h is one of the elongated lateral
sternal sclerites, which with its fellow forms a sheath for the
rest of the apparatus ; f is the median sternal sclerite ; it is
pointed and grooved on its sternal surface ; while c, one of
the lances, is a process of the tergal half of the somite. Each
lance is sharp and slender, and its tergal edge fits upon the
margin of the groove of the median style, in such a manner
as to be able to slide backward and forward upon it. The
sternal edges of the two lances meet in the middle line, and,
together with the median sternal piece, inclose a canal which
serves to convey the secretion of the poison-gland into the
wound made by the sting. In the operation of stinging, the
median piece serves as a sort of " director" for the two lances.
However, recent investigations into the development of
stings and ovipositors,' e. g., the sting of the Hive-bee, and
of the Wasp and the ovipositor of an Ichneumon-fly ( Cryptiis
migrator)^ show that while the median grooved piece and the
two sheath-pieces arise from papillse developed upon the
sternal surface of the ninth abdominal somite of the larva,
the lances are the result of the metamorphosis of papillse
seated on the sternal surface of the eighth somite ; and these
papillae are so similar to those from which the limbs are de-
veloped, that it becomes (to say the least) probable that they
represent true appendages of the somites to which they are
attached, rather than mere modifications of the sclerites of
the body-wall, as Lacaze-Duthiers supposed them to be. In
like manner, the examination of the development of the ovi-
positor of Locusta viridissima has proved that, of the three
pieces of which each half of it is composed, two are developed
from the sternum of the ninth and one from that of the eighth
somite. But the two median pieces of the ninth somite do

* Kraepclin, " Untersuchungen fiber den Ban, Mechanismus tind Entwicke-
luncrsgeschiohte des Stachels der biencTuirtiifen Thiere " {Zeitschinft fur wiss.
Zoolngie^ 1873) ; and Dewitz, " [Jeber Ban und Entwiokelunsf des Stachels und
der Legescheide " {Zeitschrift fiir wiss. Zoologie^ 1875). 8ee also the observa-
tions of Packard, " On the Development and Position of the Hymenoptera,"
1866.

THE COPULATORY ORGANS OF INSECTS. 373

not unite together to form a single piece grooved below, as
in the hymenopterous sting or ovipositor. And observations
Tv^hich I have made on the development of the gonapophyses
of Blatta lead me to the conclusion that the posterior bitid
pair are developed from the ninth and the anterior curved
pair from the eighth somite. In this case the latter will be
the homologue of the lances of the Bee-sting.

Thus it would appear that, while there can be no doubt as
to the general unity of plan of ovipositors and stings, the
view of Lacaze-Duthiers must be modified. It must be ad-
mitted that these apparatuses appertain to the eighth and
ninth somites, and not to the ninth alone ; and that there is
much reason to suspect that their chief constituent parts are
modified limbs.

The male copulatory organs * are often very complicated,
and their homologies have not yet been fully determined.
Kraepelin {I. c), who has examined the development of these
parts in the Drone, and the modifications found in hermaphro-
dite Bees, is led to the conclusion that they are developed
from the eighth and ninth somites of the abdomen, and there-
fore are the homologues of the parts of the sting in the fe-
male. In the male Jylatta, however, it is obvious that the
male copulatory apparatus belongs to a more posterior somite
than that upon which the female gonapophj^ses are developed.

The heart usually has the form of a fattened tube, closed at
its posterior end, but, in front, continued into the aorta, which
may be traced as far as the cerebral ganglia, and appears to
give oif no branches. The sides of the tube present slit-like
openings (ostia)^ which vary in number from two to nine pairs ;
and, when there are several pairs, each pair answers to a so-
mite of the abdomen. The margins of the ostia may be sim-
ple, or may be produced inward into folds, which play the part
of valves. Muscular or ligamentous fibres may extend from
the hypodermis to the dorsal aspect of the heart, and serve to
suspend it in place.

The alary muscles, which in most insects are fan-shaped,
and lie in pairs, opposite one another, on each side of the
heart, either unite in the middle line, or are inserted into a
sort of fascia, on the sternal aspect of the heart, to which
organ they are not directly attached.

1 The male Lihellnlidoe, possess a peculiar copulatory apparatus developed
upon the sternum of the second abdominal somite. The genital aperture has
the ordinan^ position, and hence, before copulation, the male has to bend the
extremity of his abdomen upward in order to load this apparatus with sperma-
tozoa.

374 THE ANATOMY OF INVERTEBRATED ANIMALS.

The septum between the pericardial cavity and the gen-
eral cavity of the abdomen thus formed is termed by Graber *
the pericardial se/)tum. From their anatomical relations,
therefore, the alary muscles can have nothing to do with the
diastole of the heart, the pulsations of which, indeed, go on
just as well when the alary muscles are cut through. Graber
throws out the very probable suggestion that the contraction
of the alary muscles causes the pericardial septum to move
toward the axis of the body, and, by thus enlarging the cavity
of the pericardium, facilitates the flow of blood to the ostia of
the heart. The same investigator ascribes a special respira-
tory function to the abundant tracheae which are distributed
to the walls of the pericardium, and which, undoubtedly, must
tend to facilitate the aeration of the returning blood.

In many insects, a septum, provided with transverse mus-
cles, overlies the abdominal nerve-cord and separates a ven-
tral blood sinus, in which the cord lies, from the abdominal
cavity. The sinus is open in front, and, as the muscles of the
septum contract rhythmically from before backward, they
tend to drive the blood which enters it to the posterior end
of the body.

In the respiratory system of insects the number of stig-
mata is observed to vary from one to ten pairs. As a rule,
none are found in the head,^ or between the head and the
first thoracic somite, and they are usually absent from the
terminal somites of the abdomen. A very common number
is nine pairs; the first being situated between the mesothorax
and the metathorax, and the rest between the followinsr somites.
Only two pairs of stigmata are found in the Lihellulidce and
JEphemeridae^ and they are seated upon the thorax. In N'epa
and Banatra there is only one pair of abdominal stigmata,
in addition to those in the thorax, and in the larvae of Tipu-
lidcB and of ITi/drophilus the stigmata are reduced to one
terminal abdominal pair. The stigmatic openings are usually
situated upon the sides of the abdomen, but in some Coleojytera
(e. g., Dytiscus) they are dorsal, and in many Hemiptera they
are situated on the ventral aspect of that region of the body.
Either the lips of the stigmatic aperture itself, or the walls
of the tracheal trunk which arises from it, are so disposed as

' "Ueber den propulsationischen Apparat der Insecten" {Zdfscliriftfur
wiss. Zooloqie, 1873), and " Ueber den pulsirenden Bauchsinus der Insecten "
{ibid., 1876).

3 Sir John Lubbock found the two spiracles of Smt/nthurus to be situated
on the under side of the head, immediately below the antennae.

THE RESPIRATORY ORGANS OF INSECTS. 375

to constitute an occlusor apparatus, provided with a muscle,
by the contraction of which communication with the external
air can be cut off. This occlusor apparatus, long ago de-
scribed in certain insects by Strauss-Durckheim, Newport,
Burmeister, Siebold, and others, has recently been specially
investigated by Landois and Thelen,^ who describe it as
usually consisting of four essential parts: the bow (Ver-
schlussbiigel), the Up (Verschlussband), the ^e^'er (Verschluss-
hebel), and the muscle. The bow is a thickening of one-half
of the circumference of the chitinous lining. The lip is formed
by the other half of the circumference, and the lever is a
chitinous process connected with one end of the bow, or with
the lip. When the lever is single, the muscle which is at-
tached to it passes over the lip and is inserted into the oppo-
site end of the bow. When it contracts, it therefore presses
the lip against the bow. When two levers are present, they
are attached to opposite ends of the lip and bow, and the
muscle extends between their extremities. The effect of its
contraction is to thrust the free edge of the lip against the
bow.

The tracheal trunk which arises from a stigma may ramify
without communicating with the rest ; but, usually, the tracheas
which proceed from each stigma enter into more or less exten-
sive anastomoses. Very commonly the main trunks of each
side give off wide anastomotic brandies, which unite and form
a longitudinal trunk on each side of the body, while transverse
trunks often connect the main tracheae of opposite sides.

In many insects, especially those which possess great
powers of flight, more or fewer of the trachese become dilated
into sacs, in which the spiral marking of the chitinous lining
is interrupted or disappears. In Bees and Flies, a vast air-sac
is thus developed, on each side of the abdomen, from the
longitudinal anastomotic trunk.

The aquatic larvas of many Orthoptera (Ephemeridmy
Agrion, Calopteryx) and ISfeuroptera^ and of some Diptera,
Lepidoptera^ and Coleoptera^ though provided with a fully-
developed tracheal system, possess no stigmata. The somites
of the abdomen or of the thorax are, however, provided with
delicate foliaceous or filamentous processes, into which
branches of the trachese enter and ramify. The air contained
in these tracheas is therefore separated from that dissolved in

» " Der Stierraenverschluss bei den Insecten." {Zeitschrift fur wissen,'
schaftliche Zoologie, 1867.)

376 THE ANATOMY OF INYERTEBRATED ANIMALS.

the water only by a very thin layer of integumentary tissue,
and an exchange of gaseous constituents between the two
readily takes place. These are often called hranchice^ but
they are obviously of a totally dijfferent nature from true
branchiae. The larvae of some Dragon-flies (JLibellula and
^scJinci) present yet another form of respiratory organ. Al-
though they possess a pair of thoracic stigmata, these appear
to have little or no functional importance, but respiration is
eifected by pumping water into and out of the rectum. The
walls of the latter are produced into six double scries of
lamellae, in the interior of which tracheae are abundantly dis-
tributed, and which play the same part as the tracheal bran-
chise just mentioned. These rectal respiratory organs, in
fact, appear to be a complicated form of the so-called "rectal
glands," which are so generallj'' met with in insects.

The chief agent of the movements of expiration and in-
spiration in insects is the abdomen, the capacity of which
may be diminished by the approximation of its terga and
sterna, and the shortening of its length by the retraction of
its posterior into its anterior somites ; while it may be en-
larged by movements in the opposite directions. When the
cavity is enlarged, air rushes in at the stigmata, and when it
is diminished, if the stigmata are open, expiration occurs ;
but, if the stigmata are shut, the effect of the expiratory act
must be to drive the air into the ultimate ramifications of the
tracheae. The movements of inspiration and expiration vary
in rapidity with the condition of the insect. In the Bee,
Newport observed that in the state of rest they were as few
as forty, but that they rose to one hundred and twenty with
muscular exertion.

The air-sacs doubtless assist flight by the diminution of
the specific gravity of the insect, which follows upon their
distention.

The sounds produced by insects ' are, in a great propor-
tion of cases, effected by the friction of hard parts of the in-
tegument one against the other. Thus the Grasshopper rubs
the femur of the hind leg against a ridge on the anterior
wing, and the chirp of the Crickets and Locusts is produced
by the friction of the elytra. The parts which thus rub to-
gether are provided with serrations and ridges, which have a
constant and characteristic disposition. The longicorn Bee-

1 Se* Landois, " Die Ton- und Stiram-Apparate der Insecten." {ZeiUclirifi
fur wise. Zoologie^ 1867.)

TEE SOUNDS PRODUCED BY INSECTS. 377

ties produce a sound by the friction of the tergum of the pro-
thorax upon a process of tliat of the mesothorax, and the
Dung-beetles by rubbing the coksq of the hind-legs against
the hinder edg-e of the third abdominal sternum. Further,
sounds are necessarily produced by the extremely rapid vibra-
tion of the wings, which characterizes the flight of many in-
sects. Landois, however, found that the thorax of a Blue-
bottle fly continued to buzz after the separation of the head,
the wings, the legs, and the abdomen. The separation of the
halteres weakened the sound but slightly. The acoustic ap-
paratus, in fact, lies in the immediate neighborhood of the
thoracic stigmata. The main trunk of the trachese dilates
into a hemispherical sac, which opens externally by the stig-
matic orifice. The sac presents a hooplike thickening, to
which are attached free chitinous folds or processes, and it is
to the vibration of these that Landois ascribes the sound.
The vocal organ of the Fly would thus appear to be a modi-
fication of the occlusor apparatus of the stigmata, just as the
oreran of voice of mammals is a modification of the occlusor
apparatus of their respiratory opening.

In the Cicadce the vocal organs are, according to Lan-
dois, the posterior thoracic stigmata. These open into cham-
bers, in the walls of which tense membranes are so disposed
as to intensify the sound by their resonance.

As in the Crustacea^ so in insects, the central nervous
system varies very much in the extent to which its compo-
nent ganglia are united together. In most Orthoptera and
Neuroptera and in many Coleoptera^ the thoracic and abdom-
inal ganglia remain distinct, and are united by double com-
missures as in JBlatta. In the Lepidoptera^ the thoracic gan-
glia have coalesced into two masses, united by double com-
missures ; while in the abdomen there are five ganglia, with
single or partially separated commissural cords. The concen-
tration goes furthest in some Diptera and in the Strepsiptera,
in which the thoracic and abdominal ganglia are fused into a
common mass.

A system of stomato-gastric nerves, similar in its general
arrangement to that of JBlatta, is very generally present.

A special system of nerves, termed respiratory or trans-
verse, is found in very many insects, both in the larval and in
the perfect condition. The principal nerves of this system
are arranged in pairs on the sternal aspect of the body, and
their outer extremities anastomose with branches of the or-
dinary peripheral nerves, and are distributed to the muscles

378 THE ANATOMY OF INVERTEBRATED ANIMALS.

of the stigmata. Their inner ends unite into a plexus, which
lies over the interval l)et\veen two of the ganglia of the cen-
tral nervous cord, and they are connected by longitudinal
cords with one another, and with these ganglia.

In insects, as in other arthropods, the branches of the
nerves which are distributed to the integument, and especially
those which pass to the bases of the larger or smaller seta3
with which the integument is provided, frequently end in
minute ganglia. Hensen has shown that in the Crustacea
similar set^e in all probability have an auditory function ; and
Leydig, Hicks, Lespes, Landois, and others, have ascribed
functions of special sensation to these structures in insects.
But whether these setae, on the antennsB or elsewhere, sub-
serve either hearing or smell, is still very doubtful ; and the
only organs which can safely be regarded as auditory in in-
sects are those which occur in Grasshoppers (Acridida^),
Crickets (Achetidce), and Locusts (Lociistidce), and which
w^ere first accurately described by Von Siebold.^ Recently,
they have been studied by Leydig, Hensen, Ranke,^ and Os-
car Schmidt,^ but it must be confessed that much obscurity
still hangs over their minute structure.

In the Acrididce, the chitinous cuticula of the metathorax
presents on each side, above the articulation of the last pair
of legs, a thin tympaniform membranous space surrounded
by a raised rim. On its inner face, the cuticular layer of the
tympaniform membrane is produced into two processes, one
of which is a slender stem ending in a hollow triangular dila-
tation. A large tracheal vesicle lies over the tympanic mem-
brane, and between its wall and the latter, a nerve derived
from the metathoracic ganglion, passes to the region occupied
by the processes, and there enlarges into a ganglion, the outer
face of w^hich, beset with numerous glassy rods arranged side
by side, is in contact with the tympaniform membrane. A
nerve arising from this ganglion passes along a groove to the
" stem " and ends in a ganglion in its dilatation. From this
ganglion certain fine filaments proceed.

In the Achetidm and Locustidoe the tibiae of the fore-legs
present similar tympaniform membranes which are easily seen
in the common Cricket, but, in other forms, become hidden

' " Archiv fur Naturgeschichte," 1864.

'"Beitrage zu der Lehre von den Uebergangs-Sinnesorganen." {ZeiU
schriftfur wiss. Zoologie^ 1875.)

3'Schmidt, "Die GenOrorgane der Heuschrecken." ("Archiv fur mikr.
Anatomic," 1875.)

THE PHOTOGENIC ORGAXS OF INSECTS. 3:^9

by the development over tliem of folds of the cuticle of the
adjacent region of the limb. Tw^o spacious tracheal sacs oc-
cupy the greater part of the cavity of the tibia, and a large
nerve ends in a ganglion in the remaining space. Upon this
ganglion a series of peculiar short rod-like bodies are set.

The compound eyes of insects differ only in detail from
those of the Crustacea.

la the ocelli, or so-called simple eyes, a sclerotic, a cornea,
a lens, a vitreous humor, and a choroid coat, have been dis-
tinguished, and the whole organ has been compared to the
vertebrate eye. But the "lens " appears to be always a mere
thickeninsr of the cuticle which constitutes the cornea, and
the so-called "vitreous humor" is partially or wholly made up
of crystalline cones analogous to those which are found in
the compound eye. In this respect the ocellus of the insect
resembles the simple eye in Arachnida and Crustacea}

Many insects, as the Glow-worm and Lantern-flies, are re-
markable for their power of emitting light.

iVccording to Schulze,^ the males of Lampyris splendidula
possess two photogenic organs, which lie on the sternal aspects
of the penultimate and antepenultimate abdominal somites.
Each is a thin, whitish plate, one face of which is in contact
with the transparent chitinous cuticula, while the other is in
relation with the abdominal nerve-cord and the viscera. The
sternal gives out much more light than the tergal face. The
photogenic plate is distinguishable into two layers, one occu-
pying its sternal and the other its tergal half. The former is
yellowish and transparent, the latter white and opaque, in
consequence of the multitude of strongly refracting granules
which it contains. Trachese and nerves enter the tergal layer,
and for the most part traverse it to terminate in the sternal
layer, which alone is luminous. Each layer is composed of
polygonal nucleated cells. The granules are doubly refrac-
tive, contain uric acid, and probably consist of urate of ammo-
nia (Kolliker). Hence the cells of the layer which contain
them are termed by Schulze the " urate cells," while he calls
the others the "parenchyma cells." The branches of the
tracheas which ramify among the parenchyma cells end, like
those of other parts of the body, in stellate nucleated cor-

» Leydis:, "Das Au^e der Gliederttiiere," 1864. Landois, "Das Raupen-
auge" {ZeltscJirift far wiss. Zoologk, IS'j'J), and " Zur Entwickelungsgeschichte
der facettirten Augen von Tenebris molitor'''' {ihii., 1867).

* " Zur Kenntiiiss der Leuchtorarane von Lampyris splendidula.'^'' (" Archiv
fur mikr. Anatomie," 1855.) See^aXso KoUiker, " Wiirzburg Phys. Med. Ge-
sellschaft," 1857.

380 THE ANATOMY OF INVERTEBRATED ANIMALS.

puscles, one process of the corpuscle passing into a ramifica-
tion of the trachea. Schulze is inclined to think that the
other processes end in parenchyma cells.

The nerves of the photogenic plates are derived from the
last abdominal ganglion ; they branch out between the paren-
chyma cells into finer and finer branches, which eventually
escape observation.

The female reproductive organs of insects consist of the
ovarian tubes, or ovarioles, with their so-called peritoneal in-
vestments, and of the oviducts, which unite into a vagina ;
while a spermatheca, and, generally, accessory glands open
into, or close to, the vagina.

The ovarioles may be few or very numerous. Each con-
sists of an external structureless memhrana propria^ within
which lies a solid columnar mass composed of cells. The an-
terior, usually tapering, end of this ovarian mass is composed
of protoplasmic substance in which nuclei are imbedded, but
in which the contours of the cells which they indicate are not
distinguishable. Further back, some of these nuclei enlarge,
become surrounded by an accumulation of protoplasm, and
constitute the primitive ova. Each primitive ovum is sepa-
rated from its fellow by a layer of nucleated protoplasm
which thus forms a capsule around it. In some insects, such
as Blatta, the capsule is hardly distinguishable in those ova
which lie between the smallest and those of middling size,
which follow the former in order from before backward. But,
in the larger ova which succeed these, the cells of the ovicap-
sule rapidly enlarge in a direction perpendicular to the sur-
face of the ovum, and constitute a very well-marked epithelial
layer. I am inclined to believe that, for some time, an addi-
tion is made to the vitellus of the e^g by these epithelial
cells, and that they, in fact, play the part of vitelligenous
cells. But however this may be, before long, a delicate struct-
ureless lamella appears on the surface of the vitellus and
incloses the egg as a vitelline membrane. The epithelial
cells of the ovicapsule next secrete from their surface a thicker,
often ornamented, layer of chitinous substance, which consti-
tutes the chorion, and the egg is complete.

The ovarian mass, therefore, as Waldeyer has justly pointed
out, corresponds with one of the epithelial tubes of the ovary
of a vertebrated animal, and the ovicapsules answer to Graa-
fian follicles.

In some insects, as Aphis, the indifferent tissue of the an-
terior end of the ovarioles gives rise not only to ova and ovi-

THE OVARIA OF INSECTS. 381

capsular epithelium, but to large viteUigenous cells. These
stay in the dilated anterior chamber of the ovarian tube.
But each ovum is originally connected by continuity of sub-
stance with one of these cells, and the pedicle of connection
may be traced even to the second and third ovum. It seems
probable, therefore, that these " viteUigenous cells," for some
time, supply material to the growing ova.

In most insects, similar viteUigenous cells are found ; but
they are situated at the anterior end of each ovicapsule, so
that, as the column of ovicapsules lengthens by the addition
of new ovicapsules to its anterior end, the viteUigenous cells
are interposed between every two ova. The vitelline mem-
brane and the chorion first invest the posterior extremity and
the sides of the ovum ; and, for some time, leave an opening
at the end of the ovum adjacent to the viteUigenous cells.
This opening is usually only partially closed, and what re-
mains of it constitutes the aperture or apertures, termed the
7nicropyle^ through which the spermatozoa enter when the
e^^ is fecundated. The viteUigenous cells usually remain
outside the ovum, and eventually undergo degeneration ; but,
in many Dijjtera, the)' become inclosed within the coats of the
ovum and their substance is merged in that of the vitellus.

Dr. A. Brandt has proposed the term panoistic for ovaries
cf the first mode, and meroistic for those of the second and
third modes of development of the ova here described. So
far as is at present known, only the Orthoptera and the Puli-
cidm possess panoistic ovaria.

The peritoneal coat of the ovarioles is a cellular struct-
ure, containing many trachese and, frequently, muscular fibres.
It is usually extended beyond the anterior end of each ovari-
ole into a filamentous process, which, after uniting with those
of the other ovarioles of the same side, is continued into the
pericardial tissue. At its opposite extremity it passes into
the walls of the oviduct, which are muscular and are lined by
an epithelium.

The development of the ovaria has been traced in Diptera
and Lepidoptera. Each ovary is, at first, a rounded mass of
indifferent tissue, from which a filiform prolongation is given
off backward ; this has not been traced into connection with
any other organ, and appears to terminate by a free end.
The mode of origin of this rudimentary, or primary, ovarium
is unknown, but the first step toward the formation of the
genital organs is the separation of the peripheral indifferent
tissue from the central portion, and the division of the latter

382 THE ANATOMY OF INVERTEBRATED ANIMALS.

into as many elongated solid cellular bodies as ovarioles are
to be formed. The peripheral cells become the peritoneal
layer. Each cellular rudiment surrounds itself with a struct-
ureless membrane, and then elongates into an ovariole, some
of the cells filling the posterior end of which then becomes
diflferentiated into the first primary ovum and its capsule, with
or without vitelligenous cells. The contents of each ovariole
must therefore be regarded as a column of generative cells,
which instead of burrowing in the stroma of an ovary, and
becoming divided into ovisacs, as in a vertebrated animal,
grows straight backward, and, as it grows, becomes divided
into ovisacs, of which the oldest and most advanced is the
hindermost.

Nothing is certainly known respecting the origin of the
vagina or the oviducts, though it may be suspected that the
posterior prolongations of the primary ovaries give rise to the
latter.

The development of the testes takes place in the same
manner as that of the ovaries, but the contents of the testic-
ular tubes become converted into spermatozoa. The origin
of the vasa deferentia is unknown.^

In most iasects, the vitellus undergoes partial yelk-divis-
ion. In some Podiiridoe^ however, complete division has been
observed. The development of the blastoderm takes place in
the same way as in other Arthropods, and the cephalic end of
the embryo terminates in two procephalic lobes. In many
insects, the periphery of the blastoderm, external to the lon-
gitudinal thickening which giv^es rise to the sternal region of
the body, and which may be termed the sternal hand (" Keim-
streif " of the German embryologists), gives off a lamina
which grows inward over the sternal face of the embryo,
and eventually forms a complete investment thereto. The
lamina may be formed by a single layer of cells, or it may,

» The account ffiven above of tlie structure of the ovarian tubes in Blatta
and Aphis is based on ray own observations, which are in pretty close accord-
ance with those of A. Brandt. " Ueber die Eirohren der Blatta {Periplaneta)
orientalist' ("Mem, de I'Acad. St.-Petersbourg," tome xxi., 1874). The liter-
ature of the subject is somewhat extensive. See especially Leydig, " Der
Eierstock und die Samentasche der Insecten " ("Nova Acta," xxxiii., 18B7);
Lubbock, " The Ova and Pseudova of Insects " (" Phil. Trans.," 1858) ; Weis-
mann, " Die_ nachembryonale Entwickelung der Musciden" {Zeitschrift fiir
wiss. Zoologie^ xiv.) ; Bessels, " Entwickelung der Sexualdriisen bei den
Lepidopteren " {Zeitschrift fiir wiss. Zoologie^ 1857); and Von Siebold,
*' Beitrage zur Parthenogenesis der Arthropodeti," 1871. The various forms
of the micropyle and the structure of the chorion are dealt with by Leuckart,
in his elaborate memoir, " Ueber die Micropyle und den feineren Bau der
Schalenhaut bei den Insekteneiem " (" MuUer's Archiv," 1855).

AGAMOGENESIS IN INSECTS. 383

from the first, be a fold of the blastoderm and thus consist of
two layers, the inner of which is continuous with the sternal
band, and the outer with the blastoderm which invests the
tergal surface of the vitellus. In the latter case, it becomes
strictly comparable to the amnion of a vertebrated animal ;
and, when the folds have united in the middle line, the invest-
ment in question is distinguishable into an outer membrane,
which answers to the lainina serosa,^ and an inner, which cor-
responds with the amnion proper of the vertebrate embryo.
In some cases, the vitelline substance fills up the interval be-
tween the lamina serosa and the amnion, so that the sternal
band and the latter form a sac plunged into the interior of
the yelk.

The development of a more or less complete amniotic in-
vestment has been observed in Orthoptera {Libelluld)^ Cole-
optera, Semiptera, Hymenoptera^ Lepidoptera^ and Dip-
tera, but it does not appear to be universal.

Agamogenesis is of frequent occurrence among insects,
and occurs under two extreme forms ; in the one, the parent
is a perfect female, while the germs have all the morpho-
logical characters of eggs, and to this the term parthenogene-
sis ought to be restricted.* In the other the parent has in-
complete female genitalia, and the germs have not the ordi-
nary characters of insect eggs.

In Coccus (jOecaniwn) hesperidiim^ in Chermes ahietis
and pini, no males have been observed; but the perfect
females produce ova, out of which only females proceed. It
is probable that many species of gall insects {Cynips) are in
the same predicament.

The unimpregnated, apterous, caterpillar-like females of
the Lepidopterous genera Psyche and Solenobia lay eggs
out of which only females issue. The males occur but rarely
and locally, and, from the impregnated eggs, males and
females issue in about equal numbers.

Leuckart discovered that the ovaries of so-called neuters
among wasps, hornets, humble-bees, and ants, often contain
more or less well-developed eggs, and that in the wasps and
humble-bees such eggs are laid and develop young, the sex
of which was not ascertained. Von Siebold has observed that
the neuters of Polistes gcdlica are distinguished from the per-

1 The excellent " Beltrasfe zur Parthenogenesis " (1871) of Von Siebold is
my chief authority for the statements in the text respecting Agamogenesis in
Insects.

384 THE ANATOMY OF INVERTEBRATED AXIMALS.

feet fertilizable female, by little more than their smaller size,
and that they possess completely developed female organs.
These neuters, or rather, small females, laid eggs which de-
veloped, and gave rise only to male Polistes. The unim-
pregnated females of a Saw-fly, Keinatus ventricosus (the
larvae of which are known as gooseberry caterpillars), regu-
larly lay eggs, which develop and produce male offspring.

The terms arrenotolcous and thelytokous have been pro-
posed by Leuckart and Von Siebold to denote those par-
thenogenetic females which produce male and female young
respectively.

In the case of the Hive-bee, it has been ascertained that
the queen either impregnates, or does not impregnate, the
eggs when they are laid. The spermatheca, in which the
spermatic fluid, introduced by the single act of copulation
which takes place, is contained, contracts as the eggs pass
along the vagina, in the former case, and remains passive in
the latter. The unimpregnated eggs give rise to males or
drones ; the impregnated eggs to females, which become
neuters with imperfect reproductive organs, or queers, with
perfect organs, according to the nutriment which they re-
ceive.

In the Ajjhides, ova deposited by the impregnated females
in the autumn are hatched in the spring, and give rise to
forms which are very generally wingless, and bring forth
living young. These may be either winged or wingless, and
are also viviparous. The number of successive viviparous
broods thus produced has no certain limit, but, so far as our
present knowledge goes, is controlled only by temperature,
and by the supply of food. Aphides kept in a warm room
and well supplied with nourishment have continued to propa-
gate viviparously for four years.

On the setting in of cold weather, or, apparently, on the
failure of nourishment alone, in some cases, males and females
are produced by the viviparous forms. The males may pos-
sess wings, or may be devoid of them. The females appear
invariably to be apterous. Copulation takes place and tho
eggs are laid.

Sometimes viviparous forms coexist w^ith the male or fe-
male forms, and some viviparous Aphides are known to hi-
bernate.^

» Huxley, " On the Agamic Reproduction and Morphology of Aphis."
("Linntean Transactions." 1857.)

The papers of M. Balbiani (" Ann. des Sciences Naturelles," 1869, 1870, and

AGAMOGEJfESIS IN APHIDES. 385

The viviparous forms diflfer essentially from the oviparous
forms in the structure of their reproductive organs. They pos-
sess neither spermatheCcTB nor coileterial glands, both of which,
as Von Siebold first demonstrated, are present in the females.
The young are developed within organs which resemble the
ovarioles of the true females in their disposition and may be
termed pseudovaries. The terminal or anterior chamber of
each pseudovarian tube is lined by an epithelium, which in-
closes a number of nucleated cells. One of the hindermost
of these cells enlarges and becomes detached from the rest as
a pseudovum. It then divides and gives rise to a cellular
mass, distinguishable into a peripheral layer of clear cells and
a central more granular substance, which becomes surrounded
by a structureless cuticula. It is this cellular mass which
gradually becomes fashioned into the body of a larval Ap)his.
A portion of the cells of which it is composed becomes con-
verted into a pseudovarium, and the development of new
pseudova commences before the young leaves the body of its
parent. It is obvious that this operation is comparable to a
kind of budding. If the pseudovum remained adherent to the
parental body, the analogy would be complete.*

The agamogenetic multiplication of Cecidomr/ia-lsLryse is
an essentially similar process. Professor Nicolas Wagner, of
Kasan,' discovered that the larvoe of a Dipterous insect be-
longing to the genus Gecidomyia, or to a closely-allied form
{3flastor)^ multiply agamogenetically in the autumn, winter,
and spring. In summer, the final terms of the successive
broods of grubs thus produced are metamorphosed into
males and females, which copulate and lay eggs. From these,
larvae which exhibit the same phenomena, emerge. In this
case, the young are all developed from germs which are found
lying loose in the perivisceral cavitv of the parent, the body
of which they destroy and burst in order to become free.

1872) should be consulted, not only on account of their richness in details, but
for the peculiar views which the author entertains respecting the nature of
the reproductive process in the Apliides.

* Leydig (" Der Eierstock und die Samentasche der Tnsecten," "Nova
Acta," 1867) affirms that, in November, he has met with Aphides in which, in
the same animal, some of the ovarian tubes contain fullv-formed ova, and others
pseudova, undergoing: their ordinary method of development. Unfortunately
no information is afforded as to whether these aphides possessed a sperma-
tbeca, aud showed evidence of impregnation or not. The occurrence of aga-
mogenesis alongside of sexual propagation is in itself nothing unprecedented,
e. g., Pyrosoma.

3 K. E. von Baer, " Bericht." (" Bulletin Acad. St.-P^tersbourg," 1863.)

IT

386 THE ANATOMY OF INVERTEBRATED ANIMALS.

Leuckart, Metschnikoff, and Ganin/ have shown that these
germs are detached from the pseudovarium, which occupies
the place of the rudimentary ovarium ordinarily found in larvae ;
and that each represents the egg-chamber of an ordinary in-
sect ovariole with its epithelial capsule, ovum, and vitelligenous
cells.

In the ordinary process of growth of an insect, from the
time it leaves the egg until it attains the adult condition, every
marked change in the outward form of the body, or of its ap-
pendages, is accompanied by a shedding of the cuticula. In
some cases the modification eflFected at each ecdysis is very
slight, and the moultings of the cuticle are numerous, amount-
ing in a species of Day-fly (Ghloeon), described by Sir John
Lubbock, to as many as twenty. In such a case as this, the
structure of the adult is gradually substituted for that of the
larva, and the organs of the larva, for the most part, pass into
those of the adult.

The like holds good of some insects which undergo meta-
morphosis, that is to say, in which a quiescent pupal condi-
tion is interposed between the active larval and the active
imaginal states. Herold and Newport have described at
length the series of changes by which the elongated gangli-
onic chain of the Lepidopterous caterpillar is converted into
the much more highly concentrated nervous system of the
Butterfly ; and Weismann has shown by what gradual steps
the apodal Corethra-lsirvn. acquires the character of the Dip-
terous imago. But, in the Flesh-flies {Musca)^ and probably
in many other members of the division of the Diptera to
which they belong, the apodal maggot, when it leaves the
egg^ carries in the interior of its bodv certain regularly ar-
ranged discoidal masses of indifi'erent tissue, which are termed
imaginal disks.^ Of these, twelve are situated in the thora-
cic region, two on each side of each thoracic segment, while
two others lie in front of the pro-thoracic disks. These imagi-
nal disks undergo little or no changfe until the larva incloses
itself in its hardened last-shed cuticle, and becomes a pupa.
But they then rapidly enlarge ; each of the sternal thoracic
disks gives rise to a leg and to its half of the sternal region of

^Leuckart "Die unsreschleolitliclie Yermelimnsr cler CeoidoTnvienlarven "
( G'ttinqer Nachrichten^ 1805) ; K. von Baer, " Uelier Prof. Nic. "Wasrner's Ent-
deckunsr," etc. (" Melansres hioloeiques tires du Bulletin de I'Acad. Imp. des
Sciences de St.-P^tersbouror," 1865).

^ 2 See the remarkable memoir of Weismann, " Die nachembryonale Ent-
"wickelung der Musciden."

THE PARASITISM OF INSECTS.

387

the thorax, while the tergal disks develop into the tergal
halves of the corresponding somites, with their appendages,
the wings and the halteres. The anterior pair of disks origi-
nate the head and proboscis of the fly. As the imaginal
disks develop, the preexisting organs contained in the head
and thorax of the larva undergo complete or partial resolu-
tion. On the other hand the abdomen of the fly is produced
by the continuous modification of the constituents of the lar-
val abdomen.

As in the Crustacea^ so in Insecta, the parasitic habit is

Fig. 111.— The left-hand figure repreeeiats an adult female of Siylops aterrim'us con-
taining two nearly hatched eg^e. and the right-hand fiL'Ure, a newly born larva of
Sfyhps on a hair of Andrfvna Trimmerana. A. ventral surface of the thorax ;
B. the abdomen: a. mandibles ; 6 labial plates and mouth ; c, vulva ; 1, 2, 3, the
three thoracic segments united. (Alter Newport.)

accompanied by extreme modification of form. In this re-
spect the Stre2Jsiptera^vf\\\Q\\ are parasitic upon Bees, present
a remarkable history. The female (Fig. Ill) has the form of
a sac with a short neck, and never leaves the body of the
Hymenopteran in which she is parasitic. The males, on the
contrary, are exceedingly active insects provided with a sin-

388 THE ANATOMY OF INVERTEBRATED ANIMALS.

gle pair of wings, which are attached to the metathorax,
while the mesothorax has a pair of twisted appendages in the
place of wings.

The larvae of both males and females when thej leave the
egg are minute active hexapod insects (Fig. Ill), with rudi-
mentiry manducatory organs, and are found creeping about
between and on the hairs with which the abdomen of their
host is provided. In this condition they are carried into the
nests of the bees, and they attack the larvae of the latter, bor-
ino" their way through the integument into the abdominal
cavity of the grub. Here they cast their cuticle and become
changed into sluggish apodal grubs, provided with a mouth,
with rudimentary jaws, and with an alimentary sac, but de-
void of an anus. About the time that the Hymenopterous
larva passes into its imago state, the Strepsipteral larva
thrusts the anterior end of its body (the so-called cephalo-
thorax) between two of the abdominal segments of the bee,
so that it projects externally. The male becomes a pupa, and
eventually makes its way out as a winged insect. The fe-
male, on the other hand, undergoes little change of outward
form, but presents an opening, which plays the part of a
vulva, and enables the male to effect the fecundation of the
eggs. These are developed within the body of the female,
and make their way out by the cleft in question.^

The Ichneumon-flies deposit their eggs within the bodies
of the larvae of other insects, and the grubs thence hatched
devour the corpus adiposum of their host. The larvae of
some of these parasites (Platygaster, Teleas), described by
Ganin,'' are extraordinarily unlike other insect larvae, and
have a certain resemblance to Copepoda.

» See Von Siebold " Ueber Strepsipteren" (" Arelnv fiir Natur^escliichte,"
1843), and Newport, "Natural History, etc., of the Oil-beetle, Melue^^ ("Linn.
Trans," 1847}.

2 Zeitschrift fur Zoologie, 1869.

CHAPTER VIII.

THE POLTZOA, THE BRACHIOPODA, AND THE MOLLUSCA.

However diverse in outward appearance and in com-
plexity of organization the multitudinous forms of animals
which have been described in the preceding four chapters
(Chap. IV. to VIL) may be, the student passes from one to
the other, by easy and natural gradations, from the simple
Turbellarian at the bottom to the most highly differentiated
Arthropod at the summit of the series. But with the higher
Crustacea^ Ai'achnida, and Insecta the scale ends ; from none
of these are we led to any higher form of animal life.

The Cuttle-fish, the Whelk, the Snail, and the other in-
numerable forms of animals with univalve, bivalve, and mul-
tivalve shells, which are commonly known as Mollusca^ are
so widely different, not only from the Arthropoda,, but from
all the higher members of the group of Worms (Chap. V.),
that any connection with these seems, at first, to be wanting.
The segmentation of the body, which is so conspicuous a fea-
ture of the greater number of the members of the series
which ends with the Arthropods, is absent ; limbs are want-
ing ; instead of the equality of the neural and haemal faces
of the bilaterally symmetrical body, and the consequent
remoteness of the oral and anal apertures, which is usual
among the Arthropods and Worms, these two faces are usual-
ly unequal. The haemal face is often produced into a longer
or shorter cone ; the anus is, as a rule, approximated to the
mouth ; and, very often, the haemal face of the body is asym-
metrical.

The higher Mollusks, in fact, form the final term of a
series of their own, which commences in the Polyzoa, with
animals which have many resemblances to the Hotifera.

The Polyzoa or Bryozoa. — In outward form these ani-
mals bear a general likeness to the Sertularian Hydrozoa^

390

THE ANATOMY OF INYERTEBRATED ANIMALS.

with which they were formerly confounded under the name
of *' Corallines." Like the Sertularians, they almost always
form compound aggregations, produced by repeated acts of
gemmation from the primitively single embryo, and have a
hard cuticular exoskeleton, which remains when the soft
parts decay. The compound organism thus formed is termed

Fig. 112.— A portion of the polyzoarium of Plumatella repens (after Allman).*

a Polyzoarium (Fig. 112), and each zooid which buds from
the common stock is a Polypide. The outer, chitinous or
calcified, cuticular exoskeleton is termed the ectocyst, and, as
the rest of the body of the polypide is contained in, or can
be retracted into, the hard case thus formed, it is commonly
termed a " cell."

The proper ectoderm, with the parietal laj'er of the meso-
derm which lines and secretes this cell, is termed the endo-
cyst. The mouth is situated on a disk, termed the lopko-
phore, at the free end of the polypide; and the margins of
the lophophore are produced into a number of richly ciliated
tentacida. At the oral aperture, the ectoderm passes into the
endodermal lining of the alimentary canal, which is almost
always divided into three portions, a long and wide pharynx,
a spacious stomach, and a narrow intestine. The latter is al-
ways bent up nearly parallel with the pharynx, and termi-
nates in an anus situated beside the mouth. As the nervous
ganglion is placed between the mouth and the anus, the flex-
ure of the intestine is neural^ and the hasmal face of the

» " Monograph of the Fresh-water Polyzoa," 1856.

' In dealing with the morphological relations of the parts of Mollusks, it is
very necessary to employ a terminology which shall be independent of the or-

THE POLYZOA.

391

bodv is developed greatly in excess of the neural face. A
wide perivisceral cavity occupies the interval between the
alimentary canal and the parietes cf the body, and sometimes

Fia. 113.— Plumatella repsns.—A single cell more masrinfied : a, ectocyst ; 5, endo-
cyst ; m, calyx at the base of the ciliated tentacula borne by the disk or lopho-
phore ; k, mouth ; /, gullet ; g g, stomach ; h, intestine ; i, anus ; 7i, muscles; w,
nervous ganglion ; z, statoblasts ; 0, funiculus. (After AUman.)

the walls of this cavity are ciliated. Ver}'- generally, the
gastric division of the alimentary canal is connected with
the parietes of the body by a sort of ligament, the funi-
culus^ or g astro-parietal hand. Circular and longitudinal mus-
cular fibres, which frequently exhibit distinct transverse stria-
tions, may be developed in the body-wall ; and there are usual-
ly special muscles for the retraction of the lophophore within
the cell, and others for the closing and opening of the oper-
cular apparatus, with which many species are provided.

dinary position of the animals. I therefore term that face of the body on
whic£ the chief nervous centres, or the pedal ganglia (when such are separately
distinguishable), are placed, neural^ and the opposite JuBmal.

392 THE ANATOMY OF INVERTERBATED ANIMALS.

The single nervous ganglion is situated, as has been
stated, between the oral and the anal apertures. In Seria-
laria^ Scrupocellaria, and some other genera, nervous cords
and plexuses connecting the ganglia of the several polypides,
and constituting what F. Mliller ^ terms a " colonial nervous
system," have been described. But it is not yet certain that
these cords and plexuses are really nerves. It is doubtful if
there are any special organs of sense, unless a lobed process
— the ejnstoma — which overhangs the mouth in many fresh-
water Polyzoa, be of this nature. The ectoderm of that region
of the body which lies immediately beneath the tentacula is
always soft and flexible ; and when the lophophore is re-
tracted, becomes invaginated, so as to form a sheath, by which
the tentacles are protected. Sometimes, as in the Ctenosto-
inata^ this sheath is surrounded by a circle of chitinous fila-
ments, which, when the tentacles are retracted, furnish a pro-
tective outer covering to them. And, sometimes, as in the
Cheilostomata^ part of the ectocyst of the polype cell is dis-
posed in such a manner as to constitute a movable lid, which

Fig. Ili. — ScriipocePMria fernx.—\ small portion of a polyzoarlum, Bhowing tlie

vibracula (a). (After Busk.) *

J " Arclilv liir Anatomie," 1860.

2 Farre, " Ohservations on the Minute Stnictiire of some of the Hisrher Forms
of Polypi" ('• Phil. Trans.," 1837\ Reichert, " Ueber Zoobotryon pellucidus'''
(" Ahh. d. konisrl. Akad. der Wissenschaften," Berlin, 1869).

3 Busk, " Catalogue of the Marine Polyzoa in the British Museum : Cheilo-
Btomata," 1852-'o4. See for this prroup Nitsche's recent important "Beitrage zur
Kenntniss der Bryozoen" {Zeitachrift f>'ir wiss. Zoologie^ 1869-'71).

* " Catalogue of Marine Polyzoa," 1852.

AVICULARIA AND VIBRACULA.

393

shuts down on the retracted poljpide. This operculum is
placed on the opposite side of the polypide to that on which
the nervous gauglion is situated.

In many genera, the cells are provided with flagelliform
appendages— the vibracula (Fig. 114). These are usually
articulated with short dilated processes of the ectocyst, and
execute constant lashing movements. In others bodies
shaped like birds' heads, with a movable mandible, and either
seated upon slender and flexible peduncles or sessile, snap
incessantly. Sometimes these last, which are termed ^avicu-
laria (Fig. 115), are present along with vibracula.

E^G. 115. — Bvqula avicularia.—A. Part of the polyzoarium viewed from the nenral
side, showing the tentacles of a polypide protruded from its cell (A); the intestine
(D and the stomach and gullet (/) ; <7. retractor muscles ; d. d, avicularia. One of
these is holding a minute worm which it has seized. In front of this is seen an
ovicell.

B. A retracted polypide withan avicularium (c?), viewed from the haemal or dorsal
side.

The dilated bases of the vibracula contain muscles by the
contraction of which the flagelliform appendage is moved.
In the avicularia, a large adductor muscle, which takes its
origin from the greater part of the inner surface of the

394 THE ANATOMY OF INVERTEBRATED ANIMALS.

"head," is attached by a slender tendon to the "mandible"
on the one side of the hinge line, while a smaller divaricator
muscle is fixed to the other side. The mechanism of adduc-
tion and divarication of the mandible is quite similar to that
by which the dorsal valve of the shell of an articulated Bra-
chiopod is moved upon the ventral valve.

Male and female reproductive organs are usually com-
bined in the same polypide. They are cellular masses, devel-
oped in the funiculus, or in the parietes of the body, whence
the ova or spermatozoa are detached into the perivisceral
cavity. They sometimes pass thence, and undergo the first
stages of their development in dilatations of the w^all of the
body, termed ovicells.

Multiplication by gemmation occurs throughout the group,
but the buds usually remain adherent to the stock. In Loxo-
soma and Pedicellina, however, the buds become detached.

Some Polyzoa multiply agamogenetically by a kind of
gemmule developed in the funiculus^ provided with a pecul-
iar shell, and termed a statohlast.

With these general characters, tlie Polyzoa present an
interesting series of modifications. They have been divided
by Nitsche into two groups — the JEiitoprocta^ in which the
anus lies within the circle of tentacles ; and the Ectoprocta,
in which it lies outside this circle. In the former division,
the genus Loxosoma^ which attaches itself to Sertularians
and to other Polyzoa^ is particularly noteworthy. It is a
small stalked animal, and the superior wider end of the body
is an obliquely truncated disk, the margins of which are elon-
gated into ten ciliated processes. The mouth is a trans-
versely elongated, slit-like aperture on the lower side of the
tentacular circlet. A long oesophagus connects this with a
globular cgecal gastric sac. From the midst of the disk, a
conical prominence, the summit of which bears the anus, is
situated. The sexes are united, the ovaries and testes being
situated on each side of the stomach, and the spermatozoa
pass directly into the ovaries. No nervous system has yet
been made out in Loxosoma. The animal is fixed by the
truncated extremity of its narrow stalk-like end ; and this
stalk contains a gland, the duct of which opens in the centre
of the face of attachment.

Loxosoma appears to multiply by budding, but the ap-

1 Kowalewsky, " Beitrage zur Anatomic und Entwickelunsfsgeschichte des
Loxosoma neapoUtanum " ("Mem. de I'Acad. de Si.-Petersbourg," 1866). Os-
car Schmidt " Die Gattung Loxosoma" ("Archiv liir mikr. Anatomic," 1875).

THE POLYZOA. 395

parent buds are reallj one of two kinds of embryos devel-
oped from the impregnated ova. The other kind of embryo
becomes a gastrula, with a large post-oral ciliated disk, like
a mesotrochal annelid larva, and its ultimate fate has not yet
been traced.

The Ectoprocta are divided into the Gymnolcemata^ which
have a circular lophophore, and no epistoma ; and the Phylac-
tolcematQj^ which possess an epistoma, and usually have the
lophophore prolonged into two lobes, so as to be horseshoe-
shaped ; whence the term hippocrepian applied to such Po-
lyzoa.

Among the Gymiiolcemata are distinguished : the Cyclo-
stomata, in which the opening of the cell is round and has no
opercular structures ; the Ctenostomata (supra)^ and the
Cheilostomata (supra).

All the Phylactol(Binata are inhabitants of fresh water ;
while all the Gyynnolcernata, except Paludlcella^ are marine.

The polyzoarium of Cristatella is free and creeps about as
a whole ; and that of Lunulites is free, at any rate in the
adult condition.

In the fresh-water Polyzoa^ the impregnated ovum gives
rise to a saccular planuliform embryo, which is covered external-
ly with cilia. From one end of this cystid^ one or more poly-
pides are developed from thickenings of the wall of the sac.

In the Gymnolfematous genera Pugula, Scrupocellaria, and
Picellaria, the embryo is ciliated, and provided with a mouth
and with eye-spots. After swimming about for some time, it
loses its cilia, fixes itself, acquires a chitinous outer coat, and
becomes a mere sac or cystid, in which a polypide is developed
by gemmation, and gives rise to the first cell of the polyzoa-
rium.

Schneider^ has shown that the anomalous Cyphonaufes,
which he considers to resemble Actinotrocha, and which is
inclosed in a bivalve shell, is the larva of Membranipora pi'
losa. It is provided with an intestine, and with largely de-
veloped ciliated motor bands. But when it attaches itself, all
these organs disappear, and the larva passes into the condi-
tion of a cystid, from which a polypide is developed, as in the
foregoing cases.

* See Bumortier and Van Beneden, " Histoire Naturelle d. Pol}7)es compo-
s6es d'eau douce " (" M^ra. de I'Aad. Jioyale de Bruxelles," 1850)'; the mono-
graph of Allman cited above ; and Nitsche's " Beitrage."

2 " Zur Entwickelungsgeschichte und svstematischen Stellung der Bryozoen
Und Gephvreen." (" Archiv fur mikr. Anat.," 1869.)

896 THE ANATOMY OF IXVERTEBRATED ANIMALS.

Hence, it has been pointed out that the characteristic poly-
pide of the ectoproctous Folyzoa is a structure developed
from the cystid, in much the same way as the Tte^^/a-head is
developed from its saccular embryo; or as the Cercaria is de-
veloped from the sporocyst, or JRedia ; the cystid of the Phy-
lactolcBmata being comparable to a sporocyst, and that of 3Iem-
hrcmipora to a Bedia. But, without altogether denying the
justice of this comparison, it may be suggested that the cys-
tid is comparable to a vesicular morula, and that the mode of
development of the alimentary canal of the polypide corre-
sponds with that of the formation of an alimentary sac by in-
vagination. If this view of the case be correct, the perivisce-
ral cavity in the Polyzoa is a blastocoele, more or less modified
by the development of the mesoderm.

The only known representative of the genus Bhahdoplen-
ra ^ is an aberrant Polyzoon which presents many interesting
peculiarities. The polyzoarium consists of a creeping stem
from which erect branches, each of which ends in a circular
aperture and constitutes the cell of a polypide, arise. The
cavity of the stem is divided by transverse septa, and its
centre is traversed by a hollow chitinous cord, which passes
through and is attached to the septa.

The lophophorc resembles that of the hippocrepian Phy-
lactolceinata in being produced into two arms, fringed with a
double series of tentacula. These arms are longer, narrower,
and more cylindrical than in any other Polyzoa, and, thus far,
approach the arms of the Prachiopoda. Furthermore, the
tentacula are confined to the arms, which are very flexible.
Betvveeen the bases of the arm.s there is a rounded or pen-
tagonal disk with raised and ciliated edges, which occupies
the place of the epistoma in the phylactolaematous Polyzoa,
The mouth is situated beneath the free margin of this disk,
on the opposite side to the anus, and to that toward which
the arms are turned. The animal is attached to the bottom
of its cell, or rather to the endosarc of the stem, by means of
a long contractile pedicle, by which its retraction is effected.
According to Sars it protrudes itself by climbing up the wall
of its tubular cell by means of the disk. Prof. Lankester's
comparison of the polypide of Phahdopleura to the embryo
Plsidium^ appears to me to be fully justified. The branchiae
of Nucula, in form and position, present no little resemblance

* See the papers of Allman and G. 0. Sars, Quarterly Journal of Micro-
icopical Scie?ice, 1869 and 1874.

3" On the Developmental History of the Mollusca." (" Phil. Trans.," 1874.)

THE BRACHIOPODA. 397

to the arms of Rhabdopleura^ though these, like the arms
of the BroGhiopoda^ are probably more strictly comparable
to the labial palpi of the Lamellibranchs.

Polyzoa occur in the fossil state from the Silurian epoch
to the present day, and the oldest forms are referable to the
groups which now exist.

The Brachiopoda. — The BracTiiopoda are all marine
animals provided with a bivalve shell, and are usually fixed
by a peduncle which passes between the two valves in the
centre of the hinge line, or the region wdiich answers to it, in
those Brachiopods which have no proper hinge. They never
multiply by gemmation, nor give rise to compound organisms.
The shell is always inequivalve and equilateral ; that is to say,
each valve is symmetrical within itself and more or less un-
like the other valve. The shell is a cuticular structure se-
creted by the ectoderm, and consists of a membranous basis,
hardened by the deposit of calcareous salts, sometimes con-
taining a large proportion of phosphate of lime {Lingulci).

In many Brachiopods, variously-formed calcareous spic-
ula, or minute plates, are found in the walls of the peri-
visceral cavity, and of the greater sinuses ; and also in the
arms and cirri, and sometimes these unite together so as to
form an almost continuous skeleton.*

The body, or rather that part of it which contains the chief
viscera, is often small relatively to the valves of the shell,
and the integumsnt is produced into two broad lobes, which
line so much of the interior of the valves as the visceral mass
does not occupy. The free edges of these lobes are thickened,
and are beset with numerous fine chitinous set£e like those
found in Annelids, and like them lodged in sacs. Between
the two lobes of the mantle, or pallium^ is the pallial cham-
ber, bounded behind the anterior wall of the visceral mass.
In the middle line, this wall presents the oral aperture, which
is seated in the midst of a wider or narrower area, the mar-
gins of which are provided with numerous ciliated tentacula.

In Argiope, the oral area occupies a large part of that lobe
of the mantle which is ordinarily termed dorsal, and its mar-
gins are simply indented by three deep sinuations. In Theci-
diwn, the sinuations are deeper, and the folds of the oral area
thus produced narrower. But in most Brachiopods the oral

1 These have been described by "Woodward, Lacaze-Duthiers. and especially
by Eudes Deslongcbamps, "Reclierches sur 1' organisation du Manteau chez les
Brachiopodes articules," 1864.

398 THE ANATOMY OF IXVERTEBRATED ANIMALS.

area is narrowed to a mere groove, and is produced on each
side of the mouth into a long spirally-coiled arm, fringed with
tentacles ; whence the name of Brachiopoda^ applied to the
group.

In this case the tentacula disappear from the anterior
margin of the oral disk in the region of the mouth, and are re-
placed by a lip-like ridge. Each arm contains a canal, wliich
ends in a sac at the side of the mouth.

In Waldheimia (Fig. 116), the two arms are united to-
gether and their distal portions coiled into a horizontal spiral.
In many genera, the margins of the oral area or arms are
fixed to processes of the dorsal valve of the shell.* In this
case the arms are not protrusible ; but, according to the ob-
servations of Morse,^ they can be straightened and extended
beyond the shell in Iihy7ichonella^ which has no brachial
skeleton.

The alimentary canal consists of an oesophagus, a stomach,
provided with hepatic follicles, and an intestine. In the uja-
jority of existing genera the latter is short, and ends in a
csecum in the middle line of the body (e. g., Waldheimia) ; in
others it is long, and opens into the pallial chamber on the
right side of the mouth (e. g., Lingula^ Discina, and Crania).

The alimentary canal is invested by an outer coat — the so-
called peritoneum — by which it is suspended, as by a mesen-
tery, in a spacious "perivisceral" cavity. The walls of this
cavity are provided with cilia, the working of which keeps up
a circulation of the contained fluid. Lateral processes of this
coat — the g astro-parietal and ileo-parietal bands — connect
the gastric and intestinal divisions of the alimentary canal
respectively, with the parietes.^

From the perivisceral cavity, sinus-like, branched prolonga-
tions extend into each lobe of the mantle, and end cascally at
its margins. The lobes of the mantle are probably, together
with the ciliated tentacula, the seat of the respiratory func-
tion. The sinuses of the pallial lobes of Lingida give rise
to numerous highly contractile, teat-like processes, or ampul-
loe. During life the circulating fluid can be seen rapidly cours-
ing into and out of each ampulla in turn (Morse, /. c, p. 33).

1 See, for excellent figures of these arranoreraents, and for the shells and ex-
ternal form of the body in general, "Woodwanl's " Manual of the Mollusca."

a " On the Systematic Position of the Brachiopoda." (" Proceedings of the
Boston Society of Natural History," 1873.)

3 Huxley, " Contributions to the Anatomy of the Brachiopoda " ("Proceed-
ings of the Koval Society," 1854); and Hancock, " On the Organization of the
Brachiopoda" '("Phil. Trans.," 1858).

THE BRACHIOPODA.

399

Fio. 116.— Lateral view of the vifcera of WaMheimia austraZis (after Hancock. " On
the Or?anization of the Brachiopoda," " Phil. Trans.," 1858). a, " dorsal " layer
of mantle; ft, "ventral " layer; c, anterior walls of the body between the mantle
lobes : (f, arras ; p, gullet ; q^ stomach with cut biliary ducts of the left side ; r, rii.'ht
hepatic mass; «, intestine endino: cascally below; v, so-called *' auricles;" o, the
right " pseudo-heart," the left being almost wholly removed ; w. pyriform vesicle
fixed at the back of the stomach; z, oesophageal ganglia; e,^", adductor; k, divari-
cator ; I, adjustor muscles ; n, peduncles.

400 THE ANATOMY OF INVERTED RATED ANIMALS.

The perivisceral cavity communicates with the pallial
chamber by at fewest two, and sometimes four {Rhynchonel-
la), tubular organs, which have been described as hearts,^ but
are now known to have no such nature.

Each of these organs is shaped like a funnel, the wide por-
tion which opens into the perivisceral cavity being much plait-
ed and folded, and separated by a constriction from the nar-
rower part, which answers to the pipe of the funnel. The lat-
ter, passing obliquely through the anterior wall of the visceral
chan)ber, ends by a small aperture in the pallial cavity. Prof.
Morse has observed the passage of the eggs through these
organs in TerehratuUna septentrionalis. They are drawn
into the open end of the funnel by the action of the cilia with
which its surface is covered, and enter the pallial cavity by
the aperture just mentioned. It is probable that these ** pseu-
do-hearts " subserve the function both of renal organs and
of genital ducts ; and that they are the homologues of the
organs of Bojanus of other moUusks, and of the segmental
organs of worms.

Between the ectoderm and the lining membrane of the
prolongations of the perivisceral cavity in the mantle, and
between the endoderm, the ectoderm, and the lining membrane
of the perivisceral cavity itself, there is an interspace broken
up into many anastomosing canals, which I conceive to rep'
resent a large part of the proper blood system.

Vesicular dilatations of the walls of these canals found at
the back of the stomach, and in some other localities, in the
Brachiopods with articulate shells, have been regai-ded as
hearts; but observations on the living animals, made by various
investigators, show that they are not contractile, and their
function is unknown. Although the existence of a direct
communication between the perivisceral chamber and the
blood canals has not been demonstrated, it is very probable
that the perivisceral chamber really forms part of the blood-
vascular system.

Muscles for the adduction and divarication of the valves
of the shell, and for effecting the other movements of the ani-
mals, are well developed in the Brachiopoda.^ They are to a
great extent striated.

* Owen, *' Lettre sur Pappareil de la circulation chez les Mollusques de la
classe des Brachiopodes." (" Annates des Sciences Naturelles," 1845.)

2 See Hancock {I. c). Owen, Introduction to Davidson's "Fossil Brachi-
opoda." ("Memoirs of the Palicontoofraphical Society," and "Transactions
of the Zoological Society of London," 1835.)

1

THE DEVELOPMENT OF THE BRACHIOPODA. 401

The nervous system of the articulated Brachiopods, in
which it has been best made out, consists of a relatively thick
ganglionic band on the ventral side of the mouth, the ends of
which are united by a commissural cord, which surrounds the
gullet, and bears two small ganglionic enlargements. The lat-
ter probably answer to the cerebral, the former to the pedal,
ganglia of the JLamellihranchiata. Immediately behind the
pedal mass, from which two large nerves to the dorsal or ante-
rior lobe of the mantle are given off, are two elongated ganglia,
connected by a commissure of their own, which possibly cor-
respond with the parieto-splanchnic ganglia of the higher Mol-
lusks. The nerves to the ventral lobe of the mantle and those
to the peduncle arise from these ganglia.

In the inarticulated Brachiopods, our knowledge of the ner-
vous system is very imperfect. In Lingula^ Professor Owen
has described two lateral nerve-cords, and the observation
has been confirmed by Gratiolet and Morse. The latter anato-
mist finds similar cords in Dlsclna, and Gratiolet has de-
scribed an oesophageal ring in Lingular

The reproductive organs are lodged in the perivisceral
cavity or its prolongations, and are apparently always con-
tained in processes of the lining membrane of that cavity.
It is not clear whether hermaphrodism is the rule or the ex-
ception. Thecidium^ however, has been shown by Lacaze-
Duthiers to be dioecious ; and, according to Morse, the sexes
are distinct in Terebratidina and Discina.

The development of the JBrachiopocla, notwithstanding
the important observations of F. Muller,'' Lacaze-Duthiers,^
and especially of Morse,* stood much in need of further eluci-
dation (especially in reg'ard to the earlier conditions of the
embrj^o), until quite recently, when the investigations of
Kowalew^sky" filled up the hiatus in our knowledge for the
genera Argiojye^ Thecidium, Terehratida,, and Terebratulina,
The Q^^ becomes converted into a vesicular morula, in which
an alimentary sac is developed by invagination, and this sac
gives off, as in Sagitta, two diverticula, w-hich become shut

* " Recherches pour servir a Pbistoire des Eracliiopocles." (" Journal de
Conch vliologie," 1860.)

2 " Beschreibunof einer Bracliiopoden-Larva." (" Archiv fur Anat.," 1860.)

3 " Histoire de la Thecidee." (" Ann. d'Hist. Nat.," 1861.)

* " On the early stages of Terehratulina septentrionalisJ'^ (" Memoirs of the
Boston Society of Natural History," 1869, and the memoir already cited).

6 Contained in a memoir, published at Moscow in 1874, for which I am in-
debted to the courtesy of the author. It is in Russian ; but I have been able to
acquaint myself with its contents, to some extent, by the aid of a friend.

402 THE ANATOMY OF INVERTEBRATED ANIMALS.

off from the alimentary canal, and are converted into the peri-
visceral csLvity, The latter, therefore, is an enterocoele. The
embryo elongates, and constrictions divide it into three seg-
ments, of which the anterior becomes fringed with long cilia,
and develops eye-spots. Thus the young Brachiopod acquires
a great resemblance to an ordinary Annelid larva. The re-
semblance is increased by the appearance of four bundles of
setag on the middle segment, which becomes produced into a
sort of hood, the free edges of which are at first turned back-
ward and bear these set^. As the larva grows, the third
segment becomes truncated at the end, and furnishes a sur-
face (provided with a shell gland ? i^ifrd), by which the larva
attaches itself. At the same time, the first, or prsestomial
segment, atrophies, and the setigerous hood developed from
the middle segment is retroverted, rapidly grows, and gives
rise to the lobes of the mantle, on which the valves of the
shell are developed.

The resemblance of the larval Brachiopod to a Polyzoon,
and especially to Xoxosoma, is striking, and full}^ bears out
the conclusion as to the affinity of the Polyzoa with the
JBrachiopoda which results from the study of their adult
structure. On the other hand, the development of the Bra-
chiopoda no less strongly testifies to their close relations with
the Worms.^

In the course of the previous pages the terms dorsal and
ventral have been employed in the sense in which the^^ are
conventionally used by conchologists. But an interesting
question, and one not easy to settle, is. What relation do these
dorsal and ventral regions of a Brachiopod bear to the neural
and hajmal regions of a Polyzoon, or to those of a Lamelli-
branch, or of a Gasteropod ?

If we compare one of the articulated Brachiopods, such as
Waldheiniia^ in its shell, with a polypide of a Cheilostoma-
tous Polyzoon in its cell, the dorsal valve will appear to an-
swer to the operculum, and the ventral valve to the cell. If
this comparison be just, the two lobes of the mantle of the
Brachiopod must both belong to the dorsal or hsemal aspect
of the body ; that which corresponds with the so-called dor-
sal valve of the shell being the anterior, and that which lines

* The acceptance of the view originally propounded by Steenstrup, and so
ably urged by Prof. Morse, respecting the affinities of the Brachiopods with
the Worms (" Proceedings of Boston Society of Natural History," 1873), does
not to my mind weaken the opinion I have always held as to their affinities
with i\iQ Polyzoa, on the one hand, and with the higher J/oW««ca, on the other.

THE BRACHIOPODA. 403

the ventral valve of the shell being the posterior lobe. And
the region of the anterior wall of the pallial cavity which lies
behind or below the mouth will answer to the neural aspect
of the Polyzoon.

On the other hand, if the segments of the body of the
larval Brachiopod are true somites, and the discoidal surface
of the hindermost corresponds with the similarly formed end
of the larva of Lacinularia^ as Prof. Morse suggests, the
dorsal lobe of the mantle will, as before, represent part of
the haemal surface of the body, but the ventral lobe will be-
long to its neural surface — and can no longer properly be
termed mantle, but will rather answer to the foot of one of
the higher Mollusca.

The Brachiopoda are distinguishable into two groups,
the Articulata and the Inartlculata, In the Articulata, the
two valves are united by a hinge, and the ventral valve is
usually provided with teeth, which are received in sockets
of the dorsal valve. The gullet ascends in the middle line
toward the dorsal v^alve, and the intestine descends toward the
opposite, or ventral, valve, and there ends in a cascum. The
dorsal valve often gives rise to spiral or looped shelly pro-
cesses to which the arms are attached. The valves are
brought together by a pair of adductor muscles, which pass
directly from valve to valve ; and they are separated by di-
varicator muscles, which run obliquely from the ventral valve
to a median process (the cardinal x>^'ocess) of the hinge-line
of the dorsal valve. The impressions of the attachments of
these muscles on the inner surfaces of the valves have con-
siderable systematic importance. Very often the ventral
valve is produced into a sort of spout, through which passes
the peduncle by which the animal is attached to rocks. At
the sides of the visceral chamber the thickened edge of the
dorsal lobe of the mantle passes into that of the ventral lobe.

The substance of the shell is very often traversed by
numerous canals perpendicular to its surface, which contain
prolongations of the mantle.^

This division contains the families of (1) The Terehra-
tulidcB^ (2) the Spiriferidne, (3) the Ehynchonellidoe, (4) the
Orthidm^ and (5) the Productidce, of which the second, fourth,
and fifth are extinct and almost wholly palasozoic, no species

1 The structure of the shell has heen particularly studied by Carpenter.
("Reports of the British Association," 1844-'47, and Introduction to David-
son's " Fossil Brachiopoda.") See also Kin^, " Trans. Royal Irish Academy,"
1869.

404 THE ANATOMY OF INVERTEBRATED ANIMALS.

extending beyond the lias, while the majority of the species
of the other two families are also extinct.

The family of the Terehratulidce, which is not certainly
known to occur in formations older than the Devonian, is the
only one in which, since the end of the palaeozoic epoch,
numerous new generic types appear.^

The Jnarticulata have no hinge ; the intestine opens into
the cavity of the mantle, the margins of the lobes of which
are completely separate. Some have a long peduncle {Lin-
gula), others are fixed by a plug which passes through an
aperture or notch of one valve [Discina), or by the surface
of one valve ( Crania). There is no brachial skeleton, and the
arrangement of the muscles is in many respects different
from that which obtains in the articulated division.

Species of all these families, except the Spiriferidce,
Orthidce, and Productidce^ exist at the present day, but they
are also represented in the older palaeozoic epochs, and Lin-
gulce are among the oldest known fossils.^

The Mollusca. — The term 3follusca may be used as
a convenient denomination for the Lamellibranchiata and
Odontophora (= Gasteropoda^ Pteropoda, and Cephalopoda,
of Cuvier), which can be readily shown to be modifications oi
one fundamental plan of structure. This may be represented
by a body, symmetrical in relation to a median vertical plane,
at one end of which is the oral and at the other the anal
aperture of the alimentary canal. In the body a ventral, or
nei(ral^ face, an opposite dorsal, or hmmal, face, and a right
and left side may be distinguished. The neural face usually
gives rise to a muscular /bo^. The integument of the haemal
face is generally produced at its edges into a free fold, and
the term mantle, or pallium, is applied to the region of the
integument thus circumscribed. Between the free portion of
the mantle and the rest of the body is a cavity, the pallial
chamber, from the walls of which, processes which subserve
respiration, the hranchim, may be developed.

In the median line of the surface of the mantle of the em-
brvo a shell-gland is very generally formed, and from the
surface of the mantle a cuticular secretion, the shell, is pro-
duced,

» Suesp, " Ueber die Wohnsitze der Brachiopoden." (" Sitzb. d. "Wiener
Atad.," 1857.)

« See "Davidson's " Monographs of British Fossil Bracbiopoda," in tlio Pa-
Iseontographical Society's publications.

THE MOLLUSCA. 405

A systemic heart usually exists, and when present is situ-
ated in the middle of the posterior heemal region, and consists
of, at fewest, two chambers, an auricle and a ventricle. Arte-
rial vessels often ramify extensively through the body, but
more or fewer of the venous channels remain in the condition
of lacunae. The blood-corpuscles are colorless and nucleated.
Distinct respiratory organs may be absent, or they may take
the form of braiichi^ or pulmonary sacs. When present, they
lie in the course of the blood which is returning to the heart.
Beside the heart and the intestine are situated the renal or-
gans, which, on the one side, open externallj', and on the
other communicate with the blood system.

The nervous system consists of, at least, one pair of ganglia
{cerebral) at the sides, or on the haemal aspect of, the mouth,
and of two other pairs of oesophageal ganglia (pedal and
parieto-splanchnic). The latter are situated at the sides, or
on the neural aspect, of the alimentary canal, and are con-
nected by commissures with the former.

In the majority of the 3Iollusca, the embryo passes through
a stage in which it is provided w^ith bands of cilia or with a
simple, bifid, or multifid fold of the integument {vehun), the
edges of which are ciliated, developed on the haemal aspect of
the cephalic region of the bod}', in front of the pallial region.

The special peculiarities of the different groups of the
MoUusca result chiefly —

1. From the form of the pallial region, and the extent of

the mantel-lobes relatively to the body,

2. From the number and arrangement of the pieces of the

shell to which the mantle gives rise.

3. From the proportional size and the form of the foot

and the production, or non-production, of chitinous,
or shell}', matter by it,

4. From the development of sense-organs on the anterior

end of the body, and the absence or presence of a
distinguishable head.

5. From the disproportionate growth of the haemal region

of the body into a visceral sac, followed by a change
in the primitive direction of the intestine, and often
accompanied by asymmetrical lateral distortion.

The Lamellieeaxchiata.^ — In these MoUusks there are

1 For a description of the anatomy of a Lamellibranch in detail, the student
is referred to Huxley and Martin, " Elementary Biology," and Eolleston,
*' Forms of Animal Life."

406 THE ANATOMY OF INVERTEBRATED ANIMALS.

always two large pallial lobes, the margins of which are de-
void of setae ; and which are lateral, or right and left, in rela-
tion to the median plane. Each lobe gives rise to a piece, or
valve^ of the shell ; and to these, accessory pieces, developed
upon the median haemal face [Pholas) or the posterior end of
the mantle {Teredo)^ are in some cases added; or, in addition
to its valves, the mantle may secrete a shelly tube {Teredo^
Asi^erg ilium). The shell itself consists of superimposed
lamellae of organic matter, hardened by the deposit of calca-
reous salts. It is a cuticular excretion from the surface of the
mantle, and never presents any cellular structure. But, from
the disposition of its lamellae, and from the manner in which
the calcareous deposit takes place in them, it may present
varieties of structure which have been distinguished as nacre-
ous, prismatic, and epidermic*

The two valves are generally united over the median line
of the haemal surface of the body by an uncalcified chitinous
cuticular matter, termed the ligament^ which is usually very
elastic, and is so disposed that, when tlie valves are closed, it
is either stretched or compressed. In either case, it antago-
nizes the action of the adductor muscles, and divaricates the
valves when these muscles are relaxed. Conchologists com-
monly draw a distinction between an internal and an external
ligament ; but, in relation to the body of the animal, all liga-
ments are external, and their internality or externality is in
respect of the hinge-line^ or the line along which the edges
of the valves meet. In symmetrical, or equivalve, Lamelli-
branchs, each valve is concave internally and convex exter-
nally; it has, in fact, the form of a very depressed cone, the
apex of which, termed the umbo, is incurved and is situated
on, or projects beyond, the haemal, or, as it is termed, dorsal
edge of the valve. Moreover, it is usually inclined forw^ard,
and situated nearer the anterior than the posterior end of the
valve. Sometimes the umbonic cone is prolonged and bent
inward, or may even form a short spiral turn (Tsocardia,
Diceras), so that the valve acquires a certain resemblance to
the shell of some gasteropods. As the shell of a Lamellibranch
increases in thickness by the deposition of new layers on the
interior face of the old ones, and, in area, by the extension of
the new layers beyond the old ones, the summit of the umbo
represents the original shell of the embryo, and the outer sur-

1 See Carpenter, article " Sliell," Todd's " Cyclopaedia." Huxley, " Tegu-
mentary Organs," ibid.

THE LAMELLIBRANCHIATA.

407

face is usually marked by concentric lines of growth^ which
indicate the boundaries of the successively added new layers
of shell-substance.

Fia. 117.— Sectional diasrram of a fresh-water Mussel { Anodonta).—A^ ^.mantle, the
right lobe of which is cut away; jB, foot,; C, branchial cliamber of the mantle
cavity; D. anal chamber; /, anterior adductor muscle; 77, posterior adductor
muscle; 777, retractor muscle of the foot; «, mouth; 6, stomach; c, intestine,
the turns of which are supposed to be seen through the side-wall? of the meso-
eoma; d, rectum: e, anus; /, ventricle; g, auricle; A, ffills, except i. rieht exter-
nal orill, lar<rely cut away and turned back; A-, labial palpi; /, cerebral: m, pedal;
n. parieto-splanchnic ganglia; o, aperture of the kidney or organ of Bojanus; i?,
pericardium.

The applied edges of the two valves are very often pro-
duced into elevations and depressions which interlock with
one another. The form and arrangement of these teeth and
sockets are of much use in systematic conchology.

408 THE ANATOMY OF INVERTEBRATED ANIMALS.

The muscles which are attached to the valves, viz., the ad-
ductors, retractors of the foot, and pallial muscles, give rise
to impressions on the inner faces of the valves, which are
very obvious when the latter are removed and cleaned. With
the growth of the animal, the distance of these impressions
from the hinge-line and from one another is necessarily in-
creased, and it is not difficult in some cases (e. g., Anodo7ita)
to trace a faint triangular mark, which has its base in each
adductor impression and its apex in the umbo, and which in-
dicates the successive shiftings of position of the muscle.

In some cases (e. g., Lima) a Lamellibranch may perform
a sort of aquatic flight by the flapping of the valves of its
shell.

The hard and sharp-edged valves of the shell in Teredo
are probably the agents by which the mollusk carves its pas-
sages through the wood which it inhabits. Whether the
valves of the shell of the Pholades and Saxicavm are the in-
struments b}' which they excavate their burrows in hard rock,
or whether, as has been suggested, the foot, armed with
sand, is the boring instrument, is a question which has been
much discussed, but hardly brought to a satisfactory decision.

The haemal face of the body is either flat or slightly
arched, whence, in side view, the haemal contour is either
straight or convex. In most Lamellibranchs the body is
symmetrical in relation to the median plane, but, in those
which have inequivalve shells, as the Scallop {Pecten) and
the Oyster ( Ostrma), the one half is more convex than the
other. No Lamellibranch has a distinct head ; but, in those
which possess two adductor muscles (e. g., A9iodoraa), the
region in which the anterior adductor lies and which is situ-
ated in front of the mouth may be distinguished as the^ro-
somay from the middle region (mesosoma) which gives rise to
the foot ; while the part which lies behind the foot and con-
tains the posterior adductor may be termed the metasotna.

The foot may be rudimentary, but it is usually large, flex-
ible, and employed as an organ of locomotion. The posterior
face of the foot not uncommonly presents a gland which se-
cretes a chitinous, or shelly, substance — the hyssus.

From the sides of the mesosoma, close to the attachment
of each mantle-lobe, the branchiae project into the pallia)
cavity.

In its simplest form, the branchia of a Lamellibranch con-
sists of a stem fringed by a double series of filaments (e. g.,
JSfucula), The next degree of complication arises from these

THE LAMELLIBRAXGHIATA. 409

filaments becoming as it were doubled upon themselves at
their free ends, the reflected portions lying on the outer side
of the outer, and on the inner side of the inner, series of
primary filaments. But the free, or h^mal, ends of the re-
flected filaments contract no adhesion either with the mantle
on the outer side, or with those of the opposite gill on the
inner side. Delicate bands stretch from the primary to the
reflected filaments across the interspace which they inclose
{j\f>/tili(S, Pecten). In most Lamellibranchs the gills are
four elongated plates, each of which is in fact a long and
narrow pouch, with its open end turned toward the haemal
face of the body. Two pouches are situated on each side of
the mesosoma ; one of these pouches is internal, the other
external.^ Their walls are united by transverse septa ; they
are richly ciliated, and are perforated by numerous apertures.
As the outer wall of each pouch is united with the mantle,
and the inner with its fellow of the opposite side, behind the
foot, the whole branchial apparatus forms a sieve-like parti-
tion extended between the mantle and the foot (Fig. 117),
and thus divides the pallial cavity into a supra-hranchial
and an infra-branchial chamber. Inasmuch as the haemal
edge of the inner wall of each inner branchial pouch is, for
the greater part of its extent, not united with the mesosoma,
but only closely applied against the latter, the supra-bran-
chial and infra-branchi;d chambers may communicate by the
cleft thus formed, as well as by the apertures in the lamellar
walls of the branchial pouches. The anterior part of the
supra-branchial chamber is divided into a right and left
cavity by the interposition of the mesosoma, on the sides of
which the apertures of the renal and generative organs are
situated. The products of these organs therefore readily
pass into these right and left cavities. The posterior part of
the supra-branchial chamber, into which these two lateral
divisions open, contains the termination of the rectum, and
receives the faeces, as well as the urinary and generative prod-
ucts: it is therefore a sort of cloaca. Its external opening is
usually termed the anal opening of the mantle cavity. The
margins of this opening may be produced into a tube which
is termed the anal siphon. In front of the anal, or rather
cloacal, opening, the margins of the mantle may be com-
pletely disunited. Very frequently, however, they are con-

1 The external gill-ponch is often smaller tlian tlie internal. In species of
Lucirui^ Cytherea, and Telliao.^ only one gill-pouch, the internal, is present.

18

4:10 THE ANATOMY OF INVERTEBRATED ANIMALS.

joined, so as to leave only an opening for tlie exit of the
foot, and another behind this, which is termed the branchial
opening. The edges of this aperture may be prolonged into
a tube, which is termed the branchial siphon. When a La-
mellibranch is in its natural element and undisturbed, the
valves of the shell gape sufficiently to allow of the free en-
trance or exit of water to or from the pallial cavity ; or, when
siphons exist, they are fully protruded. The cilia with which
the branchiae are beset work in such a manner as to drive the
water from the infra -branchial chamber, through the open-
ings of the branchiae, into the supra-branchial chamber. From
hence its only way of exit is by the cloaca and the anal
siphon, when the latter exists. In order to make up for the
water thus driven out, a new supply of water enters by the
interspace between the lobes of the mantle, which bound the
infra-branchial chamber, or by the branchial siphon. These
currents may readily be made obvious by allowing a stream
of finely-divided coloring matter to pass slowly toward the
branchial siphon of a Lamellibranch. It will be seen to be
swiftly sucked in, and after a very short time a colored stream
will flow out of the anal siphon. The same agency brings
the nutritive matters suspended in the water within reach of
the labial palpi, by which they are guided to the mouth.

Whatever form the branchiae may possess, they are sup-
ported by a chitinous skeleton, in the form of a partial or
complete investment to the transverse branchial vessels.

The mouth is bounded by lips, the angles of which are
usually produced on each side into two labial palpi. Some-
times the lips are represented by a circular fold produced into
numerous tentacula [Pecten). There are no organs for the
prehension or mastication of food. A wide and short gullet
leads into a stomach surrounded by the liver, which consists of
numerous caeca united into ducts which open into the stom-
ach. Very generally a diverticulum of the pyloric end of the
stomach contains a transparent rod-like body — the crystalline
style.

The intestine usually makes many convolutions, but, finally
reaching the middle line of the dorsal region of the body, it
terminates by the anus in the posterior part of the pallial
chamber. The heart lies in the region traversed by the termi-
nition of the intestine. It consists of an auricle and a ventri-
cle, or of a ventricle and two auricles, or may be divided into
two separate auricles and ventricles {Area). Aortic trunks
distribute the colorless blood to the bod v, whence it is carried

THE LAMELLIBRANCHIATA. 411

to a large median venous sinus; from this it passes through the
walls of the renal organs to the gills, and is returned from
these to the auricular division of the heart. ^ Very generally
the ventricle invests the rectum, but in OstroEa, Teredo^ and
A.no77iia, the ventricle is quite detached from the intestine.

The renal organs, or orgaiis of Bojanus^ are usually two
in number, often more or less united together, of a dark color,
situated beneath and behind the pericardium and in front
of the posterior adductor muscle, extending forward on each
side of the mesosoma, and traversed by such numerous blood-
channels, that they have a spongy texture. The walls of the
cavernous blood-sinuses are lined with cells which secrete the
urinary matters from the blood. These take the form of cal-
careous concretions, containing uric acid. The gland commu-
nicates at one extremity with the pericardium ; at the other,
it either opens directly on to the surface of the body, or into
a vestibular cavity which has an external aperture.

In Ostrcea and Teredo the renal organ seems to be present
in only a very rudimentary form.^

The mesodermal region, between the endoderm and the
ectoderm, is for the most part occupied by vascular, connec-
tive, and muscular tissues, and by the reproductive organs,
so that there is no large perivisceral space. But there is —
1. The large median sinus already mentioned, which receives
the blood returned from all parts of the bod}^ and is com-
monly termed the vena cava. 2. A spacious pericardial
chamber which incloses tlie heart. It is in communication
with the venous system, and, consequently, directly or in-
directl}'-, with the vena cava. 3. The cavities of the renal
organs, which usually freely communicate with one another,
while they open into ttie pericardium on the one hand, and on
the exterior of the body on the other. 4. In some Lamelli-
hranchiata^ canals open on the exterior of the body, especially
on the surface of the foot. In this way the blood-system is
placed in direct, though circuitous, communication with the
surrounding water. These so-called icater-vessels communi-
cate internally with the venous system, of which, indeed,
they seem to form a part. It is probable that all these cavi-
ties, taken together, represent the perivisceral cavity, pallial
sinuses, and pseudo-hearts of a Brachiopod.

1 The circulatory orsfans of the fresh-water Mussel have been very fully de-
scribed by Langer. ('^ Denkschrifteo der Wiss. Akademie, " 1855 and 1856.)
^ ^ See, for the structure of the renal organs and many other points connected
with the anatomy of the Lamellihranchiata, the series of valuable papers of La-
caze-Duthiers. (" Annales des Sciences Naturelles," 1854 to 1861.)

412 THE ANATOMY OF INVERTEBRATED ANIMALS.

Strong bundles of muscular fibres, usually unstriated, pass
transversely from one valve of the shell to the other, and
bring them together ; while they are divaricated by the

Fio. 118. — Anoclonta. — ^Vertical and transverse section of the body through the heart;
/, ventricle ; g, auricles ; c, rectum ; p. pericardium ; A, inner, i, outer gill; (/, ves-
tibule oi q, tbe organ of Bojauua; B, foot, AA, mantle lobes.

elastic reaction of the ligament. Of such adductor muscles
there may be either one or two. When there are two (Di-
my aria), the anterior adductor lies in front, and on the hae-
mal side, of the oesophagus; while the posterior adductor lies
in front, but on the neural side, of the rectum. Hence the
alimentary canal, as a whole, lies between those two muscles.
When only one adductor muscle exists [Mono my aria), it is
the posterior.

The foot is retracted between the valves of the shell by
two or three pairs of retractor muscles, of which the anterior
and posterior pairs are usually attached to the shell, close to
the anterior and posterior adductor impressions. The pro-
traction of the foot appears to be effected by the compression
of the blood by the intrinsic muscles of the walls of the meso-
soma and of the foot itself.

Each lobe of the mantle is attached to the corresponding
valve of the shell by a series of muscular fibres, the attach-
ments of which give rise to a linear impression, which runs
from one adductor to the other, and constitutes the pallial

THE LAMELLIBRANCHIATA. 413

li7ie. When the siphons are largely developed they have re-
tractor muscles, the insertions of which are so disposed as to
cause the posterior part of the pallial line to be more or less
deeply curved or angulated. Hence the distinction of integro-
palUate and sinupalliate as applied to Lamellibranchs which
have the pallial line evenly rounded or notched.

The cerebral ganglia lie at the sides of the mouth, and
are connected by a commissure, which passes in front of it.
They give branches to the anterior region of the mantle, to
the gills, to the anterior adductor muscle, to the labial palpi,
and to the parts about the mouth. ^\iii pedal ganglia are situ-
ated in the foot; or in the corresponding region on the neu-
ral side of the alimentary canal, w^hen no foot is developed.
Each is united by a commissure with the cerebral ganglion of
the same side, and gives off branches to the muscles of the
foot. The parleto-splanchnic ganglia lie on the neural face
of the posterior adductor muscle. The long commissures which
unite them with the cerebral ganglia usually traverse the
renal organ, and lie beneath the floor of the pericardium.
Each of these ganglia gives off a nerve to the branchia of its
side, and supplies the posterior and middle part of the man-
tle. This posterior pallial nerve may anastomose with the
anterior pallial nerve from the cerebral ganglion. The gan-
glia also furnish nerves to the posterior adductor muscle, to
the heart, to the rectum, and to the muscles of the siphons,
when the latter are present. Eyes are never developed in the
cephalic region of the Lamellibranchs, but, in many (e. g.,
JPecte7i)^ numerous simple eyes terminate papillee of the mar-
gins of the mantle. Auditory sacs are almost invariably at-
tached by longer or shorter peduncles to the pedal ganglia.

The Lamellihranchiata are usually dioecious, but some-
times hermaphrodite * (e. g., Cyclas, some species of Cardium
and Pecten, Ostroia, Clavacella^ and Pandora). The genera-
tive organs are ramified glands of simple structure and simi-
lar in both sexes, the ducts of which open into, or close to,
the renal organs.

The process of yelk-division* usually gives rise to smaller

* The testes and ovaria are distinct in the hermaphrodite Pectines. In Car^
diiim serratum, adjacent creca of the sexual gland contain spermatozoa or ova,
or both products may be developed in the same ctecum. In the common Oys-"
ter the genital caeca in any given individual are found to be either almost all
ovigerous or almost all spermigerous ; and it appears probable that the pre-
dominantly male precedes the predominantly female condition. See Lacaze-
Duthiers, " Organes genitaux des Acephales Lamellibranches." (" Annales
des Sciences Naturelles" 1854.)

^ See Lov^n, Archiv fur Naturge^chichte^ 1849. De Quatrefages, " Memoires
sur I'Embryogenie desTarcts." (" Annales des Sciences Naturelles," 1849.)

414 TOE ANATOMY OF IXVERTEBRATED ANIMALS.

and larger blastomeres, of which the former, as an epiblast,
invest the latter as a hypoblast. At the cephalic end of the
embryo of most Lamellibranchs, a velum, or disk with richly
ciliated edges, and, usually, a central tuft of longer cilia, is
formed. On the dorsal face of the embryo the integument
rises into a patch with raised edges, which is the rudiment of
the mantle. The separation of the shell into two valves,
united by an uncalcified hinge, must probably be ascribed to
the manner in which the calcareous matter subsequently
added to the shell is deposited. The foot appears as a median
outgrowth of the neural face of the embryo behind the
mouth. The branchiae have, at first, the form of separate fila-
mentous processes, which are developed from the roof of the
anterior part of the pallial cavity, at the point of junction of
the mantle with the mesosoma, and gradually increase in
number from before backward. In those Lamellibranchs
which have pouchlike gills, it appears that the processes
which are first formed become the outer lamella of the inner
gill-plate, their free ends uniting together; the inner lamella
of this plate is produced by the upgrowth of a thin lamina,
which subsequently becomes perforated, from the united ends
of these processes. The inner lamella of the outer gill is
formed of branchial processes, which grow out from the at-
tached ends of the first set ; and the outer lamella of this gill
is produced in the same fashion as the inner lamella of the
inner gill.^

Recent observations tend to show that in these, as in
other Invertehrata, the nervous ganglia are modified in-
growths of the epiblast.

The simplest form of development of the Lamellihranchi-
ata has been observed in Pisidium.'^ By the process of
cleavage, the vitellus is divided into a number of equal blas-
tomeres. The morula thus formed undergoes invagination,
and is converted into a gastrula. The blastopore, or aperture
of invagination, closes, and the epiblast, or ectodermal layer
of the embryo, growing much faster than the hypoblast, or en-
dodermal layer, the latter forms a small shut sac, the primi-
tive alimentary sac (or archenteron) attached to one point of
the inner surface of the much larger ectodermal sac. The

1 Lacaze-Duthiers, " Sur le cleveloppement des branchies des Mollusques
ac^phales Lamellibranches." (" Annales des Sciences Naturelles," 4, iv.)

2 Lankester, " On the Developmental History of the Mollusca." (" Phil.
Trans.," 1874.)

THE DEVELOPMENT OF LAMELLIBRANCHS. 415

mesoblastic cells appear to be derived both from the epiblast
and the hypoblast.

The mouth is formed by a depression of the ectoderm at
the anterior end of the body, which grows toward and opens
into the archenteron. The anus is dev^eloped at the opposite
end, in the region of the primitive invagination. On the
neural face of the embryo the foot grows out, while the mantle
appears on the opposite face ; and, in the centre of the man-
tle, a transversely oblong depression lined by elongated cells
is the," shell gland." In the median line this answers to the
ligament, and, at the sides, to the middle region of the future
valves of the shell ; but the precise share, if an}", which it
takes in the formation of these parts does not appear. I^lsi-
dlum has no velum.

The development of one of the fresh-water Mussels ( JJnio
pictoruni) has recently been worked out very fully by Rabl.^
The vitellus divides into two unequal masses, of which the
larger is termed by Rabl the " vegetative " and the smaller
the " animal " cell — somewhat inconvenient names, which may
be replaced by " macromere " and " micro'mere.^'^ Each of
these becomes subdivided, partly by ordinary fission, partly,
as in the case of the macromere, by a process of budding, into
blastomeres, of which those w^hich proceed from the macromere
long remain larger and more granular than those which pro-
ceed from the micromere. The blastomeres arrange them-
selves into a hollow sphere — the blast o sphere. This is a vesic-
ular morula, composed of a single layer of blastomeres, of
which those of one hemisphere have proceeded from the micro-
mere, and those of the other from the macromere. Two blas-
tomeres of the macromeral hemisphere remain much larger
than the rest. The macromeral hemisphere next undergoes
invagination, and its invaginated part becomes the hypoblast.
The two large blastomeres just mentioned, which are disposed
symmetrically, one on each side of the median plane at the
anterior margin of the area of invagination, become inclosed
between the hypoblast and the epiblast, and by their division
give rise to the mesoblast. This last, therefore, may be re-
garded as an indirect product of the hypoblast.

The endodermal sac formed by the hypoblast now loses its
connection with the region of the embryo of which it is an
invagination, and applies itself to the anterior wall of the
body, where an involution of the ectoderm, which gives rise

* C. Eabl, " Ueber die Entwickelungsgeschichte der Malermuschel," Jena,
1876.

416 THE ANATOMY OF INERVTEBRATED ANIMALS.

to the oral cavity, takes place. The greater part of the meso-
blastic cells become the adductor muscle, which is at first sm-
gle and answers to the posterior adductor of the adult. There
seems to be no shell gland. The shell appears at first as a
membranous cuticula, continuous from side to side, and there-
fore undivided into two valves. Subsequently it becomes
calcified and bivalve. The byssus gland is developed as an
involution of the octoderm at the posterior end of the body;
and the ventral hemisphere, or that opposite the shell, be-
comes divided by a deep median fold into the two lobes pf the
mantle on which the characteristic pencil-like papillae appear.
In front of the rudimentary mouth are two ciliated depres-
sions of the ectoderm, which are possibly the rudiments of the
nervous ganglia.

In JJ)iio and Anodonta the young are hatched in the outer
gill pouches of the parent, from which they are so dissimilar
that they were at one time considered to be parasites ( Glochi-
dium). The valves of the shell are triangular, and have in-
curved and serrated apices, by the help of which the larvae,
after they leave the parent, attach themselves to fishes and
other floating bodies. In this position they undergo a sort of
metamorphosis, and eventually fall off and sink to the bottom
as minute fresh-water Mussels.

On comparing the Lamellihranchiata with the JBracMo-
poda, it is obvious that the two have, in common Avith one
another and with the A.7inelida, the ciliated or veligerous
larval form. If the shell gland is, as Mr. Lankester suggests,
the homologue of the peduncular gland of Loxosoma and of
the Brachiopod larvge, it follows that the peduncle of the
Brachiopod corresponds with the centre of the pallial surface
of the Lamellibranch, and that the so-called dorsal and ven-
tral lobes of the mantle in the Brachiopod correspond with
the anterior and posterior halves of the mantle in the Lamel-
libranch. The Brachiopod hinge will therefore be transverse
to the axis of the body, while the Lamellibranch hinge is
parallel with it. If this comparison be just, however, the
three segments of the Brachiopod larva cannot answer to the
segments of an Annelid larva, but the two posterior seg-
ments of the Brachiopod larva must represent an outgrowth
of the hasmal side of the body ; and this would correspond
very well with the arrangement of the intestine in the artic-
ulated JBrachiopGda.

In the simplest forms of the Lamellihranchiata^ as Tri-
gonia^ Nucula^ and Pecten^ the mantle-lobes are almost, or

THE LAMELLIBRANCIIIATA. 417

completely, disunited from one another and from the branchiae,
and the latter are either simple plumes or have undergone
but little modification. The hsemal face of the body is short
relatively to its vertical height.

In most Lamellibranchs the hiEmal face of the bodv is
longer; the gills are lamellar, and the mantle-lobes are united
with one another and with the gills, so as to separate a supra-
branchial from an infra-branchial chamber (Anodo7ita). In
yet others, the posterior margins of the mantle are produced
backward into short s>2y^io?is, but the mantle-lobes remain
separate for the rest of their extent ( Cardhnn) ; in others,
the siphons are greatly elongated and the ventral margins of
the mantle-lobes unite, so as to leave only a small median
aperture for the foot {Pholas). In the most modified forms,
the body becomes more and more elongated, until, in Teredo^
it is completely vermiform, and the valves of the shell cover
but a very small portion of the body.

The foot is wanting as a distinct structure in Ostraca ;
while in Cardhnn and Trigonia it is a large muscular organ,
by the aid of which the animal is able to leap for some dis-
tance. The byssus n:ay be present in the young and absent
in the adult (e. g., Anodontci). It may have the form of strong
chitinous filaments {Mytilus)^ or of a plate of horny or shelly
texture {Arca^ A?2omia). The inequality of the valves at-
tains its maximum in the JH/jjicritidce, in which one valve
may have the form of a long cylinder, or cone, while the other
is a flattened plate. ^

The shells of Lamellibranchs are among the most abun-
dant of fossil remains in all epochs of the world's history. In
the Palaeozoic formations, however, the proportion of these
mollusks relatively to the Brachiopoda is the reverse of what
obtains at the present day, the latter being very numerous,
while the Lamellibranchs are comparatively scanty. The in-
tegropalliate are far m.ore numerous than the sinupalliate
forms in the older rocks. The Hippuritldm of the Cretaceous
epoch is the only family of ancient Lamellibranchs which is
extinct at the present day, and the only one which diverges
to any considerable degree from existing forms.

The Odo:n^tophoea. — In the Mollusks which belong to
this division, the mantle, always present in the newly-hatched
young, may abort in the adult condition. It is never divided

* For an excellent account of the LamellihrancJiiata from the conchological
side, see Woodward's " Manual of the MoUusca."

418 THE ANATOMY OF INVERTEBRATED ANIMALS.

into two lobes, though it may be slit or perforated where
it forms the wall of the branchial chamber {MaliotiSy Fis-
surella).

Very generally, the prosoma bears tentacula and eyes ;
and a distinct head being thus recognizable, these Mollusks
have been named Cephalophora^ in contradistinction to the
acephalous Lamellibranchs and Brachiopods.

The mantle commonly gives rise to a shell, which may
either be a more or less calcified cuticular product of the epi-
dermis, covering the outer surface of the mantle, when it
constitutes an external shell, as in the Lamellihraiichiata
and Brachiopoda / or it may be developed within a sac in
the interior of the mantle, as an internal shell. In neither of
these cases is it ever a bivalve shell divided into two lateral
portions.* Usually it is in one piece (univalve), but in one
group, the Chitonidoe^ it consists of a number of pieces (not
exceeding eight), arranged in longitudinal series along the
middle line.

Calcareous matter is very commonly diffused, in the form
of granules, through the connective tissue, and often takes
the form of spicula (e. g., Doris),

The mesosoma is generally prolonged into a muscular foot,
which may be provided with lateral appendages, the epipodla.
And, on the haemal aspect of the posterior portion of the foot,
a chitinous or shelly plate, termed the operculum, may be
developed. This operculum appears to be the analogue, if
not the homologue, of the byssus of the Lamellibranchs, and
is certainly not homologous with either of the valves of the
shell of the latter, which are pallial structures. The edge of
the mantle forms a free fold which nearly or entirely sur-
rounds the mesosoma; and in one genus, Dentalium, the
margins of the mantle unite for the greater part of their
length : in all the rest they remain free. A space is inclosed
between the lobes of the mantle and the mesosoma. Usually
this space is much larger on one face of the body, and con-
stitutes the pallial chamber. As a rule, the branchiae are
lodged in this chamber, and the anus opens into it.

In a very few Odontophora, the symmetry of the body is
undisturbed ; that is to say, the mouth and the anus are situ-
ated at opj)03ite ends of the axis of the body, and the hfBmal

_ \ The singular bivalve plates, termed AptycJnis, which occur in the Ammo-
niiidcB^ whatever their nature may be, are obviously not homolojjous with the
shell of ordinary Mollusks, which is represented by the chambered shell of
the cephalopod.

THE ODONTOPHORA. 419

face is not produced into a visceral sac (e. g., CJiiton^ Denta-
Ihim). But, in the great majority, such a visceral sac is
formed. In the Cephalopoda it coexists with bilateral sym-
metry, inasmuch as the mantle and the anus lie in the plane
which divides the body into two similar halves. But, in most
Odontophora^ the anus is twisted to one side (usually the
right), and in many it is situated, together with the pallial
chamber in which it is contained, on the anterior face of the
body.

The mouth lies at the anterior end of the body, on the
iiEemal side of the anterior part of the foot (except in the
Cephalopjoda), It may be provided with variously-disposed
jaws, or cutting-plates, of a chitinous or calcified substance.
But the structure which is most characteristic of the Odon-
tophora, and which is absent in only very few genera (e. g.,
Tethys, Doridlum^ Mhodope), is a peculiar rasping and some-
times prehensile apparatus, the odontophore^ or, as it is often
termed, the tongue, w^hich is attached to the floor of the mouth
(Figs. 119, 120).

This apparatus consists of a skeleton / of a siibradular
Tnemhrane^ which is continuous vfhh. the lining of the oral
cavity; of i\\e radula ; and of intrinsic and extrinsic 7nus-
cles.

The skeleton is composed of two principal masses of par-
tially fibrous, or completely cartilaginous, tissue [odonto-
phoral cartilages)^ which may be more or less confluent, and
are further united together in the middle line by fibrous and
muscular tissue. Their aliterior ends and oral faces are free
and smooth, and are usually excavated so as to present a
trough-like surface to the subradular membrane, which rests
upon them. Accessory cartilages may be added to these.
Behind, the subradular membrane is continued into a longer
or shorter sac, lined by a continuation of the buccal epithe-
lium. The radula is a cuticular chitinous product of the
epithelium of the subradular membrane. It is armed with
tooth-like processes arranged in one or many series ; and ad-
ditions are constantly being made to its posterior end, which
is lodged in the sac of the subradular membrane. Thus the
teeth are replaced from behind, as fast as they are worn away
by friction against the food which they rasp, at the anterior
end of the ribbon.

The intrinsic muscles of the odontophore are attached, on
the one hand, to the posterior and under faces of the odonto-
phoral cartilages, and, on the other, to the subradular mem-

420

THE ANATOMY OF INVERTEBRATED ANIMALS.

brane, some being inserted into its posterior and lateral por-
tions, and others into its anterior extremity', after it has
turned over the anterior extremities of the principal cartilages.

Fig. 119.— Buccinum nndatum.—A, rarlnla. B, one of the transverpe rows of teeth ;
a, anterior, 6, posterior end ; c, central, I, lateral teeth. (After Woodward, "Man-
ual of the MolluBca.")

Fig. 120.— a, Trcclius clnerarnis ; the median tooth and tlie teeth of the right hnlf
of one row of the radula. *B, Ci/prcea, Europom, one row of teeth ot the' radula.
(Woodward, ibid.)

Certain of the muscular bundles are also attached to the fore-
part of the odontophoral cartilages themselves. The con-
traction of these muscles must tend to cause the subradular
membrane, and with it the radula, to travel backward and
forward over the ends of the cartilages in the fashion of a
chain-saw, and thus to rasp any body against which the teeth
may be applied. When undisturbed, the radula is concave
from side to side, and the teeth of the lateral series, being
perpendicular to the surface to which they are attached, are
inclined inward toward one another. But when the intrinsic
muscles come into action, the radula, as it passes over the
ends of the cartilages, becomes flattened, and the lateral teeth
are consequently erected or divaricated. The extrinsic mus-
cles pass from the odontophore to the lateral walls of the
head, and protract or retract the whole apparatus. They

THE ODONTOPHORA. 421

may give the protruded extremity of the radula a licking mo-
tion, which is quite independent of the cliain-saw action due
to the intrinsic muscles/

The odontophore is developed very early, and it would be
interesting to know whether it exists in the young of those
few Odontophora in which it is wanting in the adult state.

Salivary glands are very generally present in the Odonto-
phora^^ and the liver is usually large.

As in the MoUusca in general, the blood-corpuscles are
colorless and nucleated. The blood plasma is red in JPla-
norhis.

The heart may be wanting (Dentalium), or it may resemble
that of the Lamellibranchs in having two auricles ( Chiton,
Hallotis), and even in being perforated by the rectum (Ha-
llotls, Turbo, JVerlta) ; most commonly it consists of a single
auricle and a single ventricle. In the Cephalopods, it is hard
to say whether the two or four branchio-cardiac trunks which
open into the ventricle should be regarded as veins or as
auricles. An accessory " portal " heart has been described
in Doris.^ Special respiratory organs may be wanting, their
place being taken by processes of the body, or by the walls
of the mantle cavity, or by the general surface.

The branchiae, when present, are numerous lamellar pro-
cesses, or from one to four plume-like gills. Aerial respira-
tion is effected by the walls of a pulmonary sac, which is a
modification of the pallial cavity.

The presence of renal organs, in the form of one or more
sacs situated close to the heart, open to the exterior on one
side, and, on the other, in relation, usually by means of a
glandular structure, with the returning current of blood, is
very general ; and, in many cases, these renal sacs communi-

^ In my memoir " On the Morphology of the Cephalous Molhisca " (" Phil.
Trans.," 1852) I described the chain-saw action of the odontophore, as I ob-
served it in the transparent Firoloides and Atlanta^ while livius^. But, as Tro-
schel has remarked in his excellent monograph (" t)as Gebiss clerSchnecken,"
erste Lieferung, pp. 19, 20j 1856), I did not sufficiently dwell on the frequency
and importance of the licking action produced by the extrinsic muscles. I am
still of opinion, however, that this action cannot be rightly described as a
movement of the radula following secondarily upon that of the cartilages, inas-
much as it is a motion of the whole odontophore. On the other hand, it may
be, as has been suggested to me by Mr. Geddes — who at my suggestion has
undertaken a reexamination of the structure of the odontophore — that the flex-
ure of tlie anterior ends of the odontophoral cartilages, by the intrinsic mus-
cles inserted into them, plays an important part in the motion of the radula.

2 In Dolium the salivary secretion contains free sulphuric acid.

3 Hancock and Embleton, " On the Anatomv of Doris." (" Phil. Trans.,"
1852.)

422 THE ANATOMY OF INVERTEBRATED ANIMALS.

cate directly with the blood sinuses through the pericardium.
In many Pteropods and Heteropods they are rhythmically
contractile.

As in the LainelUhranchiata^ so in many Odontophora,
simple or branched canals traverse the substance of the foot
and open externally by a more or less conspicuous pore, which
is usually situated upon its inferior face. These aquiferous
canals, as they have been termed, appear, in many cases, to
open by their inner ends into the blood sinuses, and thus to
establish a direct communication between the blood and the
surrounding water. In species of JPyrula, Agassiz found that
colored fluids injected into the pore passed into and filled the
blood-vessels generally. But it may be doubted whether
these canals should be regarded as a special system of ves-
sels, rather than as blood sinuses which open externally.

The arrangement of the centres of the nervous system in
Dentalium ^ most nearly approaches that which exists in the
Lamellibranchiata. Two cerebral ganglia lie close together
on the hnemal side of the oesophagus. A long commissural
cord connects each of them with one of the pedal ganglia,
which are also closely united. A second long commissure
passes backward from the cerebral ganglia, and often presents
a ganglionic enlargement at its origin. It unites with one of
two ganglia, situated close to the anus, and connected, in
front of it, by a rather long transverse commissure. The
nerves distributed to the posterior half of the mantle are
given off from these ganglia, and those to its middle region
from the anterior end of the commissure or its ganglionic en-
largement. There seems no reason to doubt that the ganglia
close to the anus, together with the ganglionic enlargements
at the anterior ends of the commissures which connect them
with the cerebral ganglia, correspond with the parieto-
splanchnic ganglia of the Lamellib ranch s, and that the cere-
bral and pedal ganglia are the homologues of those so named
in the latter Mollusks.

In addition to this approximation of part of the gangli-
onic mass of the parieto-splanchnic system to the cerebral
ganglia, Dentalium differs from the Lamellibranchs and re-
sembles other Odontophora, in the possession of a system of
buccal nerves, which arise from the cerebral ganglia, and in
which minute ganglia are developed. The nerves which pro-

» See Lacazc-Dutliicrs, " Organisation du Dcntale."

THE ODONTOPHORA. 423

ceed from the buccal ganglia are distributed to the odonto-
phore and its muscles.

In ether Odontophora^ the two cerebral and two pedal
ganglia, with their commissures, are alwaj's to be recognized ;
but the number of the ganglia which represent the parieto-
splanchnic system may be increased, and the anterior ganglia
of this system may attain a large size, and may come into
close relation not only with the cerebral but with the pedal
ganglia.

In Lymnceiis palustris^ for example, there are five such
ganglia situated close to the cerebro-pedal ring. The most
anterior of these, on each side, is united with both the cere-
bral and the pedal ganglion of its side, and appears, indeed,
like an enlargement upon a second commissure between those
two ganglia. The ganglia which constitute the second pair
are united, in front, by a short commissure, with the preced-
ing; and, behind, with the fifth or azygos ganglion. The
second pair of ganglia give off the neives to the right and
left sides of the mantle respectively.

In Limax^ and apparently in the terrestrial Pidmonata
generally, the arrangement is essentially the same, except
that all the ganglia of the parieto splanchnic S3'stcm coalesce
into one mass, between which and the pedal ganglia the aorta
passes.

In Hallotis^ on the other hand, while the anterior parieto-
splanchnic ganglia are situated close to the pedal ganglia,
and are connected with them and with the cerebral ganglia
in such a manner as to give rise to an apparent second cere-
bro-pedal commissure, the ganglia which represent the second
pair in Lymnceus are situated at the base of the branchiae,
and are united by a long commissure with one another, and
also with the anterior paricto-splanchnic ganglia. Of the
latter commissures, that from the left branchio-pallial gan-
glion goes to the right anterior parieto-splanchnic ganglion,
and vice versa.

With respect to the position of the cerebral and pedal
ganglia in the OdontopJiora^ the commonest arrangement is
that in which the cerebral ganglia are supra-oesophageal, and
are connected by two longer or shorter commissures, on each

» Compare Lacaze-Duthiers, " Du svsteme nerveux des Mollusques gas-
teropodes pulraones aquatiques" ("Arch, de Zoologie," 1872), and ths numer-
ous figures of the arrangement of the cerebral ganglia of the nervous system
given in his memoir on the otocysts. (Ibid.)

' See Lacaze-Duthiers, " Sur le systeme nerveux de Haliotide."

424 THE ANATOMY OF INVERTEBRATED ANIMALS.

side, with the pedal and anterior parieto-splanchnic ganglia,
both of which are infra- or post-cesophageal. But in many
cases (most Kudibranchiatd) the pedal and parieto-splanch-
nic ganglia are approximated to the cerebral ganglia (the
latter being supra-oesophageal), and are united by long sub-
cesophageal commissures. In others, as in most Pteropoda,
the pedal and parieto-splanchnic ganglia are sub-cesopbageal;
while the cerebral ganglia, brought close to them, are united
by a supra-cesophageal commissure.

Accessory ganglia are frequently developed in the region
of the heart and branchiae, on the nerves of the parieto-
splanchnic system.

A complicated system of visceral nerves is distributed
over the whole length of the alimentary canal, the genital
organs, and various parts of the vascular system, in many
Odontophora.^

Two auditory vesicles usually exist, and very generally ap-
pear to be sessile upon the pedal ganglia. In the Heterojjoda,
in many JS'udibranchiata, as shown by Hancock, and in nu-
merous genera of JBranchio- and Fidmo-gasUropoda^ which
have been carefully examined by Lacaze-Duthiers,^ however,
there seems to be no doubt that the auditory nerves arise
from the cerebral ganglia, even though the vesicles may be
situated close to the pedal ganglia.

' Olfactory organs certainly exist in the Cephalopoda in
the form of saccular involutions of the integument near the
eyes ; and it is very probable that the integument of the ten-
tacula, or of the lips, may subserve the same function in the
Gasteropods.

Eyes are generally present, and are limited to two, situ-
ated in the head. They resemble the vertebrate eye in struct-
ure, so far as they possess a concave retinal expansion, and
usuallv, in front of this, a vitreous humor, lens, and cornea.

1 See especially Hancock and EmHeton, " The Anatomy of Doris." (*' Phil.

Tran?.," 1852.) -^ , . , „

2 " Otocvstes des Mollusques." (''Archives de Zoologie ExpCnmentale,"
1872.) In the memoir the origin of the acoustic nerves from the cerebral
ganglia is detennined in so many Pulmo-gadempoda {Limax^ Arion^ Testacella,
ClausiUa, Zoniteif, Helix, Succinea, Physa, Lymnoevs, Jncyhs) and Bravchio-
gaderopoda {Neritina, Paludina^ Cyclostoma^ Pileopsv>, Caiyptraa^ JSatica^
JVassa, TrocJtu.o, Murex, Cassidaria, Fttrptira, Patella, Haliotis, Philine, Aply-
sia, Lamellaria), that there is a large basis for the generalization that this
mode of oriirin is universal. Moreover, according to Lacaze-Duthiers, the
same law holds good for the Cephalopoda. Such being the case, the question
BUL'gests itself whether the connection of the nerves of the otocysts with the
pedal ganglia, which obtains universally among the Lamellibranchs, indicate*
their real or only their apparent origin.

THE ODONTOPHORA. 425

But they differ from the eyes of Vertehratay and resemble
those of other invertebrated animals, in that the structures
which answer to the rods and cones are situated on that face
of the retina which is turned toward the light, while the fibres
of the optic nerve traverse the pigment layer to reach them.

The reproductive organs of tlie OdontopJiora present very
great diversities of structure. They may be either dioecious
or monoecious, and each type of reproductive organs may pre-
sent various degrees of complexity. Of the dioecious repro-
ductive organs there are two cliief forms : the one in which
the duct of the ovarium or testis is continuous with the gland;
and the other in which the duct opens into a sac, into which
the ova or spermatozoa are set free by the dehiscence of the
follicles in which they are developed. The latter arrange-
ment is met with in the Cephalopoda ^ the former appears
to prevail among all the other dioecious Odontophora,

In these, the racemose generative gland is usually situ-
ated close to the liv^er. In the female, the oviduct ordinarily
presents a uterine dilatation toward its termination, which is
generally situated in the pallial cavity on the right side of
the body. In some rare cases (Paludina, Keritina\ a dila-
tation or a special vesicular appendage of the uterus may
serve as a vesicida seminalis / and in Pahidina, according
to Leydig, an albumen-gland opens into it.

A penis is not always present. When it exists, it is a
muscular process of the mesosoma, to which the semen may
be led from the opening of the vas deferens by a groove ; or
it may be traversed by the vas deferens which opens near, or
at, its apex.

In all the monoecious Odontophcra w^hich have as yet been
thoroughly examined, there is a generative gland termed the
ovotestis, in which both spermatozoa and ova are produced.
Only in the anomalous genus Hhodope (Kolliker) are the
spermatozoa and ova formed in distinct caeca ; in all the rest,
each caecum is hermaphrodite, the spermatozoa and the ova
being usually developed in different parts of the caecum. The
duct of the ovotestis may remain single to its termination at
the genital aperture, or become only incompletely divided
into two semicanals {^Pteropoda^ Pleuropliyllidia^ JJmhrella,
Aplysia) ; or it may become, at first partially, and then com-
pletely, divided into an oviduct and a vas deferens {JS^udi-
hranchiata^ Pleurohranrhia^ Puhnonata),

In the former case there is but one genital aperture. The
common duct usually receives the secretion of a uterine gland

426 THE ANATOMY OF INVERTEBRATED AXIM.VLS.

which may take the form of a special albumen gland, and a
spermatheca opens into it near its outer extremity; while, on
the male side, a vesicula seminalis and an eversible penis
may be added. The penis, however, may be distant from the
genital opening, and then a groove on the side of the body
leads to it [Aplysla). In the latter case there are two geni-
tal apertures, one for the male and one for the female organs,
though thev may open into a common vestibule. The penis
is an eversible involution of the integument, on which the vas
deferens opens. A prostate gland is usually connected with
the latter, and, near its opening, there may be a saccular ap-
pendage, in which a hard pointed body, the spiculum amoris,
is contained (Doris, Helicidce). An albumen-gland opens
into the uterus, and a spermatheca is connected with the
vagina.

Spermatophores, by the aid of which the spermatozoa are
transferred into the female organs, occur in the Cephalopoda,
and in the Palmonata. In the latter they are grooved bands,
or incomplete tubes of hardened mucus secreted by the penis,
which become filled with spermatozoa during copulation ;
while, in the former, they are closed cases which may have a
very complex structure.

In the great majority of the Olontophora the young
leaves the Q'g'g as a vellger, very similar to that of the Lamel-
llbranchlata. The velum usually becomes bilobed, and some-
times {Ileteropodd) its margins are produced into many ten-
taculiform processes ; and, in all Pteropoda and Pranchio-
gasteropoda, whether the adult possess a mantle and a shell
or not, the larva is provided with both, the shell being at first
a simple conical symmetrical cap, developed in the middle line
of the mantle. The eyes make their appearance behind the
velum, and the tentacles in front of or upon it.

While the course of the development of the embryo in
the Odontophora presents a general uniformity, there are
wide differences in detail.

In Paludina^ the blastomeres produced by yelk-division
are of equal size. They arrange themselves into a vesicular
morula, which undergoes invagination and becomes a gas-
trula of the simplest type. The aperture of invagination
(blastopore) becomes the anus, while the mouth is formed
by an involution of the ectoderm of the anterior end of the

» Lankester, " On the Coincidence of the Blastopore and Anus in Paludina
vivipai'a.'^ {Quarterly Journal of Microscopical Science, 1876.)

THE DEVELOPMENT OF THE ODONTOPHORA. 427

body, which extends toward and eventually opens into the
blind end of the archenteron or primitive alimentary sac. A
ciliated velum is developed on the hsemal side of the mouth,
and a " shell gland " appears in the centre of the area which
gives rise to the mantle.

In LymncBUS,^ also, cleavage ends in the production of
blastomeres of equal size, whether with or without a transi-
tory stage of inequality, and the vesicular morula undergoes
invagination to give rise to the archenteron. The blastopore
is elongated, and it appears to be likely that its ariterior and
posterior ends may coincide w4th, if they do not give rise to,
the mouth and anus respectively.

In most Olontophora^ the process of yelk-division goes
on unequally, and results in the production of large and small
blastomeres (macromeres and micromeres). The latter form
a layer which gradually extends over the macromeres and in-
closes them. Obviously, this comes to the same result as
invaorination ; and the included macromeres and their progeny
either become converted into the archenteron with its ap-
pendages, and more or less of the mesoblast, or a portion of
them may serve as food-yelk.

In the Pteropoda and Heteropoda^ and in JVassa, Natica^
and lusus^ the blastopore, or aperture circumscribed by the
edges of the micromeral layer as it grows round the macro-
meres, closes, but corresponds in position to the invagination
of the ectoderm which gives rise to the future mouth ; and
the anus is a new formation.

In such land Pahnonata as Limax^ the process of yelk-
division gives rise to macromeres and micromeres, and the
latter inclose the former. What becomes of the blastopore
is not clear, though I am inclined to think that it corresponds
in position with the mouth. The latter is seen very early as
a funnel-shaped invagination of the epiblast bounded bj' lat-
eral lips. Behind it, the foot grows out and rapidly attains
a considerable size. Its posterior extremity becomes flattened
from above downward, and converted into an orbicular ap-
pendage, the opposite walls of which are connected by retic-
ulated muscle-cells. This appendage undergoes rhythmical

'Lankester, " Observations on the Development of the Pond-Snail" {Quar-
terly Journal of Microscopical Science., 1874), and C. Rabl, " Die Ontogenie der
Siisswasser Pulmonaten" {Jen. Zeitschrift., 1875).

3 Fol, " fitudes sur le developpement des MoUusques." (" Arch, de Zoologie
experimentale," 1875, 1876.)

3 Bobretskv, " Studiea fiber die embryonale Entwickelung der Gasteropo-
den." (" Archiv f. Mikr. Anat.," 1876.)

438 THE ANATOMY OF INVERTEBRATED ANIMALS.

movements of dilatation and contraction. The macromeres
form a large mass inclosed within a spheroidal dilatation of
the greater part of the bcemal wall of the body, which deserves
the name of yelk-sac even better than the structure so named
in the Cephalopoda^ inasmuch as it more nearly corresponds,
morphologically, with the vitelline sac of vertebrated animals.
Between this sac and the foot the small remainder of the
haemal wall becomes converted into the mantle.

The walls of the vitelline sac undergo contractions which
sometimes, but not always, alternate with those of the pedal
appendage. On each side of it appears the "primitive kid-
ney," consisting of a curved elongated series of cells within
which concretions are developed, and terminating in a duct
which opens on the posterior face of the vitelline sac, close to
the mantle. The exact mode of origin of the alimentary
canal has not been made out ; but, in any case, only a very
small portion of the endodermal cells can take part in its
formation, and the archenteron is, at first, a sac which nearly
fills the small projection formed by the rudimentary mantle.
The oral involution of the ectoderm gives rise to the odon-
tophore, and extends across the base of the foot, to open,
eventually, into the archenteron.

The fold of the mantle which overhangs the iqespiratory
aperture makes its appearance very early; and, immediately
behind it, the intestine is visible as a short tube, which ex-
tends from the archenteron to the surface, but does not, at
first, open there.

As development proceeds, a movement of the macromeric
part of the vitellus takes place in exactly the opposite direc-
tion to that of the food-yelk of the Cephalopoda ; that is to
say, from the vitelline sac into the constantly enlarging foot.
The alimentary canal accompanies it, the anus alone remain-
ing in its primitive position. The constantly lengthening
alimentary canal becomes disposed in folds ; between these
the macromeric part of the vitellus, which gradually forsakes
the diminishing vitelline sac, disposes itself around the coils
of the intestine. Eventually, for the mcst part, it becomes
converted into the liver.

The rudimentary shell first makes its appearance in the
form of a few subcrystalline calcareous plates, on the inner
side of the ectoderm.*

The development of Helix is similar to that of JLiinax ;

* Compare Gegrcnbaur, " Zur Entwickeluns^sgescliichte der Laiid-Gastero-
poden." {Zeitschrift fiir Wise. Zoologie^ 1852.)

THE DEVELOPMENT OF THE ODOXTOrHORA. 429

but the intestine passes into the large visceral sac instead of
into the cavity of the mesosoma. The shell is stated by
Gegenbaur to be at first internal, as in Limax. In neither
case has the relation of the shell to the shell-gland been
determined.

The process of development appears to present a consider-
able range of variation in the I^ulnionata. Semper* states
of a species of Vaginulus, that, after the process of cleavage,
the embryo assumes the form of a cylinder, at one pole of
which the rudiments of the tentacula and of the lips appear ;
while, at the sides, a longitudinal ridge indicates the edge of
the mantle, and marks off the more convex pallial region
from the flat foot. No shell is formed.

In LymncBus^ as has been already stated, the vitellus
undergoes complete division, and the resulting vesicular
morula undergoes invagination to produce the hypoblast.
Only the middle part of the archenteron becomes the alimen-
tary canal, however. The lateral portions, which take on the
form of rounded sacs, may not improbably, as in the Bracliio-
pods, give rise to the perivisceral cavity, though this has not
been proved. The mouth is produced by the formation of an
opening in the coalesced endoderm and ectoderm, at a point
near the anterior end of tlie body. Upon each side of the
mouth a transverse ciliated ridge of the ectoderm is developed,
and represents the edge of the velum in other molluscan em-
bryos. Behind this, and on the opposite side of the embryo
to that on which the mouth is placed, a raised patch of the
ectoderm represents the mantle. The foot commences as a
papilla immediately behind the mouth. An involution of the
centre of the pallial ectoderm gives rise to a shell-gland, but
the proper shell is developed, independently of this, as a cu-
ticular secretion from the whole surface of the mantle.

Thus the embryo of JLymnceus possesses an incompletely
developed velum, and is, in all essential respects, similar to
the veligerous embryo of Lamellibranchs, Pteropods, and
Gasteropods ; while the Slugs and Land-snails have neither
the velum (unless it be represented by the anterior contrac-
tile sac) nor the external embryonic shell.

The development of the Cephalopoda is very unlike that
of other Mollusks, and will be dealt with under the head of
that group.

» " EntwickelunffsgeseWehte der AmpuUaria polita.''''

2 Lankester, " Observations on the Development of the Pond-Snail, Lym-
n(tvs stagnalis.''^ (^Quarterly Journal of Microscopical Science^ vol. xiv., New
Series.)

430

THE ANATOMY OF IN VERTEBRATED ANIMALS.

The lowest forms of the Odontophora are the PolyplacO'
phora^ or Chitonidce, and the Scaphojwda, or DentalidcB.
The bilateral symmetry of the body is completely, or almost

Pig. 121.— I. Chiton Wossnessenskii. (After Middendorf.)

II. Chiton dissected to show o, the mouth; g, the nervous rin^:; ao, the aorta; c, the
ventricle ; c', an auricle; bf, the left branchiae ; od, the oviducts. (After Cuvier.)
IIL, IV,, V. Stages of development of Chiton cinereus. (After Loven.)

completely, undisturbed, while the haemal wall is flat, or near-
ly so, and there is no visceral sac.

The Polyp lacophora. — The Chitons (Fig. 121, I.) are
elongated, slug-like animals, having the mouth at one end of
the body, and the anus at the opposite extremity. A rounded
lobe surmounts the mouth, but it bears no eyes nor tenta-
cula, and there is no definite head. The edges of the mantle
are thickened, but little prominent, so that the pallia] cavity
is not much more than an elongated groove, beneath and
internal to the thickened edge, which is sometimes beset
with setae. In the region in which these setae occur, the surface
of the mantle is covered by a thick cuticula. The setae, which

THE SCAPHOPODA, 431

may be merely cbitinous or completely calcified, or partly in
the one and partly in the other condition, are developed in
sacs lined by the cells of the ectoderm.' In the pallial groove
lie the short lamellar processes which represent the branchiae.
The shell is unlike that of any other Mollusk. It consists of
eight, transversely elongated, symmetrical pieces, arranged
one behind the other, overlapping in such a manner that the
posterior edge of the one covers the anterior edge of the
next, and articulated together. Sometimes the valves are
partially or completely inclosed in the mantle. The heart,
composed of a single median ventricle and two lateral auri-
cles, is placed in the middle line, above the rectum, at the
posterior end of the bodv. The aorta is continued forward
from its anterior end, while the auricles receive the blood
from the branchiae. In Chiton, piceus, according to Schiff,^
each auricle communicates by two openings with the ven-
tricle, and the two auricles are united behind. The repro-
ductive organ is median and symmetrical, and its two ducts
open on each side of, and not far from, the anus.

The embryo leaves the e^^ as an oval body, surrounded
near its anterior end bv a circular ciliated band, behind which
an eye-spot appears on each side (Fig. 121, III.). The seg-
ments of the shell appear while the young Chiton is still
locomotive, and the disk in front of the ciliated band becomes
converted into the lobe above the mouth (Fig. 121, IV., V.).
The Chitons have existed from the Silurian epoch to the
present day, apparently' with very little modification.

The Scaphopoda.^ — In Dentaliiun^ the shell is elongated,
conical, and curved, like an elephant's tusk, with the apex
broken oflF, and it is open at both ends. The animal has a
large mantle corresponding in form with the shell, and also
open at both ends, the margins of the anterior, larger, aper-
ture being much thickened. The mouth, placed at the extrem^
ity of a sort of cup, the margin of which is fringed with pa-
pilhie, is situated far behind the anterior opening of the man-
tle. Behind the oral cup, where the body joins the mantle,
is a transverse muscular ridge, from which proceed a great

1 Remcke, "Beitriiire zur Bildungsgescliichte der Staclieln, u. s. w." {Zeit-
%chHftfnr wU>^enschaftUche Zooloaie.)

2 Zeitschrift fi'ir tvissenschaftliehe Zooloaie, 1858.

3 A very complete and accurate account of the orsranization of Dentalium is
given in the monograph of Lacaze-Duthiers, " Histoire de I'oreranisation, du
developpement, des moeurs et des rapports zoologiques des Dentales," 1858.

432 THE ANATOMY OF INVERTEBRATED ANIMALS.

number of long tentacles. These protrude through the an-
terior opening of the mantle, and play the part of prehensile
organs. Behind and below the oral cup the very long sub-
cylindrical foot proceeds. Near its extremity are two lateral
fleshy lobes v/hich perhaps correspond with the epipodia of
other Mollusks. The oral cup leads into a buccal chamber
containing the odontophore, whence the oesophagus passes
to the stomach. The liver consists of two symmetrically-
branched divisions ; and the intestine, after becoming coiled
upon itself, ends in a prominent anal papilla, in the median
line, behind the root of the foot. There is no heart, but the
blood fills spacious sinuses. There are no special respiratory
organs distinct from the wall of the pallial cavity. The two
renal organs open one on each side of the anus. The renal
blood sinus communicates directly with the pallial cavity by
two apertures, situated close to those of the renal organs.
In the nervous system, the commissures of the parieto-
splanchnic ganglia pass directly to the cerebral ganglia, as
in the Lamellibranchs. The sexes are distinct, and the geni-
tal gland is single and symmetrical, though its duct opens
into the right renal organ. The embryo is at first surrounded
b}^ a number of ciliated rings, its anterior end presenting a
tuft of long cilia. By degrees the cilia become restricted to
the edges of a disk, into which the anterior end of the embryo
expands, and which represents the pra^-oral ciliated velum of
the Lamellibranchs. The mantle now appears on the dorsal
aspect of the body, behind this disk. Its ventral edges are
free, and it secretes a shelly plate of corresponding form.
But, as development advances, the edges of both manile and
shell unite in the median ventral line, leaving the anterior and
the posterior ends open.

The Scaphopoda are an ancient group, remains of them
occurring as far back as the Devonian epoch.

The higher Odontophora (or the Gasteropoda^ Pteropoda,
and Cephalopoda of Cuvier) fall into two divisions, according
to the structure and arrangement of the parts of the foot.
In the one division (the Gasteropoda and Pteropoda) it may
be a simple disk, or it may be divided into three portions —
an anterior (the propodhiw)^ a middle (the inesopodhmi)^
and a posterior (the metapodlum) ; and it may be still further
complicated by the development from its sides of muscular
expansions — the epipodia. But, whatever the shape of the
foot in these Mollusks, its margins are not produced into

THE GASTEROPODA AND PTEROPODA. 433

prehensile processes, and its antero-lateral portions do not
extend beyond the sides of the head, and unite in front of the
mouth.

In the other division (the Cephalopoda)^ the margins of
the foot are produced into prehensile processes or arms, and
the antero-lateral regions of the foot extend over, and unite
in front of, the mouth, in such a manner that the latter is
placed in the centre of the discoidal foot.^

In the former division — that is, in all Pteropoda — in all
those Gasteropoda which breathe the air dissolved in water
(Srayichiogasteropoda), and in some of those which breathe
air directly [Palmogasteropoda)^ the embryo is, as in the
Scaphopoda and Polyplacophora, a veliger ; or, at any rate,
it has ciliated bands which subserve locomotion. But in the
Cephalopoda no such velum is formed, and the animal ac-
quires the general characters of the adult before leaving the

A shell-gland is often, if not always, present in the em-
bryo of the higher Odontophora ; and, in all Pteropods and
Branchiogasteropods, the mantle secretes a cuticular shell,
which, however, may exist only during the larval condition.

If the arrangement of the alimentary canal in a Cephalo-
pod, or a Pteropod, be compared with that which obtains in
such a Branchiogasteropod as Atlanta^ it will be observed
that, in the former, the oesophagus enters the outgrowth of
the haemal region of the body which constitutes the visceral
sac, to reach the stomach ; and that the intestine passes, at
an acute angle with the anterior portion of the alimentary
canal, along the posterior face of the visceral sac, to end in
the pallial chamber, which is situated on the posterior face of
the body. The pedal ganglia consequently lie between lines
traversing the anterior and the posterior divisions of the ali-
mentary canal respectively ; and hence the alimentary canal
has a neural flexure^ or is bent toward the neural face of the
body.

In Atlanta^ on the other hand, the intestine, when it leaves
the stomach, passes along the anterior face of the visceral
sac, to reach the pallial cavity, which is situated on the an-
terior face of the body. Hence lines traversing the two di-
visions of the alimentary canal would inclose not the pedal

* See^ for a valuable discussion of the horaolosries of the arms and the funnel
of the Cejahalopoda^ in which the view here taken is ably, though I do not
think satisfactorily, controverted, Grenacher, "Zur Entwickelunjjsgeschichte
der Cephalopoden." {Zeitschriftfur wiss. Zoologie, 1874.)

19

434 THE ANATOMY OF INVERTEBRATED AXIMALS.

but the cerebral ganglia. In other words, the intestine is
bent in the opposite direction to that which it takes in the
Cephalopod, or has a haemal flexure.^

The haemal flexure of the intestine is very characteristic
of the Brayichiogasteropoda^ and is completed at an early
stage of their development.

In such a slightly-modified Odontophoran as Chiton, the
heart presents its normal position in the posterior region of
the haemal face of the body, and has its aortic end turned for-
ward. Although the branchiae are situated at the sides of
the body, the blood which passes through them must take a
backward course to reach the heart; and thus the branchiae
may be said to be virtually behind the heart, and the animal
is truly opisthohranchiate. It appears to be otherwise with
such a Gasteropod as J^uccimcm, in which the gills lie actual-
ly in front of the heart, and the animal is therefore said to
be prosohranchiate. It must be recollected, however, that,
strictly speaking, no Odontophoran is othisr than opistho-
hranchiate. The anus represents the morphological hinder
end of the body ; and the auricle of the heart, into which the
current of blood from the branchiae passes, is never, morpho-
logically, posterior to the branchiae.

This is perfectly obvious in the Cephalopoda. In the
position which tlie animal frequently assumes and in which it
is ordinarily represented, the gills are in front of the heart.
But if the Mollusk is placed in its morphologically correct
position with the oral face of the arms downward, it will at
once be seen that what is commonly called tlie ventral face
of the animal is the posterior half of its haemal face, and that
the heart lies, morphologically, anterior to the branchiae.

In such Branchiogasteropods as are prosohranchiate, the
gills come to lie in front of the heart in consequence of their
having followed the twisted intestine forward and to the
haemal side of the bod v.

The Pteropoda.' — In this group of small pelagic animals
there is no distinct head, the eyes and the ordinary tentacles
remaining rudimentary. Auditory sacs are attached to the
pedal ganglia. Sometimes (^Pneumodernion) two eversible

' Iluxlev, "On the Morphology of the Cephalous Mollusca." ("Phil.
Trans.," 1852.)

2 See Rang and Souleyet, " Histoire naturelle des Mollusques Ptiropodes ; "
and Gegenbaur, " Untersuchungen uber die Pteropoden und Heteropoden,"
1855.

THE PTEROPODA. 435

spinose tentacular organs are developed at the sides of the
mouth, and, in addition, two acetabuliferous tentacles take
their origin on the inner side of a cup-like hood, which sur-
rounds the anterior end of the body.^ CymhuUa is stated
to possess no radula. The epipodia are large muscular ex-
pansions, by the flapping of which the Pteropods swim ; but
the rest of the foot is always small, and often rudimentary,
in correspondence with the small size of the neural face of the
bodv.

The hsemal face, on the contrary, is always produced, as
in the Cephalopoda^ into a relatively large visceral sac ; and
in some (the Thecosomatd) this visceral sac is coextensive
•with the mantle, which is protected by a shell. In others
(Gymnosomata) the mantle early disappears, and there is
no shell. In Cymhulia^ the delicate transparent chitinous
shell is internal, and is invested by an epithelial layer derived
from the mantle. In Spirialis^ the foot bears an operculum.
Chromatophores similar to those of the Cephalopoda occur in
Tiedemannia,

In the Thecosomata^ the free lobe of the mantle, which
incloses a spacious pallial cavity, usually lies on the posterior
aspect of the visceral sac, as in the Cephalopoda, and the
rectum terminates in it, on one side of the middle line. In
these there is a simple neural flexure of the alimentary canal,
as in the Cephalopods, although the turning of the rectum to
one side destroys the symmetry of the body. In Limacina
and S^nrialis, the intestine appears to be bent round to the
anterior face of the visceral sac, the mantle-cavity accom-
panying it, so that the opening of the mantle is placed on
the anterior, instead of on the posterior, face of the visceral
sac. There are no distinct gills in the Thecosomata, but the
lining of the mantle-cavity subserves the function of respira-
tion, and is sometimes produced into folds, which doubtless
aid in the performance of that function. Processes of the
body, to which the office of gills is ascribed, are found in
some Gymnosomata (Pneumodernxon Sipongohranchia^.

The heart consists of a single auricle and a single ventricle.
The auricle lies close to the pallial cavity, and receives the
aerated blood from its walls. The ventricle is sometimes
directed forward (as in all Gymnosomata), and sometimes

'^See^ for the somewhat similar arran^^ements In Clione, Eschriclit, " Ana'
tomische Untersuchuugen tiber Clione boreulis^^'' 1858 ; and Macdonald, " On
the Zoological Characters of the Living Clio cxudata.''^ (" Traus. Eoval Society
of Edmburgh," 1863.) \ . j

436 THE ANATOMY OF INVERTEBRATED ANIMALS.

backward, sa that nearly-related forms are sometimes opistbo-
branchiate, sometimes prosobranchiate. The branches of the
aortic trunk soon terminate in lacunae, by which the blood is
conveyed back to the walls of the mantle-cavity. The renal
organ is a contractile sac with delicate walls, which opens on
one side into the pallial chamber, and on the other into the
pericardial sinus.

The Thecosomata have the principal ganglia concentrated
around the gullet — the cerebral ganglia being lateral, and
united by a long commissure.

In the Gymnosomata the ganglia are more scattered, but
the arrangement of their nervous system needs reexamina-
tion.

All the Poeropoda are provided with an ovotestis. This
is a racemose gland, in the ultimate casca of which both ova
and spermatozoa are developed. The spermatozoa make
their appsarance at the closed end of the caecum and accumu-
late in its cavity ; the ova are developed from the epithelial
tissue of the caecum, somewhat lower down; nevertheless
fecundation does not take place in the ovotestis, probably in
consequence of the ova and spermatozoa attaining maturity
at different times. The ovotestis has a single excretory duct,
the tarmination of which ma}' be provided with a receptaculum
seminis and connected with a penis.

The young of the Pteropoda leave the e^g provided with
a velum, with a rudimentary shell, and probably with an
operculum. In most of the Thecosomata the shell is re-
tained and forms the commencement of that of the adult,
while the vela disappear and the epipodia are developed.
In Cymhulia, the primary external shell is shed and the
chitinous internal shell is a secondary development. In the
Gymnosomcita, the primary shell is also cast off, but is not
replaced, and three girdles of cilia are developed on the sur-
face of the body.*

The Silurian genera TentacnUtes, Theca, Pterotheca,
Conularia^ Ecculiomphalus^ are referred to the Pteropoda^
but they differ much from all existing forms. Unquestionable
PteropodoL are not know earlier than the tertiary formations.

The Beanchiogasteropoda. — In all the members of this

' Gegenbaur, I. c. ; Krohn, " Beitrage zur Entwickelungscrescliichte der
Pteropoden und Heteropoden," 1860; and Fol, "Etudes" ("Ai*chives d©
Zool. Experimentale," 18T5 and 1876).

THE BRANCHIOGASTEROPODA. 437

group, the development of which has hitherto been studied,
the intestine becomes twisted round on to the anterior face
of the body, in such a manner that the alimentary canal has
a completely haemal flexure, even in the veligerous embryo.
Hence, in the adult, the intestine springs from the hremal or
dorsal, and not from the ventral or neural, aspect of the
stomach ; and the pallial cavity, when it exists, is placed
upon the anterior htemal face of the body.

In the embryo, the shell always makes its appearance as
a conical, symmetrical, median cap. This embryonic shell
usually persists at the apex of that of the adult, the form of
which is modeled upon that of the visceral sac, and hence,
like the latter, is usually spiral. The embryo is also very
generally, if not universally, provided with an operculum.

The shell and operculum of the embryo disappear in the
naked Branchiogasteropods ; but the primitive external shell
is sometimes replaced by an internal shell lodged in a cavity
of the mantle (e. g., Aplysia). Usually, the Branchiogastero-
pods possess a distinct head provided with a pair of tentacles
and with tw^o eyes, which may either be sessile or mounted
upon peduncles of their own.

The mouth may be armed with chitinous jaw-plates, in ad-
dition to the radula. The heart is generally composed of a
ventricle and a single auricle, but sometimes there are two
auricles.

The l^ranchiogasteropoda fall into two distinct series, of
which the one is hermaphrodite (the genital gland being an
ovotestis) and invariably opisthobranchiate ; while the other
is unisexual and usually prosobranchiate. In each series
there are some forms which are provided with a large mantle,
and others in which the mantle is altogether abortive (Nudi-
hraiichlata^ Firola). These chlamydate and achlamydate
Branchiogasteropods correspond with the Thecosomata and
Gymnosomata among the Pteropods.

The chlamydate Branchiogasteropods are usually provided
with branchiae, which either take the form of numerous la-
mellae, or of two plume-like organs, sometimes reduced to one
functional gill and a rudiment of the second. In the achlamy-
date forms true gills are usually absent, though they may be
replaced functionally by processes of the hasmal bod^^-wall.

Among the Opisthohranchiata, Phyllldia is nearly sym-
metrical, the anus being situated at the posterior end of the
body, and there is a large mantle, devoid of a shell. There

438 THE ANATOMY OF INVERTEBRATED ANIMALS.

is no pallial cavity, and the branchioe are numerous lamellae,
placed on each side of the body, between the free edge of the
mantle and the foot. In Aiylysia, the mantle is relatively
small, and possesses an internal shell ; the branchiae, the
anus, and the reproductive apertures, are placed on the right
side of the bod}'. In this genus, and in G aster opter on, there
are very large epipodial lobes, by the aid of which some
species propel themselves like Pteropods.

The Nudibranchiata have no mantle, and the anus is
usually situated on the right side of the body ; sometimes,
however, as in Doris, it is terminal. In the pelagic Fhylli-
rhoe, the foot aborts, as well as the mantle, and the body has
the form of an elongated sac.

The gastric portion of the alimentary canal becomes com-
plicated by division into several portions, some of which are
provided with chitinous or calcareous plates, or teeth, in
Aplysia, l^ulla, and other genera. In many Nudibranchs,
as Kolis, the liver is represented by a much-branched tubular
organ, the cascal ultimate ramifications of which end in the
elongated dorsal papilla?. The apices of these papillae contain
thread-cells.

In the series of the Frosohranchiata , the great majority
are not only chlamydate, but there is a spacious branchial
chamber, and the pallial wall of the body is produced into a
conical visceral sac, which contains the stomach, liver, and
genital organs. It is usually asymmetrically coiled, and is
protected by the shell. No Opisthobranch possesses a large
visceral sac of this kind. On the other hand, no Prosobranch
is, like Phyllidia, symmetrical, with the anus at the posterior
end of the body. Patella and Flssurella are nearly sym-
metrical, but the anus is anterior.

The Prosohranchiata have, at most, rudiments oi epipodia,
but the rest of the foot often acquires a much greater develop-
ment than in the Opisthobra7ichiata, and a chitinous or shelly
plate — the operculum — is frequently developed from the dor-
sal or haemal aspect of the metapodium. The differentiation
of the foot attains its highest degree in the so-called Hetero-
poda, in which the propodium, mesopodium, and metapodium
differ widely in form ; the propodium being broad and fin-like,
and constituting the chief organ of locomotion in these free-
swimming oceanic animals.

In the Limpets (Patellidce), the visceral sac forms merely

THE HETEROPODA. 439

a conical projection of the h^mal surface, and the numerous
lamellar, or filamentous, respiratory organs, are lodged be-
tween the free edges of the mantle and the sides of the body.
In the other chlamydate Frosohranohiata, except the Cyclos-
tomata, there are two plumose gills lodged in a pallial chamber
situated on the anterior face of the visceral mass, which is
usually large and spirally coiled. Sometimes, as in the di
vision of the Asjyidobranchiay the two branchiae are equal, or
nearly equal, in size. Sometimes one is so much smaller than
the other as to be nearly abortive ( Cteiiobrajichia). Ampul-
laria has a pulmonary cavity as well as gills. On the other
hand, the Gyclostomata have no branchiae, but breathe air
by means of the parietes of the pallial chamber, whence they
are ordinarily reckoned among the Fahno7iata, which they
resemble in their terrestrial habits. In many Prosohranchiata,
the wall of the branchial chamber is produced into a muscular
spout-like prolongation, termed the sipho7i, which serves to
direct the branchial current. The presence of this siphon is
usually accompanied by a notch or grooved process of the
shell, and by carnivorous habits.

In the Hiteropoda^ there is a gradual reduction of the
mantle, from Atlanta^ in which the mantle and shell have the
ordinary proportions, and the departure from the ordinary
Gasteropod type is but little greater than that observed in
Stromhus and Pteroceras, through Carinaria, in which the
mantle is much reduced, and the shell is a mere conical cap,
to Flrola, in which the mantle and shell are wanting in the
adult, and which, therefore, corresponds with the achlamydate
Fteropoda and Opisthohrancliiata.

In many genera of the Ctenohranchia, and especially
among the carnivorous forms, the mouth is situated at the
end of a long proboscis, which contains the odontophore, and
a great part of the long oesophagus. This proboscis is pro-
truded and retracted by special muscles.^

The eggs are often laid in capsules secreted by the walls
of the oviduct. In Neritina^ Purpura^ and Fuccinuyn,, each
capsule contains a considerable number of ova, but of these
only a few (one in Neritina) become embryos, and devour
the rest.^

> See the description of the proboscis of the Whelk in Cuvier's " Memoires
sur les MoUusques."

2 Koren and Daniellssen, "Recherches sur le developpement des Pectini-
branches" ("Fauna littoralis Norvegiae," ii., 1856), and Carpenter, "On the
Development of the Embryo of Purpura lapillus'''' ("Trans. Micr. Society,"
1854, and " Annals of Nat. Hist.," 1857). Claparede, " Anatomie und Entwicke-
lungsgeschichte der Ntritina fluviatilis.^'' (" Archiv fiir Anatomie," 1857.)

440 THE ANATOMY OF IXVERTEBRATED ANIMALS.

The parasitic habit which is so rare among the Mollusca
occurs in the genus Stylifer, which infests Star-fishes and
Sea-urchins, sometimes imbedding itself in the perisoma ;
and, under a very remarkable and not yet thoroughly-under-
stood form, in the singular parasite of another Echinoderm,
Synapta digitata^ termed by its discoverer, Muller, Ento-
concha mirabiUs.^

In some few of the Synaptce (not more than one, or per-
haps two, in a hundred), elongated tubular molluskigerous
sacs are found attached by one extremity to one of the intes-
tinal vessels ; while the opposite end either hangs freely into
the perivisceral cavity, or may be entangled among the bases
of the tentacles, at the cephalic extremity of the body of the
Synapta. The sac is closed, but, at its attached end, a long
invagination extends into its interior. The cavity of the sac
beyond the closed extremity of the invagination contains an
ovary ; and, beyond this, a certain number of free seminal
capsules. The ova are detached from the ovary, and under-
go their development in envelopes, each containing many
ova, which gradually fill the cavity of the molluskigerous sac.
From these ova, embryos, provided with a velum, shell, and
operculum, proceed. A large pallial cavity is soon apparent ;
but, in the most advanced stages of development observed, it
contained no branchiae.

What becomes of these larva3 is unknown, nor is it even
certain to what group of the Odontophora JEntoconcha be-
longs.

The Pulmoxata. — These are oclontophorous Mollusks
which breathe air directly, by means of a respiratory surface
furnished by the wall of the pallial cavity.

In some, such as the Peroniadoe (Fig. 123) and Yeroni-
cellidoe, the body of the slug-like animal is very nearly sym-
metrical ; the anus and the lung-sac being situated close to-
gether at the posterior extremity of the body. The mantle
is large, and extends over the whole haemal or dorsal surface.
In all the other Puhnonata, the pulmonary and the anal
apertures lie on the right side of the body, and the mantle is
provided with at least the rudiments of a shell. The pallial
region is sometimes very small in proportion to the rest of
the body, and then forms a flattened disk, as in the common
Slug ; while, in some Lhnacidce and TestacellidcBy and in the

* " Die ErzcuErnng von Schneclcen in Holotliuricn," 1852. Baur, " Ueber
Synapta digitata.'''' (" Nova Acta," xxxi., 1864.)

THE PULMOXATA. 441

J'anellidoe^ the mantle is so much reduced that they are al-
most acldamydate. In the Snails, the mantle is large and is
produced into an asymmetrically coiled visceral sac, in which
the stomach, liver, and genital gland lie. The mantle-cavity
lies on the fore-part of the sac, and the anus opens on its
margin. Thus, in all the ordinary Pulmonata^ the termina-
tion of the intestine is twisted from its normal position at the
hinder end, forward to the right dorsal, or haemal, aspect of
the body.

When the pulmonary sac is posterior, and the pallial re-
gion small, the ventricle of the heart is anterior, and the
auricle posterior, and the animal may be said to be opistho-
puhnonate. On the other hand, when the pallial region is
large, and gives rise to a visceral sac, with the concomitant
forward position of the pulmonary chamber, the auricle is
inclined more or less forward and to the right side, and the
apex of the ventricle backward and to the left side. Tiie
animal is thus more or less prosopuhnonate.

The mouth is commonly provided with a horny upper jaw,
as well as with a well-developed odontophore. Large salivary
glands are usually present.

The heart consists of a single auricle and a single ventri-
cle. The aortic trunk, which proceeds from the apex of the
latter, divides into many branches, but the venous channels
are altogether lacunar. A renal organ lies close to the pul-
monary sac in the course of the current of the returning blood.

There are usually two simple eyes, often lodged in the
summits of retractile tentacula.

The Pulmonata are hermaphrodite. The generative gland
is an ovotestis, and is composed of branched tubuli, from the
cellular contents of which both ova and spermatozoa are de-
veloped (Fig. 123, III.).

A narrow common duct leads from the ovotestis, and, soon
dilating, receives the viscid secretion of a large albumen-
gland. The much wider portion of the common duct beyond
the attachment of this gland is incompletely divided by longi-
tudinal infoldings into a sacculated, wider, and a straight,
narrower, division. The former conveys the ova, and the
latter the spermatozoa. At the end of this part of the ap-
paratus, the wider portion, which represents the oviduct,
passes into the vagina, which opens at the female genital
aperture, while the narrower portion of the common duct is
continued into a separate, narrow, vas deferens, the end of
which opens into a long invagination of the integument — the

U2

THE ANATOMY OF INVERTEBRATED ANIMALS.

penis. In Peronia, the vas deferens and the oviduct open
together by the genital aperture, and, as in some Branchio-
gasteropods, a groove, along which the seminal fluid is con-

FiG. 123. — Dinsrram exhibiting the disposition of the intestine, nervous system, etc.,
in a common Snail (Hdix). — a, mouth; ft, tooth ; c, odontophore ; d, gullet j e, its
dilatation into a sort of crop;/, stomach; g, coiled termination of the visceral
mass ; the latter is also close to the commencement of the intestine, which will be
Been to lie on the neural side of the oesophagus ; A, rectum ; t, anus ; A*, renal sac ;
;, heart; m, lung, or modified pallial chamber; n, its external aperture; o, thick
edge of the mantle united with the sides of the body; J9. foot; r, s, cerebral, pedal,
and parie to-splanchnic ganglia aggregated round the gullet.

duoted, leads to the outer opening of the eversible penis (Fig.
123, I., II.).

In connection with the female genital aperture, there is
always a spermatheca., or sac (which is sessile in the Slugs,
but in the Snails is placed at the extremity of a long duct),
for the reception of the semen of the other individual when
copulatioti takes place.

The Ilelicidm alone possess, in addition, the so-called sac
of the dart^ a short muscular bag, in which pointed ciiitinous
or calcified bodies — the splcula amorls — are formed ; and
certain glandular coeca, generally arranged in two digitate
bundles, termed mucous glands, which give rise to a milky
secretion. Sometimes prostatic glands are developed on the

THE PULMONATA.

443

vas deferens, which may be dilated in part of its course into
a vesicula seminalis.

Fig. 123.— I. Peronia Temiculata.—a, anns : pi, palmonary aperture; {7, genital aper-
ture ; fs, seminal groove : p, openins: for tlie penis.

II. Generative organs of the same animal, the ovotestis beina: omitted.— g'a?, gland
which furnishes a glairy secretion ; od, oviduct ; vd, vas deferens ; t, intestine ; a,
anus; rs, receptaculum seminis ; /), aperture of the penis; //, penis; cs, seminal
duct ; ap, glandular appendage ; m, retractor muscle of the penis. (After Kefer-
stein.)

ni. Blind end of a follicle of the OYoiQ^t^^ nf Helix pomatia. At the apex the sperma-
tozoa are seen in different stasres of development, ihe fully-formed spermatozoa
floatint; in bundles in the cavitv of the follicle. Lower down, ova are developing
in the walls of the follicle. (After Keferstein and Ehlers.)

The ova are impregnated hi^h up in the oviduct, and are
invested by a relatively very large mass of albumen and in-
closed within a thick, sometimes calcified, chorion. The mass
inclosed by the latter may be a tenth of an inch or more in
diameter, while the proper ovum may have not more than a
twelfth of that size.

There is no trustworthy evidence of the existence of the
opisthobranchiate Gasteropods" before the epoch of the Trias,
but it is to be remembered that the great majority of these
animals have no shells. Of the rest of the preceding groups
of Odontophora, representatives are known as far back as
the middle of the Palaeozoic epoch, while Pteropoda^ Setero-

444 THE ANATOMY OF INYERTEBRATED ANIMALS.

poda, and Prosohranchiata, occur in the Silurian formations.
Among the Prosohranchiata, the Patellidm and the Aspido-
hranchia are the characteristic forms of the older formations,
the Ctenobranchia appearing later, and acquiring their pres-
ent relative abundance only in the later secondary and the
tertiary epochs.

The Cephalopoda. — The bilateral symmetry which is so
obvious in the Polyplacophora and the Scaphopoda is but

Fig. 124.— a. Sepia officinalis. B, lateral view of the horny ring- of an acetabulum.

little disturbed in this group of the Odontophora. The
mouth and the anus are situated in the median plane, which
divides the body into corresponding halves ; while the bran-
chiae, two or four in number, are disposed symmetrically on
each side of this plane, as are' the brachial prolongations of
the margins of the foot. The hfemal face of the body, how-
ever, is not flat, as in the mollusks which have just been men-
tioned, but is elongated perpendicularly to the neural face, so
as to form a sort of sac, invested by the mantle. On the pos-

THE CEPHALOPODA. 445

terior, or anal, face of the sac, the mantle incloses a large
pallial cavity, in which the branchiae are protected. On the
anterior aspect of the sac, on the contrary, the mantle may
have no free edge, or, at most, forms a comparatively small
flap.'

The integument is provided with chromatophores, which
are sacs with elastic walls, full of pigment, and provided with
radiating muscles, by which they may be drawn out to a size
many times greater than that which they possess in their
contracted state. In their dilated condition, the color proper
to the contained pigment becomes plainly visible, while in
their contracted state they appear as mere dark specks. It
is to the successive expansion and contraction of these chro-
matophores that the Cephalopoda owe the peculiar play
of "shot" colors, which pass like blushes over their sur-
face in the living state. These blushes of color are especial-
ly well displayed by young Cephalopods just freed from the

But that which particularly distinguishes the Cephalo-
pod is the form and disposition of the foot. The margins
of this organ are, in fact, produced into eight or more pro-
cesses, termed arms, or hrachla / and its antero-lateral por-
tions have grown over and united in front of the mouth,
which thus comes, apparently, to be placed in the centre of
the pedal disk. Moreover, two muscular lobes which cor-
respond with the epipodia of the Pteropods and Branchio-
gasteropods, developed from the sides of the foot, unite pos-
teriorly, and, folding over, give rise to a more or less com-
pletely tubular organ, the funnel, or infundihuhon. The
open end of the funnel projects between the posterior face
of the body and the pallial wall of the branchial cavity, and
serves to conduct the water, when it is driven out of the
latter by the contraction of the mantle in ordinary expira-
tion ; and when the animal swims, the stream forcibly driven
out in this way causes it to dart swiftly backward.

The aperture of the mouth (Fig. 125, a) is provided with
a hard, chitinous beak, like that of a parrot, the two divis-
ions of which are anterior and posterior. Of these, the
anterior is always the shorter, and is overlapped by the
other.

» Cephalopods are usually described as if the oral end of the body were the
upper end, and the face on which the pallial chamber is placed Ventral — a
method which seriously interferes with the comprehension of their relations
with other Mollusks.

446

THE ANATOMY OF INVERTEBRATED ANIMALS.

Within the cavity of the mouth is an odontophore, with
its radula (Fig. 126, II.) ; and the long gullet passes back on
the middle line to open into the stomach, which is situated

Fig. 125.— Diagrammatic Pection of a female Seina.—a, Buccal mass snrronnded 'by
the lips, and shDwing the homy jaws and tongue; ft, oesophagus; c, salivary
gland" d. stomach; e, pyloric csecnm: g, the intestine; h, the anus; z, the ink-
nag; k, the place of the systemic heart; I, the liver; n, the hepatic duct of the
left Bide; o, the ovary; p, the oviduct; g, one of the apertures by which the wat^r-
chamhers are placed in communication with the exterior ; r, one of the branchise;
s, the principal ganglia agirregated round the oesophagus;/, the funnel; vi. the
mantle; sh, the internal shell, or cuttle-bone ; 1, 2, 3, 4, 5, the produced and modi-
fied margins of the foot, constituting the so-called arms of the Sepia.

toward the middle, or the end, of the mantle-sac. From the
stomach, the intestine, more or less bent upon itself, passes
toward the neural aspect of the body, and ends in the median

THE CEPHALOPODA. 447

anus. Hence the alimentary canal has a v^el) marked neural
flexure (Fig. 125).

Except in JSfautilus^ one or two pairs of salivary glands
are present (Fig. 126, I. «'). The liver (Fig. 126, I. A) is al-
ways large ; and there are two hepatic ducts (Fig. 126, 1. fM),
beset for a greater or less extent with glandular follicles, gen-
erally considered to be pancreatic in function. Very often a
large, sometimes spirally wound, caecum is developed from the
commencement of the intestine ; into this the hepatic ducts
open.

The heart (Fig. 127, c) is placed upon the posterior face
of the body on the hasmal side of the intestine, and receives
the blood by branchio-cardiac vessels, which correspond in
number with the gills, and, as they are contractile, nii^ht be
regarded as auricles. The gills themselves have no cilia, and
are, in some cases, if not always, contractile. The arteries
end in an extensively-developed capillary system, but the
venous channels retain to a greater or less extent the char-
acter of sinuses.* The venous blood, on its way back to the
heart, is gathered into a large, longitudinal sinus — the veria
cava — which lies on the posterior face of the body, clo^e to
the anterior wall of the branchial chamber, and divides into as
many afferent branchial vessels as there are gills. Each of
these vessels traverses a chamber which communicates di-
rectly with the mantle-cavity, and the wall of the vessel which
comes into contact with the water in this chamber is saccu-
lated and glandular ^ (Fig. 127, re). Each chamber, in fact,
represents a renal organ. The pericardium, and the sacs in
which the testes and ovaria are lodged, may communicate

1 Milne-Edwards, " Eecherclies Anatomiqnes et Zoologiques. Premiere Par-
tie." " Observations et Experiences sur la Circulation chez les Mollusques,"
1845.

2 On account of the transparency of the tissues in the living Loligo media,
this species affords an easy opportunity of observing the rhythmical contrac-
tions of the branchife, and their afferent and efferent vessels. For this pui^
pose the mantle should be laid open, and the nidimental glands carefully
removed. The sacculated afferent veins and the branchial hearts contract
about sixtv times a minute. The pulsations of these veins, and of the bran-
chial hearts, are not synchronous. The branchial veins, and the lamellae of
the branchife, also contract rhythmically, but I could observe no contraction in
the branchial arteries. The portion of the branchial vein which lies between
the base of the gill and the systemic ventricle is very short, and it is hard, to
sav whether it contracts independently or not. Mechanical irritation causes
contraction both of the afferent branchial veins and of the branchial hearts.

In the living Eledone cirrliosus I have o-bserved regular rhythmical con-
tractions of the vena cava itself as well as of its divisions, the sacculated affe-
rent branchial veins, of the branchial hearts, and of the branchio-cardiac ves-
sels.

448 THE ANATOMY OF INVERTEBUATED ANIMALS.

with the pallial cavity either directly or through these cham-
bers. Thus, in Sepia officinalis, Krohn ^ observed that the

m

Fig. l^i^.— Sepia officinalis.— I. The alimentary canal, with the ink-bag: mb, buccal
mass; gb, inferior buccal ganglion ; s', pot*terior salivary glands ; oe, CBSophagus;
h, liver; dh, hepatic duct ; «?, stomach ; v\ pyloric cjecum ; i, intestine ; a, anus;
bi, ink-bag ; gsp, splanchnic ganirlion on the stomach. (After Keferstein.)

n. Longitudinal and vertical section through the buccal mass : mxi, posterior beak ;
mxs, anterior beak ; 7/ibc, buccal membrane ; ml, lip ; a:, gustatory (?) organ ; rd,
radula ; z, sac of the radula ; s', salivary gland ; gl, superior buccal" ganglia. (After
Keferstein.)

III. A single transverse row of teeth from the radula. (After Troschel.)

renal chambers communicate not only with the cavities in
which the branchial hearts are lodged, but with a chamber
which contains the stomach and the spiral pyloric appendages ;
and that all these cavities are distended when air is blown
into one renal chamber. In JEledone, on the contrar37-, he
found, and I have repeated the observation, that one renal

* " Ueber das wasserfiihrende System einiger Cephalopoden." (" ArcMv
fiir Anatomie," 1839.)

THE CEPHALOPODA.

449

chamber can be fully distended without the air passing into
the other.

Fig. 1^^.— Sepia officinalis. — <?. eystemichoart ; rto,antorinr aorta ; oo\ posterior aorta ;
1, vena cava; 2, afferent branchial ves.^els; re. mual ort;ai:8; z, appendages of
thef»e vessels; 3, 4, larsje posterior veins bringintr blood to the afferent branchial
vessel? ; 5, 6, 7, efferent branchial vessels, branchial veins, -and brauchio-cardiac
or auricular trunks. (After Hunter.)

In ^aiitilics 2^ompilius there are, as Valenciennes discov-
ered, three pairs of openings which lead from the branchial
sac into chambers contained in the interior of the body. Of
these chambers there are five : the anterior and posterior
pairs are situated on each side of the rectum, and each has
its own opening ; the fifth, a very much larger chamber, has
two openings, one on each side. It is coextensive with that
part of the mantle which lies behind the insertion of the shell-
muscles and the hornv band which connects them. It is
separated from the paired chambers by their inner walls, and
these walls are traversed by the afferent branchial veins.
Appendages of these veins project on the one hand into the
paired chambers, and on the other into the single chamber.
The latter appendages are elongated papillae, while the for-
mer are lamellar. Earthy concretions, composed mainl}^ of
phosphate of lime, but which yield no trace of uric acid, are
usually found in the paired sacs.^

1 Owen, "Memoir on tlie Pearly Nautilus." Van der Hoeven, "Beitra^
zur Anatomie vom I^aiitilus pompiluis^'' (" Archiv fiir Naturgeschichte,"
1857). Huxlev, " On some Points in the Anatomy of Nautilus pompilius "
(" Proceedings of the Linnnpan Society," 1858). See also Keferstem, Bronn's
"Klassenu. Ordnungen," Bd. iii. (1862-'66), pp. 1390, 1319.

450 THE ANATOMY OF INVERTEBRATED ANIMALS.

The nervous system in the Cephalopoda^ as in other Mol-
lusca, consists of cerebral, pedal, and parieto-splanchnic gan-
glia, aggregated around the gullet, and connected by com-
missural cords. In addition to these, buccal, visceral, bran-
chial, and pallial ganglia may be developed on the nerves
which supply the buccal mass, the alimentary canal, heart,
branchia, and mantle.

In the Dibranchiata (Fig. 128), the three principal pairs
of ganglia are usually large, and so closely aggregated to-
gether that the commissures are not readily distinguishable.
The optic nerves are very large ; one or two nerves are given
off to the superior or anterior buccal ganglia, which have co-
alesced into one mass, and are united by commissures, which
encircle the oesophagus, with the coalesced inferior or pos-
terior buccal ganglia. The pedal ganglia lie on the pos-
terior side of the gullet, and supply the large nerves to the
arms, and those to the funnel, while the auditory nerves are
immediately connected with them. Each parieto-splanchnic
ganglion gives off a nerve which runs along the shell-muscles
to the anterior wall of the mantle, and there enters a large
ganglion, the ganglion stellatum, A large median branch, or
branches, from the parieto-splanchnic ganglia, accompanies
the vena cava, and is distributed to the branchiae and sexual
organs. The inferior buccal ganglion sends a recurrent nerve
along the oesophagus, which ends in a ganglion on the stom-
ach.^

The nervous svstem of KauUlus differs in some important
particulars from that of the Dibranchiata. The cerebral
ganglia are represented by a thick transverse cord, which lies
in front of the oesophagus, and from the outer angles of which
the optic and olfactory nerves are given off, while nerves to
the buccal mass proceed from its anterior edge. The pedal
ganglia lie close to the cerebral ganglia, and are united by a
slender commissure, which passes behind the gullet. They
supply all the brachial processes and the funnel with nerves,
and the short auditory nerves are connected with them. The
parieto-splanchnic ganglia are, like the cerebral ganglia, elon-
gated, and together constitute a thick cord, which, united at
each end with the cerebral ganglia, forms a hoop round the gul-
let, distinct from the pedal nerve-arch, and separated from it
by a process of the cartilaginous skeleton. The largest nerves

> See Hancock, " Anatomy of the Nervous System of Ommastrep'het.''^
(" Ann. Nat. History," 1852.)

THE CEPHALOPODA. 451

given off from these ganglia are those which go to the bran-
chiae.

Eyes, olfactory organs, and auditory sacs, are always
present. The eyes of the Cephalopoda may be lodged in
orbital cavities at the sides of the head, as in all the Dibran-
chiata ; or may be pedunculated, as in Nautilus. In the
former case, the eye is inclosed partly by the cephalic car-

ao

Fig. Vi,%— Sepia officinalis.— The nervous mass which surround? the gnllet; JV, the
cerebral; iV', the pedal ; N'', the parieto-splatichnic gandion ; ao. the aorta; oe,
the oesophagus; (/, buccal nerves ; i", nerves to the anus; Jf, pallial nerves ; j7,
superior; g', inferior buccal ganglion. (After Garner.) *

tilage, to which sometimes special orbital cartilages are add-
ed, and partly by a fibrous capsule continuous with these.
The fibrous capsule becomes transparent over the eye, and
gives rise to what is variously interpreted as the representa-
tive of the cornea, or as that of the eyelids of vertebrated ani-
mals. This transparent coat is sometimes entire, or presents
only a small perforation ( Octopus, Sepia, Jjoligo, and the
other 3Iyopsid(B of D'Orbigny) ; sometimes it has a wide
opening, through which the crystalline lens may project {Lo-
ligophes, Ommastrepsis, and the other Oigopsidm of D'Or-
bigny) ; and sometimes it is altogether absent, and the capsule
of the eye becomes an open cup {Nautilus).

In the J) ihranchiata,^ a great part of the chamber of the
capsule of the eye is occupied by the ganglion, into which the
optic nerve enlarges after entering it ; by muscles ; and by a
peculiar white glandular substance. Lining the capsule, but

1 "Trans. Linngean Society," 1836.

2 See Hensen, " Ueber das Auge einiger Ccplialopoden." {Zeitschrift fur
wissenschaftliche Zoologie^ 1865.)

452 THE AXATOMY OF INVERTEBRATED ANIMALS.

not adhering to its inner surface, in front, is the silvery tapC'
tuni^ formed of two layers. These pass into one another at
the edges of the free jDrolongation of the tapetum, which forms
the iris. Longitudinal muscular fibres are interposed between
the two layers of the tapetum. Under the tapetum is a layer
of cartilage, which forms the inner capsule of the eye, extends
as far as the iris externally, and is perforated by the fibres of
the optic nerve on its inner side. The free edge of the inner
capsule gives attachment to a thick rim of connective tissue,
containing muscular fibres. This so-called ciliary body enters
the deep groove which surrounds the lens ; the latter is, in
fact, made up of layers of structureless membrane, which are
cuticular productions of the ciliary body. In shape, the lens
is elongated in the direction of the axis of the eye, so as to
be almost a cylinder with convex ends, and thus, with its deep
equatorial groove, into which the ciliary body fits, it has a
wonderful resemblance to a Coddington lens. The vitreous
humor is a transparent fluid. The retina lines the inner cap-
sule, and may be divided into an outer and an inner stratum,
separated by a pigment layer. The inner stratum is formed
of prismatic or cylindrical rods, the outer ends of which abut
upon the pigment, while their inner ends, turned toward the
cavity of the eye, are covered by a thick hyaloid membrane.
The outer stratum contains the plexus of the fibres of the
optic nerves, and numerous cells (ganglionic), supported by
connective tissue. The terminations of the nerves, therefore,
must traverse the pigment layer to reach the rods.

It will be observed that the apparent resemblances between
the cephalopodous and the vertebrate eye are merely super-
ficial, and disappear on detailed comparison.

In Nautilus, the eye has neither cornea, lens, nor vitreous
humor, but is a mere cup, lined by the retina. The aperture
for the admission of light is exceedingly small.

The olfactory organs, the true nature of which was dis-
covered by Kolliker,^ are sometimes pits, sometimes papillas
of the integument, situated behind or above the eyes. In the
Teuthidce and SejyiadcB, they are depressions above the eyes;
in the Octopoda, they are either depressions or papillae (Ar-
gonaxda and Tremoctopus) in the same position, but nearer
the anterior face of the body. In Nautilus, they are elon-
gated, tentaculiform, and situated immediately behind the
eyes.

» " Entwickelungsgeschichte der Cephalopoden," 1841, p. 107,

THE CEPHALOPODA.

453

In the Dibraowhiata, the auditory sacs are lodged in cavi-
ties of the cephalic cartilage, and contain a single large
otolith, composed of carbonate of lime, and of rounded or
irregular but definite and characteristic form. In JS^autilus,
Dr. Macdonald discovered that the auditory sacs are attached
to the pedal ganglia, and are not lodged in the cranial cartilage.
They contain numerous otoliths.

An endoskeleton formed of true cartilage is dev^eloped in
the region of the principal ganglia, and sometimes furnishes
them with a complete investment. It gives attachment to the
most important muscles. In some Cephalopods additional
cartilages appear in the mantle and in the funnel. The mus-
cular fibres of the Ce2yhalopoda are unstriated.

The sexes are distinct, and the reproductive organs are un-
like those of other Mollusks. They consist, in both sexes
(Fig. 129), of lamellar or branched organs, the cellular con-
tents of which are metamorphosed into ova or spermatozoa,

Fig

semi

I'HQ. —Sepia officinalis.— 1. maleorfrans: t. testis; vd. vas deferens; vs, vesicula
-jDiinalis; pr, prostate; hsp.. receptacle of the spermatophores ; p, peuis with the
penital aperture. (After Duvernoy.)
II. Female genital organs: a. anus ; i, intestine; ov, ovary ; oc?'. oviducal aperture;
od, oviducal gland ; gn, nidamental gland ; gn', accessory glands. (After Milue-
Edwards.)

and which are attached to one point or line of the wall of a
chamber, which communicates with the pallial cavity by two

454 THE ANATOMY OF INVERTEBRATED ANIMALS.

symmetrically - disposed oviducts, in the females of some
species ; but in most female, and almost all male, Cephalo-
pods ^ it has only one duct, the termination of which is usually
situated on the left side, but may be near the middle line
(male JSfautilus), or even on the right side (female JSfaiitilus).
In the female, the oviduct, or oviducts, present glandular en-
largements. In addition, two lamellar nidaniental glands are
developed upon the walls of the branchial cavity, and to these
accessory glands may be added. These glands secrete a vis-
cid fluid, which invests the ova, and connects them, when laid,
into variously-shaped aggregations. In the male, a prostatic
gland furnishes the material of the cases, or spermatophores^
in which packets of spermatozoa are contained, and which
sometimes possess a very complicated structure.

In the Dihranchiata^ the spermatophores are slender
cylindrical bodies which may reach haK an inch in length.
They have an external structureless case, thinner at one end
than the other, and often ending" in a fine filament at the thin
end. Within this case, filling its thicker end, and as much as
half or two-thirds of the rest of its cavity, is a delicate sac
full of spermatozoa.

The rest of the case is occupied hy a very singular elastic
body, in form somewhat resembling the sponge of a gun with
a spiral screw turned on the handle. The enlarged "sponge"
end of this body is fastened by a delicate prolongation to the
spermatic sac, while the " handle," being too long to lie
straight, is coiled up at the end opposite to the sponge, and
then fastened to the outer case. When these bodies come
into contact with water they undergo strange contortions,
and finally, the thin end of the case giving way, the spring
frees itself, starts out of the case, and drags with it the sper-
matic sac."

In Nautilus^ according to Van der Hoeven, the spermato-
phores have a much simpler structure.

The male Cephalopods are distinguished from the females
by the asymmetry of their arms, one or more of which, on
one side, are peculiarly modified, or hectocotylized.

Some Cephalopods are devoid of any shell, but most pos-
sess a pallial shell, which is either external or internal. In
the former case, the visceral sac is lodged within that part of

^ Keferstein found two ducts in a male Eledone moscTiata.

2 For the minute structure of these curious spermatic cartridires, see Milne-
Ed wards' s elaborate essay, "Observations sur les Spermatophores des Mol-
lusques Cephalopodes." (" Annales des Sciences Naturelles," 1840.)

THE CEPHALOPODA. 455

the cavity of the shell which lies nearest its open end, and
the rest of the cavity is divided into chambers, which contain
air, by transverse sejyta. The septa are perforated, and a pro-
longation of the mantle — the siphuncle — is continued through
the series of perforations, as far as the apical chamber of the
shell. The internal shells of the Cephalopods may have
very various forms, and may even be chambered and siphun-
culated ; but, in this case, the chamber nearest the mouth of
the shell is small, and incapable of lodging the viscera.

Our knowledge of the development of the Cephalopods is
confined to that of the Dibranchiata.^ In these, the yelk
undergoes partial division, and the blastoderm, formed upon
one face of it by the smaller blastomeres, spreads gradually
over the whole ovum, inclosing the larger and more slowly-
dividing blastomeres. The mantle makes its appearance as
an elevated patch in the centre of the blastoderm, while the
future arms appear as symmetrically-disposed elevations of
the periphery, on each side of the mantle. Between these
and the edge of the mantle, two longitudinal ridges mark the
rudiments of the epipodia, while the mouth appears in the
middle line, in front of the mantle, and the anus, with the
rudiments of the gills, behind it. The rest of the blastoderm
forms the walls of a vitelline sac, inclosing the larger blasto-
meres.

The pallial surface now gradually becomes more and more
convex, the posterior margin of the mantle growing into a
free fold, which incloses the pallial chamber and covers over
the gills.

The internal shell is developed in a sac formed by an in-
volution of the ectoderm of the mantle. The epipodia unite
behind, and give rise to the funnel, while the anterolateral
portions of the foot grow over the mouth, and thus gradually
force the latter to take up a position in the centre of the neu-
ral face, instead of in front of it. The yelk-sac gradually
diminishes, and the contained blastomeres are finally taken
into the interior of the visceral sac, into w^hich the alimentary
canal is gradually drawn.

The Cephalopoda are divided into two very distinct
groups, the Tetrabranchiata and the Dihranchiata.

The Tetrabranchiata possess an external chambered si-

' KolUker, "Entwickelungsgeschiclite der Cephalopoden," 1841. Gre-
nadier, " Zur Entwickelunsrs^eseliichte der Cephalopoden " {ZeitscJirift fi'ir
wiss. Zooloqie, 1876). Lankester, "Observations on the Cephalopoda"
{Quarterly Journal of Micr. Science^ 1875J.

456

THE ANATOMY OF INVERTEBRATED ANIMALS.

pliunculated shell. The terminal chamber is much larger than
any of the rest, and the body of the animal can be almost
completely retracted into it. When, as in the only existing
genus, JVautilus^ {Fig. 130), the shell is coiled into a flat,
symmetrical spiral, its apex lies on the anterior face of the
body, and the outermost chamber, into which the whole body
can be retracted, is consequently posterior to the axis of the
helix. In JVautilus, the brachial processes are short, and pos-
sess no acetabula such as exist in the Dibranchiata, but the
margins of the foot are produced externally into a sort of
sheath, which, in front, has the form of a broad hood with a
tuberculated surface ; while, at the sides, it is divided into
many processes of unequal lengths. Behind, the halves of
the sheath are separated throughout the greater part of their
length by a wide interval, but are united above by a thick

ntx

Fig

c\e\ ch, black part of the shell under the mantle p' ; kn, process of the cartilagi-
nous skeleton into the funnel. (After Keferstein.)

muscular isthmus. The central portion of the sheath is a
broad, triangular, hood-like plate, the apex of which is free.
It contains two long, narrow cavities, each of which lodges a
tentacle. The tentacle consists of a slender stem, on which

1 Owen, " Memoir on the Pearlv Nautilus," 1832. Van derHoeven, " An-
nales des Sciences Naturelles," 1856. Keferstein in Bronn's " Klassen u.
Ordnungen."

THE TETRABRANCHIATA. 457

are set a great number of transverse plates, in such a manner
that the axis of the stem passes through the centre of the
plates. The anterior and lateral regions of the hood are
completed by two narrower processes, each of which contains a
similar tentacle, and the lateral portions of the sheath are
formed by sixteen or seventeen smaller tentaculiferous pro-
cesses, the surfaces of which are more or less distinctly an-
nulated. When the sheath is opened out, there is seen to be
attached to its inner surface, on each side, close to the reen-
tering angle between it and the lip which surrounds the beak,
and along the line of junction of the lateral part of the sheath
with the isthmus, a thin, free, quadrate lobe, which carries
twelve tentacles. The isthmus joins the posterior edges of
these outer tentaculiferous lohes^ as well as those of the two
halves of the sheath, and it exhibits on its anterior, or inner,
surface a broad area beset with delicate, close-set, curved
laminse. Two other similar, but much thicker, inner tenta-
culiferous lobes, which also carry twelve tentacles, lie be-
tween these and the lip. They are quite free from the outer
tentaculiferous lobes, and unite with the sheath only above
and behind. Like the halves of the sheath, these two lobes
are united behind by a thick isthmus, the surface of which
presents a number of parallel longitudinal laminae. The
beak, which is hidden by the sheath and the lobes, is sur-
rounded by the thin circular lip already mentioned, the free
margin of vvhich is papillose. Besides these, there is a short,
conical tentaculiferous process above the pedunculate eye, and
another below it. In the male, the internal tentaculiferous
lobes are wanting, and the outer tentaculiferous lobes are
divided into two portions, an anterior which bears eight, and
a posterior with four, tentacula. On the left side, the four
tentacles of the posterior division have undergone much mod-
ification, and are converted into a peculiar organ termed the
spadix, which bears a discoidal follicular gland upon its outer
surface. There is thus a kind of hectocotvlization in the
Tetrahranchiata,

The margins of the united epipodia are not united into a
tubular funnel. They constitute a muscular membrane, nar-
row on the anterior face of the body, but becoming wide, and
folded in such a manner that its posterior edges overlap, be-
hind.

The mantle has a broad anterior fold, which covers the
anterior convexity of the shell, and the region which it thus
invests is black. The pallial chamber does not extend for
20

458 THE AXATOMY OF INVERTEBRATED ANIMALS.

more than three-fifths of the length of the body, and is there-
fore much less deep than in the Dihranchiata. The anus
opens in the middle line on the posterior wall of the pallial
cavity, close to its junction with the anterior wall. The four
branchiae are attached, two on each side of the anus, to the
posterior wall of the branchial chamber, and the inner branchia
is shorter than the outer. The nidamental glands, composed
of numerous vertical lamellse, partly covered by a fold of the
lining membrane of the pallial cavity, are situated on the
posterior wall of that cavity, almost midway between its
union with the anterior wall and its free edge. The paired
renal chambers lie immediately above them also, in the pos-
terior wall of the pallial cavity.

The buccal mass is very large, its length amounting to
one-third that of the body. The apices of the great horny
beaks are obtuse, and are coated with a calcareous deposit.
The oesophagus dilates into a wide crop and is separated by
a constriction from the stomach, the chitinous lining of which
is thick and ridged. The pyloric caecum is small and rounded,
and the intestine makes two bends upon itself before reaching
the anus. Salivary glands appear to be wanting, unless cer-
tain glandular bodies placed within the buccal mass should
be of this nature.

The liver is a loosely racemose gland, divided into four
lobes, and is lodged in the anterior part of the perivisceral
cavity. There is no ink-bag, and there are no branchial
hearts. The quadrate systemic heart is situated on the left
side of the posterior face of the body, close to the junction
of the posterior with the anterior wall of the pallial cavity.
It receives four branchio-cardiac veins ; and, attached to it, is
a pyriform sac, which, according to Keferstein, opens into the
pallial cavity.

The cartilaginous skeleton supports the pedal and parieto-
splanchnic ganglia, but does not encircle the gullet, or roof
over the cerebral ganglia. Two long processes of the skele-
ton pass into the funnel and give attachment to its muscles.
Two large shell-muscles are attached to it; and, passing up-
ward and outward, are inserted into oval chitinous patches
visible on the outer surface of the mantle, and connected to-
gether by a thin ring of the same substance (the atinidus)
which encircles the mantle.

The oviduct does not arise directly from the sac in which
the ovary is lodged, but from a distinct chamber, into w^hich
the ovarian sac opens. A large albumen-gland pours its

THE TETRABRAXCHIATA. 459

sscretion into the ovarian sac. The vas deferens similarly
takes its origin, not from the sac of the testis but from a
smaller chamber communicating therewith. The commence-
ment of the vas deferens is enlarged and glandular. Nothing
is known of the development of the Tetrabranchiata.

The only existing representatives of the Tetrabranchiata
are the different varieties of "pearly nautilus" {^Nautilus
pompilius), which are found in the southern seas, living at
the bottom at a considerable depth. The genus is one of the
oldest in existence, since it is traceable through the whole
series of fossiliferous rocks as far back as the Silurian
epoch.

Along with it, in the Palieozoic formations, occur numer-
ous closelj^-allied forms, which differ from JVautilus mainly in
the different curvature {Lituites, Gyroceras^ Trochoceras) or
straightness ( Orthoceras, Gomphoceras) of the shell, and in
the varying position, proportions, and degree of calcification
of the siphuncle.

In the middle of the Palaeozoic strata (Devonian), Tetra-
branchs {Ammonitidce) appear, in which the margins of the
septa are strongly bent, whence their edges appear as zigzag
transverse lines, folded into lobes and saddles, when the outer
layer of the shell is worn away (Goniatites, Ceratites); and,
in the Mesozoic epoch, the lobes and saddles become extreme-
ly complicated, w^hile the shells may be straight, simply
curved, or bent, or turbinated {Ammo7iites, Bacidites, Turri-
lites). The AmmonitidcB are extraordinarily numerous in
the Mesozoic epoch, but no trace of them has been found in
tertiary or quaternary formations.

Associated with Ammonites, and not unfrequently lodged
in the terminal chamber of the shell, are the so-called Ai^tychi.
These are plates of a shelly substance, three-sided, with
rounded-off angles, and applied together by their straightest
edges so as to resemble bivalve shells. They consist of two
layers, an inner and an outer, of which the inner presents
lines of growth, concentric with the angle of each plate which
is situated on that side of its broad end which is applied to
its fellow. The outer layer is composed of many laminie, and
is traversed by pores. Its free surface frequently presents
longitudinal ridges. The heart-shaped plates, undivided by
a suture, which are found in some Goniatites dind Ammo7iites,
are termed Anoptychi.

The Aptychi, when undisturbed, occupy the middle of the
posterior wall of the terminal chamber of the Ammonite, and

460 THE ANATOMY OP INVERTEBRATED ANIMALS.

have their bases toward its mouth. Nothing is certainly
known as to the nature of the Aptychi or Anaptychi^

In the Dibranchiata^ the margins of the foot are pro-
duced into not fewer than eight, nor more than ten, arms,
which are provided with acetabular or suckers. Each ace-
tabulum is a sessile or stalked cup, from the bottom of which
rises a plug, which nearly tills the cup, but can be retracted
by the action of muscular fibres attached to it. When the
margins of the acetabulum are applied to any surface, and
the plug is retracted, a partial vacuum is created, and the
acetabulum is caused to adhere to the surface by atmospheric
pressure. The edges of the acetabula are frequently strength-
ened by chitinous rings, and these may be serrated (Fig. 124,
B), and are sometimes produced into long, curved hooks.

The margins of the united epipodia are not only folded
inward, but coalesce so as to give rise to a tubular funnel,
through which the water taken into the branchial sac for
respiratory purposes is ejected. Very often, a valve which
prevents the flow of water back into tlie mantle cavity is de-
veloped within the funnel. There are two branchiae, and the
anus terminates between them in the anterior wall of the
branchial sac, on which also the nidamental glands are situ-
ated. The apices of the horny beaks are acutely pointed,
and not ensheathed in calcareous matter. The liver is usual-
ly a compact mass. A peculiar gland, which secretes an ex-
tremaly dark fluid — the so-called ink — and has the form of an
oval or pyriform sac (the inJc-har/)^ with a long duct which
opens into, or close to, the rectum, is lodged sometimes in
the liver, sometimes further back (Fig. 126, L). The ink is
ejected when the animal is alarmed, and gives rise to a dark
cloud in the water, by which its retreat is covered. There
are two branchial hearts.

The eye is lodged in an orbit and is provided with a lens.
The cartilaginous endoskeleton forms a ring surrounding the
gullet and enveloping the principal ganglia. There is usually
an internal pallial shell. It may be chambered and siphun-
culated, but in this case the last chamber is small, and hardly
larorer than the others.

The Dibranchiata are divided into the Octopoda and the
Decapoda. The Octopoda have eight arms, and possess no
pallial shell. But, in the female of one genus (Ar(/o?iauta,
the "paper Nautilus," Fig*. 131), the extremities of the an-

1 See the discussion of this question by Keferstein, in Bronn's " ThieiTeich.*'

THE DIBRANCHIATA.

461

terior pair of arms are greatly expanded, and, being turned
back over the mantle, secrete an elegant shelly structure
which covers the bodv, and serves for the attachment of the

Fig. 'ISl.—Argonaufa argo.—A, female with the expanded anns in their natural
position, embracing the shell 6; c?, the other six arms; a, the luunel. i^, ace-
tabula.

FiQ. lZ2.—Argonauta argo, male, with the Hectocotylus-tivm attached.

eggs. In this genus, and in some other Octopods ( Octopus
carina^ Tremoctojnis violaceits^ and T. §?«oyanw5), the male
is very much smaller than the female, and gives rise to a
Hectocotylus.

In Argonauta arr/o (Figs. 132, 133), it is the third arm on
the left side which becomes thus modified. At first it has
the form of a sac, within which the slender terminal part of
the arm is coiled up (Fiff. 133, B). The sac splits to give
exit to the latter (Fig. 132), and its two halves reunite on the
outer face of the base of the arm to form a chamber, which
becomes filled with spermatophores in a manner not yet un-
derstood. During sexual union the arm thus charged with

462

THE AXATOMY OF INVERTEBRATED ANIMALS.

semen is detached and left in the mantle cavity of the female
(Fig. 133, A). When first discovered it was regarded as a
parasite, and termed Trichocephalus acetabularls bj Delle

Fig. ISS.—Arf/onaufa argo.—^, male, with thehectocotylized arm inclosed in its sac;
1, 2, 3, 4, the other arms of the right side ; and 1', 2', 4', those of the kit side. A,
the hectocotylus detached.

Chiaje, while the corresponding body found in an Octopus
was called Hectocotylus octopodis by Cuvier.

In TremoctopKS^ it is the third arm on the right side
which becomes the Hectocotylus. In other Octopods^ one
or other arm is peculiarly modified, but does not become
detached or serve as a receptacle for the spermatophores.

The Decapocla have ten arms, two of which are usually
much longer than the rest, and can be protruded from, or re-
tracted into, sockets. The acetabula have horny rims, which
may take on the form of hooks.

Hectocotylization does not go further than a modification
of the form of one of the arms. There is always an internal
shell, which is either a pen, a sep)iostaire, a phragmocone, or
a combination of the latter with a pen.

1 Steenstrup, " Die Hectocotylenbildunor bei Argovuuta und Tremodopvs
erl^lart durch Beobachtuncren ahnlicher Bildungen bei den Cephalopoden."
(" Archiv fiir Naturgeschichte," 1856.)

THE DIBRANCHIATA. 463

The Teuthidce, or Squids, are characterized bj possessing
a pen. This is a lamellar, chitinous bod)^, strengthened by
one or more longitudinal ridges, which lies in a sac lodged
in the anterior wall of the body, by the lining membrane of
which it is secreted. The posterior end of the pen is com-
monly broad, and its sides may be infolded so as to form a
conical cup ( Ommastrephes).

In the Seinadce, or Cuttle-fishes, the sepiostaire, or " cuttle-
bone," which occupies the same position (Fig. 125, sA), is
composed of a broad plate answering to the pen, and likewise
infolded at its apex so as to give rise to a short cone, but cal-
cified. On the inner face of this plate a great number of deli-
cate calcified lamina?, connected by numerous short columns,
form a spongy tissue, which is full of air.^

In the Spirulidoe^ represented by the solitary genus Sin-
rula^ which is among the rarest of animals in museums,
though its shells are found piled up in countless millions on
the beaches of the islands of the Pacific, the shell is spirally
coiled and divided by septa, perforated by a siphuncle, into
chambers. The last chamber of this phragmocone, however,
is no larger than its predecessor, and the shell is held in posi-
tion by lateral processes of the mantle, which are united over
it, and probably represent the walls of the sac in which the
shell was primitively formed. The last chamber of the shell
lies in front of the axis of the helix ; the shell is therefore
coiled in the opposite direction to that of Nautilus.

In certain extinct genera (e. g., Spirulirostra)., a shell like
that of Spirula is inclosed in a dense and laminated pointed
sheath, like the hinder end of a sepiostaire, or of the pen of
an Ommastrephes,

In the Belemnitld(B (Fig. 134), which abounded in the
Mesozoic epoch, but have been extinct since that time, a
straight phragmocone is inclosed within a more or less coni-
cal, calcified, laminated structure, the guard or rostrum,
which is continued forward into a variously-shaped, usually
lamellar, pro-ostracuY)%. The pro-ostracum and the rostrum
together represent the pen in the Teuthidm.

The rare specimens of Helemnitidm in which the fossil-

1 The planes of the superimposed parallel laminre form an acute angle with
that of the principal plate of the sepiostaire. The connecting columns are
placed perpendiculariy to the laminse between which they are interposed, and
may be simple or branched. When the young Sepia leaves the ^gg^ the sepi-
ostaire already contains air.

2 Owen, ""Zoology of the Samarang," 1848.

464

THE ANATOMY OF INYERTEBRATED ANIMALS.

ized soft parts are retained, show that the arms were prO'
vided with hooks, and that there was a large ink-bag.^

Fig. 1M. — Belemnif''8, with tho remains of the body of the animal. (From a
specimen in the Museum of Practical Geology.)— a, arms with hooks ; b, head;
c, ink-bag; d, phragmocone ; e, guard.

* Huxley, " The Structure of Belemnites." (" Memoirs of the Geological
Survey of the United Kingdom," 1864.)

THE DIBRANCHIATA. 465

The genus Acanthoteuthis * {JBelemnotenthis^ Pearce) —
one of the JBelemnitidoe, in which the guard is almost rudi-
mentary, while the pro-ostracum is large and penlike — oc-
curs in the Trias, and is the earliest-known Dibranchiate
Cephalopod. The ordinary JBelemnitidce abound from the
Lias to the end of the Mesozoic period, after which they
disappear. The Sepiadm first appear in the latter half of
the Mesozoic epoch ; while the Teuthidm are represented by
genera closely allied to existing forms {Teiithopsis^ JBelern-
nosepia) as early as the Lias.

» Owen, " A Description of Certain Belemnites," etc. (" Phil. Trans.,"

18M.)

CHAPTER IX.

THE ECHIXODEEilATA.

The Echinoderms are exclusively marine animals. They
a,re always provided with a skeleton, composed of calcareous
spicula, which commonly unite into networks, and give rise
to definite skeletal plates. These generally become connect-
ed with one another by joints or sutures, but sometimes re-
main distinct. A more or less spacious peritoneal cavity
separates the walls of the body from those of the alimentary
canal. The nervous system, in those Echinoderms in which
it has been most satisfactorily made out, presents a ring,
which surrounds the gullet, and gives off radiating longitudi-
nal cords. A remarkable system of vessels, termed amhula-
cral^ which also form a ring around the gullet, is highly char-
acteristic of the E^hinoderniata. The most conspicuous and
familiarly-known Echinoderms — the Star-fishes [Asteridea),
Brittle-stars [Ophiuridea), ' Se^i-WYiihms (Echinidea)^ and
Feather-stars (Crmoidea) — have a marked radial symmetry ;
similar parts, usually to the number of five, being arranged
around a central axis ; and the body is spheroidal, discoidal,
or stellate. The Sea-cucumbers and Trepangs {Solothuridea)
are elongated and vermiform ; but the radial symmetry is
still traceable in the arrangement of the oral tentacula, the
nervous, and the ambulacral systems. It is to be remarked,
however, that, in many Echinoderms, the radial symmetry,
even in the adult, is more apparent than real ; inasmuch as a
median plane can be found, the parts on each side of which
are disposed symmetrically in relation to that plane. With
a few exceptions, the embryo leaves the Qgg as a bilaterally
symmetiical larva, provided with ciliated bands, and other-
wise similar to a worm-larva, which may be termed an Echi-
noprjediuni. The conversion of the Echinopoedium into an
Echinoderm is effected by the development of an enterocoele,

THE HOLOTHURIDEA.

467

and its conversion into the peritoneal cavity and the ambula-
cral system of vessels and nerves ; and by the metamorphosis

Fig. 135. — Diasrara exhibiting the general plan of the development of the Echino-
derm?. (After Miiller.) — A, common form whence the vermiform Holothnrid (B, W)
and the plnteiform Ophiurid or Echinid (C, C) larvae arc derived ; D, W^ younser
and more advanced stages of the Asterid (JSipinnaria) larvae ; a. month ; b, stom-
ach : c, intestine ; d, anus ; ?, ciliated band ; e', second or anterior ciliated circlet
of Bipinnaria.

of the mesoderm into radially-disposed antimeres, the result
of which is the more or less complete obliteration of the
primitive bilateral symmetry of the animal.

1. The Holothueidea. — The study of the structure of
the Echinoderms may best be commenced with the members
of this division, which, in many respects, deviate least from
such worms as the Gephyrea.

In the SynaptcB, for example (Fig. 136), the body is

iQS THE ANATOMY OF INVERTEBRATED ANIMALS.

greatly elongated and cylindrical, the mouth being placed at
one end and the anus at the other. The oral aperture is sit-
uated in the centre of a circle of tentacula, and the gullet
leads from it to an alimentary canal, without marked distinc-
tion of stomach and intestine, which extends through the
body, and is connected by a mesentery with the parietes of
the latter. The wall of the alimentary canal presents exter-
nal circular, and internal longitudinal, muscular fibres, and
its cavity is lined by a cellular endoderm.

The body-wall, or perisoma, consists of an external cellu-
lar ectoderm, cov^ering a layer of connective tissue within
which are circular and longitudinal muscular fibres. The lat-
ter are disposed in five bands, attached anteriorly to a corre-
sponding number of the pieces of a calcareous ring which
surrounds the gullet (Fig. 136, E). The separate ossicles
which compose this ring are usually ten or twelve in number,
and the five to which the longitudinal muscles are attached
are notched or perforated for the passage of the ambulacral
nerves, which proceed from the circum-oesophageal nerve to
the parietes of the body.

The integument contains numerous perforated, flat, calca-
reous plates, to which protruding anchor-like hooks of the
same substance are attached (Fig. 136, F). According to
Semper, these anchor-like bodies are developed in special
sacs with an epithelial lining.^

A spacious peritoneal caA ity lies between the parietes of
the body and the alimentary canal, and the cells which line it
are more or less extensively ciliated. Pedunculated ciliated
cups are attached to the mesentery.

The circular vessel of the ambulacral system surrounds the
gullet below the calcareous ring (Fig. 136, E, A). Posterior-
ly, it gives off various cjecal prolongations, which depend
freely into the peritoneal cavity. Some of these — the Polian
vesicles — are mere caeca ; but, in addition, there are one or
more tubular prolongations, the perforated extremities of
which are invested by a calcareous network, and are termed
the madreporic canals. Through the openings in the free
end of the madreporic canal, the interior of the ambulacral
system communicates with the peritoneal cavity. Anterior-
ly, the circular vessel gives off branches to the tentacula.
These pass between the calcareous ring on the outer side,

» See^ on this and all points relatini? to the structure of the Hnhthuridea^
the beautiful monotrraph bv Semper, " Eeisen im Archipel der Philippinen."
(" Wissenschaftliche Resultate," Bd. i. : Holothurien.)

THE HOLOTHURIDEA. 4C0

and the anterior end of the alimentary canal and the nerve-
rino- on the inner side. As each enters its tentacle, it dilates
and sends down a short ca3cal prolongation on the outer side
of the calcareous ring. The ambulacral vessels are filled with
a fluid containing numerous nucleated cells.

Contractile vessels, which accompany the intestine, and
lie on opposite sides of it, filled with a similar corpusculated
fluid, seem, notwithstanding the difference in their contents,
to represent the pseud-haemal vessels of the Annelids. These
vessels do not extend into the parietes of the body.

The nervous system consists of a ring which lies superfi-
cial to the circula/water-vessel, and from which five principal
equidistant cords proceed. These pass through the apertures
or notches in the circum-oesophageal plates already mentioned,
and each proceeds along the middle line of one of the longi-
tudinal muscular bands, to the opposite extremity of the

body.

The ambulacral nerves appear to be hollow ; or perhaps it
would be more correct to regard them as thickenings in the
wall of a neural canal, as they are in the Asteridea.^

The genital gland is single, and opens near the oral end
of the body, in the line of the attachment of the mesentery.
The branched cascal tubuli of which it is composed contain
both ova and spermatozoa, so that the Synaptm are her-
maphrodite. In the majority of the Holothuridea^ however,
the sexes are distinct.

In other Holothuridea^ the skeleton may attain a much
greater development, and even take the form of conspicuous
overlapping plates (Psolus). Moreover, the circular vessel
of the ambulacral system not only gives origin to Polian vesi-
cles, madreporic canals, and tentacular vessels, but five canals
proceed from it, pass through holes or notches in those cir-
cum-oesophageal plates to which the longitudinal muscles are
attached, together with the nerves, and run backward, along
the centre of the area occupied by these muscles, on the deep
or inner side of the longitudinal nerve. These are the radial
ambulacral vessels. In the hisrher Solothuridea^ each radial
ambulacral vessel gives off many lateral branches; these enter
contractile processes of the body-wall, which subserve loco-

1 Accordincc to Greef (" Ueber den Bau der Echinodermen," 3te Mittheilun^,
Sitzungsberichte der Gesellschaft zu Marbursf, 1872), another canal lies super-
ficial to the ambulacral nerve in the Holothuridm^ and represents the ambu-
lacral groove of the star-fishes. Teuscher, " BeitrSge zur Anatomic der Echi-
nodermen" (s/ifTiaisf-As ZafccA?**'/".^, 1876), however, maintains that this superfi-
cial canal is an artificial product.

470 THE ANATOMY OF INVERTEBRATED ANIMALS.

motion, and are the ambulacral feet^ suckers^ ov pedicels. In
accordance with the disposition of the ambulacral vessels, the
pedicels are usually disposed in five longitudinal bands, which
are the ambulacra. Sometimes {Psolus) the pedicels are sup-
pressed in two of the five ambulacra, and the other three are
disposed upon a flattened surface upon which the animal
creeps.

In the higher Holothurids, the intestine terminates in a
distinct cloaca, into which two hollow ramified organs, which
lie in the perivisceral cavity, open. The ramifications of one
of these are received between the meshes of a special plexus
of the pseud-h^emal vessels. Water is taken into, and ex-
pelled out of, the cloaca and these appendages, which, doubt-
less, subserve an excretory function, and are commonly called
respiratory trees. It seems probable that the ultimate
branches of these organs open directly into the perivisceral
cavity.*

The Cuvierian organs are simple or branched appendages
of the cloaca, the function of which is unknown. The inte-
rior of these organs is occupied by a solid substance, some-
times of a viscid nature. In some Solothuridea, the anal
aperture is provided w^ith a circlet of calcareous plates.

In many of the higher Holothurids the pseud-haemal vas-
cular system attains a great complexity, and its branches
not only extend over the alimentary canal, but, as has been
said above, closel}^ embrace one of the branched excretory
organs.

The most aberrant form of this group at present known is
the genus Rhopalodina. According to Semper, the body is
flask-shaped, and at the narrow end of the flask are two aper-
tures. One of these — the mouth — is surrounded by ten ten-
tacula ; the other, which is the anal aperture, is encircled by
ten papillse, and by as many calcareous plates. A spacious
cloacal cavity, provided with excretory organs, traverses tlie
neck of the flask, and opens by the anal aperture. The gul-
let is surrounded by a ring of ten calcareous plates. The
genital duct is situated between the cloaca and the gullet.
Ten ambulacra diverge from the centre of the enlarged aboral
end of the body, and extend, like so many meridians, to near
the commencement of the neck of the flask. In correspond-
ence with each ambulacrum is a longitudinal muscular band ;
and it is an especial peculiarity of Mhopalodina that five of

1 Semper, loc. cit., Heft iv., p. 133.

THE HOLOTHURIDEA.

471

Fts. ^BQ.—S}/napta diqifata and hihcerms (After Banr.) ^

A, ]arva wirh the bilateral ciliated band, and wheel-shaped calcareons plates:
ventral view, a, mouth and j^iHet ; b, stomach; c, intestine and anus; d, sacs
of the enterofcele (sausatre-shaped Vodies) at the sides of the stomach ; e, rudi-
ment of the ambulacra! vascular system. B, further advanced condition ot the
larva in which the oral aperture is obsolete (the so-called " pupa-stace "). and the
cilia are arranged in zones, i, tentacula; Jc, Polian vesicle ; /, the longitudinal
muscles of the bodv-wall. C. a voun? Synapta, in which the ciliated zones liave
disappeared ; with" its five tentacles and the wheel-shaped calcareous bodies at
its hinder end. m, the madrenoric canal which now opens into the cavity of the
body. D. a voun^^ Synapta whcerens with anchor-shaped calcareous spicula, ex-
cept at the hinder end of the body, where they are small and polygonal. E, lon-

» "Beitryge zur Naturgeschiclite der Synapta digitata,''^ ("Nova Acta,'*
xxxi., 1864.)

472 THE ANATOMY OF INTERTEBRATED ANIMALS.

gitudinal section of the anterior end of the body of an adult Synapta dlgitata, «,
perisoma with the longitudinal muscles and radial nerve-truolia ; 6, calcareous
plates wliich surround the gullet* c, tentacular canals ; rf, oesophagus ; e, radiat-
ing muscle? oltlie pharynx ; g^ divided ends of the circum-oral nerve; h, circular
ambulacral vessel with Polian vesicle ; i, cavity of a longitudinally divided tenta-
cle, into which a tentacular canal opens; k, generative caeca ; /, mesentery veith
the dorsal blood-vessel; w, "auditory vesicle"' on the radial nerve ; w. longitudi-
nal muscles; o. tentacular pedicels; p, oral disk. F, calcareous plate and anchor
of Synapta inhcerens.

these are attached to the anal circlet, and five to the circum-
oesophageal circlet. Until, however, it has been shown that
the circular ambulacral vessel incloses the cloaca as well as
the oesophag-us — which is highly improbable — it is justifiable
to assume that the anus of Hhopalodina is really, as in the
Crinoidea, interradial in position.

The development of the Holothuridea is extremely in-
structive. Yelk-division gives rise to a vesicular morula,
which undergoes invagination, and becomes converted into
an oval ciliated gastrula. The opening of invagination be-
comes the anus, while a mouth and gullet are produced by an
invagination of the ectoderm, near the anterior end of the
body, which unites with and opens into the blind end of the
endodermal sac, or archenteron. The completed alimentary
canal is thus composed of a gullet, a rounded stomach, and an
intestine ; and the cilia of the ectoderm usually become re-
stricted to a single band, bent upon itself, though its general
direction is transverse to the axis of the body (Fig. 135, B ;
Fig. 136, A). At a subsequent period, this single band may
be replaced by a series of hoops of cilia (Fig. 136, B). Ac-
cording to Kovvalewsky,' the embryo of Pentacta doUoluni
does not becomg ciliated at all, and that of Psolinus passes
from the condition in which the cilia are dispersed over the
surface directly into one in which it is provided with five
zones of cilia, between two of which the mouth opens. In
this condition it singularly resembles the embryo of Coma-
tula. And, indeed, in the further advanced condition of the
Psolmus, the oral end of the bodj^ surrounded by triangular
calcareous plates, within which the tentacles take tlieir origin,
has a striking resemblance to the oral end of the young Pen-
tacrinoid larva of Comatula.

The peritoneal cavity and the ambulacral vessels take their
origin,' in a very remarkable manner, from the archenteron,

» " Mem. de 1' Acad, de St.-Petersbourcr," 1868.

' See Metschnikoff, " Studien iiber die Entwickeluncr der Echinodermen und
Nemertineu" ("Mem. de PAcad. de St.-P6tersbourg," xiv., 1869) ; and espe-
cially the very satisfactory memoir of Salenka, " Zur Entwickelung der Hole-
thurien" {Zeitsckrift fiir wiss. Zoologie^ 1876).

THE DEVELOPAIENT OF THE HOLOTHURIDEA.

473

before the oesophageal invagination reaches it. The anterior
part of the archenteron gives off a cascal process which, be-
coming a sac, is constricted off from the archenteron as what
Salenka terms a vaso-perito7ieal vesicle. This vesicle changes
its position to the left side of the alimentary canal, and then
sends a narrow, duct-like diverticulum toward the dorsal re-
gion of the ectoderm, which eventually coalesces with the
latter, the cavity of the diverticulum opening on the exterior
by a rounded pore. The vaso-peritoneal vesicle now divides
into two portions, one of which — the amhulacral sac — remains
connected with the exterior by the duct, and constitutes the
foundation of the whole of the ambulacral system of vessels ;
while the other — the peritoneal sac — gives rise to the peri-
toneum. The former becomes five-lobed, grows round the
gullet, and gives rise to the tentacular and ambulacral canals
with the Polian vesicle, or vesicles ; while the duct, de-
taching itself from the dorsal wall, becomes the madreporic
canal.

The latter divides into two vesicles, which arrange them-
selves at the sides of the stomach. The stomach takes on
a more cylindrical shape, and these vesicles become the " sau-
sage-shaped bodies " (wurstformige KOrper) observed by Mul-
ler (Fig. 136, A). They gradually increase in size, and, grow-
ing round the alimentary canal, unite above and below it.
Thus a cylindrical cavity with a double wall is formed be-
tween tlie endoderra and the ectoderm. The inner wall of the

Fig. 137.— DevelopnK^t of a Holothurid. (After Miiller.)— A, early condition of the
larva {Auricula?'ia) : g, the dorsal pore of k, the ambulacral sac. B, later stacre:
c', intestine ; g, dorsal pore ; /,/', circular ambulacral vessel with its prolonga-
tions ; i, calcareous body. C, youn? Holothuria with circular ciliated bands : g,
madreporic canal ; /', Polian vesicle.

cavity applies itself to the alimentary canal, and, aided by
the mesoblastic cells which appear to be developed from the

474 THE ANATOMY OF INVERTEBRATED ANIMALS.

endoderm, becomes the muscular and peritoneal coat of that
viscus ; while the outer wall, attaching ilself to the ectoderm,
or to the mesoblastic cells which line it, is, with them, con-
verted into the muscular and peritoneal investment of the
parietes of the body. The interspace between the two is the
peritoneal cavity.

In the mean while, the body of the embryo elongates, the
tentacula are developed around the mouth, the cihated bands
disappear, and the Holothurid Echinoderm is complete.

Thus it is clear that the peritoneal cavity of the Holo-
thurid is an enterocoele, and that it answers to the perivis-
ceral cavitj^ of Sagitta, or of the Brachiopoda y and further,
that the ambulacral vessels are also modifications of the entero-
coele. Moreover, it is obvious that the structures which are
developed between the enterocoele and the ectoderm and en-
doderm, answer to those which are evolved from the meso-
blast in other animals ; and that the adult Echinoderm stands
in the same relation to the JEchinopcedimn as an Annelid does
to its embryo ; the adult form being due to the peculiar ar-
rangement of the parts developed froai the mesoblast. No
part of the Echinopaadium is cast off in the course of the de-
velopment of the Holothuridea.

2. The Asteridea. — A Star-fish is comparable to a Holo-
thurid, the ambulacra of which are restricted to its oral half,
flattened out so as to have a very short axis ; while its equa-
torial diameter is greatly increased, and produced in direc-
tions corresponding with each ambulacrum. The result would
be a disk, having the form of a pentagon, or of a five-rayed
star, with ambulacra only on that face of the disk which
bears the mouth. Hence the ambidacral, and the opposite,
or antarnhulacral, faces are of equal extent.

Most Asteridea are like five-rayed stars, but some are
pentagonal disks (Goniaster), and some fevr (Solaster) have
more than five rays. In Brisinga^ the rays are much more
different from the disk than usual, and the genus thence
acquires an outward resemblance to an Ophiarid.

All the Asteridea are provided with a skeleton made up
of plates or thick rods, composed of a dense calcareous net-
work. A deep groove, radiating from the mouth to the end
of the ray, marks the position of each ambulacrum, and the
sides of this groove are supported by two series of ambu-
lacral ossicles, which meet and articulate together in the
middle line or roof of the groove. The ambulacral nerve and

THE ASTERIDEA. 475

canal lie superficial to these ossicles. There are no oral ten-
tacula.

The five-rayed body of the commonest of British Star-
fishes,^ the Five-finger (Ui'aster^ or Asteracanthion^ ruhens)^
presents an oral face^ in the centre of which the mouth is
placed, and an opposite or ahoral face. The hardly-discern-
ible anal ajoerture is situated not exactly in the centre of this
face, but close to it. The mouth, which varies very much in
size, lies in the middle of a soft membranous oral disk, A
deep furrow, the amhulacral groove^ occupies the middle of
the oral surface of each ray, and is nearly filled by contractile
sucker-like pedicels, with circular discoidal ends, apparently
arranged in four longitudinal series. The deepest part of the
groove is at its central end, where its lining passes into the
oral membrane. The shallowest part is at its distal end, where
it terminates against a median projection, the peduncle of
the eye, on the aboral side of which is the single median
ocular tentacle. Lines drawn from the mouth along each
ambulacrum are termed radii^ and the regions occupied by
the ambulacra are said to be radial. The parts of the body
situated between the ambulacra are interradial. The lateral
walls of the ambulacral grooves of adjacent ambulacra unite
at the circumference of the oral disk, and give rise to five
interradial angles. On one side of the aboral face of the
centre of the body, between the origins of two of the rays,
and therefore interradial in position, is an oval or somewhat
pentagonal, slightly convex, porous plate, the surface of
which is covered with narrow, meandering grooves. This is
the tnadreporic tubercle^ or madreporite.

The perisoma, or wall of the bodj', upon the aboral face,
and upon the sides of the rays, is everywhere covered with
short spines. In the intervals between these, groups of deli-
cate membranous tubuli, which are closed at their free ends,
project. Small two-pronged, pincer-like bodies, the pedicel-
laricB^ are attached to the spines and to the perisoma between
them, and during life are seen to twist about and snap.

The perisoma presents, externally, a cellular ectoderm,
provided wdth a thin cuticle, which bears numerous cilia. Be-
neath this lies a mesoderm, containing connective and mus-
cular elements, in which the calcareous structures which con-
stitute the skeleton are lodged. On the inner side of the
perisoma, a ciliated epithelium lines the perivisceral cavity.

* Compare Hoffmann, '' Zur Anatomie der Asteriden." (" Niederliindischea
Archiv," Bd. ii., 1874, }

476 THE ANATOMY OF INVERTEBRATED ANIMALS.

The separate elements of which the skeleton is composed
may be divided into three groups: the ossicula, which, joined
end to end and united by connective and muscular tissues,
constitute the chief framework of the body ; the spmeSy at-
tached to the ossicula by ligamentous fibres at one end, and
free at the other; and the calcareous structures contained in
the 2)edicellarim. On the antambulacral wall of the body,
the ossicula are elongated rods of very unequal lengths,
united in such a manner as to leave polygonal, rounded, or
elongated meshes. The sides and roof of each ambulacral
groove, however, are bounded by two series of regularly-dis-
posed and similar ambulacral ossicles, which lean against one
another in the middle line above, diverge so as to inclose the
ambulacral groove, and, at their outer ends, abut upon thick,
short adanibulacral ossicles, which lie at the sides of the
groove (Fig. 139, D).

Between every two ambulacra! ossicles in the same half of
the ambulacrum there is a canal, formed by the junction of
notches in the oral and distal faces of the two ossicles. Con-
sequently there is a half -pore on the oral, and another half-
pore on the distal, face of each ossicle. The half-pore on the
oral face is always internal in position to the half-pore on the
distal face, and, as the part of the ambulacral ossicle which lies
between the two is thin, the row of pores, though it is really
single and bent in a sharp zigzag, appears at first sight to be
double. The ducts, which connect the ambulacral vesicles
with the pedicels, traverse these pores ; and the comparatively
large and very flexible and extensile pedicels are thus so
closely packed together, that they appear to form a double
row on each side of the middle of the ambulacrum.

At the circumference of the oral disk, the ossicles of the
ambulacra, diminished in size, and closely united together,
form a pentagon, the angles of which answer to the ends of
the ambulacral grooves, round the oesophagus. The con-
joined outer ends of the pair of ambulacral ossicles nearest
the mouth project on the oral face, outside the buccal mem-
brane, as five vertical crests, armed with strong spines, which
are beset with pedicel larisD. In correspondence with these,
five falciform folds of the perisoma, more or less calcified, pro-
ject into the cavity of the body. They are interradial in
position, and extend up to the aboral wall. Their inner
edges are free, and look toward the stomach ; with one of
them, the madreporic canal and the sinus which accompanies
it are closely connected.

THE ASTERIDEA. 477

The spines are more or less movably united with the
ossicula, but there are no such regular joints as are met with
in the JSchinidea, The pediceliarioe are supported upon
short, flexible peduncles. The skeleton of each consists of
two blades articulated with a basal piece. From the centre
of this, very strong adductor muscles proceed to the inner
faces of the blades, and weaker Hbres, attached to the exterior
and to the outer faces of the bases of the blades, act as
divaricators.

The gullet opens into a wide stomach produced into five
large cardiac sacSy the walls of which are subdivided into
many sacculi. Each cardiac sac is radial in position, and may
extend a short way into the cavity of the arm, to which it
corresponds. On the aboral side of these sacs the alimentary
canal suddenlj'^ narrows, and then dilates again into a shallow,
but w4de, pentagonal pyloric sac, the angles of which are
produced into five tubes. Each of these passes along the
middle of the aboral face of a ray, and divides into two
branches, which run parallel with one another through half
or two-thirds the length of the ray, and end blindly. The
branches give off numerous caecal dilatations, arranged in
pairs on opposite sides, and these hang down into the cavity
of the ray. The edges of the pentagonal pyloric sac, and the
aboral faces of its sacculated branches, are connected by
mesenteric folds with the aboral perisoma. The oral faces of
the cardiac sacs are similarly connected by pairs of mesenteric
folds with the sides of the corresponding series of ambulacral
ossicles. The aboral face of the pyloric sac presents an aper-
ture closed by projecting valvular folds, which leads into the
short tubular intestine. The latter terminates in a minute
anal pore, situated nearly in the centre of the aboral face of
the body. The intestine receives the duct of a caecum divided
into two main branches, each of which has many minor sub-
divisions. If the animal, having its mouth downward, is di-
vided into two halves, by a vertical plane passing through the
mouth, the central point of the aboral face, the madreporic
tubercle, and the middle line of the ray opposite to the tu-
bercle ; and if this ray is anterior, then the anus opens into
the left posterior interradial space, and the caeca lie partly in
this and partly in the left anterior interradial space.

The nervous * and vascular systems of the Star-fish are so

^ See Wilson, " The Nervous System of the Asterida" (" Transactions of
the Linngean Society," 1862), and the later contributions of Prof. Teuscher, cited
below.

478 THE ANATOMY OF INVERTEBRATED ANIMALS.

closely related to one another tliat they may be best consid-
ered together ; and as there is least difficulty in making out
their arrangement in the ambulacra, the study of them may
be commenced in this region.

When the suckers of an ambulacrum are carefully cut away,
a longitudinal ridge is seen to lie at the bottom of the groove
between their bases. This ridge is the amhulacral nerve.
Followed to the apex of the ray, it ends upon the eye and its
tentacle ; in the opposite direction, it reaches the oral disk,
at the periphery of which it divides, and, skirting the margins
of the disk, joins the branches formed by the bifurcation of
the adjacent ambulacral nerves, thus giving rise to a subpen-
tagonal ring round the mouth.

The eye ^ is a thick cushion-like expansion of the ectoderm
continuous with the ambulacral nerve. In it are imbedded
many clear oval bodies surrounded by pigment, which appear
to represent the crystalline cones of a compound eye.

The tentacle which lies on the aboral side of the eye re-
sembles one of the pedicels in structure, but has no terminal
sucker ; its function appears to be tactile.

In a good transverse section of one of the arms or rays of
the Star-fish, the nerve is seen to be a band-like thickening of
the ectoderm, the cells of which have become peculiarly mod-
ified, but which is continuous latterly with the ordinary ecto-
dermal covering of the pedicels. This band-like nerve consti-
tutes the superficial wall of a canal, which extends through
the whole length of the ambulacrum, and may be termed the
amhulacral Pleural canal. It is divided by a longitudinal
septum. At its oral end, as has been seen, each ambulacral
nerve, when it reaches the oral membrane, divides into two
divergent branches, which unite with the corresponding
branches of the other ambulacral nerves to form the oral ring.
Answering to the latter is a w ide circular neural canal, into
which the ambulacral neural canals open.

In the transverse section of the arm, a second and much
larger canal is seen to lie between the conjoined ends of the
ambulacral ossicles and a strong septum, containing trans-
verse fibres, which separates it from the neural canal. This
is the radial canal of the ambulacral system of vessels. At
its oral end it opens into the circunioral ambulacral vessel,
which lies close to the ossicles to which the margins of the
oral membrane are attached. From opposite sides of the

> Conf. Ilacckel, Zeitschriftfiir wiss. Zoologie^ 1860.

THE ASTERIDEA. 479

radial canal, short branches are given off, which pass between
the ambulacra! ossicles, and each opens into the neck of a
relatively large sac, with muscular walls {ambulacral vesicle),
which lies on the aboral face of the ambulacral ossicles in
the interior of the ray. The neck of the ambulacral vesicle
pisses in the opposite direction into one of the pedicels.
Thus th3 ambulacral vessel communicates with the cavities of
all the pedicels on the one hand, and with the cavity of the
circumoral ambulacral vessel on the other. Five pairs of
small eminences, consisting of caeca, which open into the cir-
cumoral vessel, are seated upon it ; and from one part of it,
opposite one of the interradial falciform folds already men-
tioned, springs a canal, which, taking a sinuous course, passes
to the aboral face, and terminates beneath the madreporio
tubercle ; this is the madreporic canal. It is not a simple
tube, but, as Sharpsy first observed, its walls are doubly in-
voluted so as partially to obstruct its cavity, and it is strength-
ened by annular calcifications. The pores of the madreporic
tubercle place the cavity of the madreporic canal in commu-
nication with the exterior, whence it follows that the cavities
of the whole ambulacral system must be directly accessible to
the sea-water in which the Star-fish lives. The madreporic
canal is invested by the lining membrane of the peritoneal
cavity. This incloses a sinus, which accompanies the madre-
poric canal, and into the interior of which a fold projects.

There is no great difficulty in ascertaining the existence of
the structures which have now been described, and all anato-
mists are agreed as to the nature of the ambulacral system. But
whether the neural canals are to be considered as a special
system of blood-vessels, and the sinus which accompanies the
madreporic canal, a heart, as is usually assumed, appears to
me to be very doubtful.^ I am disposed to think, in fact,
that not only these canals, but the circular, or rather pentag-
onal, vessel which has been described as situated on the abo-

^ Since Tiedemann's time, the presence or absence of a blood-vascular sys-
tem in the Star-fishes has been alternately asserted and denied. The recent
investigations of Greef, " Ueber den Bau der Echinodermen" ("Marburg
Sitzuagsberichte," 1871-'72), Hoffmann {I. c), and of Teuscher, " Beitrage zur
Anatomic dor Echinodermen" {Jenaische Zeitschrift^ Bd. x.), are in favor
of the existence of the " anal ring," and of an extensively ramified system of
canals, connected with it and with the neural canals. But it does not appear
to me tliat the facts, as they are now known, justify the assumption that these
canals constitute a distinct svstera of blood-vessels. Injections show that all
these canals communicate with tlie ambulacral vessels, and with the exterior,
by means of canals in the madreporic tifbercle which open partly outward,
partly into the madreporic canal, and partly into the sinus which accompa-
nies it, and communicates with the circumoral neural vessel.

480 THE ANATOMY OF IXVERTEBRATED ANIMALS.

ral face of the body, around the anus, giving off various
branches to the viscera, and communicating with the so-called
heart, are mere subdivisions of the interval between the
parietes of the body and those of the alimentary canal, aris-
ing from the disposition of the ambulacral vessels and that of
the walls of the peritoneal cavity ; both of which, as their
development shows, are the result of the metamorphosis of
saccular diverticula of the alimentary canal, which have en-
croached upon, and largely diminished, the primitive perivis-
ceral cavity which exists in the embryo.

The peritoneal cavity of the body and rays is filled with
a watery corpusculated fluid ; a similar fluid is found in the
ambulacral vessels, and probably fills all the canals which
have been described. The corpuscles are nucleated cells,
which exhibit amoeboid movements ; and the fluid so obvious-
ly represents the blood of the higher animals, that I know not
why the preposterous name of " chylaqueous fluid " should
have been invented for that which is in no sense " chyle,"
though, like other fluids of the living body, it contains a good
deal of water. As the cavities of the tubular c«ca of the
perisoma communicate freely with the general cavity, and
their walls share in the general ciliation of the lining of the
cavity, it is very probable that they may subserve the func-
tion of respiration.

The genital glands are situated in pairs, interradially, at
the junction of the body with the rays. Each gland is di-
vided into a number of elongated processes, the common base
of which is attached to the face of one of the interradial septa,
while the processes project freely into the cavities of the arms.
According to Hofi'mann and Greef, the inner cavities of the
genital processes are filled when the vascular system is inject-
ed. It is possible, therefore, that the genital glands are mere-
ly processes of the mesodermal layer, in the walls of which
the genital products are developed ; in which case there would
be a close approximation between the genital glands of the
Star-fishes and those of the Crinoids. According to Greef,
the external openings of the genital glands are visible in
JJraster, in the breeding-season ; in other Star-fishes, they are
conspicuous in the interradii of the aboral face of the body.
In Luidea, Ophidiastei\ and some other genera, the glands
extend far into the interior of the arms ; and Prof. G. O, Sars ^

» "Researches on the Structure and Affinity of the Genus Brisinga,''^ 1870.
In this important memoir the author proves that Brisinga is a true Asterid,
and not, as has been supposed, a transitional form between the Asteridea
and the Oj)hivridea,

THE DEVELOPMENT OF THE ASTERIDEA.

481

has pointed out that, in Brisinga endecacnemos, the genitalia
are numerous distinct glands, arranged in two series, one on
each side of the middle line of the central half of each ray.
Each of these ovaries or testes has a separate aperture.

In seme Star-fishes, as in some Holothurids, the embryo
passes into the Star-fish form without any free larval stage.
But, more usually, an Echinopaedium is formed in the same
way as in the Holothurians, though it presents difi'erences
in the arrangement of its ciliated bands, and especially in
their prolongation into numerous lobes or narrow processes,
as in the remarkable form originally named Mipirmaria,
(Fig. 135, D D', and Fig. 138). It has no calcareous skel-
eton.

Fig. 138.— a youn? ^.5fe?7(7 larva {Bijnnnaria, after Muller).— A, ventral, B, lateral,
views of larva (Bipinnaria). C, Bipinnaria with rudiment of the Star-fish : a,
mouth : b, oesophagus ; c. stomach ; (/, intestine ; o, anus ; x, ventral, y, dorsal,
Bide of the anterior end of the hody ; <7. rf'. ciliated bands ; h, caecal diverticulum
forminsr the rudiment of the amhulacral vascular system, and opening externally
by the pore g.

According to the observations of Prof. A. Agassiz,* which
have been confirmed by Metschnikoff and Greef, the amhu-
lacral vessels commence as diverticula of the stomach, which,
becoming detached from the alimentary canal, give rise to the
peritoneal cavity, and to all the substance of the body be-
tween the endoderm and the ectoderm. '^ A portion of one of
these diverticula, however, separates itself from the rest,

1 " Embryolosry of the Star-fish." (" Contributions to the Natural History
of the United States," v., 1S'j4.) The species, the development of which is
described in this important memoir, are Asteracanthion pallidus and A. hery-
linus.

2 Probably independently-developed ruesoblastic cells contribute to the
formation of the mesoderm, as in the Holothurids,

21

482 THE ANATOMY OF INVERTEBRATED ANIMALS.

opens externally by a pore, and becomes metamorphosed into
the ambulacral vessels. But this ambulacral diverticulum
does not surround the gullet, and consequently a new mouth
is developed in the centre of the ambulacral ring. The larval
mouth and gullet are abolished, and the greater part of the
body of the Echinopagdium is separated from that portion
which contains the stellate Echinoderm. The latter results
from the metamorphosis of the mesoderm, which is modeled
upon the different divisions of the enterocoele, and incloses
the middle portion of the alimentary canal.*

The Ophiuridea. — The brittle Stars, though they re-
semble the ordinary Star-fishes in form, differ essentially from
them, not only in the structure of their skeleton, but in the
characters of the Echinopaedium. The ambulacra are con-
fined to the oral aspect of the body, so that, as in the As-
teridea, the ambulacral and oral, the autambulacral and the
aboral surfaces, respectively coincide. The mouth is situated
in the centre of the oral face, but no grooves radiate from it
along the ambulacra, which are covered by a series of plates
of the skeleton. The alimentary canal is a simple gastric sac
without caeca, and has no intestine or anus. In contradistinc-
tion from the Star-fishes, the prolongations of the peritoneal
cavity into the rays are very narrow.

The typical Ophiuridea possess a very complete calcareous
skeleton, which, on the body, and on the exterior of the rays,
has the form of plates. On the body, the disposition of these
varies much ; but five of them, which are situated inter-
radiallv in the neis^hborhood of the mouth, are often laro^er
than the others, and are termed scuta huccalia.

Each ray contains an internal solid axis, composed of a
single series of quadrate axial ossicles (Fig. 139, C, a), each
consisting of two lateral halves united by a longitudinal
suture, and articulated together by tenon and mortice joints
upon their terminal surfaces. Each of these ossicles (which
are sometimes termed vertebral) is surrounded by four plates
— one median and antambulacral (Fig. 139, C, h), two lateral
(Fig. 139, B, o), and one median and superambulacral (Fig.
139, A, d). The lateral plates may meet in the middle line
on both the ambulacr.d and the antambulacral faces. Be-

^ Greef (Z. c.) has worked out the development of Uraster ( Aster acantJ/ion)
ruhens^ the larval form of which resembles the Bipinnaria and Brachiolaria
of Helsinc^fors, described by Miiller. Parthenogenesis appears to occur in
this Star-fish.

THE OPHIURIDEA.

483

tween the lateral plates are the apertures by which the pedi-
cels make their exit. The oral aperture is surrounded by five
oral angles^ each of which consists of five pieces. The two

Fig. 139.— a, ventral, B. lateral, views of a ray of O^hhtratextitraia. (After Muller.)
C. transverse section ; a, axial or " vertebral" ossicle of ray ; 6, antambulacral
plate ; c, lateral plate; d, ventral or superambulacral plate. D, section of a ray of
an Asterid, AstropecUn aurantiacus (after Gaudry) : a, ambulacral or " verte-
bral " ossicles ; b. adambulacral ossicles ; c, c', marginal ossicles ; d, paxillae of
antambulacral surface.

constituents of the axial ossicle which lies at the oral end of a
ray become movably articulated with one another, while
each anchyloses with an interambulacral piece. Transverse
muscles connect the two interambulacral pieces, the oral
edg-es of which are articulated with a long, narrow plate, the
torus angidaris (Fig. 140,/*). The free surface of the torus
angularis lies in the walls of a sort of vestibule in front of
the mouth. A number of short, flat processes, the paloe angu-
lares, are articulated with it, and moved by special muscles.
They doubtless perform the function of teeth. Rudimentarj''
representatives of the calcareous ring of the Solothuridea
and of the parts of the lantern of the Echinidea exist as deli-
cate calcareous plates, which lie on the circular ambulacral ves-
sel. The latter is usually provided with csecal appendages,
or Polian vesicles. The madreporic canal ends on the sur-
face of one of the scuta huccalia y the radial ambulacral ves-
sels run in the arch between the axial ossicles and the super-
ambulacral plates. The nerve lies superficial to the super-

484 THE ANATOMY OF INVERTEBRATED AXIMALS.

ambulacral vessel, but is also covered by the superambulacral
plate. A neural canal lies between the nerves and the ambu-
lacral vessels. The pedicels are tentaculiform, and have no
vesicles at tlieir bases. The genital glands are lodged in the
disk, and pour their products into the peritoneal cavity, which
communicates freely with the exterior by vertically-elongated
apertures placed interradially on its margins.* According to
Metschnikoff, Ophiolepis squamata is hermaphrodite.

The early conditions of the embryos of most Ophinridea
are similar to those of other Echinoderms, and acquire the
characteristic bilateral ciliated zone ; but in some the embryo
does not become an Echinop)oedm'm^ but passes directly into
the adult condition. Thus Krohn discovered that the embryo of
Ophiolepis ciliata is developed within the body-cavity of the
parent, to which it adheres by a kind of pedicel. "Where an
Echinopc^dium stage exists, the larva is a Fluteiis (Fig. 135,
C C). The dorsal wall of the body of the embryo exhibits a

Fig. 140. — A, OpMoIejns ciliata, oral skeleton from within (after Miiller) : a, dor-
sal marjrinal plates ; b. ventral plates ; rf, vertebral osfeicles ; e, interambulacral
pieces of oral angle ; /, torne annularis ; q, apermrep for oral tentacles ; h, posi-
tion of nervous collar ; i, impression of circular ambulacral vessel ; k, orifice in
the first ambulacral plate for the tentacular branch of the oral vessel ; o, palse angu-
lares. B, Adrophyton, onil skeleton seen from within (.after Muller): m »2,peris-
tomial plates ; other letters as in A.

median conical outofrowth : along: the course of the ciliated
band symmetrically-disposed processes are developed ; and

1 Miiller, " Ueber den Bau der Ecbinodermen " ("Abh. Berl, Akad.y
1853) ; Teupcher (I. r.) ; Simrock, " Anatomic und Schizogonie der OpJiiaetis
mrens^^ (Zeitschriftfur wiss. Zootogie^ 1876). The latter writer describes nu-
merous apparently coecal diverticula of the circular ambulacral canal, and of the
necks of the Polian vesicles {vasa amhiclacralia cavi) which traverse the peri-
toneal cavity in all directions.

THE ECHINIDEA. 485

these outgrowths are supported by a calcareous skeleton,
which is also bilaterally symtiietrical. Metschnikoff^ has
made the interesting observation that in an Ophiurid (prob-
ably Ophiothrix, fragills) the whole system of perivisceral
and ambulacral cavities arises from two bodies, one situated
on each side of the gullet, wliicli are solid, though it is possi-
ble that they may primitively have been hollow diverticula of
the archenteron. Two cellular masses become detached from
these bodies, apply themselves to the sides of the stomach,
and are converted into disks, from which the parietal and vis-
ceral walls of the peritoneal cavity take their origin. The
rest of the solid body on the left side of the gullet acquires a
vesicular character, opens by a dorsal pore, and grows round
the gullet, to give rise to the circular ambulacral vessel. The
other solid body disappears. The mouth of the Echinopce-
dium becomes that of the Ophiurid.

It cannot be doubted that these solid bodies take their
origin, in the same way as in other Echinopa^dia, from the
hypoblast ; and thus the question arises. How far does the
mesoblast thus formed differ from that which arises by the
mere outgrowth of cells from the hypoblast, as in the Dog-
fish, and how far does this case tend to render it probable
that a schizocoele is only a modification of an enterocoele ?

The Echinidea. — An ordinary Sea-urchin is comparable
to a Holothurid, with the body distended into a more or less
globular form, and with a skeleton in the form of regular
plates arranged in meridional series ; those plates which cor-
respond with the ambulacral vessels being superficial to the
latter, and consequently perforated by the canals which pass
from the ambulacral vessels to the pedicels.

In the Eohinidea^ as, for instance, in the ordinary Echinus
or Sea-urchin, the perisoma round the mouth (peristome)
is usually strengthened for some distance by irregular oral
plates. In addition, ten rounded plates are placed in pairs
close to the lip ; these support as many pedicels, and are per-
forated by. the canals of the latter. A much smaller space
around the anus (per'iproct) is similarly protected by anal
plates. The rest of the body is supported by a continuous
wall made up of distinct, more or less pentagonal plates, usu-
ally firmly united by their edges, which is called the corotia.
Of these plates there are twenty principal longitudinal series,

* " Studien fiber die Entwickeluns: dor Echinodermen und Neraertinen.'*
(" Mem. Acad. St.-Petersbourg," 1869.)

486

THE ANATOMY OF INVEKTEBRATED ANIMALS.

constituting the great mass of the corona ; and ten single
plates, which form a ring around its aboral or apical margin.
The twenty series of longitudinal plates are disposed in ten
double series — five ambulacral and five interambulacral —
alternating with one another throughout the circumference of
the corona. Each double series of plates presents a zigzag
suture in the middle line, formed hy the alternating arrange-
ment of the triangular extremities of its component elements.
The sutures between the respective series of ambulacral and
interambulacral plates, on the other hand, are less obvious

V X

Fig. 141.— Diasrram exhibiting the relations of the different systems of orqrans in an
Echinus.— a, mouth ; 6, teeth; c, lips; ^, alveoli ; «, falces ; /, auriculae; g;, re-
tractor, and A, protractor, muscles of lantern ; i, mad reporic canal ; ^, circu-
lar ambulacral vessel ; I, polian vesicle ; m, w, o, ambulacral vessel ; p, pedal vesi-
cle ; Q, g, pedicels ; r, spine ; s, tubercle to which it is articulated ; t, pedicellarise ;
u, anus ; v, madreporic tubercle ; x, ocular spot.

and more straight. Each ambulacral plate is subdivided by a
greater or less number of sutures, which traverse it obliquely,
into a corresponding number of minor plates ; and these, in-
asmuch as they are perforated by the canals or pores, which
give exit to the two vessels whereby each pedicel is placed in
communication with its basal vesicles and with the ambula-
cral vessel, are called pore-plates. Throughout the sjreater
part of the length of an ambulacrum of the common Echinus

THE ECHINIDEA. 487

spha^ra (Fig. 142, A) each ambulacral plate is thus divided
into three pore-plates, traversed altogether by six pores, or
short canals. The outer openings of these canals are arranged
close together in pairs upon little, excavated, shield-shaped
elevations, or iimhones^ sculptured on the outer or interam-
bulacral half of the face of the ambulacral plate ; but their
inner extremities are much wider apart. A pore-plate, or
subdivision of the ambulacral plate, thus corresponds with
each pair of pores, and therefore with each pedicel. Loven *
has shown that the pore-plates are the primitive ambulacral
ossicles in the Echinoidea. At its apical extremity, in fact,
the ambulacrum is composed of only two small ossicles, wdiich
meet in the middle line. Each of these primitive ambulacral
ossicles is perforated by a single or double pore for the pedi-
cel which it bears. But as, in the course of the growth of the
corona, new primitive ambulacral ossicles are added between
the ocular plate and those already formed, the latter shift
toward the oral end of the ambulacrum, and grow in corre-
spondence with the larger space which they have to fill. But
they grow unequally ; and while all retain their primitive con-
nections with the adjacent interambulacral plates, some lose,
while others retain, their median union with the correspond-
ing ossicles of the same ambulacrum. The former, therefore,
are, as it were, pushed away from the middle line by the union
of their encroaching predecessors and successors. Groups of
the prim-itive ambulacral plates, thus modified, enter into close
union, and constitute the complex ambulacral plates of the
fully-developed ambulacrum.

In the genus Cidaris, the primitive ambulacral plates en-
large, but do no coalesce into secondary ambulacral plates ;
hence the distinction between ambulacral plates and pore-
plates vanishes. The ambulacral plates are continued on the
peristome to the margins of the mouth, and here they become
somewhat altered in form, and their edges overlap.

In the living genus Asthenosoma, and in certain extinct
Echinidea (Lepidocentriis^ Echinothuria), the plates of the
corona are loosely united and overlap one another ; while, in
the extinct palaeozoic Perischoechinidoe^ there are more than
two series of interambulacral plates, those in the middle of
each interambulacrum being hexagonal.

In Echinus^ the apical extremities of the ambulacra abut
upon the five smaller of the ten single plates which surround

1 " Etudes sur les Echinoidees." (" Kongl. Svenska Vetensk-Akad. Hand-
lingar," Bd. ii., 1875.)

4:88 THE ANATOMY OF INVERTEBRATED ANIMALS.

the periproct. Each of these is perforated, and supports the
eye-spot; it is thence called an ocular plate. The apical ex-
tremities of the interambulacra, on the other hand, correspond

Fig, li2. (After MiiTler.)— A, throe ambnlacral plates of Echinus sphoRra, 6x1111)11-
ins? the siiturei? of the pore-plates of which each ambulacra] plate is composed. B,
part of the petaloid ambulacrum of a Clypeastroid.

with the five larger plates, which alternate with the ocular
plates, and, like them, are perforated. The aperture is, how-
ever, larger, and constitutes the exit for the generative prod-
ucts. One of these five genital plates is larger than the
others, and presents a peculiar porous convex surface, which
is the raadreporic tubercle or 'nxadreporite. The latter is
therefore interambulacral in position, as in the Star-fish.

Comparison with the elongated Echinoderins shows that
the madreporite lies in the right anterior interradius of the
sea-urchin, so that the anterior ambulacrum is that which lies
to the left of the madreporite, when the latter is directed for-
ward. In consequence of being able to distinguish this odd
or anterior radius, it is possible in any of the Echinidea to
separate the three anterior ambulacra, as the trivium^ from
the two posterior, the hwium ; and in the fossil genus Dy-
saster, this separation of the ambulacra into trivium and bivi-
um exists naturally. Milller has pointed out that in all the
flattened Echinidea, with a special ambulatory surface, the
latter is formed by the bivial ambulacra and interambulacra,
while, in the similarly modified Holothuridea, the animal rests
upon the trivium.

Within the circle formed by the genital and ocular plates
the periproct presents a variable number of calcifications, of
which one, the anal plate, is larger than the rest. The anus
lies excentrically, between this plate and the posterior margin
of the periproct.

With the exception of certain palaeozoic forms (JPaloechi-

THE ECHINIDEA. 4S9

nifs), tlie composition of the skeleton of the Echinidea is
alwajs essentially similar to that which has just been de-
scribed ; but the form of the body and the relative positions
of the anal and oral apertures may vary very much. In the
Echinoida ( Cldaris^ Echinus) the body is spheroidal, and
the oral and anal apertures are opposite and central, or very
nearly so. In the Clypeastroida ( Clypeaster^ Echinocyamus)
the form of the body varies from a spheroidal to an exces-
sively flattened and even lobed shape. The mouth remains
central, but the anus varies in position, from the apical sur-
face to the margin, or even to the oral surface, as in Echino-
cyamus. In the remaining division of the JEchinidea, the
Spatcmgoida {S2Jatangus^ Amphidotus^ Ananchytes)^ the
form is usually a somewhat depressed oval, and both the oral
and the anal apertures are excentric. The madreporite and
the genital and ocular plates, on the other hand, remain in
the centre of the aboral region in all the Echinidea.

The ambulacra present important variations in the three
divisions of the Echinidea, In the Echinoida thev are ho-
mogeneous^ presenting the same composition from their oral to
close to their apical extremities, and having the pores and
pedicels similar throughout. Furthermore, the ambulacra are
widest in the middle, and taper gradually to each extremity
(Echinus)^ or are of nearly the same size from one end to the
other (Cida7'is).

In many Clypeastroida, on the contrary, the oral and the
apical portions of each ambulacrum differ very widely, or are
heterogeneous. The apical moiety is usually very wide in the
middle, and tapers to a point marginally, where it joins the
oral portion. Hence there is an appearance of five petals
diverging from the apex ; and such ambulacra are called petal-
oid (Fig. 142, B). In the oral portions of the ambulacra, on
the contrary, the pores are either scattered widely over the
ambulacral, and sometimes over the interambulacral, plates,
ioTming^ pore-arece ^ or they are arranged in bands which ram-
ify over the interambulacral as well as the ambulacral plates,
giving rise to what Miiller has termed j^ore fasciae. In the
Spatangoida (Fig. 143) the ambulacra commonly present
the same heterogeneous character, but the oral portions are
not arranged in fascias ; and it not unf requently happens that
the anterior ambulacrum becomes more or less abortive, so
that only four petals are obvious on the apical surface, instead
of five.

The growth of the shell of the Echinidea is effected in

490 THE ANATOMY OF INVERTEBRATED ANIMALS.

two ways : partly by addition to the circumference of the
existing plates, partly by the interpolation of new ambu-
lacral and interambulacral plates at the apical end of each
series, between it and the ocuiar or genital plate, as the case
may be. New plates are never added to the oral extremity
of the corona proper.

The surface of the plates of the corona in the JEchlmdea
is covered with minute rounded elevations, or tubercles^ to
which are articulated the spines so characteristic of the
group. The tubercle may be either simple or marked by a
central pit, into which, and a corresponding pit on the head
of the spine, a ligament of attachment is inserted. Further-
more, capsular muscular fibres connect the neck of the spine
with the base of the tubercle, and effect the varied move-
ments of which the organ is capable. The spines of the
Echiyiidea vary very much in form and size, from the close-
set, velvety pile of Scutella, or the delicate, spoon-shaped
blades of Amphidotus^ to the long-pointed lances of Echinus
and the great clubs of Cidaris. Even on the same Echino-
derm the spines may, as in the two latter genera, vary very
much in appearance ; and it becomes necessary to distinguish
those large ones which form a continuous series from one end
of an ambulacrum or interambulacrum to the other, as pri-
mary spines, from the other less complete secondary and
tertiary series.

Loven^ has drawn attention to the existence, in all Ihe
Echinidea, except Cidaris^ of certain minute spheroidal
bodies, rarely more than yj^ of an inch long, which he terms
sphoiridea. They occur upon the ambulacral plates, and es-
pecially upon those nearest the mouth. Each contains a cal-
careous and more or less dense and glassy skeleton, which
is articulated with a corresponding tubercle, as if it were a
miniature spine. In some genera, these sphcEridea^ to which
Loven ascribes a sensory function (probably auditory), are
sunk in fossne of the plate to which they are attached.

Scattered among their spines, the Echinidea -possess 2yedi-
cellarice, which are usually provided with long, slender stems,
terminating in oval heads, divided into three jaw-like pro-
cesses. The latter are strengthened by calcareous ossicles,
which articulate with an ossicle contained in the basal part
of the head, and a calcareous rod is usually developed in the
stem,

» " Etudes sur les Echinoid^es," 1875.

THE ECHINIDEA.

491

In the Spatangoida^ when the skeleton is cleaned, its sur-
face is, in many cases {Amphidotus, ^rissus, Spatangus),
marked by one or more symmetrical bands of close-set, mi-
nute tubercles (Fig. 143, e,f, g). During life, slender spines
are attached to these tubercles, the calcareous skeleton of

Fig. liZ.—Amphidotus cordatus.—K, viewed from above ; B, from bebind : a, 5 5,
trivium, or anterior and anterolateral ambulacra ; cc, bivinm. or posterolateral am-
bulacra ; d, madreporic tubercle, surrounded by the genital apertures ; e, intra-
petajous semita : /.circumanal semita ; (/, suhanfil semita ; ^. anup ; i, intra-
seniital pores of bivial ambulacra. C, semita magnified : a, semital tubercles ; 5,
ordinary tubercles. D, semital spine: a, terminal enlarged, non-ciliated portion;
5, ciliated stem.

which is clothed with a thick coat of integument, which sud-
denly enlarges at the apex (Fig. 143, D) ; long and close-set

492 THE AXATOMY OF IXYERTEBRATED AXIMALS.

cilia cover tlie shaft of the spine, while no such structures
exist on the terminal enlargement. These bands of pecu-
liarly-modified spines are called semitoe or fascioles, /Setnitce
lie beneath and surround the anus in some genera, and are
called subanal and circumanal ^' others surround the outer
extremities of the petaloid ambulacra, and are termed peri-
petalous, or, when they encircle the inner terminations of
their ambulacra, intra2)etalous (Amjyhidotus) (Fig. 143, A, B).

If we turn to the interior of the shell of the jEchini-
dea, we find in the Echinoida that ambulacral, or sometimes
( Cidaris) interarabulacral, plates of the oral margin of the
corona are produced into five perpendicular perforated pro-
cesses, which arch over the ambulacra, and are called the au-
rlcidce.

Besides these, processes are developed from the ambula-
cral plates in Cidaris which form a sort of wall on each side
of the ambulacral canal, but do not arch over it. In Clypeas-
ter, similar processes form complete arches ; and in the flat-
tened Clypeastroid Scutella, the oral and apical walls of the
corona are united together by calcareous trabeculae, so that
the cavity of the body is restricted to a very small space.

The Spatangoida present neither Auricidoe nor other in-
ternal processes.

In the Echinidea., the oesophagus is usually distinct, but,
beyond a c^ecal diverticulum in some cases, there is no further
differentiation of the alimentary canal, which is disposed spi-
rally around the walls of the corona, and attached thereto by
a mesenterv.

In the Echinidea., the oral skeleton attains its highest
development in the so-called "Aristotle's lantern" of the
Sea-urchins (Fig. 144, B, C, D).

This apparatus consists of five hollow, wedge-shaped, cal-
careous pieces — the alveoli (Fig. 144, B, a) — each of which is
composed of two halves united together in the middle line,
while each half again consists of a superior epip)iysis, and an
inferior principal portion, united together. Each alveolus
serves as the socket for a long tooth (e), shaped somewhat
like the incisor of a Rodent, harder externally than internally,
so as always to develop a sharp edge with wear. The tooth
constantly grows from its upper extremity, while its lower
half becomes united with the wall of the alveolus. The five
alveoli, if fitted together, form a cone, the applied surfaces of
which are united by strong transverse muscular fibres, while
superiorly, the epiphyses of each pair of alveoli are connected

THE ECHINIDEA.

493

by long radial pieces — the rotulce (c) articulated with their
edges. To the inner extremity of each rotula, finally, a slen-
der arcuated rod, presenting indications of a division in the

Fig 144.— a, dentary apparatus of Clypeastrid (after Mullet) : a, alveolns ; d, rofula ;
e, tooth. B, C, D, dentary apparatus (Aristotle's lantern) of Echinus sphoera.
B, two of the five chief component parts of the lantern apposed and viewed lat-
enillv. C, lateral view, and D, back view, of a single part: a, principal piece of
alveolns; a', suture with its fellow; 6, epiphysis ; j' suture of epiphysis with
principal piece ; c, rotula ; d, radius or compass ; e, tooth.

middle of its length, is articulated, and, running outward par-
allel with the rotula, terminates in a free bifurcated extremity.
This is the radius (d).

Altogether, then, the Lantern consists of twenty principal
pieces — five teeth, five alveoli, five rotulge, and five radii — of
which the alveoli are again divisible into four pieces each,

494 THE ANATOMY OF INVERTEBRATED ANIMALS.

and the radii into two, making a total of forty pieces. In
their normal position, it must be remembered that the alveoli
and teeth are interambulacral, while the radii and rotulae are
ambulacral. Besides the interalveolar muscles already de-
scribed, this complex apparatus has protractor muscles arising
from the interambulacral region of the oral edge of the coro-
na, and inserted into the upper part of the alveoli ; slender
oblique muscles, with a similar origin, but inserted into the
radii; transverse muscles connecting the radii together ; and
retractor muscles arising from the arches of the auriculae, and
inserted into the oral ends of the alveoli.

A similar but less complex oral skeleton exists in most
Clypeastroida (Fig. 144, A), but nothing of the kind has yet
been discovered in the Spatcuigoicla.

In the Echinidea^ the circular ambulacral vessel lies be-
tween the oesophagus and the alveoli, and is usually provided
with five sacculated polian vesicles. There is a single madre-
poric canal, membranous in Eihinus^ but calcareous in Cida-
ris, which extends nearly in the axis of the body from the
circular vessel to the madreporic tubercle. Five radial ves-
sels run up the middle of the inner surface of the ambulacral
plates, which they reach by passing from the circular canal,
outward, beneath the rotulae, when these exist ; next, down-
ward, external to the interalveolar muscles ; and then, out-
ward, through the arches of the auriculae ; these give off
branches on each side to the pedicels, the bases of which open
into large ambulacral vesicles. The circular ambulacral ves-
sel of the SpatcjDigoida has no polian vesicles, and no vesic-
ular appendages ; in the Clypeasters there are many vesicu-
lar appendages, but no polian vesicles. In most Echiiioida^
all the pedicels are expanded into sucking-disks at their ex-
tremities, and are here strengthened by a calcareous plate
or plates; but, in Eohinocidaris and some other Echinoida^
the pedicels of the oral portion of the ambulacra only have
this structure, while those of the apical portion are pecti-
nated, flattened, and gill-like. Again, in the heterogeneous
ambulacra of the Clypeastroida and Spatangoida, the forms
of the pedicels vary much. Thus Miiller distinguishes four
kinds of pedicels in the Spatangoida: simple and locomo-
tive pedicels, without any sucking-disk ; locomotive pedicels,
provided with terminal suckers, and containing a skeleton ;
tactile pedicels, with papillose expanded extremities ; and
gill-like pedicels, triangular, flattened, more or less pecti-
nated lamellre. Two or three of these kinds of feet may

THE ECniNIDEA. 495

occur in any given ambulacrum, and those which lie within a
semita are always different from the others.

In the Clypeastroida^ the petaloid portions of the ambu-
lacra possess branchial pedicels, interspersed with delicate
locomotive pedicels, provided with a calcareous skeleton and
with a terminal sucker. The latter kind alone extend on to
the oral portions of the ambulacra.

The circumoral nerve of Echinus surrounds the oesoph-
agus near the mouth. It has a pentagonal form, and is
inclosed by the alveoli, between which the ambulacral nerves
pass, over the peristome and through the arches of the au-
ricula, to the ambulacra. Each ambulacral nerve is accom-
panied by a neural canal, which, however, insheathes the
nerve, and does not merely lie on its inner side.^

Tiie only known organs of sense in the Echinidea are the
pigmented *' eye-spots," developed in connection with the
ends of the ambulacral nerves.

The peritoneal space is filled by a corpusculated fluid,
which is kept constantly in motion by cilia distributed over
the parietes and the contained viscera. The aeration of this
fluid appears to be faciUtated in all the Echinoida^ except
Cidarls^ by live pairs of special branchial plumes developed
from the peristome; while, in the Clypeastroida and Spa-
tangoida^ which possess the modified pedicels commonly
termed ambulacral gills, there are no such organs.

In the Echinidea^ a circular pseud-h?emal vessel, whence
branches are given off to the genitalia, is said to surround
the anus. Tlie alimentary canal is accompanied by two ves-
sels, one on the side of the mesentery (dorsal), the other on
the free side (ventral), which communicate with a lacunar
network in its walls ; and besides these, a fusiform body run-
ning parallel with the madreporic canal, and terminating
inferiorly in a circular vessel which lies close to the circular
ambulacral vessel, around the oesophagus, has been described
as a " heart." '

The genital organs are sacculated glands, which attain a
large size in the breeding season, and open externally by the

» Teuscher, I. c.

2 According to Hoffmann's latest investi^tions, there is neither anal nor
cesophacreal circular vessel in Spatangus and Echinus. In the fonner, a distinct
anastomotic trunk connects the intestinal vessels with the circular ambulacral
vessel. In the latter, both intestinal vessels open directly into the circular
ambulacral vessel, and what has been described as a heart is really the madre-
poric canal. (" IJeber das Blutgefdss-Systcm der Echiniden," "Niederliin-
disches Archiv," Bd. i.)

496 THE ANATOilY OF LWERTEBRATED ANIMALS.

pores on the genital plates, through which their products are
extruded. Hoffmann has fcund the peritoneal fluid of the
raales full of spermatozoa.

Pig. 145.— Development of an Echinid. (After MuHer.)— A, Echinopjedium of Kchi-
nil s pulchellus in the gastrula ptaye. B, fully-developed Echinopaedium {Pluteiis)
of tlie same species: a, mouth; &,' stomach and intestine; c.anus; ^jp^, proceseegi
of the body into which prolonj^ations of the internal skeleton extend. C, the
Echinopjedium of an Echinid in which the Echinoderm is so far advanced that
the spines, pedicels, and pedicellariae are visible. D, Echinopaedium of Echinus
lividus: a, mouth ; a', pullet; b. stomach; 6', intestine; c, radimentary Echino-
derm; c', the ambulacral sac; c'\ the external opening of its duct; A A, FE, B,
the processes of the body.

In the JEchiiiidea, as in the Ophiuridea^ the Echinopje-

THE CRINOIDEA. 497

dium is a Pluteus, and has a skeleton formed of calcareous
rods, which support the processes into which the body, in the
region of the ciliated bands and elsewhere, is prolonged.

The origin of the ambulacral system, before it has the
form of a caecum with a dorsal pore, has not been made out.
The blind end of this caecum lies on the left side of the ali-
mentary canal, and is connected with a discoidal body, which
is situated on the left side of the stomach ; a similar body ap-
pears on the right side. Doubtless these discoidal bodies an-
swer to the peritoneal diverticula of the alimentary canal of
the Echinopgedium in other Echinoderms.

The blind end of the tube enlarges, and gives rise to a
rosette, whence the ambulacral vessels proceed ; and a de-
pression of the integument of the larva, forming the so-called
umbo, extends inward to this. At the bottom of the umbo,
a new mouth opens through the centre of the rosette into the
gastric cavity of the larva, the primitive oesophagus being
abolished. Tiie larval skeleton undergoes resorption, but the
rest of the Ejhinopagdium passes into the Echinoderm.^

Loven has recently drawn attention to the fact that, in
young Echinids,'^ the plates of the apical region are not only
more conspicuous in relation to the corona, but differ some-
what in their arrangement from those of the adult. Thus
the anus is at first wanting, and the anal plate, which occu-
pies the centre of the apical area, is relatively large ; it is
united by its edges with the five plates, which, imperforate in
the young, will become the genital plates in the adult. The
five ocular plates are also imperforate, and are disposed in a
circle outside that formed by the genital plates, their inter-
spaces being occupied by interambulacral plates. The apical
region of an Echinid has thus, as Loven points out, a most
striking resemblance to the calyx of a Crinoid ; the anal
plate representing the basalia, the genital plates the para-
basalia, and the ocular plates the first radialia.

The CnrN'OiDEA. — This remarkable group, which abounded
in former periods of the world's history, is represented at the

J See, in addition to the memoirs of Miiller and Metschnikoff already cited,
A. Asrassiz, " On the Embryology of EcMnoderms." (" Mem. American Acad-
emy of Sciences," 1864.)

3 The admirable monograph of A. Asrassiz, " Revision of the Echini," pub-
lished in the '' Illustrated Catalogfue of the Museum of Comparative Zoology at
Harvard College," is also full of information respecting the young states of the
Echinids.

498 THE ANATOMY OF INYERTEBRATED ANIMALS.

present day only by the genera Antedon (Co7natula), Acii-
nometra, Comaster^ Pentacrinxis^ Hhizocriiius^ and Holopus,

The first three genera are capable of locomotion, while
the next two are attached by long articulated stems to sub-
marine bodies. Molopus^ which is but imperfectly known,
appears to be fixed by a short, thick, unjointed prolongation
of its base.

FJiizocrimis lofotensis (Fig. 146), which has been very
carefully and elaborately described by Sars,^ is a small animal
which does not attain more than three inches in length, and
lives at great depths (100-300 fathoms or more) in the sea.
It consists of a relatively long, many-jointed stem, from many
of the articulations of which, branched, root-like filaments,
or c^Vr^, are given off ; at the summit of this is seated a cup-
shaped body, the calyx, from the margins of which five to seven
arms {hrachia) radiate. To each arm is attached a double
series of alternating pinnulce. The mouth is situated in the
centre of that part of the perisoma which forms the surface
of the calyx opposite to the stem. The oral aperture is cir-
cular, but five (or sometimes only four) triangular lobes of
the perisoma, with rounded free ends, project over it, and,
when shut, close it like so many valves. From the inter-
vals between these oral valves five (rarely four) grooves trav-
erse the oral surface of the calyx, and extend thence
throughout the whole length of each arm, giving offsets as
they go to the pinnules. Thus the oral surface of each arm
and of each pinnule is deeply excavated.

Between the circular lip and the oral valves, soft flexible
tentaculiform pedicels are attached in a single series. Two
pairs of pedicels correspond to every valve, each pair aris-
ing opposite the basal angle of a valve. These pedicels are
hollow, their surface is papillose, and the outer or radial pedi-
cel of each pair is very contractile. Pedicels of the same
general character are continued throughout the brachial and
pinnular grooves.

The anus is situated at the end of a conical prominence
between two of the grooves on the oral face of the calyx, and
is therefore interracUal in position (Fig. 146, III. an).

The skeleton consists of very numerous pieces resulting
from the calcification of the perisoma. In the stem they have
the form of elongated, subcylindrical, or hour-glass-shaped,
joints (articulf), the opposed faces of which are united by

J " Memoires pour servir h la connaissance (lea Crinoides vivants," 1868.

THE CRINOIDEA.

499

strong elastic ligamentous fibres. The centre of each is
traversed by a longitudinal axial canal, which extends
through the whole length of the stem and is occupied by a

n .9„/^ ^\^

Fig. lis.— Bhizoc?'invs lofotetifis. (After Sars.)

I. Hhizocrimis eniire: a, enlarged upper joint of the stem; 5, larval joints of the
stem ; c, cirri ; d, bracbia.

II. Calyx and arras, with tbe summit of the stem of a Ehizocrinns having five well-
developed brachia : a, as before ; s, first radials ; r-, r^, second and third radials ;
Z/, first brachial ; J9, p, pinnules.

in. Upper part of the stem and oral face of the calyx, viewed obliquely: i\ lower
part of visceral mass; 5i{, tentacular grooves; o, oral valves ; i, oral tentacles;
auy anus.

soft but solid substance. The distal joint of the stem is not
directly fixed to the surface to which the Crinoid is attached,
but is connected therewith by the branched cirri which pro-
ceed from it. Each cirrus has a skeleton composed of joints
or articuli, somewhat like those of the stem, and traversed by
a prolongation of the axial canal. Similar cirri are developed
from a larger or smaller number of the articuli of the distal
portion of the stem.

The proximal joints become gradually shorter in propor-
tion to their length, until they assume a discoidal form. It
appears that new articuli are continually added at that end of
the stem which lies nearest the calyx.

The summit of the stem, or the base of the calyx, is
formed by an enlarged, solid, pear-shaped ossicle, which is
probably formed by the coalescence of several articuli. Upon

500 THE ANATOMY OF INVERTEBRATED ANIMALS.

this follow five pieces (first radlalia) closely united together
and with a central piece, which probably represents the basa-
lia of other Crinoids. The first radial corresponds in direc-
tion with the origin of one of the arms, and is followed by a
second and third radial. With the third radial is articulated
the first of the brachial ossicles, which constitute the skele-
tal support of the unbranched brachia. The pinnules are also
supported by a series of elongated calcified joints, the basal
joint being articulated with a brachial ossicle and the distal
joint pointed.

The axial canal dilates in the enlarged pyriform ossicle
above mentioned ; and, from the dilatation, branches, which
traverse the radial and the pinnular ossicles, are given ofl^.
There is a calcareous plate in the substance of each oral valve,
and minute reticulated calcifications are scattered throusrh the
perisoma of the oral face of the disk.

The sides of the radial grooves are provided throughout
with a double series of oval calcareous plates — the marginal
lamelloe — which are disposed transversely to the groove, those
of opposite sides alternating with one another. They can be
erected or depressed ; and, in the latter case, overlap one an-
oiher like tiles.

In Pentacrinus^ the long stem is fixed by its distal end,
and the pentagonal ariicull of its skeleton give off", at inter-
vals, whorls of unbranched cirri. No distinct basal piece is
known, but the calyx appears to begin with the five first radi-
alia. At the third radiale^ the series bifurcates into two
series of brachialia, and these again bifurcate to give rise to
the palmaria^ which support the free arms. There are mar-
ginal lamellae along the sides of the tentacular grooves, and a
longitudinal series of calcareous ossicles occupies the floor of
each groove. The anus is situated upon an elevated inter-
radial cone.

The body of an adult Comatula [Aiitedoii) answers to the
calyx, with its brachia, in other Crinoids.

The centre of the skeleton is constituted by a large centro-
dorsal ossicle, articulated with the aboral face of which are
the numerous cirri, by which the ji?itedon ordinarily grasps
the bodies to which it adheres, though it is able, on occasion,
to swim freely about. This centro-dorsal ossicle appears to
be the homologue of the uppermost part of the stem in the
Pentacriniis. There are five divergent series of radialia^
each containing three ossicles. The first radials, or those
nearest the centro-dorsal plate, are closely adherent to one

THE CRIXOIDEA. 501

another and to the centro-dorsal plate, and are not visible on
the outer surface of the calyx. The space left between the
apices of the five first radials is occupied by a single plate,
the rosette^ which is formed by the coalescence of the five
hasalia present in the larva.

The anatomy of the soft parts of the Crinoidea has been
most thoroughly investigated in the genus Comatula [An-
tedon),^

The mouth leads, by a short, wide gullet, into a spacious
sacculated alimentary canal, which is coiled upon itself in
such a manner as to make about one turn and a half around
the axis of the body, and then terminates in the projecting
rectal cone, which, as has already been seen, is situated inter-
radially on the oral face of the calyx. Tlie central cavit}'^,
included by the coil of the alimentary canal, is occupied by a
sort of core of connective tissue, and has received the name
of columella, but it must be understood that it is not a dis-
tinct structure. Bands of connective tissue connect the outer
periphery of the alimentary canal with the perisoma.

The five triangular lobes of the perisoma, which surround
the mouth like so many valves, contain no calcareous skele-
ton in the adult Antedon. Within these lobes, attached to
the oral membrane, there is a circle of tentacula. From the
interval between each pair of oral valves, a groove radiates
outward over the surface of the calycine perisoma and speed-
ily bifurcates ; one branch goes to the oral surface of each of
the arms, and runs along it to its extremity, giving off alter-
nate lateral branches to the pinnules in its course.

These grooves are the ambulacral grooves. Their sides
are, as it were, fenced by small, lobed processes of the peri-
soma ; and, on the inner sides of these processes, groups of
minute pedicels take their origin from the sides of the floor
of the groove. A thickened band of the ectoderm occupies
the middle of the floor, and so strikingly resembles the ambu-
lacral nerve of the Star-fish that the homology of the two,

1 Carpenter, " On the Structure, Plivslologv, and Development of Comatula.''^
("Phil. Trans.," 1866.)

2 E. Perrier, '' Recherches sur I'Anatomie de la Comatula rosacea'''' ("Arch,
de Zoologie Experimentale," 1873). Semper, " Kurze anatomische Bemerkun-
een \\\)Qr Comatula''^ (" Wiirzburg Arbeiten," 1874). Ludwisr, " Zur Anatomic
der Crinoideen" {ZeitscJirift fi'ir wiss. Zool.^ 1876). Carpenter, *'0n the Struct-
ure, Physiology, and Development of Antedon" ("Proc. Royal Society,"
1876). 6reef, '" Ueber den Bau der Crinoideen " (" Marburg Sitzungsberichte,"
1876). P. H. Carpenter, " Remarks on the Anatomy of the Arms^of the Cri-
uoids" {Journal of Anat, and Physiology^ 1876).

502 THE ANATOMY OF INYERTEBRATED AXIMALS.

first asserted by Ludwig,^ cannot be doubted. Immediately
beneath it runs a small canal, discovered by Dr. Carpenter,
and termed by him the tentacular canal, which gives oS lat-
eral branches to communicate with the cavities of the pedi-
cels. A second much wider canal — the suhtentacular canal —
lies beneath this, and is divided by a longitudinal septum.
But the septum is incomplete at intervals, and thus the two
canals communicate. A third, still larger — coeliac canal — is
interposed between the floor of the subtentacular canal and
the axial skeleton of the arm.

Where the arm joins the calyx the tentacular canals run
beneath the ambulacral groove to the gullet, around which
they are united by a circular canal, from which numerous
short diverticula, resembling the vasa amhidacralia cavi in
the Ophiurids, described by Simrock (I. c), depend. The
subtentacular and coeliac canals communicate with channels
in the perivisceral tissue, on the oral or the aboral face of the
visceral mass ; and these channels appear, eventually, to
open freely into the cavities by which the columella is trav-
ersed.

In the partition between the subtentacular and the coeliac
canals there lies a cellular cord, or rachis, which can be traced
back into a reticulation of similar tissue in the visceral mass.
The genital glands contained in the pinnules are enlargements
of lateral branches of this rachis. But the rachis is appar-
ently only an extension of the mesodermal tissue of the vis-
ceral mass, comparable to that in which the genitalia are
lodged in the Star-fishes ; and the multiplication of the geni-
tal glands may be regarded as a further extension of the
structure which obtains in Brisinga. Thus it would seem
that the position of the genital glands in the Crinoids is not
so anomalous as it at first appears to be.

The centro-dorsal tubercle contains a cavity with which
the canals which traverse the ossicula of the cirri, the calyx,
the brachia, and the pinnules communicate. This cavity was
considered by Miiller to be a heart. It proves, however, to
be largely filled by solid tissue, which is continued not only
into all the canals which traverse the ossicula, but also into
the columella, or tissue which occupies the centre of the coils
of the alimentary canal.

Dr. Carpenter ^ is of opinion that so much of this axial

1 ZeifseTirift fur wiss. Zooloqie, 1876.

3 '' Proceedings of the Eoyal Society," 1876.

THE DEVELOPMENT OF THE CRINOIDEA. 503

tissue as occupies the cavity of the central tubercle, and is
continued throughout the ossicula of the calyx and arms, is
the proper central organ of the nervous system; founding
this opinion partly upon the fact that, when this mass is irri-
tated in a living Antedon, a sudden contraction of all the
muscles of the arms takes place, and partly upon the distri-
bution of the ultimate ramiHcations of the axial tissue in the
arms. Greef, on the contrary,^ affirms that all these tracts can
be injected, and retains the name of " heart " for the cavity
of the centro-dorsal tubercle.

The perisoma of the oral surface of Comatula exhibits a
great number of minute circular pores, with thickened cellu-
lar margins. Greef has discovered that these are the external
apertures of canals, with ciliated walls, which open into the
body-cavity, and readily allow fluids to pass into, or out of,
that cavitv.

Each mature ovary of Ajitedon has a distinct aperture,
through which the ova are discharged, and to which they ad-
here for some days like bunches of grapes. The testis devel-
ops no special aperture, but the spermatozoa appear to be
discharged by dehiscence of the integument.

Since the discovery by Vaughan Thompson that Comatida
passes through a Pentacrinoid larval condition, the develop-
ment of the free Crinoids has been the subject of various in-
vestigations,'* and the following results may be regarded as
established :

Complete ye'k-division takes place. The morula acquires
an oval form, and develops four hoop-like bands of cilia,
with a tuft of cilia at the hinder end. Between the third and
fourth bands of cilia, counting from the anterior end of the
Echinopasdium, the blastoderm becomes invaginated, and
gives rise to an archenteron. In the interspace between this
blind sac, the wall of which is the hypoblast, and the epiblast,
constituted by the rest of the blastoderm, a mesoblast com-
posed of reticulated cells makes its appearance. The blasto-
pore closes, while the archenteron detaches itself from its
attachment to the posterior ventral face of the larva, and be-
comes connected with an oesophageal involution formed at its
anterior end. Tlie archenteron next throws out three diver-
ticula, of which two are lateral and one is ventral. The lat-

1 " Ueber das Herz der Crinoideen" (" Marburg Satzunersbericbte," 18T6\

2 5^^ Wyx-ille-Thompson ("Phil. Trans.," 1865^, Metschnikotf ("Bulletin
de I'Acad. Imp. des Sciences de St.-Petersbourg," 1871), and especially GOtte
(" Archivfar Mikroskopische Anatomie," 1876j,

504 THE ANATOMY OF INVERTEBRATED ANIMALS.

eral diverticula enlarge, and apply themselves to the rest of
the archenteron, now become the intestine, from which they
are soon completely shut off, and converted into peritoneal
sacs. The left sac thus formed lies on the ventral side of the
intestine, the right sac on its dorsal side. The walls of the
two sacs become applied together, and form a circular mesen-
tery. The peritoneal sac of the aboral side sends a pro-
cess into the hinder end of the body, which has begun to
elongate, in order to give rise to the stem of the Pentacrinoid
form.

The third, or ventral, diverticulum is shut off from the
alimentary canal much later than the other two. It grows
round the mouth, and gives rise to the circular ambulacral
vessel, whence the tentacular canals are given off.

Ten plates, each consisting of a calcareous network, and
arranged in two rows of five each, next appear in the sub-
stance of the Echinopaedium around the alimentary canal.
From the centre of the posterior row, eight calcareous rings
extend through the length of the body of the larva, inclosing
the backward prolongation of the aboral peritoneal sac ; and
the series terminates by a broad, discoidal network, which lies
on one side of the posterior end of the larva. This discoidal
plate is that which occupies the attached end of the stem of
the future Crinoid ; the rings become the stem^ and the two
circles of plates the basal and oral ossicula of the calyx, re-
spectively. As the stem elongates, new rings {^articuli) are
added at the junction of the stem with the calyx.

The larva now fixes itself bv the discoidal end of its stalk,
which becomes relatively longer and narrower ; while the
part of the body which contains the basal and oral plates, and
is to be converted into the calvx, remains thick and short.
Its broad end becomes five-lobed, each lobe answering to an
oral plate. These plates separate like the petals of a flower-
bud, and discover, in the centre, the wide, permanent oral
aperture. Between the margins of this and the oral plates,
tentaculiform pedicels, at first only five, but eventually ar-
ranged in groups of three, between every pair of oral plates,
make their appearance.

The alimentary cavity is still a mere sac, without intestine
or anus.

Five radial plates next appear in the wall of the calyx, be-
tween the basal and the oral plates, and alternating with both ;
and, in correspondence with them, the arms grow out as rap-
idly-elongating processes, in which the other radials are sue-

THE AFFINITIES OF THE ECHINODERMATA. 505

cessively developed. The entire zone of the calyx, which is
occupied by the origins of the arms, at the same time widens,
so that the oral plates, which remain round the mouth, and
the basal plates, which encircle the stem, become widely sep-
arated. The intestine grows out as a diverticulum of the
alimentary cavity, and opens on an interradial elevation of the
calyx, in which an anal plate is developed. The young Echi-
noderm has now passed into the stalked Pentacrinoid stage.

In Comatula^ the oral and anal plates disappear altogether,
and the basals, coalescing into the rosette, are hidden by
the first radials, on the one hand, and the centro-dorsal tuber-
cle, which represents coalesced joints cf the stem, on the
other. The arms bifurcate and acquire their pinnules ; and
the calyx, with its appendages, eventually becomes detached
from its stem as a free Cotnatula. In the existing stalked
Crinoids, such as Pentacrinus^ on the other hand, the seg-
ments of the stem acquire whorls of cirri, at intervals, and no
such modification of the uppermost segments into a centro-
dorsal tubercle takes place.

On comparing the facts of structure and development
which have now been ascertained in the five existing groups
of the Echinodermata,, it is obvious that they are modifications
of one fundamental plan. The segmented vitellus gives rise
to a ciliated morula, and this, by a process of invagination, is
converted into a gastrula, the blastopore of which usually be-
comes the anus. A mouth and gullet are added, as new for-
mations, by invagination of the epiblast. The embryo normally
becomes a free Echinopoedium, which has a complete alimen-
tary canal, and is bilaterally symmetrical. The cilia of its
ectoderm dispose themselves, in one or more bands, which
surround the body ; and, while retaining a bilateral sym-
metry, become variously modified. In the Solothnridea^ As-
teridea, and Crinoidea., the larva is vermiform, and has no
skeleton ; in the Echinidea and the Ophiuridea it becomes
pluteiform, and develops a special spicular skeleton.

If an Echinopaedium were to attain reproductive organs,
and reproduce its kind, I think that it cannot he doubted that
its nearest allies would be found among the Turhellaria^ the
Motif era, the Gephyrea^ and the Enter opneusta.^ But that

1 In a report upon the " Researches of Prof. Miiller into the Anatomy and
Development of the Echinoderms," published in the AniiaU of Natural His-
tory for July, 1851, I drew attention to the affinities of the Echinoderms -yvith
the "Worms; and iu a paper on Lacinularia socialise le&d before the Micro-

22

50G THE ANATOMY OF INVERTEBRATED ANIMALS.

which characterizes the Echinodermata is the fact that the
alimentary canal of the Echinop^edium gives rise to an en-
teroGcele, which again is subdivided into two systems of cav-
ities, one ambulacral and the other peritoneal, and that the
mesoblast becomes modified in accordance with the arrange-
ment of these systems. The enterocoele may be formed by
one diverticulum or by three. In the former case, the first
formed becomes subdivided into three, of which one is ante-
rior, and two lateral, as in the latter case. The lateral di-
verticula give rise to the peritoneal cavity and its lining ;
the medium diverticulum is converted into the circular ambu-
lacral vessel and its dependencies ; and it is in consequence
of the radiating disposition of the latter, and of the nerves
and muscles which are related to it, that the Echinoderm pos-
sesses so much radial symmetry as it displays. It is clear,
therefore, that the radial symmetry of the Echinoderm results
from the secondary modification of an animal, which is primi-
tively bilaterally symmetrical ; and that the apparently radi-
ate Echinus^ or Star-fish, is a specially modified " Worm '*
(using that terra in its widest sense), in the same sense as the
apparently radiate Coronula is a modified Arthropod.

Haeckel goes farther than this, and supposes that each ray
of a Star-fish or Opbiurid, for example, represents a Worm,
and that the Echinoderm consists of coalesced vermiform
buds, developed in the interior of the Echinopaedium. I
must confess my inability to see that this hypothesis is sup-
ported by valid reasons. On the contrary, the more closely
one compares the structure of the ray of an Echinoderm with
the body of any known Annelid, the more difficult does it ap-
pear to me to be to find any real likeness between the two.

In order to find any analogy for the production of the
Echinoderm within the Echinopaedium, on the contrary, it ap-
pears to me that we must look to the lower, and not to the
higher, morphological types. Among the Hydrozoa^ nothing
is commoner than the distribution of the functions of life be-
tween two distinct zooids, one of which alone develops repro-
ductive organs. In the former — the hydranth — radial sym-

scopical Society in the same year, I expressed the view that the Eotifera " are
the permanent forms of Echinoderm larv:^, and hold the same relation to the
Echinoderms that the Hydriform Polypi hold to the Medusae," and that they
"connect the Echinoderms with the'Nematid;ie and the Nematoid Worms."
When they were published, those who did not ignore these views, ridiculed
thena. Nevertheless, though somewhat crudely expressed, I think it will be
admitted that they have been substantially justified by the progress of knowl-
edge during the last Quarter of a century.

THE AFFIXITIES OF THE ECHINODERMATA. 507

metry is often hardly discernible (e.g., CalycophoridcB)\ in the
latter — the medusoid — it is very marked, and especially char-
acterizes the arrangement of the gastro-vascular canals, which
are offshoots of the alimentary cavity, and, if they became
shut off therefrom, would answer to the enterocoele of the
Echinoderm.

Suppose that, from a hydranth such as that of a Dlphyes^
a medusoid were developed, and that, instead of projecting
from the exterior of the body, it remained hypodermic, spread-
ing out between the ectoderm and the endoderm of the hy-
droid, and consequently superinducing a very marked radial
symmetry upon it. The resulting form would give us a
Coelenterate which would be a close analogue of an Echino-
derm.

In a certain sense, an Actinozoon may be fairly regarded
as such a combination of a hydroid with its medusoid ; and,
hence, it must be conceded that the parallel between the gas-
tro-vascular system of the CtenopKora and the ambulacral
system of the Echinoderms, instituted by the elder Agassiz,
was well worthy of consideration. Shut off the gastro-vascu-
lar canals of a Cydippe from the alimentary canal, and they
become an enterocoele, of which the prolongations along the
stomach may be compared with the peritoneal sacs, and those
beneath the paddles with the ambulacral vessels of the Echino-
derm.

But there is a long step between the admission of the force
of these analogies, and the conclusion that the Echinoderms
and the Codenterata are so closely allied as to be properly
associated in one natural assembla2:e of " Radiate " animals.
On the contrary, the Echinoderm, by its Echinopaedium stage,
shows an advance in oro;anization far bevond anvthins: known
in the Codenterata / and in the highly-characteristic mode of
development of its enterocoele (the elucidation of which in
the " Star-fishes," by Prof. A. Agassiz, is the most important
advance in our knowledge of the Echinoderms made since
the time of Miiller), the Echinoderm agrees with the higher,
and not with the lower, Metazoa.

Ecliinodermata abound in the fossil state. Calcareous
plates, referred to the Solothiiridea, occur in the Mesozoic
rocks, but are not known earlier. The Star-fishes are met
with in the older Palaeozoic strata, under forms very similar
to some of those which now exist. The Echinidea abound
from the Upper Silurian [Paloechinus) onward. The Palieo-

508 THE ANATOMY OF IXVERTEBRATED ANIMALS.

zoic forms are spherical, and have multiple interambulacral
plates and simple ambulacra. Echinidea of the modern type
appear in the Mesozoic strata — the Echinoida first, while the
Spatangoida and Clypeastroida are of later date. This order
of occurrence agrees with the embryonic deTelopment of the
two latter groups, which are more nearly spherical when young
than subsequently.

The Crinoidea abound in the Palaeozoic and older Meso-
zoic rocks, graduall}'' diminishing in number in later forma-
tions. The oldest appear to have all been stalked, and of
peculiar and extinct types.

Three groups are wholly extinct, and are unknown in
strata newer than the Carboniferous formation. These are
the Cyst idea, the Edinoasterida, and the Blastoidea,

The Ctstidea. — In their general characters the Cystidea
come very near the Crinoids. Cryptocriniis^ the simplest
form of the group, possesses a calyx supported on a stem, and
composed of five bascdia, five parahasalia, and five radialia.
An interradial aperture is surrounded by a cone of small
plates, termed the pyramid. The antambulacral surface has
no pores, but these were present in other genera, and sometimes
are scattered irregularly ( Caryocrinus) ; sometimes disposed
in pairs (Sphceronites) ; while sometimes they take the form
of parallel slits arranged in " pectinated rhombs." The arms
were free {Comarocystites), or recurved and closely applied
to the calyx. They bore pinnules, which, in consequence of
the non-development of the arms, were sometimes sessile on
the radialia. In the species with recurved arms, the latter
simulate calycine ambulacra. There is an aperture placed in
the centre of the calyx at the point of convergence of the
ambulacra ; another small one on one side of this ; and, third-
ly, the aperture of the pyramid. The first of these is com-
monly regarded as the mouth, the second as the anus, the
third as the reproductive aperture.

The Cystidea would, on this interpretation, differ from all
other Echinodermata, except the Edrioasterida and Solo-
thuridea, in the genital outlet being single ; but around the
central aperture five pores are seen, in some species at least,
to which a genital function has been ascribed. In any case,
the Cystidea would appear to come very close to the Cri-
noidea.

The Edrioasterida. — This group contains several genera

THE BLASTOIDEA. 509

of extinct Echinoderms {Eclrioaster, Agelacrinites, Semicys-
tites), which, in general form, somewhat resemble what the
Asterid Gonmster would be if its angles w^ere rounded ofP.
Like the Cystldea^ they possess an interambulacral^9yra/?ziV^,
but they differ from them in that they have ambulacra per-
forated by canals which open directly into the cavity of the
calyx, and that they possess no arms. The Edrioasterida
have no stem, but seem to have been attached by the abo-
ral face of the body.

The Blastoidea. — In Pentremites^ the representative of
this order, the ambulacral and antambulacral regions are
nearly on an equality : the body is prismatic or subcylin-
drical. Tlie pedunculated calyx is composed of three basal
plates, two of which are double. The aboral plates receive
in their intervals five plates deeply cleft above. In the clefts
lie the apices of the ambulacra, the oral portions of which are
included between the five deltoid interradial pieces which
surround the mouth. The cleft plates are not radials, but
portions of the perisomatic skeleton of the aboral region.
Surrounding the central, probably oral, aperture, are four
double pores, and a fifth divided into three. The median of
these three seems to be anal, the others and the paired pores
being genital. Each ambulacrum is lanceolate in form, and
presents superficially a double row of ossicles, which meet in
the middle line and support pinnules at their outer extremi-
ties ; beneath them lies a single plate, perhaps the homologue
of the vertebral ossicles in the Ophiurldea / beneath it again
are parallel canals, the nature of which is unknown.

CHAPTER X.

THE TUNICATA OJR ASCIDIOIDA.

This remarkable and, in many respects, isolated group of
marine animals contains both simple and composite, tixed and
free, organisms. None attain a length of more than a few
inches, and some are minute and almost microscopic.

The simplest members of the group, and those the struct-
ure of which is most readily comprehensible, are the Appen-
dicularice y minute pelagic organisms, which are found in all
latitudes, and are propelled, like tadpoles, by the flapping of
a long caudal appendage at the surface of the sea.

Appendicularia flabellum (Fig. 147) has an ovoid or flask-
shaped body (^), one-sixth to one-fourth of an inch in length.
The appendage (B) is from three to four times as long as the
bodv, to one face of which it is attaclied near, but not at, the

%j J 7 7

posterior extremity. It is flattened, and is supported by a
firm central axis, which may be termed the iirochord (Fig.
147, I). The greater part of the bod\' is usually invested by
a structureless gelatinous substance, but, on its rounded
hinder extremity, this ceases to be distinguishable from the
ectoderm.

On the caudal appendage the polygonal contours of the
cells of which the ectoderm is composed are plainly discern-
ible.

The mouth has an overlianging lip. It leads into a large
pharyngeal sac, the walls of which are formed by the endo-
derm. Posteriorly this sac narrows into the oesophagus,
which bends toward the haemal side of the body, and then
opens into a spacious stomach, w^hich takes a transverse direc-
tion, and is divided into two lobes, a right and a left.

From the left lobe the intestine arises, and, bending in-
ward, turns abruptly forward in the middle line, where it
terminates midway between the oral aperture and the attach-
ment of the caudal appendage. The intestine, therefore, has

APPENDICULARIA FLABELLUM.

611

a haemal flexure. In the middle of its haemal aspect the en-
doderm of the pharyngeal cavity is raised into a fold, which
projects into the blood-cavity contained between the endo-

PiG. Ml.— Appendicular ia flahdlvm.

I. The entire animal, wiih the caudal appendage in its ordinarj' position, or turned
forward.

II. Side view of the body, with the caudal appendage forcibly bent backward.

.4, the body; 5, tlie caudal appendage; a, oral aperture; b, the pharynx; <:, an
atrial opening; cf, the corresponding stigma, with its cilia ; e, anus; /, rectum ;
g, oesophagus; A, ?, stomach ; k, testis; ^, nrochord ; m, cellular patch at the
side (»f the oral end of the body; n, endostyle ; /?, ganglion ; q, ciliated sac; r,
otocyst ; s, posterior nerve with its ganglia, i; en, endoderm ; ec, ectoderm.

derm and ectoderm. The walls of the bottom of the fold are
thicker than the rest, so that, viewed sideways, it has the
aspect of a hollow cylinder. This is the endostyle.^ (Fig.
147, n.)

1 So described and named in my " Observations upon the Anatomy and
Physiology of Salpa and Pyrosoma*, together with Eeraarks upon Doliolmn
and Appendicularia." ("Phil. Trans.," 1851.) In 1856, however, I stated:
'' "With regard to the endostyle, I have nothing important to add to ray pre-
vious account, except that 1 believe it to be here, as in other Ascidians, the
optical expression of the thickened bottom of a fold or groove of the branchial
sac." ( Quarterly Journal of Microscopical Science, April, 1856.) In my memoir
on Pyrosoma ("Linn. Trans.," 1860, p. 205), the endostyle is stated to be "in

512 THE ANATOMY OF INVERTEBRATED ANIMALS.

The endoderm of the pharynx is ciliated, and the cilia are
especially large over a narrow tract, or peripharyngeal hand^
which encircles the oral aperture at the level of the anterior
end of the endostyle, and is continued back, as a hypopha-
ryngeal hand^ along the middle of the neural face of the
pharynx to the oesophageal opening.

On each side of the endostyle, the posterior part of the
haemal wall of the pharynx presents two oval apertures or
stigmata (Fig. 147, <^), encircled by cells, which are provided
with very long and active cilia. Each stigma leads into a
funnel-shaped atrial canal, the open end of which terminates
beside the rectum.* (Fig. 147, c.)

The heart is a large sac, which exhibits rapid peristaltic
contractions, and is placed transversely between the two
lobes of the stomach. In the species which I observed no
blood-corpuscles could be seen, and the direction of the pul-
sations of the heart was not reversed at intervals, as it is in
the Ascidians in general. M. Fol,'' however, states that, in
other Appendicidarice, the reversal of the contractions of the
heart takes place. Like myself, he has been unable to dis-
cover any blood-corpuscles. There are no distinct vessels,
but the colorless fluid which takes the place of blood makes
its way through the interspaces between the ectoderm and
endoderm and the various viscera.

The nervous system consists of a ganglion (Fig. 147, p)
situated nearly opposite the anterior end of the endostyle ;
in front, this gives off the nerves to the sides of the mouth,
while, behind, it is continued into a long cord (s), which runs
back beside the oesophagus, and between the lobes of the
stomach, to the base of the appendage. It then passes along
one side of the urochord to its extremity, giving off nerves
at intervals. At the origins of these nerves aggregations of
ganglionic cells are situated. (Fig. 147, t.) The most an-
terior of these ganglia is the largest.^

reality a longitudinal fold or diverticulum of the middle of the hoeraal wall of
the pharynx, which projects as a vertical ridge into the hromal sinus, but re-
mains in free communication with the pharynx by a cleft upon its neural side."

1 These stigmata were first described by Gegenbaur ("Bemerkungen uber
die Organisation der Appendicularien," Zeitschrift fur wiss. Zoologie^ 1855),
who supposed that they communicated with canals olPthe interior of' the bodv.
However, by feedin.g Appendimlarice with indigo, I demonstrated the commu-
nication of these stigmatic funnels with the exterior of the body. {Quarterly
Journal of Microscopical Science, I. c.)

2 " Etudes sur les Appendiculaires," 1872,

3 Quarterly Journal of Microscopical Science, 1856, pp. 8, 9. M. Fol, who
j&nds the same arrangement in other Appendicularict, counts this as the second
ganglion of the nervous system, and states that a fine canal traverses both the
ganglia and the longitudinal nerve.

APPENDICULARIA FLABELLUM. 513

A rounded octocyst containing a spherical otolith is at-
tached to the ganglion, and a small ciliated sac, which opens
into the pharynx, is in close relation with it (Fig. 147, r, q),
M. Fol describes a number of fine tactile set^e situated around
the oral aperture.

The urochord, which constitutes the axial skeleton of the
appendage, is transparent, rounded at each end, and bounded
by a delicate membrane. The remains of the cells of which
it is composed are to be seen in it, here and there, as ramified
corpuscles lodged in its periphery.

The only muscles hitherto observed in Apj^endlcularia
are two sheets of striped fibres interposed between the uro-
chord and the cellular ectoderm of the appendage.

The reproductive organs occupy the rounded projection
formed by the posterior part of the body behind the digestive
canal. The testis (Fig. 147, 1c) is a large cellular mass which
fi.lls the greater part of the cavity of this projection in the
adult. When fully formed, it is resolved into spermatozoa
with rod-like heads about yoVo" ^^ ^^ inch long and very fine
filiform tails. They escape by the dehiscence of the testis.

I have never met with Ajjpeiidicularice containing ova,
nor do any other observers, except M. Fol, appear to have
been more fortunate. The latter, however, states that these
animals are hermaphrodite (Oiko^yleicra dioica apparently is
dioecious), and that the ovary is developed later than the
testis.*

Two singular rounded patches of a cellular structure (Fig.
147, II. m) are interposed between the ectoderm and the en-
doderm on each side of the anterior end of the endostyle.
Similar bodies occur in other Ascidians, but their function is
unknown.

One of the strangest peculiarities of the Appe7idicularice
is the power which they possess of excreting from the surface
of the ectoderm, with extreme rapidity, a mucilaginous cu-
ticular investment, in the interior of which, as in a spacious
case, the whole body is lodged. This is what was originally
described by Mertens as the " house " of the Appendicularia.

^ T must confess that M. Fol's figures and deseriptions of the ovary and ova
are not satisfactory to me, and his dismissal of the subjectof their development
in the following^ paragraph is tantalizing :

" Le developpement, que j'ai pu suivre jusqu'a la formation de la larve, ne
me parut differer en rien de celui des Ascidies ; et comme d' autre part la peti-
tesse de ces oeufs et la difficult^ qu'on a de les obtenir les rendent peu favo-
rables a I'^tude, je n'ai pas jug^ a propos d'approfondir davantage ce sujet."
(I. c, p. 1.)

514 THE ANATOMY OF INVERTEBRATED ANIMALS.

It is obviously the homologue of the test of other Ascidians,
which is often adherent to the ectoderm by only two or three
points ; but no cellulose has been discovered in it. Accord-
ing to M. Fol, who has studied the formation of the "house"
with great care, the Appendicidarim have no proper test, and
what I have described as the structureless gelatinous invest-
ment of the anterior part of the body is the commencement
of the " house." It increases, assumes a peculiar fibrous
structure, and in the course of an hour, in a vigorous animal,
it is separated as an envelope in which the whole body is
capable of free movement. In front, it presents two funnel-
shaped apertures supported by a fibrous trellis-work, which
lead down to the cavity in which the body is contained. A
spacious median chamber allows of the free motion of the tail.
After a few hours the animal deserts its test and forms an-
other.

In the great majority of those Tiinicata which are fixed
in the adult state, the young leave the egg in an active lar-
val condition, and resemble Appendicidaria in being pro-
pelled by a muscular appendage in the axis of which lies an
urochord. The body and appendage, however, are invested by
a coat, or test, impregnated with cellulose, and the former
presents some important structural differences from that of
Appendicidaria. After a free existence of a certain dura-
tion, the body of the larva fixes itself, the appendage withers
awa^^, and the young animal assumes the ordinary form of a
fixed Ascidian. It may remain simple, or it may develop
buds and give rise to a compound organism or Ascidiariurn,
consisting of many Ascidiozooids united together.

All the fixed Tunicates present two, more or less closely
approximated, apertures : one, oral, leads into the alimentary
cavity ; the other, atrial, opens into a chamber, the atriicm^
into which the faeces and genital products are poured. During
life, when these apertures are open, a current sets into the
oral and out of the atrial opening. But if the animal is irri-
tated, the sudden contraction of the muscular walls of its
body causes the water contained in the brachial and atrial
cavities to squirt out in two jets, while both apertures are
speedily closed.

The apertures are much farther apart in some forms than
in others, and in certain of the BotryllidcB they are almost
terminal. In the pelagic genera Pyrosoma (Fig. 150), Dolio-
lum (Fig. 151), and Salpa (Fig. 152), the atrial and oral aper-

THE TUNICATA.

515

tures are at opposite ends of the longest diameter of the
body ; and, in the two latter, locomotion is effected by the
contraction of transverse muscular bands, which drives the

Fig. li8.—PhaUuHa mentula.— The test is removed, and hardly more of the animal
represented than would be seen in a longitudinal section: a, oral aperture; b,
ganglion; c, circlet of tentacles ; d, branchial sac— the three rows of apertures in
Its upper part indicate, but do not represent, the stigmata; e, the lan^uets ; /, the
oesophageal openins; ; g, the stomach ; h, the intestine ; i, the anus ; AT, the atrium;
I, the atrial aperture ; m, the endostyle ; w, the heart.

water out of the one aperture or the other, and causes the
body to be propelled in the opposite direction.

When one of the simple fixed Ascidians, such as a Phal-
lusia (Fig. 148) or a Cynthia, is laid open by a section car-

516 THE ANATOMY OF INVERTEBRATED ANIMALS.

ried through the oral opening-, at right angles to a transverse
plane passing through its centre, the mouth is found to open
into a large pharyngeal dilatation, termed the branchial sac
(Fig. 148, d). A series of simple or pinnatifid tentacles (Fig.
148, c) is seen encircling the oral aperture at some little dis-
tance within the margin of the lip, which is usually divided,
like that of the atrial opening, into four or six lobes. Imme-
diately behind the tentacular circlet is a ciliated pliaryngeal
band.

On that side of the branchial cavity which is farthest away
from the atrial opening, a pair of delicate lip-like folds ex-
tend, parallel with one another, from the peripfiaryngeal band
along the middle line of the branchial sac as far as the open-
ing of the oesophagus at the opposite end of the branchial
sac. The interspace between these leads into a fold of the
endoderm, lined by a thick epithelium and forming the endo-
style, and, in the middle line of t\\e peripharyngeal hand^ on
the same side as the atrial aperture, there is a tubercular ele-
vation, which contains a ciliated cavity, and answers to the
ciliated sac of Appendicidaria. The walls of this sac are va-
riously folded, and, consequently, the surface of the tubercle
presents a more or less complicated pattern. Continued back-
ward in the middle line as far as the oesophageal aperture on
this side of the branchial sac, there are sometimes one, some-
times two, longitudinal lamellae — the hypopharyngeal folds j
or there may be merely a ridge surmounted bv a series of ten-
tacles, termed languets (Fig. 148, e). The languet which is
nearest the ciliated sac is often the lar2:est of the series. Be-
hind the peripharyngeal band, the lateral walls of the pharj-n-
geal, or branchial, sac are perforated by small elongated ap-
ertures— the stlgtnata — the edges of which are fringed with
long cilia ; and, by means of these apertures, the cavity of
the sac commimicares with the atrium.

The stigmata are arranged in transverse rows, and are
usually very numerous. The reticulated wall of the branchial
sac may be strengthened by longitudinal lamellae, or it may
be raised into few and distant, or many and close-set, folds.
In some cases paf)ill?e of a complicated form are developed
from the inner surface of the sac, and its outer wall is always
connected by vascular trabeculjc with the parietal wall of the
atrium. In some cases [Molgida), the stigmata, instead of
being elongated meshes, are coiled spirally. The atrial cham-
ber (Fig. 148, Jc), into which the branchial stigmata open, is
shown by laying it open from the atrial aperture, in the same

THE TUXICATA. 617

way as the brancliial chamber was laid open from the oral ap-
erture. The atrial opening is thus seen to lead into a cavity,
interposed between the branchial sac and the parietes and
lined upon all sides by a delicate membrane (the third tunic
of Milne-Edwards) like a peritoneum. This membrane has a
parietal and a visceral layer. The former is continued from
the atrial aperture on to the parietes of the body to the level
of the peripharyngeal band in one direction, to a line parallel
with the endostyle in another, and to the alimentary and
genital viscera in a third direction. From these various lines
it is reflected on the branchial sac, of which it forms the outer
wall. At the margins of the stigmata it is continuous with
the endoderm of the pharj^nx, and, at the aperture of the rec-
tum, w4th the endoderm of the intestine. Thus the atrial
membrane forms a bilobed sac, one lobe extending on each
side of the pharynx, and opens outward by the atrial aper-
ture ; it communicates by the stigmata with the interior of
the branchial sac, and, by the anal and genital openings, it
receives the ffeces and genital products. The current which
sets in at the oral and out at the atrial aperture is set in mo-
tion by the cilia of the stigmata.

The atrium of the higher Ascidians differs from that of
Appendicularia^ not only in extent, but in being single and
not double ; and in its single aperture being placed upon the
neural aspect of tlie body close to the ganglion, while the
atrial funnels of Appendicidaria open upon the hsemal aspect
of the body. The development of the higher Tunicata^ how-
ever, shows that the peculiarities of the atrium in them are
of secondary origin ; and that, to begin with, there are two
distinct atria, as in Appendicidaria,

The oesophageal aperture is usually surrounded by a raised
lip, and the short and wide oesophagus leads into a dilated
stomach, whence a shorter or longer intestine proceeds. The
alimentary canal is always bent upon itself in such a manner
that the anus terminates on the neural side of the body, in
the atrial chamber.

In Clavelina^ Amouroticium, Didemniim^ Syntethys, and
most of the compound Ascidians, the greater part of the ali-
mentary canal lies altogether beyond the branchial sac, in a
backward prolongation of the body which has been termed
the ahdo7nen, and is often longer than all the rest of the
body ; the alimentary canal forming a long loop, and the di-
rection of the axis of the branchial sac being continued by
that of the gullet, stomach, and first half of the intestine. In

518 THE ANATOMY OF IXVERTEBRATED ANIMALS.

the JBotryUldm, however, the stomach is bent at right angles
upon the gullet, as in Append Icidaria / the intestine almost
immediately turns forward, and then, turning sharply upon
itself, passes forward parallel with the hinder part of the
branchial sac, on one side of which it opens into the atrium.

A similar arrangement obtains in Perophora, but the
branchial sac extends backward for a short distance on one
side of the stomach. In the solitary Ascidians the stomach
lies sometimes altogether behind the branchial sac {Pelonaia,
some Phallusioe) ; but, usually, the branchial sac extends so
far back that the whole alimentary canal lies on one, usually
the right, side of it. In Phallusia monachus^ the hinder end
of the branchial sac is recurved, and the oesophageal opening
looks backward to the fundus of the sac, instead of forward
to the mouth.

In many Ascidians a strong fold of the endoderm of the
intestine projects into its interior, as in Lamellibranchs and
in the Earthworm, where such a fold constitutes the so-called
typhlosole.

In the pelagic Tunicates, Salpa., Pyrosoma^ and Doliolum^
I found a system of fine tubules ^ which ramify over the in-
testine and are eventually gathered together into a duct which
terminates in the stomach. An apparatus of the same nature
exists in Phallusia^ Cynthia., Molgula^ Perophora, Botryllus,
Botrylloides, Clavellna, Aplidum^ and Dideninum^ and I
have little doubt that it is hepatic in its function. In some
Gynthice, however, there is a follicular liver of the ordinary
character, which opens into the stomach by several ducts.

In some PhallusioB, the alimentary canal is coated by a
very peculiar tissue, consisting of innumerable spherical sacs
containing a yellow concretionary matter. In Molgula (and
in the Ascldia vitrea of Van Beneden) an oval sac containing
concretions lies close to the genital gland, on one side of the
body. As these concretions have been shown by Kupfer ' to
contain uric acid, the organ must be regarded as renal in

' Savignv seems first to have observed this oraran, as -would appear from his
account oi Diazona (" Memoires sur les Animaux sans vertebres," p. 176), and
the description of Plate 12. Lister mentions and figures it in PerovTiora C" Phil.
Trans '' 1834). ^ ^

2 " Reports of the British Association," 1852. Hancock, " On the Anatorav
and Physiology of the Tunicata.^^ (" Journal of the Linnajan Society," vol. ix.)
The development of these tubules from the stomach was traced by Krohn in
Phaliuam^ and by myself in Pijronoma.

3 " Zur Etitwickeiung der einfachen Ascidien." (" Archiv fur Mikr. Ana-
tomic," 1872.)

THE TUXICATA. 519

function. M. Lacaze-Duthiers ' terms this sac an "organ of
Bojanus ;" but, as he admits, no opening is discoverable : it
would probably be more correct, therefore, to regard it as
the representative of the glandular part of the organ of Bo-
janus.'

The heart is an elongated sac open at each end, lodged
near the stomach, and close to the hinder extremity of the
branchial sac. After a certain number of contractions in
one direction, it stops and contracts for the same number of
times in the opposite direction. The course of the circula-
tion is thus reversed with great regularity. The blood is a
clear fluid, containing colorless corpuscles.

Respiration is efl"ected in the walls of the branchial sac
through which the blood is driven. The supply of aerated
water is kept up by the currents already mentioned, which
subserve the ingestion of food, the respiratory process, and
the ejection of effete matters, as well as the expulsion of the
generative products. The test in which the body is inclosed
is sometimes closely adherent to the surface of the ectoderm,
but sometimes is united with it onlv at the oral and atrial
apertures, and by prolongations of the body. In consistency
it presents every variety, from soft and gelatinous, to dense
and hard like cartilage, or tough like fibrous tissue. In some
cases the exterior of the test is covered with horny spines,
tubercles, or even with regularly-disposed plates ( Chelysonia).

In texture, the test may present merely a homogeneous
matrix, in which cells like connective-tissue corpuscles may
be scattered ; or it may resemble cartilage [Phallusia) or
fibrous tissue. In most cases it is non-vascular ; but, some-
times, tubular prolongations of the ectoderm, divided by a
median septum and containing blood, enter it at one point,
and thence branch out through its substance.

In the Chevreulius of Lacaze-Duthiers,' the test is some-
what like a snuff-box with a movable lid. There is no hinge,
however, but the substance of the lid is continuous with that
of the rest of the test along the line of junction. And the
elasticity of this part causes the lid to stand open, unless it is
shut by the contraction of two adductor muscles which are
attached to it.

1 "Les A scicTies simples des Cotes de France." ("Archives de Zooloarie
experimentale," 1874.) M. Lacaze-Duthiers has obtained murexide by heating
this substance with nitric acid.

' There is a close resemblance between the cells of which this organ is com-
posed and those which constitute the primitive kidney in the Pulmonata»

s " Annales des Sciences Naturelles," 1865.

520 THE ANATOMY OF IXVERTEBRATED ANIMALS.

Chk

Fig. I40.—Phann.<nam(77vjninafa.~-Ycinon& stashes in the development of the larvaj.

C After Kowalewsky.^i
I. The vesicular morula, flattened and about to undergo invagination :fh. hlastoccele.

The large blastomeres conetitnfe the hypoblast, the small one? the epihlapt.
n. The gastnila with the blastopore, or opening of invagination, eo: en, the blasto-

mere? which conctitnte the rudiment of the urochord; dd, the remaining blasto-

meres of the hypoblast.
ni. A more advanced embryo: ch, dd, a? before; c. the epiblast ; n, the nervous

layer of the neural cavity, which is now open only in front near ch.
rV". An embryo with tlie caudal appendage distinct. The nerve-tube n is complete,

and the mu:^cle-cells m are distinguishable.

* ""Weitere Studicn fiber die Entwickelung der einfachen Ascidien."
(" Archiv furMikr. Anat.," 1871.)

THE DEVELOPMENT OF THE TUNICATA. 521

V. The body of a larva as it escapes from the egir : a, the eye ; gb, the eaccular an-
terior end ol the central nervous apparatus into which the otolith projects ; L'g,
Jim, itri tubular backward prolongation ; C/is, cells of the urochord ; o, mouth; kl,
atrial aperture ; /, opening at the anterior end of the central nervous apparatus,
by which it coiumuuicates with the alimentary cavity ; d. commencemem of tne
oesophagus and stomach ; m, blood-corpuscles ; /ij), papillae by which the larva

attaches itself. , ^ ,

VI. The body and the commencement of the caudal appendage of a free larva two
days old: en, endostyle; k^, branchial sac ; Ika, 2ks, nranchial stigmata ; bb. en-
trance into the blood-siuus between them; d, iutesiiue; 6, blood-corpudclee; klm,
atrial aperture.

The reproductive organs of the two sexes are united.
Usually, the testis and the ovary have the form of racemose
glands situated in the loop formed by the intestine ; or be-
yond it, when the "abdomen" is long; and their ducts run
parallel with one another, to open close together beside the
anus. In many of the simple Ascidians, however, the repro-
ductive organs are lodged in the lateral walls of the atrial
cavity, and their ducts are distant from the anus ; and, some-
times, there are many distinct genital glands.

In some genera, e. g., Phallusia,, each e^g is surrounded
by an ovicapsule, formed by the coalescence of cells of the
epithelial lining of the ovary, and these cells may grow out
into processes which give the fully-formed egg a stellate ap-
psarance.

Complete yelk-division takes place, and the morula under-
goes invagination (Fig. 149, I., TI.). A longitudinal depres-
sion of the eniblast, extendins* forward from the marorins of
the aperture of invagination, next makes its appearance ;
and, deepening, gives rise to an involution, the edges of
which unite, and thus shut off a tubular portion of the epi-
blast. This is the rudiment of the nervous ganglion (Fig.
149, III.). The aperture of inv^agination closes, and an out-
growth of the body gives rise to the caudal appendage, into
which the urochord, formed bv the coalescence of certain cells
of the hypoblast, extends (Fig. 149, IV.). The sac of the
hypoblast becomes divided into its branchial, oesophageal,
gastric, and intestinal portions, and the mouth is formed by
the perforation of a spot in which the h3''poblast and the epi-
blast cohere (Fig. 149, VI.). The atrial cavity is formed by
two involutions of the ectoderm, which extend inward and
apply themselves to the lateral and neural walls of the bran-
chial sac (Fig. 149, VI.). Their originally separate apertures
eventually coalesce into one,^ The atrial tunic thus formed,

^ * In 1852 Krohn discovered the fact tliat the larva of Phallusia is provided
with two distinct symmetrically-disposed openings, by which the orisfinally
separate atria open outward ; and that the two eventually coalesce into the sin-

522 THE ANATOMY OF IXYERTEBRATED ANIMALS.

and the walls of the branchial sac, coalesce and become per-
forated, m order to give rise to the stigmata.

The test appears, at first, to be a cuticular secretion of
the epiblast, and to derive its cellular elements from the wan-
dering into its substance of cells derived from the epiblast.

In 3Iolgula tubulosa^ Kupfer and Lacaze-Duthiers have
observed that the fecundated eggs are expelled from the
atrial cavity, and almost immediately become fixed to the
surface on which they fall. Yelk-division takes place, and,
after four nearly equal blastomeres are formed, much smaller
ones are developed from one face of these, and increase until
they constitute a blastodermic layer around the larger blasto-
meres, which undergo a slower division. The alimentary
cavity is formed by invagination. The embryos leave the
egg as voal bodies, capable of undergoing considerable but
slow changes of form, and devoid of any caudal appendage.
Each embryo rapidly invests itself with a transparent test,
throws out several tubular prolongations of the ectoderm,
and finally passes into the adult condition. Although no tail
is developed, a cellular mass is to be seen in the same posi-
tion as that occupied by the remains of this appendage, when
it has undergone its retrogressive metamorphosis, in the As-
cidians with caudate larvae. The atrial aperture is single at
its first appearance, and no larval sensory organs are devel-
oped.

In the compound or social Tuyiicata, many ascidiozoSids,
which are united by a common test into an ascidiarium, are
produced by gemmation from a solitary metamorphosed larva.

Sometimes, as in Clavellna and Perophora^ the parent
ascidiozo(3ids give rise to creeping stolons, from which branches,

^le atrium of the adult. Kowalewsky, Fol, and later observers, afrree tliat
these openinsrs and the atrial sacs are formed hy two involutions of the ecto-
derm, which applv themselves to the sides of the pharynx, and coalesce with
it at the points which become perforated by the stigmata; of which, in Plial-
lusia, there are at first but two on each side. If this is a true account of the
oricrin of the atrium, the atrial membrane is obviously part of the ectoderm,
and its cavity is analo,'Xous to the pallial cavity of a mollusk.

On the other hand, MetschnikofF and Kowalewsky agree that in the buds
of Botri/llus, and other ascidians which multiply by gemmation, the two primi-
tively distinct atrial cavities are portions of the alimentary sac, which become
shut off from it, and subsequently open outward.

Metschnikoff (" Entwickeluriesgeschichtliche Beitrage." "Bulletin de
I'Acad, St.-Petersbourg," xiii.) therefore compares the atrium to the entero-
co^le of Echinodcrms. Renewed observations specially directed to this point,
which is of great morphological importance, are much needed. If the atrial
cavitv is really an enterocoele, it will answer to the perivisceral cavity of the
Brachiopoda., the pseudo-hearts of which will correspond with the primitive
atrial aperture.

PYROSOMA.

523

Fig. 150. — Pyrosoma giganievm.—l. A vertical section of the wall of the Apcidiarium
near the cloacal aperture and includinir its lip. II. The youngest condition of a
bud before the ectoderm is elevated. III., IV., V. Further stages of the develop-
ment of a bud. VI. A fully-formed bud with a second ascidiozoOid in course of
development from its peduncle.

524 THE ANATOMY OF INVERTEBRATED ANIMALS.

VII. A foetus with the blastoderm divided into five seinneTits, of which the cyatho-
zo5id (I.) 19 the largest. VIII. A foetus, tlie aBCidiozo5ids of which half encircle
the base of the cyaihoz )5id. IX. Fcetii?, the most advanced etage observed. The
remains of the conjoined cyathozooid and ovisac are hidden by the circle of
ascidiozooids. -■,,.■,

The letters have the same signification in all the figures, a, test; a^, labial process ;
a*, lip of the cloacal aperture; a^, ceils of the enibryoi;ic test ; e, oral aperture ;
p\ atrial aperture ; ^, endostyle ; P, Z^, branchial sac and siidrmata; r, heart; z-^,
stolons of the adult ascidiarium ; r*, stolons of the embryonic ascidiarium ; s, ovi-
sac; t, testis ; M, w', ovum; w', peduncle of a bud; x, the alimentary portion of
the endoderai entering into a bud; x^, its generative portion; x^, the ectoderm
entering into a bud ; «, the ceieoblast ; z, ganglion.

I., II., III., IV., V. Segments of the blastoderm. I. CyathozcOid. IV.-V. Ascidio-
' zooids. B, mouth of the cyathozooid.

which develop new ascidiozooids, are given off at intervals ;
but, more commonly, the ascidiarium is massive, and the as-
cidiozooids retain no permanent connection with one another.
In the BotnjllidrB^ the zooids are arranged in whorls around
a common central cavity, or cloaca, into which the atria of
all the members of the whorl open. In Pyrosoma, which is
a sort of floating Botryllus, the process of budding is highly
instructive, as it exempliiies the manner in which gemmation
occurs in the Tanicata in general.^

The ascidiarium of Pyrosoma (Fig. 150, I.) has the form
of a hollow cylinder, rounded and closed at one end, truncated
and open at the other, formed of a firm transparent test, in
which the zooids are arranged in whorls. Their oral apertures
open on the exterior surface, and their atrial apertures into
the interior of the cylinder. The haemal aspect of each zooid
is turned toward the closed end of the cylinder. The bran-
chial sac has the ordinary structure, and each zooid is provid-
ed with a testis and with an ovisac, containing a single ovum.

E^ery zooid multiplies by gemmation from a region of the
body which lies immediately behind the extremity of the en-
dostyle. Close to the heart, attached to a short caecal process
of the endoderm which constitutes the extremity of the endo-
style, and which I have termed the endostylic cone, is a cellu-
lar mass — the remains of that mass of indifferent tissue which
I have called the generative blastema, and from which the gen-
erative organs of the gemmiparous zooid have been developed
(Fig. 150, II.). The endostylic cone elongates, and, curving
toward the haemal side of the bodj^, applies itself closely to the
ecoderm (Fig. 150, III.). The latter grows out into a conical
elevation, which projects into the surrounding substance of
the test, and contains a mass of mesoblastic cells, one of which

* Huxley, " Anatomy and Development of Pyrosoma.'''* (" Trans. Linnaean
Society," i860.) Kowalewsky {I. c, infra^ p. 616).

THE BUDDING AXD FISSION OF ASCIDIANS. 525

{u') has already taken on the character of an ovum, and is
surrounded by a rudimental ovisac. The conical bud elongates
and dilates at its extremity, and the dilatation gradually takes
on the form of a new zooid united by a narrow neck, or pe-
duncle, with the parent (Fig. 150, IV.). The endostylic cone
gives rise to the whole alimentary canal of the bud, while the
ectoderm of the latter proceeds from the ectoderm, and its
ovisac and testis from the mesoblastic cells, of the parent.
Thus the organs of the bud are all the direct product of the
corresponding parts, or of the primitive layers of the germ
from which they are derived, in the parent.*

After the terminal bud is formed, a second is usually de-
veloped immediately below it (Fig. 150, VI.) by the growth
of the ectoderm, endodermal axis, and mesoblastic cells of the
peduncle ; and it would appear that this process is frequently
repeated. The fully-formed bud becomes detached, and takes
its place among the other zooids in the test, there to repeat
the process of gemmation.

The observations of Krolin, MetschnikofF, and Kowalewsky,
have shown that two components enter into the buds of ascid-
ians in general ; first, an outer la3'er consisting of the ecto-
derm of the region in which the budding takes place, and,
secondly, an inner layer derived from the endoderm of the
branchial sac (Perop/iora) ; or, as in J^otryllus^ according to
MetschnikofF, from the atrial tunic' To these must be added

^ In my second memoir on Pyrosoma C' Trans. Linn, Society," xxiii., p.
211) I have said:

" Gemmation does not take place in Pyrosoma as in so many of the lower
animals (e. g., the Bydrozoa and Polyzoa^ or Salpa and Clavelina, amonsr the
ascidians), by the outgrowth of a process of the body-wall, whose primarily "whol-
ly indifferent parietes become differentiated into the organs of the bud; but,
from the lirst, several components, derived from as many distinct parts of the
parental organism, are distinsruishable in it, and each component is the source
of certain parts of the new being, and of them only. Thus the body-wall or
external tunic of the parent gives rise to the external tunic of the bud ; while a
process of the endostvlic cone of the parent is converted into the alimentary
tract of the bud, and the reproductive organs of the latter are furnished by a
part of that tissue whence the reproductive organs of the parent took their
origin."

As will appear further on. however, recent investisrations show that the
•whole process of budding in the srreat majority of the Tf/nicata, and at any rate
the first steps of that process in Salpa, are essentially similar to those in Pyro-
soma ; and it remains to be seen whether there is any difference in other As-
cidians. And as resrards even the ffydrozoa, the expression that the parietes
of a bud are at first " wholly indifferent " in structure is not quite accurate, in-
asmuch as they are composed of an ectodermal and an endodermal layer, which
are continuous with those of the parent, and give rise to homologous orsrans.

^ If, as some observations tend to show, the atrial tunic itself is a diver-
ticulum of the primitive endoderm, this case would form no exception to the
general law of budding in the Tunicata.

526 THE ANATOMY OF INVERTEBRATED ANIMALS,

a third component, derived from the indifferent tissue, out of
which the reproductive organs of the parent have been de-
veloped.

In Amouroxccium proliferum^ agamic multiplication takes
place when the larva has fixed itself and grown into a soli-
tary ascidian. The long post-abdomen (as the prolongation
of the abdomen beyond the alimentary canal is termed) sepa-
rates itself from the body, carrying with it the heart, and
divides into a number of segments which rise to the summit
of the test of the parent, range themselves around it, and be-
come converted into independent zooids. The parent devel-
ops a new heart and post-abdomen. The process appears to
be repeated in the post-abdomina of the new zooids. The
post-abdomen is a process of the ectoderm, the inner cavity
of which is divided by a septum into two chambers, contain-
ing many fatty cells. The septum itself incloses a cavity,
and there appears to be no doubt that it is a prolongation of
the pharyngeal sac. When the segments of the post-abdo-
men develop, the cavity of the anterior end of the septum
dilates and divides, as in Didemnum^ into three chambers, of
which the median becomes the branchial sac, and the lateral
the atrial chambers. The rest remains as the septum of the
post-abdomen of the foetus, and its cavity at first communi-
cates with the branchial sac, between the endostyle and the
oesophageal aperture.

Kowalewsky ^ has observed the formation of buds from
free cellular masses in the common test of Didemniirn styli-
ferum; the origin of these masses is undetermined. They
multiply by division, after the rudiments of the alimentary
cavity and of the reproductive organs have made their ap-
pearance. , The alimentary cavity gives off a process whence
the oesophagus, stomach, and intestine are developed, and
then becomes divided by longitudinal partitions into three
chambers, a median and two lateral. The latter give rise to
the lateral chambers of the atrium, which subsequently open
into one another on the neural side of the body, and finally
communicate with the exterior by a median atrial opening.

Gegenbaur' has described the detachment of the ova of
a species oi Didemmim into the substance of the common test,
where they are developed into caudate larvae provided with
an eye. Before the development of the larva is nearly com-

2 " Ueber die Kcospung der Ascidien." (" Archiv fiir Mikr. Anatomie,"
1874.)

1 " Ueber Didemnum gelatinosumy (" ArcLiv far Anat.," 18G2.)

THE DEVELOPMENT OF BOTRYLLUS. 527

plete, a zoold is formed from it, so tliat, at one time, the em-
bryo appears to have two branchial sacs.

Metschnikoff ^ and Krohn'^ have shown that the caudate
larvge of Botryllus are not composite, as Savigny and Sars
supposed, but that the bodies imagined by these observers to
be buds are simply diverticula of the ectoderm, and become
converted into the vascular processes, which ramify through
the common test, and commonly end in dilatations. In the
adult, the buds are developed, one, or sometimes two, at a
time, at the sides of the body, and consist of an outer layer,
derived from the ectoderm, and an inner layer, which, accord-
ing to Matschnikoff, proceeds from the atrial tunic. From
the inner layer the alimentary canal of the bud proceeds, and
between the inner and the outer layers the rudiments of the
genitalia appear. The ovaria advance toward their develop-
ment much more rapidly than the testes. The zooids thus
developed, as they enlarge, rise to the surface, taking the
place of those from which they proceed and which die away.
The ova are impregnated from without, and undergo their de-
velopment in the atrium of the parent. Subsequently the
testes attain their full development ; and, at the same time,
the buds are formed which will give rise to a third generation,
supplanting the second.

After the larva (which may be called A) has attached it-
self, the first sets of zooids which are developed are sexless.
The first bud arises on the right side of the body of the larva
(A) in the neighborhood of the heart ; as it increases in size,
the parent withers away, and the zoOid (B) thus developed
takes its place. Two buds, a right and a left, are developed
from (B) and become zooids (C, C), B disappearing. The two
zouids (C, C) are so disposed that their atrial ends are close
together, and their oral ends turned away from one another.
These each develop two lateral buds, which become four
zooids (D, D, D, D). The zooids (C, C) disappear as before, and
their successors arrange themselves in a circle. Each of these
develops two, or sometimes three, lateral buds ; these grow
into zooids, which supplant their predecessors, and are them-
selves, in turn, supplanted.

Every new system of the later successions is, at first, de-

1 " Entwickelungrssrescbichtliclie Beitr^ge." (" Bulletin de 1' Academie des
Sciences de St.-Petersbourg," xiii, 1868\

2 " Ueber die Fortpflanzungsverhaltnisse bei den Botrvlliden " (" Arcbiv
fir Natursreschichte," 1869). " Ueber die fraheste Bilduno- der Botryllea-
Stocke" \^ibid.).

528 THE ANATOMY OF INVERTEBRATED ANIMALS.

void of a common cloaca; and the zooids which compose it
may arrange themselves into one or several circles, each of
which then acquires its cloaca.

It thus appears that, in l^otryllus^ the ascidiozooid which
results from the metamorphosis of the caudate larva serves
merely as a kind of stock, from whence the other zooids
which build up the ascidiarium proceed ; and this leads to
the still more singular process of development in Pyrosoma^
in which the first-formed embryo attains only an imperfect
development, and disappears after having given rise to four
ascidiozo5ids.

In Pyrosoina^ the ovisac is attached by a short oviduct to
the walls of the atrium, into which it eventually opens, and
thus allows of the entrance of the spermatozoa.

Of the process of yelk-division I could see nothing in my
specimen, which was preserved in spirit, but it has since been
traced, in fresh specimens, by Kowalewsky,' who compares it
to that which takes place in osseous fishes. The result is the
formation of an elongated, flattened blastoderm, which occu-
pies one pole of the ^^^^ and is converted into what I termed
the cyathozooid, which is shown by Kowalcwsky to be a sort
of rudimentary ascidian (Fig. 150, VIII.). From this a pro-
longation or stolon is given ofi", which becomes divided by lat-
eral constrictions into four portions, each of w^hich gives rise
to a complete ascidiozooid. As these increase in size, they coil
themselves round the cyathozooid, with tlieir oral openings
outward and their cloacal openings inward, and thus lay the
foundation of a new ascidiarium (Fig. 150, VIII.). The cyatho-
zooid eventually disappears, and its place is occupied by the
central cloacal cavity (Fig. 150, IX.). Thus, in Pyrosoma^ the
usual first stage of an Ascidian — the caudate larva — is abor-
tive, and serves only to found the colony by the buds which
are developed fiom it.

In the pelagic genus Poliolum ' the cycle of life of the
species is represented by distinct sexual and sexless forms.
The ^^^ produced by the sexual form (A)' gives rise to a
caudate larva which passes into the first sexless form (B) ;
this gives off from the neural side of the body an outgrowth

' " Fcber die Entwickclungs<Teschichte der Pyrosoma." (" Archiv fur Mikr,
Anatomie," 187;''.')

2 Huxley, " Remarks upon Appendicylnrin and Dofiohimy (" Phil.
Trans.," 1851.) Krohn, " Ueber die Gattunc Doliolvm.'''' (" Archiv fur Natur-
geschichte," 1852.) Geprenbaur, " Ueber die Entwickelung von Doliolum.^*
{ZeiUchrift fi'ir wiss. Zoologie, 1853.)

8 Keferstcin and Ehlers, " Zoologische Beitrrge," 1861.

THE DOLIOLUM. 529

or stolon, from whicli buds are developed. These buds are
arrauged in three rows, two lateral and one median, and grow
into zooids of two different forms, of which the median may
be indicated by C m, the lateral by C /. All these zooids are
detached, and swim about as independent organisms. What
becomes of the lateral zooids (C /) is unknown. But the
median zooids give off a stolon from the haemal side of the
body on which buds are developed, which pass into the sex-
ual form (A).

The sexual zooid (A) (Fig. 151) is shaped like a cask with
an opening at each end ; these are the oral and cloacal aper-
tures. According to Keferstein and Ehlers there is no test,
the outer wall of the body being formed, as in most Appen-
dicularicBj by the ectoderm. Eight muscular bands encircle
the body, and by their contractions expel the water from
either the oral or the cloacal ends. The body is thus pro-
pelled either backward or forward. The branchial sac is
much simplified. In Doliolicni Mulleri^ the atrial cavity does
not extend further forward than the hinder end of the wide
pharynx, and this is perforated only by two rows of stigmata,
four or five in each. In Doliolwn denticulatuin (Fig. 151),
on the other hand, the atrial cavity extends forward at the
sides of the pharynx, both on the haemal and the neural side,
and the stigmata are numerous and vertically elongated.

An opening in the middle line of the haemal face of the

Fig. 15^.— Doliolum d&nficvlahim.— a, ganglion: c. endostyle ; c?, oral opening; g,
oesophagus; i, stomach; /, intestine; p,j/, testis; r, heart; t^t, nauscles.

pharynx leads, by a short gullet, into a dilated stomach,
whence the slender intestine proceeds to terminate in the
atrial cavity. The nervous ganglion is situated in the third
intermuscular space in D. denticulatum. There is a ciliated
sac, but no auditory organ, in the sexual form. The testis
23

530 THE ANATOMY OF INVERTEBRATED ANIMALS.

is a long tube (Fig. 151, jO,p), which lies on one side of the
hnemal face of the body and opens on a papilla in the atrium.
The ovary, small, rounded, and situated close to the hinder
end of the testis, contains many ova. According to Kefer-
stein and Ehlers, the ova and spermatozoa appear often to
become ripe at the same time.

The first sexless zooid (B) resembles A in general form,
but has nine muscle-rings. The long stolon, which trails in
the water, is attached in the seventh intermuscular space to
the middle of the neural face of the body. The stigmata are
arranged as in the form A, of Doliolum Mulleri^ and one of
the anterolateral nerves terminates in an otolithic sac. It is
spherical, and contains a single otolith.

The zooids produced by the lateral buds of the stolon (C T)
have wide oral apertures, and the body is shaped somewhat
like the bowl of a spoon. They possess neither auditory or-
gans nor genital organs, nor have they been observed to de-
velop buds. The median zoOids (C rri) closely resemble the
sexual zooids. The stalk by which each is attached, and the
insertion of which is in the middle line of the haemal face in
the sixth intermuscular space, remains as a prominence after
the animal is set free ; and, from the base of this prominence,
buds are developed which take on a sexual form (A).

In the Salpoe, the divergence from the ordinary Tunicata
reaches its maximum. The oral and atrial openings are situ-
ated at opposite extremities of the body, as in Pyrosoma and
Doliolum ; and the branchial stigmata are represented by
wide vacuities at the sides of the branchial sac, the walls of
which are thus represented only by the epipharyngeal folds
on the one side, and a narrow trabecula, which occupies the
region of the hypopharyngeal band, on the other side. The
relatively small alimentary and reproductive viscera are some-
times a2r2:reo:ated into a mass — the so-called nucleus — at the
posterior end of the haemal side of the body. The chief mus-
cular bands, by the contraction of which the water is driven
out of the branchial and atrial apertures, and the propulsion
of the animal is effected, are transverse, but do not form com-
plete hoops, as in DoUolmn,

In all the Salpce^ each species is represented by two sets
of zooids, the one sexual and the other sexless. The sexual
zooids are produced by budding from a stolon, which is given
off from the body of the sexless form in the immediate neigh-
borhood of the heart. When the sexual zooids thus formed
are detacher], they are at first connected into chains of vari-

SALPA DEMOCRATICA-MUCROXATA. 531

ous forms, but these eventually break up, and the constituent
zooids are set free. Fig. 152 shows the two zooids of the
species Salpa democratica-inucronata^ viz., the sexless zouid,
Salpa democratica (Fig. 152, L), and the free sexual zo5id,
/Salpa mucro?iata (Fig. 152, 11.).

The recent investigations of Dr. Todaro,* in accordance
with those of Kowalewsky, show that the stolon is formed,
as in Pyrosoma, by the conjunction of a process of the endo-
derm which forms the extremity of the endostyle, with an
outgrowth of the ectoderm, and with certain cells of the meso-
blast. But, according to Todaro, there is this essential differ-
ence : the young Salpce, which make their appearance in
double series along the stolon, are developed altogether from
the mesoblastic cells. These cells, in fact, besome aggre-
gated into masses, of which four are arranged in the circum-
ference of each segment into which the stolon is divided ; and
two of these masses, one on each side of each segment, are
converted into young Saljym by a process analogous to that
by which a morula becomes an embryo. If this account of
the matter be correct, the agamic development of the Salpce
would rather resemble that of the germ masses of the sporo-
cysts of Trernatoda^ or the pseud-ova of insects, than ordinary
budding.

Each sexual zooid possesses a testis and a single ovum.
The latter is contained in an ovarian follicle, the slender duct
of which is attached to the wall of the atrium and opens into
the atrial cavity. The testis attains its full growth and func-
tional perfection only after the ovum has undergone develop-
ment. It follows, therefore, that impregnation must be ef-
fected by the spermatozoa of some other zooid. The sexless
form which is developed from the egg goes through the early
stages of its development in the atrial cavity of the parent, to
the walls of which it is attached by a peduncle (Fig. 152,
III.), the centre of which is occupied by a diverticulum of the
vascular canals of the parent, inclosed within a cup-shaped
cavity in free communication with the blood-sinuses of the
foetus. It is, in fact, a true placenta ; and, during life, the
independence of the foetal and maternal circulations is readily
observed, as the blood-corpuscles of the two organisms course
through their respective channels.

The early stages of the development of the embryo Salp)a
have been investigated by numerous observers, most recently

1 «

Sopra lo Sviluppo e I'Anatoinia delle Salpe," 1S75,

532 THE ANATOMY OF INVERTEBRATED ANIMALS.

by Kowalewsky/ Todaro, Brooks,' and Salensky.' The ob-
servations of the last-named author relate chietly to Sal2^a

Fig. 152. — Salpa democraticamucronata. — T. Sdlpa democrafica. U. Salpa mvcro-
nata. III. A fcetal Salpa democratica attached by its placenta to the wall of the
atrial cavity of a Salpa mucronata. IV. Part of the stolou of Salpa democratica
with attached Salpa mucronata buds.

The letters have the same signification throujjhout : a, oral ; 6, atrial orifices ; c,
endosiyle; (Z, ganglion ; e, hypopharyngeal band in a eo-cailecl " brancliia;"j^
lan^uet; g, heart ; A, gemmiparoiis stolon ; t, visceral mass or nnclens ; k, mus-
cniar bands ; m, placenta ; n, blood-sinus ; q, ovisac and ovum ; t, stomach; w,
ciliated sac ; ce, celeoblast ; a, ectoderm and test ; j3, endoderm.

democratlca-imicronata, and his account of the process ap-
pears to me to be the most satisfactory.

* " Nachrichten der Koni<^lichen Gesellschaft zu Gottin^en," 1868.

2 " Bulletin of the Museum of Comparative Zoology," r^o. 14.

3 Zdtschrij'tfur whs. Zoologie^ 1876.

THE DEYELOPMEXT OF THE SILPJE. 533

The egg is impregnated in the ovarian follicle, as in Pyro-
soma ; and the oviduct, shortening, gradually draws the ova-
rian follicle, with its contents, into a sort of incubatory pouch,
which is a diverticulum of the wall of the atrium, and pro-
jects into the atrial cavity.

For distinction's sake the incubatory pouch may be termed
the ovicyst. As the oviduct shortens, it widens, and consti-
tutes, together with the ovarian follicle, a single uterine sac,
the outer or oviducal half of which applies itself to the wall
of the ovicyst, while the inner half contains the ovum. The
vitellus undergoes complete division, and the superficial layer
of bListomeres constitutes itself into an epiblast, investing
the solid mass formed by the other blastomeres, w^iich repre-
sent the hypoblast. A mesoblastic laj^er subsequently ap-
pears between the two. The nervous ganglion results from
an inv^olution of the epiblast, while the branchial sac, the
alimentary canal, and the asrium, are the product of the sub-
division of a cavity which appears in the midst of the hypo-
blast. The maternal and the foetal parts of the placenta
arise, respectively, from the wall of the ovarian sac, and from
certain large blastomeres on the adjacent hagmal face of the
embryo.

Todaro ao^rees with other observers in statins: that the
vitellus undergoes division, and that a small celled blastoderm
invests the large remaining cells, which he terms the germinal
mass. But his acoount of the further stages of development
is very diflferent. A circular thickening of the blastoderm
separates the hemisphere which is directed outward from that
which is turned inward, and gives rise to a lamellar outgrowth.
It is, at first, directed toward the inner end of the ovisac,
having reached the bottom of which, it becomes reflected ;
and the reflected portions lining the inner wall of the ovisac,
and meeting over the outer hemisphere, form a sort of am-
niotic investment of the embryo. It is the cavity left be-
tween this " amnion " and the inner hemisphere of the blasto-
derm, which becomes the parental blood-sinus. An involu-
tion of the outer hemisphere of the blastoderm gives rise to
the alimentary canal, which becomes shut ofl^, as the endoderm,
from the remaining blastoderm, which constitutes the ecto-
derm. A mass of cells which appears in the middle of the
outer half of the embryo, between the alimentary sac and the
ectoderm, and which has only a transitory existence, is re-
garded by Todaro as the representative of the urochord.

CHAPTER XI.

THE PEEIPATIDEA, THE MYZOSTOMATA, THE EXTEROPXEUSTA,
THE CH^TOGXATHA, THE XEMATOIDEA, THE PHYSEMARIA,
THE ACAXTHOCEPHALA, AXD THE DICYEMIDA.

I HAVE reserved for discussion in this chapter the JPeripa-
tidea^ which have heretofore been referred by most authors to
the Annelida / and certain groups of the lower Metazoa^ the
precise morphological relations of which are as yet uncertain,
although it is pretty clear that several of them are allied with
the lower Annelida^ the Hotifera, and the Turbellaria. They
are, for the most part, totally devoid of segmentation ; while
the Clicetognatha and the Myzostomata alone present any
structures resembling limbs, though the nature of these is
doubtful. So far as the nervous system is clearly made out,
it exhibits no such chain of post-oral ganglia as characterizes
the hi^rher worms.

&

The Peripatidea. — At p. 225, I have referred this group
to the Arthropoda^ Mr. Moseley's memoir on Peripatus ^ hav-
ing left no doubt upon my mind, that he had satisfactorily
proved the justice of the surmise respecting its affinities
originally made by Gervais. It is only recently, however,
that I have been able, thanks to Mr. Moseley, to examine one
or two specimens of Peripatus Novm Zelaniae^ and to satisfy
myself of the main point, namely, the existence of the tra-
cheal system which he has described.

Of the ^enus Peripatus several species are now known,
from the West Indies, South America, the Cape of Good
Hope, and New Zealand, where they are found among the de-
caying wood of damp and warm localities. They have the

»" Philosophical Transactions," 18H. See^ also, the valuable memoir of
Grube, " Ueber den Bau von Peripatus Edwardsii " (" Archiv far Anatomie,"
1853).

THE PERIPATIDEA. 535

curious habit of throwing out a vreh of viscid filaments when
handled or otherwise irritated.

The head is distinct, and is provided with a pair of many-
jointed antenna-like tentacula and two simple eyes. The
mouth, situated upon the under surface of the head, is sur-
rounded by a prominent lip, which incloses a pair of jaws,
each of which is terminated by tw^o curved chitinous claws,
similar to those of the feet. On each side of the mouth, the
head supports a short obscurely -jointed " oral papilla," which
is somewhat like one of the feet, but is devoid of claws and
perforated at its extremity. The head is followed by an un-
segmented body produced laterally into paired appendages,
which vary in number from fourteen to more than thirty, ac-
cording to the species ; and each of these appendages is in-
distinctly articulated, the terminal joint being provided with
two small curved claws.

The anus is terminal, and the genital aperture is situated
on a papillae, a little distance in front of the anus, on the
neural or ventral face of the bod3\

The alimentary canal commences by an ovoid muscular
pharynx. The oesophagus, continued from this, gradually
dilates into a wide and long stomach, from which a very short
intestine is continued to the anus, situated at the posterior
end of the body. There are no Malpighian ca^ca. Two very
large ramified tubular glands, which secrete the viscid matter
of which the web is composed, lie at the sides of the alimen-
tary canal, and open outward by the perforations of the oral
papilla?. A vessel occupies the middle line of the dorsal
body-wall, and is probably a heart.

The respiratory organs are the trachea discovered by Mr.
Moseley. The numerous pores, or stigmata, from whick the
tracheae take their origin, are scattered all over the surface
of the body, one row being median and ventral. Each stigma
is the outward termination of a short, wide tube, which, at
its opposite end, branches out into a pencil of fine tracheae,
which rarely divide, and are distributed in great abundance
to the viscera. They are very delicate tubes, which often
take an undulating course, and are rarely more than tj-oVo"
of an inch in diameter. In optical section, their walls have
a finely-beaded appearance, as if from the presence of trans-
verse thickenino-s, thouo:h distinct transverse markino;s are
rarely to be seen.

The nervous system, as Milce-Edwards discovered, con-
sists of two ganglia in the head, closely united above the

536 THE ANATOMY OF INVERTEBRATED ANIMALS.

oesophagus. From each of these a relatively stout longitudi-
nal cord proceeds, overlying the bases of the feet (and hence
widely separated from its fellow), to the posterior extremity
of the body. As Grube has stated, there are no distinct
ganglia on this cord. On the contrary, ganglionic cells ap-
pear to be pretty evenly distributed along its ventral face,
throughout its length ; and nerves, which pass transversely
outward and inward, are given off from opposite sides of it
at short intervals. Grube has shown that many of the
branches that take the latter direction are commissures be-
tween the two cords.

The muscles of Peripcitus are not striated, which is a
curious exception to its generally well-marked arthropod char-
acteristics.

Mr. Moseley has proved that the sexes are distinct. The
ovary is small, divided by a median septum into two lobes,
and lies beneath the alimentary canal. The oviduct, at first
single, divides into two branches, which are long, and, pos-
teriorly, present uterine dilatations. They then unite, and ter-
minate by a short vagina on the ventral aspect of the rectum.
The testes are ovate bodies, each with a cascal appendage.
The lonof and coiled vasa deferentia unite into a common
duct, which opens in the same position as in the female. The
ova are developed within the uterine dilatations of the ovi-
ducts.*

Mr. Moseley has made out the chief points in the develop-
mental history of Perii^atus.

In an early condition, the embr3'o is very like that of a
Scorpion, but is folded upon itself, so that the ventral aspects
of the anterior and posterior halves of the body are turned
toward one another. As in the Scorpion, there is a pair of
large procephalic lobes, succeeded by a series of segments,
from the sides of which, processes — the rudiments of the
limbs — bud out. The procephalic lobes give rise to a kind
of hood, the lateral angles of which extend over the bases of
the first pair of limbs, and join with those of the second pair,
which are the oral papilla? of the adult. The first pair of
limbs thus become inclosed within the hood (tlie margins of
which form the suctorial lip of the adult), and developing
two chitinous claws uoon their extremities, like those of the

' One of tlao specimens wliicb I examined was a precrnant female, but the
viscera were glued together, apparently by the action of the spirit in which it
had been preserved, in such a manner, that little could be made of their struc-
ture or of that of the embryos.

THE MYZOSTOMATA. 537

other limbs, they are converted into the jaws of the adult
animal. It is remarkable that the antennae are developed
from the anterior part of the procephalic. lobes ; while the
chelicerse of the Scorpion appear at the posterior margin of
these lobes, in a position corresponding with that of the first
pair of limbs, or jaws, of Peripatus,

It is obvious that whether we consider the appendages,
the respiratory and reproductive systems, or the development
of the embryo, Peripatus is a true ArthrojDod, apparently
nearly allied to the suctorial Myriapoda,

The Myzostomata. — The genus 3Iyzostomuin^ compre-
hends certain small animals, the largest species not exceeding
one-fifth of an inch in length, which are parasitic upon the
Feather-stars. The body has the form of a flattened oval disk,
the surface of which is ciliated, while its margins may be
produced into as many as twenty short filamentous processes
or cirri. Within the margin of the ventral face are eight
suckers, four on each side, and, internal to these again, are
ten short conical " feet," five on each side ; each of these
lodges two strong setse, which can be protracted and re-
tracted in the same way as those of Annelids. Just within
the middle of the anterior margin lies a rounded aperture,
through which a muscular proboscis, the free end of which is
beset W'ith papillae, can be protruded, A straight alimentary
canal runs throufrh the bodv, and terminates in a sort of
cloaca, which opens in the middle line on the posterior mar-
gin. From each side of the alimentary canal long ramified
c£eca are given off.

No vessels or organs of circulation have been discovered.
All that is known of the nervous system is an elongated gan-
glionic mass, from which branches are given off on each side,
situated in the middle line of the ventral face of the body.

The sexes are combined in the same individual. The
acini of the generative glands are scattered through the body.
Those of the testes pour their contents into ducts, which
unite together and open by a separate vas deferens on each
side of the body, about the middle of its ventral face. The
two oviducts convey the ova to the cloacal chamber.

The development of JMyzosto'nium has been worked out
by Semper and by Metschnikoff.' The vitellus undergoes

1 Set Loven, " ArcMv fiir Katurgeschichte," 1842.

2 Semper, " Zur Anatomie und Entwickelungsgeschiclite der Gattung My-
zostoraum." {Zeitschrift fiir xoiss. Zoologie^ 1875.) "Zur Entwickelungsge-
schichte von Myzostom'um." (Ibid., 1866.)

538 THE ANATOMY OF INVERTEBRATED ANIMALS.

complete division, and the embryo leaves the egg as an oval
morula, covered with vibratile cilia. In the next stage ob-
served, the embryo is cylindroidal, and is provided with a
mouth at one end and an anus at the other. The commence-
ment of the straight and simple alimentary canal has the
form of a muscular bulb or proboscis. There are two pairs
of rudimentary appendages, each containing two setffi. The
number of the setigerous appendages increases up to five
pairs, and the intestine begins to show indications of diver-
ticula; but, in the latest stage observed, the cirri had not
made their appearance, and the body was still comparatively
narrow.

Metschnikoff regards ^fyzostomuin as a parasitic form of
a polychgetous Annelid ; and there is much to be said in
favor of this suggestion ; though, in some respects, it rather
approaches the Hirudinea.

The presence of cilia on the surface of the body and of
protractile seta3 in the parapodia excludes Myzostomnm
from the Arthropoda j while Metschnikoff has justly com-
pared its larval state with that of Syllis. Sufficient doubt,
however, still adheres to the determination of the true place
of Myzostcmum, to lead me to discuss it apart from the An-
nelids.

The Enteeopneusta. — The very singular animal JBalano-
glossus, which is the only known example of this group, is
an elongated, apodal, soft-bodied worm, with the mouth at
one end of the body and the anus at the other (Fig. 153,
III.). The mouth is surrounded by a sort of collar, or promi-
nent lip, within the margin of which springs a long probos-
cidiform median appendage, which is hollow within and has a
terminal pore. On the same side as that from which the pro-
boscis springs, the anterior region of the body presents an
elongated, somewhat flattened area, bounded by raised longi-
tudinal folds. On each side of this area is a longitudinal se-
ries of apertures — the branchial apertures. The latter com-
municate with saccular dilatations of the anterior part of the
alimentary canal, and these branchial sacs are supported by
a peculiar skeleton.

No nervous system, nor any organs of sense, have vet been
certainly made out.

According to Kowalewsky,' who was the first to elucidate

» " Anatomie des Balanoglossus." (" M(jm. de I'Acad. Imp. de St.-Peters-
bourg,'' 18G6.)

THE DEVELOPMENT OF BALANOGLOSSUS.

539

the true nature of Salanoglossus, the vascular system con-
sists of a dorsal and a ventral vessel. At the posterior end
of the branchial region the former divides into a superior and
an inferior dorsal, and two lateral, trunks. The superior
trunk passes forward, and, at the anterior end of the body,
divides into two descending branches, which unite with the
ventral trunk. The inferior dorsal trunk supplies the bran-
chiae, of which the lateral trunks are the eflFerent vessels.

For the pharyngeal branchiae of JBalanoglossus, the only
parallels to be found are among the Tunicata and the Verte-
brata. On the other hand, the larval form of this anomalous
creature is generally Annelidan or Turbellarian, with very
close and special resemblances to the Echinop^edia of some
J5c'A inoderm ata.

The young of Balanoglossris was first observed by Miiller,
who called it Tornaria, and regarded it (as did all succeeding
observers until its true nature was discovered) as an Echino-
derm-larva, on account of its extraordinary resemblance to
the larvae of Star-fishes (Fig. 153, I.).

Fig. 15^.— Balanofflossus. (After A. Agassiz.)

I. The Tornaria larva, Bide-view (about tV of an inch Ion?): a, anns; 5, vessels
leadincf to the dorsal pore (d) from w, the eac of the water-vascnlar system ; w',
prolongation of the sac; h, heart; i, intestine; s, stomach; o, cesopliasfus: m,
mouth; u, u', lobes of the alimentary canal; mb, muscular band running from
the eye-speck (e) to the water-vascnlar sac.

n. A young Balatioglossiis—leitQTB as before, except g, the first formed branchial
stigmata.

ni. A more advanced Balanog'.ossus : e, the collar; p, the proboscis.

It is an elongated ovoid body, provided with three bands
of cilia, one of which is prae-oral, while the other two are

540 THE ANATOMY OF INVERTEBRATED ANIMALS.

post-oral. Of the latter, one, at the posterior end, is circular,
while the other is inclined obliquely to the axis of the body,
so that anteriorly and superiorly it reaches the anterior ex-
tremity, -while posteriorly it occupies nearly the middle of the
body. On the ventral face a deep groove separates it from
the prge-oral ciliated band, and in this groove the mouth is
situated. The margins of the prag-oral and post-oral ciliated
bands are deeply sinuated, and they come into contact in the
median dorsal line. A wide gullet leads from the mouth, and
opens into the gastro-intesiinal portion of the alimentary
canal, which passes backward in the middle line to terminate
in the anus, at the hinder end of the body. About the middle
of the dorsal face of the body there is a circular pore (Fig.
153, I. d), whence a canal leads to a rounded sac which lies
on the junction between the gullet and the stomach. The
sac gives off two lateral short diverticula, which embrace the
oesophagus. A delicate band, apparently of a muscular na-
ture, connects the summit of the water-sac with that part of
the dorsal aspect of the body at which the prse-oral and post-
oral ciliated bands unite. Here two eye-spots are developed.
A constriction seoarates a rounded o:astric from a tubular in-
testinal division of the alimentary canal. Diverticula of the
gastro-intestinal part of the alimentary canal give rise to two
pairs of discoidal bodies, from which, apparentl}", the meso-
blast and the perivisceral cavity of the Jialanoglossus are de-
veloped.

From the sides of the oesophagus a series of diverticula
are given off, which unite with the ectoderm, open externally,
and become the gill-pouches. When only two of these bran-
chial apertures are formed, they are said by Metschnikoff to
have a striking resemblance to those of Aijpendicularia, A
pulsating vesicle — the so-called " heart " — makes its appear-
ance close to the water-sac. The anterior end of the body,
in front of the mouth, now elongates, and is converted into
the proboscis ; while the post-oral region loses its ciliated
bands, and, lengthenin<x, becomes the long body of the adult
worm.^ (Fig. 153, IT.,"^!!!.)

The Ch^togxatha. — The genus Sar/itfa,^ which is the

• See Agassiz, " The History of Balanoqlossus and Tornaria " (" Memoir?*
of the American Academy of Arts and Sciences," 1873); and Metschnikoff,
" Untersuchunofen iiber die Metamorphose einiger Seethiere." {ZeitscJirift
fur wiss. Zoologie, xx., 1870).

2 See Busk, Quarter^/ journal of Microscopical Science^ 1856. Leuckait and
Pagenstecher, " Archiv liir Anatomic," 1858.

THE CH^TOGNATHA. 541

only member of tliis group, comprises several species of small
animals which are found swimming at the surface of the ocean
in all parts of the world. Although the whole structure and
course of development of Sagitta are now very w^ell known,
its true affinities are not definitely settled. Anatomically, it
approaches the Nematoid worms and the oligochaetous Anne-
lids in some respects ; but its development presents pecul-
iarities which are as yet unknown among these animals, while
they occur among the JBracMopoda and the Ec}dnoderinata,
The body of Sagitta (Fig. 154), rarely more than an inch
long, is elongated, subcylindrical, and unsegmented ; it is en-
larged at one end into a rounded head, while at the other it
tapers to a point. There are no parapodial appendages, but
the chitinous cuticle is produced into a finely-striated lateral
fin on each side of the body and tail, and into delicate setse.
On each side of the head there are a number of strong, curved,
claw-like, chitinous processes, which can he laterally divari-
cated and approximated, and serve as jaws. Between them
is the mouth ; and at the sides of the mouth are four sets of
short but strong spines. The mouth leads into a simple and
straight intestine, which opens by an anus situated on the
ventral face of the body, where the tapering caudal region
commences. A dorsal and a ventral mesenteric band connect
the intestine with the wall of the body, and divide the peri-
visceral cavity into two chambers. Beneath the ectoderm
lies a layer of longitudinal, striated, muscular fibres. The
nervous system consists of a large oval ganglion, which lies
in the middle of the ventral wall of the body, and sends off
anteriorly two commissural cords, which unite with a supra-
cesophageal ganglion. Among other branches, this gives off
two to the dorsal side of the head ; these dilate at their ex-
tremities into spheroidal ganglia on which the eyes rest. The
ovaries are elongated tubular organs, which lie one on each
side of the intestine, attached to the parietes of the body.
Their ciliated ducts open close to the vent and are provided
with dilatations which serve as receptacula seminis. Behind
the anus the mesenteric laminae unite and form a vertical
partition, which divides the cavity of the caudal part of the
body into two chambers. On the lateral walls of these, cellu-
lar masses are developed, which become detached, and, float-
ing freely in the perivisceral fluid, are developed into sper-
matozoa. The latter escape by spout-like lateral ducts, the
dilated bases of which may be regarded as vesicula3 semi-
nales.

542 THE AXATOMY OF INVERTEBRATED ANIMALS.

Thus far, although the organization of Sagitta is very
peculiar, it presents analogies both with the Nematoidea and
with the Annelida, But its development, as described by

\—,0—^a

"i

\u%

Fig. \^^,—SaQitta Mpiinctata.—a, the head, with its eyes and appenda2:c8 ; ft, the
anus ; c, the ovary ; d, iho testicular chambers.

Kowalewsky,* is, in some respects, unlike anything at pres-
ent known in either of these groups. Yelk-division takes
place as usual, and converts the eggs into a vesicular morula,

^ " M^moires de 1' Academic Imperiale des Sciences de St.-Pctersbourff,"

1871.

THE DEVELOPMENT OF SAGITTA. 543

with a large cleavage cavity, or blastocoele. One face of the
vesicle thus constituted now becomes invaginated, with the
effect of gradually obliterating the blastocoele, and converting
the spherical single-walled sac into a hemispherical, double-
walled, cup-shaped gastrula. The cavity of the cup is the
future digestive cavity ; the layer of invaginated blasto-
dermic cells which lines this cavity is the hypoblast, which
will become the endoderni ; and the outer layer of cells is
the epiblast, and will become the ectoderm. In this condi-
tion the embryo resembles that of the Leech in its early
state. The embryo elongates, and the aperture of invagi-
nation, or blastopore, eventually ceases to be discernible.
Whether it becomes the anus, or whether the anal aperture
is formed anew, is not certain. The nervous ganglia result
from the moditication of cells of the ectoderm. The anterior
end of the primitive alimentary cavity, or archenteron, is at
first closed. It soon sends out an enlargement on each side,
so that the archenteron is divided into a central and two lat-
eral divisions. The central division opens externally and an-
teriorly by the development of the oral aperture ; and, as the
body elongates, it becomes the tubular intestine. The lat-
eral diverticula at first communicate with it, but they are
eventually shut off, and constitute the right and left perivis-
ceral cavities, their walls becoming converted into the cellu-
lar and muscular lining of those cavities. It results, from the
mode of development of the perivisceral cavity of Sagitta,
that this cavity, like the perivisceral cavity of the Brachio-
pods, and the " peritoneal " cavity of the Echinoderms, is an
enterocoele, comparable to that of the Hydrozoa and Actino-
zoa / but which, instead of remaining in communication with
the alimentary cavity, is shut off from it, its wall becoming
the mesoderm, and its cavity the perivisceral cavity.^

Nothing of this kind is known to occur in the Turhellaria,
Annelida, JVematoidea, or Rotifera ; but when a perivisceral
cavity exists in these animals, it appears always to result from

1 KowaleAvsky's account of the development o^ Sagitta has been confirmed
by Butschli,* who has further determined the origm of the reproductive or-
gans, which arise as outgrowths from the hypoblast"; and the division of each
primitive enterocoele into two sacs — one forthe head and another for the body. ^
It appears probable that the latter becomes subdivided by a transverse parti-
tion between the ovnry and testis. Bi'itschli suggests that the segmentation
of the mesoblast, which forms the walls of the enterocoele, is a point of approxi-
mation between Sagitta and the Annelids.

*"Zar Entwickelangs^eschiclite der Sa;;itta." (ZeitscJirift fur tuiss. Zoologie,
1873.)

644 THE ANATOMY OF INYERTEBRATED ANIMALS.

the excavation of the, at first, solid mesoblast. The perivis-
ceral cavity thus developed is what I liave termed a schizocoele*
But whether there is any fundamental dijBference between an
enterocoele and a schizoccele is a matter for further inquiry. I
have referred above (p. 485) to the case of an Ophiurid, in
which the hollow diverticula of the archenteron, characteristic
of the Echinoderms, are represented by solid outgrowths of
the hypoblast. From this condition there would appear to be
an easy transition to that presented by the embryos of those
Oligochceta and Hlrudinea^ in which, though the mesoblast
is a product of the hypoblast, it contains no continuation of
the alimentary cavity, but eventually splits into a visceral
and a parietal layer, the interval between which is the peri-
visceral cavity; and there is much probability in Kowalew-
sky's suggestion that the longitudinal bands (Keimstrei-
fen) in which the mesoblast makes its appearance may be
homologous with the diverticula of the alimentary cavity of
Sagitta.

In this, case, the schizoccele will be an advance upon the
enterocoele, and the development of the perivisceral cavity
in Sagitta may represent the primitive mode of development
of all invertebrate perivisceral cavities. On the other hand,
it must be remembered that between the endoderm and the
ectoderm, in the disk of a Medusa, or in the body of a Cte-
nophoran or Turbellarian, there is a gelatinous mesoderm
which occupies the position of the primitive blastocoele.
Now, this mesoderm may be, and probably is, a product of
the endoderm ; but any cavities which appear in it, such, for
example, as the water-vascular canals of the Turhellaria, can
have nothing to do with an enterocoele.

Again, in the Timicata, as w^e have seen, the atrium is a
kind of " perivisceral cavity," which is formed either by an
invagination of the ectoderm, in which case it may be termed
an epiccele j or else it is a true enterocoele. Assuming the
former alternative, for the moment, to be that which ought to
be adopted, what is called a ''perivisceral cavity " may be one
of four things :

1. A cavity within the mesoblast, more or less represent-
ing the primitive hlastocoele,

2. A diverticulum of the digestive cavit}^, which has be-
come shut oif from that cavity {enterocoele).

3. A solid outgrowth, representing such a diverticulum,
in which the cavity appears only late (modified enterocoele or
schizoccele).

THE NEMATOIDEA.

545

4. A cavity formed by invagination of tlie ectoderm
(epiccele).

And whether any given perivisceral cavity belongs to one
or other of these types can only be determined by working

out its development.

The Nematoidea. — The "Thread-worms" have elongated,
rounded bodies, which usually taper toward one or both ends;
they are not divided into segments, and they are devoid of
limbs, though they may occasionally be provided with seti-
f orm spines or papillae. In Desrnoscolex^ the papillae and setse
acquire an almost Annelidan aspect, and the annulation of
the body is much more distinct than in any other Nematoid
Worm.

Fig. loo.—Angulllula brevispinus. (After Clans.) i

I. Male. II. Female. III. Female gt-nital organs. IV. Seminal coi-pusclcs in dif-
ferent stages of development.

a, oesophagus ; a', chitinized oral capsnle ; c, gastric, and d, rectal, portion of the
alimentary canal. A, anus; gg', anterior and posterior thickenings with their
commissures ; G, sexual aperture ; F, fatty-looking irland ; r, dilatation of the
uterus, serving as a receptacniura seminis ; D, nricelhilar cutaneous glands at the
anal extremity ; Z>^ glandular mass, wiih its excretory duct above the gizzard ;
ov, ovarium ; 7>, testis ; S, seminal corpuscles.

» " TJeher einige in Humus lebende Anguillulincn." {ZeUschrift fur win.
Zoologicy xii.)

546 THE ANATOMY OF INVERTEBRATED ANIMALS.

The outermost layer of the body is a dense chilinous
cuticula, usually divisible into several layers. These layers
may be fibrillated, the direction of the fibrillation being dif-
ferent in tiie successive layers. Cilia are found neither on the
surface, nor elsewhere, at any period of life. The mouth is
situated at one extremity of the body, the anus at, or near,
the other end. The first portion of the alimentary canal is
a thick-walled pharynx, lined hy a continuation of the chiti-
nous layer of the integument, which may be raised up into
ridges or tooth-like prominences. Transverse fibres, appar-
entlv of a muscular nature, radiate from the lininor of the
pharynx through its thick wall, and probably serve to dilate
its cavity. A straight and simple tubular alimentary canal,
without any distinction into stomach and intestine, extends
through the axis of the body, a narrow oesophageal portion
usually connecting it with the pharynx.

The endoderm, or wall of the alimentary canal, consists
of a single layer of cells, disposed in few or many longitu-
dinal series ; and lined, both internally and externally, by a
cuticular layer. On each side, the intestine is fixed through
its w^hole length to the " lateral area," to be described below.
The cuticle, which lines the inner faces of the endodermal
cells, and circumscribes the digestive cavity, appears, on verti-
cal section, to be divided into rods, which are possibly merely
the intervals of minute vertical pores. In some cases, muscu-
lar fibres invest the posterior portion of the intestine.

Beneath the layers of the chitinous cuticle there is a
proper integument, or ectoderm, internal to which again is
a single layer of longitudinally-disposed muscles, which may
or may not be divided into distinct series of "muscle-cells."
The space between these and the outer face of the intestine
is occupied by a spongv or fibrous substance, which must
probably be regarded as a kind of connective tissue. The
muscles and this tissue, taken together, constitute the meso-
derm.

In the typical JVematoidea, the muscular layer does not
form a complete investment of the body, but is interrupted
along four equidistant longitudinal lines. One of them is
termed dorsal, the opposite ventral, and both these are very
narrow. The other two are much broader, and are termed
the lateral areas. They often (Fig. 156) present two or more
series of conspicuous nuclei, and each is traversed by a canal
with well-defined contractile walls and clear contents. Op-
posite the junction of the oesophageal with the gastric por-

THE NEMATOIDEA.

547

tion of the alimentary canal, each of these lateral canals passes
inward and toward the mid-ventral line, and, joining with its
fellow, opens by a pore on the exterior. In some cases, con-
tinuations of the lateral canals extend forw^ard into the head.
A ring of fibres and nerve-cells surrounds the gullet, about

Fig. 156. — Oxyitris.—a, month; 5, pharynx; c, commencement of into f tine, and d,
its termiuation. The inteimediate portion is not figured, e, j:euital aperture : /,
opening of vessels ; g, their receptacle; h. one of (he vess^els ; i, cellular matter
enveloping them. A portion of one of the contractile vessele is lepresented more
highly magnified in ihe upper figure.

the level of the opening of the water-vascular sj^stera, and
gives off filaments forward to the head, and backward to the
muscles and to the lateral area ; while two cords pass back,
along the dorsal and ventral median lines, to the hinder end
of the body. In the males of some species, nervous ganglia
have been observed in the neighborhood of the sac of the
spicula.* Organs of sense are not certainly known to exist,
unless the pigmented spots on the nervous ring of some free
Nematoids have this character.

The J^ematoidea are for the most part dioecious. In the
females, the reproductive aperture is usually placed toward
the centre of the body ; in the males, it is always situated at
or near the posterior extremity.

The female apparatus (Fig. 155, III.) consists of a vagina,
with which is connected a single, or double, elongated, tubu«
lar, organ, which tapers to a point at its blind extremity, and
is at once ovarium, oviduct, and uterus. The caecal end is

^ ^ The question of the structure and disposition of the nervous system in the
Nematoidea is, perhaps, not even yet completely decided ; but there is much
evidence in favor of vrh at is here stated. See Leuckart, "Die menschlichen
Parapiten;" the monograph of Schneider, cited beloAv ; and especially Biitschli,
"Beitrage zur Kenntniss des Nerveusystems der Nematoden" ("Archivfiir
Mikr. Anatomie," 1873).

548 THE ANATOMY OF INVERTEBRATED ANIMALS.

occupied by a nucleated protoplasmic mass. Further on, this
mass becomes differentiated into an axile cord of protoplasmic
substance — the rhachis — and peripheral masses, each contain-
ing a nucleus and connected by a stalk with the rhachis,
which are the developing ova. Still further on, in the ovi-
ducal portion of the tube, the ova become free ; while, in the
uterine portion, they are impreg'nated, and acquire a hard,
often ornamented, shell.

The testis is, generally, a single caecal tube, in the blind
end of which cells are developed, much in the same way as in
the ovary : they become free in that part of the tube which
plays the part of a vas deferens. Contrary to what happens
in most animals, these spermatozoa retain the character of
cells, and may even exhibit amoeboid movements. The defer-
ential end of the testicular tube opens into a sac close to the
anus, from the dorsal wall of which one or two curved chiti-
nous spicula are developed. These are introduced into the
vulva of the female when copulation takes place, and appear
to distend it, in order to allow of the free passage of the sem-
inal corpuscles into the vagina, and tlience into the uterus.
In the female ora-ans, the seminal cells underoro further
changes, and eventually enter into, and coalesce with, the
substance of the ova.

Yelk-division follows impregnation. The oval morula bo-
comes indented on one side, and the embryo, as it grows,
folds itself in accordance with this indentation. In most, it
would appear that the central cells of the solid morula are
differentiated from the rest to form the endoderm, which thus
arises by delamination. But B'dtschli * has recently shown
that the morula, which results from the division of the vitellus
of GucuUaniis elegans^ has the form of a flattened plate, com-
posed of two layers of blastomeres, the blastocoele being re-
duced to a mere fissure. The lamellar blastoderm next be-
comes concave on one side, convex on the other, and passes
into the gastrula form. The blastopore, at first very wide,
gradually narrows, and appears to be converted into the oral
opening of the worm. The mesoblast takes its origin frbm
certain cells of the hypoblast, which lie close to the mouth,
and grow thence toward the caudal extremity. The resem-
blance of this developmental process to that of Lumbricus is
obvious.

1 " Zur Entwickelunsrsorescluclite des Cucnllanus eJegans.'^'* {ZeiUclirift filr
wiss. Zoologie^ 1876.) Hallez (" Rovue dcs Sciences Naturelles," 1877) Has'ob-
served a similar process in Anguillula aceti^ but he denies that the blastopore
becomes the mouth.

THE DEVELOPMENT OF THE NEMATOIDEA. 549

The female reproductive apparatus is, at jSrst, represented
by a solid cellular body which lies in the mesoderm ; though
whether it originally belongs to this, or to the ectoderm, or
to the endoderm, is not clear. The cellular body acquires a
tubular form, and eventually opens externally by uniting
with an inward process of the ectoderm, which gives rise to
the vagina.

The young cast their cuticle twice — first, when they leave
the Qggt and, again, when they acquire their sexual organs.

The Kernatoidea have been divided into three principal
groups ^ — Polymyaria^ Meromyaria^ and Holoniyaria — char-
acterized by the nature of their muscular system.

In the Poly my aria, the muscles of the parietes of the
body are divided into many series, each made up of many
" muscle-cells.'* In the Meromyaria there are only eight
longitudinal series of such muscle-cells, two between each
lateral area and the dorsal and ventral lines respectively. In
the Holomyaria the muscles are not divided into series of
muscle-cells.

The first two divisions contain only such genera as an-
swer to the general description just given ; but, in the IIolo-
inyaria^ there are included several aberrant forms. Thus,
Trichocephalus has no lateral areas; Ichthyonema has no
anus ; Mermis has no anus, and the alimentary canal is rudi-
mentary, though it possesses the lateral areas, and the males
have spicula. Gordins has no lateral areas, and only the
ventral line ; the alimentary canal is reduced to a rudiment,
without either oral or anal aperture, and the male has no
spicula. In both these genera the anterior ends of the em-
bryos are provided with spines, which aid them to bore their
way into the bodies of the insects on which they are para-
sitic. In Sphcerularia the alimentary canal is similarly rudi-
mentary, and Sir John Lubbock discovered that the small
male becomes permanently adherent to the female.

Some Nematoidea (e. g., Leptodera^ Pelodera) live in
water or damp earth, and are never actually parasitic; but
the}' require abundant nitrogenous food in order to develop
their sexual organs, and hence they are found in the sexual

1 Schneider, " Monosrrapliie der Nematoden," 1866. See also Bastian,
"Monotrraph of the Aneruillulidfe'' (" Trans. Linnaean Societv," 1865>; and,
*' On the Anatomy and Phvsioloory of the Nematoids " (" Phil. Trans.," 1866) ;
and several memoirs by Biitschli. The latter affirms that the muscles are as
much made up of muscle-cells in the Holomyaria^ as in the rest. (" Giebt es
Holomvarier ? " ZeitschriftfiJr wm. Zoologie^ 1873.)

550 THE ANATOMY OF INVERTEBRATED ANIMALS.

state only among putrefying vegetable or anin^al matters.
The sexless worms, whirh live in moist earth, are at once at-
tracted hy nutriment, such as a few drops of milk/ Here
they multiply with great rapidity as long as the store of food
lasts ; but, when it is exhausted, the last-hatched young-
wander away. In the course of their wanderings, the em-
bryos enter into their larval condition ; but, before doing so,
they become twice as large as those which attain the larval
state in putrefj'ing substances. The embryonic cuticle be-
comes thickened, and its oral and anal apertures closed, so
that it forms a cyst for the larva. The larva, however, is not
restrained by this cyst from moving about and continuing its
wanderings, though, at length, it passes into a quiescent con-
dition. Its inner substance, at the same time, becomes dark
by transmitted light, in consequence of the accumulation of
small fatty granules ; and, if this state of things lasts long,
the larva dies. If the larvae should dry up, the circumstance
tends to their preservation. The embryonic cuticle is sepa-
rated, and forms a protective cyst ; and, when moistened, the
larvae resume their vital activity.

Nematoid w^orms belonging to naturally free and nonpara-
sitic genera may enter, and become encysted in, worms and
slugs ; but they only attain their sexual state when their
host dies, and they are nourished by the products of its putre-
faction.

Anffinllula scandens, the Nematoid which infests and gives
rise to a diseased condition of the ears of wheat, is a true
parasite. The young are hatched from the eggs laid by the
parent in the infected ear, and there become encj^sted. When
the wheat dies down, the larvae are set free, and wander on
the moist earth, until they meet with young wheat plants, up
which they creep, and lodge themselves in the developing
ears. Here they acquire the sexual condition, nourishing
themselves at the expense of the inflorescence, which becomes
modified into a kind of gall.

Most Nematoids found in the alimentary canal of animals
are parasitic in the sexual state, but have a longer or shorter
period of freedom as larvae or as eggs. But some, as Cucul-
laniis elegans, are parasitic both in the sexless and the sexual
condition ; inhabiting Cyclops^ while in the former state, and
sundry fresh-water fishes, particularly the Perch, in the latter.

Trichina spiralis ' acquires its sexual state in the alimen-

» Schneider, ?. c, pp. 362-' 3.

8 Leuckart, " Untersuchungen iiber Trichina spiralis,^^ 1866.

THE NEMATOIDEA. 551

tary canal of Man, of the Pig, and other mammals ; but the
young, set free in the alimentary canal, bore their way through
its walls, and enter the fibres of the voluntary muscles, in
which they become encysted in the sexless state. If the flesh
thus trichinized be eaten, the Trichinm are set free, acquire
their sexual state in the alimentary canal, and the thousands
of embryos which are developed immediately bore their way
into the extra-alimentary tissues of their host.

The insect parasites, Gordlus and Jfermis, are sexless
so long as they are parasitic ; but, when they have attained
their full growth, they leave the body of their host, acquire
sexual organs, copulate, and lay eggs. From these, embryos
proceed, which bore their way into the bodies of insects.

It has been stated that the Nematoidea are, for the most
part, dioecious. Schneider has, however, discovered certain
species of the nonparasitic genera, Leptodera and Pelodera,
which always have the external appearance of females, but in
the ovarian tubes of which spermatozoa are developed, and
impregnation takes place. This was placed beyond doubt by
isolating embryos of these Nematoids, and tracing out the
development of the spermatozoa, which result from the sub-
division of the first cells developed from the rhachis. After
a time, the development of spermatozoa ceases, and the cells
separated from the rhachis become ova, which are impregnated
by the already formed spermatozoa. These Nematoidea are
probably the most complete and necessary hermaphrodites
known in the animal kingdom.

Ascaris nigrovenosa is parasitic in the lungs of Frogs and
Toads, and attains a length of three-quarters of an inch. It
has the characters of a female, and no male has ever been met
with, but spermatozoa are developed in the ovaries in the same
manner as in the preceding forms.

The eggs of this Ascaris are discharged, and the embryos
find their way into the intestines of the Amphibian in which
they are parasitic. Here they become males and females
which are very much smaller than the hermaphrodite form
(not exceeding one-twentieth of an inch in length), and other-
wise different from it. They are evacuated with the f;isces of
the frog, and passing into damp earth or mud, the females
give rise to a few eggs. Embryos are developed from these
eggs within the body of the mother, the organs of which they
destroy, until her cuticle forms a mere case for them. The
free embryos, introduced into the Frog's mouth, pass into the
lungs, and take on the characters of the large hermaphrodite

552 THE ANATOMY OF INVERTEBRATED ANIMALS.

forms. It is not unlikely that the Guinea worm {Filaria
medinoisis), which infests the integument of Man in hot cli-
mates, may answer to the hermaphrodite stage of a similarly
dimorphous Nematoid, though its multiplication has hitherto
been supposed to take j^lace agamogenetically.

The many points of resemblance between the JSTematoi-
dea, the Oligochceta, and the Polychceta, have been indicated
by Schneider. They differ, however, from these no less than
from the Turhellaria and liotifera, in possessing only longi-
tudinal parietal muscles. In this respect they agree with
B>hampliogordius and Polygordius (united by Schneider into
the group of Gymnotoma),^ which are segmented worms,
devoid of setie, but possessing- mesenteries, segmental organs,
and pseud-hcemal vessels. Polygordius has a telotrochous
larva, and in its development, as in other respects, it is ex-
traordinarily like a polychaetous Annelid.

Butschli,* on the other hand, dwells upon the connection
between the JSfematoidea and the G aster otricha (see Chap.
IV., p. 170) and Atricha [Echinoderes)^ which he includes in
the group of Nematorhynclio.^ on the one side, and the lower
Arthropods, such as the Tardigrada, on the other.

The Physemaria. — Since the completion of the third
chapter of this work, Haeckel ^ has published an account of
certain low Metazoa^ constituting the two genera, Haliphy-
sema and Gastrophysema^ which had previously been con-
founded, partly with the Sponges and partly with the Pro-
tozoa.

These are minute marine bodies, having the form of cups
with longer or shorter stalks, by which they are attached.
The cavity of the cup into which the wide or narrow oral
opening leads is either simple (Haliphysema) or divided by
circular constrictions into two or more communicating cham-
bers ( GastropliyseY)%d). The wall is composed of two layers,
an ectoderm and an endoderm — the latter being formed by a
single layer of flagellate cells, like those of sponges ; and a
series of larger flagellate cells are disposed in a spiral, on the
inner face of the endoderm near the mouth. The ectoderm
is a syncytium, which attaches foreign bodies, such as sponge

^ See supra ^ p. 165, nnU.

2 " Untersuchungen liber freilebende Nematoden und die Gattung ChmtO'
votus.''^ {Zeitschrift fur wm. Zoologle^'' 1876.) See also Ludwig, " Ueber
die Ordnung Gastrotricha " (ibid.).

3 " Biologische Studien," Heft 2, 1877.

THE ACANTHOCEPHALA. 553

spicula or skeletons of Forarninifera^ to itself, and thus be-
comes prov'ided with an adventitious skeleton, the nature of
which varies in different species, but is constant for each.
Reproduction is effected by ova, which are said to be modi-
fied cells of the endoderm. In Gastrophysenia, the endo-
derm of the innermost chamber alone gives rise to ova. The
place of development of the spermatozoa has not been made
out.

Yelk-division is complete and regular, and gives rise to a
vesicula morula (arcliiblastula of Haeckel), each cell of which
is provided with a flagellate cilium. A gastrula arises by in-
vagination, but the final stages of development have not
been made out.

As Haeckel points out, the Thysemaria are obviously re-
lated, on the one hand to the Porifera^ and on the other to
the Ccelenterata ; in fact, they very nearly represent the
morphological common plan of which these two groups are
modifications.

The Acaxthocephala. — In their sexual state the para-
sites which constitute the genus Echinorhynchus inhabit the
various classes of the Vertehrata, while they are found in
the Invertehrata only in a sexless condition.

The Echinorhynchus of the Flounder (Fig. 157), the
structure of which may serve as an illustration of that of the
group, inhabits the rectum of that fish, which it pierces in
such a manner that the anterior extremity or head projects,
inclosed within a cyst, upon the peritoneal surface, while the
body hangs freely into the cavity of the intestine. Where
the worm traverses the wall of the rectum it presents a
much constricted neck (Fig. 157,/"). It would appear that,
eventually, the Echinorhynchi completely pass out of the
intestine, as they are found inclosed in detached cjsts lying
in the peritoneal cavity. The anterior extremity of the Echi-
norhynchus is produced into a short cylindrical proboscis,
covered with many rows of recurved hooks, and, behind this,
it forms a dilatation, in which the integument and the mus-
cular coat are separated by a considerable interval. The
body, behind the constricted neck, which separates it from
this anterior dilatation, has a thick, yellowish outer wall,
between which and the inner muscular tunic lies a system, of
vessels, consisting of two longitudinal trunks, connected by
a network of anastomosing canals.

These canals do not appear to possess distinct walls, nor
24

554 THE ANATOMY OF INYERTEBRATED ANIMALS.

are any cilia visible in them ; but the minute molecules which
float in the clear fluid which they contain are driven to and
fro, apparently by the contraction of the body. Inferiorly,

Fig. 157. — Echinorhynchus.—K.. Diagram exhibiting the relative position of the or-
gans : a, proboscis ; 6, its stem ; c, anterior enlargement of the body ; /, neck or
constriction between the anterior enlargement and the rest of tbe body, e?/ e,
posterior " funnel ;" gr, meniscus ; A, superior oblique tubular bands; yfc, inferior
muscles of the proboscis : /, m, genitalia ; o, penis, or vulva. B. Lower extrem-
ity of the stem of the proboscis : a, ganglion ; h, vascular space ; c?, outer coat ;
c, inner wall ; e, tubular band, with the nerve ; A,/, muscular bands ; ^, suspen-
Borium of the genitalia. C. Part of the female genitalia : a, ovary ; h b, ducts
leading from ovary to uterus, spermiducts (?); c, open mouth of oviduct; d, e,
uterus and vagina.

the vessels all terminate in blind canals, disposed around the
margin of the posterior funnel. Internal to the vessel lies a
double layer of anastomosing muscular fibrils, the external of
which are circular, while the internal are longitudinal.* The
cavity of the body is filled with a fluid, in which the ova, or
spermatozoa, float, and, at its anterior extremity, two elon-
gated oval bodies depend from the parietes, and hang freely in
it. These are the lemniscl / they are traversed by vessels
continuous with those of the parietes. The axis of the pro-
boscis is continued downward into an elongated subcylindrical
stem, rounded below, which hangs down like a handle into
the cavity of the bod}'. The extremity of the stem is con-
nected by broad retractor muscles with the parietes, and

' See, for an account of the remarkable structure of these muscles, Schneider,
" Ueber den Bau der Acanthocephalen." (" Archiv fiir Anatomic," 1868.)

THE ACANTHOCEPHALA. 555

gives attachment to the suspensory ligament of the repro-
ductive apparatus (Fig. 157, B). Two other bands are at-
tached a little above these, and run obliquely forward to the
parietes ; they are not mere muscles, as they are ordinarily
described, but contain a wide vessel, continuous with a large
sinus, which separates the axile portion of the stem of the
proboscis from its investing coat. In the axis of the stem of
the proboscis is the oval ganglion, which sends ofiP some small
branches upward, and two larger lateral trunks, which can
be followed into the vessels of the oblique bands ; and, in
other species, have been traced to the walls of the body and
to the genital openings. Two ganglia have been found by
Schneider in this region in the males.

There is no mouth or alimentary canal in Echinorhynchus,
the animal being probably nourished by imbibition through
the walls of the body. The reproductive organs are, both in
the male and in the female, attached by a suspensory liga-
ment to the extremity of the proboscis, and extend thence,
through the axis of the body, to the posterior extremity.
Here they open in a papilla at the bottom of a funnel-shaped
terminal dilatation of the body, which exists both in the male
and in the female, though it is much more marked, and sepa-
rated by a constricted neck from the body, in the former.
Oa each side of the papilla is an organ which has much the
appearance of a sucker, but which is apparently noncontrac-
tile, while the funnel itself undergoes constant and rhythmi-
cal contractions.

In the male the testes are two oval sacs, one behind the
other, connected by vasa deferentia, often provided with pe-
culiar accessory glands, with the genital outlet, which is pro-
vided with a long penis. In the female the ovary is a single,
long, thin-walled, cylindrical tube, the anterior end of which
is usually empty for a short distance. Further back, clear,
pale, rounded masses appear, containing cavities in which cor-
puscles, like the germinal spots of ova, lie. More posteriorly
still, these masses become elliptical, and are surrounded by a
membranous coat, which gradually thickens, and gives rise at
each end to a spiral filament which surrounds the inclosed
e^g. The ova thus constituted then pass into the cavity of
th3 body, where they accumulate in great numbers ; but, in
this species, I have not found the free floating ovarian masses
described in other Echinorhynchi. From the lower end of
the ovarium two short oviducts, or rather spermiducts, arise,
and almost immediately unite into a sort of uterus, which is

556 THE ANATOMY OF INVERTEBRATED AXIMALS.

continued into the vagina (Fig. 157, C). The uterus passes
above into a short, open, funnel-shaped canal, which lies be-
tween the two oviducts (Fig. 157, C c), and, according to Von
Siebold, takes in the ova from the perivisceral cavity by a pe-
culiar swallowing action.

The embryos of the different species of Echinorhynchi
vary somewhat in structure. Von Siebold has described those
of E. gigcts^ which are provided w4th hooks disposed like those
of the Gestoidea^ but only four in number. Sexless JEchino-
rhyjichi have been found in Cyclops and in the muscles of
fishes. Leuckart states that they acquire sexual organs in
the alimentary canal of Gadus lota. The same excellent ob-
server has succeeded in tracing the development of Echino-
rhynchus proteus, a common parasite of many river fishes, es-
pecially the Perch. ^ What appeared to be the sexless con-
dition of the same Echinorhynchus had previously been seen
by Leuckart in Gamraarus pulex. Into water containing
specimens of this Crustacean, ova from E. proteus were trans-
ferred. After a few days these ova could easily be detected
in the digestive tube of the Gaimnarus, while numerous em-
bryos, escaped from the egg-shell, were found within the ap-
pendages of the Crustacean.

Each ovum has two coats — an outer, albuminous, and an
inner, chitinous. The first is digested in its progress through
the alimentary canal ; the second is afterward ruptured by
the embryo, which bores through the intestinal walls into the
cavity of the body, and is thence conve^'ed to the site proper
for its development.

The body of the embryo is somewhat fusiform in shape,
and consists of a colorless, transparent parenchyma, protected
by a cuticle. The parenchyma may be resolved into an outer,
homogeneous, contractile layer, and a semi-fluid medullary
substance. Within this is lodged an ovoid, central mass,
made up of large, hig'hly-refracting granules. Isolated gran-
ules of the same kind may also be found scattered throughout
the soft medullary substance. At its posterior end the em-
bryo tapers to a point, while its opposite extremity is obliquely
truncated toward the ventral aspect. On this oblique surface
may be observed two series of straight spines, five (rarely six)

1 " Ueber Echinorhynehns " (" Gottinc^er Nachrichten," 1862). Results of
further investlsrations and a history of the subject are contained in Leuckart's
" Prooframm," " De statu et embryonalietlarvali Echinorhynchorura eorumque
inetamorphosi," 1873; and, further, in the concludins: part of "Die mensch-
lichen Parasiten," 1876, which has reached me too late for use in this place.

THE DEVELOPMENT OF ECHIXORHYNCIIUS. 557

in each. The two series meet near the middle line to form
an arch, the central and largest spine constituting its summit.
Two short, ridge-like elevations of the cuticle, close to the
middle line, separate the spines on either side from one an-
other. Behind, the peripheral layer gives rise to a knob-like
process.

At the end of fourteen days, the embryo is found to have
increased much in size, but presents few changes of form.
The anterior extremity displays two rounded elevations, the
spines retaining their original position. The peripheral layer
has become thicker and more distinct ; its knob-like process
has by this time disappeared. The central mass, now much
larger, has assumed a spherical figure. No longer granular,
it is seen to be composed of numerous pale cells, which con-
tinue rapidly to increase.

During the third week, numbers of large yellow granules
begin to appear within the outer Ia3^er of the embryo. No
other changes, save those of growth, take place in its walls :
but the central mass, still continuing to enlarge, gradually
puts on the aspect of a young Eddnorhynchus. This mode
of development has been compared by Leuckart to that of
certain Echinoderms, or to the production of the Nemeitid
larva within its pilidhon.

The first part to become differentiated is the cavity of the
future proboscis, which appears as a transparent lenticular
vesicle at the anterior end of the spherical mass. Behind this
are soon seen rudiments of the central axis and its contained
ganglion ; and the suspensorial ligament, with the reproduc-
tive organs, are, at the same time, marked out. The muscles
of the outer wall have also commenced their development.
Next, the central region of the young jLchinorhynchus rapid-
ly elongates ; its walls become thinner, and, separating from
the included structures, show the first trace of the visceral
cavity. About this time distinctions of sex first make them-
selves evident. The posterior end of the body undergoes a
disproportionate increase of size, the muscles become mere
distinct, and the rudimentary generative organs are clearly
manifest. At length the young Echinorliynchus occupies
almost the Vv^hole interior of the embryo, the walls of which
have, meanwhile, undergone but slight histological change.
The spines, however, have disappeared, together, it would
seem, with the cuticle to which they were attached. No rup-
ture of the other embryonic structures takes place, but they
gradually attach themselves to the body of the contained

558 THE ANATOMY OF INVERTEBRATED ANIMALS.

Echinorhynchus^ becoming closely fitted to its surface, and
apparently persisting throughout its entire life. The devel-
opment of the Echinorhynchus now approaches completion.
The lemnisci appear. Hooks arise on the surface of the pro-
boscis, not, as might be supposed, from its outer cuticle, but
from specially modified cells of an inner membrane. The in-
ternal organs begin to assume their final aspect. The ex-
ternal form of the adult organism is rather slowly reached,
and a few changes which take place after transference of the
Echinorhynchus to its final host have yet to be observed.

The Acanthocephala undoubtedly present certain resem-
blances to the N^ematoidea, and more particularly to the Gor-
diacea, but the fundamental differences in the structure of
the muscular and nervous system, and in that of the repro-
ductive organs, are so great, that it is impossible to regard
them as Nematoids which have underg-one a retroo^ressive
metamorphosis. In their case, as in that of the Cestoidea
and that of the Dicyemida^ it is, I think, desirable to keep
one's mind open to the possibility that anenterous parasites
are not necessarily modifications of free, enterate ancestors.

The DiCYEiiiDA. — In 1830, Krohn discovered certain cili-
ated filiform parasites in the renal organs of Cephalopods,
to which KoUiker subsequently gave the name of Dicyema,
Recently, these strange organisms have been made the subject
of renewed investigation by E. van Beneden, from whose
elaborate memoir ^ I take the followinor account of their
structure :

The body of a Dicyema (Fig. 158, 1.) consists of one large,
cylindrical, or more or less fusiform, axicd cell^ which extends
from the slightly-enlarged head-end, by which the animal is
attached, to its posterior extremity, and is invested by a
single layer of relatively small flattened corticcd cells. These
are arranged, like a pavement epithelium, around the axial
cell, their edges being juxtaposed ; they are nucleated, and
their free surfaces are ciliated. There is no interspace be-
tween the cortical cells and the axial cell, and the organism
is a simple cell-aggregate, devoid of connective, muscular, or
nervous tissues.

The cortical cells which invest the anterior or head-end
of the Dicyema hav^e peculiar characters, and are distin-
guished as the polar cells. They are arranged in such a

1 " Rechcrclics sur les Dicyemides." (" Bulletin de I'Acad. Eoyale de Bel-
gique," 1876.)

THE DICYEMIDA.

559

manner that the head is bilaterally symmetrical. Sometimes
the polar cells constitute the whole of the cephalic enlarge-
ment ; but, in others, cells of the adjacent part of the body

Pig. 158.— Dicyema. — I. Z>. typus. The larsre papillae of the cortical layer and the
germs in the Interior of the axial cell are noticeable.

II. 7>. typus. Difl'erent stages of the development of a vermiform germ.

III. Infusoriform embryo found free in the renal organs of Eledo/te rfwschata. trented
with osmic acid: p, the urn; ca, its capsule; s, its lid; i, muliiuucleate cells in its
interior. (After Van Beneden, /. c.)

(parapolar cells) contribute to the investment of the head.
Strongly-refracting globules and rods accumulate in some of
the ectodermal cells, and cause them to project in the form
of papilljB.

The axial cell is a mass of protoplasm. Its relatively
dense outer la3'er passes into a central reticulation, in the
midst of which there is a large oval nucleus.

Reproduction takes place by the formation of germs, and
the development of embryo from them, in the axial cell. The
embryos are of two kinds, the one vermiform, the other in-
fusoriform, and are not met with in the same Dicyema, but in
individuals of somewhat different characters. Those which
give rise to the vermiform embryos are termed Neinatogena^
while the others are named Mhomhogetia.

In the N^ematogena, the germs arise in the protoplasmic
reticulum of the axial cell, and, at first, are minute spherical
bodies, each of which is provided with a nucleus. This germ-
cell divides into two, and each of these again becoming bi-
sected, four cells are produced, of which one remains undi-
vided, while the rest go on dividing. The former enlarges,
and gives rise to an axial cell, around which the other cells
arrange themselves, until eventually they inclose it. Before

560 THE ANATOMY OF IXVERTEBRATED ANIMALS.

they meet, they surround an opening through which one end
of the axial cell protrudes. This corresponds with the oral
pole.

Before the young Dicyema thus developed leaves the
body, which it generally does by traversing the oral pole
(though it may make its way out through the parietes), two
embryos of the same kind appear within its axial cell.

Thus the nematogenous Dicyema gives rise by agamo-
genetic process to new Dicyemas.

In the Hhoinhogena the germs are developed in from two
to five special nucleated parent cells, the origin of which is
not known. They are found imbedded in the protoplasm of
the axial cell, and the germs are developed endogenously
from the protoplasm of the parent cell, the nucleus of which
remains unchanged. The germs undergo division, and be-
come spheroidal bodies composed of two kinds of cells, small
and large. Each of these bodies is converted into an infu-
soriform, bilaterally symmetrical embryo, which consists of
an urn, a ciliated body, and two refractive bodies.

The urn, situated on the ventral side of the embryo, is
composed of a capsule, a lid, and contents.

The latter are four granular masses, each of which con-
tains many nuclei, and eventually becomes covered with cilia.
The refractive bodies take their origin in two adjacent cells.
Thev partially cover the urn in front, and form the largest
portion of the dorsal face of the embryo. The ciliated body
consists of ciliated cells, and forms the caudal portion of the
embryo.

While the vermiform embryo becomes a Dicyema in the
body of the Cephalopod on which its parent is parasitic, the
infusoriform embryo is set free, and probably serves as the
means by which the parasite is transmitted from one Cepha-
lopod to another.

Professor E. van Beneden compares the cortical layer of a
Dicyema to the ectoderm, and the axial cell to the endoderm
of a Metazoon ; and the mode of production of the embryo
to the process of epiboly in the Metazoa. But, from the
complete absence of any mesoblastic layer, he proposes to
establish a new division of Mesozoa, intermediate between
the Protozoa and the Metazoa, for the Dicyemida.

CHAPTER Xri.

THE TAXOIS'OMY OF IXVERTEBEATED ANIMALS.

The grouping of the various kinds of invertebrated ani-
mals which has been adopted in the preceding pages is to be
regarded merely as a temporary arrangement. Each chapter,
from the second to the tenth, is devoted to a series of forms,
the morphological relations of which are more or less obvi-
ous, while Chapter XI. is reserved partly for such groups as
do not readily find a place in any of the series which precede
them ; and, partly, for such as have been established since
this work was commenced.

Oar knowledge of the anatomy, and especially of the
development, of the Invertebrata is increasing with such pro-
digious rapidity, that the views of Taxonomists in regard to
the proper manner of expressing that knowledge by classifi-
cation are undergoing, and, for some time to come, are likely
to undergo, incessant modifications.

To the beginner, who is apt to make the mistake of look-
ing upon classification as the foundation and essence of mor-
phology, instead of what it really is, the superstructure and
outcome thereof, this state of things is distressing. Every
hand-book presents him with a different system of classifica-
tion, and he may, not unnaturally, despair of finding any
stability in a science, the most general results of which are
capable of being stated in such very diff"erent ways. If, how-
ever, the student will attend to the facts which constitute the
subject-matter of classifications, rather than to the modes of
generalizing them which are expressed in taxonomic systems,
he will find that, however apparently divergent these systems
juay be, they have a great deal in common.

It Is possible to divide invertebrated animals into a certain
number of groups, each of which will be admitted by everj^
morphologist to be in itself a perfectly natural assemblage.
That is to say, all the forms thus associated together will re-

5G2 THE ANATOMY OF INVERTEBRATED ANIMALS.

serable one another, and will differ from all other animals in
certain respects. Each such assemblage is, in fact, a " nat-
ural order " in the sense in which that word is used by bota-
nists ; and, although the number of these natural orders may
be increased by the discovery of new forms, or diminished by
the ascertainment of closer bonds of union than are at present
known to exist between the orders already discriminated,
yet, the morphological types which they represent will al-
ways remain ; and, therefore, the knowledge of their charac-
ters, once acquired, will be a permanent possession.

It is not needful that these natural orders should be mor-
phologically, still less numerically, equivalent ; and, in form-
ing them, it is more important that similarities should not be
neglected, than that differences should be overlooked. Those
w^hich have been recognized in the preceding pages are enu-
merated in the following list, arranged in sections correspond-
ing with the chapters in which thev are discussed. Under
the head of each section I shall proceed to make such obser-
vations as have been suggested to me by new information or
by further reflection, during the progress of this work.

Section I. — 3Ionera \Foraiiiinifer(i\ [ITeliozoajy JRadio-
laria, Protoplasta^ Gregarinidm^ Catallacta^ Infusoria [ Ojki-
linina, Ciliata^ Flagellata, Tentaculiferd].

Section II. — Porifera, Sydrozoa, Coralligena [Cte7io-
pJiorci],

Section III. — Turhellaria^ Rotifer a l^N'ematorhynchd]^
Trematoda, Cestoidea.

Section IV. — Hirudinea, OUgocJioeta, Polychceta, Gephy-
rea.

Section V. — Crustacea^ Arachnida \Pycnogonida., Tardl-
grada^ Pentastomidd], 3Iyriapoda, Insecta,

Section VI. — Polyzoa, Brachiopoda^ Lamellihranchiata,
Odontophora.

Section VII. — JEchinodermata,

Section VIII. — Tunicata.

Section IX. — Peripatidea, 3Tyzostomata^ Enter opneusta^
Chmtognatha^ Nematoidea^ Physemaria, Acanthocephala^
Dicyemida.

Section I. — In the commencement of Chapter II., I have
expressed a doubt as to the validity of the distinction of the
groups contained in this section by the presence or absence

THE HELIOZOA. 5^3

of a nucleus, and the recent investigations of Schulze * and
Hertwig " have justified my hesitation. These observers have,
in fact, demonstrated the existence of one or more nuclei in
many Foraminifera {Entosalenia, Polystomella, JRotalia,
Textularia, some Miliolidce). These nuclei may be simple
or multiple ; in the latter case, they have no special relation
to the cameration of the skeleton, and they are single in the
young.

The discovery of the nuclei was effected by treating the
Foraminifera in which they were found in a special manner;
and, considering the negative results at which the best ob-
servers of the Foraminifera have hitherto arrived, and the
fact that the other Monera have not been investigated by
the same methods, it will probably be wise to consider the
question of the nonexistence of a nucleus in them as an open
one.

Hertwig proposes to include all the Rhizopods which are
invested by a coat of chitin, or by siliceous or arenaceous par-
ticles, or which possess a skeleton, under the head of Thala-
mophora y but the name of Foraminifera is now so widely
accepted and so long established that I cannot but think that
the better course is to retain it.

I have included the Acti^ioj^hryida and the similar forms
found in fresh water, and provided with Radiolarian skele-
tons, with the marine Madiolaria.

Hertwig and Lesser,^ however, in their important mono-
graph upon the Rhizopods, have stated reasons for separating
the former as a distinct group (the Seliozoa of Haeckel),
though their conclusion that there are, at present, no grounds
for assuming even a remote relation between the Heliozoa
and the Hadiolaria {I. c, p. 159) appears to me to have no
sufficient warranty.

The Heliozoa are defined by these authors to be unicellu-
lar organisms, which occasionally become multicellular, or at
any rate multinucleate, by the multiplication of the nucleus.
They are usually spheroidal and free, but some are fixed by
means of a stalk. In most, the protoplasm of which they con-

1 " Rhizopoden-Studien, VI." (" ArcMv fur Mikr. Anatornie," 1876.)

2 " Bemerkungen zur Organisation uud systematisclien Stellung der Forami-
niferen." {Jenaische Zeitschrift^ 1876.)

3 " Ueber Rhizopoden und denselben nahestehenden Organismen." (" Ar-
chiv fiir Mikr. Anat.," Bd. x., Supplementheft, 1866.) Full references to the
literature of the subject will be found in this memoir and in Dr. Carpenter's
" Introduction to the Study of the Foraminifera," 1862.

564 THE ANATOMY OF IXVERTEBRATED ANIMALS.

sist is differentiated into a cortical and a medullary substance
(ectosarc and endosarc). Tiie sharpness of demarcation of
the ectosarc from the endosarc varies. In Actinophrys sol
the two pass, imperceptibly, one into the other ; in Actino-
sphaerium^ the change from the ectosarc into the endosarc
takes place within a narrow zone, everywhere equidistant
from the centre. The line of separation between the endo-
sarc and the ectosarc is best defined in the Acanthocystidce^
Heterophryidce, etc., but it arises only from a differentiation
of the protoplasm, and not from the development of a defi-
nite membranous investment around the endosarc. The nu-
clei lie in the endosarc. When only one exists it is usually
eccentric, and, when there are many, they are scattered irreg-
ularly. The ectosarc contains contractile, and sometimes non-
contractile, vacuoles, which last may also be met with in the
endosarc. The pseudopodia are thin, filiform, and radiate
from the body ; sometimes their surface presents moving
granules. They rarely branch or anastomose. In many cases
they present an axial substance which may be traced as far as
the endosarc. The sileoious skeleton may consist of separate
spicula or form a continuous shell.

The Heliozoa propagate by simple division wither with-
out previous encystation ; and the products of division may
or may not become encysted. They may either pass directly
into the adult state or become monadiform active larva?, pro-
vided with two flagella, a nucleus and contractile vesicle,
which in course of time develop into the parent form.^

A completely new light has been thrown upon the vexed
qu3stion of the supposed sexual method of reproduction of

» As this chapter was passingf through the press, Hcrtwi<y's monograph " Zur
flistolo^^ie der Radiolarien " has come into my hands. The Radiolaria are de-
fined as Khizopods with pointed, branched, usually anastomosini? and granular
pseudopodia, which proceed from a protoplasmic body inclosing either nu-
merous small heterogeneous nuclei, or a sincjle larger highly-differentiated ve-
sicular nucleus. The protaplasm of the body is further separated into a pe-
ripheral non-nucleated and a central nucleated portion, by a membranous capsule
with porous walls. The capsule is invested by a homogeneous gelatinous sub-
stance; the extracapsular protoplasm usually contains numerous vellow cells.

Propagation is effected (probably always) by the breakinor up of the bodv
into unicellular monadiform embryos provided with a sinsrleflasrellum. As a
result of these investigations, Hertwig admits that the Radiolaria and the
Heliozoa are closely allied, and even suggests that the name of Radiolaria
should apply to both groups, which would then form the subdivisions of Hdrozoa
and Cytophora. The Radiolaria {(hjfopJiora^ are distinguished into Collozoa
(with numerous small nuclei) and Collida with a single highly-differentiated
nucleus.

THE KEPRODUCTIOX OF THE INFUSOHIA. 565

Infusoria by the investigations of Engelmann,^ Biitschli,'
and Hertwig,^ the results of whose observations may be
summed up as follows :

1. The so-called acinetiforra embryos are parasites.

2. The rod-like bodies occasionally observed in the endo-
blast are also parasites, and probably JBacteria.

3. The globular so-called germs in the VorticelUdce and
the bodies termed "ovules" by Balbiani have nothing to do
with reproduction.

4. In the VorticelUdce, when conjugation takes place, the
endoplasts of both individuals break up into a number of
fragments. These become mixed up in the common body
which results from conjugation. The endoplast of the latter
results from the gradual union of many smaller particles
which make their appearance in the endo?arc. Whether
they are identical with the fragments into which the endo-
plasts of the conjugated individuals have divided, is not cer-
tain.

5. When Infusoria which possess an cndoplastule, as
well as an endoplast, conjugate, both of these structures un-
dergo division ; and the eudoplastule, before division, ac-
quires the striated structure and spindle shar)e, which has ob-
tained for it the name of " seminal capsule."

6. The fmal result of conjugation is the appearance in
each of the individuals which have undergone conjugation
of the endoplast and endoplastule (either single or multiple)
which characterize the species.

It does not appear that there is any positive proof that
the striated endoplastule, or endoplastules, of the conjugated
individuals are or are not exchanged. From Biitschli's obser-
vations on Stylo7iichia mytilus, he concludes that the endo-
plast divides into four fragments ; that these round them-
selves off into the so-called " ovules " of Balbiani, and are
expelled from the body ; while, of the four striated endoplas-
tules into which the endoplastules which exist before fecunda-
tion divide, one is converted into a large transparent body,
and, dividing, gives rise to the two new endoplasts which ap-
pear in the Stylonichice, after their separation. Two of the
others become the new endoplastules ; while one, apparently

^ " Ueber Entwiclcelunsr und Fortpflanzting der Infusorien." (" Slorpho-
logisches Jahrbuch," 187G.)

2 " Mittheilunoren fiber die Conjugation der Infusorien und die Zellthei-
lung." (ZeitschHft fur wiss. Zoolo'gie^ 1875.)

3 " Ueber Podopkrya gemmipara^ nebst Bemerkuntren zum Bau und zur
SYstematischen Stcllung tier Acinetcn." ("Morph. Jahrbuch," 1876.)

566 THE ANATOMY OF INVERTEBRATED ANIMALS.

undergoing retrogressive metamorpliosis, is cast out of the
body.

From these facts, and from the circumstance that the en-
doplastules of Infusoria^ which are merely dividing, acquire
the striated structure, it must be concluded that the ascrip-
tion of a spermatozoal nature to the striae of the modified en-
doplastules is not warranted. And the remarkable observa-
tions of Biitschli, Strassburger,* Van Beneden, and Hertwig,"
on the changes which take place in the nuclei of both animal
and vegetable cells, which are undergoing division, or are pre-
paring for fecundation, seem to leave no doubt as to the jus-
tice of this negative conclusion. In such cells the nucleus
becomes elongated and assumes a striated appearance, so as
to resemble in a very striking manner the so-called " seminal
capsule " of the Infusoria. Nevertheless, it is still possible
thafc the conjugation of the Infusoria may be a true sexual
process ; and that a portion of the divided endoplastules of
each may play the part of the spermatic corpuscle ; the con-
jugation of which with the nucleus of the ovum appears, from
recent researches, to constitute the essence of the act of im-
pregnation.

"With the proof that the " acinetiform embryos " of the
Infusoria ciliata are parasites, the view of the relations of
the Tentaculifera with the Ciliata^ suggested at p. 101,
ceases to be exactly tenable. Nevertheless, the resemblance
of the ciliated young Aeinetce to the simpler forms of the
Ciliata is so close that they may still be said to be modifi-
cations of a common type. Hertwig ^ has made the interest-
ing observation that, in some Aeinetce, the tentacula are of
two kinds : those of the one kind are the characteristic suc-
torial organs, while those of the other kind are simply pre-
hensile, and have a structure very similar to that of the pre-
hensile pseudopodia of the Actinophryidoe. The same au-
thor shows that the ciliated germs do not arise from the en-
doplast alone, but that a portion of the protoplasm of the
body invests each division of the endoplast. In fact, the pro-
cess by which these germs are developed is altogether similar
to ordinary cell-division.

1 " Ueber Zellbiklung und Zclltheilung," 1876.

2"Beitri1fire zur Kenntniss und Bildung. Befruchtung und Theilung des
thierischen Eies." ("Morphologisches Jahrouch," 1876.)

3 " Ueber PodnpJirya gemmipara nehst Bemerkungen zum Bau und zur sy-
Btematischen Stellung der Acineten." (" Morphologisches Jahrbuch," 1876.)

THE DEVELOPMENT OF SPONGES. 567

The Opalini?ia must clearly be arranged among the Infu-
soria. Stein regards them as simply the lowest forms of the
IToIotncha, but it will probably be safer to consider them as
a distinct group, standing in somewhat the same relation to
the Clliata as the Gregarmidce do to the Amoebce.

Sectio]^' II. — The elucidation of the problem of the mode
of development of the Sponges has been greatly advanced
by the investigations of Oscar Schmidt/ Schulze/ and espe-
cially of Barrels,^ which confirm the assertion of Metschnikoff
that the vesicular morula which constitutes the early condi-
tion of the sponge-embryo consists of blastomeres of two
kinds; those of the one-half of the spheroidal or flattened
embryo being elongated and flagellate ; those of the other,
rounded, granular, and nonciliated. Schulze and Barrois
have independently ascertained that the latter region some-
times undergoes partial invagination ; and that a cup-shaped
body is produced, composed of an epiblast formed of flagel-
late cells and a hypoblast of spheroidal, non-ciliated cells.
Thus the " gastrula " stage of Haeckel may exist, though it
is not formed by delamination, as he supposed, but by invagi-
nation. But it appears that this gastrula-stage does not
always occur, and that when it does, it is transitory, in so far
as the hypoblastic cells subsequently enlarge, protrude be-
yond the epiblastic cells, and give rise to the free ovate em-
bryo formed of a ciliated and nonciliated half, which has so
often been observed. According to Barrois's observations,
this free swimming larva fixes itself by its nonciliated hypo-
blastic half, and the hypoblastic cells are invested by those of
the epiblast, which thus constitute the whole outer covering
of the young sponge. The central cavity of the sponge,
which represents the archenteron, arises in the midst of the
included hypoblastic cells, while the osculum is a secondary
opening, formed apparently by an invagination of the ecto-
derm, and has nothing to do with the primitive blastopore.
Thus even the simplest sponge has passed bej'ond the gas-
trula-stage.

Schulze has made the important discovery that, in S^-

^ 1 " Zur Orientirung liber die Entwickelung der Spongien " {Zeifschriftfi/r
wiss. Zoologie^ 18751; and " Nochmals die "Gastrula der Kalkschwiimme "
(" Archiv f. Mikr. Anat.," 1876).

2_"Ueber den Bau und die Entwickelung von Sycandra raphanus'''' {Zeit-
Bchrlft fur wiss. Zoologie, 1875) ; and " Zur Entwickelungsgeschichte von Sy-
caudra" (ibid., 1876).

3 " Annalea des Sciences Naturelles," 1876.

568 THE ANATOMY OF INVERTEBRATED AXIMALS.

candra raphcmiis, there is a layer of flattened cells external
to the syncytium ; whence the latter may rather be regarded
as the equivalent of the mesoderm than of the ectoderm of
the Godenterata. And the observations of Barrois on other
calcareous sponges tend to the same conclusion. The care-
ful investigations of the last-named writer have not enabled
him to discover spermatozoa in any sponge, and he finds that
the ova, when they are first discernible, are situated in the
syncytium or mesoderm, and not in the endoderm. In the
free larvae of the calcareous sponges an equatorial zone of
rounded equal-sized blastomeres is interposed between the
ciliated, or epiblastic, and the nonciiiated, or hypoblastic,
hemisphere ; and it appears probable that these cells repre-
sent a mesoblast, and give origin to the mesoderm. The
embryo in this condition has a very interesting resemblance
to that of Clepsine^ in the stage in which the epiblast occu-
pies one face of the embryo, and the hypoblast, formed of
three very large blastomeres, the opposite face ; while an in-
complete zone of six or eight large blastomeres, which are
eventually inclosed by the epiblast, surrounds the margins of
the latter.

At p. 135, I have quoted Haeckel's account of a pro-
cess of JEntogastric gemmation in Cannarina hastata of an
altogether anomalous character.

F. E. Schulze ^ has lately investigated specimens of Gery-
onia hexaphylla provided with entogastric processes beset
with budding Giiniyim^ and he proves that, in this case, at
any rate, the phenomenon is one of parasitism. The stem
from which the buds proceed, in fact, is not a process of the
body of the Geryonia, but is simply attached to the wall of
the gastric chamber of the latter. It is hollow, and its cavi-
ty is lined by an endodermal epithelium. The Gunina buds
are not developed from the epithelium which covers the stem
and represents its ectoderm, but commence in the ordinary
way, as csecal diverticula of the wall of the stem, the apices
of which soon open to form the hydranth of a medusoid, the
disk of which results from the outgrowth of the base of the
hydranth. In all probability the larva of the Gunina enters
the gastric cavity of the Geryonia as a planula ; and, attach-
ino: itself to the wall, c:rows out into a stolon whence the me-
dusoids bud.

^ "Ueber die Cuninen-Knospeniihren im Maeren v. Geryonien." (" Mit-
theilungen des Naturwissenschaftlichcn Vcreines." Gratz, 1875.)

THE DEVELOPMEXT OF CLEPSINE. 569

It may be suspected that the other cases of supposed en-
togastric proliferation will prove to be susceptible of a similar
explanation.

Although^ as I have endeavored to show, the Ctenophora
are readily reducible to the general plan of the Acti?iozoa., yet,
considering their many peculiar characters, I think it is ad-
visable to separate them from the CoralUgena^ as a distinct
natural order.

Moreover, the Physemaria must undoubtedly be placed
in this section, which will, therefore, consist of the following
natural orders : Fhysemaria, Porifera, Ilydrozoa., Coralli-
gena^ Ctenophora,

Section III. — I concur in the proposal of Biitschli * to es-
tablish a group, Kematorhynclia^ for the genera Chatonotus^
Echinoderes^ and their allies, to which reference is made at p.
101. The Kernatorhyncha are divisible into the Gastrotricha ^
(Chc^tonotiis, Chcetura^ Cephalklium^ Ichthydhnn^ Turhanel-
la, Semidasys, and Pasydites), which are ciliated on the
ventral surface of the body, and the Atricha (Echinoderes)^
which possess no cilia. Biitschli finds two convoluted water-
vessels analogous to those of the Potifera, but apparently
not ciliated, in Choetonotus.

Sectiox IY. — Our knowledge of the development of the
Ulrudinea has received an important addition in the " Me-
m.oire sur le developpement embryogenique des Hirudinees,"
by M. C. Robin ; who, among other important contributions
to embryology, has rectified some important errors of Rathke
resj^ecting the early stages of the development of Clepsine.
I have found the description and figures of the various stages
of cleavage, and of the steps by which the blastoderm is con-
verted into the young Clepsine^ given in this memoir, to be
exceedingly accurate.

The whole process in Clepsine is very similar to that wdiich
has been described in Euaxes by Kowalewsky,^ and shares
with it the remarkable peculiarity that the first-formed por-
tion of the blastoderm becomes the haemal region of the body.

1 " Untersiicliuno^en fiber frellebende Neraatoden und die Gattung Ckaitono-
tus.'''' {Zeitschrift^fiir wiss. Zoolo(fie^\%'l'o,)

2 See H. Ludwig, " Ueber die Orduung Gastrotriclia." {Zeitschrift fur wiss.
Zoologw, 1876.)

3 " Embrvolooriscbe Studien an Wurmem und Arthropoden." (" Mem. do
1' Acad. Imp." de St.-Fetersbourg," 1871.)

570 THE ANATOMY OF IXVERTEBRATED ANIMALS.

As this blastodermic disk grows, its margins thicken and give
rise to two germ-bands (Keimstreifen). These gradually ap-
proximate and eventually unite upon the opposite face of the
ovum. As the chain of ganglia is the product of the differen-
tiation of the epiblast of the germ-bands, it follows that it is
formed by the union of two primarily distinct nerve-tracts,
which move round from the hsemal to the neural aspect of the
bodv ; and thus the arrano-ement of the nervous trunks in
Malacohdella ^ may be regarded as expressive of a condition
which is transitory in Clepsine and Euaxes,

Many years ago "^ I directed my attention to the fact that
" the development of a Mollusk commences on the haemal
side and spreads round to the neural side, thus reversing the
process in Articulata and Vertebrata ; " and it is very inter-
esting, considering the many curious points of approximation
between the A.nnelida and the Mollusca which are now com-
ing to light, to observe that certain Annelids present this
especially MoUuscan peculiarity.^ As Yon Baer long ago
pointed out, there is a striking likeness between the foot of a
Gasteropod and the suctorial disk of one of the Ilirudinea,
The so-called jaws of the Leeches (the " teeth " of which, I may
observe in passing, are calcified) are curiously similar to an
odontophore devoid of cartilages, the representative of the
radula being supported on a muscular cushion.

The statement at p. 215, that " no calcareous skeleton
is found in any of the 6^5/)Ayrea," ceasss to be true since the

» According to Semper's recently-published statements, Malacohdella is a
true Nematoid, and not a Leech. C' Die Verwandtschaftsbeziehungen der
ges^liederten Thiere," " Arbeiten aus d. Zoolo^isch-zootomischen Institut in
Wiirzburg," Bd. iii., 1876.) The memoir here cited is full of important obser-
vations respecting the structure of the nervous system in the Annelida ; the
agamogenetic multiplication oi Nais and CTiodog aster ; and the development of
the organs of these Annelids.

Moreover, the autlior discusses very fully the relation of the Annelidan with
the vertebrate types of organization. 1 do not propose to touch upon this subject
in the present volume ; but I may remark that the evidence upon which the
identification of the structures termed " Kiemengangwulste " and their products
with the branchial apparatus of vertebrate animals is founded, appears to me
to be whollv inadequate to bear out the conclusions deduced from it.

«"0n the Morphology of the Cephalous Mollusca." ("Phil. Trans.,"
1852, p. 45 and note.')

3 The mode of development of the central nervous system in Euaxes and
Clepsine offers many points of interest. Not the least important of them is the
obvious similarity (to which attention has already been directed by Semper)
between the germ-bands of Clepsine when they have united throughout the
greater part of their length, but surround the blastopore behind, and the Am-
phibian embryo with its dorsal ridires, which have exactlv similar relations.
{See., for example, Fig. 40, in Plate IIL of GOtte's work, " Die Entwickelungs-
geschichte der Unke.")

CH.ETODERMA, NEOMEXIA, AND CHITON. 5 71

discovery of L. Graff,^ that the minute spines of Chcetoderma
are calcified. It is a further peculiarity of this genus that
two distinct nerve-cords proceed from the cerebral ganglia
parallel with one another on each side of the body, in the
place of the single median nerve-cord of other members of the
group.

Dr. Jhering ^ has directed attention to certain points of
resemblance between Ghcetodernia^ with the allied genus
Neoruenia^ and the Chitons, especially in the arrangement of
the trunks of the nervous system ; and he proposes to unite
the three into a group of Amphineura — thus separating the
Chitons from the Mollusca altogether.

Section V. — I regret that I have been unable to make use
of Claus's recently-published important contributions to the
history of the development of the CrustacecC

Section VI. — The thorouofh examination of the structure
of Pedicellina and Loxosonia by Nitsche * has shown that
the differences between the ectoproctous and the endoproctous
Polyzoa are of a more fundamental character than had been
suspected. In the Ectoprocta^ in fact, the endocyst consists
of two layers, an outer and an inner, of which the former is
the representative of the ectoderm in other animals. The lat-
ter lines the wall of the " perivisceral cavity," and is reflected
thence, like a peritoneal tunic, over the tentacular sheath and
into the interior of the tentacula, whence it is continued on to
the alimentary canal, of which it forms the external invest-
ment. The endoderm, which lines the alimentary canal, is,
of course, continuous, through the oral opening, with the ec-
toderm.

In the Endoprocta^ on the contrary, the endocyst is com-
posed of only one layer, and the endoderm of the alimentary
canal has no second or external coat. The " perivisceral
cavity," or interspace between the endoderm and ectoderm,
is occupied by ramified mesodermal cells.

Thus the Endoprocta present a structure as simple as that

* " Anatomie des Ckoefoderma nitidulumy (Zeitschrift fur wiss. Zoologie,
1870.)

2 " Vergleichende Anatomie das Nervensysteras der Mollusken," 1877.

3 " Untersuchunsren zur Erfor.schuug der genealogischen Ginindlage des
Crustaceensystein<<," 1876.

* " Beitrage zur Kenntniss der Brvozoen." ( ZeUftcJirift fur wiss. Zoologie^
1870 and 1875.) Compare Barrois (" 'Comptes Eendus," 1875).

573 THE AXATOMY OF INVERTEBRATED ANIMALS.

of Nematoid worms ; while the Ectoprocta^ in possessing a
perivisceral cavity with a special lining, the inner surface of
which may be ciliated, are, so far, comparable to Brachiopods
or Echinoderms.

Unfortunately, our knowledge of the embryonic develop-
ment of the ectoproctous Polyzoa does not enable us to de-
termine with certainty the nature of this perivisceral cavity,
and of the layer which bounds it. Nitsche shows that the
saccular cystid, which results from the first developmental
changes of the embryo in the Phylactolamata, is composed
of two layers, which correspond with those of the endocyst
in the adult ; and, further, that the polypide (alimentary
canal, tentacula, and ganglion) results from an ingrowth of
the outer layer of the endocyst, wdiich pushes before it an in-
volution of the inner layer. The latter gives rise to the re-
flected " peritoneum."

But I am not aware that there is any evidence which
proves conclusively the manner in which these tv^ o layers of
the embryonic endocyst take their origin, or with wdiat layers
of the ordinary embryo they are homologous. If we make
the ordinary assumption that the inner or peritoneal layer of
the endocyst is the partial or complete homologue of the hy-
poblast in other animals, it follows that the perivisceral cav-
ity of the Ectoprocta is really an enterocoele, as it is in the
JBrachiopGda. The only other alternative appears to be the
supposition that the inner layer of tlie endocyst is a meso-
blast, differentiated from the germ earlier than the hypoblast;
in which case the perivisceral cavity will be a schizocoele.

Dr. Jhering's work on the nervous system of the Molhisca^
to which 1 have already referred, contains a number of valu-
able anatomical details, and especially gives a better account
of the structure of the nervous system of Chiton than has
hitherto existed.^

1 In addition to a great variety of surprising phylogenic speculations, Ur.
Jhering puts forward tlie novel niorphological views tliat the respiratory sac
of the Pulmonata {Nephropneusta^ Jhering) is morphologically a sort of urinary
bladder, and that the ganglia whence the arm-nerves of the Cephalopoda arise
are cerebral, and not pedal. The arms are thus parts of the head, and only the
funnel represents the foot of Gastcropods.

I do not presume to rebel against the authoritative censure of my memoir on
the " Morphology of the Mollusca," published now live-and-twenty years ago,
which is pronounced by Dr. Jhering. Nevertheless, I may remark that, had
he condescended to pay" attention to what is said respecting the flexure of the
intestine in Mollusks in that antiquated production, he would not have com-
mitted himself to the publication of the two diagrams— one of a Cephalopod
and the other of a Pteropod— each with its alimentary canal twisted after a
fashion of which Nature knows nothing, which illustrate, though they hardly
adorn, page 272 of his work.

THE HIGHER GROUPS. 573

There is no invertebrated animal at present known which
cannot at once be referred to one or other of the natural or-
ders which have been discussed in the preceding pages. The
next question which arises is, How far are these groups sus-
ceptible of arrangement into assemblages of a higher order,
distinofuished from all others by certain common characters ?

It is universally admitted that the Insecta, Mijriapoda^
Arachnida^ Crustacea, Pycnogonida, and Tardigrada, form
such an assemblage, termed the Arthropoda, and character-
ized by the segmentation of the body ; the chitinous cuticula ;
the absence of cilia upon, or in, the body at any period of life ;
the segmentation of the central nervous system, and its per-
foration by the gullet ; and the presence (with the possible
exception of the Irlloblta) of limbs, which, almoit always,
are themselves subdivided into joints. The reasons for in-
cluding the Peripatidea in this division have been given in
Chapter XL; and, though the Pentastomida must be regarded
as hardly within the limits of the definition, I think that, tak-
ing: into account the strans-e modifications which are under-
gone by the parasitic Crustacea and Arachnida, it is not
needful to depart from the ordinary practice of associating
them with the Arthropoda.

The L'xmellibranchiata and the Odontophora constitute
another very well marked division, the Mollusca, the char-
acters of which have been discussed in Chapter VIII.

The proposal to separate the Polyplacophora from the
Mollusca, to which I have already referred, appears to me to
be devoid of any justification. The resemblances between
certain Gephyrea, such as Choetoderma and Neoinenia^ and
the Polyplacophora, are accompanied by wide differences ;
and even if these resemblances are to be regarded as evi-
dences of affinity, some considerations, such as the restriction
of the branchiae to the hinder part of the body, and the reduc-
tion of the foot in Chitonellus, rather lead to the suggestion
that ChcBtoderma and JSTeonienia may be extremely modified
Mollusks, allied to the Polyplacophora.

As to the supposition that the resemblances between the
Nudihranchiata and the Turhellaria indicate a direct affin-
ity between these groups, it seems to be forgotten that the
Nudihranchiata are all, when young, unmistakable Gastero-
pods provided with mantle and shell. Their adult structure
is as little evidence of any Turbellarian affinities as that of
Lern'Ba is proof of its being allied to the worms rather than
to the Crustacea,

574 THE ANATOMY OF INVERTEBRATED ANIMALS.

The Physemaria^ the Porifera^ the Hijclrozoa, the Coral-
ligena^ and the Ctenophora, are obviously modifications of
the same fundamental plan. I think it is convenient to re-
tain the well-established name of Coelenterata for the last
three orders, which are much more closely related to one
another than to the other two. Haeckel's proposal to apply
the old name of Zoophyta to the whole division appears to
me to be well worthy of adoption. The inconvenience of
using a term the connotations of which have varied some-
what widely since it was first invented, is probably less than
that which would attend the invention of a new name.

The Monera^ Foraminifera^ Heliozoa, Padiolaria, Pro-
toplasta, Gregarinida, Catallacta^ and Infusoria {Opali'
9iina, Ciliata, Tentaculifera, Flagellata)^ again, are so close-
ly united together that the difficulty is to distinguish the
less differentiated forms of each from one another. They
constitute the division of the Protozoa, the common charac-
ters of which have been given in Chapter II.

If there were no invertebrated animals besides those in-
cluded under these four divisions of Artheopoda, Mollusca,
Zoophyta, and Protozoa, the task of classification would be
very easy, and each of the higher divisions would be sharply
defined from the others. But a vast residuum remains to be
considered ; and it is with the attempt to arrange these resid-
ual orders into higher groups that the difficulties of the Tax-
onomist commence.

The Polychceta and the OUgochoeta, the Hirudinea and
the Gephyrea, resemble one another generally in the seg-
mentation of the body, indicated at least by the serially mul-
tigangliate nervous centres ; * in the presence of cilia and of
segmental organs ; and in the nature of the larvae, which are
set free when their embryos are hatched in an early stage of
development. And, although no one of these characters is
of universal occurrence (cilia, for example, being absent in
most adult Hirudinea)^ yet they are found in such association
that the accepted arrangement of these four groups (to which,
though not without some hesitation, I add the 3Iyzostomata)
into the division of the Annelida is undoubtedly very con-
venient.

The Trematoda, the Turhellaria, and the Potifera, form

* This character is wanting in most Gepliyrea, which, as I have remarked
at p. 218, incline in many respects toward the next division, and especially
toward the Botifera and Nematorhyncha,

THE HIGHER GROUPS. 575

another very natural assemblage. But it must be admitted
that the highest forms of this division are separated by no
very sharp line of demarcation from the A?inelida / while
the simplest Turhellaria are almost on a level with the JPhy-
semaria and the lower Hydrozoa. Even a l^lanaria is com-
parable to a free zoophyte ; its proboscis may be likened to
the hydranth of a Medusa^ the prolongation of the alimen-
tary sac to the gastro-vascular canals, the central nervous sys-
tem, with its lateral prolongations, to the marginal ganglia
and nerves. The water-vascular system and the com.plication
of the reproductive organs, indeed, afiford clear marks of dis-
tinction ; but both of these systems vary indefinitely in the
degree of their development within the limits of the Turhel-
laria.

On the other hand, the connection of the Hirudinea by
such forms as Malacohdella with the Turhellaria and Trenia-
toda is very close ; Polygordius appears to be a transitional
form between the Turhellaria and the Polychc^ta ; wdiile the
JRotifera^ in many respects, represent larval forms of the
PolycJiata and of the Gephyrea.

The Cestoidea are usually regarded as anenterous Trema-
toda, in which case, of course, they must be associated with
the latter.

I propose to establish a division of Tkichoscolices for
the natural orders now enumerated, in order to discriminate
the morphological type which they exemplify from that of
the Nematoscolices, containing the N^ematoidea, which are
as remarkable for the universal absence of cilia as the former
are for their presence ; and which are further so clearly dis-
tinguished by the arrangement of their nervous and muscu-
lar systems and of their water-vessels ; and by their ecdysis.

The connection between the two divisions by way of the
Nematorhy7icha and the Potifera is undoubtedly very inti-
mate, and there is almost as much reason to arrange the JVe-
matorhyncha with the Trichoscolices, as with the Nematosco-
lices. On the whole, however, I think that, notwithstanding
tlie cilia of the Gastrotricha, the closest affinities of the
N'einatorhyncha are with the N^ematoidea, and I therefore
place them among the J^ematoscolices.

But I may remark, once for all, that the attempt to estab-
lish sharply-defined, large divisions of the animal kingdom is
futile. The progress of knowledge every day renders it
more and more clear that morphological groups are compara-
ble to distributional provinces ; each, however well marked

576 THE ANATOMY 01^ INVERTEBRATED ANIMALS.

may be its characteristic features, shades off at its margins
into some other group ; and the object of classification is
simply to bring into prominence the morphological types
which embody these characteristic features.

It appears to me impossible to compare the structure and
the larval conditions of a Polyzoon with those of a Brachio-
pod, without arriving at the conclusion that they are more
closely allied with one another than they are with any third
group. Nevertheless, the Polyzoa approach the JRotifera,
and the l^rachiopoda the Annelida^ on the one side ; while
on the other they present unmistakable affinities with the low-
er Mollusca. At the same time the weight of the resemblances
between the Polyzoa and the Tanicata^ which led Milne-
Edwards to the establishment of the group of " Molluscoides "
(adopted by myself under the title of 3Iolluscoida), has been
much lessened by the progress of investigation.

I conceive that we may best keep these resemblances and
differences in view by associating the Polyzoa and the Pra-
chiopoda into a division apart, for which I propose the name
of Malacoscolices ; in order to indicate its relations wath
the Worms on the one side and with the Mollusca on the
other.

The Tunicata are absolutely distinguished from all other
invertebrated animals except Palanoglossus^ by the perfora-
tion of the pharynx and its conversion into a respiratory
organ. ^

At first sight there appears to be little ground for the
approximation of groups apparently so widely different as the
Tunicata and the Enteropneusta. But the extraordinary
similarity in the structure of the perforated pharyngeal sac in
the larvDS of Tunicates and of Palanoglossus is a fact of
great morphological weight. An ecaudate Appendlcularia
of those species which have the alimentary canal nearly
straight, would be marvelously like a larval Palanoglossus,
which is again little more than a specially modified Turbella-
rian. I think, therefore, that the Tunicata and the Entero-
pmusta may properly constitute a division of Pharyxgo-

PJ^^EUSTA.

1 I have alluded above to the structures described by Semper in some Oli-
gochceta and in SaheUa. I do not doubt the accuracy of the description ; but
it does not lead me to conclude that the structures in question are homologoua
with either Vertebrate, Enteropneustal, or Tunicate branchiaj.

THE ANENTEROUS INVERTEBRATA. 577

The Tunicate Pharyngopneusta, with their caudate larvae,
may be supposed to stand in the same relation to the Turbel-
lariform Pharyngopneustciy as the Trematoda^ with their cer-
cariform larvae, to the Turbellaria,

Another very well marked division is that of the Echino-
DERMATA, the characteristics and relations of which have
been fully discussed in Chapter IX

Although the structure and development of Sagitta have
now been as thoroughly elucidated as those of any animal,
the proper Taxonomic place of the Choetognatha is still an
unsolved problem. The issues, however, appear to be nar-
rowed to these : either they belong to the Aniielida, or to the
N^ematoscolices^ or to the Trichoscolices ; or the Chceto-
gjiatha are to be regarded as an independent division, allied
to all these, and perhaps to the lower Arthropoda. I am dis-
posed to adopt the last view, chiefly on the ground of the
mode of development of Sagitta^ which is unlike anything at
present known to occur in Annelida, TrichoscoUces, JS^ema-
toscolices, or Arthropoda,

The Acanthocephala are hardly less anomalous than the
Chmtognatha, Taking into account the Gordiacea and the
characters of the proboscis in the Kematorhyncha^ there is
undoubtedly room for the suggestion that they are specially-
modified anenterous Nematoscolices^ and should be classed
among the latter. But here, as in the case of the Cestoidea^
there are many difficulties in the way of accounting for these
anenterous forms by the supposition that they are the results
of a retrogressive metamorphosis of enterate animals.

This question of the true relations of the anenterous in-
vertebrates— by which I mean not only those which, like the
male Rotifers, have no functional alimentary canal in the
adult condition ; but those which, like the Cestoidea and the
Acanthocephala^ never exhibit a trace of an alimentary canal,
even in the embryo ; which is usually dealt with so summarily
by the assumption of retrogressive metamorphosis — acquires
still more importance, when we attempt to determine the
Taxonomic place of the Dicyemida.

Prof. E. van Beneden has proved that these parasites can-
not be dismissed, sans fagon, as retrogressively metamor-
phosed " worms ; " and though I am not disposed to attach
much weight to the absence of a mesoderm, on which Van
25

578 THE ANATOMY OF INVERTEBRATED ANIMALS.

Beneden insists as a distinction between the Dicyemida and
the Metazoa^ the manner in which the contents of the axial
cell give rise to germs is so completely unlike anything which
is known to obtain in the 3fetazoa, as, to my mind, to justify
the separation of the Dicyemida from the whole of this divi-
sion. On the other hand, the similarity of their development
to the formation of metazoic embryos by epiboly, as com-
pletely divides the Dicyemida from all the Protozoa. It
must be recollected that the changes which are undergone by
the ciliated embryos are still to be discovered ; but, provision-
ally, I am disposed to agree with Van Beneden, that the Di-
cyemida should be regarded as the representatives of a dis-
tinct division, the Mesozoa, intermediate between the Pro-
tozoa and the Metazoa, And without distinctly pledging
myself to any such view, I yet think it is worth while to
throw out the suggestion that the Cestoidea, if not the
A-canthocephala , niaj^ be modifications of the same type,
differing from the Dicyemida in the development of a meso-
derm, but resembling them in the total absence of an alimen-
tary apparatus.

The Sekial Relations of the Invertebeata. — When
the various groups of invertebrate animals are compared, it is
obvious that they present very different degrees of morpho-
logical complexity ; whence they may be considered as terms
in a graduated progression, in w^hich the place of each group
corresponds broadly with the degree of its differentiation.
The lowest Protozoa will occupy one extreme of such a pro-
gression, the Arthropoda and the Mollusca the other, while
the remaining groups fall into intermediate places. On at-
tempting to carry out this serial arrangement into detail,
however, it will be found that no single series will suffice to
express the facts, but that, starting from the lowest Protozoa,
we are led along various lines, none of which, as far as our
present knowledge enables us to judge, can be traced, with-
out interruption, throughout the whole length of the scale.

If we assume, in the absence of proof to the contrary, that
the Monera have the simplicity of structure ascribed to them
by Haeckel, then, on comparing the Endoplastica with the
Monera, the different groups of the former appear to be re-
lated to those of the latter division, as if they were similar
forms complicated by the addition of one or many nuclei.
Protogenes may thus be considered as the root of the Foram-
iniferal series, Protamoeba of the Protoplasta, Myxastrum

THE SERIAL RELATIONS OF INVERTEBRATA.

579

of the Gregarinidce^ Yampyrella of the Heliozoa, Protomo-
7ias of the Flagellata. A Moneran, ciliated over its whole
surface, which might stand in the same relation to the Ojm-
linina, Catallacta, Tentaculifera, Ciliata^ is at present un-
known. The Protozoa thus fall into the following series :

Peotozoa.

T. II. III.

Protogenes. Protamoeba. Myxastrum.

IV.

Vampyrella,

Foraminifera. Protoplasta. Gregarinidce.

Heliozoa,

I

Radiolaria.

V.

I

Tent aculif era.

VI.
?

Catallaeta,

I

Ojxdinina.

VII.

Protomoiias,

Flagellata,

Ciliata.

I am unable to trace any one of these series of modifica-
tions further; that is to say, to find forms which actually
bridge over the interval between any one of them and the
Metazoa^ though it is easy enough to imagine what such forms
might be. The spheroidal free-swimming monad aggregates,
such as TTvella and Polytoma^ and 3Iagosphciera itself, are,
in many respects, comparable to Physemarian or Poriferan
embryos ; while an animal Volvox would be a sort of perma-
nent vesicular morula. So, one of the higher Infusoria^ if it
became multinucleate, like an Opalina^ would approach the
lowest Turhellaria.

The axial cell of a I>lcyeina^ from the protoplasm of which
its ciliated and nonciliated germs are produced, is, to a cer-
tain extent, comparable to the capsule of a Radiolarian ;
while, on the other hand, a Radiolarian with a multinucle-
ate cortical layer would approach the structure of Dicyema.
And if what is at present known of Dicyema gives a just
conception of the essential points of its entire history, it un-
doubtedly, as E. van Beneden has suggested, represents a
type intermediate between the Protozoa and the Metazoa,

580 THE ANATOMY OF INVERTEBRATED ANIMALS.

though it can hardlj be said to fill up the hiatus between
them.

In our farther search after the serial relations of animals,
■we must therefore start afresh from the lowest 3fetazoa.
Here a Zoophytic Series is very well marked ; commencing
with the Physemaria^ and thence diverging, on the one
hand, to the Porifera, and, on the other, to the Coelenterata,
with the highest forms of which this series comes to an end.

A second gradation, which may be termed the Anntjloid
Series, is represented by the Trichoscolices and the Anne^
licla. The lowest Turhellaria are upon nearly the same level
of organization as the Hydrozoa, It would be hard to dis-
tinguish an aproctous Turbellarian, devoid of a ganglion and
water-vessels, from a free-swimming nontentaculate Hj^dro-
zoon. On the other hand, as I have alread}^ pointed out, the
line of demarcation between the higher TrichoscoUces and
the Annelida is very indistinct, and we may expect it to be
speedily obliterated by the progress of discovery.

A third gradation is constituted by the N'ematoscolices and
the Arthropoda, The lowest Nematoidea possess no higher
organization than the lowest Turhellaria and the Rotifer a.

The JSfematorhyncha, whether they are really transitional
forms between the Nematoidea and the Arthropoda or not,
at any rate indicate the road by which the transition may be
effected ; and I am much inclined to think that the Chceto-
gnatha may occup}'^ a place in this series. The oral armature
of Sagitta may be regarded as a modification of the oral
spines of Echinoderes, and its nervous system is as much
Arthropodal as is that of the Pentastoniida, This may be
called the Arthrozoic Series.

A fourth series is that w^hich I shall term the Malacozoic
Series. It includes the Malacoscolices and the Mollusca.
The entoproctous Polyzoa form the lowest term of this series.
The resemblances of the Polyzoa with the Potifera (e. g.,
with Stephanoceros) have often been remarked, and, indeed,
insisted upon, with too little regard to the diff*erences which
are established bv the water-vessels and the peculiar pharyn-
geal armature of the Rotifers. Nevertheless, these resem-
blances are important as far as they go, and in grade of or-
ganization the two groups are much upon the same level. On

THE SERIAL EELATIONS OF INVERTEBRATA.

581

tbe other hand, the comparison of a Polyzoon with a larval
Lamellibranch or Gasteropod, or with a Pteropod, leaves no
doubt in my mind that the JIalacoscolices have the same rela-
tion to the jHolluaca, as the IVichoscolices to the Annelida.

A fifth gradation is presented by the Tunicata and the
Entcropneusta, which constitute the PHAEYNGOP^iTEUSTAL
Series. I do not regard the Enteropneusta as of distinctly
lower organization than the Tunicata^ but rather as a col-
lateral group; and I conceive it to be probable that some lower
forms, connecting the Enter opneusta and the Tunicata with
one another and with the Trichoseolices, will yet be found.
How^ever this may be, Appendicular ia presents a grade of
organization but little higher than that of the Polyzoa,

A sixth gradation is represented by the Echixodermal
Series. Like the foregoing, this series at present stands
isolated,^ no annectent forms between the Echinoderms and
higher or lower groups being known. On the ground of the
uniformity of character of the larvjB of the Echinoderms,
however, there can be little doubt that, if ever such forms are
discovered, they will prove to be allied to the Gephyrea^ the
Trichoscolices^ and the Enter opneusta.

Thus the study of the gradations of structure among the
Metazoa leads to the conclusion that they fall into six series,
which may be arranged in the following tabular shape :

Series.

n.

Echijs^odeemal.

Ccelenterata, Echinodermata. Enter opneusta, Tunicata,

Porifera.

Physemaria,

I.

ZOOPHYTIC.

III.

Pharyxgopneustal.

IV.

Malacozoic.
3follusca.

Malacoscolices.

Y.

Ani^^uloid.
Annelida.

Trichoscolices.

VI.

Arthrozoic.

Arthropoda.
ChoBtognatha (?).
Nematoscolices.

' I say, at present, masraucli as the characters of the nervous system sharp-
ly separate the most vermiform of the Echinoderms from the most Echinoderm-
like Gephyrea.

5S2 THE ANATOMY OF INVERTEBRITED ANIMALS.

The lowest known term of the Arthrozoic series is a Nema-
toid worm ; that of the Annuloid series is a low Turbellarian
or Rotifer ; that of the Malacozoic series is an entoproctous
Polyzoon ; that of the Pharyngopneustal series is probably
most nearly exemplified by the young larva of ^aianogloS'
sus ; that of the Echinodermal series by the vermiform J^chi-
nopcBcllum.

Bat the differences between one of the simpler Nematoid
worms, an aproctous Turbellarian, a Rotifer, an Echinopasdium,
and a Pedicellina^ are relatively so small, that all six series
may be said to converge toward a common form ; and that
common form, when the special characters of each group are
eliminated, and the alimentary canal is reduced to its primi-
tive aproctous condition, would be exceedingly similar to a
Physemarian.

Hence the consideration of the gradations of structure
which are presented by the various series of Invertebrated
animals, irresistibly leads to the conclusion that the whole of
the Me,tazoa may be conceived as diverse modifications of a
common fundamental plan.

The Serial Relatioxs or the Ixvertebrata com-
pared WITH THE Results of Embryology. — The conception
of the unity of organization of the I/ivertebrata thus reached,
so far as it is based upon the comparison of adult structures,
is purely ideal ; and the study of the development of individ-
ual animals is alone competent to decide the question wdiether
this ideal unity has a foundation in objective fact. But the
history of the development of animals appertaining to every
group of the Invertebrata which has been given, bears out
the statement which is made in the Introduction, that the
ideal unity has such a foundation in fact ; inasmuch as all
these animals commence their existence under the same
form — that, namely, of a simple protoplasmic body, the ovum
or germ.

In the Introduction I have said that, '' among the lowest
forms of animal life, the protoplasmic mass which represents
the morphological unit may be, as in the lowest plants, devoid
of a nucleus " (p. 18), Howev^er, as I have remarked at the
commencement of this chapter, until the search for the nucleus
has been instituted afresh, with the help of such methods as
have recently proved its existence in the Foraminifera, I
think it will be wise to entertain a doubt whether any of the
Monera are really devoid of this amount of structural differ-

TUE RESULTS OF EMBRYOLOGY. 533

entiation ; and the tendency of recent investigations appears
to render it very questionable whether the nucleus of the
ovum ever really disappears, whatever may be the modiiica-
tions undergone by the germinal vesicle and its contents. I
shall, therefore, assume provisionally, that the primary form
of every animal is a nucleated protoplasmic body, cytode, or
cell^ in the most general acceptation of the latter term.

Whether the primary cytode possesses a nucleus or not,
the important fact remains that, in its earliest condition,
every invertebrated animal, if it were competent to lead an
independent existence, would be classed among the Protozoa,

The first change which takes place in the development of
the embryo from the primitive cytode, or impregnated ovum,
in all the Metazoa, is its division ; and the simplest form of
division results in the formation of a spheroidal or discoidal
mass of equal, or subequal, derivative cytodes, the blasto-
meres. Next, the morula, thus formed, generally acquires a
central cavit}', the blastocoele, and becomes a hollow vesicle,
the hlastospliere, the wall of which, composed of a single layer
of blastomeres, is the blastoderm.

The blastomeres of the blastoderm next undergo differen-
tiation into two kinds, distinguished by their internal activi-
ties, if not by their outward form. Of these the one set con-
stitute the epiblast, the others the hypoblast. The further
changes of the embryo are the consequences of the tendencies
toward further modification resident in the epiblastic and hy-
poblastic blastomeres respectively. Each of these is, as it
were, a germ, whence certain parts of the adult organism will
be evolved.

Every series of the Invertebrata has now yielded a num-
ber of examples of the further modification of the blastosphere
by the process of invagination, or emboly, the result of which
is that the hypoblast becomes more or less completely inclosed
within the epiblast. The invagination is accompanied by the
diminution, or even abolition, of the blastocoele, and the for-
mation of a cavity inclosed within the hypoblast, which is the
archentero7i, or primitive alimentary cavity. The opening
left by the approximated edges of the epiblast, when the pro-
cess of invagination is completed, and by which the archente-
ron commimicates with the exterior, is the blastopore. In
this state the embryo is a gastrula.

It very commonly happens that the process of develop-
ment is modified by an inequality in the size of the blasto-
meres ; which inequality may be manifest from the bisection

584 THE ANATOMY OF INVERTEBRATED ANIMALS.

of the ovum, or may appear later. In this case, it usually
happens that the smaller and more rapidly-dividing blasto-
meres belong to the epiblast, and the larger and more slowly
dividing to the hypoblast. Moreover, no blastoccele may
arise, and the process of inclusion of the hypoblast within the
epiblast may have the appearance of the growth of the latter
over the former, or what is termed epiboly j while the archen-
teron may not be formed within the hypoblast till very late.

When, in cases of epiboly, the blastoderm is small in rela-
tion to the vitellus, the epiblast and hypoblast, at their first
appearance, necessarily adapt themselves to the surface of the
yelk ; and thus the gastrula, instead of having the form of a
deep cup, becomes more or less flattened and discoidal.

I am inclined to believe that all the various processes b}'
which the gastrula or its equivalent are produced, are reduci-
ble to epiboly and emboly. Even when the epiblast and the
hypoblast appear to be formed by delaminatio7i, or the split-
ting into two layers of cells of a primitively single-layered
blastoderm, there seems little doubt that what happens is
either the very early inclusion of the hypoblastic blastomeres
within those which give rise to the epiblast, or a very late
and inconspicuous ingrowth, or invagination, of the hypoblas-
tic region of the blastoderm.

If we employ the term gastrula in the broad sense defined
above, it may be truly said that every metazoon passes through
the gastrula stage in the course of its development. The
question whether the mode of development of the gastrula
by emboly is primitive, and that by epiboly secondary ; or
whether epiboly is primary and emboly secondary; or whether
the two processes have originated independently, is of sec-
ondary importance, and belongs to the debatable ground of
phylogeny.'

The meaning of the differentiation of the aggregate of
cytodes, of which the body of a simple m.etazoon is composed,
into a hypoblastic, or endodermal, and an epiblastic, or ecto-
dermal, group, is to be sought in the physiological division of
labor, which is the primary source of morphological changes.
It is a separation of the aggregate of morphological units into
one set with a specially nutritive, and another set with a spe-
cially motor and protective, function. It is quite possible to
conceive of an adult metazoon having the structure of a sponge-

* Compare Ilaeckel, "Studien zur Gastr»a-Theorie," in Lis " Biologische
Studien," 1877.

THE RESULTS OF EMBRYOLOGY. 585

embryo ; moving by its ectodermal hemisphere, and feeding
by its endodermal hemisphere.

The next advance in organization of such a raetazoon
would doubtless consist in the more complete extension of
the protective layer over the nutritive layer, with due pro-
vision for the access of the surrounding medium to the latter.
It is obvious that this advance might be effected in either of
two ways : the one by emboly, the other by epiboly. In the
former, the blastopore would be left as the aperture of com-
munication of the endoderm with the exterior; and the result
would be the formation of an archmostomatous gastrula, such
as that which is supposed by Haeckel to be the primitive form
of the metazoon. In the latter, the blastopore would com-
pletely close up, and a new aperture or apertures must be
formed in the ectoderm to subserve the ingestion of nutri-
ment. The resulting organism would be a deuterostomatous
gastrula.

Undoubtedly it seems natural to suppose that the first
process preceded the second, in order of evolution ; but the
proof that it did so is at present wanting. And, however
this may be, the progress of inquiry seems to throw more and.
more doubt upon many cases of the supposed persistence of
the blastopore as the mouth. It is certain that, in the great
majority of invertebrated animals, the blastopore either be-
comes the anus, or closes up ; and renewed observations are
needed to determine the limits within which the archseostoma-
tous condition prevails.

The blastocoele of the gastrula may be obliterated by the
approximation of the epiblast and the hypoblast, or it may
persist and constitute the perienteron^ or primitive perivis-
ceral cavity.

Those animals which, in their adult condition, most nearly
represent simple gastrula with obliterated blastocoele, are the
JPhysemaria and Hydra^ cup-shaped bodies with an oval
opening at one end, the walls of which are made up simply
of an ectoderm and an endoderm.^

In the great majority of the Metazoa^ a further advance in
complication is effected by the appearance, between the epi-
blast and the hypoblast, of cytodes, either isolatedly or in a
continuous layer, which constitute the mesoblast, and eventu-
ally are converted into mesodermal structures. The origin

* I do not think that Kleinenberof's fibres in Hydra strictly represent a
mesoderm, though they occupy the position of one.

586 THE ANATOMY OF INVERTEBRATED ANIMALS.

of these is still a matter of doubt, but in many cases it ap-
pears to be unquestionable that they are derived from the
hypoblast.

The perienteron, mere or less interrupted and broken up
by the constituents of the mesoblast, may give rise directly to
the perivisceral space, or channels, of the adult, which thus
constitute a schizocoele. It is hardly doubtful, I think, that
the perivisceral cavity takes its origin in this manner in the
JRotifera^ the entoproctous Polyzoa., the Echinopsedia of the
Echinoderms, the Tunicata, and the Nematoidea.

On the other hand, in many Invertehrata, one or more di-
verticula of the archenteron extend into the perienteron and
its contained mesoblast. Sometimes, as in the Coelenterata,
these remain connected with the alimentary cavity through-
out life, and are termed gastrovascular canals. In other cases
{Echinodermata, Brachiopoda^ Choetognatha) they become
shut ofiP: their cavities constitute a variouslv-modified entero-
coele ; and their walls give rise, along with the primitive
mesoblastic elements, to the mesoderm.

To which of these two possible sources of the mesoderm,
the mesodermal structures of the Annelida and the Arthro-
poda, which so very generally take on the form of two longi-
tudinal germ-bands in the embryo, and subsequently undergo
segmentation, are to be referred, is a very interesting, but, as
yet, unsolved problem. It is possible that they are solid rep-
resentatives of the hollow diverticula which, in other animals,
give rise to the enterocoele ; in which case the perivisceral
cavity in these animals will be a virtual enterocoele. On the
other hand, they may merely represent the cells of the meso-
blast of the entoproctous Polyzoa and of the Echinopaedia,
and their perivisceral cavity would then be a schizccoele. But
it is needless to pursue this topic further ; enough has been
said to show conclusively that, however different one inver-
tebrated animal may be from another, the study of develop-
ment proves that each, when traced back through its embry-
onic states, approaches the earlier stages of all the rest ; or,
in other words, that all start from a common morphological
type, and even in their extremest divergence retain traces of
their primitive unity.

It is very important to remark that these morphological
generalizations, so far as they are correctly made, are simple
statements of fact, and have nothing to do with any specula-
tions respecting the manner in which the invertebrated ani-

PALEONTOLOGY AND PHYLOGENY. 587

mals with which we are acquainted have come into exist-
ence. They will remain true, so far as they are true at all,
even if it should be proved that every animal species has
come into existence by itself and without reference to any
other. On the other hand, if there are independent grounds
for a belief in evolution, the facts of morphology not only
present no difficulty in the way of the hypothesis of the evo-
lution of the Invertehrata from a common origin, but readily
adapt themselves to it.

Hence the numerous phylogenic hypotheses which have
of late come into existence, and of which it may be said that
all are valuable, so far as they suggest new lines of investi-
gation, and that few have any other significance. I do not
desire to add to the number of these hypotheses ; and I will
only venture to remark that, in the absence of any adequate
palseontological history of the Invertehrata^ any attempt to
construct their Phylogeny must be mere speculation.

But the oldest portion of the geological record does not
furnish a single example of a fossil which we have any rea-
sonable grounds for supposing to be the representative of the
earliest form of any one of the series of invertebrated ani-
mals ; nor any means of checking our imaginations of what
may have been, by evidence of what has been, the early his-
tory of invertebrate life on the globe.

Already indications are not wanting that the vast multi-
tude of fossil Arthropods, Mollusks, Echinoderms, and Zoo-
phytes, now known, will yield satisfactory evidence of the
filiation of successive forms, when the investigations of pa-
laeontologists are not merely actuated by the desire to dis-
cover geological time-marks and to multiply species, but are
guided by that perception of the importance of morphological
facts w'hich can only be conferred by a large and thorough
acquaintance with anatomy and embryology. But, under
this aspect, the palaeontology of the Invertehrata has yet to
be created.

INDEX.

Abdominalta, 260.
Abiognesis, 38.
Abiological sciences, 9.
Acanthobde'.la, 1S9.
Acanthocephala, 55?, 5TT.
Acanthoteuthis, 465.
Acarina. 329.
Achetidse, 378.
Achtheres, 241.
Acineta mystacina, 94.
Acinet£e, 89. 93, 94, 99-101.
AcrididiB, 37S.
Actinia, 5:3, 140, 141, 151.

holsatica, 139.

Actinidae, 140,141, 145.

Actinophrys, 83, 80, 93.

Actinosphaerium Kichhom i, S3, 85.

Actinozoa, 55, 110, 137, 14u, 1 19, 150.

Actinula, 132.

uEglnidae, 136.

^thalium septicum, 13,

jfitiology, 16, 37.

Agamogenesis, 31, 34, 383.

Aglaophenia (Plumularidae), 120.

Air-breathing Arthropoda, 320.

Alaurina, 157.

Albertia, 170.

Alcippe lampas, 201.

Alcyonium, 143.

Alectoromorph.T, 69.

Alarae, 14, 20, 32. 97.

Alimentary apparatus, 56.

Allantoic, 67.

Alternation of generations, 36, 6T.

Ambulacral vessels, 54.

Ametabola. 361.

Amtnonitidae. 459.

Ammothea pycnogonoides, 332.

Amnion, 67.

Amoeba radiosa, 8^).

sphaerococcus. 86.

Amcebse, 13, S6, 103.
Amouroucium prolifernm, 526.
Amphibia, 53, 59, 64, 70, 86.
Amphidiscus, 108.

Amphidotus cordatus, 491.
Amphioxus, 57, 59.
Amphipoda, 313.
Amphithoe, 311.
AmpuUaria, 60.
Anatomy. 16, 17.
Anenterbus invertebrates, 577.
Anguillula brevispinus, 545.

scandens, 550.

Animals, characters, 44: morphology, 4T;

physiology, 54 ; natural orders, 562.
Anisonema, 90.
Annelida, 50, 51, 66. 164. 171, 193, 206, 20T,

212, 219, 416, 575. 580, 5b6.
Annuloid series, 5S0.
Annulose differentiation, 59.
Anodonta, 407. 408, 412, 416. 417.
Anoraia. 411, 417.
Anomura. 293, 294.
Anthophysa, 90.
Antinoptirys Eichhornii, 13.
Antipatbidae, 14.\ 146.
Aphis, 36. .37. 3S0-884.

pelargonii. 364.

Aphroditidas, 210.

Apoda, 260.

Aporosa. 146, 147, 150, 153.

Appendages, 20.

Appendicularia, 53, 510-518, 576.

flabellum, 511.

Aprocta, 158.

Aptychus, 459, 460.

Apus, 220, 223, 226. 242-245.

cancriformis, 243.

glaciali<5. 245,

Arachnida. 59, 221, 224, 320, 573.

Araneina, 326.

Area. 417.

Arcella, 86.

Arcti?ca, 334.

Argonauta argo, 461, 462.

Argulus, 241.

" Aristotle's lantern," 492.

Arthrojastra, 320.

Arthropoda. 21, 32, 36. 52. 53, 57, 64, 66, 198,

206, 219-225, 320, 573, 574, 580, 586.
Artlirozoic series, 580,

590

IXDEX

Articulata, 403,
Ascaris nigrovenosa, 551.
Ascetta primordialis, 104.
Ascidians, 45, 63.
Ascidioida. 510.
Ascones, lOG, 110.
Ascula, 106.
AseUas, 313, SIT.
Aspergillum, 406.
Aspidobranchia, 444.
Aspidogaster conchiola, 172-178.
Astacus, 66, 219, 264-293.

fluviatilis, 266.

Asterldae, 466, 474, 475.
Astr£ea calycularis, 146.
A tax Bonzi, 330.
Athorybia, 130.

rosacea, 127-129.

Atolls, 151.
Atrocha, 213.
Aurelia aurita, 122.
Avicularia, 393.

Q

Bacteria, 12-14, 38.
Balanidis, 260.

Balanoglossus, 53, 539, 570, 582.
Balantidium, 97, 9S.
Balanus, 254-260.

balanoides, 258

Bees, 33, 34.

Beetles, 366.

Belemnitidifi, 463-4G5.

Beryx, 40.

Bicosoeca, 90.

Bilharzia, 178.

Biogenesis, 40.

Biology, principles, 9 ; divisions, 16.

Bipinnaria, 481.

Blastoderm, 382.

Biastoidea, 509.

Blastomere, 20, 22, 28, 32, 34, 48, 317, 415.

Blastosphere, 415.

Blastostyle, 119.

Blatta, 343, 349, 856, 371, 873, 377, 380, 3S2.

orientalis, 846, 357, 360.

Blood and circulatory apparatus, 56.
Bojanus, organs of, 52, 57, 61, 66, 411.
Bombus, 369. 372.
Bothriocephalus, 187.

latus, 184.

Botryllida^, 514, 518, 522, 524-528.

Botrytis Bassiana, 45.

Brachionus, 168, 169.

Brachiopoda, 389, 896. 397. 417, 586.

Brachyura, 281, 293, 294, 295.

Branchellion, 189.

Branchiae, 58.

Branchiogasteropoda, 424, 433, 434, 436, 437.

Branchiopoda, 242.

Branchipus, 247, 248, 249.

Brisinga, 4S0, 481.

Bryozoa, 389.

Buccinum, 434.

undatum, 420.

Bucephalus polymorphus, 181.
Bngula avicularia, 3^3.
Butterflies, 366.

Calcispoxgi^, 104-110.

Caligus, 241.

Oalycophoridae, 86, 117, 123-131.

Cambium layer, 21,

Campanularia, 119.

Campanularidse, 117, 118.

Oampodea staphyliuus, 362.

Capitella, 200.

Caprella, 313.

Carcinus moenas, 295, 302, 303.

Cardium, 417.

Carmarina, 115, 135.

Caryophyllaeus, 182.

Catallacta, 89, 101, 574.

Caulerpa. 48.

Cecidomyia, 385.

Cells, 17, 21, 28, 31, 32.

Cell- wall, 18.

Centipedes, 344.

Cephalopoda, 64, 66, 404, 418, 419, 425-435,
444,445.

Cephea ocellata, 124-126.

Cercariae, 53, lSO-182.

Cereanthus, 145.

Cestoidea, 56, 157, 182, 575.

Cestracion, 69.

Cetacea, 69.

Chaetoderma, f 71.

Chaetogaster, 193, 194.

Cha'tognatha, 540, 577, 586.

"Challenger" expedition, 68, 7rt, 79, 81.

Changes, cyclical, in living matter, 10.

Chara, 21.

Chemical composition of Uving matter. 9.

Chick, 19.

Chilodon, 98.

Chilognatha, 337, 338.

Chilopoda, 337.

Chitonidae, 430, 481, 434, 571, 572.

Chlamydomonas, 46.
Chloraima, 210.
Chlorophyll, 45, 97.
Chondracanthus gibbosus, 237-241.
Chromatophores, 445.
Cicadae, 365, 377.
Cidaris, 487.
Cilia, 29, 73.
Ciliata, 93-101.
Circulatory apparatus, 56.
Cirripedia, 221, 253.
Cladocera, 242.

Classification of living forms, 23.
Clepsine, 190-192,568-571.
Climate in relation to animal life, 69.
Cliona, 107.
Clionidae, 110.
Clypeastroida. 4S9.
Cockroach, 348.
Codonoeca, 90.
Codonellida, 98.
Codosiga, 90.

Coelenterata, 45, 50, 51, 56, 102, 109, 110, 115,
574,586. ''..,.

Coenurus, 185.

Cold, action of; on living matter, 12.

Coleochaete, 46.

Coleoptera, 366, 375. 377.

Collembola, 220, 262, 303.

INDEX.

591

Collosphapra, 85.
Colpoda, 62, 95, 96, 9S.
Comatula, 36, 37.

(Antedon), 500.

Conjugation. 31, 74.
Contractile tissue, 29.

vacuole, 73.

Copepoda. 63, 234. 230, 300-302.

Coral ligena, 188, 574.

Corallines, 390.

Corallite, 139.

Corallium rubrum, 144.

Corals, 110.

Cordylophora, 36, 37.

Coryne, 118.

Crayfish, 276, 284-286.

Crickets, 376, 378.

Crlnoidea, 466, 497.

Crocodilia, 26.

Crustacea, 21, 26, 58, 65, 222-225, 573.

Cryptogamia, 32.

Crytophialus, 261.

Ctenobrauchia, 439, 444.

Ctenophora, 53, 63, 68, 138. 153, 162, 192, 574.

Cucullanus elegans, 548, 550.

Cuma Eathkii, 308.

Cumacea. 264, 308.

Cunina, 568.

rbododactyla, 136.

Cyamus, 313.
Cyan sea, 36.

capillata, 134.

Cyclops, 234, 235, 550.
Cyclostomata, 439.
Cvdippe (PleurobracMa), 155.
Cymothoa, 314, 317.
Cymothoadas, 316.
Cynthia, 299.
C'ypraea Europaea, 420.
Cypris, 251, 252.
Cystic worm, 186.
Cysticercus, 184, 186.
Cystidea, .508.
Cythere, 251, 252.

D

Dalmanttes, 227.
Daphnia, 247, 248.
Decapoda, 462.

Deep-sea fauna, 26, 68, 80, 81.
Dendroccela, 160.
Dentalidre, 430, 431.
Dentalium, 422.
Dermis (enderon), 55.
DesmidiiE, 89.
Development, 17, 19, 66.
Diatomaceae, 12, 75, 80, 69.
Dibranchiata, 460-4C6.
Diceras, 406.
Dicorj^ne conferta, 119.
Dictyocystae, 76.
Dictyocystida, 98.
Dicyema, 579.
Dicyemida, 558, 577, 578.
Didemnum styliferum, 526i.
Dldinium, 98.

serpula, 13.

Differentiation, 20.

Dimyaria, 412.

Diphydae, 131.

Diphyes appendiculata, 126.

Diphyllidea, 1557.

Diphyozooid, 126, 131.

Diplozoon paradoxum, 33, 182.

Dipnoi, 60.

Diporpa, 33, 182.

Diptera, 366, 375, 381.

Discophora, 118, 121, 132-135.

Disintegration of living mattei', 10.

Distoma, 179.

Distribution. 16, 24-26, 67-69.

Dog-louse, 187.

Dogs, retrieving of, 35.

Doliolum, 514, 518, 523.

denticulatum, 529.

" Double circulation,'" 60.
Dragon-flies, 221.
Dysteria, 98.

IE

Eakthworm. 193.
Echeneibothrium, 187.
Echinidea, 56, 466, 4S5, 4SS, 489.
Echinococcus, 184.

veterinorum, 185.

Echinoderes, 171.

Echinodermal series. 581, 582.

Echinodermata, 26, 30, 53-55, 466, 577, 586.

Echinoida, 489.

Echinopaedium, 54. 406, 481, 505, 582.

Echinorhynchus, 553.

Echinus, 486.

sphsra, 487, 488.

Ectoderm, 55, 56.
Ectoprocta, 394, 571, 572.
Ectosarc, 74.
Edrioasterida, 508.
Edriophthalmia, 310.
Elytron, 204.
Embryology, 42, 50, 582.
Empusa, 45.
Endoparasites, 182.
Endoplast, 48, 74.
Endoplastica, 73, 82.
Endoprocta, 571.
Endosarc, 74.
Endostyle, 511.

Enteropneusta, 59, 538, 576, 581.
Entoconcha mirabilis, 440.
Entogastrlc gemmation, 135, 568.
Entomostraca, 224, 234.
Entoprocta, 394.
Eozoon, 72.

canadense, 82.

Epiblast, 21, 51.
Epidermis (ectoderm), 56.
Epigenesis, 19.
Epimera, 268.
Epizoa, 237.
Equidae, 26.
Ergasilus, 241.
EristaUs floreus. 868.
Errantia, 206, 207.
Ervilia, 98.
Estheria, 248-250.
Euaxes, 199, 569, 570.

593

INDEX.

Euglena, 12,

viridis, 90.

Euphausia, 307.
Euplectelia, 110.
Eurypterida, 232, 234.
Eurypterus remipes, 233.
Evolution, 40.

F

Families, 23.

Fauna, oldest known, 72. See Fossils.

Faunae, dissimilar, 24.

Fecundation, 33, 34.

Ferns, 21.

Fibrospongia, 109, 110.

Fishes, 59-65.

Fish-lice, 237.

Fission, 31.

Flagellata, 89-95.

Flagellum, 73.

Fleas, 366.

Flies, 366,

Florae, dissimilar, 24.

Floridese, 33.

Flower-buds, 21.

Food-vacuole, 90.

Food-yelk, 32.

Foraminifera, 48, 63, 77-82, 85, 86, 574.

Fossils ; succession of species, distribution,

etc.. 24, 25, 40, 43, 6S, 71, 81, 136, 158, 225,

253, 263, 310, 317, 342, 416, 413, 459, 403,

438, 507.

Cambrian, 82.

carboniferous, 508.

cretaclous, 82, 110, 417.

devonian, 250, 432, 459.

laurentian, 82.

lias, 465.

limestone, 81, 82, 153.

mummulitic, 82.

silui-ian, 82, 137, 232, 251, 397, 431, 436,

507.

Solenhofen slate, 136.

trias, 465.

Fowl, 33.

Fringing'-reefs, 150.
Functions, 27, 54.
Fungi, 12, 20, 27, 32, 38, 45.
Fungidap, 146, 151.

G

Galeodes, 326.

Gamogenesis, 32.

Ganoids, 60.

Gasteropoda, 404, 432.

Gasterostomum, 1 82.

Gasterotricha, 170.

Gastraea, 51.

Gastrophysema, 107, 110.

Gastrula, 107.

Gecarcinus, 295.

Gemmation, 30, 525.

Generation, 30-32.

Genus, 23.

Gephyrea, 59, 189, 215, 570, 573, 574.

Geryonia, 568.

Geryonidae, 117,135.

Glass-crabs, 308.

Globigerina, 40, 79-82.
Glossocodon, 115.
Gnathites, 224, 236.
Gomphonema, 75, 96.
Gonodactylus, 319.
Graafian follicles, 66, 3S0.
Graptolites. 137.
Gregarina, 73, 87, 88.

gigantea. 88, 89, 96.

Gregarinidae, 86-88, 574.
Gromia, 78.
Gromidae, 79.

Growth of animals and plants, 10.
Gymnolaemata, 895.
Gymnophthalmata, 118.
Gymnosomata, 435-487.
Gyrodactylus, 1S2.

£1

Haltotis, 423.

Haliphysema. 107, 110.

Halisarca, 107, IKt.

Heat, effect of, on living matter, 11.

Hectoeotylus, 461.

Helicidae, 442.

Heliopora caerulea, 148.

Heliozoa, 563, 574.

IlelLx, 428, 442.

pomatia, 443.

Hemiptera, 365, 366.
Hereditary transmission, 35, 41,
Hermaphrodites, 182, 192, 198, 224, 413, 442,

484,551.
Heterom orphan, G9.
Heteronereis, 215.
Heteropoda, 424, 426, 439.
Heterotricha, 95.
Hexacoralla, 145-147.
Hippuritidae, 417.
Hirudmea, 189, 190, 192, 194, 213. 569, 574,

575.
Hirudo medicinalis, 190, 191.
Histology, 16.
Histriobdella, 189-192.
Holomyaria, 540.
Holothuria, 153.
Holothnridea, 59, 466-468.
Holotricha, 95.
Homarus, 66.
Humming of insects, 223.
Hyalonema, 110.
Hydatina senta. 168.
Hydra, 56, 62. 63, 115, 118, 585.
Hydractina. 65, 115.
Hydranth. 110.
Hydrophilus piceus. 363.
Hydrophora. 118, 132.
Hydrophyllia, 117.
Hydrosoma, 116, 117.
Hydrotheca, 117.
H^drozoa, 36, 65, 107, 110, I32-13S, 153-155,

574, 575.
Hymenoptera, 369.
Hopoblitst, 21. 51.
Ilypotricha, 95.

ICHTHTOPSTDA, 57.

Idoteidie, 315.

INDEX.

593

Imperforata, 79.

Impregnation, 81.

Inarticulata. 403. „ „„ „^ «^

Infusoria, 12, 20, 33, 45, 48, 74, 77, 89-91, 94,

99-105, 157, 505, 574.

ciliata. 89.

flagellata, 89, 90.

tentaculiiera. 89.

Insecta, 21, 59, 67. 224, 316,342, 372-386, 5i3.

Insectivorous plants, 44.

Integumentary organs, 55.

Invertebrata, morphological types among,

49.
Isocardia, 406.
Isopoda, 313.
lulus, 342.
Ixodes ricinus, 830.

Janelltd^, 441.
Jaws, 50.
Jelly-fishes, 110.

LABitTM, in insects, 203.

Lacinularia, 108-171.

Laemodipoda, 313, 316.

Lamellibranchiata, 404, 405, 573.

Lampyris splendidula, 379.

Laomedea, 120.

Larvs, 66, IW, 166, 169, 170, 179, 213, 214,

247, 253, 319, a32, 339, 365, 371, 375, 385-

388, 402. 403, 467, 550, 568, 582.
Leeches, 189, 570.
Lepadidae. 260.
Lepas, 254-260.

australis. 25S.

Lepidoptera, 366, 375, 377, 381.

Leptoplana, 162.

Lerna?. 241.

Lerna?odiscn8 porcellanae, 262.

Leucifer, 299.

Leucones, 106, 110.

Lice, 363.

Lieberkuhnia, 78.

Ligula, 182.

Lima. 408.

Liraacidse, 440.

Limax, 423. 427-429.

Limnetis. 247-249.

brachyurus, 249.

Limpets, 438.
Limulus. 22S-235, 823.

moluccanus. 228.

polyphemus, 231.

Lineus, 1 65.

Linguatula. 320, 334.

Lingula, 397,401.

Lithooysts. 115.

Lituitidae. 79.

Living matter, properties of, 9^2.

Lobster, 264.

Locustidse, 378.

Loxosomma, 894, 416. 571.

Lucernaria, 122, 123, 132, 137, 138.

Lumbricus, 193-195.

Lungs, 59.

Lymnspus, 427, 429.

palustris, 423.

Macbobiotus Schultzei, 333.
Macrostomum, 158-160.
Macrura, 281, 293-299.
Madrepores, 147.
Madreporite, 4^9.
Magosphaera, 89.
Malacobdella, 189-192, 570, 575.
Malacoscolices, 576, 580.
Malacostraca, 224, 264.
Malacozoic series, 5S0.
Mallophaga, 352, 363.
Manubrium, 116.
Mastigopods, 73.
Meandrina. 151.

MedusiP, 36. 37, 110, 115-113, 575.
Megalopa, 802, .803.
Meromyaria. 549.
Merostomata, 224, 227.
Mesoblast, 21.
Mesoderm, 55, 56.

Mesotrocha. 214.

Mesozoa, 578.

Metabola, 361.

Metamorphosis, 66, 386.

Mecazoa, 48, 51-58, 102, 110, 166, 171, 578, 582,
5S3, 585.

Microstomum, 163.

Miliolidae, 79.

Millepores, 147, 148, 151, 153.

Millipedes, 337.

Mites. 329.

Moisture, effect of, on living matter, 11.

Molar motion, 27.

Mollusca, 55-61, 76, 889, 404, 572, 573, 580,
.581.

Mopads. 38, 39. 46. 77, 85. 89, 90, 103.

Monera, 73. 77. 85, 574, 578, 582.

Monomyaria, 412.

Monostomum mutabile. 179.

Morphological species, 22.

Morphology, 16.

^lorula, 48.

Moths. 33.

Mucor, 88.

Munna, 310.

Muscular tissue. 27, 29.

Mussel, 407, 415'

Mygale Blondii, 828.

csementaria, 827.

Myriapoda, 59, 224, 842, 573.

Mysis, 291, 299, 303, 317.

M.vtilus, 409, 417.

Myxastrum 77, 86.

MjTcodictyum, 75, 77.

Myxomycetes, 13, 46, 86.

Myxopods, 73, 75, 82, 83.

Myxospongiae, 109, 110.

Myzostomata, 537, 574.

N

Nair, 193, 194.

Naked-eyed medusae. 118.

Nauplius, 234, 287, 247, 253, 258-263, 300-802,

307.317,331,333.
Nautilus. 64, 65. 69, 447-4C5.
Nebalia. 242, 247, 248.
Nematoidea, 32, 545, 575.
Nematophores, 119.

594

INDEX.

Nematorhyncha, 509, 575.
I^euiatoscolices, 575, 5b0.
Nemcrtidae, 105, 100.
Neomenia, 571.
Jfephelis, 192.
Nerve, 27, 29.
Nervous system, 61,
Neuroptera, 300, 876, 377.
Noctiluca, 40, 90-92.
Notodelphys, 241, 242.
Notommata tardigrada, 170.
Nova Scotian coal fossils, 842.
Nucleus, 17. 48, 74.
Nucula, 408, 416.
Nndibranchiata, 424, 437, 438.
Nullipore, 15l), 151.
Nunimulites, 79.
Nyctotherus, 97, 98.

OCTOCORALLA, 143, 146.

Octopoda, 452, 460.

Odontophora, 404, 417, 419, 573.

Olipochasta, 60, 189, 198, 198, 199, 213, 574.

Onchidura, 60.

Oniscidie, 816.

Oniscus, 219, 313.

Ophiodes, 119.

Ophiolepis ciliata. 4S4.

Ophiuridea, 406, 482.

Ophrydidae, 100.

Opistliobranchiata, 437, 438.

Opisthomum, 159.

Orbulina, 79.

Orders, 23.

" Organized," meaning of, 15.

Organs, 28, 64, 65.

Origin of living matter, 41.

" Origin of Species," 37.

Orthidse, 403, 404.

Orthoptera, 864, 306, 875, 377, 381, 383.

Ossicnla auditus. 65.

Ostracoda, 221, 251,

Ostraea, 408-417.

Oviparous animals. 67.

Ovoviviparous animals. 67.

Oxidation, waste of living matter by, 10.

Oxyuris, 547.

Oyster, 408.

Pagttrid^, 299.

Palseocyclus, 153.

Palaeontology. See Fossils.

Pallnurus vulgaris, 293.

Paludina, 180, 181, 425, 426.

Pangenesis, 41 .

Paramoecium, 48, 96-100.

Parasites, 45, 67, 171, 181-1 F3, 187, 237, 241,

242, 253, 262, 263, 314, 331, 334, 863, 387,

416,440,650,553.
Pasteur's experiments, 12, 13.
Patellidae. 438, 444.
Pecten, 408-416.
Pectostraca, 253.
Pedalion, 170, 171.
Pedicellina, 571, 582.
Pedicels, 64.

Pediculina, 362, 8C8.
Pelagia, 132, 138.
Peltogaster paguri, 262.
Peneus, 80U-8U2.
Peniciliium, 45.
Pennatulidae, 145.
Pentacriuus. 600.
Pentastomida, 226, 284, 573.
Pentastonmm taenioides, 385.
Pentremites, 509.
Perennibranchiata. 58.
Perforata, 79, 146-150, 158.
Peridineae, 70, 93.
Peripatidea, 226, 534, 573.
Peritricha, 96, 96, 100.
Perla nigra, 304.
Peroniadae, 440.
Peronia verruculata, 443.
Peronospora, 46, 46.
Phalangidse, 326.
Phallusia, 616, 620.
Pharyngopneusta, 577.
Pbaryngopneustal series, 581.
Pholas, 406, 408, 417.
Phoronis, 217, 218.
Phrosina, 815.
Phrynidae, 826.
Phylactolaemata, 572.
Phyllodoce, 214.

viridis, 211.

Phyllopoda, 242.
Phyllosomata, 307, 303.
Phylogeny, 42.
Physalia, 115, 12^ 130.
Physemaria, 552, 574, 575, 585.
Physiology, 9, 16, 26.
Physophoridae, 117. 128-132.
Pilidium gyrans, 165, 166.
Pisidium, 414. 416.
Placenta, 67, 101.
Piagiostome fishes. 67.
Planaria, 161, 192, 575.

dioica. 161.

Plants, 31-88, 44, 68.
Pleurobranchia, 155.
Pleurodictyon, 153.
Plumatella repens, 300, 391.
Plumularidae, 118.
Pocillopora, 148.
Podophrya fixa, 94.
Podophthalmia, 230, 264.
Poduridae, 882.
Poecilopoda, 228.
Pohan vesicles, 468, 483.
Polvarthra, 171.
Polycelis, 192.

laevigata, 161.

Polychaeta, 66, 189, 199, 200, 207, 574, 575.

Polycistina, 83.

Polygordius. 575.

Polykricos, 96.

Polymyaria, 549.

Polynue, 200.

lunulata, 210.

squamata, 200-207, 212.

Polyophthalmus, 200, 212.
Polypes, 31, 62, 110.
Polypide, 890.
PolVpite, 110.
Polyplacophora, 430, 433, 573.

INDEX.

595

Polyzoa, ^3, 56, 380, 571, 5T2, 576.

Polyzoarium, 3J0.

Pontellidae, 237, 313.

Porifera, 51, 55, 62, l.j2, lOD, 574.

Porpita, 51, 127.

Porites, 151.

Prawn, 300.

Priapulus, 216.

Primordial utricle, IS.

Proctucha, 15S, 102.

Prod ictidae, 403, 404.

Proglottis, 136.

Prosobranchiata. 438, 430, 414.

Protamoeba, 75, S6-SJ.

Protein, 9.

Proteolepas bivincta, 261.

Proteus, 60.

animalcules, 86.

Protococcus, 12, 17.

Protogenes, 75-7S, 35, 83.

ProLoraonas, 77, 85, 89.

Protomyxa aurantiaca, 76.

Protoplasm, 9, 14,

Protoplasta, 86, 574.

Protozoa, 33, 47, 48, 54, 61, 62, 73, 102, 103,

103, 514.
Protozoic series, 579.
Protula, 207, 215.

Dysteri, 2o8.

Provinces of distribution, 24.
Pseud-hiiemal system, 57.
Pseudo-filaria, b3, 89.
Pseudo-navicella, 88.
Pseudophyliidea, 137.
Pseudopodia, 29, 73, 75.
Pseudo-scorpions, 326.
Psychology, 9.

Ptcropoda, 53, 68, 401, 424, 432-434.
Ptery^otus, 233.
Puhcidae, 331.

Palmogasteropoda, 59, 424, 433.
Pulmonary sacs, 59.
Pulmonata, 423-429, 440.
Pulvinularia, 79, 80.
Pupipara, 367.
Pycnogonida, 331, 573.
Pyrosoma, 514, 528.
giganteum, 523.

R

Eadiolari.v, 12, 46-43, 68, S0-S3, 93, 564, 574.

Eedia, 179, 131.

Reef-builders, 149.

Eenierinae. 103.

Keproductive system, 65.

Respiratory system, 53.

Rliabdocoela, 159.

Rhabdopleura, 396, 397.

Rhachis. 543.

Rhizocephala, 2.53, 263.

Rhizocrinus lofotensis, 498.

Rhizostomidae, 124.

Rbodope, 425.

Rhopalodina, 470.

Rhynchocoele turbellaria, 163.

Rhynchonella. 398, 400.

Rhynchonellidae, 403.

Rock-builders, 81, 82.

Rotalia, 78.

Eotifera. 12, 56, 89, 157, 162, 166, 574.
Rugcsa, 148, 149, 153.

Sacgttlina purpurea, 232.
Sagitta. 6S, 542, 577.
Salpa. 36, 514, 518, 531, 532.
Salpingoeca, 90.
Sarsia prolifera, 1 20.
Saxicava, 408.
Scallop. 408.

Scalpellum vulgare, 260-262.
Scaphopoda, 430-433.
Schizocoele, 51, 52, 219.
Schizopoda, 293, 299.
Scolopendra borbonica, 337.

Ilopei, 338.

Scorpio afer, 321.

Scorpions. 59, 320, 323, 324.

Scrupocellaria ferox, 3J2.

Scyllarus, 294.

Sea anemones, 113.

Sensitive plant, 44.

Sensory organs, 27.

Sepiadae, 4i6,452, 463, 465.

Sepia officinalis, 446, 453.

Serial relations of invertebrata, 578,

Serpulidaj, 207,214.

Sertularidae, 117, 118.

Shrimp, 3v»3.

Siphonophora, 118, 127, 133.

Sipunculiis nudus, 216-218.

Snail, 442.

Solenhofen slates, 136.

Somatopleure, 57.

Somites, 200.

Sounds from insects, 376.

Spat.angoida, 439.

Spermatophores. 451.

Sphaeromidae, 315.

Sphierozoum ovodimare, 84, 85.

punctatum, 84.

Sphinx ligustri, 367.
Spiders, 326.
Spiriferidae, 403.

Spirillum volutans, 12.
Spirorbis. 207.

Spirostomum, 97.

Spirulidae, 463.

Splanchnopleure, 56, 5T.

Spongida, 102, 567.

Sponsrilla, 104, 111.

fluviatilis, 103, 107.

Sporocysts, 182.

Springs, hot, living things in, 14.

Squids, 46:3.

Squilla, 312, 319.

scabricauda, 318, 319.

Star-fish, 474.

Stentor, 100.

Stephanoceros, 167, 169, 170.

Sternaspis, 21.5.

Stigmata, 59, 325, 389, 356, 374.

Stings of insects, 372.

Stomatopoda, 237. 317.

Stone corals, 147, 149.

Strepsiptera, 371, 3b7.

Strombidium, 94.

" Struggle for existence," 80.

596

INDEX.

Stylifer, 440.
Stylonychia, 100.
Stylops aterrimus, 88T.
Bun-animalcule, b2.
Sundew. 44.

'• Survival of the fittest," 41.
Sycandra raphauus, 5(58.
Sycon, 110.

ciiiatum. 100.

Syllis, 214, 215.

vittata, 211.

Synapta, 440, 467.

digituta and Lnhserens, 471.

Syncytium, 105.
Syrphus ribesii, 363.

Tabttlata, 146, 148, 150.

Taenia, 163-1 BS.

Tape-worms, 182.

Tardigrada, 2-25, 334, 573.

Taxonomy, 16, 22, 561.

Teeth, 56.

Tegumentary system, 55.

Telotrocha, 164.

Temperature in relation to living matter,

11, 39.
Tentacula, 51.
Tentacuiifera, 93.
Terebratula, 40.

psittacea, 40.

Terebratulidifi, 404.

Terebratulina septentrionalis, 400.

Teredo, 406, 417.

Testacellidae, 440.

Tetrabranchiata, 455.

Tetraphvllidea, 187.

Tetrarhvnchus, 183, 187.

Tetrastemma, 163, 164.

Teuthidae, 452, 463, 465.

Thecosomata, 435-437.

Thysanopoda, 299.

Thysanura, 220, 362, 363.

Ticks, 329.

Tissues, 17.

Tomopteris, 207.

Torquatella, 98.

Torula, 38.

Tracheae, 59.

Tracheo-branchiae, 221.

Trachynemata, 135.

Tradescantia hair, 75.

Trematoda, 53, 56, 157, 171-173, 182-188, 190,

574, 575.
Tremoctopus, 462.
Triarthra, 170,171.
Trichina, 550. 551.
Trichocvsts, 97.
Trichodida;, 100,
Trichodina grandinella, 96.
Trichoscolices, 575, 580, 581.
Trigonia, 69. 416.
TrUobita, 220, 224, 225, 573.

Trochus cinerarias, 420.

Tubicola, 207.

Tubifex, 193, 199.

Tubipora, 146.

Tubu'.aridae, 118, 132.

Tunicata, 53, 55, 59, 67, 510. 576, 581.

Turbellaria, 45, 51, 56-61, 65, 157, 573-575.

Tylos, 315.

Types, morphological, 49.

Typhlosole, 196, 518.

U

Unio pictoruin, 415.
Uropoietic system, 61.

YAGiXFLrs, 429.

Yampyrella, 75, 96,

Vanessa atalanta, 367.

Velella, 127.

Ventriculites, 110,

Veronicellidse, 440.

Vertebrata and Invertebrata, 49.

Vibracula, 3U3.

Vibrionidae, 89.

Vital force, 15. 16.

Viviparous animals, 67.

Volvocincae, 89.

Volvox, 89, 579.

Vorticellidae, 5, 33, 48, 62, 61, 91-101.

"Waldheimia, 898.

australis, 399.

"Waste of living matter, 10.
Water in living matter, 10.
Willsia, 120, 121.
Wolffian duct, 61.

X

Xtpmosttra, 228, 232.

Yeast, 12.

Yeast-plant, 45.

Z

ZoJEX. 302, 303.

Zoanthidae, 145, 146.

Zoanthodeme, 138.

Zoological chronology and geography, 70,

Zoophyta, 574.

Zoophytic series, 5S0

ThoiDas H. Hflxley's forks.

Science and Culture, and other Essays. I2ma

Cloth, price, $1.50.

The Crayfish : An Introduction to the Study of Zoology.
With 82 Illustrations, (*' International Scientific Series.")
12mo. Cloth, $1.75.

Science Primers : Introductory. l8mo. Flexible Cloth,

45 cents.

Man's Place in Nature. l2mo. Cloth, $1.25.
On the Origin of Species. i2mo. Cloth, $i.oo.
More Criticisms on Darwin, and Administrative

Nihilism. 12mo. Limp cloth, so cents.

Manual of the Anatomy of Vertebrated Animals.

Illustrated. 12mo. Cloth, $2.50.

Manual of the Anatomy of Invertebrated Animals.

12mo. $2.50.

Lay Sermons, Addresses, and Reviews. l2mo.

Cloth, $1.75.

Critiques and Addresses. i2mo. Cloth, $1.50.

American Addresses ; with a Lecture on the Study of
Biology. 12mo. Cloth, $1.25.

Physiography: An Introduction to the Study of Nature.
With Illustrations and Colored Plates. 12mo. Cloth, $2.50.

Huxley and Youmans's Elements of Physiology

and Hygiene. By T. H. Huxley and W. J. Youman^

12mo. $1.50.

For sale by all booksellers ; or sent by mail, post-paid, on receipt of
price.

D. APPLETON & CO., Publishers,

1. 3. & 5 Bond Street, Njsw York.

Charles Darwin's Works.

Origin of Species by Means of Natural Selection, or the
Preservation of Favored Races in the Struggle for
Life. New and revised edition, with Additions. 12mo. Cloth,

$2.00.

Descent of Man, and Selection in Relation to Sex. With
many Illustrations. A new edition. 12n.o. Cloth, |3.00.

Journal of Researches into the Natural History and Geol
ogy of the Countries visited during the Voyage of H.
M. S. Beagle round the World. A new edition. 12mo.
Cloth, $2.00.

Emotional Expressions of Man and the Lower Animals.

12mo. Cloth, $3.50.

The Variations of Animals and Plants under Domestioa*
tion* \V' ith a Preface, by Professor Asa. Gray. 2 vols. I'dus-
trated. Cloth, $5.00.

Insectivorous Plants. 12mo. Cloth, $2.00.

Movements and Habits of Climbing Plants. "With Ulustra.
tions. 12ino. Cloth, $1.25.

The Various Contrivances by which Orchids are Fertil-
ized by Insects* Revised edition, with Illustrations. 12mo.

Cloth, $I.'75.

The Effects of Cross and Self Fertilization in the Vegeta-
ble Kingdom. 12mo. Cloth, $2.00.

Different Forms of Flowers on Plants of the same Species.

With Illustrations. 12mo. Cloth, $1.50.

The Power of Movement in Plants. By Charles Darwin,
LL. D., F. R. S., assisted by Francis Darwin. With Illustrations.
12mo. Cloth, $2.00.

The Formation of Vegetable Mould, through the Action
of Worms. With Observations on their Habits. With Illustra-
tions. 12mo. Cloth, $1.50.

fl'or sale by all booksellers ; or sent by mail, post-paid^ on receipt of price.

Now York : D. APPLETON & CO.. 1, 3. & 5 Bond Street.

JOHN TYNDALL'S WORKS.

Essays on the Floating Matter of the Air, in Re-
lation to Putrefaction and Infection. 12mo, cloth, $1.50.

On Forms of Water^ in Clouds, Rivers, Ice, and Glaciers.
With 35 Illustrations. 12mo, cloth, $1.50.

Heat as a Mode of Motion. New edition. i2rao,

cloth, $2.50.

On Sound : A Course of Eight Lectures delivered at the
Royal Institution of Great Britain. Illustrated. 12mo.
New edition. Cloth, $2.00.

Fragments of Science for Unscientific People. i2mo.

New revised and enlarged edition. Cloth, $2.50.

Light and Electricity. i2mo, cloth, $1.25.
Lessons in Electricity, 1875-76. i2mo, cloth, $i.oo.
Hours of Exercise in the Alps. With Illustrations.

12mo, cloth, $2.00.

Faraday as a Discoverer. A Memoir. i2mo, cloth,

$1.00.

Contributions to Molecular Physics in the Domain of

Radiant Heat. $5.00.

Six Lectures on Light. Delivered in America in 1872-
'73. With an Appendix and numerous Illustrations.
Cloth, $1.50.

Farewell Banquet, given to Professor Tyndall, at Del-
monico's, New York, February 4, 1873. Paper, 50 cents.

Address delivered before the British Association, assembled at
Belfast. Revised, with Additions, by the author, since the
Delivery. 12mo, paper, 50 cents.

New York : D. APPLETON & CO., 1,3, & 5 Bond Street.

ERNST HAECKEL'S WORKS.

THE HISTORY OF CREATION;

Or, The Development of the Earth and its Inhabitants by the
Action of Natural Causes. A Popular Exposition of the Doc-
trine of Evolution in general, and of that of Darwin, Goelhe, and
Lamarck in particular. From the German of Ernst Haeckel, Pro-
fessor in the University of Jena. The translation revised by Pro-
fessor E. Ray Lankester, M. A., F. R. S., Fellow of Exeter College,
Oxford. Illustrated with Lithographic Plates. In two vols., 12mo.
Cloth, $5.00.

THE EVOLUTION OF MAN.

A Popular Exposition of the Principal Points of Human Ontogeny
and Phylogeny. From the German of Ernst Haeckel, Professor in
the University of Jena, author of " The History of Creation," etc.
With numerous Illustrations. In two vols., 12mo. Cloth. Price,

$5.00.

" In this excellent translation of Professor Haeckel's work, the Eng-
lish reader has access to the latest doctrines of the Continental school of
evolution, in its application to the history of man. It is in Germany, be-
yond any other European country, that the impulse given by Darwin
twenty years ago to the theory of evolution has influenced the whole
tenor of philosophical opinion. There may be, and are, differences in
the degree to which the doctrine may be held capable of extension into
the domain of mind and morals ; but there is no denying, in scientific
circles at least, that as regards the physical history of organic nature
much has been done toward making good a continuous scheme of being."
— London Saturday Review.

FREEDOM IN SCIENCE AND TEACHING.

From the German of Ernst Haeckel. With a Prefatory Note by
T. H. Huxley, F. R. S. 12mo. $1.00.

For sale by all booksellers; or sent by mail, postpaid, on receipt of price.

New York : D. APPLETON k CO., 1, 3, & 5 Bond Street.

^X\ >■

AaJK ^