Full text of "Lectures on the elements of comparative anatomy"

See other formats

THE ELKMENTS OF COMPARATIVE ANATOMY.

LECTURES

ON THE

ELEMENTS OF COMPARATIVE ANATOMY.

THOMAS HENRY HUXLEY, F.R.S.,

PROFESSOR OF NATURAL HISTORY, ROYAL SCHOOL OF JUNES ; AND PROFESSOR OF COMPARATIVE ANATOMY AST)
PHYSIOLOGY TO THE ROYAL COLLEGE OF SURGEONS OF ENGLAND.

7E25S

■■■■■B

— ^— ZT

~~^

: D

SSSS^S cO

~ P_3

~ IMHi^^^^

» 1— 1

KM^BHH 1. 1

■™i^™ r- )

1^=

ON THE CLASSIFICATLON OF ANIMALS

AND

ON THE VERTEBRATE SKULL.

LONDON:
JOHN CHURCHILL AND SONS, NEW BURLINGTON STREET.

MDCCCLXIV.

33 H- ^

LONDON: a>RINTEI) BY WILLIAM CLOWES AND SONS, STA3IKORD STREET

AM) CHARING CROSS.

DEDICATION.

-*o*-

To GEORGE BUSK, Esq., F.R.S.

My dear Busk,

I beg leave to dedicate these Lectures of mine to you,
partly as my old and staunch friend, and partly as a member
of the Council of the Royal College of Surgeons, and therefore
a representative of the body to which I am indebted for the
opportunity of delivering them.

Ever

Yours very faithfully,

T. H. HUXLEY.

The Royal School of Mines, Jermyn Street.
March 1th. 18G4.

PREFACE.

The present work contains, substantially, the Lectures which
I delivered, in the spring of 1863, at the Royal College of
Surgeons of England, in discharge of my duty as Hunterian
Professor of Comparative Anatomy and Physiology to the
College. I purpose, should I continue to hold that honourable
office, to publish the substance of subsequent courses in a
similar manner ; and by that process to bring out, eventually,
a comprehensive, though condensed, systematic work on Com-
parative Anatomy. In intention, therefore, the volume now
before the reader is the first of a series, to be followed in
due order by a second, " On Man and the other Primates ;" a
third, on the remaining Mammalia, and so on. Whether this
intention will ever be fully carried out depends on so many
contingencies, however, that I have thought it the better
course to let each volume remain, in form, independent of
the rest.

I have much pleasure in expressing my obligations to the
President and Council of the Royal College of Surgeons for the
unrestricted access which I have enjoyed to the abundant
anatomical treasures contained in their Museum and its store-
rooms, to which my Lectures are indebted for many of their
most instructive illustrations.

b 2

VI PREFACE.

Nor can I conclude without a word of thanks to the young
artist, Mr. W. H. Wesley, to whose skill, painstaking, and
thorough faithfulness, the excellence of most of those illustra-
tions is due. It will be understood that, when the contrary is
not expressly stated, the drawings are original.

The Lectures differ, by the omission of the inaugural Dis-
course, which would have been out of place in this volume of
the work I contemplate ; by additions to others ; and by a
change of arrangement, from those which were published in

the " Medical Times and Gazette."

T. H. H.

London, March 7th, 1864.

CONTENTS.

-»o*-

LECTURES I. to VI. ON THE CLASSIFICATION OF ANIMALS

LECTURE I.

PAGE

THE GREGARINIDA, RHIZOPODA, SfONGIDA, AND INFUSORIA .... 1

LECTURE II.

THE HYDROZOA, ACTINOZOA, POLYZOA, BRACHIOPODA, ASCIDIOIDA,
LAMELLIBRANCH1ATA, BRANCHIOGASTEROPODA, PULMOGASTERO
PODA, PTEROPODA, AND CEPHALOPODA 20

LECTURE III.

THE ECHINODERMATA, SCOLECIDA, ANNELIDA, CRUSTACEA, ARACH-

NIDA, MYRIAPODA, AND INSECTA 42

LECTURE IV.

THE VERTEBRATA ; OR PISCES, AMPHIBIA, REPTILIA, AVES, AND MAM-
MALIA 58

CONTEXTS.

LECTURE V.

PAGE

THE ARRANGEMENT OF THE CLASSES INTO LARGER GROUPS . . . 73

LECTURE VI.

THE SUBDIVISIONS OF THE MAMMALIA LARGER THAN ORDERS . . 87

LECTURES VIL to XIV. ON THE VERTEBRATE SKULL.

LECTURE VII.

THE STRUCTURE OF THE HUMAN SKULL 113

LECTURE VIII.

THE DEVELOPMENT OF THE HUMAN SKULL 136

LECTURE JX.

THF SKULL OF THE PIKE COMPARED STRUCTURALLY AND DEVELOP-
MENTALLY WITH THAT OF MAN 1H2

NOCTURE X.

TDK SKULLS OF FISHES 180

LECTURE XI.

THE SKULLS OF FISHES AND AMPHIBIA 202

CONTENTS. XI

LECTURE XII.

PAGR

THE SKULLS OK KEIT1LIA AND AVIS 211)

LECTURE XIII.

THE SKULLS OF MAMMALIA 245

LECTURE XIV.

NIK THKORY OF TIIK VERTEBRATE SKULL .... ... 278

LECTURES

ON THE

ELEMENTS OF COMPARATIVE ANATOMY.

LECTURE I.

ON THE CLASSIFICATION OF ANIMALS.

THE GREGARINIDA, RIIIZOPODA, SPONGIDA, AND INFUSORIA.

By the classification of any series of objects, is meant the
actual, or ideal, arrangement together of those which are like
and the separation of those which are unlike ; the purpose of
this arrangement being to facilitate the operations of the mind
in clearly conceiving and retaining in the memory, the cha-
racters of the objects in question.

Thus, there may be as many classifications of any series of
natural, or of other, bodies, as they have properties or relations
to one another, or to other things ; or, again, as there are
modes in which they may be regarded by the mind : so that,
with respect to such classification as we are here concerned
with, it might be more proper to speak of a classification than
of the classification of the animal kingdom.

The preparations in the galleries of the Museum of this
College are arranged upon the basis laid clown by John Hunter,
whose original collection was intended to illustrate the modihca-
tions which the great physiological apparatuses undergo in the

2 ON CLASSIFICATION.

animal series : the classification which he adopted is a classifica-
tion by organs, and, as such, it is admirably adapted to the
needs of the comparative physiologist.

But the student of the geographical distribution of animals,
regarding animated creatures, not as diverse modifications of the
great physiological mechanism, but in relation to one another,
to plants and to telluric conditions, would, with equal propriety,
dispose of the contents of a Zoological Museum in a totally
different manner ; basing his classification, not upon organs, but
on distributional assemblages. And the pure palaeontologist,
looking at life from yet another distinct point of view, would
associate animal remains together on neither of these principles,
but would group them according to the order of their succession
in Time.

Again, that classification which I propose to discuss in the
present Lectures, is different from all of these : it is meant to
subserve the comprehension and recollection of the facts of
animal structure ; and, as such, it is based upon purely structural
considerations, and may be designated a Morphological Classifica-
tion. I shall have to consider animals, not as physiological
apparatuses merely ; not as related to other forms of life and to
climatal conditions ; not as successive tenants of the earth ;
but as fabrics, each of which is built upon a certain plan.

It is possible and conceivable that every animal should have
been constructed upon a plan of its own, having no resemblance
whatsoever to the plan of any other animal. For any reason we
can discover to the contrary, that combination of natural forces
which we term Life mi^ht have resulted from, or been mani-
fested by, a series of infinitely diverse structures : nor, indeed,
would anything in the nature of the case lead us to suspect a
community of organization between animals so different in
habit and in appearance as a porpoise and a gazelle, an eagle
and a crocodile, or a butterfly and a lobster. Had animals been
thus independently organized, each working out its life by a
mechanism peculiar to itself, such a classification as that
which is now under contemplation would obviously be impossi-
ble ; a morphological, or structural, classification plainly implying
morphological, or structural, resemblances in the things classified.

GREGARINIDA, RHIZOPODA, SrONGIDA, AND INFUSORIA. 3

As a matter of fact, however, no such mutual independence
of animal forms exists in nature. On the contrary, the different
members of the animal kingdom, from the highest to the
lowest, are marvellously interconnected. Every animal has a
something in common with all its fellows : much, with many of
them ; more, with a few ; and, usually, so much with several,
that it differs but little from them.

Now, a morphological classification is a statement of these
gradations of likeness which are observable in animal structures,
and its objects and uses are manifold. In the first place, it
strives to throw our knowledge of the facts which underlie, and
are the cause of, the similarities discerned into the fewest possible
general propositions — subordinated to one another, according to
their greater or less degree of generality ; and in this way
it answers the purpose of a memoria technica, without which the
mind would be incompetent to grasp and retain the multifarious
details of anatomical science.

But there is a second and even more important aspect of
morphological classification. Every group in that classification
is such in virtue of certain structural characters, which are not
only common to the members of that group, but distinguish it
from all others ; and the statement of these constitutes the defi-
nition of the group.

Thus, among animals with vertebra?, the class Mammalia is
definable as those which have two occipital condyles, with a
well-ossified, basi-occipital ; which have each ramus of the man-
dible composed of a single piece of bone and articulated with
the squamosal element of the skull ; and which possess mammaB
and non-nucleated reel blood-corpuscles.

But this statement of the characters of the class Mammalia is
something more than an arbitrary definition. It does not merely
mean that naturalists agree to call such and such animals Mam-
malia : but it expresses, firstly, a generalization based upon, and
constantly verified by, very wide experience ; and, secondly, a
belief arising out of that generalization. The generalization is
that, in nature, the structures mentioned are always found asso-
ciated together : the belief is, that they always have been, and
always will be, found so associated. In other words, the defini-

13 2

4 ON CLASSIFICATION.

tion of the class Mammalia is a statement of a law of correlation,
or coexistence, of animal structures, from which the most im-
portant conclusions are deducible.

For example : if a fragmentary fossil be discovered, consisting
of no more than a ramus of a mandible and that part of the skull
with which it articulated, a knowledge of this law may enable
the palaeontologist to affirm, with great confidence, that the
animal of which it formed a part suckled its young and had
non-nucleated red blood-corpuscles ; and to predict that should
the back part of that skull be discovered, it mil exhibit two
occipital condyles and a well-ossified basi-occipital bone.

Deductions of this kind, such as that made by Cuvier in the
famous case of the fossil opossum of Montmartre, have often
been verified, and are well calculated to impress the vulgar
imagination ; so that they have taken rank as the triumphs of
the anatomist. But it should carefully be borne in mind, that,
like all merely empirical laws, which rest upon a comparatively
narrow observational basis, the reasoning from them may at any
time break down. If Cuvier, for example, had had to do with a
fossil Thylacinus instead of a fossil Opossum, he would not have
found the marsupial bones, though the inflected angle of the jaw7
would have been obvious enough. And so, though, practically,
any one who met with a characteristically mammalian jaw
would be justified in expecting to find the characteristically
mammalian occiput associated with it ; yet, he would be a bold
man indeed, who should strictly assert the belief which is im-
plied in this expectation, viz., that at no period of the world's
history did animals exist which combined a mammalian occiput
with a reptilian jaw, or vice versa.

Not that it is to be supposed that the correlations of struc-
ture expressed by these empirical laws are in any sense acci-
dental, or other than links in the general chain of causes and
effects. Doubtless there is some very good reason why the
characteristic occiput of a Mammal should be found in association
with mammae and non-nucleated blood-corpuscles ; but it is one
tiling to admit the causal connexion of these phenomena with
one another, or with some third ; and another thing to affirm
that we have any knowledge of that causal connexion, or that

GREGARINIDA, RHIZOPODA, SPONGIDA, AND INFUSORIA. 5

physiological science, in its present state, furnishes us with any
means of reasoning from the one to the other.

Cuvier, the more servile of whose imitators are fond of citing
his mistaken doctrines as to the nature of the methods of palaeon-
tology against the conclusions of logic and of common sense, has
put this so strongly that I cannot refrain from quoting his words.*

" But I doubt if any one would have divined, if untaught by
observation, that all ruminants have the foot cleft, and that they
alone have it. I doubt if any one would have divined that there
are frontal horns only in this class : that those among them
which have sharp canines for the most part lack horns.

" However, since these relations are constant, they must have
some sufficient cause ; but since we are ignorant of it, we must
make good the defect of the theory by means of observation •
it enables us to establish empirical laws, which become almost
as certain as rational laws when they rest on sufficiently re-
peated observations ; so that now, whoso sees merely the print of
a cleft foot may conclude that the animal which left this im-
pression ruminated, and this conclusion is as certain as any
other in physics or morals. This footprint alone, then, yields to
him who observes it, the form of the teeth, the form of the jaws,
the form of the vertebras, the form of all the bones of the legs, of
the thighs, of the shoulders, and of the pelvis of the animal which
has passed by : it is a surer mark than all those of Zadig."

Morphological classification, then, acquires its highest im-
portance as a statement of the empirical laws of the correlation of
structures ; and its value is in proportion to the precision and
the comprehensiveness with which those laws, the definitions of
the groups adopted in the classification, are stated. So that,
in attempting to arrive at clear notions concerning classification,
the first point is to ascertain whether any, and if so, what groups
of animals can be established, the members of which shall be at
once united together and separated from those of all other
groups, by well-defined structural characters. And it will be
most convenient to commence the inquiry with groups of that
order which are commonly called Classes, and which are enume-
* ' Oftscmens fossilcs," ed. -ime, tome \r, p. 164.

ON CLASSIFICATION.

rated in an order and arrangement, the purpose of which will
appear more fully by and by, in the following table.

TABLE OF THE CLASSES OF THE
ANIMAL KINGDOM.

The Limits of the Four Cuvierian Sub-Kingdoms are indicated

by the Brackets and Dotted Line.

Kadi ata.

Gregarinida. Infusoria.

Scolecida (?).

PJiizopoda (?).
Spjongida.

Echinodermata.

Annelida.

; Hydrozoa.

Crustacea.

; Actinozoa.

Arachnida.

> ARTICUL;

'■■ Polyzoa.

Myriapoda.
Insecta.

Brachiopoda.
Ascidioida.

Lamellibranchiata.

Br anchiog aster opoda.
Pulmog aster opoda.
Pteropoda.
Cephalopoda.

' MOLLUSCA.

Pisces.
Amphibia.
Pieptilia.
Aves.
Mammalia. ,

> Vertebrata.

GKEGAKINJLDA, UHIZOPODA, SrONGIDA, AND INFUSORIA. I

It is not necessary for my purpose that the groups which are
named on the preceding table should be absolutely and precisely
equivalent one to another ; it is sufficient that the sum of them is
the whole of the Animal Kingdom, and that each of them em-
braces one of the principal types, or plans of modification, of
animal form ; so that, if we have a precise knowledge of that
which constitutes the typical structure of each of these groups,
we shall have, so far, an exhaustive knowledge of the Animal
Kingdom.

I shall endeavour, then, to define — or, where definition is
not yet possible, to describe a typical example of — these various
groups. Subsequently, I shall take up some of those further
classificatory questions which are open to discussion ; inquiring
how far we can group these classes into larger assemblages, with
definite and constant characters ; and, on the other hand, how
far the existing subdivisions of the classes are well based or
otherwise. But the essential matter, in the first place, is to be
quite clear about the different classes, and to have a distinct
knowledge of all the sharply-definable modifications of animal
structure which are discernible in the animal kingdom.

The first class of which I shall speak is the group of the
Gregakinida. These are among the simplest animal forms of
which we have any knowledge. They are the inhabitants of the
bodies for the most part of invertebrate, but also of vertebrate,
animals ; and they are commonly to be found in abundance in
the alimentary canal of the common cockroach, and in earth-
worms. They are all microscopic, and any one of them, leaving
minor modifications aside, may be said to consist of a sac, com-
posed of a more or less structureless, not very well-defined mem-
brane, containing a soft semi-fluid substance, in the midst, or at
one end, of which lies a delicate vesicle ; in the centre of the
latter is* a more solid particle. (Fig. 1, A.) No doubt many
persons will be struck with the close resemblance of the struc-
ture of this body to that which is possessed by an ovum. You
might take the more solid particle to be the representative of
the germinal spot, and the vesicle to be that of the germinal
vesicle ; while the semi-fluid sarcodic contents might be re-
garded as the yelk, and the outer membrane as the vitelline

ON CLASSIFICATION.

membrane. I do not wish to strain the analogy too far, but it
is, at any rate, interesting to observe this close morphological
resemblance between one of the lowest of animals and that form
in which all the higher animals commence their existence. It
is a very remarkable characteristic of this group, that there is
no separation of the body into distinct layers, or into cellular
elements. The Gregarinida are devoid of mouths and of
digestive apparatus, living entirely by imbibition of the juices

Fig. 1. — A, Gregarina of the earthworm (after Lieberkuhn) ; B, encysted; C, D, with
the contents divided into pseudo-navicellae ; E, F, free pseudo-navicellae ; G, H, free
amcebiform contents of the latter.

of the animal in whose intestine, or body cavity, they are con-
tained. The most conspicuous of those phenomena, which we
ordinarily regard as signs of life, which they exhibit, is a certain
contraction and expansion along different diameters, the body
slowly narrowing, and then lengthening, in various directions.
Under certain circumstances (though the conditions of the
change are not thoroughly understood), it is observed that one of
these Gregarinida, whatever its form may be, will convert itself
into a well-rounded sac, the outer membrane ceasing to exhibit

GREGARINIDA, RHIZOPODA, SPONGIDA, AND INFUSORIA. (J

any longer those movements of which I spoke, and becoming
coated by a structureless investment, or " cyst" (Fig. 1, B).

The substance of the body contained within the cyst next
undergoes a singular change. The central nucleus and the
vesicle disappear ; after a time, the mass breaks up into a series
of rounded portions and, then, each of those rounded portions
elongates, and, becoming slightly pointed at each end, consti-
tutes a little body which has been called a " Pseudo-navicella"
from its resemblance to the Diatomaceous Navieula or Navi-
cella (Fig. 1, C, D). Next, the capsule bursts and the Pseudo-
navicellm (Fig. 1, E, F) are scattered and passed out of the
body of the animal which they inhabit. Though, of course, a
great number of them are destroyed, some, at any rate, are
devoured by other animals; and, when that is the case, the little
particle of protein substance which is inclosed within the
Pseudo-navicella is set free from its shell, and exhibits much
more lively movements than before, thrusting out processes in
various directions, and drawing them in again, and, in fact, closely
resembling one of those animalcules which have been called
Amoehse (Fig. 1, H). The young Amoebiform Gregarina grows,
increases in size, and at length assumes the structure which it
had at first. That, in substance, is all that we know of this
lowest division of animal life. But it will be observed, there is
a hiatus in our knowledge. We cannot say that we know the
whole nature and mode of existence of this, or any other animal,
until we have traced it to its sexual state ; but, at present, we
know nothing whatever of this condition among the Gregarinm ;
so that in reasoning about them we must always exercise a cer-
tain reticence, not knowing how far we may have to modify our
opinions by the discovery of the sexual state hereafter.

The process of becoming encysted, preceded or accompanied
very often by the mutual apposition of two Gregarinw, was for-
merly imagined to correspond with what is termed among plants
" conjugation," — a process which in some cases, at any rate,
appears to be of a sexual nature. But the discovery that a
single Gregarina may become encysted and break up into
Pseudo-navicellce seems to negative this analogy.

But now, leaving this, I pass on to the next class — that which

ON CLASSIFICATION.

is indicated in the table as the Ehizopoda. I have put a
query against it, as I shall have to return to it as another of
those respecting which our knowledge is incomplete. And at
this moment I merely direct attention to the salient and cha-
racteristic features of the whole group (Fig. 2).

Fig. 2.

Fig. 2. — A, B, Free and encysted conditions of an Amceba (after Auerbach) ; E, a Fora-
minifer (Rotalia) with extended pseudopodia ; D, its shell in section (after Schulze).

It seems difficult to imagine a stage of organization lower
than that of Gregarinicla, and yet many of the Bhizopoda are
still simpler. Nor is there any group of the animal kingdom
which more admirably illustrates a very well-founded doctrine,
and one which was often advocated by Hunter himself, that life
is the cause and not the consequence of organization ; for, in
these lowest forms of animal life, there is absolutely nothing-
worthy of the name of organization to be discovered by the micro-
scopist, though assisted by the beautiful instruments that are
now constructed. In the substance of many of these creatures,
nothing is to be discerned but a mass of jelly, which might be
represented by a little particle of thin glue Nut that it corre-

GREGARINIDA, KHIZOPODA, 8PONG1DA, AND INFUSORIA. 11

spends with the latter in composition, but it has that texture and
sort of aspect; it is structureless and organless, and without
definitely formed parts. Nevertheless, it possesses all the
essential properties and characters of vitality ; it is produced
from a body like itself; it is capable of assimilating nourish-
ment, and of exerting movements. Nay, more, it can produce
a shell ; a structure, in many cases, of extraordinary complexity
and most singular beauty (Fig. 2, D).

That this particle of jelly is capable of combining physical
forces in such a manner as to give rise to those exquisite and
almost mathematically-arranged structures — being itself struc-
tureless and without permanent distinction or separation of
parts — is, to my mind, a fact of the profoundest significance.

Though a Rhizopod is not permanently organized, however,
it can hardly be said to be devoid of organs ; for the name of
the group is derived from the power which these animals possess
of throwing out processes of their substance, which are called
" pseudopodia," and are sometimes very slender and of great
length (Fig. 2, E), sometimes broad and lobe-like (Fig. 2, A).
These processes may flow into one another, so as to form a net-
work, and they may, commonly, be thrust out from any part
of the body and retracted into it again.

If you watch one of these animals alive, you see it thrusting
out, first one and then another of its pseudopodia, exhibiting
changes of form comparable to those which the colourless corpus-
cles of the human blood present. The movements of these Rhizo-
pods are quite of the same character, only they are much more
extensive and effect locomotion. The creature also feeds itself
by means of its pseudopodia, which attach themselves to nutritive
particles, and then draw them into the substance of the body.
There is neither ingestive nor egestive aperture, neither special
motor nor prehensile organs, but the pseudopodia perform each
function as it may be required.

But here, again, we labour under an imperfection of knowledge.
For, although it is quite certain that the Bhizopocla may multiply
by division of their substance — in a way somewhat analogous to
that which I detailed when speaking of the Gregarinida — yet, as
in that case, we have no knowledge of any true sexual process. It

12 ON CLASSIFICATION.

is a most remarkable circumstance that though these animals are
abundant, and are constantly under observation, we are still in
doubt upon that essential point, — still uncertain whether there
may not be some phase in the cycle of vital phenomena of the
Ehizojooda with which we are unacquainted; and, under these
circumstances, a perfect definition of the class cannot even be
attempted.

The next division is the group of the Spongida, which exist
under such multitudinous forms in both salt and fresh waters.
Up to the last few years Ave were in the same case, with respect
to this class, as with the Gresarinida and the Rhizopoda. Some
zoologists even have been anxious to relegate the sponges to the
vegetable kingdom ; but the botanists, who understood their
business, refused to have anything to do with the intruders.
And the botanists were quite right ; for the discoveries of late
years have not left the slightest doubt that the sponges are
animal organisms, and animal organisms, too, of a very con-
siderable amount of complexity, if we may regard as complex
a structure which results from the building up and massing
together of a number of similar parts.

The great majority of the sponges form a skeleton, which is
composed of fibres of a horny texture, strengthened by needles,
or spicula, of silicious, or of calcareous, matter ; and this frame-
work is so connected together as to form a kind of fibrous
skeleton. This, however, is not the essential part of the animal,
which is to be sought in that gelatinous substance, which invests
the fibres of the skeleton during life, and is traversed by canals
which open upon the surface of the sponge, directly or indirectly,
by many minute, and fewer large, apertures.

If I may reduce a sponge to its simplest expression — taking
the common Spongilla, for example, of our fresh waters, — the
structure — removing all complexities, and not troubling our-
selves with the skeleton, because that has nothing to do with
what we are now considering — may be represented by the
diagram (A, Fig. 3). There is a thin superficial laver (a)
formed entirely of a number of the so-called sponge particles, or
ultimate components of the living substance of the sponge each
of which is similar to an Amoeba, and contains a nucleus. These

GREGARINIDA, RHIZOPODA, BPONGIDA, AND INFIsouiA. 13

l'V. 3.

m ■ ' i

■

■<-"

-?,'■'■''-■.
/ ■

Fig. 3. — A, Hypothetical section ot a Spongilla ; a, supeincial layer; 6, inhalent aper-
tures; c, ciliated chambers; d, an exhalent aperture; c, deeper substance of the
sponge. The arrows indicate the direction of the currents. B, A small sponge with
a single exhalent aperture, seen from above (after Lieberkiihn); a, inhalent apertures ;
c, ciliated chambers ; d, exhalent aperture. C, a ciliated chamber. D, a free-
swimming ciliated embryo.

14 ON CLASSIFICATION.

are all conjoined in a single layer, so as to form a continuous
lamellar membrane, which constitutes the outer and superficial
laver of the bodv. Beneath this is a wide cavity, communi-
eating with the exterior by means of minute holes in the super-
ficial layer (b), and, of course, filled with water. The cavity
separates the superficial layer of the sponge from its deeper
substance, which is of the same character as the superficial layer,
being made up of a number of aggregated sponge particles, each
of which has a nucleus and is competent to throw out numerous
pseudopodial prolongations if detached. While the living sponge
is contained in water, a great number of currents of water set in
to the wide cavity beneath a, a, through the minute apertures
(b), which have thence been termed " inhalent."

In the floor of the cavity, there are a number of apertures
which lead into canals ramifying in the deep layer, and eventu-
ally ending in the floors of certain comparatively lofty funnels, or
craters. The top of each of these presents one of those larger
and less numerous apertures, which have been referred to as
existing on the surface of the sponge, and which are fitly termed
" exhalent " apertures. For, as Dr. Grant discovered, many years
ago, strong, though minute, currents of water are constantly
flowing out of these large apertures ; being fed by the currents
which as constantly set in, by the small apertures and through
the superficial cavity, into the canals of the deeper substance.
The cause of this very singular system of currents remained for
a long time unknown. It was rendered intelligible by Dr.
Bowerbank's discovery of the existence of vibratile cilia in the
genus Gvantia, but it is only quite recently that the precise
nature of the arrangement of the apparatus which gives rise to
these currents in ordinary sponges, has been made out by
Lieberkuhn and by Carter. The canals which enter the deep
substance of the sponge become dilated into spheroidal chambers,
lined with sponge particles (Fig. 3, A, c, C), each of which is
provided with a vibratile cilium ; and as all these cilia work
in one direction — towards the crater — they sweep the water out
in that direction, and its place is taken by fresh water, which
flows in through the small apertures and through the super-
ficial chamber. The currents of water carry alone: such mat-

GEEGARINIDA, RH120P0DA, SPONGIDA, AND [NPUSORIA. 15

ters as are suspended in them, and these are appropriated by
the sponge particles lining the passages, in just the same way
as any one of the Rhizopoda appropriates the particles of food
it finds in the water about itself. So that we must not compare
this system of apertures and canals to so many mouths and
intestines ; but the sponge represents a kind of subaqueous city,
where the people are arranged about the streets and roads, in
such a manner, that each can easily appropriate his food from
the water as it passes along.

In the sponges two reproductive processes are known to
occur : the one of them, asexual, corresponding with the en-
cysting process of the Gregarinicla ; and the other, truly sexual,
and answering to the congress of the male and female elements
in the higher animals. In the common fresh- water Spongilla,
towards the autumn, the deeper layer of the sponge becomes full
of exceedingly small bodies, sometimes called " seeds " or " gem-
mules," which are spheroidal, and have, at one point, an open-
ing. Every one of these bags — in the walls of which are ar-
ranged a great number of very singular spicula, each resembling
two toothed wheels joined by an axle — is, in point of fact, a mass
of sponge particles which has set itself apart — gone into winter
quarters, so to speak — and becoming quite quiescent, encysts
itself and remains still. The whole Spongilla dies down, and
the seeds, inclosed in their case, remain uninjured through the
winter. When the spring arrives, the encysted masses within
the " seed," stimulated by the altered temperature of the wrater,
creep out of their nests, and straightway grow up into Spongillse
like that from which they proceeded.

But there is, in addition, a true sexual process, which goes
on during the summer months. Individual sponge particles
become quiescent, and take on the character of ova ; while, in
other parts, particular sponge particles fill with granules, the
latter eventually becoming converted into spermatozoa.

These sacs burst and some of the spermatozoa, coming into
contact with the ova, impregnate them. The ova develop and
grow into ciliated germs (D, fig. 3), which make their way out,
and, after swimming about for a while, settle themselves down
and grow up into Spongillse.

ON CLASSIFICATION.

Now that we know the whole cycle of the life of the sponges,
and the characters which in ay be demonstrated to be common
to the whole of this important and remarkable class, I do not
think any one who is acquainted with the organization or the
functions of plants, will be inclined to admit that the Spongida
have the slightest real affinity with any division of the vegetable
kingdom.

The next group to be considered is the division of the Infu-
soria; and here, again, within the last few years, prodigious
strides have been made in our knowledge of the subject. Al-
though the Infusoria have been favourite studies for many vears,
still it is only quite recently that the cycle of life of these animals
has been made almost completely known, and that we have become
acquainted with the true sexual process as it occurs in them.

Fig. 4.
A B c

j. :ete .//r' .- ' > i -d£» .«-/• t>'.\V- "o^

Fig. 4. — Paramoecium bursaria (after Stein) : A, The animal viewed from the dorsal side ;
a, cortical layer of the body ; 6, "nucleus;" c, contractile chamber; eld', matters
taken in as food ; e, chlorophyll granules.

B, The animal viewed from the ventral side ; a, depression leading to b, mouth ;
c, gullet; d, " nucleus;" d', "nucleolus;" <?, central sarcode. In both these figures
the arrows indicate the direction of the circulation of the sarcode.

C, Paramoecium dividing transversely; a af, contractile spaces; h b' ', "nucleus"
dividing; c <■', "nucleoli."

The different species of the genus Paramoecium are very
common among the microscopic inhabitants of our fresh waters,
swimming about by means of the vibratile cilia with which the
whole surface of their bodies is covered ; and the structure which
essentially characterises these animals is probably that which is
common to the whole of the Infusoria, so that an account of the

GREGARINTDA, BHIZOPODA, srONGIDA, AND INFUSORIA. 17

leading structural features of Paramoecium is, in effect, a
definition of those of the group.

Imagine a delicate, slipper-shaped body inclosed within a
structureless membrane, or cuticula, which is formed as an excre-
tion upon its outer surface. At one point (Fig. 4, B a) the body
exhibits a slight depression, leading into a sort of little funnel
(b c) coated by a continuation of the same cuticular investment,
which stops short at the bottom of the funnel. The whole of the
bag formed by the cuticula is lined by a soft layer of gelatinous
matter, or " sarcode," which is called the " cortical " layer
(Fig. 4, A a) ; while inside that, and passing into it quite gradu-
ally, there being no sharp line of demarcation between the two,
is a semi-fluid substance, which occupies the whole of the
central region of the body. Neither in the cuticle, the cortical
layer, nor the central substance, has any anatomist yet dis-
covered a differentiation into cellular layers, nor any trace of
that histological composition which we meet with in the tissues
of the higher animals ; so that here is another case of complex
vital phenomena proceeding from a substance which, in a histo-
logical sense, is structureless.

At two points of the body (Fig. 4, A c c) the substance
of the cortical layer exhibits a remarkable power of contrac-
tion and dilatation. If you watch one of those points, the
sarcode suddenly seems to open like a window and, for a
while, a clear space is visible, which then, quite suddenly, shuts
again. After a little time the same diastole and systole are
repeated. As the systole takes place, it is possible, occasionally,
to discern certain radiating canals, which extend from the
cavities into the surrounding sarcode, and disappear again
before diastole occurs. There is no doubt that the clear space
is a chamber filled with fluid in the cortical layer, and since
good observers maintain that there is an aperture of communica-
tion, through the cuticula, between the ' contractile chamber '
and the exterior, this fluid can be little more than water. Per-
haps the whole should be regarded as a respiratory or secretory
mechanism : in one shape or another, it is eminently charac-
teristic of the Infusoria. Besides this singular apparatus, there
lies embedded in another part of the cortical layer a solid mass,

18 ON CLASSIFICATION.

of an elongated oval shape (Fig. 4, A B d), which has been
called the " nucleus," though it must be carefully distinguished
from the " nucleus " of a cell. Upon one side of this, and, as it
Avere, stuck on to it, is a little rounded body (Fig. 4, B d'),
which has received the name of the " nucleolus." The animal
swims about, driven bv the vibration of its cilia, and whatever
nutriment may be floating in the water is appropriated by
means of the current which is caused to set continually into the
short gullet by the cilia which line that tube.

But it is a singular circumstance that these animals have an
alimentary canal consisting of a mere gullet, open at the bottom,
and leading into no stomach or intestine, but opening directly
into the soft central mass of sarcode. The nutritious matters
passing down the gullet, and then into the central more fluid sub-
stance, become surrounded by spheroids of clear liquid (Fig. 4,
Ad), consisting apparently of the water swallowed with them, so
that a well-fed Paramoecium exhibits a number of cavities, each
containing a little mass of nutritious particles. Hence formerly
arose the notion that these animals possess a number of stomachs.
It was not unnaturally imagined that each of the cavities in
question was a distinct stomach ; but it has since been dis-
covered that the outer layer of the sarcode is, by means of some
unknown mechanism, kept in a state of constant rotation ; so
that the supposed stomachs may be seen to undergo a regular
circulation up one side of the body and down the other. And
this circumstance, if there were no other arguments on the same
side, is sufficient to negative the supposition that the food-
containing spaces are stomachs ; for it is impossible to imagine
any kind of anatomical arrangement which shall permit true
dilatations of an alimentary canal to rotate in any such manner.
Fsecal matters are extruded from an anus, which is situated not
far from the mouth, but is invisible when not in use. It is an
interesting and important character of the Infusoria, in general,
that, under some circumstances, they become quiescent and
throw out a structureless cyst around their bodies. The Infu-
sorium then not unfrequently divides and subdivides, and, the
cyst bursting, gives rise to a number of separate Infusoria.

The remarkable powers of multiplication by fission and

GREGARINIDA, RHIZOPODA, SPONG1DA, AND INFUSORIA. 19

gemmation which many of the group exhibit are well known ;
but within the last few years the investigations of Midler,
Balbiani, Stein, and others, have shown that these minute
creatures possess a true process of sexual multiplication, and
that the sexual organs are those which have been denominated
' nucleus " and " nucleolus." The nucleus is the true ovary —
the nucleolus, the testis, in Paramoecium. At particular times,
the latter increases very much in size, and its substance is
broken up into rod -like bodies, which represent spermatozoa.
Two Infusoria, in this condition, become conjoined, and the
nucleolus (now converted into a spermatic capsule) of each passes
into the body of the other. The spermatic filaments are said to
enter the nucleus, which then enlarges, and either divides into,
or gives off, a number of rounded germs, which become oval
ciliated bodies provided with long processes. These make their
way out of the body, and, it is believed, are metamorphosed
directly into young Paramoecia. But, perhaps, further informa-
tion is required before we can be quite certain on this point.

0 2

LECTURE II.

ON THE CLASSIFICATION OF ANIMALS.

THE HYDEOZOA, ACTINOZOA, POLYZOA, BRACHIOPODA, ASGI-
DIOIDA, LAMELLIBRANCHIATA, BRANCHIOGASTEROPODA,
PULMOGASTEROPODA, PTEROPODA, AND CEPHALOPODA.

In giving an account of the lowest forms of animal life in the
preceding Lecture, I have substituted for a definition of each
class, a description of the structure of some particular member
of that class, or of the organic features which are most obviously
characteristic of the class ; because, in hardly any of those groups
has the structure of many, and widely different, members
been thoroughly and exhaustively worked out.

I entertain little doubt, however, that the main features of
the description of Spongilla might substantially be taken as a
definition of the Sjwngida, and those of the description of Para-
moecium, as a definition of the Infusoria. On the other hand,
we possess no such complete knowledge of the vital cycle of
any Gregarina or Rhizopod ; and neither description nor defini-
tion of the corresponding classes, of a thoroughly satisfactory
kind, is attainable.

No such difficulties beset us in studying the next class, the
Hydkozoa (comprising the Hydroid polypes and the Medasse),
which may be defined with as much precision as any group in
the Animal Kingdom.

All the Hydrozoa exhibit a definite histological structure,
their tissues primarily presenting that kind of organization
which has been called cellular. Again, the body always ex-
hibits a separation into at least two distinct layers of tissue —

THE HYDROZOA.

an outer and an inner — which have been termed, respectively,
ectoderm and endoderm. The endoderm is that layer which
lines the inner cavities of the body, from the mouth inwards;
the ectoderm is that which forms its external covering.

These two layers are shown in the accompanying diagram-

's. 5.

Fig. 5. — Diagrams illustrative of the mutual relations of the Hydrozoa. — 1. Hydra.
2. Sertularian. 3. Diphyes. 4. Physophorid. 5. Lucemaria. a. Ectocyst. b. Endo-
cyst. c. Their enclosed cavity.

P. Tentacles. N. Natatorial organ. T. Ccenosarc. B. Bract. C. Cell. S.
Polypite or digestive cavity. G. Reproductive organ. A. Air vesicle. F. Float.

I., II., III., IV. represent the successive stages of development of a Medusiform
zooid or reproductive organ.

matic sections of the leading forms of Hydrozoa, the ectoderm
being represented by the thin line with the adjacent clear
space, the endoderm by the thick dark line.

22 ON CLASSIFICATION.

A third distinctive character of the Hydrozoa is, that the
digestive cavity communicates directly, by a wide aperture, with
the general cavity of the body ; the one, in fact, passing by
direct continuity into the other. Furthermore, the digestive
sac is not in any way included in the substance of the rest of
the body, but stands out independently, so that the outer wall
of the digestive cavity is in direct contact with the water in
which the animal lives, and there is no perivisceral chamber.
The like is true of the reproductive organs, which may vary
very much in form, but have the common peculiarity of being
developed as outward processes of the body wall, so that their
external surfaces are directly in contact with the surrounding
medium.

No nervous system has yet been discovered in any of these
animals. The majority of them seize their prey by means of
tentacula developed either around the mouth, or from the walls
of the digestive cavity, or from the body wall ; and these tenta-
cles, as well as other parts of the body, are provided with
those peculiar weapons of offence which have been termed
" thread-cells."

The class of the Actinozoa contains those animals which are
familiar to us as Sea-anemones and Coral-polypes, by the latter
of which, in many parts of the world, those huge reefs, which are
so well known to navigators, are constructed. It embraces the
Sea-pens and the Red coral, and those creatures which are
known to us under the names of Beroe, Cydippe, Pleurobrachia,
&c, transparent, beautifully symmetrical, free-swimming ani-
mals, provided with eight rows of longitudinally-disposed large
cilia. In all these animals we find a great uniformity of struc-
ture, and their plan of construction is quite as readily definable
as that of the preceding class, with which they exhibit a close
affinity. Like the majority of the Hydrozoa, most Actinozoa
have their mouths surrounded by tentacles; and there is the
same primary distinction of the body into two cellular layers —
the ectoderm and the encloderm — though, in the adult forms of
the more highly organized Actinozoa, these primitive layers be-
come further differentiated into bundles of definitely disposed
muscular fibres, and even into nerves and ganglia.

THE ACTINO/OA.

As in the Hyclrozoa, again, the alimentary canal communi-
cates freely, and by a wide aperture, with the general cavity of the
body ; but the whole of the Adinozoa, polype-like as they are in
external appearance, differ from the Hydrozoa by a very im-
portant further progress towards complexity. We found that in
the Hydrozoa the digestive cavity was completely outside the
general cavity of the body, the digestive portion of the organism
being continued into, and not in any way contained within, the
part which contains the general cavity. But if you make a
vertical section of a sea anemone (Fig. 6), you will find that
the alimentary cavity — as freely open at the bottom as in the
Hydrozoa — is enclosed within a part of the body which contains
a prolongation of the general cavity. If you could suppose the
stomach of a Hydrozoon thrust into that part of the body with
which it is continuous, so that the walls of the body should rise
round it and form a sort of outside case, containing a prolon-
gation of the general cavity, the Hydrozoon would be converted
into an Actinozoon.

Fio-. 6.

Fig. 6. — Perpendicular section of Actinia holsatica (after Frey and Leuckart) ; a, mouth ;
6, alimentary cavity ; c, common cavity; d, intermesenteric chambers; e, cord con-
taining thread-cells at the edge ; /, the mesentery ; g, reproductive organ ; h, tentacle.

The prolongation of the general cavity thus produced,
which, as it surrounds the chief viscus, may be termed the
" perivisceral cavity ' (d), receives the products of digestion
mixed with niuch sea-water ; and the nutritive fluid, which fills
the perivisceral cavity and its ramifications, plays the same part

24 ON CLASSIFICATION.

as the blood of the move highly organized animals. The gastric
chamber of the Actinozoa does not lie free in the interior of the
body, but is connected to the sides of it by means of mem-
branous partitions, the so-called "mesenteries" (/), which pass
radially from the stomach to the side walls of the body, and so
divide the " perivisceral cavity " into a number of chambers,
which communicate with the bases of the tentacles. In the
whole of the Hydrozoa the reproductive organs were attached to
the exterior of the body, and projected from it. In the whole
of the Actinozoa, on the other hand, the reproductive organs
(of which both sexes are frequently combined in the same indi-
vidual) are internal, inasmuch as they are situated in the
substance of the mesenteries (g).

These are the universal and distinctive characters of the
Actinozoa. That some are simple and some are compound
organisms ; that some are fixed and some free swimmers ; that
many are soft, while a great number are provided with very
dense skeletons ; that some possess a rudimentary nervous
system, while the majority have as yet afforded no trace of any
such structure, are secondary circumstances in no way affecting
the problem before us, which is, to find a diagnostic definition of
the group.

Notwithstanding the invariably minute size of the organisms
which constitute the next class on the list — the Polyzoa — they
exhibit a very great advance in complexity of structure. In
such a compound Polyzoon as the Sea-mat, or Fhistra, the entire
surface of the foliaceous expansion, on being examined by the
microscope, will be found to be beset with an infinitude of
minute apertures leading into little chambers, out of each of
which, when the animal was living and active, multitudes of
little creatures might be seen protruding the oral extremities
of their bodies. The ends of the branches of the freshwater
genus PlumateJIa, represented in Fig. 6, present a similar spec-
tacle. Each mouth is surrounded by a circlet of tentacles ; and,
as every tentacle is fringed with long and active vibratile cilia,
lashing the water towards the mouth, hundreds and thousands
of little Maelstroms are created, each tending to suck down
such nutritious bodies, living or dead, as come within its range.

THE POLYZOA. 25

Tho mouth (Fig. 8) leads into a long and wide pharyngeal
and oesophageal tube, which opens, below, into a definite stomach.
From this is continued a distinct intestine, which bends upon

Fig. 7.

Fig. 7. — Plumatella repens, a freshwater Polyzoon, magnified (after Allman).

itself towards the oral end of the body, so as to form a sharp
angle, and then terminates upon the outer surface near the
mouth ; so that we have here, for the first time in our ascend-
ing survey of the Animal Kingdom, an animal possessing a
complete intestine, not only structurally separated from the
general substance of the body, and provided with permanent
apertures, as in the Hydrozoa and Actinozoa, but completely
shut off from the perivisceral cavity, and in direct communica-
tion only with the external medium. All the Polyzoa possess
a nervous system, the characters and position of which are
very well defined. It consists of a single ganglion (Fig. 8, w),
placed between the oral and the anal apertures, and sending-
off nerves in various directions. It has been affirmed that,
in some Polyzoa, there is a more extended system of nerves by
which the various zooids of the compound organism are placed
in communication ; but of that we want further evidence. In
these animals no heart has been discovered as yet, the matters
which result from digestion percolating through the walls of the
intestine, and becoming mixed with the perivisceral fluid. One
of the structural characters which I have mentioned is exceed-
ingly important. As I have said, the intestine is not straight,

ON CLASSIFICATION.

but is bent upon itself (Fig. 8), and the direction of flexure is
such that the nervous ganglion, which corresponds with those
called "pedal" in Lamellibvanchiata, is placed in the re-entering
angle between the gullet and the rectum. In order to express

Fig. 8.

Fig. 8. — Phimatella repens, a single cell more magnified ; m, calyx at the base of the
ciliated tentacula borne by the disk or lophophore ; k, gullet; g g, stomach; /<, in-
testine; i, anu> ; w, nervous ganglion (after Allman).

this relation of the nervous system to the alimentary canal, the
flexure of the latter has been called " neural " — the side of the
body on which the principal ganglion is placed, and towards
which the intestine is bent, being the " neural " side. AYhatever
our terminology, however, the great point is to remember that
the structural relation which it expresses is constant throughout
the Polyzoa.

In the next division, the Bkachiopoda, which are animals
differing very much in external appearance from the Polyzoa,

THE BRACHIOPODA. 27

we shall find, nevertheless, a singular fundamental resemblance
of internal structure to the latter. All known Polyzoa are com-
pound animals, that is to say, the product of every ovum gives
rise, by gemmation, to great assemblages of partially indepen-
dent organisms, or zooids. The Brachiopoda, on the contrary,
are all simple, the product of each ovum not giving rise to others
by gemmation. All the Brachiopoda possess a bivalve shell — a
shell composed of two, more or less horny, or calcified, pieces,
which are capable of a certain range of motion on one another,
and are very commonly articulated together by teeth and sockets.
The proper body, which is small when compared with the
size of the shell, has its dorsal integument produced into broad
membranous expansions, which line the interior of the valves of
the shell, and are called the lobes of the mantle, or " pallium."
The aperture of the mouth is situated in the middle line, between
the pallial lobes, and, on each side of it, is a longer or shorter
prolongation of the body, provided with ciliated tentacula. It
is from the presence of these " arms " that the class has received
its name. The tentaculate oral disk of a Plumatella is already
horse-shoe shaped (Figs. 7 and 8) ; suppose each eras of the
horse-shoe to be pulled out to a much greater length, and
tentaculated "arms" would be produced, closely resembling
those of the Brachiopoda.

The mouth leads into a gullet which is directed towards, or
lies along, that side of the body from which one lobe of the
mantle, the anterior, is continued ; the gullet opens into a
stomach, provided with a well-developed liver ; and from the
stomach., an intestine proceeds, which is directed towards, or
along, that side of the body from which the other lobe of the
mantle proceeds ; and then either, as I pointed out some years
ago,* ends, blindly, in the middle line (Fig. 9), or else ter-
minates in a distinct anus between the pallial lobes.

* Professor Owen, in the second edition of his lectures on the " Comparative
Anatomy and Physiology of the Invertebrate Animals," published in 1855, thought
it not unbecoming to sneer at this discovery. " There may be blindness somewhere,
but I think not at the termination of the intestine of Terebratula." — L. c, p. 403.
As my statements have subsequently been fully borne out by Mr. Albany Hancock
and by M. Laeaze Duthiers— two of the best minute anatomists of the day— T trust
Mr. Owen is now fully satisfied as to where the " blindness" really was, in 1855.

ON CLASSIFICATION.

The -principal ganglionic mass is situated behind and below
the mouth, in the re-entering angle between the gullet and the
rectum ; in other words, the intestine has, as in the Polyzoa, a
neural flexure (Fig. 9). In all Brachiopoda which have been

= * 5

~ i - - —

l "*2 o r~> «

-^ --a

°* »o £ £> >
bi

Tr.

carefully dissected, a singular system of cavities and canals situ-
ated in the interior of the body, but in free communication with
the surrounding medium, has been discovered. This, which I

THE BKACIIIOrODA. 29

shall term the " atrial ' system, from its close correspondence
with the system of cavities which has received the same name
in the Ascidians, has been wrongfully regarded as a part of the
true vascular system, and the organs by which it is placed in
communication with the exterior have been described as
" hearts." There are sometimes two and sometimes four of
these " pseudo-hearts ': situated in that part of the body wall
which helps to bound the pallial chamber. Each pseudo-heart
is divided into a narrow, elongated, external portion (the so-
called " ventricle "), which communicates, as Mr. Hancock has
proved, by a small apical aperture with the pallial cavity ; and
a broad, funnel-shaped inner division (the so-called " auricle "),
communicating, on the one hand, by a constricted neck with the
so-called " ventricle ;" and, on the other, by a wide, patent mouth,
with a chamber which occupies most of the cavity of the body
proper, and sends more or less branched diverticula into the
pallial lobes. These have been described as parts of the blood
vascular system ; and arterial trunks, which have no existence,
have been imagined to connect the apices of the ventricles with
vascular networks of a similarly mythical character, supposed
to open into the branched diverticula.

In fact, as Mr. Hancock has so well shown in his splendid
and exhaustive memoir, published in the Philosophical Trans-
actions for 1857, the true vascular system is completely distinct
from this remarkable series of " atrial ' chambers and canals,
the function of which would appear to be to convey away excre-
tory matters and the products of the reproductive organs, which
are developed in various parts of the walls of the atrial system.

The precise characters of the true vascular system of the
Brachiopocla probably require still further elaboration than they
have yet received ; and the same may be said, notwithstanding
the valuable contributions of F. M tiller and of Lacaze Duthiers,
of their development ; but the shell, the pallial lobes, the in-
testine, and the nervous and the atrial systems, afford characters
amply sufficient to define the class.

The next great division is that of the Ascidioida, which, like
the Brachiopoda, are marine animals, and are very common all
over the world ; the more ordinary forms of them being always

30 ON CLASSIFICATION.

easily recognisable by the circumstance that their external inte-
gument is provided with two prominent, adjacent apertures, so
that they look very much like double-necked jars (Fig. 10). At

first sight you might hardly suspect the
animal nature of one of these singular or-
ganisms, when freshly taken from the sea ;
but if you touch it, the stream of water which
it squirts out of each aperture reveals the
existence of a great contractile power within ;
and dissection proves that this power is
exerted by an organism of a very consider-
able degree of complexity. Of the two aper-
tures, the one which serves as a mouth is
often — but not always — surrounded bv a
circlet of tentacles (Fig. 11, c). It invariably
Fag. 10 —Phaihma men- ieac[g mt0 an exceedingly dilated pharynx,

tula ; a, oral ; b, atrial # ,

aperture ; c, base of the sides of which are, more or less exten-
sively, perforated. The gullet comes off
from the end of the pharynx, and then dilates into the
stomach, from which an intestine, usually of considerable
length, is continued to the anal aperture. The latter is almost
always situated within a chamber, which opens, externally, by
.that second aperture upon the exterior of the test, to which I
referred just now. Furthermore, in all Ascidians which I have
examined, the first bend of the intestine takes place in such a
manner that, if the intestine continued to preserve the same
direction, it would end on the opposite side of the mouth to the
nervous ganglion (Fig. 11) ; in other words, the nervous gan-
glion would not be situated in the re-entering angle between the
gullet and the rectum, but on the opposite side of the gullet to
that angle. Therefore, the flexure of the intestine is not neural,
as in the Polijzoa ; but as, on the contrary, the intestine is
primarily bent towards the heart side of the body, its flexure may
be termed " haemal." And this haemal flexure of the intestine
in the Ascidians thus constitutes an important element in the
definition of the class.

In these animals there is an atrial system, the development
of which is even more extraordinary than in tho Pohjzon. The

THE ASCI 1)101 DA.

:;i

second aperture to which I have refen
continued into a large cavity, lined by
a membrane, which is reflected, like a
serous sac, on the viscera, and consti-
tutes what is called the " third tunic,"
or k< peritoneum." From the chamber
which lies immediately within the
second aperture (k, Fig. 11) it is re-
flected over both sides of the pharynx,
extending, towards its dorsal part, very
nearly as far as that structure which
has been termed the "endostyle':
(m, Fig. 11). It then passes from the
sides of the pharynx to the body walls,
on which the right and left lamellae
become continuous, so as to form the
lining of the chamber (h), into which
the second aperture (b) leads, or the
" atrial chamber." Posteriorly, or at
the opposite end of the atrial chamber
to its aperture, its lining membrane
(the " atrial tunic ") is reflected to a
greater or less extent over the intes-
tine and circulatory organs, some-
times inclosing each of their parts
in distinct plications (as in the genus
Phallusia), sometimes merely passing
over them, and limiting the blood
sinus in which they are contained
(as in Clavelina, &c). Where the
atrial tunic is reflected over the sides
of the pharynx, the two enter into
more or less close union, and the sur-
faces of contact become perforated by
larger or smaller, more or less nume-
rous, apertures. Thus the cavity of
the pharynx acquires a free commu-
nication with that of the atrium ; and,

ed (fc,Figs. 10 and 11) is
Fig. 11.
a

Fig.

1 1 . — Phallusia mentula ; the
test removed, and haidly more
of the animal drawn than
would be seen in a longi-
tudinal section, a, oral aper-
ture ; b, atrial aperture ; c,
circlet of tentacles ; d, pha-
ryngeal, or branchial, sac : the
three rows of apertures in its
upper part indicate, but do not
represent, the pharyngo-atrial
apertures ; c, the languets : a
series of tongue-shaped pro*
cesses which project into the
branchial sac ; /, oesophageal
opening ; g, stomach ; h, in-
testine performing its haemal
flexure ; i, anus
/. ganglion
heart.

h, atrium ;
m, endostyle ; n,

32 ON CLASSIFICATION.

as the margins of the pharyngo-atrial apertures are fringed with
cilia, working towards the interior of the body, a current is
produced, which sets in at the oral aperture, and out by the
atrial opening, and maybe readily observed in a living Ascidian.

The Ascidians possess a distinct heart, but of a very simple
construction, seeing that it is merely an incomplete muscular
tube, open at each end, and devoid of valves. Functionally, it
is not less remarkable than structurally ; for, in the great majority
of Ascidians, if not in all, it exhibits a regular alternation in the
order of the peristaltic contractions of its muscular substance,
which has no parallel in the Animal Kingdom. The result of
this reversal in the direction of the contractions of the heart is a
corresponding periodical reversal of the course of the circulation
of the blood, so that the two ends of the heart are alternately
arterial and venous.

The perforated pharynx performs the function of a branchial
apparatus, the blood contained in its sieve-like walls being
subjected-to the action of constant currents of aerated water. All
Ascidians possess a single nervous ganglion placed upon one
side of the oral aperture (/, Fig. 11), and, in all known genera
but Appendicularia, it is situated between the oral and atrial
apertures, and, indeed, between the oral and anal aj)ertures ;
for, in all genera but that mentioned, the intestine, after it has
made its haemal bend, curves clown towards the neural side of the
body, and opens into the atrium on that side of the body, and
behind the nervous ganglion.

The outer integument of the Ascidians secretes upon its
surface, not a calcareous shell, but a case or " test," which may
vary in consistence from jelly to hard leather or horn. And it
is not one of the least remarkable characteristics of the group
that this test is rendered solid, by impregnation with a substance
identical in all respects with that " cellulose ': which is the
proximate principle of woody fibre, and forms the chief part of
the skeleton of plants. Before the discoveries of late years had
made us familiar with the production of vegetable proximate
principles by the metamorphosis of animal tissues, this circum-
stance was justly regarded as one of the most remarkable facts
of comparative physiology.

THE LAMEIiLTlSl!ANCIIIATA.

The last common and distinctive peculiarity oi* the Ascidians
which I have to mention, is one which acquires importance only
from its constancy. The middle of the haemal wall of the
pharynx, from near the oral to the oesophageal end, in all these
animals, is pushed out into a longitudinal fold, the bottom of
which projects into a blood sinus, and has a much thickened
epithelial lining. Viewed from one side, the bottom of the fold
consequently appears like a hollow rod, and has been termed
the " endostyle ': (m, Fig. 11). The functions of this structure
are unknown, but it has been noticed in all genera of Ascidians
hitherto examined.

In the next group, that of the Lamellibkanchiata, which
comprises those creatures which we know as mussels, cockles,
and scollops, and all the fabricators of what are commonly
known as bivalve shells (except the Brachiopoda), wTe meet with

Fiff. 12.

Fig. 12. — Sectional diagram of a freshwater mussel (Anodon). A A, mantle, the right
lobe of which is cut away ; B, foot ; C, branchial chamber of the mantle cavity ; D,
anal chamber; /, anterior adductor muscle; II, posterior adductor muscle;
a, mouth; 6, stomach; c, intestine, the turns of which are supposed to be seen
through the side walls of the foot ; d, rectum ; e, anus; /, ventricle ; fj, auricle ; h,
gills, except i, right external gill, largely cut away and turned back; k, labial palpi :
/, cerebral ; m, pedal ; n, parieto-splanchnic ganglia ; o, aperture of the organ of
Eojanus ; p, pericardium.

an important advance in organization. In all these animals, the
body, as is exemplified by the diagram (Fig. 12) of a freshwater

ON CLASSIFICATION.

mussel (Anodon), is included within a mantle or " pallium," which
is formed by a prolongation of the dorsal integument, — a structure
in principle quite similar to that which Ave met with in the
Brachiopoda. But there is this essential difference between the
two, — that whereas, in the Brachiopoda, the mantle lobes corre-
sponded with the anterior and posterior regions of the body, in
the LamellibrancMata thev answer to the right and left halves of
the body. The intestine, which always terminates by a definite
anus between the mantle lobes, at the posterior end of the body,
has its first flexure neural. There is always a wTell-developed
heart, which is much more complex than that of the Ascidians
or Brachiopods, being divided into distinct auricular and ven-
tricular chambers. Commonly, there are two auricles and one
ventricle, as in Anodon ; but in other LamellibrancMata, such as
the oyster, there is a single auricle and a single ventricle, and in
some exceptional cases there are two auricles and two ventricles,
forming two distinct hearts. Distinct respiratory organs, which
usually have the form of lamellae (as the name of the class

Fig. 13.

Fig. 13. — Anodon, vertical and transverse section of the body through the heart ; /, ven-

tricle; g, auricles; c, rectum; p, pericardium; /i, inner, i, outer gil
of Bojanus ; B, foot; A A, mantle lobes.

oq,

organ

implies), are found in all Lamellibranchiata, and are situated
upon each side of the body, in a chamber which extends bet \\< ■« n

THE BRANCHIOGASTEROPODA. 35

the foot and the mantle lobes in front, and between the mantle
lobes posteriorly (Fig. 1 3). The branchial organs may consist of
distinct filaments, or of plates composed of tubular rods support-
ing a network of blood-vessels, and covered with cilia, by the
action of which they are constantly bathed by currents of water.

The nervous system presents a no less distinct advance than
the other organs. All Lamellibranchs possess at least three
pairs of principal ganglia — a cerebral pair at the sides of the
mouth, a pedal pair in the foot, and a third pair on the under
surface of the posterior adductor muscle, which are commonly
called " branchial," but which, as they supply not only branchial,
but visceral and pallial filaments, may more properly be termed
" parieto-splanchnic." Three sets of commissural filaments con-
nect the cerebral ganglia with one another, with the pedal, and
with the parieto-splanchnic ganglia. The inter-cerebral commis-
sures surround the mouth, and the other two pairs of cords
extend respectively, from the cerebral to the pedal, and from
the cerebral to the parieto-splanchnic ganglia.

Finally, there is always, in these animals, an external shell,
which is formed as an excretion from the surface of the lobes of
the mantle, and is composed of layers of animal matter hardened
by deposit of carbonate of lime, which may or may not take a
definite form, and so give rise to " prismatic ': and " nacreous "
substance. As the lobes are right and left, so the valves of the
shell are right and left, and differ altogether from the valves of
the shell of the Brachiopoda, which are anterior and posterior.
The valves of the shell can be brought together by adductor
muscles. Of these one (Fig. 12, II) always exists, posteriorly,-
on the neural side of the intestine. A second (Fig. 12, I) is
commonly found anteriorly to the mouth, on the haemal side of
the intestine.

The hiatus between the next class, which is termed Bkan-
chiogasteropoda in the table, and that just defined, is consider-
able, though not quite so well marked as that between the
Ascidians and the Lamellibr and data. This group, which con-
tains the whelks, periwinkles, sea-slugs, and the Heteropoda of
Cuvier, consists of animals which, like the Lamellibranchs, pos-
sess (in their young state, at any rate) a mantle ; a foot, which

d 2

36 ON CLASSIFICATION.

is the chief organ of locomotion ; and three principal pairs of
ganglia — cerebral, pedal, and parieto-splanchnic. When they
are provided with a heart, which is usually the case, it is divided
into auricular and ventricular chambers ; but the mantle, instead
of beins: divided into two lateral lobes, is continuous round the
bodv, and when it secretes a shell from its surface, that shell is
either in a single piece, or the pieces are repeated from before
backwards, and not on each side of the median line. The shell
of a Branchiogasteropod may, therefore, be univalve, or com-
posed of a single conical piece, straight or coiled ; or it may be
multivalve — formed of a number of segments succeeding one
another antero-posteriorly ; but it is never bivalve.

Sometimes a shelly, horny, or fibrous secretion takes place
from the foot, giving rise to a structure resembling the byssus
of some Lamellibranchs ; it is the so-called " operculum," which
serves to protect the animal when retracted into its shell ; but as
the operculum is developed from the foot and not from the mantle,
it can obviouslv have no homology with the valves of either a
Brachiopod or a Lamellibranch. The Branchiogasteropoda
(Fig. 14) commonly possess a distinct head, provided with a
pair of tentacles and a single pair of eyes, both of which are
supplied with nerves from the cerebral ganglia. Cephalic eyes
of this kind are not known in the Lamellibrancliiata. But the
characters which most definitely distinguish the Brancliio-
gasteropoda are to be found in the alimentary canal. The
cavity of the mouth is invariably provided with an organ which
is usually, though not very properly, called the tongue, and
which might more appropriately be denominated the " odonto-
phore." It consists essentially of a cartilaginous cushion, sup-
porting, as on a pulley, an elastic strap, which bears a long-
series of transversely disposed teeth. The ends of the strap are
connected with muscles attached to the upper and lower surface
of the hinder extremities of the cartilaginous cushions ; and
these muscles, by their alternate contractions, cause the toothed
strap to work, backwards and forwards, over the end of the pulley
formed by its anterior end. The strap consequently acts, after
the fashion of a chain-saw, upon any substance to which it is
applied, and the resulting wear and tear of its anterior teeth

THE r.KANCIIKHJASTKKOl'ODA.

are made good by the incessant development of new teeth in
the secreting sac in which the hinder end of the strap is Lodged.
Besides the chain-saw-like motion of the strap, the odontophore
may be capable of a licking or scraping action as a whole.

Fig. 14. — Section ot a female whelk (Buccinum). The organs marked t and A are removed
from their proper places ; the others are seen in situ, a, mouth ; b, gullet ; c, head ; d ,
foot ; e, operculum ; /, free part of the mantle ; g, that part which invests the
visceral mass lodged within the shell ; h, a gland of unknown function connected with
the gullet ; i, crop ; k, stomach ; /, intestine; m, rectum ; n, heart ; o, aperture of the
renal organ; r, mucous gland developed from the wall of the mantle cavity; s,
oviduct; t, salivary gland. The arrows indicate the position of the branchiae. The
cerebral, pedal, and parieto-splanchnic ganglia closely surround the gullet, and the
latter send off a long ganglionated cord towards the heart and branchiae.

The other peculiarity of the alimentary canal of the
Branchiogasterojooda is that it is always bent upon itself, at first,
not to the neural, but to the haemal, or heart side of the body —
the rectum very commonly opening into the mantle cavity,
above the cephalic portion of the body.

In most Branchiogasterojpoda the foot is a broad, flat, mus-
cular body, without any distinct division of parts ; but in some
forms, such, as the Heteropoda of Cuvier, it is divided into three
very well-marked portions — an anterior, a middle, and a pos-
terior, which are termed respectively the propodium, mesopo-

ON CLASSIFICATION.

dium, and metcvpodium ;* while the Aphjsise, in which the foot
proper has the ordinary composition, exhibit processes from the
lateral and upper surfaces of that organ, having the form of
great muscular lobes, which serve as a sort of aquatic wings to
some sj^ecies, and are termed epipodia.

The Br anchiog aster opoda are such of the Gasteropoda of
Cuvier as breathe water either by means of the thin wall of the
mantle cavity [Atlanta, e.g.), or by special pallial branchiae
(Peethiibranchiata, Tectibranchiata, &c), or by certain parts of
the integument of the body (Nudibranchiata) more or less

Fie. 15.

15. — Diagram exhibiting the disposition of the intestine, nervous system, &c, in a
common snail (Helix), a, mouth ; b, tooth ; c, odontophore ; d, gullet ; e, its dilata-
tion into a sort of crop; /, stomach ; g, coiled termination of the visceral mass; the
letter is also close to the commencement of the intestine, which will be seen to lie
"ruler the oesophagus, and not over it as in the whelk ; h, rectum ; *, anus : k, renal
sac; I, heart; m, lung, or modified pallial chamber; n, its external aperture; o,
thick edge of the mantle united with the sides of the body; p, foot; r, cerebral,
pedal, and parieto-splanchnic ganglia aggregated round the gullet.

specially modified. The next class, the Pulmogasterofoda,!
on the other hand, are the Pulmonate Gasteropoda of Cuvier,

* I proposed these terms in a Memoir in the " Morphology of the Cephalous
M"llusca," published in the Phdozopldcal Transactions for 1853.

i I adopt these convenient names at the suggestion of my friend Profe.—<(i
Greene, of Queen's College, Cork.

THE TULMOGASTEROPODA AND PTEROPODA. 39

the snails and slugs, which agree with the Branchiogastero-
pocla in the general characters of their body, mantle, nervous
and respiratory systems, and in possessing an odontophorc ;
but differ from them, not only in breathing air by means of
the thin lining of the pallial chamber, but, as I believe, by
the direction of the flexure of their intestine. A careful dis-
section of a common snail, for example (Fig. 15), will prove
that, though the anus is situated in the same way as in the
Brancldogasteropoda, on the dorsal or haemal side of the body,
the primary bend of the intestine is not to the htemal, but to
the neural, side, the eventual termination of the intestine on
the haamal side being the result of a second change in its
direction.

How far this neural flexure of the intestine really prevails
among the Pulmo-gasteropods is a question which must be
decided by more extensive investigations than I have as yet
been enabled to carrv out.

The members of the class Pteropoda are small, or even
minute, molluscs ; all marine in habit, and for the most part
pelagic, or swimmers at the surface of deep seas. Like the two
preceding groups, they possess three principal pairs of ganglia,
an odontophore, a mantle, which is not divided into two lobes,
and which secretes a univalve shell, if any. But the propodium,
mesopodium, and metapodium are usually rudimentary, and
locomotion is almost wholly effected by the epipodia, which are
enormously developed, and, in most of the genera, perform the
office of aquatic wings still more efficiently than those of the
Aplysize. Furthermore, the intestine is flexed towards the
neural side of the body ; and the head, with the organs of sight,
are usually quite rudimentary. I include in this group not
only Criseis, Cleodora, Hyal&a, Pneumodermon, &c, but also
the aberrant genus Dentalium*

The last class of this series is that of the Cephalopoda —
comprising the Poulpes, the Cuttle-fishes, the Squids, and the
pearly Nautilus ; a group definable by most marked and distinct
characters from all the preceding, though it resembles them in

* Dentalium resembles the Pteropoda in its rudimentary head, the neural flexure
of its intestine, its epipodial lobes, and the character of its larva.

4()

ON CLASSIFICATION.

fundamental characters. Thus, the mantle is related to the
body, as in Pterojwla and Gasteropoda ; when an external shell
exists it is composed of a single piece, and the Cephalopocls have

Vis. Iti.

Fig. 16. — Diagrammatic section of a female Cephalopod (Sepia officinalis), a, Buccal mass
surrounded by the lips, and showing the horny jaws and tongue ; 6, oesophagus ; c,
salivary gland; d, stomach ; e, pyloric caecum ; /, the funnel ; g, the intestine ; h, the
anus ; i, the ink-hag; k, the place of the systemic heart; Z, the liver ; 'n, the hepatic
duct of the left side ; o, the ovary ; p, the oviduct ; q, one of the apertures by which
the atrial system, or water-chambers, are placed in communication with the exterior ;
/•, one of the branchiae ; s, the principal ganglia aggregated round the oesophagus ; in, the
mantle ; sh, the internal shell, or cuttle-bone. 1, 2, 3, 4, 5, the produced and modi-
lied margins of the foot, constituting the so-called "arms" of the Sepia.

an odontophore constructed upon just the same principle as
that of the other classes. The nervous system, the foot, and
the epipodia exhibit the same primary relations as in these
groups, and there is a distinct head, with ordinarily well-

THE OEPHALOBODA. 41

developed optic und olfactory organs. That which essentially
characterises the Cephalopoda, in fact, is simply the manner in
which, in the course of development (as Kolliker long since
proved), the margins of the foot proper and the epipodia become
modified and change their relations. The margins of the foot
are produced into more or less numerous tentacular appendages,
often provided with singularly constructed suckers or acetabula ;
and the anterolateral parts of each side of the foot extend
forwards beyond the head, uniting with it and with one another ;
so that, at length, the mouth, from having been situated, as
usual, above the anterior margin of the foot, comes to be placed
in the midst of it. The two epipodia, on the other hand, unite
posteriorly above the foot, and where they coalesce, give rise
either to a folded muscular expansion, the edges of which are
simply in apposition, as in Nautilus ; or to an elongated flexible
tube, the apex of which projects beyond the margin of the
mantle (Fig. 16, /), called the "funnel' or " infundibulum,"
as in the dibranchiate Cephalopoda.

The Cephalopoda present a vast number of the most inte-
resting features, to which it would be necessary to devote much
attention if we were studying all the organic peculiarities mani-
fested by the class ; but it is in the characters of foot and of the
epipoclium that the definition of the class must be chiefly sought.
In addition, the flexure of the intestine is, in all Cephalopods,
neural ; and the mouth is always provided with a horny or more
or less calcified beak, like that of a parrot, composed of two
curved pieces, which move in the median antero-posterior plane
of the body ; and one of which, that on the neural side, is always
longer than the other.

LECTURE III.

ON THE CLASSIFICATION OF ANIMALS.

THE ECHINODEEMATA, SCOLECIDA, ANNELIDA, CRUSTACEA,
ARACHNIDA, MYRIAPODA, AND INSECTA.

Hitherto, it lias not been a matter of very great difficulty to
discover the characters in which the members of the various
classes, which have passed under our notice, agree with one
another and differ from the members of all other classes. But
to-day we shall be met, at the outset of our studies, by a large
series of organisms which present us with much greater ob-
stacles,— the result, in a great measure, of imperfect knowledge.

The first group on the list — the Echixodeemata — com-
prises the star-fishes, sea-urchins, sea-cucumbers, trepangs, and
feather-stars — known technically as Asteridea, Echinidea, Holo-
thuridea, Opliiuridea, Crinoidea, &c, — marine animals which
differ vastly in external appearance, though they all, in the
adult state, present a more or less definitely radiate arrange-
ment of some parts of their organization.

That which most remarkably distinguishes the Eehinoder-
mata is the nature of the embryo, and the strange character of
the process by which the adult form is originated by a secondary
development within that embryo.

In the great majority* of the Echinodermata, the develop-

* In Opldolepis squarrn if a and Ecliinaster sejiositu*, the larva appears to attain
only a very imperfect state of development before the appearance of the echinoderm
body; and careful re-examination is required to decide how far the larvae of these
animals are truly bilaterally symmetrical.

THE ECHINODEKMATA.

ment of which has been examined, the impregnated egg gives
rise to a free-swimming, ovoid, ciliated embryo, the cilia of
which soon become restricted to, and, at the same time, largely
developed upon, one, two, or more bands, which are disposed
either transversely, or more or less obliquely to the longitudinal
axis of the body, but which are, in any case, bilaterally sym-
metrical (Fig. 17).

Ficr. 17.

Fig. 17. — Diagram exhibiting the general plan of the development of the Echinoderms
(after Miiller). — A. Common form whence the Holothurid (B, B'j and Ophiurid or
Echinid (C, C') larva) are derived. D, D'. Younger and more advanced stages of the
Asterid (Bipinnarici) larvae, a'. Mouth, b. Stomach, c, Intestine. cL Anus.
e. Ciliated band. <?'. Second or anterior ciliated circlet.

The parts of the body which carry the ciliated band, or
bands, often become developed into processes, which correspond
upon each side of the body, and thus render its bilateral sym-
metry more marked (Fig. 17, C, J)'). And, in the larvse of some
Echinidea and Ojihiuridea, other bilaterally symmetrical processes
are developed from parts of the body which do not lie in the
course of the ciliated bands.

ON CLASSIFICATION.

The larvae of Aster idea and Holothuridea are devoid of anv
continuous skeleton, but those of Ophiuridea and Ecliinidea
possess a very remarkable bilaterally symmetrical, continuous,
calcareous skeleton, which extends into, and supports the pro-
cesses of the body (Fig. 20).

A distinctly defined alimentary canal early makes its ap-
pearance in these Echinoderrn larvae. It is divided into a well-
marked oral and oesophageal portion, a globular stomach, and a
short intestine terminating in an anal aperture (Figs. 17 and 18).
All the parts of the alimentary canal are disposed in a longi-
tudinal and vertical plane, dividing the larval body into two
symmetrical halves ; but the oesophageal and intestinal portions
are so disposed as to make an angle, open towards the ventral
side, with one another. No nervous, or other organs, besides
those indicated, have as yet been discovered in these larvae.

Jl.

Fig. 18. — A young Asterid larva (after Miiller). — A. Ventral. B. Lateral views of the
larva. C. Commencing rudiment of the starfish. a. Mouth, b. (Esophagus,
c. Stomach, c'. Intestine, o. Anus. x. Anterior, and y, principal ciliated band.
h. Cascal diverticulum, forming the rudiment of the ambulacral vascular system, and
opening externally by the pore, g. k. Perisoma of the starfish.

FiL'. 19.

Fig. 19.

■Development of a Holotkurid (after Miiller), — A. Early condition of larva.
B, C. Later stages. /, g, It, the ambulacral vascular system.

THE ECIIINODKU.MATA.

Fig. 20. — Development of an Echinid larva (after Miiller). — A. earliest, and B, later
condition of larva. C. The Echinid imago developed within and nearly obliterating
the larva.

After swimming about in this condition for a while, the
larva begins to show the first signs of those changes by which
it is converted into the adult Echinoderm. An involution of
the integument takes place upon one side of the dorsal region
of the body, so as to give rise to a csecal tube, which gradually
elongates inwards, and eventually reaches a mass of formative
matter, or blastema, aggregated upon one side of the stomach.
Within this, the end of the tube becomes converted into a
circular vessel, from which trunks pass off, radially, through the
enlarging blastema. The latter, gradually expanding, gives rise
in the Echiniclea, the Asteridea, the Ojrfiiuridea, and the Crinoidea,
to the body-wall of the adult; the larval body and skeleton
(when the latter exists), with more or less of the primitive
intestine, being either cast off as a whole, or disappearing, or
becoming incorporated with the secondary development, while
a new mouth is developed in the centre of the ring formed by
the circular vessel. The vessels which radiate from the latter
give off diverticula to communicate with the cavities of
numerous processes of the body — the so-called feet — which are
the chief locomotive organs of the adult. The radiating and
circular vessels, with all their appendages, constitute what is
known as the " ambulacral system ;" and, in Asterids and
Echinids, this remarkable system of vessels remains in commu-
nication with the exterior of the body by canals, connected with
perforated portions of the external skeleton — the so-called
" madreporic canals " and " tubercles." In Ophiurids the per-
sistence of any such communication of the ambulacral system

46 ON CLASSIFICATION.

with the exterior is doubtful, aud still more so in Crinoids.
In Holothurids no such communication obtains, the madreporic
canals and their tubercles depending freely from the circular
canal into the perivisceral cavity.

Whether the larva possessed a skeleton or not, the adult
Echinoderm presents a calcareous framework which is developed
quite independently of that of the larva. This skeleton may
be composed of mere detached spicula, or plates, as in the
Holothurids; or of definitely disposed ossicula, or regular
plates, as in other Echinoderms. In the latter case its parts
are always disposed with a certain reference to the disposition
of the ambulacral svstem, and hence have a more or less dis-
tinctly radiate arrangement. It might be expected, in fact,
that the arrangement of the organs of support should follow
more or less closely that of the chief organs of movement of the
adult Echinoderm, and it is not surprising to find the nervous
system similarly related. It is, in all adult Echinoderms, a
ring-like, or polygonal, ganglionated cord, situated superficially
to that part of the ambulacral system which surrounds the
mouth, and sending prolongations parallel with, and superficial
to, the radiating ambulacral trunks.

The reproductive organs of the Echinoderms, which usually
open upon, or between, parts of the radially disposed skeleton,
commonly partake of the radial symmetry of that skeleton ; but
they have no such radial symmetry in the Holothuridea.

The alimentary canal of the adult Echinoderm is still less
dependent upon the skeleton, and only in one group, the
Asteridea, exhibits anything approaching a radiate disposition.
^Yhere skeletal elements are developed around the mouth or
gullet, however, they have a radial disposition ; as, e. g., the parts
of the so-called " lantern of Aristotle."

The vascular system which exists in many, if not all, adult
Echinoderms, but the true nature of which is by no means
understood at present, is closely related both to the alimentary
and to the ambulacral systems, and partakes of the disposition
of both.

No Echinoderm whatsoever has its organs, internal or ex-
ternal, disposed with that absolute and perfect radial symmetry

THE SCOLECIJ'A. 47

which is exhibited by a Medusa, the tendency towards that kind
of symmetry being always disturbed, either by the disposition of
the alimentary canal, or by that of some part of the ambulacra!
apparatus. Very often, as in the Spatangoid sea-urchins, and in
many Holothuriclea, the ambulacra! and nervous systems alone
exhibit traces of a radial arrangement; and in the larval state,
as we have seen, radial symmetry is totally absent, the young
Echinoderm exhibiting as complete a bilateral symmetry as
Annelids, or Insects.

Nothing can be more definite, it appears to me, than the
clas.s Echinodermata, the leading characteristics of which have
just been enumerated ; but it is a very difficult matter to say
whether the seven groups, some of considerable extent, which
are massed under the next head, that of Scolecida, are rightly
associated into one class, or should be divided into several.
The seven groups to which I refer are the Piotifera (or Wheel-
animalcules), the Turbellaria, the Trematoda (or flukes), the
Tseniada (or tapeworms), the Nemcdoidea (or threadworms), the
Acantliocephala, and the Gordiacea. Of these, five are com-
posed of animals parasitic upon others; and exhibiting the
anomalies of structure and of development which might be ex-
pected from creatures living under such exceptional conditions.

There is one peculiarity of organic structure which the first
four of these groups certainly have in common ; they all present
what is termed the " water-vascular system," — a remarkable
set of vessels which communicate with the exterior by means of
one, or more, apertures situated upon the surface of the body,
and branch out, more or less extensively, into its substance.

In the 'Rotifer a the external aperture of the water- vascular
system is single, and situated at the hinder end of the body ;
it communicates with a large, rhythmically contractile, sac,
whence two trunks proceed, which usually give off short lateral
branches, and terminate in the ciliated " trochal disk " of the
Rotifer, in the middle of which its mouth is placed. Both the
lateral offshoots and the terminal branches contain vibratile
cilia. The Trematode and Tsenioid worms have a similar, but
usually much more ramified apparatus ; and it is interesting to
observe that, in these animals, as in the Asjpidogaster conchicola

ON CLASSIFICATION.

(Fig. 21), the water-vascular system becomes divided into two
distinct portions, one with contractile and non-ciliated walls,
the other with non-contractile and ciliated walls. In some

Fi- 21,

Fig 21. — A. Water-vascular system of Aspidogaster conchicoht ; a. terminal pore:
6, lateral contractile vessels ; c, lateral ciliated trunks, that of the left side shaded ;
d, dilatation of this trunk ; B, one of the larger, and C, one of the smaller, cili-
ated vessels.

Turhellaria the apertures of the water-vascular apparatus are
multiple ; while it would seem that in others, as the Nemertidte,
the apparatus becomes shut externally in the adult state, and
consists mainlv, if not exclusively; of contractile vessels. The
difficulties of observation are here, however, very great, and
I would be understood to express this opinion with all due
caution.

In none of these animals has any other set of vessels than
those which appertain to the water-vascular system (if I am
right in my view of the vessels of the Nemertidm) been ob-
served, nor has any trace of a true heart been noticed. The
nervous system consists of one, or two, closely approximated
ganglia.

This sum of common characters appears to me to demand
the union of the Rotifer a, Turhellaria, Trematoda, and Tseniada
into one great assemblage. Ought the Nematoid worms to be
grouped with them ? If the system of canals, in some cases
contractile, which open externally near the anterior part of
the body (Fig. 22), and were originally observed by Yon
Siebold, and since by myself and others, are to be regarded as
homologous with the water-vessels of the Trematoda, this ques-

THE SCOLECIDA.

tion must, I think, be answered in the affirmative. It is almost
the only system of organs in the Nematoidea which gives us a
definite zoological criterion, the condition of the nervous
system in these animals being still, notwithstanding the many
inquiries which have been made into the subject, a matter of
great doubt.

o- 22

Fig. 22. — Oxyuris. — a. Mouth, h. Pharynx, c. Commencement of intestine, and d its
termination ; the intermediate portion is not figured, e. Genital aperture. /.
Opening of vessels, g. Their receptacle, h. One of the vessels, i. Cellular matter
enveloping them. A portion of one of the contractile vessels is represented above,
more highly magnified.

In habit and feature, the Gordiacea, filiform parasites which
inhabit the bodies of insects, and leave their hosts only to breed,
resemble the Nematoidea so much that I can hardly doubt that
their systematic place must be close to that of the latter ; but
positive evidence is almost wanting on this head, the extant
accounts of the minute anatomy of these animals not having
received that kind of confirmation which is desirable.

The structure of the Acanthoeejrfmla, comprising the for-
midable Echinorhynchus (Fig. 23) and its allies, is, on the other
hand, pretty clearly made out. They are vermiform parasites,
like the Txmiada, devoid of any mouth or alimentary canal,
but provided with a proboscis armed with recurved hooks.
The proboscis is supported within by a sort of rod-like handle,
whence a cord is continued, to which the reproductive organs
are attached. A single ganglion is seated in the "handle"
of the proboscis. Immediately beneath the integument lies
a series of reticulated canals containing a clear fluid, and it

ON CLASSIFICATION.

is difficult to see with what these can correspond if not with
some modification of the water-vascular system.*

Fig. 23.

Fig. 23. — The Echinorhynchus of the Flounder. — A. Diagram exhibiting the relative
position of the organs, a. Proboscis, 6. Its stem. c. Anterior enlargement, d.
Body. e. Posterior " funnel." /. Neck. g. Meniscus, h. Superior oblique tubu-
lar bands, k. Inferior muscles of the proboscis. /, m. Genitalia, o. Penis, or
vulva. B. Lower extremity of the stem of the proboscis, a. Ganglion, b. Inter-
space, d. Outer coat. c. Inner wall, e. Tubular band, with the nerve h. f. Mus-
cular bands, g. Suspensorium of the genitalia. C. Part of the female genitalia.
a. Ovary. 6 6. Ducts leading from ovary to uterus ( sperm id ucts ?). c. Open mouth

of oviduct.

d, e.

Uterus and vagina.

Leaving the division provisionally termed Scolecida in this
confessedly unsatisfactory state, I pass on to the Anxlltda, a
class of large extent, containing the leech, the earthworm, the
Sijmnculus, the lobworm, the seamouse and Polynoe (Fig. 24),
the Serjmla, and the Spirorbis.

All the members of this class possess a nervous system,
which consists of a longitudinal series of ganglia, situated along
one side of the body, and is traversed anteriorly by the
oesophagus, the prse-cesophageal, or so-called " cerebral," ganglia

* The recently published investigations of Leuckart, while they demonstrate still
more clearly the close affinity which exists between the Acanthocephala and the
Txniada— by proving the adult worm to arise by secondary growth within a hooked
embryo, in the former case as in the latter — leave some doubt upon the nature of
the reticulated canals. According to Leuckart, they are the remains of the cavity
which primitively lies between the wall of the embryo and the contained rudiment
of the adult Acanthocephalan body.

THE ANNELTDA.

being connected by lateral commissural cords with the post-
cesophageal ganglia.

In many of these animals the body is divided into segments,
each of which corresponds with a single pair of ganglia of the
chain, and each of these segments may be provided with a pair
of lateral appendages ; but the appendages are never articulated ;
and are never so modified, as to be converted into masticatory
i organs, in the cephalic region of the body.

No Annelid ever possesses a heart comparable to the heart
of a Crustacean, or Insect ; but a system of vessels, with more
or less extensively contractile walls, containing a clear fluid,
usually red or green in colour, and, in some rare cases only,
corpusculated, is very generally developed, and sends pro-
longations into the respiratory organs, where such exist. This
has been termed the "pseudo-haemal" system; and I have

Fig. 24.

Fig. 24. — Pohjnoc squamata.

A. Viewed from above and enlarged. a,b. Feelers. c,d. Cirri, e. Elytra. /. Space left

between the two posterior elytra, rj. Seta? and fimbriae of the elytra.

B. Posterior extremity, inferior view, the appendages of the left side being omitted.

h. Inferior tubercle.

C. Section of half a somite with elytron, t. Notopodium, h. Xeuropodium.

D. Section of half a somite with cirrus.

E 2

ON CLASSIFICATION.

thought it probable that these " pseudo-haemal " vessels are ex-
treme modifications of organs homologous with the water-vessels
of the Scolecida. As M. de Quatrefages has clearly shown, it is
the perivisceral cavity with its contents that, in these animals,
answers to the true blood-svstem of the Crustacea and Insects.

The embryos of Annelids are very generally ciliated, and
vibratile cilia are commonly, if not universally, developed in
some part or other of their organization. In both these
respects they present a most marked contrast to the succeeding
classes.

Fig. 25.

1v-

Fig. 25. — Amphithoe, an amphipodous Crustacean. — Lateral view (A), longitudinal and ver-
tical section, detached appendages and stomach (C, D). The numbers I' to XX' indicate
the appendages of the corresponding somites, r. Lostrum. t. Telson. lb. Labium.
st. Roof of the head, or cephalostegite. os. Oostegite. Br. Branchiae. Stomach
opened from above (D), and viewed laterally (C). a, b, c. Different parts of the armature.

THE ARACHNIDA. 53

J n the Crustacea (Fig. 25), the body is distinguishable
into a variable number of "somites," or definite segments, each
of which may be, and some of which always are, provided with a
single pair of articulated appendages. The latter proposition is
true of all existing Crustacea: whether it also held good of
the long extinct Trilobites, is a question which we have no
means of deciding. In most Crustacea, and, probably in all,
one or more pairs of appendages are so modified as to subserve
manducation. A pair of ganglia is primitively developed in
each somite, and the gullet passes between two successive pairs
of ganglia, as in the Annelida.

No trace of a water-vascular system, nor of any vascular
system similar to that of the Annelida, is to be found in any
Crustacean. All Crustacea which possess definite respiratory
organs have branchiae, or outward processes of the wall of the
body, adapted for respiring air by means of water ; the terres-
trial Isopoda, some of which exhibit a curious rudimentary
representation of a tracheal system, forming no real exception
to this rule. When they are provided with a circulatory organ,
it is situated on the opposite side of the alimentary canal
to the principal chain of ganglia of the nervous system ; and
communicates, by valvular apertures, with the surrounding-
venous sinus — the so-called " pericardium."

The Crustacea vary through such a wide range of organiza-
tion that I doubt if any other anatomical proposition, in addition
to those which I have mentioned, except the presence of a
chitinous integument and the absence of cilia, can be enunciated,
which shall be true of all the members of the group.

It is this extreme elasticity, if I may so speak, of the
crustacean type which renders the construction of any defini-
tion of the Crustacea, which shall include all its members and
exclude the next class, the Arachnida, so difficult. For the
Spiders, Scorpions, Mites, and Ticks, which constitute this class,
possess all the characters which have been just stated to be
common to the Crustacea save one ; when they are provided
with distinct respiratory organs, in fact, these are not external
branchiae, adapted for breathing aerated water, but are a sort of
involution of the integument in the form of tracheal tubes, or

ON CLASSIFICATION.

pulmonary sacs, fitted for the breathing of air directly. But
then many of the lower Arachnida, like the lower Crustacea, are
devoid of special respiratory organs, and so the diagnostic
character fails to be of service.

Fig. 26.

VI VII VIII IX X

Fig. 26. — Diagrammatic section of a Scorpion, the locomotive members being cut away.
", Mouth leading into the pharyngeal pump. The large labrum lies above the mouth,
and at the side of it are the bases of the large chelae, or mandibles, IV., and above
them the chelicerse, or antennas. VI. to XX. Somites of the body. T, Telson ; 6, intes-
tine ; c, anus ; d, indicates the position of the heart ; e, the pulmonary sacs ; /, a
line indicating the position of the ganglionic chain ; g, the cerebral ganglia.

The following common characters of the Arachnida, how-
ever, help out our diagnosis in practice. They 'neverj possess
more than four pairs of locomotive limbs, and the somites of
the abdomen, even when the latter is well developed, are not
provided with limbs. Again, in the higher Arachnida (Fig. 26),
as in the higher Crustacea, the body is composed of twenty
somites, six of which are allotted to the head ; but, in the former
class, one of the two normal pairs of antennae is never developed,
and the eyes are always sessile, while, in the highest Crustacea,
the eyes are mounted upon moveable peduncles, and both pairs
of antennae are developed.

THE MYRIAPODA.

Fig. 27.

Fig. 27. — Anterior part of the body of Scolopendra Hopei (after Newport). — A, Anterior
part of the body from above ; B, from below ; A, head proper ; B, anterior thoracic
somites ; a, antennae; C, antenna?, labrum, and mandibles (iv') from below ; D, under
view of head, with the two pairs of maxilla} (v' VI1) covering the foregoing.

The Myriapoda (Fig. 27) have the chitinous integument ;
the body divided into somites, provided with articulated ap-
pendages ; and nervous and circulatory organs constructed
upon a similar plan to those of the former groups. The body
consists of more than twenty somites, and those which corre-
spond with the abdomen of Arachnida are provided with loco-
motive limbs.

The head consists of at least five, and probably of six, coa-
lescent and modified somites, and some of the anterior segments
of the body are, in many genera, coalescent, and have their
appendages specially modified to subserve prehension. The
respiratory organs are tracheae, which open by stigmata upon
the surface of the body, and the walls of which are strengthened
by chit in, so disposed as readily to pull out into a spirally coiled
filament.

The Insecta, lastly, have respiratory organs like those of
the Myriapoda, with a nervous and a circulatory system dis-

ON CLASSIFICATION.

posed essentially as in this and the two preceding classes. But
the total number of somites of the body never exceeds twenty.

Fig. 28.

Fiff. 29.

Fig. 29. — Longitudinal and vertical
section of the abdomen of a
male Cockroach (Blattci). — 1,
2, 3, 4, &c., terga and sterna of
the abdomen ; t, testis ; v,
aperture of the vas deferens ;
A, anus.

THE INSECTA. 57

Of these five certainly, and six probably, constitute the head,
which possesses a pair of antennae, a pair of mandibles, and two
pairs of maxillae; the hinder pair of which arc coalescent, and
form the organ called the " labium."

Three, or perhaps, [in some cases, more, somites unite and
become specially modified to form the thorax, to which the
three pairs of locomotive limbs, characteristic of perfect insects,*
are attached.

Two additional pairs of locomotive organs — the wings —
are developed, in most insects, from the tergal walls of the
second and third thoracic somites. No locomotive limbs are
ever developed from the abdomen of the adult insect, but the
ventral portions of the abdominal somites, from the eighth
backwards, are often metamorphosed into apparatuses ancillary
to the generative function (Figs. 28 and 29).

* The female Stylops is stated to possess no thoracic limbs.

LECTURE IV,

ON THE CLASSIFICATION OF ANIMALS.

THE VEETEBEATA; OE PISCES, AMPHIBIA, EEPTILIA, AVES, AND

MAMMALIA.

In the rapid survey of the animal kingdom with which we have
been occupied in the preceding lectures, I have, for reasons
which will be obvious by and by, taken group by group, and
considered each separately upon its own merits, without attempt-
ing to say anything of the characteristics of the larger divisions
into which these classes may be arranged. That is a point to
which I shall return on a future occasion.

But with those animals which are called " vertebrated,"
such a course as this would involve a great and unprofitable ex-
penditure of time and much repetition ; because the five groups
of animals which pass under this name — the classes Pisces, Am-
phibia, Reptilia, Aves, and Mammalia — are obviously united and
bound together by many common characteristics, and are well
known to be so connected. I shall commence the present
lecture, therefore, by enumerating the most important of those
structural peculiarities which these five great divisions exhibit
in common.

In the animals to which our attention has hitherto been
confined, the external, or integumentary and parietal, portion
of the blastoderm never becomes developed into more than a
single saccular, or tubular, investment, which incloses all the
viscera. So that if we make a transverse section of any one of
the.-' animals endowed with a sufficiently high organization

tjik VEi;iT,r,i;.\TA.

to possess a nervous system and ;i heart, that section mav be
represented diagrammatically as in Fig. 30 (I.), where 1* repre-

l-V. 30.

.Ar-

il, m

jyt-

Fig. 30. — Diagrams representing generalised sections of* one of the higher I a vertebrates
(I. II.), and of a Vertebrate (III. IV.); I. III. transverse, II. IV. longitudinal sec-
tion. A, alimentary canal; H, heart; P, parietes of the body; P', parietes of the
neural canal; N, nervous centres of Invertebrate; N1, sympathetic, and N2, cerebro-
spinal centres of Vertebrate ; ch, notochord ; M, mouth.

sents the parietes or wall of the body, A the alimentary canal, H
the heart, and N the nervous centres. It will be observed that the
alimentary canal is in the middle, the principal centres of the
nervous system upon one side of it, and the heart upon the
other. In none of these animals, again, would you discover, in
the embryonic state, any partition, formed by the original ex-
ternal parietes of the body, between the nervous centres and the
alimentary canal.

But, in the five vertebrate classes, the parietal portion of the
blastoderm of the embryo always becomes raised up, upon each
side of the middle line, into a ridge, so that a long groove is
formed between the parallel ridges thus developed ; and the
margins of these, eventually uniting with one another, constitute
a second tube parallel with the first, by a modification of the
inner walls of which the vertebrate cerebro-spinal nervous
centres are developed. Hence it follows that, after any verte-
brated animal has passed through the very earliest stages of its
development, it is not a single, but a double tube, and the two

GO ON CLASSIFICATION.

tubes are separated by a partition which was, primitively, a part
of the external parietes of the body, but which now lies, in a
central position, between the cerebro-spinal nervous centres and
the alimentary canal. Hence, a transverse section of any verte-
brated animal may be represented clia grammatically by Fig. 30
(III.), wThere, for the most part, the letters have the same sig-nifi-
cation as in the fore^'oiiis; case, but where P' denotes the
second, or cerebro-spinal, tube. The visceral tube (P) contains,
as in the case of the invertebrate animal, the alimentarv canal,
the heart, and certain nervous centres, belonging to the so-called
sympathetic system. This nervous system and the heart are
situated upon opposite sides of the alimentary canal, the sympa-
thetic corresponding in position, and in forming a double chain
of ganglia, with the chief nervous centres of the invertebrate ; so
that the cerebro-spinal tube appears to be a super-addition, — a
something not represented in the invertebrate series. The
formation of the cerebro-spinal tube of vertebrates, in the
manner which I have described to von, is a well-established
fact ; nor do I entertain any doubt that the cerebro-spinal
centres, viz., the brain and the spinal cord of vertebrates, are the
result of a modification of that serous layer of the germ which
is continuous elsewhere with the epidermis. Two years ago,
I took some pains to verify the remarkable discoveries of
Remak in relation to this point, and, so far as the chick is
concerned, his statements appeared to me to be fully borne
out. But, as Von Baer long ago suggested, it is a necessary
result of these facts that there can be no comparison between
the cerebro-spinal nervous centres of the Verttbrata and the
ganglionated nervous centres of the Invertebrata, and the
homologues of the latter must probably be suught in the sym-
pathetic.

Doubtless in close connection with this profound difference
between the chief nervous centres of the vertebrate and the
invertebrate is another remarkable structural contrast. In all
the higher invertebrates, with a well-developed nervous system,
the latter is perforated by the gullet, so that the mouth is
situated upon the same side of the body as the principal masses
• •I' the nervous system, and some of the ganglia of the latter lie

THE VERTEBRATA. 61

in front of, and others behind, the oesophagus. A longitudinal
vertical section of such an animal, therefore, may be represented
by Fig. 30 (II.).

A similar section of a vertebrated animal shows, on the con-
trary, the chief centre of the nervous system not to be perforated
by the oesophagus; the latter turning away from it and opening
upon the opposite side of the body (Fig. 30, IV.).

Another structure sharply distinctive of the vertebrate
classes is the " chorda dorsal is ' or " notochord," an organ of
which no trace has yet been discovered in any of the inverte-
brates, though it invariably exists, in early embryonic life
at least, in every vertebrate. Before the cerebro-spinal canal is
complete, in fact, the substance of the centre of its floor, be-
neath the primitive median line of the embryo, becomes differ-
entiated into a rod-like cellular structure, which tapers to both
its extremities ; and, in a histological sense, remains com-
paratively stationary, while the adjacent embryonic tissues are
undergoing the most rapid and varied metamorphoses.

To these great differences between vertebrates and in-
vertebrates, in their early condition, many others might be
added. In all Vertebrata that part of the wall of the body
which lies at the sides of, and immediately behind the mouth,
exhibits a series of thickenings parallel with one another and
transverse to the axis of the bodv, which mav be five, or more,
in number, and are termed the "visceral arches." The inter-
spaces between these arches becoming thinner and thinner, are
at length perforated by corresponding clefts, which place the
cavity of the pharynx in free communication with the exterior.
Nothing corresponding with these arches and clefts is known in
the Invertebrata.

A vertebrated animal may be devoid of articulated limbs,
and it never possesses more than two pair. These limbs always
have an internal skeleton, to which the muscles moving the
limbs are attached. Whenever an invertebrated animal possesses
articulated limbs, the skeleton to which the muscles are at-
tached is external, or is connected writh an external body
skeleton.

When an invertebrated animal possesses organs of mastica-

62 ON CLASSIFICATION.

tion, these are either hard productions of the alimentary mucous
membrane, or are modified limbs. In the latter case there
may be many pairs of them — numerous Crustacea, for example,
have eight pairs of limbs devoted to this function. In no verte-
brated animal, on the other hand, are limbs so modified and
functionally applied, the jaws being always parts of the cephalic
parietes specially metamorphosed, and totally distinct in their
nature from the limbs. All vertebrated animals, finally, possess
a distinct vascular system, containing blood with suspended
corpuscles of one kind, or of two, or even three, distinct kinds.
In all, save one, there is a single valvular heart — the vessels of
the exception, Amphioxus, possessing numerous contractile dila-
tations. All vertebrates possess a " hepatic portal system," the
blood of the alimentary canal never being wholly returned
directly to the heart by the ordinary veins, but being more or
less largely collected into a trunk, the " portal vein," which
ramifies through and supplies the liver.

These are the most important characters by which the verte-
brate classes are distinguished, as a whole, from the other classes
of the animal kingdom ; and their number and importance go a
long way to justify the step taken by Lamarck when he divided
the animal kingdom into the two primary subdivisions of Verte-
brata and Invertebrafa.

If we seek now to construct definitions of the first two classes
of the Vertebrata, Pisces and Amphibia, we shall meet with
some difficulties, arising partly from the wide variations ob-
servable in the structure of fishes, and partly from the close
affinity which exists between them and the Amphibia.

No fish exhibits any trace of that temj^orary appendage of
the embryo of the higher vertebrates which is termed an amnion,
nor can any fish be said to possess an allantois, though the
urinary bladder of fishes may possibly be a rudiment of that
structure. The posterior visceral clefts and arches * of fishes
persist throughout life, and are usually more numerous than in
other vertebrates ; while upon, or in connexion with, them are

* The relation of the perforated pharynx of Ampkioxus to the visceral arches and
clefts is not known.

THE AMPHIBIA. 63

developed villi, or lamellae, which subserve the respiratory
function.

Median fins, formed by prolongations of the integument,
supported by one or other kind of skeleton, are very character-
istic of fishes, and it is questionable if any fish exists altogether
devoid of the system of median fin-rays and their supports,
which have been termed inter-spinous bones and cartilages.
On the other hand, no vertebrate animal, other than a fish, is
known to possess them.

When the limbs, or pectoral and ventral fins, of fishes are de-
veloped, they always exhibit a more or less complete fringe of
fin-rays. ISo amphibian is known to possess such rays in its
lateral appendages, but there is some reason to believe that the
extinct Ichtliyosauria may have been provided with them.

In most fishes, the nasal sacs do not communicate directly
with the cavity of the mouth, but the Myxinoids and Lepiclosiren
are exceptions to this rule.

The blood-corpuscles of fishes are always nucleated, and are
commonly red, but by a singular exception those of Aiwphioxus
(the Lancelet, which is an exception to most rules of piscine
organization) are colourless.

Almost all fishes have the heart divided into two auricles
and one ventricle ; but Amphioxus, as I have previously stated,
is devoid of any special heart, being provided instead with a
number of contractile, vascular dilatations ; while Lejndosiren
possesses two auricles, and, at the same time, is provided with
true lung's.

It is useless therefore to appeal to the olfactory organ, the
blood, the heart, or the respiratory organs, for characters at once
universally applicable to, and diagnostic of, fishes.

The Amphibia (or Batracliians, Salamandroids, Coecilise, and
Labyrinthodonts) resemble fishes, and differ from all other verte-
brates in the entire absence of an amnion, and in having only
the urinary bladder to represent the allantois. They have red
nucleated blood-corpuscles. Yet again they resemble fishes, and
differ from all other vertebrates in the fact that filaments exer-
cising a respiratory function, or branchiae, are developed from
their visceral arches during a longer or shorter period.

04 ON CLASSIFICATION.

Xone are known to be provided with median fins supported
by fin-rays, and their limbs are never fringed with fin-rays.

Furthermore, in all Am/phibia which possess limbs, the skele-
ton of these limbs is divisible into parts which obviously corre-
spond with those found in the higher vertebrates. That is to
say, in the fore limbs there are cartilages, or bones, answering in
their essential characters and arrangement to the humerus, radius
and ulna, carpus, metacarpus, and phalanges ; and, in the hind
limb, to the femur, tibia and fibula, tarsus, metatarsus, and pha-
langes of the higher vertebrates. This is the case in no fish ;
for, whether fishes possess parts corresponding with the humerus,
radius and ulna, &c, or not, certain it is that the elements of
their limb skeletons are very differently disposed from the ar-
rangement which obtains in Amphibia and in higher vertebrates.

In all Amphibia the skull articulates with the spinal column
by two condyles, and the basi-occipital remains unossified.
Furthermore, the cranial peduncle, or suspensorium, to which
the lower jaw is articulated, gives attachment to the hyoidean
apparatus.

These last are characters by which the Amphibia are sharply
distinguished from the higher vertebrates.

There is a striking contrast between the close affinity of the
fish and the amphibian and the wide separation of the Amphibia
from the succeeding classes, all of which possess, in the embry-
onic state, a well-developed amnion and allantois, the latter almost
always taking on, directly or indirectly, a respiratory function.

The amnion is a sac filled with fluid which envelopes and
shelters the embryo, during its slow assumption of the condition
in which it is competent to breathe and receive food from with-
out. The mode of its formation is shown in the accompanying
figures of the early stages of development of the common fowl.
Fig. 31, A, represents the first step in the differentiation of the
embryo from the central portion of the blastoderm — that thin
membranous cellular expansion which lies on the surface of the
yelk where we see the cicatricula, or " tread." A well-defined,
though shallow, straight groove, the " primitive groove," bounded
at the sides by a slight elevation of the blastoderm, indicating the

T1IK AMPIII1UA.

position of the future longitudinal axis of the body of the chick.
Soon, the lateral boundaries of this groove, in what will become

Fig. 31.

Fig. 31. — Development of the Chick.

A. First rudiment of the embryo ; a, its cephalic ; 6, its caudal end ; c, primitive groove.

B. The embryo further advanced; a, b, c, as before; d, the dorsal laminae developed in

the cephalic region only, and nearly uniting in the middle line; e, the proto -vertebra;.

C. Letters as before. The dorsal laminae have united throughout the greater part of the

cephalic region, and are beginning to unite in the anterior spinal region.

D. Embryo further advanced (second day), the dorsal laminae having united throughout

nearly their whole length. The proto- vertebrae have increased in number, and the
omphalo-meseraic veins,/, are visible.
The embryos are drawn of the same absolute length, but it will be understood that the
older embryos are, in nature, longer than the vounrer.

ON CLASSIFICATION.

the anterior region of the body, grow up into plates — the dorsal
laminae (Fig. 31, B) ; and these dorsal laminae, at length
uniting, inclose the future cerebrospinal cavity (Fig. 31, C, D).
The blastoderm, beyond the region at which the dorsal laminae
are developed, grows downwards to form the ventral laminae, and
where the margins of these pass into the general blastoderm,
the outer serous, or epidermic, layer rises up into a fold, which
encircles the whole embryo ; and the anterior and posterior parts
of this fold growing more rapidly than the lateral portions, form
a kind of hood for the cephalic and caudal ends of the body
respectively (Fig. 32, E). The margins of the hoods and of

Fig. 32.

e m

■ ■

F-t-n

Am.

Fig. 321 — Development of the Chick.

E. Embryo at the third day ; g, heart ; h, eye ; i, ear ; k, visceral arches and clefts ;

I, m, anterior and posterior folds of the amnion, which have not yet united over the
body; 1,2, 3, first, second, and third cerebral vesicles; la, vesicle of the third
ventricle.

F. Chick at the fifth day ; n, o, rudiments of the anterior and posterior extremities ; Am,

amnion ; All, allantois ; Urn, umbilical vesicle.

G. Under view of the head of the embryo F, the first visceral arch being cut away.

THE REPTILIA. (57

their lateral continuations at length meet over the middle line

of the body, and there coalesce: so that the body is covered for
a while by a double sac, the inner layer of which is formed by
that wall of the fold of the serous layer which is inferior, or
nearest to the body of the embryo ; while the outer layer is
formed by that wall which is superior, or furthest from the body
of the embryo. The outer layer eventually disapj^ears as a dis-
tinct structure, while the inner remains as the amnion. From
the mode of formation which has been described, it results that
the amnion is a shut sac, enveloping the body of the embryo ;
and is continuous, on the ventral side of the body, with the in-
tegument of a region which eventuallv becomes the umbilicus
(Fig. 32, F).

The allantois is developed much later than the amnion,
neither from the serous nor from the mucous (or epidermic and
epithelial) layers of the germ, but from that intermediate stratum
whence the bones, muscles, and vessels are evolved. It arises,
as a solid mass, from the under part of the body of the embryo,
behind the primitive intestinal cavity ; and, enlarging, becomes
a vesicle, which rapidly increases in size, envelopes the whole
embryo, and, being abundantly supplied with arterial vessels
from the aorta, serves as the great instrument of respiration
during foetal life ; the porosity of the egg-shell allowing the
allantoic blood to exchange its excess of carbonic acid for
oxygen by osmosis.

The amnion and the external part of the allantois are thrown
off at birth.

That which has just been stated respecting the development
and characters of the amnion and allantois of the chick is true
not only of all Birds, but of all Beptilia.

All embryonic Reptilia are provided with an amnion and an
allantois, like those just described in the foetal fowl. In the
embryonic state, also, they possess visceral arches and clefts, but
no respiratory tufts are ever developed in the arches, nor are
reptiles endowed with an apparatus for breathing the air dis-
solved in wrater at any period of their existence. The skull of
all Reptilia is articulated with the vertebral column by a single
condyle, into which the ossified basi-occipital enters largely

f 2

ON CLASSIFICATION.

(Fig. 33). Each ramus of the lower jaw is composed of a number
of pieces, and articulates with the skull, not directly, but by the

Fig. 33. — The occipital condyle of a Crocodile's skull viewed from behind. — B.O, Basi-
occipital ; E.O, Ex-occipital; S.O, Supra-occipital.

intervention of a bone — the os quadratum — with which the
hyoidean apparatus is not immediately connected (Fig. 34).

Fig. 34.

Fig. 34.-

-The skull of a Lizard (Cyclodus). — D D, Dentary piece of the lower jaw ;
Qu, Os quadratum; Sq, Squamosal.

The fore-limb of Eeptiles never takes the form of a wing, such
as is seen in Birds ; the " wing " of the remarkable extinct flying-
reptiles, the Pterodadyles, being constructed on a totally dif-
ferent principle from that of a bird. In no known reptile, again,
are the metatarsal and tarsal bones ancliylosed into one bone.

In all Rejitilia the greater and lesser circulations are directly
connected together, within, or in the immediate neighbourhood
of, the heart ; so that the aorta, which is formed by the union
of two arches, contains a mixture of venous and arterial blood.

THE AVES. (',!)

The blood is cold, and the majority of the blood-corpuscles are
red, oval, and nucleated. The bronchial tubes are not connected
at the surface of the lungs with terminal saccular dilatations, or
air-vesicles. When, as is ordinarily the case, the superficial
layers of the epidermis of Reptiles are converted into horn, the
corneous matter takes the form of broad plates, or of overlapping
scales, neither plates nor scales being developed within pouches
of the integument.

The class of Aves consists of animals so essentially similar
to Reptiles in all the most essential features of their organization,
that Birds ma}^ be said to be merely an extremely modified and
aberrant Reptilian type.

As I have already stated, they possess an amnion and a respi-
ratory allantois, and the visceral arches never develop branchial
appendages. The skull is articulated with the vertebral column
by a single condyle, into which the ossified basi-occipital enters
largely. Each ramus of the lower jaw, composed, as in Reptiles,
of a number of pieces, is connected with the skull by an os
quadratum, to which the hyoidean apparatus is not suspended.

In no existing bird does the terminal division of the fore-limb
possess more than two digits terminated by claws, and the meta-
carpal bones are commonly anchylosed together, so that the
" manus " is of little use, save as a support for feathers.

In the hind limb of all birds the tarsal and metatarsal bones
become more or less completely fixed, and the latter, anchylosed
together, so as to form a single osseous mass, the " tarso-meta-
tarsus."

The greater and lesser circulations of birds are completely
separate, and there is only one aortic arch, the right. The right
ventricle has a muscular vah^e. The blood is hot, hotter on the
average than that of any other vertebrates, and the majority of
the blood-corpuscles are oval, red, and nucleated. The bron-
chial tubes open upon the surface of the lungs into air-sacs,
which differ in number and in development in different birds.
Lastly, the integument of birds is always provided with horny
appendages, which result from the conversion into horn of the
cells of the outer layer of the epidermis. But the majority of

70 ON CLASSIFICATION.

these appendages, which are termed " feathers," do not take the
form of mere plates developed upon the surface of the skin, but
are evolved within sacs from the surfaces of conical papilla of
the dermis. The external surface of the dermal papilla, whence
a feather is to be developed, is provided upon its dorsal surface
with a median groove, which becomes shallower towards the
apex of the papilla. From this median groove lateral furrows
proceed at an open angle, and passing round upon the under
surface of the papilla, become shallower, until, in the middle line,
opposite the dorsal median groove, they become obsolete. Minor
grooves run at right angles to the lateral furrows. Hence the
surface of the papilla has the character of a kind of mould, and
if it were repeatedly dipped in such a substance as a solution of
gelatine, and withdrawn to cool until its whole surface was covered
with an even coat of that substance, it is clear that the gelatinous
coat would be thickest at the basal or anterior end of the median
groove, at the median ends of the lateral furrows, and at those ends
of the minor grooves which open into them ; while it would be very
thin at the apices of the median and lateral grooves, and between
the ends of the minor grooves. If, therefore, the hollow cone of
gelatine, removed from its mould, were stretched from within ; or
if its thinnest parts became weak by drying ; it would tend to give
way, along the inferior median line, opposite the rod-like cast of
the median groove and between the ends of the casts of the
lateral furrows, as well, as between each of the minor grooves,
and the hollow cone would expand into a flat feather-like struc-
ture with a median shaft, as a " vane " formed of " barbs " and
" barbules." In point of fact, in the development of a feather
such a cast of the dermal papilla is formed, though not in
gelatine, but in the horny epidermic layer developed upon the
mould, and, as this is thrust outwards, it opens out in the
manner just described. After a certain period of growth tli;1
papilla of the feather ceases to be grooved, and a continuous
horny cylinder is formed, which constitutes the " quill."

Between Aves and Mammalia there is a hiatus, not perhaps,
in some respects, quite so wide as that between Amphibia and
Reptilia, but still very considerable.

All Mammals possess an amnion of an essentially similar

THE MAMMALIA. 7]

character to that of Birds and Keptiles, and all have an allantois.
But the latter either ceases to exist after a very early period of
foetal life, or else it is " placentiferous," and serves as the means
of intercommunication between the parent and the offspring.
Of the nature and characters of the "placenta" developed in the
majority of the Mammalia I shall speak more particularly by
and by. For the present, I pass it over as a structure not uni-
versally characteristic of the class.

The visceral arches are, throughout life, as completely devoid
of branchial appendages in Mammals, as in Birds and Keptiles.
In the skull, the basi-occipital is well ossified, and, with the ex-
occipitals, enters into the formation of the cranio-spinal articula-
tion ; the occipital condyle thus formed, however, is not single,
as in Reptiles and Birds, but double, and the atlas has corre-
sponding articular facets.

J-.0

Fig. 35. — The occipital condyles of a Dog's skull viewed from behind. — Signification of

the letters as in Fig. 33.

Each ramus of the lower jaw is composed of only a single
piece, and this articulates directly with the squamosal bone
of the skull, and not with the representative of the quadrate
bone.

The greater and lesser circulations of Mammals are as com-
pletely distinct as in Birds, and there is but a single aortic arch,
the left. The majority of the blood-corpuscles are red, free
nuclei, and these are always discoidal, and usually circular in
form. The blood is hot. There is a complete diaphragm, and
none of the bronchi end in air-sacs.

ON CLASSIFICATION.

Some part or other of the integument of all Mammals exhibits
"hairs" — horny modifications of the epidermis — which so far
resemble feathers, that they are developed upon papillae inclosed
within sacs ; but, on the other hand, differ from the horny ap-
pendages of birds, in not splitting up as they are protruded, in
the fashion so characteristic of feathers.

Fig. 36.

Kijr, 36. — The skull of a Dog. — D. Ramus of the lower jaw; Sq. Squamosal.

Finally, all Mammals are provided with organs for the secre-
tion of a fluid which subserves the nourishment of the young
after birth. The fluid is milk ; the organs are the so-called
"mammary" glands, and may probably be regarded as an ex-
treme modification of the cutaneous sebaceous glands. These
glands are aggregated into two or more masses, disposed upon each
side of the median line of the ventral surface of the body ; and,
in almost all Mammals, the aggregated ducts of each mass open
upon an elevation of the skin common to all — the nipple or
teat. To this the mouth of the newly-born Mammal is applied,
and from it, either by suction on the part of the young, or by
the compressive action of a special muscle on the part of the
parent, the nutritive fluid makes its way into the stomach of the
former.

7:;

LECTURE V.

ON THE CLASSIFICATION OF ANIMALS.

ON THE ARRANGEMENT OF THE CLASSES INTO LARGER

GROUPS.

Having now arrived at the end of the list of classes, and having
endeavoured to furnish you with a statement of the structural
features common to, and characteristic of, each class, it will be
my next object to discuss the relations of these classes one to
another, and to inquire how far they present such common
characters as will enable us to group them into larger divisions.

And, to commence with the highest classes, it is clear that
the Mammalia, Aves, and Reptilia are united together by certain
very striking features of their development. All possess an
amnion and an allantois, and are devoid, throughout life, of any
apparatus for breathing the air which is dissolved in water. In
other words, they constitute what has been termed the "pro-
vince " of Abranchiate Vertebrata, in contradistinction to
Pisces and Amphibia, which possess no amnion, nor allantois
(or at most a rudimentary one), and, being always provided at a
certain period, if not throughout life, with branchiae, have been
called Branchiate Yertebbata.

The abranchiate, however, form a far less homogeneous as-
semblage than the branchiate Vertebrata — Mammals being so
stroDgly separated from Reptiles and Birds that I am disposed
to regard them as constituting one of three primary divisions, or
provinces, of the Vertebrata. The structure of the occipital
condyles, the structure and mode of articulation of the mandi-

74 ON CLASSIFICATION.

bular rami, the presence of mammary glands, and the non-
nucleated red blood-corpuscles appear to separate Mammals as
widely from Birds and Reptiles as the latter are separated from
Amphibia and Fishes.

Thus the classes of the Vertebrata are capable of being-
grouped into three provinces : (I.) the Ichthyoids (comprising
Fishes and Amphibia), defined by the presence of branchiae
at some period of existence, the absence of an amnion, the
absence, or rudimentary development, of the allantois, nucle-
ated blood-corpuscles, and, as will be seen by and by, a para-
sphenoid in the skull ; (II.) the Sauroids, defined by the absence
of branchiae at all periods of existence, the presence of a well-
developed amnion and allantois, a single occipital condyle,
a complex mandibular ramus articulated to the skull by a
quadrate bone, nucleated blood-corpuscles and no parasphe-
noid, comprising Reptiles and Birds ; and (III.) the Mammals,
devoid of branchiae and with an amnion and an allantois, but with
two occipital condyles and a well-developed basi-occipital and no
parasphenoid ; a simple mandibular ramus articulated with the
squamosal and not with the quadratum, with mammary glands and
with red non-nucleated blood-corpuscles.*

These five classes, whether divided into two or three pro-
vinces, again, present so many characters, already enumerated,
by winch they resemble one another, and differ from all other
animals, that, by universal consent, they are admitted to form
the group of Vertebrata. which takes its place as one of the
primary divisions or " sub-kingdoms '' of the Animal Kingdom.

The next four classes — Insecta, Myriapocla, Arachnida, Crus-
tacea— without doubt also present so many characters in common
as to form a very natural assemblage. All are provided with
articulated limbs attached to a segmented bodv-skeleton — the
latter, like the skeleton of the limbs, being an " exoskeleton,"
or a hardening of that layer which corresponds with the outer
part of the epidermis of Vertebrates. In all, at any rate in the
embryonic condition, the nervous system is composed of a double

* To these may he added the absence of the corpus cullo&um in the brain of
Sauroids and its presence in Mammals. See note p. 89.

the AirniKopouA. 75

chain of ganglia, united by longitudinal commissures, and the
gullet passes between two of these commissures. No one of the
members of these four classes is known to possess vibratile cilia.
The great majority of these animals have a distinct heart, pro-
vided with valvular apertures, which are in communication with
a perivisceral cavity containing corpusculated blood. But the
Cirripedia and the Ostracoda among Crustaceans, and many of
the Mites among Arachnida, have as yet yielded no trace of
distinct circulatory organs, so that the nature of these organs
cannot be taken as a universal character of the larger group we
are seeking ; still less can such a character be found in the
respiratory organs, which vary widely in character, and are
often totally absent as distinct structures. Some years ago I
endeavoured to show* that a striking uniformity of compo-
sition is to be found in the heads of, at any rate, the more highly
organized members of these four classes, and that, typically, the
head of a Crustacean, an Arachnid, a Myriapod, or an Insect is
composed of six somites (or segments corresponding with those
of the body) and their appendages, the latter being modified so
as to serve the purpose of sensory and manducatory organs. I
believe this doctrine to be substantially correct ; and that,
leaving all hypothetical suppositions aside, the head of any
animal belonging to these classes may be demonstrated to
contain never fewer than four, and never more than six somites
with their appendages ; but, until this view has received confir-
mation from other workers, I shall not venture to put forward
any statement based upon it as part of the definition of the
large group or " province ' containing the four classes above
mentioned, which has received from some naturalists the name
of Akticulata, from others that of Arthropoda, the latter
being perhaps the more distinctive and better appellation.

The members of the class Annelida present marked differ-
ences from all the Arthropoda, but resemble them in at least
one important particular ; and that is, the arrangement of the
nervous system, which constitutes a ganglionated double chain,
traversed at one point by the oesophagus. In almost all other

* "On the Agamic Reproduction and Morphology of Aphis" Transactions of
the L/iiiui-an Society, vol. xxii.

76 ON CLASSIFICATION.

respects, Annelids differ widely from Arthropods. It may
be doubted whether any Annelid is devoid of cilia in some part
or other of its organization, and cilia constitute the most im-
portant organs of locomotion in the embryos of many. No
Annelid possesses a heart communicating by valvular apertures
with the perivisceral cavity, none have articulated limbs, and
none possess a head composed of even four modified somites.

Most Annelids are provided with that peculiar system of
vessels termed " pseudo-hrenial ;" but, in some, that system has
not yet been discovered.

In endeavouring to separate from among invertebrated
animals a first large group, comparable to the Vertebrata, it
appears to me that the resemblances between the Annelida and
the Arthropoda outweigh the differences ; and that the characters
of the nervous system and the frequently segmented body,
with imperfect lateral appendages, of the former, necessitate
their assemblage with the Arthropoda into one great division, or
" sub-kingdom," of Annulosa.

But what of the Echinodermata and the Scohcida ? Should
both these great classes be also ranged under the Annulosa ;
or do they belong to different sub-kingdoms ; or, if they belong
to the same, should they constitute a sub-kingdom of their own ?

I will endeavour to reply to these questions in succession.
Whether these two groups belong to the Annulosa or not, must
depend upon whether they possess any characters in common
with the Arthropoda and Annelida other than those which they
have in common with all animals. I can find none of any
great moment. No Echinoderm, or Scolecid, has a definitely
segmented body or bilaterally disposed successive pairs of
appendages. None of these animals lias a longitudinal chain

of inmirlia.

On the other hand, there is much resemblance between the
ciliated larvae of some Scolecids and Echinoderms, and those of
Annelids ; and the form of the body of many Scolecids is so
similar to that of one of the most familiar of Annelids, as to
have earned for both them and the Annelids the common title
of "worms." Nor must it be forgotten that, in the Annelids
there seem to be representatives <»[ that singular system of

THE ANNULOIDA. 77

vessels which attains so large a development as the " water-
vascular " apparatus in many Scolecids.

Whatever value may be attached to these resemblances, it
must, I think, be admitted that, in the present state of our
knowledge, it is impossible to affirm anything absolutely common
to, and yet diagnostic of, all Annidosa and all Echinoderms and
Scolecids. On the other hand, there can be no doubt as to the
many and singular resemblances which unite the Scolecids and
the Echinoderms together. The nervous system of the Echi-
noderm may present considerable differences from that of a
Trematode or Rotifer, but it must be recollected that the com-
parison is not a fair one, seeing that the mouth and gullet of an
Echinoderm, round which its nervous ganglia are arranged, are
not, strictly speaking, the same as the parts so named in a
Rotifer, but are new developments.

And it is exactly in that anomalous method of development
of the Echinoderm within its larva, which is so characteristic
of the whole group of Echinodermata, that this class exhibits its
strong alliance with the Scolecida ; the Turbellaria and Tseniada
exhibiting the only approach to the method of Echinoderm
development known in the xlnimal Kingdom.

A singular larva studied by Johannes Miiller, in one of his
many fruitful visits to the seashore, and termed by him Pilidium,
has furnished, in the hands of subsequent observers (more espe-
cially Krohn, Leuckart, and Pagenstecher), ample proof that
a Nemertes (a genus of Turbellaria) may be developed in a
manner altogether similar to that in which an Echinoderm
takes its origin.

The Pilidium (Fig. 37) is a small, helmet-shaped larva,
with a long flagellum attached like a plume to the summit of
the helmet, the edges and side lobes of which are richly ciliated.
A simple alimentary sac opens upon the under surface of the
body between the lobes (Fig. 37, a).

In this condition, the larva swinis about freely ; but, after a
while, a mass of formative matter appears upon one side of
the alimentary canal, and, elongating gradually, takes on a
worm-like figure. Eventually it grows round the alimentary
canal, and, appropriating it, detaches itself from the Pilidium

ON CLASSIFICATION,

as a Nemertid — provided with the characteristic proboscis, and
the other organs of that group of Turbellaria.

Fig. 37.

Fig. 37. — Pilidium gyrans (after Leuckart and Pagenstecher).

A. Young Pilidium ; a, alimentary canal ; b, rudiment of the Nemertid.

B. Pilidium with a more advanced Nemertid.

C. Newly-freed Nemertid.

Many Trematoda, and all Tsenioid Scolecicla, again, present
an essentially similar process of internal gemmation, in virtue of
which either a separate offspring arises, or an adult is developed
within an embryonic form ; but in these cases the appropria-
tion of the intestine of the primary by that of the secondary
form, which renders the ordinary development of the Echinoderm
so striking, does not occur.

In discussing the characters of the Echinodermata, I have

THE ANNULOIDA AND ODONTOPIIORA. 79

described at length the ambulacral system ; and, in speaking of
the Scolecida, I have no less insisted upon the peculiarities of
the " water-vascular system." But it is impossible to compare
these two systems of vessels without being struck by their
similarity. Each is a system of canals, opening externally, and
ciliated within ; and the circumstance that the two apparatuses
are turned to different purposes in two distinct groups of the
animal kingdom, seems to me no more to militate against their
homology, than the respiratory function of the limbs of Phyl-
lopod Crustacea militates against the homology of these limbs
with the purely locomotive appendages of other Crustaceans.

Thus it appears that the Ecliinodermata and the Scole-
eida are so closely connected that they can by no means
be placed in separate sub-kingdoms ; and in the course of
studying the other sub-kingdoms it will be quite obvious that,
unless they are to occupy an independent position, there is no
place for them anywhere, save among the Annulosa. I have
hitherto been accustomed to consider them, under the name of
the Annuloida, as a division of this sub-kingdom ; but until
some structural character can be discovered by which all the
Annuhida agree with the Annidosa, and differ from other
animals, I am much inclined to think it would conduce to the
formation of clear conceptions in zoology if the Annuhida were
regarded as a distinct primary division of the Animal Kingdom.

If we now turn to the other column of classes of invertebrate
animals (supra, p. 6), the four last on the list, viz., Cephalopoda,
Pteropjoda, Pulmogasteropoda, and Branchiogasteropoda, have a
number of well-marked characters in common. In all, the
nervous system is composed of three principal pairs of ganglia —
cerebral, pedal, and parieto-splanchnic — united by commissures.
All possess that remarkable buccal apparatus, the odontophore, —
whence I have ventured to propose the name of Odontophoka
for the group. The circulatory and respiratory organs vary a
good deal, but none are provided with double lamellar gills upon
each side of the body.

The Lamellibr and data stand in somewhat the same relation
to the Odoidophora as the Annelida to the Arthropoda. The

80 ON CLASSIFICATION.

Lamellibranchs have the three fundamental pairs of ganglia of
the Odontopliora, but they possess no trace of the odontophore.
Furthermore, they are all provided with bivalve external pallial
shells, the valves being right and left in relation to the body.
No shell of this kind is found in any of the Odoniopliora.
Almost all Lamellibranchs, lastly, have a pair of lamellar gills
on each side of the body, and all are provided with auriculate
hearts. No doubt the Odontophora and the Lamellibranchiata
properly form parts of one and the same sub-kingdom, Mol-
lusca, and the three classes which follow, viz., the Aseidioida,
BracMopoda, and Polyzoa, are usually included in the same sub-
kingdom.

But the difficulty of framing a definition which shall include
the last-named classes with the Lamellibranchiata and Odontophora
is almost as great as in the parallel case of the Annuloida and
Annulosa ; while, on the other hand, the Ascidians, Brachiopods,
and Polyzoa exhibit many features in common. Thus the ner-
vous system is greatly simplified in all three classes, consisting,
in the Aseidioida and Polyzoa, of a single ganglion, sending
perhaps a commissural cord round the gullet. In the Brachiopoda
the chief ganglia, which appear to be the homologues of the pedal
ganglia of the higher mollusks, and are connected by a circuni-
cesophageal cord, are combined with accessory ganglia, but these
do not seem to be identifiable with the pedal or the parieto-
splanchnic ganglia.

Again, the fact that the heart, when present, is of a simple
tubular, or saccular, character, and is devoid of any separation
into auricle and ventricle, constitutes a wide difference between
these three classes and the higher Mollusks. On the other
hand, these classes, which may be convenientlv denominated
Molluscolda, resemble one another in the fact (so far as I am
aware there is only one exception, Appendicular ia) that the
mouth is provided with ciliated tentacula, disposed in a circle,
or in a horse-shoe shape, or fringing long arms ; that it leads
into a large, and sometimes an exceedingly large, pharynx ; and
that in two of the three, at least, that system of cavities commu-
nicating with the exterior, which has been called the "atrial
system," is greatly developed.

• THE MOLLUSCOIDA AND CCELENTERATA. 81

1 cannot doubt, then, that the Molluscoida form a natural
assemblage ; but, until the precise characters, it' any exist, which
unite them with the Mollusca proper can be clearly defined, I
am inclined to think it might be better, as in the case of the
Annuloicla, to recognise them as a separate diviskm of the
Animal Kingdom.

The next two classes — the Actinozoa and the Hydrozoa — con-
stitute one of the most natural divisions of the Animal Kingdom
— the Ccelenterata of Frev and Leuckart. In all these ani-
mals, the substance of the body is differentiated into those histo-
logical elements which have been termed cells, and the latter
are primarily disposed in two layers, an external and an internal,
constituting the " ectoderm " and " endoderm."

Among animals which possess this histological structure, the
Ccelenterata stand alone, in having an alimentary canal, which
is open at its inner end and communicates freely, by means of
this aperture, with the general cavity of the body. In a large
proportion of these animals the prehensile organs are hollow
tentacles, disposed in a circle around the mouth, and all (unless
the Ctenophora should prove to be a partial exception to the
rule) are provided with very remarkable organs of offence and
defence, termed " thread cells ' or " nematocysts." These,
when well exhibited, as, for example, by the common freshwater
polype {Hydra), are oval, elastic sacs, containing a long coiled
filament, barbed at its base, and serrated along the edges.
When fully developed, the sacs are tensely filled with fluid, and
the slightest touch is sufficient to cause the retroversion of the
filament, which then projects beyond the sac for a distance,
which is not uncommonly equal to many times the length of
the latter. These fine filaments readily penetrate any delicate
animal tissue with which they are brought into contact, and
cause great irritation in the human skin when they are of large
size. Nor can it be doubted that they exert a similarly noxious
influence upon the aquatic animals which are seized by, and
serve as prey to, the Actinozoa and Hydrozoa. Characteristic as
these organs are of the Ccelenterates, however, it must not be
imagined that they are absolutely peculiar to the sub-kingdom ;
for some nudibranchiate Molhisea, such as Eolis, are armed with

82 ON CLASSIFICATION.

similar weapons, and the integument of certain Turhellaria, and
even of some Infusoria, is provided with bodies which seem to
be of a not altogether dissimilar character.

No Coelenterate possesses any circulatory organs, unless the
cilia which line the general cavity of the body can be regarded
as such ; and a nervous system has, at present, been clearly made
out only in the Ctenojrfwra. Here its central mass occupies a
position which is very unlike that in which the principal masses
of the central nervous system are found in other invertebrate
animals, being situated upon that side of the body which is dia-
metrically opposed to the mouth.

Whatever extension our knowledge of the nervous apparatus
of the Ccelenterates may, and not improbably will, receive from
future investigators, the positive characters afforded by the histo-
logical features of their substance, and the free opening of their
alimentary canal into the general cavity of the body, are such
as to separate them, as a sub-kingdom, as sharply defined and
devoid of transitional forms as that of the Vertebrata, from the
rest of the Animal Kingdom.

Great difficulties stand in the way of any satisfactory group-
ing of the remaining classes, if we are determined to remain true
to the principle that the definition of a group shall hold good
of all members of that group, and not of any others, — a prin-
ciple which lies at the foundation of all sound classification.

In possessing cilia, as locomotive and ingestive organs ; in
being provided with a contractile water receptacle with canals
proceeding from it (in some cases at any rate) into the substance
of the body ; in their tendency to become encysted and assume
a resting condition, the Infusoria undoubtedly exhibit analogies
with the lower Annuhida, such as the Turhellaria, Bofifera, and
Trematoda.

But the entire absence, so far as our present knowledge
goes, of a nervous system, the abrupt termination of the gullet
in a central semi-fluid sarcodic mass, and the very peculiar cha-
racters of the reproductive organs, of the Infusoria, separate
them widely from the Annuhida, though it seems to me not-
improbable that the gap may hereafter be considerably dimi-
nished by observation of the lower forms of Turhellaria.

THE INFUSOBIA AND PROTOZOA. 83

At present the Infusoria are usually regarded as forming
part of the same sub-kingdom as the Spongioid, Bhizopoda, and
Gregarinida, and as closely allied to them. But, so far as I am
aware, no definition can be framed which will yield characters at
once common to, and distinctive of, all these four groups ; while
recent discoveries tend to widen so greatly the hiatus between
the Infusoria and the other three classes, that I greatly doubt if
the sub-kingdom Protozoa can be retained in its old sense.

But if the Infusoria be excluded from it, the remaining
groups, notwithstanding the imperfection of our knowledge
regarding some of them, exhibit a considerable community of
partly negative and partly positive characters.

The Spongida, Bhizopoda, and Gregarinida^ in fact, are all
devoid of any definite oral aperture ; a considerable extent, and
sometimes the whole, of the outer surface of the body acting as
an ingestive apparatus. Furthermore, the bodies of these ani-
mals, or the constituent particles of the compound aggregations,
such as the Sponges, exhibit incessant changes of form — the
body wall being pushed out at one point and drawn in at another
— to such an extent, in some cases, as to give rise to long lobate,
or filamentous, processes, which are termed " pseudopodia,"

Finally, all these classes agree in the absence of any well-
defined organs of reproduction, innervation, or blood circulation.

In my first lecture upon Classification, I passed very briefly
over the class Bhizopoda, intending to return to the discussion of
its limits, and of the value of its subdivisions, when discussing
the subdivisions of classes generally. But as time will not
permit me to enter at any length upon the greater part of this
branch of my subject, I will content myself with briefly stating
the conclusions at which I have arrived from a careful study of
the extant literature of the subject, combined with some old
investigations of my own.

It appears that three, or perhaps four, types of structure
obtain among the Bhizopoda —

1st. That of the Amoebae— Bhizopods with usually short
pseudopodia, a nucleus, and a contractile vesicle.

2nd. That of the Foraminifera — Bhizopods devoid of nuclei

G 2

ON CLASSIFICATION.

and of contractile vesicles, and, for the most part, with long
pseudopodia, which commonly run into one another and become
reticulated.

3rd. That of the Thalassicolhe, provided with structureless
cysts containing cellular elements and sarcode, and surrounded
by a layer of sarcode, giving off pseudopodia, which commonly
stand out like rays, but may and do run into one another, and
so form networks.

Fiff. 38.

- i> .

;p3\JLa

Fig. 38. — Sphcerozoum ovodimare (after Haeckel), one of the Thalassicollcr .

While a fourth type of structure is probably furnished hy
those anomalous creatures, the Acinette, the radiating processes
of which serve as suctorial tubes down which the juices of their
prey are conveyed.

That the Bhizopoda are divisible into at least three groups,
corresponding to the three first-mentioned types of organiza-
tion, seems to me unquestionable ; but it is another matter, and
one on which I offer no opinion, what should be the exact
limits of these groups, and what denominations we ought to
employ for them. And it must be recollected that, so long as
naturalists are unacquainted with the sexual method of repro-

THE EIGHT PRIMARY GROUPS OF ANIMALS. 85

duction of these animals, they are, to a certain extent, working
in the dark.

In conclusion, I may sum up the results of this lecture by
stating that, in the present state of our knowledge, the whole
Animal Kingdom is divisible into eight primary categories or
groups, no two of which are susceptible, in the present state of
knowledge, of being defined by characters which shall be at
once common and diagnostic.

These groups are the —

Vertebrata.
mollusca. annulosa.

molluscoida. annuloida.

c(elenterata. infusoria.

Protozoa.

I leave aside altogether the question of the equivalency of
these groups ; and, as I have already stated, I entertain some
doubts regarding the permanency of one — the Infusoria — as a
distinct primary division. Nor, in view of the many analogies
between the Mollusca and the Molluscoida, the Annulosa and
the Annuloida, do I think it very improbable that, hereafter,
some common and distinctive characters may possibly be dis-
covered which shall unite these pairs respectively. But the
discoveries which shall effect this simplification have not yet
been made, and our classification should express not anticipa-
tions, but facts.

I have not thought it necessary or expedient, thus far, to
enter into any criticism of the views of other naturalists, or to
point out in what respect I have departed from my own earlier
opinions. But Cuvier's system of classification has taken such
deep root, and is so widely used, that I feel bound, in conclu-
sion, to point out how for the present attempt to express in a
condensed form the general results of comparative anatomy
departs from that embodied in the opening pages of the " Kegne
Animal."

The departure is very nearly in the ratio of the progress of
knowledge since Cuvier's time. The limits of the highest

86 ON CLASSIFICATION.

group, and of the more highly organized classes of the lower
divisions, with which he was so well acquainted, remain as he
left them ; while the lower groups, of which he knew least, and
which he threw into one great heterogenous assemblage, — the
Radiata, — have been altogether remodelled and rearranged.
Milne-Edwards demonstrated the necessity of removing the
Polyzoa from the radiate mob, and associating them with the
lower Mollusks. Frey and Leuckart demonstrated the sub-
regnal distinctness of the Coelenterata. Yon Siebold and his
school separated the Protozoa, and others have completed the
work of disintegration by erecting the Scolecida into a primary
division, of Vermes, and making the Echinodermata into an-
other. Whatever form the classification of the Animal King-
dom may eventually take, the Cmrierian Radiata is, in my
judgment, effectually abolished : but the term is still so fre-
quently used, that I have marked out those classes of which it
consisted in the diagram of the Animal Kingdom (p. 6), so
that you may not be at a loss to understand the sense in which
it is employed.

LECTURE VI.

ON THE CLASSIFICATION OF ANIMALS.

THE SUBDIVISIONS OF THE MAMMALIA LARGER THAN

ORDERS.

In my last lecture I endeavoured to point out the grounds upon
which naturalists have arrived at the conclusion that the classes
of the Animal Kingdom may be arranged together in larger
groups or divisions, such as have been termed "provinces " and
" sub-kingdoms." If the time at my disposal for the considera-
tion of Classification permitted me to do so, I should now, in
the logical order of my discourse, take the opposite course ; and
turning again to the list of classes, I should endeavour to indi-
cate in what manner they must be subdivided into sub-classes,
orders, and lesser divisions. But it is needless to sav that such
a task as this would require many lectures, while I have only
one to dispose of; and I propose to devote that one to a con-
sideration of the classification of that class, which is in many
respects the most interesting and the most important of any in
the Animal Kingdom, — the class Mammalia.

A great many systems of classification of the Mammalia have
been proposed, but, as any one may imagine from the nature of
the case, only those which have been published within the last
forty or fifty years, or since our knowledge of the anatomy of
these animals has approached completeness, have now any
scientific standing-ground. I do not propose to go into the
history of those older systems, which laboured more or less
under the disqualification of being based upon imperfect know-

SS ON CLASSIFICATION.

ledge, but I shall direct your attention at once to that important
step towards dividing the Mammalia into large groups, which
was taken by the eminent French anatomist, M. de Blainville,
so far back as the year 1816. M. de Blainville pointed out that
the Mammalia might be divided into three primary groups,
according to the character of their reproductive organs, espe-
cially the reproductive organs of the female. He divided them
into " Ornithodelphes," " Didelphes," " Monodelphes ;" or, as we
might term them, Ornithodelphia, Didelphia, Monodel-
phia. Xow, 1 do not mean to assert that M. de Blainville
defined these different groups in a manner altogether satis-
factory, or strictly in accordance with all the subsequently dis-
covered facts of science, but his great knowledge and acute
intuition led him to perceive that the groups thus named were
truly natural divisions of the Mammalia. And the enlargement
of our knowledge by subsequent investigation seems to me, in
the main, only to have confirmed De Blainville's views.

The division of the Ornithodelphia comprises those two
remarkable genera of Mammals, as isolated in geographical
distribution as in structure, — Ornithorhynclius and Echidna, —
which constitute the order Monotremata.

In these animals the angle of the lower jaw is not inflected,
and the jaws are devoid of true teeth, one of the two genera only
(Ornithorhynclius) possessing horny plates in the place of teeth.
The coracoid bone extends from the scapula to the sternum,
with which it is articulated, as in birds and most reptiles, and,
as in many of the latter, there is an episternal bone. There is
no marsupial pouch, though bones wrongly termed "marsupial"
are connected with the pelvis. But it is to the structure of the
female reproductive organs that the Ornithodelphia owe their
name. The oviducts, enlarged below into uterine pouches, but
opening separately from one another, as in oviparous vertebrates,
debouch, not into a distinct vagina, but into a cloacal chamber,
common to the urinary and genital products and to the faeces.
The testes of the male are abdominal in position throughout
life, and the vasa deferentia open into the cloaca, and not into a
distinct urethral passage. The penis is indeed traversed by an

THE ORNITHODELPHIA AND DIDELPHIA. 89

urethral canal, but it is open and interrupted at the root of that
organ. Tn both sexes, the ureters pour the renal secretion, not
into the bladder, which is connected with the upper extremity
of the cloaca, but into the latter cavity itself.

In the brain, the corpus eallosum is inconspicuous, though
the question how far it can properly be said to be absent
requires much more thorough investigation than it has yet
received.* We are but very imperfectly acquainted with the
reproductive processes of these animals, but it is asserted that
the young are devoid of a placenta. The mammary gland has
no nipple.

Like the Ornithodelpliia, the division Didelphia contains
but a single order, the Marsupialia, the great majority of which,
like the Oriiitlwdelphia, inhabit Australia. They almost all
have the angle of the lower jaw inflected, and all possess true
teeth. The coracoid is, as in the higher Mammals, anchylosed
with the scapula, and is not articulated with the sternum. All
have the so-called " marsupial ' bones or cartilages — ossifica-
tions, or chondrifications, of the internal tendon of the external
oblique muscle of the abdomen — and the females of almost all
possess a fold of the skin of the abdomen above the pubis, con-
stituting a " marsupium" or pouch, within which the young are
nourished and protected in their early, helpless condition.

The oviducts open into vaginse, which are more or less com-
pletely divided into two separate passages. The testes of the

* For a number of years I have entertained the gravest doubts respecting the
accuracy of the doctrine put forth now nearly thirty years ago by Professor Owen,
and almost universally received, that the corpus eallosum is absent in Monotremes
and Marsupials, and at one time I began to collect materials for the thorough in-
vestigation of the question ; but other occupations intervened, and the plan was
never carried out. Nevertheless, I have always expressed myself cautiously 011
this subject, and, as the text shows, I was particularly guarded when delivering the
present lecture. At that time, in fact, I was well aware that my friend Mr. Flower
had commenced a series of inquiries into the question, and such results as he had
then obtained tended greatly to the increase of my scepticism. Mr. Flower has
since been good enough to go carefully with me over the large series of drawings
and preparations which he has made ; and I am prepared 1o express my entire
concurrence in his conclusion that the corpus eallosum exists, distinctly developed,
though not so well as in monodelphous, or placental, Mammals, in both the Bidvl-
phia and the Ornithodelphia.

90 ON CLASSIFICATION.

males are lodged in a scrotum, which is suspended in front of
the penis ; and the vasa deferentia open into a complete and
continuous urethra, which is also the passage by which the urine
escapes from the bladder, and is perfectly distinct from the pas-
sage for the faeces, though the anus and the termination of the
urethro-sexual canal are embraced by the same sphincter.

The corpus callosum is comparatively small, as in the Orni-
thodelpliia.

It is stated that the allantois of the embryo is arrested in its
development, and gives rise to no placenta. The umbilical sac
is said to acquire a large proportional size ; but whether it plays
the part of a placenta for the short period of intra-uterine life,
or not, is unknown.

The young are born of very small size, and in a singularly
imperfect condition ; but being transferred to the marsupium,
and becoming attached to a long nipple, they are supplied with
milk until they are able to provide for themselves — the milk
being, at first, forced into their mouths by the action of a muscle
spread over the mammary gland.

In the Monodelphia, the angle of the lower jaw is not
inflected, and they may or may not be provided with teeth.
They never possess " marsupial " bones. The uterine dilatation
of the oviducts is always considerable, and whether they have
common or distinct apertures, the vagina is a single tube,
though it may be partially divided by a septum. The testes
may vary much in position ; but, if they are lodged in a scrotal
pouch, it is never pendulous by a narrow neck in front of the
penis, as in the Didelphia.

The urinary bladder opens into a distinct urethra, which,
directly or indirectly, receives the vasa deferentia in the male.

The corpus callosum is very variable in its development,
commonly attaining a much larger size than in the preceding
groups ; the optic lobes are divided into four portions.

The young are nourished within the uterus until such time
as they are competent to suck milk from the teats of the parent,
to which end the chorion always develops processes or villi,
which are well supplied with vessels brought to them by the
allantois. These processes becoming interlaced more or less

THE rLACENTALIA A XI) 1MPLACH\TTALIA.

closely with corresponding vascular developments of the wall of

the uterus (and so forming a " placenta "), an interchange of con-
stituents takes place between the foetal and the maternal blood,
through the separating walls of the fcetal and maternal vessels.
In this manner, throughout its prolonged intra-uterine life, the
Monodelphian foetus is supplied with nourishment and gets rid
of its effete products.

As the three groups instituted by De Blainville are capable
of being thus clearly differentiated one from the other, the
distinctions between them having been only more and more
clearly brought out by the subsequent progress of knowledge, I
can see no ground for refusing to adopt his classification, or for
denying him that credit to which he is fairly entitled for appre-
hending these distinctions. Certainly, the later proposition, to
divide Mammals into two great groups only, Placentcdia and
Implacentalia, cannot be regarded as any improvement upon
De Blainville's system, as it ignores the important fact that the

Fig. 39.

Fig.

39. — Diagrammatic section of a human pregnant uterus, with the contained ovum
(Longet). n, uterus; I, oviduct; c, cervix uteri; du, decidua uteri; dr, decidua
reflexa; ds, decidua serotina ; ch, chorion ; am, amnion ; ai, allantois ; nb, umbilical
vesicle; z, villi which form the foetal part of the placenta; z' , villi over the rest of
the chorion, which take no part in the placental function in man.

92 ON CLASSIFICATION.

two divisions of the Implacentalia are separated by characters
of fully as great importance as those which distinguish the
Placentalia and Implacentalia.

But whether the Omithodeljphia and the BidelpJiia are re-
garded (as I believe they ought to be) as two of the three
primary separate " sub-classes " of the class Mammalia, or
whether they are looked upon only as subdivisions of the Im-
placentalia, there is no doubt that they are, and will remain, dis-
tinct natural assemblages, the subdivisions of which present no
very great difficulties.

It is otherwise with the sub-class Monoclelpliia— which con-
tains at least a dozen orders, the arrangement of which into
groups, not only in detail, but in principle, is, and long has been,
a subject of much difficulty, and consequently of controversy.

Sir Everard Home* is commonly quoted as the originator
of one of the two systems of classification in vogue at the
present day ; but his vague statements and confused notions
respecting the varying characters of the placenta of the
Monodeljihia hardly entitle him to that honour, which, in my
opinion, belongs rather to that eminent man, Karl Ernst von
Baer, of whom it can be truly said that he has touched no sub-
ject without throwing a flood of light upon it. Towards the
end of his famous essay, " Untersuchungen iiber die Gefas-
verbindung zwischen Mutter und Frucht," published in 1828,
the following passage occurs : —

" In the first place, I have taken pains to show that the ova
of mammals are only variations of a single type ; and if we
except the ova of the Marsupials, concerning which I can
form no judgment, all consist of the same parts ; all have a
placenta; and, in all, some portion of the chorion is smooth.
The foetal placenta consists everywhere of the same elements,
but offers the most remarkable differences in its external dis-
position. It is either —

1. Merely applied to the maternal placenta, and

(a) continuous and zone-like. First form.

(b) divided into many parts. Second form.

* " Comparative Anatomy," vol. iii.

VON BAER AND BSCHRICHT. 93

Or 2. It and the maternal placenta grow together, and they
Lie,

(a) in a zone round the egg. Third form.

(b) at one end of it. Fourth form.

These differences, however, are developed gradually, and, at
first, are less marked."

The first form, described in the text of the work, is that
met with in the pig. It is what is now commonly termed a
diffused placenta; but Yon Baer, more accurate than most of
his successors, indicates the confinement of the placental villi
to the middle of the chorion — its prolonged poles remaining
bare — by the term " giirtelformig," zone-like. The second form
is that exemplified by the cow and sheep, the cotyledonary
placenta. The third is the carnivorous placenta, termed
zonular. The fourth is the placenta of man, called now-a-days
discoidal.

The most important circumstance pointed out by Von Baer,
however, is one which has been greatly overlooked, if not
wholly ignored, in subsequent discussions — the fact that the
differences in the form of mammalian placentae are subsidiary if
compared with their differences in structure, more particularly
in regard to the extent to which a maternal element enters into
their composition.

Eschricht, in the admirable memoir, " De Organis quae
Kespirationi et Nutritioni Foetus Mammalium inserviunt," which
he published in 1837, repeats the ideas of Von Baer, apparently
without being aware of the fact, and enlarges upon them as
follows (p. 30) : —

" Restat, ut succinctam expositionem Mammalium afferamus
secundum varias quae in iis observantur, placentae formas.

" A ceteris omnibus mammalibus Marsupialia et Monotre-
mata separanda sunt, quibus nulla est placenta. Caetera omnia
in duas familias dividencla, quarum alteri placenta uterina
caduca, alteri non caduca est. Huic Mammalia primata et
ungulata omnia adnumeranda sunt, inter quae Ruminantia ob
singularem cotyledorum formam caeteris opponi possunt.

" In mammalibus placentam uterinam caducam habentibus

94 ON CLASSIFICATION.

tres mihi occurrere viclentur placentae typi, quorum primus
gliribus, secundus feris, tertios simiis et homini proprius est,"

In this passage Mammals are clearly divided, in the first
place, into placental and im placental ; and the former are then
subdivided into those which have a non-caducous and those
which have a caducous uterine placenta. The Cetacea and
Ungulate Mammals constitute the former group ; the Rodents,
Carnivores, Apes, and Men the latter.

In 1813, an accomplished English zoologist, Mr. Water-
house, published a highly instructive paper on the " Classifi-
cation of the Mammalia,"* in which the following passage
occurs : —

"Taking the general form of the Brain into consideration,
the placental Mammalia would appear divisible into two
sections : first, those in which the cerebrum is generally of a
rounded form, obtuse in front and provided with distinct con-
volutions ; and secondly, those in which the cerebrum is desti-
tute of convolutions, or nearly so, and usually contracted
in front. The first division would contain the Quadrumana,
Camivora, Cetacea, Pacliydermata, and Piuminantia, and the
second would contain the Cheiroptera, Insectivora, Edentata,
and Podentia"

But although Mr. "Waterhouse puts forward thus clearly
the facts upon which a cerebral classification of the Mammalia
might be based, he immediately afterwards, with his customary
judgment, expresses great doubt as to the value of any such
classification.

" But are we in a condition to take for a basis of classifica-
tion of the Mammalia the structure of the brain ? I think
not, though, in the case of the Marsupialia, it has afforded
characters serving to separate that from other sections, and to
indicate its proper position in the system. I am not prepared
to follow those naturalists who would, in the present state of
information, take this organ as one of primary importance in
the distribution of the orders of the placental series of Mammals.
I cannot adopt the two great sections of this series as apparently

* "Annals and Magazine of Natural Historv," 1843, vol. xii. ]>. 399.

WATERHOUSE AND MILNE-EDWARDS. 95

indicated by the smooth and anteriorly contracted cerebrum on
the one hand, and the convoluted cerebrum, with its rounded
anterior portion, on the other. Were I to do so, I should find it
necessary to remove some of the Lemurs from their group in the
highest order of the first section, and to place them in the
second section."

In the succeeding year, 1844:, M. Milne-Edwards, one of the
most distinguished physiologists and zoologists of modern
France, proposed, in a highly philosophical paper upon
zoological classification in general,* a method of subdividing
the Mammalia, essentially similar to that put forward incident-
ally by Yon Baer and Eschricht, but lacking, as I conceive,
what is the great merit of the latter writers, namely, the clear
perception of the classificatory value of the intimate structure of
the placenta and the entrance, or not, of a decidual uterine
element into its composition. M. Milne-Edwards dwells with
great force (as Mr. Waterhouse had previously done) upon the
closeness of the general structural affinities which unite the
Rodentia, Insectivora, Cheirojrfera, Quadrumana, and Bimana
of Cuvier together, and shows that these affinities are denoted
by the discoid placenta which they possess in common.

The diffused placenta (under which head the cotyledonary
placenta is included) is stated to be the characteristic of the
Ruminantia, Paehi/dermata, Edentata, and Cetacea ; while,
lastly, the " Camivora and seals (Amphibies) are distinguished
from all the rest by their zonular placenta."

The singular genus Hyrax, which Cuvier endeavoured to
prove to be a true Pachyderm, is considered by M. Milne-
Edwards to form one of the series of Mammals with a zonular
placenta ; and to represent, in that series, the Pachyderms in the
series with diffuse placentation, and the Rodents, in the series
with discoid al placentation.

M. Gervais, in France, and M. Vogt, in Germany, have
adopted the placental classification of Milne-Edwards ; while,
in 1857, Mr. Waterhouse's proposed, but immediately rejected,
cerebral classification was substantially revived by Professor

* " Annales des Sciences naturelleSi" Seine 3. Tome 1. " Considerations
sur quelques Principes relatifa a la Classification naturelle des Animaux."

96 ON CLASSIFICATION.

Owen, in his paper " On the Characters, Principles of Division,
and Primary Groups of the class Mammalia," published in the
Journal of the Linnsean Society ; though it should be added
that Professor Owen made certain additions to the nucleus
furnished by Mr. Waterhouse, which are unquestionably
original.

Thus the " Lissencephala " of Professor Owen is simply a new
name for the group of Mammals (" in which the cerebrum is
destitute of convolutions, or nearly so") indicated by Mr.
AVaterhouse ; and " Gyrencejihala" is a like verbal equivalent
for Mr. Waterhouse's group of 3 [animals characterised by
having the brain provided with distinct convolutions. But
Mr. Waterhouse does not mention Man at all, while Professor
Owen creates a new sub-class, Arehencephala, for the genus
Homo, and substitutes the name " Lyencephala " for Implacentalia,
formerly applied to the Omithodelpliia and Biclelpliia.

In attempting to decide between the various classifications
thus presented to us, the canons by which our judgment must
be guided are simple enough. It is obvious, in the first place,
that the definition of a group, whether that definition be based
on cerebral or on placental characters, must be true, as a matter
of fact, if anv value is to be attached to the classification of
which that definition forms a part.

And, in the second place, it is clear that the definition of each
group must be distinctive, that is to say, it must not include the
members of other groups.

Applying the second canon to the classification last mentioned,
it appears to me to collapse at once.

The sub-class Lissencejihala, for example, is thus defined : —

"The corpus callosum is present, but connects cerebral hemi-
spheres as little advanced in bulk or outward character as in the
preceding sub-class ; the cerebrum leaving both the olfactory
lobes and cerebellum exposed, and being commonly smooth, or
with few and simple convolutions in a very small proportion,
composed of the largest members of the group. The Mammals
so characterised constitute the sub-class IAssencephala" —
L. c, p. 14.

THE ARCHENOEPHALA. 97

On the other hand, the sub-class Gyrencephala receives the
following definition : —

" The third leading modification of the Mammalian cerebrum
is such an increase in its relative size, that it extends over more
or less of the cerebellum, and generally more or less over the
olfactory lobes. - Save in very few exceptional cases of the
smaller and inferior forms of the Quadrumana, the superficies
are folded into more or less numerous gyri, or convolutions,
whence the name Gyrencephala, which I propose for the third
sub-class of Mammalia" — L. c, p. 18.

I am quite unable to see what these so-called definitions de-
fine. If, for example, we place the brains of an Ant-eater, or of
a Capybara, side by side with that of a Genett — the two former
being Lissencephala, the latter one of the Gyrencephala — either
" definition ' will apply equally well to either of the three
brains. All three have slightly convoluted brains ; in all three
the olfactory lobes and cerebellum are more or less uncovered ;
and nothing in the definitions of the sub-classes of this " cerebral
classification " would enable an anatomist to say that any one of
these three brains belonged to one sub-class rather than another.

Since Mr. Waterhouse pointed out the fact, no one has
doubted that, as a general rule, the brains of the so-called " Gyren-
cephala" are more convoluted, size for size, than those of the
" Lissencephala ;" and the relations of the size and the zoological
position of an animal to the characters of its cerebral surface
have long since been well discussed by Gratiolet, Dareste, and
others. But it is exactly because the rule is only a general
one, and has many exceptions, that the degree of cerebral convo-
lution must be rejected as the basis of the definition of any
large group of Mammals.

Thus far, we meet, in Professor Owen's definitions, with a
certain foundation in fact, though it may not be such as is fitted
to afford ground for classification : but the group " Archencephala"
is in a more unfortunate position. Our first canon comes into
operation, and we must reject it, because the statements respect-
ing matters of fact in its definition are untrue. The words stand

thus : —

98 ON CLASSIFICATION.

" In man the brain presents an ascensive step in develop-
ment, higher and more strongly marked than that by which the
preceding snb-class was distinguished from the one below it.
Not only do the cerebral hemispheres overlap the olfactory lobes
and cerebellum, but they extend in advance of the one and fur-
ther back than the other. Their posterior development is so
marked, that anatomists have assigned to that part the character
of a third lobe ; it is peculiar to the genus Homo, and equally
peculiar is the ' posterior horn of the lateral ventricle/ and the
' hippocampus minor,' which characterises the hind lobe of each
hemisphere." — L. c, pp. 19, 20.

These are the assertions which have been repeated over and
over again during the last few years ; but, thanks to the exer-
tions of the able Conservator of your Museum, it is in my
power to lay before you visible and tangible facts, which prove
these assertions to be wholly devoid of foundation.

The third lobe, characterised bv extending further back
than the cerebellum, is said to be "peculiar to the genus
Homo."

I place before yon casts of the cranial cavity, accurately re-
presenting the relative positions of the parts of the brain of a
Gorilla, of a Chimpanzee, of an Orang, of a Cynocephalus ; and
you observe that the posterior, or third lobe, of each projects
further back than the cerebellum, in just the same sense as a
man's can be said to do so ; and in some cases, as in the baboon,
to a much greater extent.

The assertion that the third lobe, as defined by Professor
Owen, is " peculiar to man," is therefore demonstrably contrary
to fact.

" Equally peculiar is the posterior horn of the lateral ventricle."

Side by side upon the table are two dissections, made in the
same way, the one of the brain of an Orang-utan, the other that
of a man, taken at hazard by Mr. Flower, who has been good
enough to dissect both (Fig. 40).

Every one in this theatre, I imagine, can see perfectly well
that the Orang has a posterior cornu, which, in proportion to the
size of its brain, is just as long and nearly as much incurved

THE THIRD LOBE AND POSTERIOR CORNU.

as that of the man, while it is a good deal wider at its com-
mencement.

Fig. 4«>.

Fig. 40.— Figures [reduced to the same scale] of the dissected brains of a Man and of an
Orang which were exhibited in the theatre of the Royal College of Surgeons. — C1,
anterior cornu ; C2, descending cornu ; C3, posterior cornu ; H, hippocampus major ;
h, hippocampus minor.

H 2

100 ON CLASSIFICATION.

Ill fact, even if the posterior cornu had not been demon-
strated (as it has now been) in the brain of numerous genera of
Apes, this one example would sufficiently demonstrate the asser-
tion, that the posterior cornu is " peculiar to the genus Homo"
to be simply untrue.

Lastly, as regards the hippocampus minor — which is also said
to be " peculiar to man " — that structure is, as you perceive, as
distinct in the Orang's as in the man's brain, so that the third
term of the definition of the " Archencejyhala " is as contradic-
tory to plain fact as the other two (Fig. 40).

Even were the posterior lobe, the posterior cornu, and the
hippocampus minor peculiar to man, as supposed by the definer
of the sub-class " Archencephala" instead of being, as they
really are, structures far better developed in some of the lower
apes than in him, their classificatory value would be extremely
doubtful, seeing that they are among the most variable of struc-
tures in the human brain. The casts upon the table of a Tartar's
and of an Australian brain-case will demonstrate to you how in-
significant may be the projection of the posterior lobe in one man
and how great it may be in another. While the practical anato-
mists and demonstrators whom I address will be familiar with the
singular variability of the posterior cornu and the hippocampus
minor — structures which, without any assignable cause, or
noticeable modification of the structure, or of the functions, of
the brain, may present every degree of development, from
absence to great size.

So little, indeed, is any zoological value to be attached to
such a character as the degree of projection of the posterior
lobe, that closely allied apes present us with most singular dif-
ferences in this respect. Thus the group of South American
monkeys which comprises the Squirrel monkey (Chrysothrix),
the posterior lobes of whose brain project beyond the cerebellum
far more than they do in man, contains also the Howling monkev
(Mycetes), in which the posterior lobes cannot be said to project
at all. And within the last two days, Mr. Flower has discovered
(and the cast upon the table enables me to demonstrate the fact
to you) that in, at any rate, one species of Gibbon, the Siamang
(Hylobates Syndactylies) the cerebellum projects behind the pos-

THE HUMAN PLACENTA. 101

terior lobes, while, in the three other genera of anthropoid apes,
the posterior lobes of the cerebrum project behind the cere-
bellum.*

The latest form of the " cerebral " classification of the Mam-
malia having thus been shown to be devoid of any sound
foundation, I proceed to inquire whether the " placental'
classification does, or does not, stand upon a more secure basis,
if we take, not merely, with Milne-Edwards, the form of the
placenta, but with Von Baer and Eschricht, its structure, into
account. It is a well-established fact that two very distinct
types of placenta are to be met with in the Monodetyhia, and
that, at the present moment, we have no knowledge of any
transitional forms between these two types. The first of these
types is that exhibited by the human placenta, the second by
that of the pig or horse.

From the commencement of gestation, the superficial sub-
stance of the mucous membrane of the human uterus undergoes
a rapid growth and textural modification, becoming converted
into the so-called " decidua" While the ovum is yet small,
this decidua is separable into three portions, — the decidua vera,
which lines the general cavity of the uterus ; the decidua
reflexa, which immediately invests the ovum ; and the decidua
serotina, a layer of especial thickness, developed in contiguity
with those chorionic villi which persist and become converted
into the foetal placenta. The decidua reflexa may be regarded
as an outgrowth of the decidua vera ; the decidua serotina as
a special development of a part of the decidua vera. At first,
the villi of the chorion are loosely implanted into correspond-
ing depressions of the decidua ; but, eventually, the chorionic
part of the placenta becomes closely united with, and bound to,
the uterine decidua, so that the foetal and maternal structures
form one inseparable mass.

In the meanwhile, the deeper substance of the uterine

* See Mr. Flower's paper " On the Brain of the Siamang," Natural History
Review, April, 1863. " This peculiarity of the Siamang's brain is due to two
causes — firstly, the large development of the cerebellum ; secondly, and I shall
afterwards show, mainly, to the actual shortness of the posterior or occipital lobe
of the cerebrum." — L. c, p. 282.

102

ON CLASSIFICATION.

mucous membrane, in the region of the placenta, is traversed
by numerous arterial and venous trunks, which carry the blood
to and from the Dlacenta ; and the layer of decidua into which
the chorionic villi do not penetrate acquires a cavernous, or
cellular, structure from becoming burrowed, as it were, by the
innumerable sinuses into which these arterial and venous trunks

Fig. 41

Fig. 41. — Section of the Human Uterus and Placenta at the thirtieth week of pregnancy.
(After Ecker.) — A, umbilical cord ; B, chorion ; C, the foetal villi separated by pro-
cesses of; D, cavernous decidua; E, F, G, wall of the uterus.

open. In the process of parturition, the decidua serotina splits
through this cellular layer, and the superficial part of it comes
away with the umbilical cord, together with the fcetal mem-
branes and the rest of the decidua ; while the deeper layer,
undergoing fatty degeneration and resolution, is more or less
completely brought away with the locliia, and gives place to
a new mucous membrane, which is developed throughout the
rest of the uterus, during pregnancy; but, possibly, arises only
after delivery over the placental area.

In the Pig the placenta is an infinitely simpler structure.

DECIDUATE AND NON-DECIDUATE PLACENTA. 103

No " decidua " is developed ; the elevations and depressions of
the nn impregnated uterus simply acquire a greater size and
vascularity during pregnancy, and cohere closely with the
chorionic villi, which do not become restricted to one spot, but
are developed from all parts of the chorion, except its poles,
and remain persistent in the broad zone thus formed through-
out foetal life. The cohesion of the foetal and maternal
placenta?, however, is overcome by slight maceration or post-
mortem change ; and, at parturition, the foetal villi are simply
drawn out, like fingers from a glove, no vascular substance
of the mother being thrown off.

The process by which the mucous membrane of the uterus
returns to its unimpregnated condition after parturition in the
pig has not been traced.

The extreme cases of placentation exhibited by man and
by the Pig may be termed, with Yon Baer and Eschricht.
from the character of the maternal placenta, " caducous " and
" non-caducous," or, from the degree of cohesion of the two
placentae in parturition, " coherent ,! and "incoherent;" or,
what perhaps would be better still, the two Mammals may
be spoken of as "deciduate" and " non-cleciduate."* But,
whatever terms be employed, the question for the classifier
is to inquire what mammals correspond with Man and what
with the Pig, and whether the groups of deciduate and non-
deciduate Monodelpliia thus formed, are natural groups, or,
in other words, contain such orders as can be shown, on other
grounds, to be affined.

With respect to the deciduate Monodelphia, it is certain that
the apes agree, in the main, with man in placental, as in other
important characters ; and, so far as has hitherto been observed
(though our knowledge of the placentation of the Lemurs is
very defective), their placenta? differ from those of Man only in
presenting a more marked lobation — a character which occurs
as a variety in Man.

* It is, of course, by no means intended to suggest by these terms, that the
homologue of the decidua does not exist in the " non-deciduate " Mammals. The
mucous membrane of the uterus becomes hypertrophied during pregnancy in both
the deciduate and the non-deciduate Mammals ; but it is thrown off, and so gives
rise to a " decidua " only in the one of these two groups.

104

ON CLASSIFICATION.

The Cheiroptera, Insectivora, and Kodentia agree with Man
in possessing a placenta which is not only as much " discoidal,"
allowance being made for the shorter curve of the uterine walls,
as his, but also entirely resembles his in being developed in
conjunction with a decidua. This decidua always corresponds
to at least the decidua serotina of Man ; frequently there is a
well-developed decidua reflexa* How far a decidua vera can
be said to be developed is doubtful.

I am well aware that these statements are in direct opposi-
tion to some that have been very confidently put forward.
Thus, Professor Owen, in arguing against the views enter-
tained by Milne-Edwards and Gervais, makes the following
assertions : —

" The degree of resemblance in outward form between the
placenta of the Eat or Hare, on the one hand, and the Mycetes
and Macacus on the other, seems to me to be more than
counterbalanced by the difference of structure. The pedunculate
and cotyloid placenta of the Bat consists of foetal parts exclu-

Fiff. 42.

Fig. 42. — Magnified view of a section of the placenta and uterus of a pregnant Rat.

sively ; the maternal areolar portion is as distinct from it as it is
in the cotyledon of the Ruminant, and is a persistent structure of

* See upon this subject the recently-published valuable essay of Eeichert :
"Beitrage zur Entwickelungs-geschichte des Meerschweinchens." Reichert finds
a complete, or almost complete, decidua reflexa in Rats, Mice, Guinea-pigs, and
Bats; while in Rabbits, Hares, and Camivora, the decidua reflexa only partially
surrounds the ovum.

THE PLACENTA OF THE PAT. 105

the uterus. The discoid placenta of the monkey includes a large
proportion of maternal cellular structure, which comes away with
the foetal portion. The difference in the organic interblending
of the circulatory organs of mother and offspring, between the
Rodent ia and Quadrumana, is of much more real importance
than the degree of superficial similarity." — L. c, p. 16, note.

Led by the extraordinary contradictions of some of the
best-known facts of embryology* contained in the passage

* Eschricht described the placenta of the Rat with great precision, as the follow-
ing extract will show, six-and-twenty years ago. Is it possible that a hasty perusal
of a passage which I have put into italics below should have misled any of his
successors into supposing that Rodents have persistent cotyledons like those of
Ruminants?— " Placenta fere circularis erat ; diametrus longior sex lineas cum
dimidia explebat, brevior sex lineas. Superficies ovo obversa nonnihil concava
erat, externa autem sat convexa, ita ut a margine ad centrum sensim in tumulum
surgeret et crassitudinem placentas ad duas lineas augeret. In superficie ejus
concava, ovo obversa, tres distingui potuerunt regiones ; circum centrum, a vasis
umbilicalibus perforatum, laminula ilia pertenui obtecta erat, quam vasa umbili-
calia hie circumdare jam observatum est. Circa peripheriam scabrosa apparuit,
quasi fila seu vasa abrupta fuissent, in media regione autem inter periphericam et
ceutralem levis erat tunica sat distincta vestita. Superficies placentas externa con-
vexa duas regiones obtulit. Peripheric propior insequalis ad axem longitudinalem
ovi latior erat, uuam et quartam lineas partem explens ; ad latera angustior
quartam modo lineas partem explens. Sic media et lasvis hujus superficiei regio
ovoidea quidem erat ut tota placenta, sed cum axe majore transverse ei insidens.
In cumulo hujus superficiei permagnum vas ab utero centrum placentas perforavit.
Haud procul a centro quinque vasa minora intervallis sat asqualibus uterum cum
placenta jungebant. Superficiei uterinas proxima placentas pars in lamellas facile
dividebatur, profundior autem pars eundem ilium laminarum innumerabilium
contextual exhibuit quem in placenta felina fusius supra descripsi. Laminulas a
centro ad peripheriam sat regulariter ordinatas esse observare mihi visus sum.
Inter laminulas illas eandem esse alternationem laminularum foetalium et uterina-
rum vix dubitare potui, prassertim quum non modo vasa umbilicalia sed ab altera
parte sex vasa majora ab utero placentam ingredi vidissem. At uterum examinans
nova orta est dubitatio. Uteri cornua antequam ova excisa fuissent, moniliformia
apparuerunt, et inter singula ova fortiter constricta. Ita non modo numerus sed
etiam forma ovorum extus apparuit quid quod etiam placentas facillime obser-
vari videbantur. Intumescentias, ova includentes, ipsas quidem oviformes erant,
sed ad marginem superiorem ad insertionem ligamentorum latorum et ante
eadem singulas tumuli speciem prasbebant, placentam aperte indicantis. Itaque
placentas in utero muris ratti non modo quoad fceturn, sed etiam quoad uterum
certum occupat locum, et embryones omnes in una serie in oppositis uteri
cornibus symmetrice collocati sunt, utrum autem capitibus prasviis abdomini-
busque abdomen matris respicientibus, an clunibus prasviis dorsisque abdomini
matris obversis inquirere neglexi. Verum tumuli placentas indicantcs non ipsas
fuerunt placentas. Ovis excisis iidem tumuli in utero remanebant ut tubercula
fusca tres lineas crassa, diametrum quinque linearum exhibentia. Quod quum

100 ON CLASSIFICATION.

I have italicised to look into the matter afresh, I have found that
the assertions made therein respecting the placenta of the Eat
are as completely contrary to fact as are those respecting the brain
of the apes which I have already cited from the same author.

Figure 42 represents a section of the uterus, chorion, and
partially-injected placenta of a foetal Eat, one inch and a
quarter long, taken in a direction perpendicular to the long
axis of the uterine cornu. a is the mesometrium traversed by
a large uterine vein ; b is the wall of the uterus becoming
looser in texture and traversed by large venous channels in its
inner substance, c; d is a decidual layer of the uterus of a
cavernous structure, whence vascular processes are continued
towards the chorionic surface of the placenta. A large vein (i)
passes directly from the decidual layer (d), and the uterine
sinuses beneath it, to near the chorionic surface of the placenta,
beneath which it branches out horizontally. The chorion (/),
rendered vascular over its non-placental part by the ornpha-
lomeseraic vessels (&), only begins to exhibit villous processes
and folds at the point (g). These outermost villi appear to
me to be free; but, more internally, they become closely
connected with the upper surface of the placenta; and over
the central third of the foetal face of the placenta, the umbilical
vessels (I) ramify in a radiating fashion, and send prolongations

observassein, initio tota mea de structura placental lnammalium unguiculatoruin
sententia labefacta est. Gliribus si ut ruminantibics pars uterina placentas in partu
non abstruderetur, mirum sane fuisset, etsi theoria de placentas structura in uni versum
eo nihil caperet detrimenti. Quum vero superficiem uterinam placentas lasvem
observassem, et idem de superficie ipsius uteri ei obversa nunc observarem, tota
theoria pasne refelli mihi videbatur, secundum quam alternatio quasdam utriusque
systematis adsit neccssr est. Attamen mox intellexi corpus illud uterinum non
ipsam partem uterinam placentas esse. Supra jam de vase majore centrali et
quinque minoribus circumstantibus sermo erat, quas placentas superficiem uterinam
transibant. Eadem vasajam in ipsius uteri superficie observavam et in apertoerat
riiiuificationem eorum in laminubis ipsius placentas fieri. Corpus uterinum trans-
cissum cellulasplurimas exhibuit, ni tailor, sanguinis coagulati nonnihil etiamtunc
continentes. Sic constructum eidem functioni inservire mihi videtur ac similes
uteri humani cellulas vel sinus venosi ; nee placentam murinam aliter a placenta
iVlina discrepare video nisi quod vasa uterina pauciora sed sat magna partem
uterinam intrant. Quibus observationibus ductus etiam glirioi placentas ex

PASTE FCSTALI ET I'TERINA CADTJCA HAC ET ILLA LAMIXIEIS IXNTMEIUS ALTKKXAXTIBUS

< ii.mpositas esse ])utabo donee melioribus observationibus refutatus fuero."

ZONULAU, DISCOIDAL, AND DIFFUSE PLACENTiE. 107

down between the decidual lamellae. The slightest traction
exerted upon the cord causes the placenta to separate along the
line e, m, m, e, bringing with it, of course, the cup-shaped
deeidua, d*

It is obvious, from the above description, that the " pedun-
culate and cotyloid '' placenta of the Rat does not " consist
of foetal parts exclusively," but that, on the contrary, as
Eschricht has so well pointed out, " the organic interblending
of the circulatory organs of mother and offspring " is as com-
plete in the Rat as in Man ; and that, therefore, the concluding
paragraph of the citation from Professor Owen's paper ought to
be reversed.

The Camivora develop a well-marked deeidua, and their

* My friend Professor Eolleston has made the following statements in a paper
which will shortly appear iu the Zoological Society's Transactions.

1. The Rat's afterbirth consists of a saucer-shaped deciduous serotina, and a
button shaped placenta proper. Afterbirths made up of these two elements may
be found in the stomachs of animals of this species after parturition, as they, like
many other Mammals below the Simiadm, devour them. Under these circum-
stances, the two constituent factors of the afterbirth may either be found in their
normal connection, or they may be separated one from the other.

2. The non-deciduous part of the serotina forms in the Rat, after parturition, a
hernial protrusion into the mesometrium, which has been mistaken for a developing
ovum (see Hunterian Catalogue, Phys. Ser. Prep. 3466) ; just as the homologous
structures in the human subject form a hernial protrusion into the cavity of the
uterus, which may persist as a more or less elevated area for several years. (Cf.
Robin, Mem. Imp. Acad. Med., torn, xxv., p. 137.)

3. The homologue of the saucer-shaped deciduous serotina of the Rat is, in the
human subject, the thin layer of laminated albuminous tissue, which, in a placenta
• xpelled without suffering much violence, is seen clothing its uterine surface. It
is smaller, relatively, to the other structures concerned in the nutrition of the foetus,
in the human than in any other species. It is more easily demonstrable in the
Monkey (Macacus Nemestrinus, e. g.), as being a more coherent and stouter
membrane than in Man. It is, however, here still a condensed and mem-
branous structure, as compared with its pulpy homologues in Camivora, Insectivora,
and Rodents.

■i. In early periods of utero-gestation in the common Shrew and Hedgehog, the
deciduous serotina is a very much larger structure than the placenta proper,
which it entirely covers, except on the fcetal aspect. But in the Tenrec, near the
full time, the deeidua serotina is of but wafer-thickness.

.*). Dr. Matthews Duncan and M. Robin have shown that the muscular coat of
the uterus is never left denuded after parturition in the human subject. The
same remark holds good in the case of the " deciduate " Mammalia, in all of which
a more or less modified mucous tissue, the " non-deciduous serotina,'' is left, after
parturition, upon the utero-placental area, from which the deciduous serotina and
placenta proper have been separated as "afterbirth."

108

ON CLASSIFICATION.

placenta in all genera which have been examined (except the
Polecat, according to Von Baer) has the form of a complete zone,

Fig. 4:i

Fig. 43. — Foetal kitten, with its membranes and placenta. The latter is seen from within,
the chorion and allantois being open and everted. — Am, amnion ; All, allantois ; PI,
placenta ; Um, umbilical vesicle.

(From a preparation in the Museum of the Royal College of Surgeons.)

or broad girdle, surrounding the middle of the chorion and
leaving the poles bare (Fig. 43).

Thus Man ; the Apes, or so called Quadrumana ; the
hisectivora ; the Cheiroptera ; the Pioclentia, to which the lowest
apes present so many remarkable approximations ; and the
Carnivora (united into one group with the hisectivora by
Cuvier) are all as closely connected by their placental structure
as they are by their general affinities.

With the Pig, on the other hand, all the Artiodactyla, all the
P erissodactyla (save one, taking the group in its ordinarily
received sense) and all the Cetaeea which have been studied,
agree in developing no decidua, or, in other words, in the fact
that no vascular maternal parts are thrown off during parturi-
tion. But considerable differences are observed in the details

THE PLACENT.E OF RUMINANTS.

109

of the disposition of the foetal villi, and of the parts of the
uterus which receive them. Thus, in the Horse, Camel, and
Cetacea the villi are scattered, as in the Pig, and the placenta is
said to be diffuse ; while, in almost all true Ruminants, the foetal

Fig. 44.

Fig. 44. — Uterus of a Cow in the middle of pregnancy laid open. — V, vagina ; U, uterus ;
Ch, chorion ; C1, uterine cotyledons ; C'2, foetal cotyledons (after Colin).

Fig;. 45.

Ch A. Co

jj Coz Co

Fig. 45. — A. Horn of the Uterus of a pregnant Ewe, laid open to show, Ch, the chorion ;
with Co, the cotyledons.

B. Diagrammatic section of a Cotyledon. — U, uterine wall ; Co1, uterine cup of the coty-
ledon ; Co2, chorionic villous tuft of the cotyledon.

''From a preparation in the Museum of the Royal College of Surgeons.)

110

ON CLASSIFICATION.

villi are gathered into bunches, or cotyledons, which in the
Sheep (Fig. 45) are convex, and are received into cups of the
mucous membrane of the uterus ; while in the Cow. on the con-
trary, they are concave, and fit upon corresponding convexities
of the uterus (Figs. 44 and 46).

Fig. 46.

Fig. 46. — A foetal cotyledon, C2, half separated from the maternal cotyledon, C\ of a Cow.

C'h, chorion. U, uterus (after Colin).

No one, probably, would be inclined to object to the associa-
tion of the orders just mentioned into one great division of the
Monodeljjhia, characterised by its placental structure. But such
grouping leaves several important points for discussion. The
Elephant, as Professor Owen* has shown, has a zonary placenta,
and the genus Hyrax has been known since the time of Home
to be in like case. Hence, as the elephants are commonly sup-
posed to be closely allied with the Pachydermata, which possess
diffuse, non-deciduate placentae, and as Hyrax is now generally,
if not universally, admitted into the same order as the Horse,
which has a diffuse, non-deciduate placenta, it is argued that
placental characters do not indicate natural affinities. A ques-
tion, indeed, arises, which has not been answered by those who
have described the placentae of Elephas and Hyrax. Is the
placenta of these animals simply a zone-like arrangement of villi

* "Description of the Foetal Membranes and Placenta of the Elephant.'"
Philosophical Transactions, 1857.

THE PLACENT/E OF THE EDENTATA. J 1 I

or cotyledons, in connection with which no decidua is developed,
or is it a true deciduate placenta, resembling that of the Car-
nivora in the essentials of its internal structure as in its external
form ? Recent investigation lias convinced me, that, in both
these animals, the placenta is as truly deciduate as that of a
Rodent ; so that most unquestionably, if the placental method of
classification is to be adopted, both Elephas and Hyrax must go
into the same primary division of the Monodelphia as the
Rodentia and Carnivora,

But do these facts really present obstacles to the placental
system of classification ?

So far as the case of the Elephants is concerned, I must con-
fess that I see no difficulty in the way of an arrangement which
unites the Proboscidea more closely with the Rodentia than with
the Artiodactyla and Perissodactyla, the singular ties which unite
the Elephants with the Rodents having been a matter of common
remark since the clavs of Cuvier.

In the absence of any definite knowledge of the placental
structure of Rhinoceros and Tapir us* it would, perhaps, be pre-
mature to discuss the position of Hyrax, as determined by its
placenta ; but if it should eventually appear, as is very probable,
that Rhinoceros, like Tajpirus and Equus, has a diffuse, non-deci-
duate placenta, I should have no hesitation in regarding Hyrax
as the type of a distinct order of deciduate Monodelphous Mam-
malia, Hyrax, in fact, hangs by Rhinoceros mainly by the
pattern of its molar teeth, — a character which affords anything
but a safe guide to affinity in many cases, t

Concerning the placentation of the Sirenia we have no in-
formation.

Among the Edentata, the Sloths have presented a cotyledonary
placenta, and the Armadillos have been affirmed to possess a
discoidal one. I am not aware that the minute structure of the
placenta has been examined in either of these groups, but I am

* Home's description of the foetal membranes of the Tapir is very poor, but
Bauer's beautiful figures show clearly that the villi are diffuse, as in the Horse.

f See, in reference to this point, the late Professor A. Wagner's excellent remarks
on Cuvier's exaggeration of the Ehinocerotic affinities of Hyrax, in Schreber's
" Saugethiere." Supp. Band, Abtli. iv. p. 307.

112 ON CLASSIFICATION.

indebted to Dr. Sharpey for valuable information respecting the
placental structure of Manis. The surface of the chorion is
covered with fine reticulating ridges, interrupted here and there
by round bald spots, giving it an alveolar aspect, something like
the inside of the human gall-bladder, but finer. The inner sur-
face of the uterus exhibits fine low ridges or villi, not reticulating
quite so much. The chorion presents a band, free from villi,
running longitudinally along its concavity, and there is a corre-
sponding bald space on the surface of the uterus. The ridges of
the chorion start from the margins of the bald stripe, and run
round the ovum. The umbilical vesicle is fusiform. This is
clearly a non-deciduate placenta, and the cotyledonary form of
that of the Sloth leads me to entertain little doubt that it belongs
to the same category.

Admitting all these difficulties and gaps in our information, it
still appears to me that the features of the placenta afford by far
the best characters which have yet been proposed for classifying
the Monodelphous Mammalia, especially if the concomitant
modifications of the other foetal appendages, such as the allan-
tois and yelk-sac, be taken into account. And it must be
recollected that any difficulties offered by the placental method
attach with equal force to the systems of classification based upon
cerebral characters which have hitherto been propounded. If
any objections, on the ground of general affinities, are offered
to the association of Elejrfias, Hyrax, Felis, and Cercopithecus in
the same primary mammalian division of deciduate Monodelphia,
they are not removed by constructing that primary division upon
other principles, and calling it Gijrenceplxala.

118

LECTURE VII.

ON THE VERTEBRATE SKULL.

THE STRUCTURE OF THE HUMAN SKULL.

The human skull is by no means one of the simplest examples
of a vertebrate cranium which can be studied, nor is the com-
prehension of its structure easy ; but, as all vertebrate anatomy
has started from the investigation of human organization, and
the terms osteologists use are derived from those which were
originally applied to definite parts of the organism of man, a
careful investigation of the fundamental structure of man's
skull, becomes an indispensable preliminary to the establish-
ment of anything like a sound comparative nomenclature, or
general theory, of the Vertebrate Skull.

Viewed from without (Fig. 47), the human cranium exhibits
a multiplicity of bones, united together, partly by sutures,
partly by anchylosis, partly by moveable joints, and partly by
ligaments ; and the study of the boundaries and connections of
these bones, apart from any reference to the plan discoverable
in the whole construction, is the subject of the topographical
anatomist, to whom one constantly observed fact of structure is
as valuable as another. The morphologist, on the other hand,
without casting the slightest slur upon the valuable labours of
the topographer, endeavours to seek out those connections and
arrangements of the bony elements of the complex whole which
are fundamental, and underlie all the rest ; and which are to
the craniologist that which physical geography is to the student
of geographical science.

114

ON THE VERTEBRATE SKULL.

Perhaps no method of investigating the structure of the
skull conduces so much towards the attainment of a clear
understanding of this sort of architectural anatomy, as the
study of sections, made along planes which have a definite
relation to the principal axes of the skull.

Ficr. 47.

Fig. 47. — Diagrammatic side view of a Human Skull. — Fr. Frontal. Pa. Parietal. S.O.
Supra-occipital. S.O*. Squama occipitis above the torciilar Herophili ami lateral
sinuses. As. Alisphenoid. fity. Portio squamosa of the temporal bone. M. Mastoid
process and pars mastoidea. Ty. Tympanic. St. Styloid process. Nd. Nasal. /..
Lachrymal. Ju. Jugal, or Malar. Pmx. Premaxilla. Mas. Maxilla. Mn. Mandible.
////. ITvoid. in. Malleus, i. incus. [These letters will bear the same signification
throughout the series of figures of crania.]

THE STRUCTURE OF THE HUMAN SKULL.

I I f,

If a vortical and transverse section be taken through the
cranium, in such a manner that the plane of the section shall
traverse both external auditory meatuses, the skull will be
divided into two unequal portions — an anterior, larger, and a
posterior, smaller. The former, if viewed from behind, will
present the appearance represented in Fig. 48.

Fig. 48.

Fig. 48. — Anterior half of the skull ot a young person (six or seven years of age) trans-
versely bisected. The temporal bone (27) on each side is left in outline, and the
contour of the alisphenoid is supposed to be seen through it. — IT, optic foramina be-
tween the roots of the orbito-sphenoids ; V, foramen ovale for the third division of
the trigeminal; N indicates the nasal chamber; Mx is placed in the buccal chamber.

A stout median floor (B8) whence lateral continuations
(AS) are prolonged to meet an arched roof (Pa), divides a
capacious upper chamber, which, during life, lodged a part of
the brain, from a lower chamber, formed by the bones of the
face. This lower chamber itself is again separable into two

i 2

116

ON THE VERTEBRATE SKULL.

parts, — an upper, divided into two by a median septum, the
nasal passages ; and a lower, the oral cavity .

The posterior portion of the bisected skull (Fig. 49) presents,
in like manner, a strong floor (BO) and a large upper chamber
for the lodgment of parts of the brain ; but the lower chamber
seems at first to be absent in the skeleton, being represented,
in fact, only by the styloid processes (St), the so-called stylo-
hyoid ligaments, and the hyoidean bone (Hij) which is suspended
by these ligaments to the skull.

Fig. 49.

Fig. 49. — The posterior half of the transversely bisected skull, Fig. 48. — B.O., the basi-
occipital ; E.O., E.O., the ex-occipitals ; 7', the temporal l>one left in outline ; O.F.,
occipital foramen; VII., canal for the portio dura and poitio mollis; IX., foramen
for the ninth or hypoglossal nerve

A longitudinal and vertical section of the skull (Fig. 50)
enables us to observe the same relations of the parts from
another point of view. The central bones (BO, BS, PS, Eth.,
Vo), which lie between the arches of the brain-case above, and
the arches of the face below, are, in such a section, found 1<»

THE STHUOTUKE OF THE III MAN SKULL.

117

constitute a continuous series, from the occipital foramen to
the anterior extremity of the nasal passage, which, as it forms
the common centre or axis, not only for the bones of the brain-

Fig. 50.

iWA%]

Fisr. 50. — Longitudinal and veitical section of a Human Skull. — * The sella turcica. An.
The position of the superior and posterior vertical semicircular canals. /., II., V.,
VIII., IX. The exit of the olfactory, optic, third division of the fifth, eighth, and
ninth nerves. Vo., the Vomer.

case or cranium proper, but also for those of the face, may be
termed the Cranio-facial axis.

It will be useful to divide this axis into two portions, — a

118

ON THE VERTEBRATE SKULL.

posterior basi-cranial (BO, B8, PS), which forms the centre of
the floor of the proper cranial cavity; and an anterior, hasi-
facial (Eth., To.), which constitutes the axis of the front part of
the face.

Fig. 51.

Fig, 51. — Front view of the skull, the halves of which are shown in Figs. 48 and 49. —
N, nasal chamber ; Or, orbit. The nasal bones are removed, and so much of the
upper and lower jaws as is necessary to show the permanent teeth.

Three pairs of chambers, destined for the lodgment of the
organs of the higher senses, are placed symmetrically upon
each side of the double bony box thus described. Of these,
two pair are best seen in a front view of the skull (Fig. 51), the
inner pair being the olfactory, or nasal chambers (X), the outer
pair, the orbits (Or). The other pair are better displayed in
the transverse sections, Fig. 48 and Fig. 49, and are formed by
the temporal bones of anatomists (T, Tl), and especially by the
petrous and mastoid portions of those bones.

There is an obvious difference between the relations of these
3ensory chambers to the contained sensory organ, in two of

THE STRUCTURE OF THE HUMAN SKULL. 119

those chambers as compared with the third. The sensory
apparatuses of the nose and of the ear are firmly fixed to, or
within, the bony chambers in which they are lodged. That of
the eye, on the other hand, is freely moveable within the orbit.
An axis, upper and lower arches, chambers for the sensory
organs, — such are, speaking generally, the components of the
skull. The special study of these components may be best
commenced from the cranio-facial axis. Viewed either from
above (Fig. 52) or from below (Fig. 53), the cranio-facial axis is
seen to be depressed, or flattened from above down wards,
behind, and thick and nearly quadrate in the middle ; while,
in front, it is so much compressed, or flattened from side to
side, that it takes the shape of a thin vertical plate. In such a
young skull as that from which the Figures 52 and 53 are taken,
the depressed hindermost division of the axis is united with the
rest, and with the bones EO, EO, only by synchondroses ; and
is readily separable, in the dry skull, as a distinct bone, which
is termed the Basi-oecijntal (B 0). This basi-occipital furnishes
the front boundary of the occipital foramen ; and its postero-
lateral parts, where they abut against the bones EO, contribute,

\*y

Fig. 52. — Cranio-facial axis and lateral elements of the superior arches of a human skull
viewed from above. — a, the spheno-occipital synchondrosis ; 6, the ethmo-sphenoid
synchondrosis ; c, the tuberculum sella, indicating the line of demarcation between
the basi-sphenoid and the presphenoid ; d, the linjulce sphcnoidalcs.

120 ON THE VEKTEBRATE SKULL.

to a small extent, to the formation of the two occipital condyles.
In the adult skull the basi-occipital anchyloses completely with
the ex-occipital, on the one hand, and with the next bone of the
basi-cranial axis on the other, so that the saw must be called to
our aid in order to demonstrate the bone.

Fig. 5o. — Cranio-facial axis and lateral elements of the superior arches (as in Fig. 52), with
the pterygoid bones, and without the vomer, viewed from below. — c, jiuTction of the
basi-sphenoid and presphenoid with the internasal cartilage ; C.S., cornua sphenoidalia,
or bones of Berlin.

From the synchondrosis a to the point b, in even so young
a skull as that here represented, the basi-cranial axis is formed
by one continuous ossification, the Basi-sphenoid bone, excavated
superiorly (Figs. 50 and 52) by a saddle-shaped cavity, the sella
turcica, which lodges the pituitary body, — an organ of no great
physiological moment, so far as we know, but of first-rate mor-
phological significance.

On each side of the hinder part of the sella turcica, the
basi-sphenoid presents a groove for the internal carotid artery,
and this groove is completed in front and externally, by tin
osseous mass, tapering from behind forwards, the lingula
sphenoidalis, which lies between the basi-sphenoid and ali-
sphenuid. At the front part of the sella, separating it from the

THE STRUCTURE OF THE HUMAN SKULL. 121

depression Cor the optic commissure, there is a transverse ridge,
the iuberculum selhe* The region between the synchondrosis
and the tuberculum is the upper surface of the basi-sphenoid.
Its under-surface (Fig. 53) exhibits a median, wedge-shaped

portion, terminating abruptly at the point e, on each side of
which are stuck on, as it were, two delicate bones, shaped
somewhat like sugar-bags, with their wide and open ends
directed forwards and their apices backwards. These are the
bones of Berlin, or cornua sphenoidalia, which do not properly
belong to the basi-sphenoid, but coalesce with it in the course of
growth.

From the tuberculum sellw (c) to the point (b) in the upper
view (Fig. 52), and from the point e, to b of the lower view
(Fig. 53), the middle region of the cranio-facial axis belongs
to a third bone, the presphenoid (PS) which terminates the
basi-cranial axis.

I say terminates the basi-cranial axis, because the appear-
ance of a continuation forwards of that axis by the crista galli,
or upper margin of the lamina perpendicular is of the ethmoid
(see Fig. 50), is altogether fallacious, depending, as it does, upon
a special peculiarity of the highest Mammalian skulls, which
arises from the vast development of the cerebral hemispheres.
In the great majority of Mammalia below the Apes, in fact, the
free edge of the lamina perpendicular is is not horizontal, but
greatly inclined, or even vertical ; and in these cases the whole
lamina plainly appears to be, what it really always is, beyond,
or anterior to, the floor of the brain-case ; while the true basi-
cranial bones are parts of the floor of the brain-case.

During foetal life, the basi-sphenoid and presphenoid are
united only by synchondrosis, traces ol which may even be
discovered (as Virchow has shown) as late as the thirteenth
year, or later. Even before birth the two bones become
anchylosed superiorly, their junction being marked by the
tuberculum sellse ; and the remains of the synchondrosis extend

* Where tie terms employed in our ordinary handbooks of Human Anatomy
do not suffice for my purpose, I adopt those used by Henle in his classical
"Handbuch der Systematischen Anatomic des Menschen," a work of great ac-
curacy and comprehensiveness, now in course of publication.

122 ON THE VERTEBRATE SKULL.

obliquely from this spot, downwards and forwards, to the
point e (Fig. 53) on the uncler-surface of the axis, where its car-
tilage becomes continuous with the osseocartilaginous inter-
nasal septum.

It is this osseo-cartilaginous septum between the two nasal
cavities, the upper free edge of which constitutes the crista
galli, while the lower free edge supports the septum nariwm,
which terminates the hasi-facial axis.

All the upper and middle part of this septum is formed by
a thin osseous plate, the lamina perjjenclicularis of human
anatomy, or true Ethmoid (Eth.), which abuts, in front, upon
the frontal and nasal bones ; behind, upon the presphenoid ;
and below, upon a rod-like mass of cartilage, which becomes
connected with the septum narium and the prem axillary bones
anteriorly and inferiorly, and is obliterated with age.

The inferior and posterior part of the septum is constituted
by a bone with a gutter-like upper and anterior boundary,
which embraces the whole rounded inferior and posterior edge
of the cartilage in question, and thus extends from the under-
surface of the basi-sphenoid, posteriorly and superiorly, to the
middle of the roof of the bony palate, anteriorly and inferiorly.
This bone is the Vomer (Vo., Fig. 50).

Thus there are three bones in the hasi-cranial axis, — the
basi-occipital, basi-sphenoid, and presphenoid ; and there are
two bones in the hasi-facial axis, — the ethmoid and the vomer ;
the essential difference between these two sets of bones being
that the former constitute the middle part of the floor of
the brain- case, while the latter are altogether excluded there-
from.

^Ye may now turn to the upper arches of the skull, or those
bones which form the Avails and roof of the brain-case. In
the young skull from which the Figures 52 and 53 are taken, the
postero-lateral margins of the basi-occipital are united with
the rest of the occipital bone, only by synchondrosis. The
parts of the latter which are thus united with the basi-occipital,
and which limit the sides of the great occipital foramen, are
primitively distinct bones, -the Ex-occipitals (Eo.); while the

THE STRUCTURE OF THE HUMAN SKULL. 12o

squamous part, which bounds the posterior segment of the
foramen, is known as the Supra-occipital (So, So). All these
bones, eventually becoming anchylosed together, form the
occipital bone of the human anatomist ; or what we may term
the first, posterior, or Occipital segment of the skull.

From the sides of the basi-sphenoid, external to the lingulm,
two wide processes, well-known as the "greater wings of the
sphenoid ': or AUsphenoids (AS) spring, and unite sutnrally
with the expanded Parietal bones (Pa), which form the dome-
like crown of the skull, and unite in the middle line in the
sagittal suture. In this way a second, middle, or Parietal
segment of the skull is distinguishable.

In like manner, the presphenoid passes, on each side, into
the smaller processes, the "lesser wings of the sphenoid,"
alee minores, or wings of Ingrassias ; which, on account of their
relations to the orbits, have been well named the Orbito-
splienoids (OS). And these, externally and anteriorly, unite by
suture with the arched and expanded Frontal bones (Fr),
originally double, and separated by a median frontal suture,
which ordinarily early disappears. These bones not only meet
in front, but send in processes which roof over the orbits and
unite with the free anterior edges of the orbito-sphenoids, thus
leaving only a long and narrow vacuity, on each side of the
crista galli, and in front of the presphenoid.

The presphenoid, the orbito-sphenoid, and the frontals are
the constituents of the third, anterior, or Frontal segment of the
skull.

It will be observed, however, that this enumeration of the
bones of the three great segments of the skull does not account
for all the distinct osseous elements, which enter, directly and
indirectly, into its boundaries. If all the bones mentioned
are put together, there still remain four considerable vacuities ;
two small, already mentioned, in the proper front wall of the
skull, on each side of the crista galli; and one on each side,
posteriorly, between the occipital and parietal segments, of
very much larger size, and extremely irregular form. The
anterior vacuities are filled up by those spongy osseous masses,
united with the laiui na perpendicular is in the adult skull, which

124 ON THE VERTEBRATE SKULL.

are called " lateral masses of the Ethmoid." or " superior
and middle spongy bones," and more immediately by the per-
forated cribriform plate, which allows of the passage of the
filaments of the olfactory nerve, and connects these lateral
masses with the lamina perjpenclicularis, or proper ethmoid.
Looking at the bones which form the immediate walls of the
upper and middle part of the nasal chambers, with reference
only to the olfactory organs, we might say, in fact, that the
anterior vacuity of the cranium proper is stopped by the
ossified walls of the olfactory sacs, consisting of the ethmoid
and vomer in the middle line, of the superior and middle
spongy bones (or so-called lateral masses of the ethmoid)
supero-laterally, of the inferior turbinal bones infero-laterally.
And to these ossifications must be added, as members of the
olfactory group, the bones of Bertin, posteriorly and superiorly,
and the nasal bones, anteriorly and superiorly.

The great posterior vacuity on each side is filled up by the
Temporal hone, which consists of a very considerable number of
distinct elements, only distinguishable by dissection and by
the study of development in Man, but which remain perma-
nently distinct, and undergo very strange metamorphoses in
many of the lower Vertebrates. Some of these constituents
of the temporal bone, such as the squamous portion or Squa-
mosal (Sq.), and the Malleus, Incus, and Stapes, are discrimi-
nated by the student of ordinary human anatomy ; but there
are many others which he is not in the hab't of regarding as dis-
tinct osseous elements. rlhus the bony ''external auditory
meatus" is primitively a distinct bone, termed Tympanic (Ty.)
on account of its affording the frame in which almost the whole
of the tympanic membrane is set. The Styloid process (St.)
is originally a distinct bone. And, lastly, the pars petrosa and
pars mastoidea of human anatomy are, in reality, made up of
three distinct ossifications, of which I shall have to say more
presently, but which I shall speak of for the present under the
collective name of the Periotic bones, because they immediately
surround the organ of hearing:.

Not merely the periotic, but also the squamosal and tym-
panic bones are so closely related to the auditory organ, that

THE STRUCTURE OF THE HUMAN SKILL, 125

the posterolateral apertures of the cranium may be said to
be stopped by the osseous chambers of the auditory organ,
in the same way as the anterior apertures are closed by the
osseous chambers of the olfactory organs. As the eye is con-
tained only in a mobile fibrous capsule, the sclerotic, the
apertures which lead to the orbit — the spheno-orbital fissures
and the optic foramina — are not closed by any special bones
pertaining to the sensory organ lodged therein.

Thus the brain-case may be said to be composed of three
superior arches connected respectively with the three divisions
of the basi-cranial axis, and of two pair — an anterior and a
posterior — of bony sense capsules interposed between these
arches. A middle, third pair of sense capsules is not repre-
sented by bone in the cranial walls.

In like manner, the face may be resolved into a series of
bones, occurring in pairs from before backwards, and forming
more or less well-defined lower arches, some of which embrace
the nasal cavity, being placed in front of, or above, the oral
aperture, while others enclose the buccal chamber, and are
situated behind and below the oral aperture. Of the former,
fre-oral bones, there are four pairs — the Premaxillte (Pmx.),
the MaxiUm (Mx.), the Palatines (PI), and the Pterygoids (Pt.).

The Premaxillw, which lodge the upper incisor teeth, so
early lose their distinctness in man, by becoming anchylosed
with the maxillary bones (at any rate externally and anteriorly)
that they are rarely recognised as separate bones. Neverthe-
less, a suture extending upon the bony palate from the pos-
terior margin of the alveolus of the outer incisors to the
incisive foramen, very commonly persists, as an indication of
the primitive separation of these bones. The most important
character of the premaxilla?, regarded morphologically, is, that
they are connected, superiorily, with the anterior termination of
the cranio-facial axis, and that this connection is a primary one.
Each premaxilla passes from its inner end, which is united with
the axis, outwards and backwards, and two of the other three
pair of pre-oral bones have similar relations to the cranio-facial
axis. The anterior of these are the Palatine bones ; the inner,
or sphenoidal, processes of which are connected with the basi-

12G ON THE VERTEBRATE SKULL.

sphenoid and with the vomer ; while the outer, or orbital,
processes articulate with the so-called lateral masses of the
ethmoid and with the maxilla; so that the upper part of each
palatine bone is directed, from the cranio-facial axis, with which
its inner end is connected, outwards and forwards (Fig. 54). The
third pair of bones, the Pterygoids, are the internal pterygoid
processes, — bones which are originally quite distinct from the
sphenoid, while the external pterygoid processes are of a very
different character, being mere outgrowths of the alisphenoids.
These are connected with the basi-sphenoid (or rather with the
lingulce sjihenoidales) , above, and, in front, with the palatines,
while their planes are directed backwards and somewhat out-
wards. The fourth pair of pre-oral bones — the Maxillee — are
connected in front and internally with the premaxillae, and
behind and internally with the palatines, but they nowhere come
into direct contact with the cranio-facial axis, at least primarily.
I make the latter qualification because the vomer articulates
with the superior surface of the palatine plates of the rnaxillae,
and it may be said that, in this way, the maxillae do unite with
the cranio-facial axis. This articulation, however, has nothing
to do with the primitive connections of the bones, but depends
upon a modification of the maxillae peculiar to the higher
vertebrata. The bony apertures — called " posterior nares " — in
Man, for example, are structures of a totally different character
from, and superadded to, what are called the posterior nares
in a frog, or ordinary lizard, or bird. In these lower Verte-
brates, the posterior nares are apertures, bounded on the inner
side, by the vomer ; on the outer side and behind, by the
palatine bones; in front, by the premaxillae and maxillae. In
Man, on the other hand, the apertures so called are limited, it is
true, on the inner side by the vomer, and on the outer side
by the palatine bones ; but they are also bounded below and in
front by the palatine bones, and the premaxillae and maxillae
have nothing to do with them. On looking closely into the
matter, however, it will be found that that region of the palatine
which forms the outer and inferior boundary of the posterior
nares of Man is a something which has no representative in the
lower Vertebrate.

THE STRUCTURE OF THE HUMAN SKULL.

127

But if, with a fine saw, the greater part of the perpendicular
plate of the palatines, and the corresponding part of the maxil-

laries, and, with those, their palatine plates, bo cut away,
leaving only the premaxillae, vomer, and upper parts of the
maxillary and palatine bones ; it will be found that hinder
nares are left, which entirely correspond with the " posterior
Dares " of a bird or of an amphibian ; that is to say, they are
passages between the vomer in the middle line, the premaxillae
and maxilla? in front and externally, and the palatines ex-
ternally and behind.

Fio-. 54.

b'ig. 54. — The base of a human skull — the nasal, ethmoid, vomerine, maxillary, palatine, and
pterygoid bones being cut through horizontally, and their lower portions removed.
The entire right maxilla is taken away. The posterior pair of letters, JViV, are situated
in the median nares, which are incomplete, in fiont, in consequence of the removal of
the premaxillae.

In fact, the apertures of the nasal chamber into the mouth,
thus artificially exposed, are those which originally exist in
Man and the higher Vertebrata ; but the downward growth of
the maxilla into its alveolar process, and of the palatine bone
into its perpendicular plate, together with the production inwards

128 ON THE VERTEBRATE SKULL.

of the palatine plates of these bones, which eventually unite with
the vomer, give rise to the apertures, which are ordinarily
called posterior nares. So that in Man, for example, there are
three pairs of " nares :" — the external, situated between the
anterior end of the internasal septum, the nasal bones, and the
premaxilla?, as in the lower Vertebrates ; the median, between
the vomer, the palatines, and the premaxillae, which correspond
with the posterior nares of the lower Vertebrates ; and the
posterior, between the vomer, internally, and the palatines
above, at the sides, and below, which are peculiar to the higher
Vertebrates.

And, to return to the maxilla, we find that it reallv differs
altogether from the other pre-oral bones, and is, as it were,
fastened on to the outer sides of the premaxillary and palatine
bones, without having any primary direct connection with the
cranio-facial axis.

The post-oral bones surround the buccal cavity, and form
two distinct arches — the mandibular and the liyoidean. Neither
of these arches is directly connected with the cranio-facial
axis, nor with the segments of the brain-case, but both are
suspended to different parts of the temporal bone, which is so
singularly intercalated between the middle and posterior of those
segments.

The lower jaw or Mandible (Mn) consists of two rami, anchy-
losed at the symphysis, and each consisting of a single piece, the
condyle of which articulates with the squamosal.

The Hyoid bone (Hy), composed ofits body and two pairs of
cornua, does not articulate directly with the temporal bone,
but ligaments connect it with the styloid processes, and these
last bones unite with the posterior part of the periotic capsules.

Thus, the natural connections of the bones by no means
allow of the separation of the walls of the lower chambers of
the human skull into a series of arches springing from, and
corresponding with, the axial parts, as wTe found to be the case
with the walls of the upper chambers.

If the temporal bone be detached, the hyoidean and man-
dibular arches come with it, and exhibit no connection with
the occipital or the parietal segments. Indeed, the latter is

THE STRUCTURE OF THE HUMAN SKULL. 129

preoccupied by the pterygoid and the palatine, both of which
are connected with the basi-sphenoid (at least with the lingulse),
while the anterior part of the palatine is also connected, in the
adult state, with the presphenoid, by the intermediation of the
cornua sphenoidalia.

Two bones yet remain to be mentioned which come neither
into the category of axial bones, nor of superior or inferior arch
bones, nor, strictly speaking, of sense-capsule bones. These are
the Lachrymal (L)9 intercalated between the nasal, maxillary,
and lateral mass of the ethmoid, and serving to lodge the con-
duit which places the orbit and the nasal cavity in communi-
cation ; and the Jugal or Malar (Ju), which connects the bones
of the orbital chamber with the squamosal element of the tem-
poral bone.

The skull, thus composed, serves as a protection to the organs
which are lodged within it, and which are of as great importance
in their morphological, as in their physiological, aspect.

The cerebral hemispheres and cerebellum, with their de-
pendent parts, fill the cranial cavity, the lower lateral margin of
the posterior cerebral lobes corresponding with the torcular
Herophili and the lateral sinuses, on the inner surface of the
occipital bone ; or, in other words, with the line of attachment
of the tentorium. Certain axial parts of the brain have definite
relations to the axial parts of the cranium. Thus, the medulla
oblongata lies upon the basi-occipital. The pituitary body
rests upon the upper surface of the basi-sphenoid, this bone con-
stituting the chief part of the front as well as of the hinder wall
of the sella turcica. The chiasma of the optic nerves rests upon
the hinder portion of the upper face of the presphenoid, and the
peduncles of the olfactory nerves upon the front portion of that
face. The termination of the axial parts of the brain in the
lamina terminalis of the third ventricle corresponds pretty
nearly with the termination of the basi-cranial axis in the ante-
rior extremity of the presphenoid.

Not less important are the relations of many of the cerebral
nerves to the lateral elements of the arches of the brain-case.

The filaments of the olfactory nerves pass out through the

130 ON THE VERTEBRATE SKULL.

cribriform plates, leaving the ethmoid proper, or lamina perpen-
dicularis, upon their inner side, and the lateral masses of the eth-
moid, or superior and middle spongy bones, upon their outer sides.

The optic nerves pass out through the optic foramina, situated
between the roots of the orbito-sphenoids, from the chiasma,
which rests, as has just been stated, upon the posterior and upper
part of the presphenoid. Hence it follows, that the presphenoid
lies in front of, and between, the optic nerves, which embrace
it, as in a fork, from behind.

The third and fourth pairs are not of so much morphological
importance that I need dwell upon them, but the trigeminal
affords first-rate cranial landmarks bv its nasal branch and its

whole third division. The nasal nerve enters the orbit by the
foramen Jacerum anterius, passes to the inner side of the eye,
and then, traversing the anterior of the two ethmoidal foramina,
perforates the " lateral mass of the ethmoid," and entering the
cavity of the bony cranium, though it always lies beneath the
dura mater, skirts the olfactory aperture, and passes out into
the nasal cavity, by an aperture in the front part of the cribri-
form plate. We shall find this irregular perforation of the
" lateral mass of the ethmoid," bv the nasal division of the fifth
nerve, to be an excellent guide to the determination of the
homologue of the bone in the lower Vertebrata.

The third division of the trigeminal traverses the foramen
ovale in the posterior part of the alisphenoid, so that it makes
its exit behind the greater part of that bone, and altogether in
front of the periotic bone.

The portio dura enters the internal auditory foramen in the
periotic mass, runs along its canal, situated above the fenestra
ovalis, and eventually passes out by the stylo-mastoid foramen.
It therefore perforates the fore part of the periotic, passino- in
front of the membranous labyrinth. The portio mollis also enters
the periotic bone by the internal auditory foramen, and it termi-
nates in the membranous labyrinth.

The eighth pair passes out through the foramen lacerum
posterius completely behind the periotic (which thus lies between
the exits of the fifth and of the eighth pairs), and in front of
the ex-occipitals.

THE STRUCTURE OF THE HUMAN SKULL.

i;;i

The ninth pair perforates the ex-occipitals in front of the

condyles.

With regard to the relations of the nerves to the inferior
arches of the skull, only one circumstance calls for particular
notice, — the distribution of the terminal divisions of the portio
dura. This nerve divides, as it is about to leave the temporal
bone, into two portions, the larger of which passes out by the
stylo-mastoid foramen, and, besides giving off many other
branches, supplies certain muscles of the hyoidean apparatus.

The smaller division of the nerve, of comparatively insignifi-
cant size — the chorda tynvpani — returns to the tympanic cavity,
crosses it, and leaving it by an aperture internal to, and above
the tympanic element, runs down upon the inner side of the
lower jaw. In Man, the great development of the facial muscles
gives a predominance to the branches of the portio dura which
supply them ; but, in the lower Vertebrates, the nerve becomes
more and- more completely represented by simple mandibular
and hyoidean divisions, corresponding respectively with the
chorda tyw/pani and the branches distributed to the stylo-hyoid
and digastric.

In the preceding description of the architecture of the human
skull, I have, as far as possible, avoided complicating the gene-
ral view of its structure which I have desired to give, by enter-

il EjiO.

Fig. 55. — Human left temporal bone, half the natural size. — a b, posterior root of the
zygomatic process ; e, middle root ; /, anterior root ; b, post-auditory fossa ; m i, long

_* _.. .riL. n_ i ,f .L. :„.,..„

processes of the malleus and of the incus.

K 2

132 ON THE VERTEBRATE SKULL.

ing into any details which were not strictly necessary ; but there
remains one part of the cranium — the temporal bone — the struc-
ture of which must be carefully and thoroughly investigated, if
we desire to understand the modifications undergone by the
bones which correspond with its constituent elements in other
Vertebrata.

Viewed from without, the temporal bone presents the well-
known pars squamosa (Sq.) and pars mastoidea (ill), in the
re-entering angle between which, the tympanic element (Ty.) is
fixed (Fig. 55).

No suture separates the pars squamosa from the pars mas-
toidea, but the posterior limits of the former are indicated, in the
first place, by the curved ascending portion of the posterior root
of the zygoma (a b), which bounds the attachment of the tempo-
ral muscle ; and secondly, by a curved ridge, convex backwards
and differently defined in different subjects — the mar go tijmpa-
nieus of Henle — which passes downwards, behind the auditory
meatus, until it cuts the contour of the tympanic bone. Near
the upper end of this ridge, or " post-auditory process," is an
elongated " post -auditory fossa " (b), more marked in old than
in young subjects.

The portion of the squamosal element, the free edge of which
terminates in this ridge, forms an arch, of which the posterior
pillar constitutes the posterior and upper wall of the auditory
meatus, while the anterior pillar forms the front boundary of the
glenoid cavity. The centre of the arch is interrupted by the
middle root of the zygoma (e), or " the post-glenoidal process " of
the squamosal, which runs, as a wedge-shaped ridge, transversely
to the span of the arch.

The upper edge of the anterior wall of the gutter-shaped
tympanic bone (which forms the hinder boundary of the glenoid
cavity), unites with this riclge, crossing its direction obliquely
inwards and forwards. Beyond the ridge it is no longer united
with the squamosal, but, keeping its oblique direction, crosses
rather to the inner side of the lower edge of that bone, and
leaves the Glaserian fissure between the squamosal and itself.

A section taken through both the external and the internal
auditory meatuses (Fig. 56) shows that this arched plate of the

THE STRUCTURE OF THE HUMAN SKULL.

L33

squamosal is interposed between the upper half of the tympanic
and the upper parts of the pars petrosa and pars mastoidea, the
depth of the interposed squamosal being greatest posteriorly,
while it diminishes to nothing anteriorly.

The upper region of the pars petrosa, however, does not
directly abut, by its thick mass, against the squamosal, but by
a thin horizontal plate, which roofs over the tympanum, the

Fig. 56.

Fig. 56. — Views of the petrous and tympanic portions of the right temporal bone of the
skull represented in Figs. 48 and 49, magnified two diameters. — A, the anterior half of
the bone; B, its posterior half; I.M., internal meatus; E.M, external meatus; o,
processus cochleariformis ; b, chamber in which the heads of the malleus and incus
lie ; c, groove for the tympanic membrane ; S. V.C., superior vertical semicircular
canal ; ex., external semicircular canal; p.c, posterior vertical semicircular canal ;
Co, cochlea; JP, Pyramid ; F.O,, fenestra ovalis ; IT/, canal for portia dura ; VII1,
for poitio mollis.

134 ON THE VERTEBRATE SKULL.

Eustachian tube, and the antrum mastoideum, and is the tegmen
tympani.*

The lower region of the pars petrosa in like manner gives off
a thicker and shorter plate, which forms the floor of the Eusta-
chian tube and the outer or inferior boundary of the carotid
canal, in front ; the floor of the tympanum, in the middle ; and
then, becoming gradually thicker, constitutes the lower boun-
dary of the antrum mastoideum. It is with the outer edge of
this inferior, or floor-plate, of the tympanum that the lower por-
tion of tympanic bone becomes anchylosed. The inner wall is
of course constituted by the outer surface of the more massive
part of the pars petrosa. Thus, the roof and part of the floor of
the tympanum are formed by the superior and inferior prolonga-
tions of the pars petrosa, while the outer wall of the tympanum
is constituted above by the squamosal, and below by the tym-
panic. A section taken vertically and transversely to the axis
of the skull through the middle of the fenestra ovalis, in the way
described above, shows that the squamosal limits, externally, an
upper chamber of the tympanum (b, Fig. 56), which is nearly as
deep as, and is wider than, the lower division, bounded externally
by the tympanic membrane and tympanic bone (Fig. 56). It is in
this upper chamber that the heads of the malleus and incus are
lodged, the handle of the one and the long process of the other,
only, depending into the proper tympanic cavity. Hence, in
looking into the tympanum from without (Fig. 55) when the ear-
bones are in situ, only these processes are seen, the heads of
both malleus and incus being hidden by the arched plate of the
squamosal.

Thus, the tympanum is formed by a very complicated adjust-
ment of bony elements, and we shall by and by see reason to
believe that it is even more complex than it now appears to be,
inasmuch as the so-called pars petrosa will prove to be composed
of two distinct elements ; an inferior, opistliotic, bone, containing
the lower part of the cochlea, and a superior, pro-otic, sheltering
the greater part of the vestibule, the upper part of the cochlea,
the anterior vertical semicircular canal, part of the posterior
vertical canal., and the external semicircular canal.

* It lios immediately beneath the letters Pr.O., Fig. 56. A.

THE STRUCTURE OF THE HUMAN SKULL, 135

Behind the posterior boundary of the squamosal, constituted
by the two diverging lines above described (Fig. 55), lies all that
portion of the temporal bone which is known as the pars mas-
toiclea. But, as I shall have occasion to demonstrate, when ex-
plaining the mode of development of the temporal bone, this
pars mastoidea is, in reality, made up of extensions of two of the
primitive constituents of the pars petrosa, and of a third element,
the epiotic. The posterior margin of the squamosal, as above
described, may be said roughly to form two sides of a parallelo-
gram. The third side is the thick part of the upper edge of the
pars mastoidea, corresponding with the termination of the upper
and anterior surface of the ptars petrosa on the inner side of
the bone. If a fourth side is made bv an imasinarv line con-
necting the ends of the others, the bony surface which lies above
and in front of the line will, as nearly as possible, belong to the
pro-otic element, while that which lies below and behind it, in-
cluding the mastoid process, appertains to the epiotic. On the
other hand, a certain amount of the pars mastoidea internal to
the digastric groove belongs to the opisthotie.

13G

LECTURE VIII.

ON THE STRUCTUEE OF THE SKULL.

THE DEVELOPMENT OF THE HUMAN SKULL.

As might be expected from the nature of the case, it lias not
yet been possible to obtain a series of human embryos, in every
stage of development, sufficiently large to enable embryologists
to work out all the details of the formation of the human skull.
But all higher vertebrate embryos so nearly follow one and the
same type of early developmental modification, that we may
reason, with perfect confidence, from the analogy of the lower
Vertebrates to man, and fill up the blanks of our observations
of human embryos by investigations of the chick, the dog, the
rabbit, or the pig.

In the chick,* the first indication of the body of the embryo
is an elongated, elevated area of the blastoderm, the axis of
which is traversed by a linear groove. The one end of the
elongated area is wider and more distinctly raised up from the
rest of the blastoderm, than the other : it is the cephalic end
(Fig. 31, A, a), and the linear groove stops short of the rounded
extremity of this part of the elevated area. A peculiar cellular
cylinder, tapering oft' at each end, the notochord, is soon dis-
cerned occupying the bottom of this groove, beneath the outer,
serous, or neuro-epidermic layer of the germ.

A laminar outgrowth of the convex summits of the ridges
Avhich bound the primitive groove now takes place, in that part
of the embryo, which will eventually become the middle region

* See Lecture IV., pp. G4— 6G.

THE DEVELOPMENT OF THE FOWL'S SKULL. L37

of the head; and the dorsal laminse, thus produced, extending
forwards and backwards, like parapets, upon each side of the
primitive groove, lay the foundations of the lateral walls, not
only of the skull, but of the spinal column.

Very early, however, the boundary line between skull and
spinal column is laid down, by the appearance in the substance
of the bases of the dorsal lamina) and the adjacent middle layer
of the blastoderm, of the first pair of those quadrate masses
of condensed tissue, the jproto-vertebrte (" Urwirbel ': of the
German writers), which are the foundations, not only of the
bodies of the vertebra?, but of the spinal muscles and ganglia.
The proto-vertebrse increase in number from before backwards ;
and, at length, extend through the whole range of the spinal
column, while none ever make their appearance in the region
which will be converted into the skull.

The edges of the dorsal laminse now unite, the coalescence
taking place first in the middle cephalic region, and extending
thence backwards and forwards ; at the same time, the cephalic
canal becomes separated into three distinct dilatations, or cere-
bral vesicles, of which the anterior is by far the most marked
(Fig. 57, A, I, II, III).

The rudimentary cranial cavity next becomes bent upon
itself in such a manner, that the longitudinal axis of the first
cerebral vesicle takes a direction at right angles to the axis of
the third, and of the spinal canal generally. In consequence of
this change, the middle cerebral vesicle occupies the summit
of the angulation, and becomes the most anterior point of the
whole body (Fig. 57, C, D).

The bend thus produced is the cranial flexure. It results
in the division of the floor of the cranial cavity into two parts,
an anterior and a posterior, which are at right angles to one
another (Fig. 57, C, D, E). Hitherto, no trace of the noto-
chord has been observed in the anterior division, that structure
ending in a point behind the flexure (Fig. 57, D, E, li).

As development proceeds, the anterior cerebral vesicle be-
comes divided into two portions, — an anterior, the vesicle of the
cerebral hemispheres (J") ; and a posterior, the vesicle of the
third ventricle (I ). In the upper wall of the vesicle of the third

Vigr. 57.

a-Srt

<(J

1D rvi

I 1^

3 K H V

iffl

\1 f*\'<: I ,|J W

;

Fig. 57. — Successive stages of the development of the head of a chick. — /, IT, TIT,
first, second, and third cerebral vesicles ; T", vesicle of the cerebral hemispheres; /'',

THE DEVELOPMENT OF THE FOWl/S SKULL. 139

ventricle the rudimentary pineal gland (e) makes its appearance
in the middle line. From the middle of the lower wall grows
out a process, the infundibuluni, terminating in a glandular
appendage, the pituitary body, which last is lodged in the deep
fossa situated in the floor of the anterior division of the skull,
immediately in front of, and beneath, the termination of the
notochord (Fig. 57, B, D, cl).

The three pairs of sensory organs appertaining to the higher
senses, — the nasal sacs, the eyes, and the ears, — arise as simple
caecal involutions of the external integument of the head of the
embryo. That such is the case, so far as the olfactory sacs
are concerned, is obvious ; and it is not difficult to observe that
the lens and the anterior chamber of the eye are produced in
a perfectly similar manner. It is not so easy to see that the
labyrinth of the ear arises in this way, as the sac resulting from
the involution of the integument is small, and remains open
but a very short time (Fig. 57, C, b). But I have so frequently
verified Huschke's and Remak's statement that it does so arise,
that I entertain no doubt whatever of the fact,* The outer
ends of the olfactory sacs remain open, but those of the ocular
and auditory sacs rapidly close up, and shut off their contents
from all direct communication with the exterior. The olfactorv
nerve is developed from the anterior division of the anterior cere-
bral vesicle. The optic nerve is primarily developed from the
posterior division of that vesicle, its connection with the middle
vesicle (which eventually gives rise to the corpora quadrigemina)

vesicle of the third ventricle ; a, rudiments of the eyes and optic nerves ; b, of the
ears ; g, of the olfactory organs ; d, the infundibuluni ; e, the pineal gland ; c, proto-
vertebrae; h, notochord; 1, 2, 3, 4, 5, visceral arches; V, VII, VIII, the
trigeminal, portio dura, and eighth pair of cranial nerves ; k, the fronto-nasal process ;
I, the maxillary process ; x, the first visceral cleft.

A, B, upper and under views of the head of a chick at the end of the second day.

C, side view at the third day.

D, side view at sevent3T-five hours.

E, side view of the head of a chick at the fifth day, which has been subjected to
slight pressure.

F, head of a chick at the sixth day, viewed from below.
F1, the cartilaginous cranium of the same.

V, pituitary space ; tr, trabecule ; Qu, quadrate cartilage ; Sc, semicircular canals ;
Co, cochlea.

G, head of a chick at the seventh day, from below.

* Sec also Kolliker's " Entwiokelimgs Geschiolito," p. 300, et seq.

140 ON THE STRUCTURE OF THE SKULL.

being only secondary. The auditory nerve is developed in the
blastoderm adjacent to the third cerebral vesicle, so that the
three pairs of sense-capsules do not correspond with the three
primary cerebral vesicles.

While these changes have been going on in the proper
cranial portion of the embryo, the rudiments of the face have
made their appearance under a very singular guise. As the
homologues of the dorsal laminae in the head have grown up-
wards to inclose the cephalic cavity, so, plates, which correspond
with the visceral laminae of the trunk, have grown downwards,
to constitute the posterior walls of the buccal, pharyngeal, and
cervical regions. These visceral plates, however, do not remain
entire and undivided, as do those of the trunk, but grooves
appear in them, directed transversely to the axis of the trunk,
and, the grooves deepening, eventually become converted into
slits — the visceral clefts — which open into the pharyngeal cavity,
and bound corresponding visceral arches. The first slit is
situated immediately below and in front of the auditory sac,
and separates the first and second visceral arches, — the anterior
boundary of the former being determined by the edges of a
depression of the integument which wall eventually become the
buccal cavity (Fig. 57, C). A third, fourth, and fifth visceral
arch are developed in successive order behind the first and
second (Fig. 57, D, and Fig. 32, F) ; but as they are of no great
moment in reference to the human skull, our attention may be
confined to the latter.

It is particularly worthy of notice that, from the moment at
which it is discernible as a distinct part, the root of the first
visceral arch passes into the rudimentary cranium below, and in
front of, the forepart of the auditory sac, while the root of the
second is attached below and behind that sac. We shall find
that the parts developed within these arches retain the same posi-
tion in the adult state ; so that any hypothesis which involves
the supposition of an extensive change of place of these parts in
the course of development is, ipso facto, unworthy of consideration.

Both the first and second visceral arches are connected with
that part of the cranium which lies behind the flexure ; but the
inflected portion of the cranium in front of the bend exhibits,

THE DEVELOPMENT OF THE FOWL'S SKULL. 141

on each side, running from the root of the first visceral arch
beneath the eye to the nasal sac, a ridge or elevation, which is
called the maxillary process, and might be regarded as a visceral
arch of the anterior division of the skull, from the base of which
it is developed (Fig. 32, G, and I, Fig. 57, F).

Lastly, the middle part of the floor of the anterior cerebral
vesicle, between the nasal sacs, thickens and gives rise to a
broad, flat median process, with an expanded extremity, the
terminal contour of which is excavated and slightly produced
at the angles — the fronto-nasal process (Fig. 57, F, h).

At first, the cranium and all its arches are membranous, or
composed of mere indifferent tissue, with the exception of the
axial notochord; but, very early, chondrification commences.
The indifferent tissue surrounding the notochord (the " invest-
ing mass" of Kathke) (Fig. 57, 0, D,/), is converted into carti-
lage, and the same histological change takes place in the Avails
of the auditory capsules, and around the foramen magnum ; the
cartilage stops in the middle line, behind the pituitary body,
but sends two processes, one on each side of that body, into the
floor of the anterior division of the skull (Fig. 57, F1, tr). These
processes, the trabecule cranii, of Kathke, unite in front, and
the cartilage formed by their union ends in the fronto-nasal
process. The roof of the skull, and the greater part of its side-
walls, except in the region of the foramen magnum, are, at first,
entirely membranous. Chondrification next takes place in the
visceral arches ; a rod of that substance, which coalesces with
its fellow in the middle line, being formed in the axis of the
several arches on each side.

Purposing to return to the visceral arches by and by, I
shall now trace out the modifications which are undergone by
the chondro-membranous brain-case. In the occipital region,
and about the auditory capsules, which early attain a very large
proportional size, the cartilage extends for some distance upon
the infero-lateral parietes of the skull ; on the floor of the
posterior division of the skull it thickens notably, and forms a
sort of model of the future basi-occipital and basi-sphenoidal
regions, the interspace between the trabecule becoming rapidly
obliterated and converted into the floor of the pituitary fossa.

142 OX THE STRUCTURE OF THE SKULL.

In front, the coalesced trabecular become changed into a plate
of cartilage, compressed from side to side, which occupies the
middle of the gradually-narrowing fronto-nasal process, as the
ethmo-vomerine, or internasal, cartilage.

From the sides of the basi-sphenoid cartilaginous plates are
developed, which foreshadow the form and relations of the
alisphenoids ; at the sides of the presphenoidal region of the
cartilage, similar plates represent the orbito-sphenoicls. In
front of these the upper part of the internasal, or ethmoidal, car-
tilage passes laterally into broad deflected cartilaginous lamellar,
which curve round the olfactory sacs, and occupy the places of
the lateral masses of the ethmoid and the inferior turbinal bones.

Thus far the terms of my description are almost as appli-
cable to the embryonic cranium of Man as to that of the chick.

The human cranium has been observed forming part of an
open groove ; it undergoes a flexure, and develops visceral
arches altogether similar to those of the chick, nor is there any
reason to doubt that the organs of sense are developed in the
same manner. The very earliest condition of the cartilaginous
cranium of the human embryo has not been observed ; but, at
the beginning of the second month, it consists wholly of carti-
lage and of membrane, disposed in a manner which differs only
in detail from that seen in the chick. Thus the occipital fora-
men is surrounded by cartilage, continuous with that which
extends through the basi-sphenoidal, presphenoidal, and ethmoidal
regions to the anterior end of the face. The alisphenoids and
orbito-sphenoids are represented by cartilage, and cartilaginous
plates arch clown from the summit of the internasal cartilage,
on each side, to form the substratum on which, the nasal bom ss
and, in which, the spongy bones will be developed. That part
of the cranial cartilage which lodges the auditory organ is
exceedingly large, and constitutes, not only an oval capsule for
the membranous labyrinth, but sends back a continuation which
fills the space corresponding to the pars mastoidea, and extends
somewhat higher than it beneath the parietal region of the
skull. All the upper part of the cranium is and remains simply
membranous.

THE DEVELOPMENT OF THE HUMAN SKULL.

143

The relations of the regions of the chondro-cranium thus
formed to the parts of the brain and to the exits of the nerves
are the same as those which are observed in the bones which
they prefigure.

When these bones begin to be developed, some of them
make their appearance in the cartilage of the embryonic skull,
some in the perichondrium, others in the membranous roof
which is continuous with the perichondrium.

A single ossification appears around the notochord in the
basi-occipital region, and lays the foundation for the basi-
occipital bone. The ex-occipitals commence as single centres of
ossification in the cartilage bounding the sides of the foramen
magnum. The supra-occipital (SO) is developed from two
ossifications in the cartilage above the foramen magnum, and

:■ A
7Z\

6
a-1

■m

N»''

* ' ■

is. ■''«! f

■Ik-

w ■■...■,

0. 1 h&~~ rfJ

■h

Fig. 58. — Successive embryonic conditions of the human head (after Ecker). — I, II, the
first and second cerebral vesicles. 1, 2, 3, 4, the visceral arches; a, the maxillary-
process ; 6, the eye ; nf, the middle nasofrontal process ; c, the lateral naso-frontal
process ; t, the tongue ; an, the outer part of the first visceral cleft, which eventually
gives rise to the external auditory meatus.

A, at three weeks.

B, at five weeks.

C, at six weeks.

D, at seven weeks.

E, at eight weeks.

F, outline side view of E.

144

ON THE STRUCTURE OF THE SKULL.

from two others which appear, not in cartilage, but in the mem-
branous roof of the skull above the limits of the cartilage, and
so give rise to that part of the squama occipitis marked SO'.

The basi-sphenoid is developed from two centres of ossifica-
tion which appear in the floor of the sella turcica, but speedily
coalesce into one. Two separate centres of ossification appear
in the cartilage between these and the alisphenoids, and form
the lingulm sphenoidales. Each alisphenoid is developed from a

Fie. 59.

Fig. 59. — A, upper, and B, under view or" the basis cranii and periotic cartilage of a
human fcetus eight inches long. The alisphenoidal and immediately adjacent parts of
the basi-sphenoid are omitted. The cartilage is darkly shaded, while the osseous de-
posits are left white, or but lightly shaded. The greater part of the supra-occipital
is cut away. The clear spaces close to the dotted lines leading from t.tj are apertures
in the cartilage. The epiotic classification has not yet appeared, and the pro-otic and
opisthotic ossifications are quite distinct on the right side.

THE DEVELOPMENT OF THE HUMAN SKULL. 145

single centre in its cartilaginous predecessor, but the parietals

are the result, not of the ossification of cartilage, but of that
of the membrane which roofs in the skull. Each has its own
centre of ossification in this membrane.

The presphenoid arises by two separate centres of ossification,
one on the inner side of each optic foramen. (Fig. 59, P.S.)
These centres coalesce with the orbito-sphenoids of their own
side before thev unite with one another.* The osseous orbito-

* The mode of ossification of the sphenoid bone is one of the most difficult ques-
tions in osteogenesis. Meckel has worked out the problem at great length in his
"Archiv," Bd. i., and thus sums up his results in the " Handbuch der Menschlichen
Anatomie," Bd. ii., pp. 102-4 : —

" In the third month, the first osseous nuclei appear in the two great wings, and
soon afterwards the internal pterygoid processes begin to ossify as separate bones.
Next, a third pair of ossifications appears hi the external circumference of the alse
minores ; and then, about the fourth month, a seventh and eighth nucleus, which lie
side by side in the body of the sphenoid. In the fifth month is formed, alongside
this fourth pair, a fifth, between it and the great wings. Upon tins the two median
nuclei of the body coalesce. Soon arises a sixth nucleus, on the inner side of the
optic foramen, and then a seventh appears between this and the fourth, so that, about
the beginning of the seventh month, the sphenoid consists of thirteen separate bony
nuclei, since, notwithstanding seven pahs have arisen, the two primary nuclei of
the body early coalesced into one.

"From this time forth the number of the nuclei diminishes still more con-
siderably by coalescence. Those nuclei coalesce earliest which give rise to the
portions of the sphenoid, which persist in a separate state longest. The fourth, fifth,
and seventh pair soon unite into one piece ; the first and second, coalescing on each
side, constitute two other pieces ; the third and sixth two others ; whereby, in the
eighth month, the sphenoid consists of five pieces — the two greater wings, the lesser
wings, and the body. Somewhat later the two lesser wings coalesce into one, and
the sphenoid now consists of four pieces ; thereupon the body and the anterior pieces
unite, so that in the fully-formed foetus the sphenoid consists of three pieces, the
greater wings and internal pterygoid jn'ocesses being still distinct ; but in the first
month after birth these three pieces unite into one."'

The fifth pah of ossifications here mentioned are the lingidse ; the sixth, those
which give rise to the presphenoid. Meckel's seventh pair of ossifications, which
arise between the fourth (basi-Sphenoidal) and the sixth (presphenoidal), and are
said, in the "Archiv, ' to coalesce first with one another, and then with the basi-
splienoid, appear not to have been observed by other anatomists. I have not seen
them, and they are not mentioned by Virchow, the latest writer on the subject.
Virchow writes ( " Entwickelung des Schadel-grundes," 1857;: — "The posterior
sphenoid arises (if we leave out of consideration the internal pterygoid processes
which are developed as separate and independent bones), according to most authors,
from three nuclei, but, according to my observations, from six. Two of these belong
to the alse magnae (alse temporales), or lateral arches (Bogenstiicken, of the parietal
vertebra. They arise in the third month, and the external pterygoids are produced
by direct outgrowths from them. In the third month, I also find two other centres

146 ON THE STRUCTURE OF THE SKULL.

sphenoids arise each by one centre in the corresponding cartilages.
The frontals, on the other hand, are developed, like the parietals,
each from one centre in the membranous roof of the skull.

Thus we arrive at the singular result that, while all the bones
of the basi-cranial axis, and all the lateral bones of the three

of ossification which belong to the apices of the lingulse, and are separated by dis-
tinct layers of cartilage from the others. The ossification of the lingida is almost
complete in the fourth month, and its size is out of all relation to the dimensions
of the other parts. It is a thick, obtusely-cylindrical process, which coalesces
primarily with the body, and has nothing to do with the alas. The lingula is there-
fore similar to an anterior or inferior transverse process (Parapophysis, Owen) ; and
the sulcus caroticus, notwithstanding its position in the inner side of the lingida,
resembles an open foramen vertebrcde. However, Arnold's opinion that the Vidian
canal answers to the canal for the vertebral artery, notwithstanding it is placed on
the inner side of the lingula, deserves the careful attention of comparative anato-
mists. The ossification of the body begins in the third month, exactly under the
pituitary fossa, which is already preformed in cartilage. Kerckring was the first to
point out that here the adjacent osseous centres at first arise, and that they unite
and form a biscuit-shaped mass in the fifth month. Once he saw this ' semilunula '
even in the middle of the third month. Kolliker and I myself have met with it in
foetuses of three months. Other observers, as Nesbitt and Mayer, speak of a single
centre in the third month, and in the fourth of two centres, which must be re-
garded as the result of the erroneous combination of different individual cases. I
find constantly, in the beginning of the third month, two nuclei, which arise near
the upper surface in the anterior wall of the pituitary fossa, and are separated by a
broad layer of cartilage. Very soon, however, only a single osseous mass is present
in the interior of the body, which extends through the whole thickness of the
cartilage, while anteriorly and posteriorly it is still enveloped in cartilage. In a
foetus 19 centimetres [7J inches] long, I saw the simple osseous nucleus in the
bottom of the sella, as a transverse plate which had not yet united with the lingula."
" The anterior sphenoid is developed by the gradual coalescence of four osseous
centres, of which again two belong to the body and one to each of the lesser wings.
The latter are developed earlier than the former. They commence early in the
third month, in the anterior clinoid processes, which are quite thick and osseous
at a time when everything else in the anterior sphenoid is hyaline cartilage, and
therefore are quite similar to the lingutes. From this point ossification proirn sss -
rapidly, at last ere. pint;- round the circumference of the optic foramen to the body of
the ala and to its anterior root. About the fifth month the lesser wing is completely
solid in all parts. On the other hand, the nuclei in the body mostly appear some-
what later, usually in the fourth month, and at the inner edge of the optic foramen,
so that they are at first separated by a tolerably broad median lamella of cartilage,
which is continued into the ethmoid cartilage and septum narium. A union now'
very si ion takes place between the centres of the body and those in the lesser wings,

so that the optic foramen is surrounded by bone Later, at times, as il

app< ars, as early as the fifth month, the two lateral masses unite into a larger central
piece, which is free superiorly, while below and anteriorly, in the middle line, it is
surrounded by broad masses of cartilage." — Virchow, he. r-it., pp. 15-18.

THE DEVELOPMENT OF THE HUMAN SKULL. I 17

cranial arches, arc primarily developed in cartilage, only one of
the superior elements of these arches — the supra-occipital (SO)
— is so ; while the upper or " interparietal ' portion of the
squama ocevpitis (SO1) and the two other pairs of superior
elements of the arches are developed altogether from mem-
brane.

Fig. 60.

Aa.jjt^' '.

S.O*

M a MO.

KiiT. 60. — Longitudinal and vertical section of the basis cranii of a foetus somewhat older
than the foregoing (Fig. 59). The basi-sphenoidal and presphenoidal centres have
coalesced ; but they and the basi -occipital are severally separated by wide interstices
of cartilage, of which the whole ethmoidal region is still constituted.

The ethmoid is developed from a single centre, arising in
the internasal cartilage. Its so-called lateral masses, with the
two upper spongy bones, are likewise developed each from a
single centre within the superior part of the inflected lateral
cartilages which wall in the olfactory sacs. The inferior tur-
binals are ossifications of the lower parts of these cartilages.
But the nasal bones are developed within the perichondrium,
which is continuous with the membrane in which the frontal
bones are developed, and the vomer is produced within the
perichondrium on the under-surface of the internasal septum.
The bones of Berlin are also said to be developed from mem-
brane— the perichondrium of the presphenoidal cartilage, or the
walls of the olfactory sacs.

The development of the temporal bone is particularly worthy
of attentive consideration. The squamosal and the tympanic
elements are developed in membrane, and, at first, lie perfectly
loose in this membrane, upon the outer side of the periotic
cartilage. The tympanic is a delicate ring, open above; the
squamosal is a mere rod, the zygoma, with an expanded posterior

L 2

148 ON THE STRUCTUKE OF THE SKULL.

end, the squama temjioris. The periotic mass, the styloid ele-
ment, and the auditory ossicles are all preformed in cartilage.

The manner in which the cartilaginous capsule, which has
the form of the subsequent periotic bones and lodges the mem-
branous labyrinth, becomes ossified, has been much misunder-
stood ; and as it is a point of vital importance in comparing
the skull of man with that of the lower Vertebrata, I shall enter
into some detail regarding it, as a matter of fact and as a matter
of anatomical history.

Nearly two hundred years ago, Kerckringius, in his excellent
" Osteogenia Fcetuum" (1670), laid the foundation for a proper
understanding of this process : —

" Quarto mense mirum visu, quaru cito et quanta perfectione
os squamosum magnani partem factum sit osseum. Os petrosum
jam rubicunda cartilagine signavit cavitatis suae formam organo-
rum auditus capacem, nihil tamen adhuc prae se fert osseum,
praeterquaru unam in longitudinem protensum crassiusculam et
inaequalein lineam, annulo seu circulo, antea nominato, sub-
jectam, et paulo longius protensam. Os itaque temporum hoc
mense tribus constat ossiculis ; annulo scilicet, osse squamoso,
et illo jam commemorato.

" Quinto mense os squamosum ita adauctum est ut os synci-
pitis fere, os autem cuneiform e omnino attigerit. Ossis petrosi
pars ilia quae processum mammillarem constituit, terna de novo
acquisivit ossicula : unum pyriforma, acutiore, sui parte squa-
moso annectitur ; alterum, scutum ovale referens, magnitudine
priori vix cedens, media oartilagme ab eo separatur ; uti et
tertium ab utroque, quamvis hoc magnitudine neutri sit aequipa-
randum, vix aciculae majoris caput adaequans; sunt autem eo
situ et ordine collocata, quern tabula foetus v. mensium, usurpata
oculis facilius ad mentern quam verba transmittet." ....

" Constat ergo os temporum hoc quinto mense sex clistinctis
ossiculis ; osse videlicet squamoso, annulo, osse internam cavi-
tatem efformante et tribus notabilibus quae hoc mense exorta
esse diximus." — L. c, pp. 222, 223.

The explanation of the third figure in the thirty-fifth plate, referred to in this pas-
sage, runs thus : — " Tria in osse petroso ossicula ostendit, e, e, e. Tria petrosi
ossis distinct;! ossic-ula."

THE DEVELOPMENT OF THE HUMAN SKULL. 149

" Sexto mense pyriforme et ovale scutiforme coaluerunt in
unura , tertium nonnihil auctum est magnitudine." — L. c, p.
224.

The third figure of the thirty-sixth plate exhibits the condition thus described, and

the explanation is :— " Bina in osse petroso ossicula ostendit. D, ossis petrosi
pars quae jam ex duobus coaluit ; c, tertium ossis petrosi ossiculum."

" Septiino mense jam tertium illud ossiculum cluobns mense
superiore inter se coalitis accessit

" Nihil ergo de mense octavo nonoque addendum, nisi quod
ne turn quidem foetus ullum habeat processum mamillarem, et
quod adhuc insigni cartilagine distet os petrosum ab occipitis
et syncipitis ossibus." — L. c, p. 224.

The temporal hone of a seven months' foetus is represented in Plate xxxvii., Fig. 2,
with the explanation : — " Qua? primo tria, deinde bina, fuerunt in petroso ossi-
cula, jam in unum coaluisse, ostendit. C, ossis petrosi substantia, ex tribus
jam ssepe dictis in unum eoalita."

Cassebohm (" Tractatus quatuor de Aure Humana," 1734,
pp. 19 and 45 ; " Tractatus Quintus," 1735, p. 15) discovered that
the little linear ossification mentioned in the first extract from
Kerckringius is developed in the immediate vicinity of the
fenestra rotunda, eventually surrounds it, and extends upon the
base of the pars jpetrosa. But the first definite light thrown
upon the signification of Kerckringius' " Tria ossa ': is in the
following extract from Meckel's " Handbuch der Vergleichenden
Anatomie" (1820. Bd. iv., p. 49), though Meckel does not take
the trouble to refer to and explain the older observer's state-
ments : —

" 4. Bony lahyrinth. — In investigating the formation of the
bony labyrinth, the origin of the bony substance of the petrous
bone is very carefully to be distinguished from that of the
labyrinth itself. The former begins earlier than the latter,
according to the ordinary type of ossification, by the develop-
ment of a loose, soft, reticulated tissue in the previously existing
homogeneous cartilage, and extends from before backwards.

" The first part to ossify, about the end of the third month, is
the circumference of the fenestra rotunda, which is remarkable by
reason of the analogy of the fenestra rotunda to the tympanic

150 ON THE STRUCTURE OF THE SKULL.

annulus. The ossification begins above, descends posteriorly,
and, after a ring has been formed in this manner, extends
forwards.

" At the same time arises a proper centre of ossification,
completely separated from this, at the external end of the
superior vertical canal.

" After this, a third little scale is produced, opposite about
the middle of the internal vertical semicircular canal.

" At the same time ossification extends swiftly backwards
and downwards from the first piece, so as to give rise to the
floor of the labyrinth.

" The second piece increases in size still more rapidly, so
that, soon, the whole vertical semicircular canal, with the ex-
ception of its lower concave surface, is ossified. Simultane-
ously ossification is continued from its inner end over the inner
surface of the petrous bone, surrounds the internal auditory
meatus, penetrates into it, and so forms the floor of the cochlea.

"The horizontal semicircular canal begins to ossify, on its
outer side, in the fifth month, by elongation of the bone form-
ing the upper vertical semicircular canal: tins is continued
backwards, from without and belowT, round the horizontal semi-
circular canal. At least, I could discover no proper osseous
centre for this canal, and it seems merely to become inclosed by
the increase of the first and second."

All this is accurate, but, unfortunately, Meckel goes on to
say, at page 5 1 of the work cited, that " the osseous labyrinth is
at first perfectly separate from the bony mass of the petrous bone
which surrounds it, is developed earlier than it, and is provided
with quite a smooth surface, though the two lie close together ;"
and that " the bony labyrinth arises independently of the osseous
substance of the petrous bone."

How Meckel arrived at this conclusion I do not know ; but
it is certainly erroneous, and it has been the means of creating
a great deal of unsound speculation as to the ossified labyrinth
being a something distinct from the proper pars petrosa.

It is further singular that, in this passage, Meckel not only,
as I have said above, makes no reference to Kerckringius, but
that he does not attempt to refer the regions of the pars petrosa

THE DEVELOPMENT OF THE HUMAN SKULL. 151

and mastoidea to their separate origins. This is tlie more re-
markable as, in his well-known paper on the "Ossification of the
Vertebral and Cranial Bones" (Meckel's " Archiv," 1 815), p. 636,
he states expressly that the mastoid process arises from a special
(••'litre. Possibly the omission arose from Meckel's supposing
that the exterior of the periotic mass is developed distinctly from
the proper bony labyrinth.

Hallmann, in his well-known work, " Die Vergleichende
Osteologie des Schlafenbeins " (1837), does not cite the account
given by Meckel, and does not really improve upon the views of
Kerckringius.

" In man, after, in the first place, the squamosal and then
the annulus tijmpanicus are formed, the os petrosum and mastoi-
ds um is still a common cartilage, which fills, externally, the gap
between the squamosal, the parietal, the supra-occipital and the
ex-occipital. When, in the fourth month, the cochlea and a
part of the semicircular canals, viz., the upper canal and the
anterior crus of the external canal, already consist of porous
bony substance, while the ossification of the posterior canal
(and probably of the posterior crus of the external canal) has
not proceeded so far ; the pars mastoidea appears as a single or
double nodule of the size of a millet-seed, which is deposited
upon the arch of the posterior canal, contributes to its ossifica-
tion, and now soon spreads over the whole cartilage, the four
neighbouring bones growing towards it. In Nos. 2543 and 9420
of the Berlin Museum, the insertion of this nodule upon the
petrous bone is quite distinct. This osseous centre appears in
the dry skeleton as an oval nodule, which I could easily scratch
off without injuring the canals, which proves that it arises as a
separate part."

Lastly, Kolliker, in his recently published " Entwickelungs
Geschichte" (1861), sums up the present state of our know-
ledge respecting the ossification of the periotic cartilage as
follows (p. 320) :—

"The ossification of the labyrinth does not appear to have
been investigated since the time of Cassebohm (' Tract, de Aure
Hum.,' Hal. et Magdeb., 1734 and 1735) and J. Fr. Meckel
('Handb. d'Anat,' iv. p. 42, et seq.), which seems to be the

152 ON THE STRUCTURE OF THE SKULL.

reason why certain incorrect statements are repeated year after
year in almost all handbooks. It is not the case that the
external part of the pyramid of the petrosal bone and the
labyrinth ossify separately, nor is it true that ossification begins
as a thin crust on the Avail of the labyrinth ; on the contrary,
ossification commences in the whole thickness of the wall of the
labvrinth ; in such a manner, however, that it appears exter-
nally sooner than internally, and the whole pyramid becomes
ossified from centres which make their appearance first upon the
cartilaginous semicircular canals and the cochlea. The number
of these is, as has been rightly stated, three — one on the first
turn of the cochlea, and one on each of the upper and posterior
semicircular canals, whence, by degrees, the whole pars petrosa,
together with the cartilaginous pars mastoidea, which is united
with it, ossifies in a manner, the details of which would not
especially interest you. On the other hand, I do not agree
in the statements that have been made as to the time at which
this ossification arises. ^Neither in the third, nor in the fourth
month, as is commonly stated, is there a trace of ossification ;
in fact, I have found the entire pyramid cartilaginous in an
embryo five inches long at the eighteenth week, or, in the
middle of the fifth month. Only at the end of the fifth, and
especially in the sixth month, do the osseous deposits commence,
but these increase very rapidly. In the sixth month, however,
one meets with nothing: but a beautiful reticulated cartilage
ossification, and, as yet, no indication of true bone, which onlv
arises, in the later months, from the periosteum of the labyrinth
and from the external periosteum, whilst, contemporaneously,
the internal cartilage ossification is reabsorbed and is replaced
by a vascular true bone, which, by degrees, becomes finely
spongy. The Modiolus and Lamina spiralis, in the sixth month,
are still quite membranous, and only ossify at the end of fastal
life, without ever having been cartilaginous."

There is no doubt that the statement of Meckel, confirmed
by Kolliker, that the periotic cartilage ossifies from three
centres, is perfectly correct ; there is no doubt, further, as
Meckel, followed less clearly by Hallmann, has affirmed, that
one of these centres gives rise to the future mastoid process ;

THE DEVELOPMENT OF THE HUMAN SKULL. 1 T>3

but it is equally indubitable that Kerckringius' original state-
ment is true, and may be readily verified in the dry skulls
of foetuses of the age he mentions. The beautiful series of
human foetuses presented by Mr. MacMurdo, in the Museum
of the Royal College of Surgeons, enable one easily to reconcile
the teachings of the older and the later observers, when taken
in conjunction with the study of the same parts in wet prepa-
rations.

Fig. 61", A, represents the periotic capsule of a human foetus
live and a quarter inches long.

One ossification in the cartilage (Op.O.) is seen surrounding
the fenestra rotunda (F.B.), and extending a little way upon
the promontory. A second, very small, quadrate ossification
(Pr.O.) is situated at the outer end of the superior vertical
semicircular canal, and apparently extends into the carti-
laginous tegmen tympani. There is no other ossification in
the cartilage than these two. As the upper part of the periotic
mass in man answers to the front part, and as the lower part
corresponds to the hind part of the same mass in the majority
of the Vertebrata, I term the ossification on the superior vertical
semicircular canal the pro-otic bone, that on the cochlea the
opisthotic bone.

In some dry foetal skulls of this age the opisthotic ossifi-
cation only is seen, just as it is described by Kerckringius,
who seems not to have observed the pro-otic ossification at
this period.

The pro-otic ossification rapidly extends, as Meckel states, over
the superior vertical semicircular canal (see Fig. 59, A, p. 144),
and reaching its posterior end, it includes the front and upper
part of the posterior vertical canal ; while, from the outer end
of the anterior vertical canal, or the primitive centre, a mass
of bone extends backwards in the periotic cartilage and, in the
dry skull, appears conspicuously immediately behind the edge of
the squamosal. (Pr.O., Fig. 61, B.) This part of it is, in
fact, that one of the " tria ossicula " of which Kerckringius says,
" pyriforma, acutiore sui parte, squamoso annectitur."

The opisthotic ossification likewise extends backwards and,
its hinder extremity becoming apparent in the dry skull behind

154

ON THE .STRUCTURE OF THE SKULL.

the tympanic, is Kerckringius' ossicle, " vix aciculee majoris
caput adsequans." (Fig. 61, B, Op.O.)

Fig. 61.

F.R.

tfk Sf

dfi.o.

Fig. 61. — Development of the temporal bono.

A, from a foetus b\ inches long, showing the commencing pro-otic and opisthotic ossifica-

tion.

B, from a fetus 8| inches long. The ossification in the tegmen tympani is no longer visible

from without, but its continuation backwards over the superior, and part of the
posterior, vertical semicircular canal is visible behind the squamosal. The epiotic ossifi-
cation has made its appearance, and the hinder extremity of the opisthotic ossification
appears behind the tympanic as the " third ossicle " of Kerckringius.

C, from a foetus 10£ inches long, the " tria ossicula " beginning to unite into the pars

mastoidea.
1), from a foetus 10^ inches long, the tria ossicula anchylosed.
/•'./.'., foramen rotundum.

■, superior semicircular canal.

THE DEVELOPMENT OF THE HUMAN SKULL. 155

Lastly, the third ossicle, " scutum ovale referens," is that
developed upon the posterior part of the posterior vertical
semicircular canal, which gives rise to the mastoid process.
(Fig. 61, B, Ep.O.)

Thus, in a foetus between the fifth and sixth months, the
"pars mastoidea ': exhibits the appearance represented in
Fig. 61, B. Its upper part is cartilaginous, but its lower part
is occupied by the three " ossicula ' of Kerckringius, which
have now come into contact, and begun to unite, though their
primitive contours are perfectly distinct.

The "pars mastoidea" of human anatomy is therefore not
a single bone, but one, the " scutum ovale," combined with
parts of two others; and as the "scutum ovale" is certainly
the homologue of the bone I have termed Epiotic in the ovi-
parous Vertebrata* I propose to get rid of the confusing term
" mastoid " altogether, and to call the specially " mastoid " part
of the pars mastoidea, Epiotic.

Of the three periotic bones thus developed, the pro-otic
gives rise to most of the pars petrosa, which is visible in the
interior of the skull (Fig. 59, A), investing, as it does, the
roof of the cochlea, the superior, and part of the posterior,
vertical semicircular canals, the internal auditory meatus, and
forming the tegmen tympani. To it, in addition, is due the
upper half of the circumference of the fenestra ovalis, and a con-
siderable portion of the pars mastoidea, as has been stated above.

The opisthotic bone constitutes all the pars petrosa visible
on the base of the skull, furnishes the floor of the cochlea,
surrounds the fenestra rotunda, and contributes half the contour
of the fenestra ovalis ; gives rise to the carotid canal by develop-
ing a lamella of bone, which gradually wraps itself round the
carotid, and so converts the primitive groove for the vessel
into a complete tube, at the same time furnishing the inner
part of its floor to the tympanum.

The lower edge of the squamosal is at first nearly straight,

* Croonian Lecture. Proceedings of the Royal Society, 1858. In the absence of
a sufficient knowledge of the development of the human temporal bone, I followed
Hallmann in identifying the opisthotic of oviparous vertebrates with the mastoid
of Mammals at the lime this Lecture was delivered.

156 ON THE STRUCTURE OF THE SKULL.

but it soon sends a curved process downwards behind the
auditory meatus and between the tympanic ring and the
periotic bones. In the foetal skull represented in Fig. 61, D,
it is obvious that this process corresponds with the Margo tym-
panicus or post-auditory process of the adult temporal bone ; and
the manner in which the hinder end of the pro-otic ossification
is fitted in between it and the representative of the ascending
part of the posterior root of the zygoma is very well shown.

The tympanic bone is at first a delicate ring, interrupted
above, and with tapering ends, wrhich approach one another
very closely. The anterior end is thicker than the posterior,
however, exhibiting a sort of flange, or internal process, which
corresponds in position with the middle root of the zygoma, and
eventually unites with it. The lower arched part of the tym-
panic ring becomes anchylosed with the floor of the tympanum,
while its posterior and upper end unites with the squamosal.

In the process of ossification thus commenced and advancing
in the foetal cranium, certain centres, at first distinct, unite, and
become hard to distinguish from one another even before birth.

At this period a considerable interval of cartilage separates
the basi-occipital from the basi-sphenoid ; but the latter has,
as at a, Fig. 62, A, become firmly united with the presphenoid,
though traces of the original separation, and remains of the
primitive cartilage, are readily discernible.

The ex-occipitals are still distinct from the supra- and basi-
occipital, and the alisphenoids are only suturally united with
the lingiilee splienoidales, which are still large in comparison
with the basi-sphenoid, though they very early unite with
it. The orbito-sphenoid and the presphenoid are completely
anchylosed together by the superior root of the former, but the
inferior root of the orbito-sphenoid, or middle clinoid process,
abuts against the basi-sphenoid. (Fig. 63.)

In the temporal bone — the pro-otic, opisthotic, and epiotic
are indistinguishably united into the pars petrosa and pars
mastoidea. The latter and the squamosal are firmly united,
but the petro-squamosal suture between the tegmen tympani
of the former and the squamosal bone is obvious. The tym-

THE DEVELOPMENT OF THE HUMAN SKULL.

157

panic bono, still little more than a mere ring, is firmly ancliy-
losed with the squamosal and with the opisthotic portion of the
jmrs petrosa, but the indication of the primitive distinctness of
the two latter can be readily traced. (Fig. 62, C.)

Fig. 62.

Pmx-

Vo. a B£. B.O.

BO. S.O.

Fig. 62. — The human cranium at birth. — A, vertical and longitudinal section of the basal
hall' of the cranium ; B, upper, and C, under, view of the same preparation.

It is only after birth, and with the gradual advance towards
adult years, that the spheno-occipital and the spheno-ethmoid
synchondroses are obliterated, and the vomer becoming anchy-
losed with the ethmoid, the whole cranio-facial axis is fused
into one bone, to which the ex-occipitals and supra-occipital,

158

ON THE STRUCTURE OF THE SKULL.

the alisphenoids and orbito-sphenoids, add themselves by a
similar obliteration of the primitive separations. By addition
of bony matter to its free margin, more especially to that
of its lower part, the tympanic bone becomes converted into
the gutter-like external auditory meatus. The epiotic grows
out, inferiorly, into the mastoid process. The cavity beneath
the bony arch in which the superior vertical semicircular canal
is lodged, at first filled only by a plug of dura mater, becomes
obliterated by bone.

Fig. 63.

JP.S.

■gXi! '■■

\ s — ~^ ' ■■ '- "'

B.S.

BS^i

Fig. 6.3. — The basi-sphenoid and presphenoid, with the orbito-sphenoids of a human skull at
birth. — A, viewed from above; B, from the right side; C, from below. T.S., tuber-
culum sellce ; L.S., IAngula sphenoidalis ; a, basi-presphenoidal synchondrosis.

The basi-sphenoid acquires larger dimensions in proportion
to the lingulse sphenoidales, and the posterior clinoid processes,
at first cartilaginous, become completely ossified. The bones of
Bert in unite with the under-surface of the presphenoid, and the
latter becomes almost obliterated, or converted into a mere
vertical lamina of bone, by the extension of the olfactory
chambers backwards to give rise to the sphenoidal sinuses.

The lateral masses of the ethmoid become anchylosed with
the lamina perpendicularis, and form one bone — the ethmoid of
human anatomy.

THE DEVELOPMENT OF THE HUMAN BKULL. 159

Of the facial bones, the premaxilla is developed within that
part of the naso-frontal process which forms the anterior
boundary of the mouth. The maxilla, the palatine, and ptery-
goid bones are produced within the maxillary process — the
former from its external, the latter from its internal part. The
internal pterygoid is, even before birth, united with the external
pterygoid, the latter being simply an outgrowth downwards of the
alisphenoid. None of these bones are at present known to be
developed from cartilage, and the lachrymal and jugal are,
similarly, membrane bones:

The cartilaginous rods within the first and second visceral
arches undergo very remarkable changes. That of the first
arch becomes modified into an upper portion, the future incus,
and a lower portion, articulated with this, the future malleus,
from which the rest of the cartilage is continued, as " Meckel's
cartilage," along the inner side of the visceral arch (Fig. 64).

The ineudal and malleal portions of the cartilage are, at
first, proportionally very large, but their growth soon becomes
arrested, and, a centre of ossification appearing in each, they
become the incus and malleus. As the root of the first visceral
arch is close to the outer and front part of the periotic capsule,
so the incus and malleus have a corresponding position, and
the tympanic bone, which is developed around the circumfer-
ence of the modified first visceral cleft (which becomes con-
verted into the auditory meatus, the tympanum, and the
Eustachian tube), necessarily lies outside them, so that Meckel's
cartilage passes between the tympanic bone and the periotic
capsule, in its course from the malleus forwards and down-
wards. In front, the tympanic circlet marks the limit of its
ossification. So far, it constitutes the p'ocessus gracilis (Pg.,
Fig. 64), while, beyond this point, it eventually becomes ob-
literated. Very early, however, ossification takes place in the
membrane of the first visceral arch, adjacent to the middle
of the cartilage, and extending upwards towards the squamosal
bone and, downwards and inwards, towards the symphysis,
lavs the foundation for each ramus of the lower jaw. The
lower jaw, therefore, arises from membrane, and is not pre-
formed in cartilage.

160

ON THE STRUCTURE OF THE SKULL.

The axis of the second visceral arch becomes converted
above into the stapes, below into the styloid cartilage, the
stylohyoid ligament, and the lesser cornua of the hyoid bone,
the body and greater cornua of which are developed from
the third visceral arch. Between the styloid cartilage and

Fig. 64.

Fig. 64. — Part of the skull ot a human foetus at about the sixth month, dissected to show
the auditory ossicles and Meckel's cartilage, Mck. Gl, the glenoid cavity. The
margo tympanicus and adjacent parts of the squamosal are represented as if they were
transparent, in order to show the position of the malleus (m) and incus (ij. The
tympanic bone (Tty) is merely indicated. Co., the cochlea.

the stapes it is modified so as to form the stapedius muscle.
A centre of ossification appears in the styloid cartilage, and,
extending upwards and downwards, gives rise to the pyramid
and the styloid process.

Some authors, however, give a somewhat different account
of the metamorphoses of the cartilaginous axes of the first and
second visceral arches to that which I have detailed, and which
is based chiefly upon the researches of Meckel, Rathke, and
Eeichert. Thus Gunther,* while he agrees with Reichert
that the cartilaginous axis of the first visceral arch divides
into three portions, the uppermost of which (that which is
primitively connected with the skull) early disappears, while
the middle and lower become converted respectively into the
incus and the malleus with Meckel's cartilage, differs from

* Beobachtungen iiber die Entwickelung de.s Ochororgans bei Menschen unci
hoheren Saugetliu ion. 1842.

THE DEVELOPMENT OF THE HUMAN SKULL. 161

him regarding the origin of the stapes. According to this

writer : — " The middle division of the cartilaginous axis applies
itself to the vesicular cartilaginous labyrinth, and when it comes
into contact with the labyrinth, it sends out a small nodule,
which is received by a pit, the future fenestra ovalis." The
nodule grows out into a process, the lower part of which be-
coming bent on the upper, and eventually articulated, is con-
verted into the stapes, while from the upper part originates the
long process of the incus.

The auditory ossicles are at first altogether outside the
tympanic cavity ; and as the latter enlarges, its mucous mem-
brane is reflected around the ossicles. The deposit of osseous
matter for each ossicle commences in the perichondrium, and
the stapes has three ossific centres, independently of that for the
os oroiculare.

It can hardly be doubted that there is much yet to be
learned respecting the first steps in the development of the
ossicula auditus ; but the investigation is one fraught with
difficulties.*

* See Magitot et Robin, " Cartilage de Meckel." Ann. des Sc. Nat. Se. IVC.
tome xviii.

162

LECTURE IX.

ON THE STRUCTURE OF THE SKULL.

THE SKULL OF THE PIKE COMPAEED STRUCTURALLY AND
DEVELOPMENTALLY WITH THAT OF MAN.

In the series of animals possessing a bony skeleton, osseous
fishes and man mav be regarded as the extreme terms ; and
I now select the skull of an osseous fish — the Pike — for com-
parison with that of Man. Whatever community of structure
obtains between these must be expected to persist throughout
the intermediate terms ; while the differences between them
will be more or less completely bridged over by the subsequent
study of the skulls of the lower Mammals, Birds, Reptiles, and
Amphibians.

At first sight, the skull of a pike (Fig. 65) presents the most

Fig. 65.
S.Or

I Oh,
At An SxjPr.O/x -Bra '

Fig. 65. — Side view of the skull of a Pike (after Agassiz). — Prf, prefrontal ; H.M., hyo-
mandibular bone; Op., operculum ; S.Op., sub-operculum ; I.Op, inter-operculum ;
Pr.Op, pre-operculum ; Brg., branchiostegal rays; Sy, symplectic; Mb, meta-ptery-
goid ; PI, palato-pterygoid arch ; Qu., quad rat um ; Ar, articular ; An, angular ; J9,
dentary ; S.Or., supra-orbital bone.

TUK STRUCTURE OF THE PIKE'S SKULL. 163

striking dissimilarity to that of a man. The skull proper is
flattened, narrow, and elongated, its vertical height and transverse
diameter bring insignificant when compared with its antero-
posterior length, the predominance of which is due chiefly to
the disproportionate enlargement of the anterior half of the
cranio-facial axis, i. e., the presphenoidal and ethmo- vomerine
regions. The brain-case is relatively very small and much
depressed, instead of presenting the capacious dome of the
human skull, while, on the other hand, the facial apparatus is
very large and complex, and its components are almost all
moveable upon the skull. Another circumstance, which at
once strikes the observer, is the fact that the lower jaw is not,
as in Man, articulated directly with the skull ; but is con-
nected with the latter by the intermediation of a complex,
mobile, suspensorial apparatus (Fig. 65, H.M. to Qu.)} which
articulates with the skull above and with the lower jaw below.
A part of the same apparatus gives attachment to the hyoidean
arch, and to the bones of the gill covers.

A certain fundamental resemblance may, however, be readily
traced beneath these external differences. Thus, if a transverse
and vertical section be taken through the pike's skull, so as to
traverse the organ of hearing, and to divide the suspensorium
longitudinally into two parts, the posterior and anterior moieties
of the skull will present the appearances represented in Figs.
66 and 67. The posterior segment (Fig. 66) is obviously
comparable with the corresponding segment of the human skull
(Fig. 49), consisting, as it does, of a floor, with an upper arch,
which, in the recent state, inclosed part of the brain, and with
a lower arch formed by the various parts of the hyoidean
apparatus.

Furthermore, certain of the bones (Ep.O, Op.O, &c.) which
enter into the composition of the upper arch are especially
related, as in the corresponding section of the human skull, to
the organ of hearing, and it is with some of these that the
inferior arch is connected.

The anterior segment (Fig. 67) presents a similar general
correspondence with the corresponding segment (Fig. 48) of the
Man's skull. That is to say, there is a floor with which is con-

m 2

164

ON THE STRUCTURE OF THE SKULL.

nected an upper arch, forming part of the brain-case, and
a lower arch which enters into the composition of the face.
The sides of this arch in the sectional view are partly constituted
(compare Fig. 48) by bones specially connected with the audi-
tory apparatus, and the peduncle of the lower arch is articulated
with these bones. The chamber contained within the lower
arch, however, differs from that seen in the section of the
human skull in that it is entirely devoted to the buccal cavity,
and is not subdivided by processes of the palatine and maxillary
bones into an upper, nasal, and a lower, oral, passage.

Fig. 6G.

EpQjif.c. S.p Jis.o. Efi-O.

Fig. 66. — The posterior segment of the skull of a Pike which has been vertically and
transversely bisected. The bones of the inferior arch are represented diagrammati-
eally. The epiotic, opisthotic, pro-otic, and squamosal bones are left unshaded, as in
the corresponding section of the human skull, p.s.c, e.s.c, arrows indicating the
positions of the posterior and external semicircular canals; x, parasphenoid.

The comparison of the transverse sections of the Pike's with
those of the Man's skull thus enables us to perceive certain
resemblances between the two. In each there is an axis, upper
and lower arches ; in each the section separates the bones which
lodge the auditory organs ; and the most apparent difference
between the two is the vastly greater proportionate size of the
periotic bones in the Tike.

THE STRUCTURE OF THE PIKE'S SKULL. 165

The comparison of the longitudinal section of the Man's
skull with that of the Pike (Fig. 68) confirms the conclusions
arrived at from the study of the transverse sections. A "cranio-

Fig. 67.

» a.s'.c

Fig. 67. — The anterior segment of the skull represented in Fig. 66. — Mn, mandible.
a.s.c, arrow indicating the position of the anterior semicircular canal. The letters
B, 8, one on each side of the basi-sphenoid, are seen through the canal for the orbital
muscles. The pro-otic bone is left unshaded. In this, and in the preceding figure,
the dotted shading indicates cartilage ; but, as the drawings were made from a dry-
skull, it must be remembered that the whole of the cartilage entering into the
cranium is not represented.

facial axis," composed partly of bone and partly of cartilage,
extends from the occipital foramen to the anterior extremity of
the snout of the fish. The posterior part of this constitutes the
floor of the cranial cavity, and is the basi-cranial axis. The
anterior part, excluded from the cranial cavity, is, as in Man, the
basi-facial axis.

Again, as in Man, three pair of chambers, destined for the
lodgment of the organs of the higher senses, are placed sym-
metrically upon the sides of the Pike's skull. The olfactory
chambers are situated just in front of Prf., in Fig. 65, and the
orbits beneath S.Or., while the auditory organs are inclosed
within the posterior bony walls of the brain-case, as indicated
in the transverse sections. And, as in Man, the olfactory and

166 ON THE STRUCTURE OF THE SKULL.

auditory apparatuses are fixed within their chambers, while the
eye is freely moveable within the orbit.

Thus, for the Pike, I may repeat the phraseology which I
employed in giving a general description of the skull of Man.
It consists of an axis, of upper and lower arches, and of chambers
for the sensorv organs.

The next point is to ascertain how far this correspondence,
thus traced generally, extends into the details of the composition
of the skull ; and here we may conveniently begin, as before,
with the study of the cranio-facial axis.

Viewed as a whole, this axis is rounded and thick behind,
com pressed from side to side in the median region, and thickened
and depressed in front. It is composed, as I have said, partly
of bone and partly of cartilage. Behind, it consists of a single
well-ossified mass (B.O.), which offers, posteriorly, a deeply ex-
cavated conical articular facet, quite similar to that presented
by the body of the first vertebra, with which it articulates.
Anteriorly, it is also excavated in the middle, its conical cavity
terminating the canal for the orbital muscles behind. Its upper
face forms the hinder part of the floor of the cranial cavity and
the inferior boundary of the occipital foramen. Its lower face is
bevelled off in front, and articulates with the hinder part of the
upper face of the bone x, Fig. 68.

Laterally and posteriorly, it articulates with the bones (E. 0.),
which constitute the lateral boundaries of the occipital foramen ;
while, laterally and anteriorly, its deeply-excavated surface is
free, and forms part of the deep chamber in which the sacculus
of the auditory organ is lodged. The greater part of this bone
is solidly ossified throughout, but its conical anterior cavity is
lined by a thin shell of bone, which is separated by a continuous
layer of cartilage, thicker above than below, from the rest of the
osseous mass.

In a longitudinal section (Fig. 68) of a fresh Pike's skull, the
upper part of this layer of cartilage is readily seen, and can be
traced without interruption, from the axis of the bone under
description as far forwards as the posterior margin of the pitui-
tary fossa, and therefore, for a long distance in front of the
anterior termination of the bone B.O. The layer of cartilage

THK STRUCTURE OF THE PIKES SKULL.

16r

bends down at the sides, and so enters into the lateral Avails of
the cavity for the orbital muscles. The cartilage, however, does
not immediately constitute the floor of the skull, or the roof
and side walls of the canal for the orbital muscles, seeing that
it is coated over, on both its faces, by bony matter, which is con-
tinuous with that forming the inner and the outer faces of the
bone Pr.O.

Although there can be no doubt, then, that the cartilaginous
lamella in question forms part of the basi-cranial axis, it does
not, strictly speaking, form part of the floor of that skull, being
shut out therefrom by the extension over it of the ossifications
(Pr.O.), towards the middle line. Leaving these ossifications out
of consideration, however, it may be said that the free edge of the
middle part of the cartilaginous lamella forms the posterior
boundary of the fossa for the pituitary body, which dips down,
surrounded by membrane, through the centre of the canal for
the orbital muscles, and rests upon the concave surface of an
elevation of the bone x at P, Fig. QS. Immediately in front of

Fig. 68.

$fl. p\.(Y.

s.va

-~E.O.

B.O.

.^A

Tc.

Fig. 68.— Longitudinal and vertical section of a fresh Pike's skull. The cut surface ot
cartilage is dotted. For S. V.G., P. V.G., read a.s.c, p.s.c, as in Figs. 66 and 67.

this elevation cartilage reappears, and extends, as an inter-
orbital, ethmoidal, and internasal septum, to the end of the
snout. The cranial cavity rapidly narrows above the cartila-
ginous inter-orbital septum, and ends where the olfactory lobes
abut against the olfactory sacs. It appears to terminate much
sooner, however ; for the olfactory lobes, after running parallel
with one another for some distance, diverge, and become sepa-

168

ON THE STRUCTURE OF THE SKULL.

rated by a plate of cartilage, which corresponds to a certain
extent with the crista galli of the human skull.

Fig. 69.

Fig. 69. — Side and upper views of a Pike's skull (after Agassiz). — a, the articular facet
for the hyomandibular bone; x, the " parasphenoid ;" y, the true basi-sphenoid ;
z, the alisphenoid.

Immediately in front of the pituitary fossa a thin plate-like
ossification, y, is developed in the cartilage, and this plate sends
off backwards and a little upwards, upon each side, a process
which is connected posteriorly with the cranial floor. These
two processes consequently lie at the sides of the pituitary
fossa, and the "Y-shaped bone," as it has been well termed,
thus furnishes part of the front and side walls of that fossa.

The next ossification to be noted in the cranio-facial axis of
the Pike is the great bone x (Figs. 66 to 69), which stretches,
like a splint, along the greater part of the length of the base of
the skull.

The lower face of the hinder half of this bone is free, while
that of its front half is covered by the bone, Vo. The upper
face of its hinder half articulates, at first, with the lower surface
of B.O., but is then free for some distance, forming the floor of
the canal for the orbital muscles, and articulates by expanded
aliform processes of its sides with the lateral walls of that canal-
At the front part of the canal it exhibits the elevation which
forms the floor of the pituitary fossa, and then, depressed at the
sides, but exhibiting a median superior ridge, it underlies \\u-
inter-orbital and ethmoidal cartilages.

THE STRUCTURE OF THE PIKE'S SKULL. 169

The last ossification of the craniofacial axis is a depressed
bone Vo, thicker in front than behind, which fits on to the under-
surface of the anterior half of the bone just described, and ex-
tends beyond it to the front end of the snout. The under-surface
of this bone is free, enters into the middle of the roof of the
palate, and bears teeth.

In comparing the cranio-facial axis of the Pike with that of
Man, two pair of bones appear, at once, to correspond so closely
that no reasonable doubt can be entertained as to their homology.
These are the posterior and anterior bones of the series in each
case. The former, in its relation to the spinal column, to the
medulla oblongata, and to the lateral arches of the skull of the
Pike, is precisely comparable with the basi-occipital of Man ;
while the anterior bone as exactly answers to the vomer of man ;
except that the fish, being devoid of any communication between
the olfactory chambers and the cavity of the mo7dh, the vomer
has a different form, and has of course no relation to nasal
passages.

Again, it seems obvious that the ethmoid is represented only
by cartilage, as in the foetal state of the human skull, for there
is no ossification in that portion of the cranio-facial axis which
lies between the olfactory sacs.

And the like appears to be true of the presphenoid, for all
that vertical plate-like portion of the cranio-facial axis which
lies between the orbits, and beneath the peduncles of the
olfactory lobes, and in front of the crossing of the optic nerves,
is merely cartilaginous.

The Y-shaped bone forms part of the front and side walls
of the pituitary fossa, and its upper prolongations are connected
behind with the bones Pr.O, and with the floor of the cranial
cavity. In this floor, the long cartilaginous plate, already men-
tioned, constitutes the hinder boundary of the fossa, and sepa-
rates the Y-shaped bone from the basi-occipital. Now, the
proper basi-sphenoid (that is to say, the central ossification
taken apart from the linguist) forms the front boundary of the
pituitary fossa in Man, but extends obliquely downwards in
front of it as the stem of the Y-shaped bone does in the Pike.

170 ON THE STRUCTURE OF THE SKULL.

Furthermore, in the foetal human skull, the basi-sphenoid con-
tributes nothing towards the posterior boundary of the pituitary
fossa, which is formed by the long cartilaginous synchondrosis
which connects the rudimentary basi-sphenoid with the basi-
occipital. I identify the lamella of cartilage which I have
described in the Pike with this synchondrosal portion of the
foetal human basis cranii.

But the basi-occipital, basi-sphenoid, presphenoid, ethmoid,
and vomer of Man being now accounted for in the Pike's cranio-
facial axis, what, it may be said, is the nature of the bone x to
which the term "basi-sphenoid " is commonly applied?

It differs from any of the ossifications of the basi-sphenoidal
cartilage in Man, not only by extending backwards beneath
the basi-occipital, but by stretching forwards, beneath the pre-
sphenoidal and ethmoidal cartilages, to within a short distance of
the anterior extremity of the cranium ; and in the still more
important chcumstance that it is an ossification within the
perichondrium, which can be stripped off, in skulls which have
been macerated, or steeped for a short time in boiling water,
without injury to the cartilage upon which it is developed.

Mr. Parker has shown, in his valuable paper on Balseniceps*
that the so-called basi-sphenoid of birds is developed from
three ossifications, a central one, the true basi-sphenoid, and two
lateral and inferior centres — the "basi-temporals" (Parker),
which appear to correspond with the lingulse of Man.

The thought readily arises that the single bone x may cor-
respond with these two basi-temporal ossifications. The latter,
however, appear to be cartilage ossifications like the lingulse
themselves : and, upon the whole, I think it will be safer, at any
rate for the present, to regard the bone x as peculiar to the
branchiate Vertebrata, and to confer upon it the special name
of " para&phenoid."

Connected with the bones of the basi-cranial axis are upper
arches, and, as in Man, the liindermost of these arches consists
of three elements, two of which are lateral and one superior.
Each lateral bone articulates below with the basi-sphenoid, and
forms the lateral boundary of the occipital foramen. Above, it

* Transactions of the Zoological Society, vol. iv.

THE STPUCTUUE OF THE PIKE S SKULL.

171

unites with its fellow for a short distance, and so constitutes the
upper boundary of that foramen, shutting out the superior bone
from any share in its formation. Except in this point, it will be
observed that the three bones quite correspond with the ex-occi-
pitals and supra-occipital (S.O.) of Man. The ex-occipital (E.O.),
however, further differs from that of Man in that it is perforated
and not merely notched, by the foramen for the eighth pair, and
that it is produced in front of, and external to, this foramen, so
as to enter largely into the chamber which lodges the lower and
posterior part of the organ of hearing. Furthermore, there is no
perforation for any hypoglossal nerve, that nerve not being
represented in a distinct form in the fish.

Fig. 70. — The basal and lateral bones of the skull of a Pike viewed from above. The squa-
mosal and the three periotic bones are left unshaded. P, the pituitary fossa.

In the Pike, no bony wall separates the membranous laby-
rinth from the cavity of the skull, the periotic ossifications being
all situated, as they are when they first appear in Man, upon the
outer side of the capsule of the labyrinth : and this capsule is
still less complete than that of the human foetus, seeing that
its inner wall is not even cartilaginous, but remains in the
condition of membrane. Notwithstanding the comparatively
incomplete condition of the periotic bones of the fish, however.

172 ON THE STRUCTURE OF THE SKULL.

they are as clearly and distinctly identifiable as any bones of
the skull.

The large bone, Pr.O., which occupies a great part of the
wall of the cranial cavity, in front of the ex-occipital, has its
front margin deeply notched for the exit of the third division
of the trigeminal. It presents a foramen through which the
portio dura passes ; it protects the anterior part of the vestibule
and the anterior vertical semicircular canal. In other words,
it has exactly those relations to the auditory organ and to the
cranial nerves which especially characterise the pro-otic ossi-
fication of Man — which, it will be recollected, also protects the
anterior part of the organ of hearing, lies behind the exit of the
third division of the trigeminal, and is perforated by the portio
dura.

In minor respects, on the other hand, the pro-otic of the
Pike differs from that of Man ; as, for example, in its vast pro-
portional size ; in its remaining distinct from the other periotic
bones ; in the wide ossification which extends from it over the
basal cartilage of the skull towards the middle line ; and in
coming into contact with the ex-occipital and basi-occipital
behind. But none of these modifications really interfere with
the homology of the bone — which we shall find to be identi-
fiable by the same essential characters throughout the vertebrate
series.

The epiotic element is not less distinctly recognisable. The
upper and lower crura of the great posterior vertical semi-
circular canal traverse notches in the supra-occipital and ex-
occipital respectively, but the summit of the arch of the canal
is inclosed within a distinct conical ossification, the " external
occipital " of Cuvier. This ossification is, in fact, perched upon
the posterior vertical semicircular canal, just as the human
epiotic ossification is perched upon the summit of the arch of
the same canal when it first appears ; and if the semicircular
canals of man were to grow in the same proportion to the brain-
case as those of the fish, the epiotic would be carried out as far,
and would .leave a considerable space between the pro-otic and
itself, into which the adjacent supra-occipital and ex-occipital
bones might intrude, as they do in the fish.

THE STRUCTURE OF THE PIKE'S SKULL. 173

The third of the periotic ossifications — the opisthotic — seems
at first not to be discoverable in the Pike. But in some speci-
mens of this fish, and in a great many other fishes, there is a
distinct bone (which is particularly large and conspicuous in the
Gadidce) connected below and behind with the ex-occipital, in
front with the pro-otic, internally and behind with the epiotic,
and externally with the squamosal. It enters especially into the
outer and posterior wall of the labyrinth, and protects a great
part of the external semicircular canal, sharing this function
with the pro-otic.

As there is neither fenestra ovalis nor rotunda in the fish,
and as it is by no means certain whether the sacculus does or
does not correspond with the cochlea of the higher Vertebrata,
some of the best marks by which the opisthotic may be identi-
fied are wanting ; but the relations of this bone to the other
periotic ossifications seem to me to be decisive as to its real
nature.

The periotic bones being thus identified, they are all elimi-
nated from comparison with the proper supero-lateral consti-
tuents of the cranial arches. And there remains only one bone
in the lateral walls of the Pike's cranial cavity which can answer
to any of these, which is that marked z in Figure QS, A.S. in the
other figures.

This bone has the form of a triangle, with its apex turned
downwards. The hinder side of the triangle abuts against the
anterior margin of the pro-otic, and closes the trigeminal notch
in that bone anteriorly. The front margin ends in the cartila-
ginous side walls of the skull ; the apex approaches, but stops a
little short of, the lateral wing of the Y-shaped bone, or basi-
sphenoid. The relations of this bone are therefore essentially
those of the alisphenoid, though I think it quite possible that
the orbito-sphenoicl may, to a certain extent, be represented by
its anterior portion.

The bones which enter into the roof of the skull (Fig. 69)
remain for consideration. Of these, the supra-occipital, which
has no additional constituent comparable to 8.0' in Man, has
already been mentioned. It articulates in front with two very
large and long bones (Fr), separated by a median suture, which

174 ON THE STRUCTURE OF THE SKULL.

narrow in front of the orbits and end in a point beyond the
nostrils, and bnt a short distance from the extremity of the
snout. The supra-occipital lies between, and separates two other
comparatively small and insignificant bones (Pa.), which are
situated between the posterior edges of the parietals and the
epiotics, and, strictly speaking, do not enter into the roof of the
cranial cavity at all. Of these two pairs of bones, the anterior
represent the frontals of Man, and the posterior his parietals.
The position and proportions of the bones are, indeed, remark-
ably altered ; but we shall find by and by, that these very vari-
able cranial elements undergo almost as great changes of propor-
tion and relation even within the limits of the Mammalian class.

The three bones which correspond with the pars petrosa and
pars mastoidea have already been identified. In Man another
element, the squamosal, situated above and external to the
pro-otic and opisthotic enters into the composition of the tem-
poral bone. In the Pike there is a corresponding bone, which
forms the external and posterior angle of the skull, and lies
above and external to the pro-otic and opisthotic, being usually
anchvlosed with the latter. The under and outer surface of this
squamosal bone contributes towards the formation of the articular
facet for the suspensory apparatus of the lower jaw. There
appears to be no ossification in the ethmoidal cartilage, which
answers to the lamina perpendicular is of the ethmoid. But,
separating the orbits from the nasal chambers, there is on each
side of the frontals, and partially overlapped by them, a bone
which helps to bound the hinder wall of the nasal chamber,
which lies external to the olfactory nerve, and which is in imme-
diate relation w7ith the nasal division of the trigeminal nerve.
This is the bone termed " pre-frontal " by Cuvier, and it obviously
corresponds with the lateral mass of the ethmoid in Man, which,
in like manner, enters into the wall of the olfactory chamber, lies
external to the olfactory nerves, and is traversed by the nasal
division of the fifth.

Thus far the bones entering into the composition of the Pike's
cranium (with the exception of the " parasphenoid ") have been
identified without much difficulty with those met with in Man.
Put there remain several others which seem to be without human

THE STRUCTURE OF THE PIKE'S SKULL. 175

homologues. These are, firstly, the bones called post-frontal,
Pt.f., which form the posterior, superior, and external angles of
the orbits, and are wedged in between the alisphenoids and the
pro-otics ; secondly, the bones marked 1 and 2, developed upon
the ethmoidal cartilage external to the points of the frontals.
The pair 1,1, which immediately overhang the external nares,
are probably to be regarded as the nasals of Man ; but the nature
of the second pair, 2,2, which lie internal to them, and extend to
the end of the snout, is doubtful. Still less does there appear
any reason to identify the bones 3,3, which are minute tri-
angular ossifications in the substance of the cartilage between
the bones 2,2 and the vomer, with any which exist in man. I
consider them to be peculiar to the fish.

And now to sum up, in a few words, the structure of the
brain-case of the Fish. We find, as in Man, a posterior, occi-
pital, segment, consisting of basi-occipital, ex-occipital, and supra-
occipital ; a middle, parietal, segment, consisting, as in Man, of
a basi-sphenoid, alisphenoids, and parietals, but in which the
latter, in consequence of the disproportionate size of the frontals,
are thrown far back out of connection with the alisphenoids ;
and, finally, an anterior, or, frontal, segment, of which only the
frontals are separately distinguishable in the osseous state.
The orbito-sphenoids and the presphenoid are alike represented
only by cartilage and membrane, unless, indeed, as has been
suggested, a part of the alisphenoid may take the place of the
former bones.

Of the bony elements connected with the sense-capsules in
Man, the pro-otic, opisthotic, and epiotic, together with the
squamosal, have been clearly identified in the Pike ; as have the
pre-frontals and the vomer.

But certain bones present in Man have not been recognised
in the Fish ; while, on the other hand, certain bones present in
the Fish appear to have no representatives in Man.

Thus, while the study of the cranial structure of the Man
and the Pike reveals a fundamental identity of composition
between the two, it demonstrates the existence of a no less
marked diversity, each type exhibiting structures and combina-
tions peculiar to itself.

176 ON THE STRUCTURE OF THE SKULL.

The principal bones which surround the oral cavity in the
Pike are disposed, as in Man, in pairs, some being in front of
and above the oral aperture, while others are behind and below
that aperture; and they inclose the buccal and pharyngeal
chambers.

Fig. 71.

KM. /

Fjcr. 71. — Palato-quadrate arch, with the hyomandibnlar and symplectic, the articular
piece of the lower jaw, and Meckel's cartilage, of the Pike, seen from the inner side.
a, the cartilage interposed between the hyomandibnlar and the symplectic; b, that
which serves as a pedicle for the pterygopalatine arch ; c, process of the hyoman-
dibular with which the operculum articulates ; d, head of the hyomandibular which
articulates with the side of the skull.

The anterior pair of pre-oral bones (Pmx. Fig. 65), small, and
beset with teeth upon their under-surfaces, are connected with the
vomer and the termination of the cartilaginous rostrum formed
by the internasal septum. They obviously answer to the human
pre-maxillse. An elongated bone (Mx.), which bears no teeth,
is connected anteriorly with the pre-m axilla, and, lying external
to the other pre-oral bones, forms the boundary of the gape.
Its homology with the maxilla of Man appears to be unquestion-
able. A second smaller bone is connected with the posterior
part of the upper edge of the maxilla, and is usually regarded
as a subdivision of it.

Behind the pre-maxillse, and internal to the maxillae, in the
situation occupied by the palatine and pterygoid bones in Man.
the Pike has an osseous arch of much greater complexity and
somewhat different connections.

The summit of this " palato-quadrate '; arch is moveably
articulated, by a cartilaginous pedicle, with the outer surface of
the pre-frontal process of the skull. The anterior crus of the

THE STRUCTURE OF THE PIKE'S SKULL. 177

arch stretches forwards, parallel with the vomer, to the pre-
maxilla; its posterior eras extends backwards, and, spreading
out, ends in an upper (Mpt.) and a lower (Qu.) prolongation.

Five bones enter into the composition of the arch — two
median, two posterior, and one anterior. The median bones are
so disposed that their anterior ends embrace the lower part of
the cartilaginous pedicle (b), the one lying more external, and
the other internal, to the latter. The more external has been
called " ecto pterygoid. ," '* It exhibits a short, ascending process,
running up towards the pre-frontal, and strengthening the carti-
laginous pedicle ; an anterior process which articulates with the
bone PL ; and a posterior arched prolongation, which descends in
front of the bone Qu., and articulates with its anterior edge.
The internal bone, called " entopterygoid ' (EjiL), is a nearly
straight, flat bone, the anterior half of which is applied, like a
splint, to the inner face of the ectopterygoid, while its broader
posterior face is similarly adjusted, above, to the bone Mpt, and,
below, to the bone Qu. The two last-mentioned " posterior ':
bones of the arch are termed respectively the " metapterygoid '
and the " quadrate " bones. The former is a broad, four-sided
bone, convex upon its inner surface, which presents a raised,
curved ridge, beneath which the entopterygoid is received
anteriorly and the hyomandibular posteriorly. It is connected
below with the quadrate bone, and, behind, it overlaps the hyo-
mandibular (H.M.) and the symplectic (Sy.).

The os quadratum {Qu.), so termed, not on account of its
form, which is triangular, but by reason of its identity with a
bone called by the same name in Birds and Keptiles, presents
inferiorly an articular head, with an elongated articular surface,
convex from before backwards, for the lower jaw. On the
posterior part of its inner surface it has a deep groove, directed
from above downwards, very nearly parallel to its posterior edge.
Into this the symplectic is received.

The anterior bone (Pa.), lastly, is elongated and flattened,
and bears teeth upon its lower surface. It is received ante-

* Not to multiply names unnecessarily, T adopt this term, which involves no
theoretical implications. It must be carefully borne in mind, however, that this
"ectopterygoid " has nothing to do with the " external pterygoid" process of Man.

178 ON THE STRUCTURE OF THE SKULL.

riorly into the re-entering angle between the vomer and the
pre-maxilla and maxilla. It is usually regarded as the palatine
bone.

Before attempting to discuss the homologies of these several
constituents of the palato-quadrate arch in the fish, it will be
necessary to take into account the nature and arrangement of
its post-oral facial bones.

In Man, the post-oral bones are arranged in two arches — the
mandibular and the hyoidean ; or, more strictly speaking, since
the hyoidean arch is really composed of two, indicated by its
lesser and greater cornua, there are three post-oral arches.

In the Pike, the lower jaw forms a mandibular arch, obviously
corresponding in a general way with that of Man ; behind this
follows a second arch, answering to the styloid processes, stylo-
hyoid ligaments, and cornua minora of the human hyoid, but
with much greater masses of bone entering into its composition ;
and this is succeeded by no fewer than five other arches, the four
anterior of which, supporting the gill filaments, are termed the
" branchial arches" while the last pair, which cany no branchiae,
and are much smaller than the others, are called the " inferior
pharyngeal bones"

The symphysis of the lower jaw is formed by the ligamentous
union of two bones, which carry the inferior teeth of the Pike,
and correspond respectively w7ith the rami of the human man-
dible. But, besides these dentary bones (D, Fig. 65), each half of
the lower jaw of the Pike has two other constituents, which are
not represented in the human lowrer jaw. One of these is a
small bone, which forms the lower part of the angular process of
the jaw. It is termed the os annulare, or angular piece (An,
Fig. 65). The other is a large triangular bone, which fits in
between the dentary and the angular, and is termed the articular
(os articular e, Ar.), because on its upper surface it bears a con-
cave articular fossa, into which the condyle of the quadrate bone
is received (Fig. 71). The cartilage which partially forms the
walls of this fossa is continued into a long tapering rod, which
lies upon the inner surface of the articular and of the dentary,
and terminates in a point shortly before reaching the symphysis
(Mch., Fig. 71).

THE STRUCTURE OF THE TIKE'S SKULL. 170

Tin's rod of cartilage affords a safe basis upon which to found
a homological argumentation. For it most certainly corresponds
with Meckel's cartilage in the human foetus, and the dentary
bone lies outside it, in just the same way as the dentigerous
ramus of the human mandible lies outside Meckel's cartilage.
But the articular bone is an ossification in and around the proxi-
mal end of Meckel's cartilage in the Pike, just as the malleus is
an ossification in and around the proximal end of Meckel's car-
tilage in the human foetus ; and the os quadratum is related to
the os articular e of the fish in the same way as the incus is
related to the malleus.

Hence it is to be concluded, in the absence of any evidence
to the contrary, that the articular piece of the Pike's lower jaw
answers to the malleus, and the quadrate bone to the incus.

I am not aware that any evidence can be adduced against
this view ; but, on the other hand, the relations of the parts
thus identified to the portio dura of the seventh nerve, in Man
and in the Fish, seem to me to afford it much support.

The portio dura in the former perforates the pars petrosa,
and, after skirting the inner wall of the tympanum, external to
the labyrinth, leaves the skull by the stylo-mastoid foramen.
Before it does so, however, it gives off a recurrent branch, the
chorda tijmpani, which takes a very singular course — passing
between the pyramid, which is the upper end of the hyoidean
arch, and the tympanic bone, entering the tympanum, crossing
the auditory ossicles to make its way out at the front wall of the
tympanum, between the tympanic and the squamosal, then
uniting with the gustatory division of the trigeminal, and passing
down along the inner side of the ramus of the mandible with it,
until eventually it leaves it to become connected with the sub-
maxillary ganglion.

The principal portion of the portio dura, on the other hand,
makes its way out by the stylo-mastoid foramen, and is dis-
tributed to the facial muscles, some comparatively insignificant
branches only, being furnished to the levators of the hyoidean
apparatus and depressors of the lowrer jaw. But, as has been
already stated, the facial muscles, so important and largely de-
veloped in Man, become insignificant in the lower Vertebrates,

n 2

180 ON THE STRUCTURE OF THE SKULL.

and are not represented at all in the Fish. Hence, in the latter,
we- might expect to find only mandibular and hyoidean branches
of the portio dura corresponding with the chorda tympani on
the one hand, and the stylo-hyoidean and digastric branches, on
the other, in Man. And this is really the case. For the portio
dura of the Pike, which leaves the skull by a special foramen in
the pro-otic bone, traverses the hyoruandibular bone, and then
divides into two branches, one of which runs backwards to the
hyoidean arch ; while the other, directed forwards and down-
wards, passes to the inner side of the quadrate bone, and over
its articulation with the articulare to the inner surface of the
lower jaw, along which it runs to the extremity of the ramus.
This last branch is obviously the representative of the chorda,
tympani, and its relations to the quadrate and articular bones
are, it will be observed, very similar to those which the corre-
sponding nerve has to the incus and malleus in Man.

Holding fast, then, by this determination of the homologies
of the articulare and quadratum, what is the nature of the other
bones entering into the palato-quadrate arch ?

The metapterygoid may perhaps answer to the os orbiculare.
The manner of its connection with the quadrate (incus) suggests
this view, for which, however, I cannot pretend to offer any
positive proof.

That the other three bones answer in a general way to the
pterygopalatine bones of Man is certain. The pterygoid of
Man, it is true, is in no way connected with the incus, while both
bones EcpA and JEpt. are united with the quadratum. But this
is in reality no difficulty, for we shall find that, in the higher
oviparous Vertebrata, the os quadratum is very generally con-
nected with a bone which is universally admitted to correspond
with the pterygoid of Man.

Again, both the palatine and the pterygoid bones of Man
are articulated with the base of the skull, while the palato-
pterygoid arch of the Fish is not directly connected with any
of the basi-cranial bones ; but, in many of the higher Vertebrata,
the ptery go-palatine arch is almost as free of the base of the
skull as in the Fish.

No doubt, then, the palato-pterygoid bones of the Fish,

THE STRUCTURE OF THE PIKE'S SKULL. 181

taken together, answer to the palato-pterygoid bones of the Man ;
but it is a very difficult matter to identity the separate con-
stituents of the two arches.

One of the most striking features of the palatine bone, not
only in Man, but in the Vertebrata generally, is its articulation
with the pre-frontal, or lateral mass of the ethmoid. If, guided
by this character, we seek for the homologue of the palatine in
the Fish, the so-called " ectopterygoid ' alone satisfies the con-
ditions. But if this bone be the homologue of the true palatine,
the bone PL must be regarded as a dismemberment, or subdivi-
sion of the palatine,* and the entopterygoid will take the
place of the true pterygoid.

The palato-quadrate arch, with the lower jaw, is immediately
suspended to the skull only by the articulation of the carti-
laginous pedicle b (Fig. 71) with the pre-frontal, none of the
posterior elements of the arch being directly articulated with
the skull. They are indirectly united with the latter, however,
by two very remarkable bones, the Hyomandibular (H.M.) and
the Symplectic (Sy.).

The os hyomandibular e is a broad flattened bone, somewhat
constricted in the middle, and divided below into an anterior
and a posterior process. The upper convex edge of the bone
(d, Fig. 71) fits into an elongated, concave, glenoidal fossa
bounded by the squamosal, opisthotic, and pro-otic bones, and
swings freely therein, in a plane perpendicular to the longi-
tudinal axis of the skull. The large anterior inferior process
articulates by its anterior edge and outer face with the meta-
pterygoicl, while below it is united by a persistent synchondrosis
with the irregular styliform bone, the SympJedic, which is firmly
fitted into the groove already described upon the inner face of
the quadrate bone.

The connection thus established between the hyomandibular
and the symplectic, is strengthened externally by the firm ap-
position of a curved elongated bone, the Pre-operculum, to the
hyomandibular above and to the quadrate bone below.

* Looking upon Fa. and Ecpt. as one bone homologous with the palatine of
Man, it will he found that iu osseous Fishes the separation between them takes
place sometimes in front of the pre-frontal articulation, as in the Pike, sometimes
behind it, as in the Cod and most bony fishes.

182 ON THE STRUCTURE OF THE SKULL.

The hyoidean arch consists of two median bones, — an an-
terior, the " entoghssal" which supports the tongue ; and a
posterior, the " urohyal" Its lateral cornua are formed by
four bones, two small (basi-hyal), and two large (ejri and
cerato-liyals) on each side, the latter supporting the "bran-
chiostea'al rays'! on which the branchiosteoal membrane is
spread out ; and the upper of the two larger bones is con-
nected with the synchondrosis between the hyomandibular and
symplectic by a styliform bone — the stylo-hyal (Fig. 71, By.).
Thus, the hyomandibular may be regarded as common to the
mandibular and the hyoidean arches, supporting the former,
indirectly, by means of the symplectic, and the latter directly,
by means of the stylo-hyal.

The stylo-hyal very probably corresponds with the styloid
process and pyramid of Man, but it is difficult to find any very
sure footing for our interpretations beyond this point.

The manner in which the symplectic is connected, on the
one hand, with the representative of the incus, and, on the other
hand, with that of the styloid process and pyramid, is strongly
suggestive of a relation between this bone and the stapes. But
it must no less be admitted that similar arguments might be
used in favour of the stapedial character of the hyomandibular
bone, the articulation of which with the pro-otic and opisthotic
might be compared with the fitting in of the stapes into the
fenestra ovalis, which is bounded by these two bones ; or again,
plausible arguments might be brought forward in favour of the
view that the hyomandibular, at any rate, is a bone special to
fishes. At present, it may be well merely to indicate these
various possibilities, as the study of development has hardly
been carried sufficiently far to enable us to decide in favour of
one rather than of another.

Each of the four anterior branchial arches is composed of
four bones, and the branchial arches of opposite sides are united
by connecting cartilages and median ossifications. The anterior
or first arch, which corresponds with the greater conm of the
hyoid of Man, is fixed to the pro-otic bone, between the exits of
the trigeminal and the portio dura, by cartilage. The succeeding
arches have no osseous or cartilaginous representatives in Man.

THE STRUCTURE OF THE PIKE'S SKULL. 183

The branchiostegal rays attached to the epi-hyal and cerato-hyal
are in like case.

Three bones, the operculum {Op.), sub-opereulum (S.Op.), and
inter-operculum (I.Op.\ are developed within the membranous
gill-cover, and serve to strengthen it (Fig. 65). The operculum
is articulated with the posterior and inferior process of the
hyomandibular bone ; the inter-operculum is connected by liga-
ment with the angular piece of the jaw, the sub-operculum lies
between the two.

The gill-cover is developed from the outer surface of the
second visceral arch, and corresponds with the concha of the
ear in the human subject ; and as the latter part contains no
osseous elements, it is obviously in vain to seek for the homo-
logues of these bones in Man.

The pre-operculum, which, as I have stated above, binds
together the hyomandibular and the quadrate bone externally,
has been compared with the tympanic bone of Man, and the
position of the bone and its relations to the representatives
of the ossicula auditus are certainly not altogether unfavourable
to this view.

These are the most important bones in the Pike's skull, but
several yet remain for consideration.

Thus there is a small, oval, supra-orbital ossicle (S.Or.,
Fig. 65) attached to the outer margin of the frontal, above the
orbit, and an inverted arch of sub-orbital bones which bound the
orbital cavity externally and inferiorly. The sub-orbital series
consists of a large anterior bone, which lies beside the nasal, and
of five or six smaller bones, the hindermost of which is con-
nected with, or attached close to, the post-frontal.

Finally, in the Pike, a forked bone, the supra-seapula, sus-
pends the scapular arch to the apices of the squamosal and
epiotic bones. This bone, it need hardly be said, is without a
distinct osseous representative in Man.

The merely anatomical comparison of the facial bones of the
Pike with those of Man thus leads to a conclusion very similar to
that attained by the examination of the bones of the skull proper.
There is a certain identity of fundamental plan upon which

184 ON THE STRUCTURE OF THE SKULL.

special structural peculiarities are superadded in each case.
Both types of skull exhibit many bones in common, but in each
type, some of these bones acquire special arrangements and very
different relative magnitudes ; and each type exhibits bones
peculiar to itself, the number of those present in the Fish and
absent in the Man, being very much greater than of those pre-
sent in the Man and absent in the Fish. As might be expected,
the study of the development of the Fish's cranium brings out
into still stronger light the fundamental resemblances of its
structure with that of the higher Vertebrates. The primitive
groove makes its appearance on the blastoderm, and becomes
converted into a canal by the arching over and coalescence of
the dorsal laminae. The anterior part of the canal dilates and
becomes subdivided into cerebral vesicles. The notochord ap-
pears and terminates, in front, in a point behind the pituitary
body ; while round its apex, that bend of the primitive cranium
takes place which constitutes the cephalic flexure. The organs
of sense make their appearance in the same regions, and the
visceral arches and clefts are developed in the same way. But
a greater number of them appear, and the posterior ones,
instead of vanishing, give rise to the branchial skeleton and
branchial clefts. The mandible is developed in the first visceral
arch, and the hyoid apparatus in the second, as in Man ; but
the details of the mode of origin of the hvomandibular and sym-
plectic, of the palatine and maxillary apparatus, and of the
naso-frontal process, have not been as yet worked out with
sufficient thoroughness to enable us to determine with certainty
the homologies of all the resulting parts.

The cranium is at first wholly membranous, but after a time
it becomes partially chondrified in the same way as in the
higher Vertebrates (Fig. 72). Cartilage appears in the base of
the skull upon each side of the notochord, and surrounds the
great auditory capsules. Anteriorly it divides into two processes,
the trabecular cranii (Tr.), which separate so as to inclose the
pituitary fossa (P), and reunite, in front of it, to form the ethmo-
vomerine rostrum. From the floor of the skull, at the front and
lateral part of each auditory capsule, a cartilaginous process
(1IM.) is given off, and passing downwards and forwards ends in

THE DEVELOPMENT OF THE SKULLS OF FISHES.

185

a free styliform process, which lies parallel with, and is bound by
connective tissue to, the free hinder cms of an inverted arch of
cartilage, the anterior cms of which passes into the pre-frontal
region of the ethmo-vomerine cartilage. The centre of this

Fig. 72. — The cartilaginous and partially ossified crania of young Sticklebacks (Gaster-
osteus). — A, in a very early, B, in a more advanced condition, from above ; C,
viewed from the side; Ch., notochord ; Au., auditory capsules; P, pituitary fossa;
Tr, trabecular ; x. parasphenoid ; H.M., Sy.f Qu., indicate not only the bones, but
the pre-existing cartilages.

palato-quadrate arch is prolonged into a process {Qu.), which
articulates with the cartilaginous ramus of the mandible, while
the upper part of the cartilage (HM. Sy.) gives attachment to
the cartilaginous hyoid {Hy.).

This is the earliest condition of the cartilaginous cranium of
the osseous fish that has yet been observed ; but it can hardly
be doubted that the hyomandibular and palato-quadrate carti-
lages have already deviated considerably from their primitive
condition, and it would be a matter of great interest to ascertain
whether these cartilages are primitively continuous ; or whether,

186

ON THE STRUCTURE OF THE SKULL.

on the other hand, the hyomandibular altogether belongs to the
second visceral arch, while the hinder crus of the palato-qnadrate
belongs to the first, but has become detached from its primitive
connection with the basis cranii.

The basi-occipital originates as an ossification, which imme-
diately surrounds and incloses the end of the notochord, and
extends into the adjacent cartilage. The ex-occipital is de-
veloped within the substance of the cartilaginous cranium on
each side of the basi-occipital. The parasphenoid, on the other
hand, is developed as a superficial ossification in the perichon-
drium of the base of the skull, and extends in front of, and
behind, the pituitary fossa in this membrane. The pre-maxillae
and maxillae have no cartilaginous predecessors, nor have the
dentary and angular pieces of the lower jaw. The palatine is
developed around, if not in, the anterior crus of the palato-

Fig. 73.

AS. US Tr:()

Fig. 7."). — The cartilaginous cranium of a Pike, with its intrinsic ossifications viewed, A,
from above; B, from below; C, from the left side. N, N, nasal fossa;; I. Or,
inter-orbital septum ; a, groove for a median ridge of the parasphenoid ; b, canal for
the orbital muscles.

THE DEVELOPMENT OF THE SKULLS OF FISHES. 187

quadrate arch ; the metapterygoid in the same relation to its
posterior eras ; the quadrate bone, in its inferior process. The
symplectic is a cortical ossification of the styliform part of the
hyoinandibular cartilage, the ossification of the rest of the latter
giving rise to the hyoinandibular bone itself (Fig. 72).

In many osseous fishes, such as the Carp, the cartilaginous
cranium disappears, with age, as completely as it does in Man ;
but, in the Pike, it not only persists, but grows and enlarges
with age, so that the relations of the cranial bones to cartilage,
or to membrane, can be investigated at any period of life.

If the skull of an adult Pike be macerated, or, better, steeped
for a short time in boiling water, a number of the cranial bones
will separate with great ease from a sort of model of the skull
chiefly composed of cartilage.

This " cartilaginous skull " forms a complete roof over the
cranial cavity (Fig. 73, A), whence it is continued, without in-
terruption, to the anterior end of the cranium, forming the narrow
inter-orbital septum {I. Or.) and the broad internasal rostrum
(ML), and giving rise to two antorbital processes (Prf.), which
separate the orbits from the nasal chambers, and are perforated
by the olfactory nerve, and by the nasal division of the fifth.

The inter-orbital cartilage is interrupted by an oval space
filled with membrane, just in front of the basi-sphenoid, so that
it is continued to the lower end of that bone only by a slender
cartilaginous rod, which passes into the stem of the Y-shaped
basi-sphenoid (Fig. 73, C).

The cartilaginous basis of the skull, therefore, is not con-
tinued back along the floor of the canal for the orbital muscles.
The roof of the orbital canal contains cartilage in the middle
line, which is almost completely hidden in front by the ex-
tension towards one another of the horizontal laminae of the
pro-otic bones. The under-surface of the inter-orbital septum
and of the greater part of the cartilaginous rostrum is marked
by a deep groove (a, Fig. 73, B), into which a median ridge of
the parasphenoid is received.

The bones which, being developed in perichondrium, are
easily removed from the macerated skull, are the parietals, the
frontals, the bones 1.1. and 2.2. (Fig. 69), the squamosals (when

188 ON THE STRUCTURE OF THE SKULL.

these are not anchylosed with the opisthotic bones),* the vomer,
and the parasphenoid.

The bones which, as ossifications of the substance of the
cartilaginous cranium itself, are not thus separable, are the basi-,
ex-, and supra-occipitals, the three periotic bones, the ali-
sphenoids, the ba si-sphenoid, the post-frontals, the pre-frontals,
the bones 3.3 (Figs. 69 and 73).

Thus, in a certain sense, the adult skull of the Pike may be
said to represent, in a persistent form, a condition of the skull
which is transitory in 31 an.

Let the sides of the human foetal cartilaginous cranium grow
up and unite in the roof of the skull ; let the pre-sphenoidal,
ethmoidal, and internasal portions be greatly elongated ; let no
distinct ossification take place in the pre-sphenoidal and orbito-
sphenoidal regions, or in the part answering to the lamina
perpendicularis, while the basi-sphenoidal ossification remains
very small, and that cranium would put on the most important
and striking characters of that of the Pike.

* How far the bone which I have marked Sq. in the skulls of Fishes is really a
membrane bone and the homologue of the squamosal of Keptiles, Birds, and
Mammals, is a question which needs thorough re-investigation. Mr. Parker is of
opinion that it is really a cartilage bone and the homologue, not of the squamosal,
but of an independent ossification, which he finds well developed in the periotic
Capsule of the Mole and Shrew, and terms the "pterotic."

18!)

LECTURE X.

ON THE STRUCTURE OF THE SKULL.

THE SKULLS OF FISHES.

It Las been seen that the skull of Man and that of the Pike
agree in passing, in a similar order, through similar develop-
mental stages. Each, at first, is a membranous cranium, its
walls being composed of indifferent tissue, with the exception
of that small part of its base which is occupied by the notochord.
The greater part of the substance of each becomes chondrified,
and thus that cartilaginous cranium is produced, which is a
temporary structure in the Man, but a persistent one in the
Fish. Neither in the membranous, nor in the cartilaginous
state, does the cranium of either Man or Fish present any trace
of that segmentation which becomes obvious in the third con-
dition, when, by the deposit of calcareous salts around certain
centres, either in the cartilaginous cranium or the adjacent
membrane, the bony cranium is developed.

These three conditions of the skull are manifested, in the
same order of succession, by all vertebrate skulls which become
completely ossified ; but the crania of many vertebrated animals
remain throughout life in the second state, or in a condition
intermediate between that and the third, while the skull of one
of the Vertebrata persists in a state which can only be regarded
as a modification of the membranous cranium. Hence I shall
proceed to describe the leading modifications of the Vertebrate
Skull under these heads: — A. The membranous cranium. B.
The cartilaginous cranium. C. The cartilaginous cranium, with

190

ON THE STRUCTURE OF THE SKULL.

superadded membrane bones, but no cartilage bones. D. The
osseous cranium.

The three first-mentioned kinds of skull are met with only
among fishes; Amphibia, Beptilia, Aves, and Mammalia in-
variably possessing* a larger or smaller number of cranial bones
developed in cartilage.

A. The membranous cranium.

The only animal, at present known, which comes under
this category is that singular fish, the lowest of all Vertebrata,
Amphioxus lanceolatus (Fig. 74). The notoehord (Ch), sur-

Fig. 74.

Fig. 74. — Skull of Amphioxus lanceolatus (after Quatrefages). — a, Position of olfactory (?)
sac ; h, optic nerves ; c, fifth (?) pair ; d, spinal nerves ; c, representatives of neural
spines ; /, g, oral skeleton ; Ch, notoehord ; My, spinal chord, or myelon.

rounded by a merely membranous sheath, extends very nearly
to the anterior pointed extremity of the body. The myelon, or
spinal chord {My), occupies the ordinary position above it, in a
canal formed by upward processes of the membranous sheath,
and gives off the spinal nerves, d d, on each side. Quadrate
masses of somewhat denser tissue, e e, seem faintly to represent
neural spines. Just above the anterior boundary of the mouth,

THE SKULLS OF FISHES. 191

but far behind the .anterior end of the notochord, the myelon,
dilating very slightly, suddenly terminates, and with it, the
neural canal. The lateral muscles are divided into segments
corresponding with the pairs of spinal nerves, and the most
anterior of these segments is situated just behind the slightly
dilated chamber of the neural canal which contains the cor-
respondingly enlarged end of the nervous axis. The latter is all
that represents the brain, and the chamber is the skull.

A ciliated sac placed at b, in connection with the upper
surface of the brain, has been considered to be the olfactory
organ of this fish, but it is possible the sac may simply represent
the pineal body ; optic nerves (b) are given off to the rudimen-
tary eyes, and the branches (c) appear to be analogous in function
to the fifth pair. But no pituitary body has been recognised,
and, what is still more singular, there is no trace of auditory
sacs. A cartilaginous ring, provided with tentacular prolonga-
tions (/, g), surrounds the mouth, and there is a singular
branchial skeleton more like that of an Ascidian than any
ordinary vertebrate structure ; but neither of these structures
probably have anything to do with the true cranial or facial
skeleton.

It will be observed that this very remarkable skull, if it can
be properly so called, is not strictly comparable to an arrest of
development of a higher vertebrate skull ; the notochord ex-
tending far beyond the end of the cranium, which it never does
in any embryonic condition of a higher Vertebrate.

B. The Cartilaginous Cranium.

Of this there are three forms : in the first (a) there is no
mandible ; in the second (b) the mandible is present, and the
suspensory apparatus by which it is connected with the skull
forms one mass with the latter ; in the third (c) the mandible
is also present, but the suspensory apparatus by which it is con-
nected with the skull is freely moveable.

a. The cartilaginous cranium without a mandible.

This kind of cartilaginous cranium is found only among
the Marsij)obranchii, or Lampreys and Hags, and a descrip-

192

OX THE STRUCTURE OF THE SKULL.

tion of its characters in the former will suffice to illustrate its
nature. Fig. 75 represents a vertical and longitudinal section

Fig. 75.

Fig. 75. — Vertical and longitudinal section of the anterior part of the body of a Lamprey
(Petrom;/zon marinus). — A, the cranium, with its contained brain ; a, section of the
edge of the cartilage marked a in Fig. 76 ; Olf, the entrance to the olfactory chamber,
which is prolonged into the caeca 1 pouch, o ; Ph, the pharynx ; Br, the branchial
channel, with the inner apertures ot the branchial sacs ; 31, the cavity of the mouth,
with its homy teeth ; 2, the cartilage which supports the tongue ; 3, the oral ring.

of the anterior part of the body of the large Sea Lamprey (Petro-
myzon marinus), and gives a very good notion of the excessively
minute proportions of the proper skull (A) to the rest of the
body in this animal. A and B (Fig. 76) are lateral and superior
views of the skull with its accessory cartilages, separated from
the soft parts. The notochord (Gli) is, as in AmjjJiioxus, ex-
ceedingly large, and is surrounded by a merely membranous
sheath, from which prolongations are given off above to form the
sides of the small neural canal. In the walls of this canal, carti-
laginous rods, which represent neural arches, are developed, and
it dilates more distinctly in the head than in Ampliioxus,
though the cranial cavity is still very minute. The myelon
also undergoes a very distinct enlargement as it enters the
cavity, and all the typical divisions of the vertebrate ence-
phalon are recognisable in the brain thus formed.

The notochord terminates in a point immediately behind
the pituitary body. As it approaches the cranium, a rod of
cartilage (I, Fig. 76, A) is developed on each side in the lateral
parts of its sheath, and gives attachment to the branchial
skeleton (in) ; still more anteriorly two other cartilaginous fila-

THE SKULLS OF FISHES.

] 93

mentS (/.') appear, side by side, in the under region of the sheath.
These pass into the hinder part of the proper cranium, which is
a sort of cartilaginous box, closed in front, and through the
greater part of its roof, only by membrane, but complete behind,
where it arches over the myelon, and is perforated by tho
occipital foramen. The postero-lateral parts of this cranium
are dilated to give rise to the two oval auditory capsules (c), and
beneath these they are produced into two processes, h and /,
which have a common base, but diverge from one another
below. The process h gives attachment to a cartilage which is
connected with that supporting the tongue (i). The process/,
on the other hand, passing downwards and forwards, becomes

Fig. 76.

Fig. 70. — A, the skull of a Lamprey viewed from the side; B, from above (after Miiller).
— a, the ethmo-vomerine plate; 6, the olfactory capsule ; c, the auditory capsule;
d, the neural arches of the spinal column ; e, the palato -pterygoid portion ; /, the
hyomandibular and symplectic portion, and g, the quadrate portion of the sub-ocular
arch ; h, stylo-hyal process ; t, lingual cartilage ; k, inferior, /, lateral prolongation
of the cranial cartilage; 1, 2, 3, accessory labial cartilages.

continuous at g, with another bar of cartilage e, which is con-
nected with the antero-lateral part of the skull beneath the

194 ON THE STRUCTURE OF THE SKULL.

olfactory capsule. The eye lies over the triangular space
inclosed between the sides of the skull and these two processes,
so that e, g, f may be termed the sub-ocular arch.

If the skull is viewed from below, the processes e and e
of opposite sides are seen to be continued into one another by a
transverse band of cartilage, which forms the proper anterior
boundary of the skull. The front edge of this band, which
Midler calls the " hard palate," articulates with the broad and
expanded cartilaginous plate (a). The common roots of the
processes f and h are also continued into a " basi-occipital '
plate of cartilage, but, between this plate and the " hard palate,"
there is an oval space through which the neck of the long olfac-
tory csecuni (o, Fig. 75) passes. This caecum, therefore, separates
the front part of the floor of the cranial cavity, which is simply
membranous, from the so-called " hard palate." On com-
paring this skull with that of the embryonic fish (see Fig. 72), h
obviously answers to the stylo-hyal cartilage ; /, to the ascending
posterior cms of the palato-quadrate inverted arch and the hyo-
mandibular cartilage ; e, to the ascending and anterior crus of
the same. It is true that no natural division of the arch into
palato-quadrate and hyomandibular (and symplectic) portions
occurs in the lamprey, but this is only one of several respects in
which the Marsipobranchs resemble Amphibia rather than osseous
fishes. The inverted cartilaginous arch which gives attachment
to the hyoidean and mandibular apparatuses of a tadpole is
strictly comparable to the arch (e, g, f) in the lamprey. The
margins of the oval space upon the base of the skull answer to
the divergent trabecular cranii, and the plate a to the ethmo-
vomerine cartilage. The remarkable and apparently anoma-
lous separation of the basis cranii into an upper membranous
and a lower cartilaginous part, by the interposition of the back
ward prolongation of the olfactory chamber, seems to me to
be comparable to that separation of the upper and lower walls
of the pre-sphenoid, basi-sphenoid, and even of the basi-occipital,
by a backward extension of the olfactory cavities, which takes
place in so many of the Mammalia. On the other hand, I doubt
whether the accessory buccal cartilages, 1, 2, 3, &c, can be
strictly compared with anything in other fishes, though some of

THE SKULLS OF FISHES.

195

them are doubtless, as Miiller has suggested, the analogues of
labial cartilages.

b. The cartilaginous cranium with a mandible and a fixed
suspensorium.

The Hohcepliali, or Chimseroid fishes (Chimsera and Callo-
rliijnehus) present this type of cranial organization. In accord-
ance with the large development of the brain, the skull of these
fishes has attained a great advance in dimensions over the spinal

Fig. 77.

Fig. 77. — Skull of Callorhynchus Antarcticus (after Miiller). — a, anterior tooth of the
upper jaw ; c, posterior tooth ; b, mandibular tooth ; d, e, /, g, h, i, k, I, m, accessory
labial, nasal, and rostral cartilages ; n, quadrate portion of the sub-ocular plate which
supports the hyoid (o) and the mandible {Mn) ; p, the representatives of branchiostegal
rays ; q, the branchial arches ; Au, auditory region ; Or., orbit ; V1, nasal division of
the fifth nerve.

column, and presents a large internal chamber. It is a con-
tinuous cartilaginous mass, without any superior aperture of suf-
ficient size to deserve the name of a fontanelle, in the base of
which the notochord does not persist, and which is definitely
articulated by two lateral convex facets and a median concave
surface on the hinder margin of its floor {A, Fig. 78) with the
anterior segment of the spinal column.

The skull is high and compressed from side to side ; pos-
teriorly, it exhibits, on each side, an enlargement (Au), which
lodges the auditory organ. In front of these are the large
orbits (Or.), separated by a thin membranous inter-orbital septum
(LOr.), which is unlike the inter-orbital septum usually met with,

o 2

196

O^ THE STRUCTURE OF THE SKULL.

in that it lies above, and not below, the forward continuation of
the cranial cavity (Fig. 78). Usually, an inter-orbital septum is
formed by the compressed floor of the skull ; here it is constituted
by the compressed roof. Two chambers for the olfactory sacs
(N, Na) terminate the skull anteriorly and inferiorly ; and they,
the lips and the anterior part of the snout, are protected and
supported by a number of accessory cartilages (d to m).

Below the auditory and orbital regions, and in front of the
latter as far as the nasal capsules, the base of the skull gives off
a broad cartilaginous sub-ocular plate (C, D, Fig. 78), the two

Fig. 78.

Fig. 78. — Vertical section of the skull of Chimoera monstrosa without the labial and nasal
cartilages. — A, the basi-occipital region ; P, the pituitary fossa ; Na, the partition
between the two olfactory sacs ; B, the alveolus for the anterior upper jaw tooth ;
I. Or., the inter-orbital septum ; asc, psc, anterior and posterior vertical semicircular
canals ; /., II. , V., VIII., exits of the olfactory, optic, fifth, and eighth pairs of
cerebral nerves.

edges of which, sloping towards one another, bring it to a point at
n (Fig. 77 ; D, Fig. 78). With this part of the plate the mandible
(Mn) is articulated, while to the middle of its posterior margin
(D, G, Fig. 78) the hyoiclean apparatus (o, Fig. 77) is attached.

A vertical section of the skull (Fig. 78) shows that the proper
cranial cavity consists of a large posterior chamber, divided by a
long and comparatively narrow neck from a much smaller, but
still large, anterior chamber. The latter contains the olfactory
lobes, and presents on each side, in front, a sort of cribriform
plate, through which the filaments of the olfactory nerve pass

THE SKULLS OF FISHES. 197

to the nasal sacs. The commencement of the narrow neck is
perforated on each side by the optic foramina (II.). The hinder
dilatation contains the mass of the brain, and, on each side,
chambers for the auditory organs, which communicate with it,
are situated. The posterior edge of the inter-orbital septum
bounds this chamber in front, above the " neck." In front of
the anterior boundary of the inter-orbital septum, and above
the olfactory division of the skull cavity, there is a curious
chamber filled with fatty matter, and open in front and behind,
which is traversed by the nasal division of the fifth nerve.

Miiller well says, " The skull of Chimasra is most like that
of a tadpole ; " * but if we interpret the former strictly by the
latter, as I believe ought to be done, the results will be some-
what different from those at which Miiller arrives. The plate
C, D answers precisely to the sub-ocular arch of the lamprey and
to the corresponding arch in the tadpole's skull, though it is
chondrified throughout, and not perforated by a large a]3erture,
as in the two latter animals. But, admitting this, the further
development of the frog proves that the sub-ocular arch answers
to the common suspensorium of the hyoid and mandible, and to
the palatine, pterygoid, and quadrate bones ; and that it has
nothing to do with the maxilla or premaxilla. The large
posterior upper jaw teeth of the Chimseroids (c), therefore, being
attached to the under surface of the anterior part of the sub-
ocular plate, must be palatine or palato-pterygoid teeth. The
small anterior teeth (a), on the other hand, are fitted into fossae,
or alveoli (B), which are situated immediately under the floor of
the nasal chambers, in the vomerine region of the skull, and
must be regarded as vomerine teeth — as, indeed, Cuvier sug-
gested. On the other hand, I think Midler's view that the
cartilages (d, e, &c.) are accessory labial cartilages, and not, as
Cuvier supposed, representatives of preniaxillae, in axillae, &c,
has everything in its favour.

c. The cartilaginous cranium, with a mandible and with a
moveable suspensorium.

This form of cranium is met with in the sharks and rays,
and may be illustrated by an account of that of Squat ina, the
* Vergleichende Anatomic tier Myxinoiden. Erster Tlieil, p. 159.

Fig. 79.

Figs. 79 and 80.— The skull of Squatina viewed from above (Fig. 79), and from the side
(Fig. 80). — a, vomerine region ; b, prefrontal ; c, post-orbital ; d, post-auditory pro-
cesses; c, occipital condyles; /, occipital foramen; (J, suspensorium ; A, upper den-
tigerous arch; *", k, I, labial cartilages; Afn, mandible; An, auditory chamber; Or,
orbit; 2V, nasal chamber ; Op, opercular cartilaginous iilaments ; Br, branchiostegal
rays ; Hy, hyoideah arch.

THE SKULLS OF FISHES. 199

monk fish (Figs. 79 and 80). The form of the skull is here the
exact converse of that observed in the Chimaoroids, being exceed-
ingly broad and depressed, instead of high and compressed.
The surface of the cranium is encrusted with a pavement of
minute ossicles, arising from the ossification of the superficial
layer of the cartilage. Behind, the basilar region of the skull
presents two lateral articular surfaces to the front part of the
spinal column, and exhibits the wide lateral expansions for the
auditory organs (An) ; at the sides of the skull, in front of these,
lie the orbits (Or) bounded behind by the post-orbital processes
(c), and by the antorbital, or prefrontal, processes (b) anteriorly.
The latter divide the orbits from the nasal chambers (N), the
apertures of which look downwards. The prefrontal processes
are continued, on the inner sides of the nasal chambers, into a
broad plate, emarginate anteriorly, which terminates the floor of
the skull, and corresponds with the ethmo-vomerine part of the
cartilaginous skull of the human foetus or of the pike. The
anterior part of the roof of the skull is not directly continued
into the upper surface of the plate, but ends in a deeply concave
edge ; the vacuity, or fontanelle, is occupied by fibrous tissue in
the recent state. Small apertures upon the roof of the occipital
region communicate with the auditory chambers. The upper
end of a stout prismatic cartilage (g) is moveably articulated with
the outer wrall of the auditory prominence. The lower end of
this cartilage is united by ligaments behind to the hyoidean arch
(Sy), and in front to the upper and lower dentigerous arches
(h and Mri). Each of these arches is composed of two pieces
united in a median symphysis, and the under surf^cCo of the
outer and posterior ends of the upper arch are articulated with
the upper surfaces of the outer and posterior ends of the lowrer
arch. The upper arch is, in addition, articulated with the under
surface of the prefrontal region of the skull. Three cartilages
(i, h, I), connected together by ligaments, lie outside the denti-
gerous arches, two, on each side, being superior and one inferior.
Furthermore, cartilaginous filaments (Op) are attached to the
hinder edge of the prismatic cartilage (d), and to the hyoidean
arch (Br).

The interpretation of the cartilages (i, k, 1) has been a matter

200 ON THE STRUCTURE OF THE SKULL.

of much controversy. In a remarkable essay published in the
first volume of the " Memoires du Museum," Cuvier proposed to
consider the upper dentigerous arch (h) as the homologue of the
palatine and pterygoid bones of osseous fishes, the cartilages
(i, k) as the premaxilla and maxilla. The suspensorium (g) he
considered to be the homologue of the hyomandibular, symplectic,
and metapterygoid. The lower dentigerous arch (Mn) was ob-
viously the mandible.

On this latter point all anatomists are agreed ; but, in his
famous " Comparative Anatomy of the Myxinoid Fishes,"
Johannes Miiller — guided, like Cuvier, by purely anatomical
considerations, and bv what I have elsewhere termed the
method of gradation — proposed a totally different interpretation
of the other parts. According to this view, i, h, and I are
merely labial cartilages, and therefore do not represent the pre-
maxilla and maxilla. Again, Cuvier had greatly relied upon
the absence of any parts on the inner side of h which could
answer to palatine or pterygoid elements, in arguing that h
itself represents them. But Miiller adduced his own and
llenle's observations to prove that in a great many Plagiostomes,
particularly the Bays, such cartilages, situated on the inner side
of the upper dentigerous arch, do occur, and thus arrived, by a
line of argumentation precisely as legitimate as that of Cuvier,
at the exactly opposite result, — that h represents the premaxilla
and maxilla, and not the palatine or pterygoid.

The fact that these opposing viewTs were entertained by men
-like Cuvier and Miiller is evidence that each had much in its
favour ; but, in truth, neither was free from grave difficulties.
Thus neither accounted for the articulation of the mandible
with the upper dentigerous arch, — a relation into which the
mandible never enters either with the palatine, or with the
maxilla, in the vertebrate series; and as Miiller himself is
forced to admit that some of the cartilages on the inner side
of the upper dentigerous arch are accessory, why should not all

be so?

This is just one of those cases in which the study of develop-
ment manifests its full importance, and decides, at once, problems
which, without it, might be the subjects of interminable discus-

THE SKULLS OF FTSI1ES. 201

sion. A comparison of the skull of the monk fish with that of the
embryonic osseous fish (Fig. 72, C) seems to me to demonstrate
beyond question, that the upper dentigerous arch (h) corre-
sponds with the palato-quadrate cartilage of the embryo,* and
that the suspensorium {g) equally corresponds with the hyoman-
dibular and symplectic cartilage. But in this case Cuvier's view
of the upper dentigerous arch must be regarded as a singularly
near approximation to the truth, for it certainly answers to the
palatine and pterygoid ; though, in addition, it contains the
representatives of the quadrate and metapterygoid bones of the
osseous fish. And his opinion regarding the nature of the sus-
pensorium was still nearer to what I believe to be right. On the
other hand, I think it very probable, though not certain, that,
as Miiller supposed, the cartilages (i, h, I) are merely labial, and
that these fishes have no representatives of the premaxilla and
maxilla. But the so-called palatine and pterygoid cartilages of
Miiller, if the view I take is correct, are as much accessory parts
as the spiracular cartilages, and, like them, have no representa-
tives in osseous fishes.

* Eatlike arrived at tills conclusion also, on developmental grounds, in 1839. See
his " Vierter Bericht," quoted in the last Lecture of this work.

202

LECTUEE XL

ON THE STRUCTURE OF THE SKULL.

THE SKULLS OF FISHES AND AMPHIBIA.

C. The cranium, consisting chiefly of cartilage and without
cartilage bones, but ivith superadded membrane bones.

The skulls of the chondrosteous Ganoids, the Sturgeons, and
Spatularise exemplify this type of structure, which forms a most
interesting transitional link between the skull of Plagiostomes
and the skull of ordinary osseous fishes.

Spatularia has a completely cartilaginous skull, produced in
front into a great beak, flattened from above downwards. The
cartilaginous representatives of, at fewest, seven of the anterior

Fig. 81.

An Ji

Fig. 81. — Side view of the skull of Spatularia with the anterior (asc) and posterior (psc
vertical semicircular canals exposed. — An, the auditory chamber: Or, the orbit with
the eye ; N, the nasal sac ; Hy, the hyoidean apparatus ; Br, the representatives of
tin; branchiostegal rays; Op, the operculum ; Mn, the mandible.

THE SKULLS OF FISHES AND AMPHIBIA. 203

vertebra) of the spinal column coalesce into one mass with one
another and with the skull. The notochord, extremely large in
the spinal column, rapidly diminishes in size as it enters the
skull, and, becoming a mere thread, terminates behind the
pituitary fossa. The auditory organs are contained in large
postero-lateral projections of the cranial mass, with the outer
sides of which the suspensoria are connected. The base of the
skull is protected by a long parasphenoid, which extends back
under the anterior part of the spinal column ; in the dorsal
region it presents an anterior and a posterior pair of perichon-
drial ossifications, separated by oblong laminae from lateral bony
plates of the same character, but the homology of these bones
with those in the roof of the Teleostean skull is not, to my mind,
satisfactorily made out.*

The suspensorial apparatus of Spatularia consists of a single
bone (A), compressed from above downwards superiorly, and
from side to side inferiorly, with a superior and an inferior car-
tilaginous epiphysis ; to the lower cartilaginous epiphysis the
operculum (Op) is attached, and a short thick prismatic cartilage
(B) is united by ligament with, and can play freely upon, its
anterior and inferior angle. Posteriorly the lower end of this
cartilage (B) is connected by ligament with the hyoidean arch
(Hy), w7hich consists of two portions on each side ; a small upper
piece, with which the flat bone (Br), representing a branchio-
stegal ray, is connected ; and a long lower ramus, the middle
third of which is bony, while the two ends are cartilaginous.

Anteriorly, the lower end of the inferior suspensorial carti-
lage (B) is united by ligaments to two cartilaginous semi-arches
(D and Mn), of which the upper (D) is articulated by a trans-
versely convex head with a concavity of the lower (Mn). The
upper semi-arch is ligamentously united to its fellow in the
middle line, and is suspended by ligamentous fibres to the under
part of the prefrontal region of the skull. A long flat bone (E),
the hinder end of wdiich is cut off in the specimen figured, lies
on the outer side of the cartilage (D), and extends to the middle
line. A second long flat bone is closely applied to the inner

* See " Spatnlariarum Anatomiam descripsit Tabulaquc iUustravit Albertus

Wagner." Berolini, 1848.

204 ON THE STRUCTURE OF THE SKULL.

surface of the cartilage and follows its curves, from its hinder to
its anterior extremity, overlapping and folding over the upper
edge of the anterior three-fifths of the cartilage. Between the
hinder part of E, here cut away, and D, is a space occupied by
the levator muscle of the lower jaw.

The mandibular cartilage extends to the symphysis, and is
coated externally, and partially embraced by, the flat bone (Hn),
the greater part of the upper edge of which bears teeth.

On comparing these parts with those of the corresponding
apparatus in the embryonic fish (Fig. 72), it becomes clear that
the pieces A and B answer to the hyomandibular and symplectic,
taken together. Indeed, at first sight, A, supporting as it does
the operculum, seems to answer to the hyomandibular, and B to
the symplectic itself ; but then it may be suggested that the
hyoidean apparatus is attached at the distal end of B, and not
between it and A, as it would be if the two corresponded, re-
spectively, to the hyomandibular and symplectic.

The cartilage D obviously answers to the palato-quadrate
arch, and that of the lower jaw to Meckel's cartilage. The fact
that a levator muscle of the lower jaw passes between E and
1) seems to prove the former to correspond with a maxilla;
in which case the internal bone would be a sort of palato-
pterygoid, similar to that we shall meet with in Lejndosiren.

The skull of the Sturgeon (Aceipenser), like that of Spatu-
laria, is greatly enlarged, posteriorly, by the coalescence with it,
and with one another, of six or seven of the anterior vertebra?.
In front, it is prolonged into a triangular snout or beak (c, Fig.
82 ; a, Fig. 83), the wide base of which is formed by the ant-
orbital, or prefrontal, prominences which separate the olfactory
chambers from the orbits. Behind the latter are the two great
projections (c, Fig. 83) which contain the auditory organs ; and
behind these again, and separated from them by a deep lateral
fossa, are two wing-like processes (b, Fig. 82), which are directed
outwards and obliquely backwards, and proceed, not from the
walls of the cranium proper, but from those of the spinal column,
where it joins the skull. At this point there is, in the cranio-
spinal cartilage of both the Sturgeon and the Sjxdularia, a great
dilatation of the neural canal, which is closed above only by a

THE SKULLS OF FISHES AND AMFHIIilA.

205

membranous fontanelle. The skull proper lias no such fonta-
nelle. There is a well-marked pituitary fossa, and the noto-
chord, very thick in the spinal column, tapers to a thread as it
enters the base of the skull, and ends belvind this fossa.

The bones which are developed in relation with this cartila-
ginous cranium in the base of the skull are, — a great parasphe-
noid which extends back under the coalesced anterior vertebra?,
and forwards to the level of the nasal cavities ; and a slender
median bone in front of this, which underlies the rostral pro-
longation (e), and appears to represent the vomer.

No distinct ossifications protect the lateral walls of the skull,
but the bones marked F (Fig. 82) send down processes for a
short distance, and the parasphenoid gives off transverse pro-
longations upwards and outwards, from each side of the middle
of its length, as in most fishes.

The roof of the skull presents a number of distinct flat
ossifications, no one of which involves the subjacent cartilages,
and which vary very much in contour and extent in different
specimens. The general arrangement is, however, fairly repre-
sented by the accompanying figure (Fig. 82).

Fig. 82.

Fig. 82. — The cartilaginous skull of a Sturgeon, with the cranial bones. The former is
shaded, and is supposed to be seen through the latter, which are left unshaded. — a,
ridge fbimed by the spinous processes of the anterior vertebras, which have coalesced
with one another and with the skull ; b, lateral wing-like processes ; c, rostrum ; An,
position of the auditory organs ; Net, position of the nasal sacs.

Of these bones, the pairs C, G and D, D clearly represent,
both in position and character, the parietal and frontal bones
of the Pike, while F, F similarly correspond with the squamosals
of that fish.

In position, again, E answers to the ethmoid, H, H to

206 ON THE STRUCTURE OF THE SKULL.

the prefrontals, G, G to the post-frontal, B, B to the epiotics,
and A to the supra-occipital of the Pike. But every one of
these is a membrane bone, and not, as are the corresponding
elements of the Pike's skull, a cartilage bone.

These bones are therefore, strictly speaking, the analogues,
and not the homologues, of the bones to which they appear to
answer in the Pike, though hitherto no distinction has been
drawn between the two.

K and I, 1 are bones which do not properly belong to the
skull, but which, as happens among many Siluroid Teleostei,
are anchylosed with the cranium. K is the most anterior
of the median dermal scutes, and 1, 1 are the supra-scapular
bones. The letters L, L indicate the scapular bones move-
ably united with these last.

The suspensorium of the Sturgeon consists of a large, irre-
gularly-prismatic body, composed of a bony central piece (/, Fig.
83) with two cartilaginous epiphyses, the lower of which (g) is
much the longer, and is connected by ligament with another
cylindrical cartilage (Ji), while the upper articulates with the outer
and under part of the auditory capsule. Bather above the letter
h, the lower cartilage gives attachment to a cartilaginous nodule
with which the principal piece of the hyoidean arch is connected.

At its distal end the cylindrical cartilage (h) is united by
ligaments to the two apparatuses h and Mn. Of these, h may
be described as a rhomboid, composed partly of cartilage and
partly of bone, and so bent as to assume a transversely arched
form. Its outer angles present convex articular meets to con-
cavities on the lower arch (Mn), which last, composed of a
single bony ramus on each side, is undoubtedly the mandible.

Fig. 83.

Fig. S3. — Side view of the cartilaginous cranium of Accipenser (after Miiller). — a,

truin ; 6, nasal chamber; Or, orbit; c, region of the auditory organ ; /, g, h, sus-
pensorium ; k, maxillo-palatine apparatus ; Mn, mandible.

THE SKULLS OF FISHES AND AMPHIBIA. 207

The cartilaginous basis of h is strengthened by eight bones,
four on each side. Of these, two lie altogether external to the
cartilage, and leave between themselves and it an interspace, in
which the levator muscle of the lower jaw lies.

The other pair consist, firstly, of a large bone, which lies,
for the most part, internal and inferior to the cartilage, and
extends from the inner side of the articular process for the lower
jaw, upwards and inwards, to meet its fellow, posteriorly ; for-
wards, to articulate with the anterior of the external bones.
And, secondly, of a small bone fitted on to the anterior and
external edge of this, and to the inferior surface of the anterior
external bone. The whole apparatus, h, is very loosely con-
nected with the skull, so that it is capable of being protracted
and retracted with great freedom.

The general relations of this singular mechanism to the
manducatory organs of ordinary Teleostean fishes appears to be
rendered evident by the same method as that which has been
employed to demonstrate the nature of the jaws of the Plagio-
stomes. The osseo-cartilaginous structure, 7c, answers to the
palato-quadrate arch of the Sharks and Kays, or to the palato-
quadrate cartilage of the embryonic fish ; and /, g, h corre-
spond with the undivided suspensorium of the Sharks and Rays,
and with the hyomandibular and symplectic cartilages of the
embryo Teleostean. Furthermore, on comparing h with the
maxillary apparatus of Sjmtularia, the cartilaginous basis appears
to answer to the cartilages (D, D) of that fish joined together ;
while the anterior outer bone in the Sturgeon is the equivalent
of the bone E, and may be regarded as a maxilla. The two
internal bones correspond with the inner bone of the Spatu-
laria's jaw. The Sturgeon, however, more nearly approaches
ordinary fishes in the development of an anterior or palatine
element, distinct from the posterior or pterygoid element. As
for the small external bone, which passes obliquely from the end
of the maxilla to the outer surface of the cartilage, it is possibly
a quadrato-jugal.

D. The cranium consisting of cartilage to a greater or less
eitent, but with cartilage bones as well as membrane bones. — ■

208 ON THE STRUCTURE OF THE SKULL.

The class Pisces presents us with a complete series of these
crania, from the least ossified forms, which possess only one or
two pairs of cartilage bones m the walls of the cranium, to the

' Fig. 84.

Mn y

Fig. 84. — Lepidosiren. A, the parieto-frontal bone ; B, the supra-orbital ; C, the nasal ;
D, the palato-pterygoid ; E, the vomerine teeth ; E.O., the ex-occipital ; Mn, the
mandible ; Hy, the hyoid ; Br, the branchiostegal ray ; Op, the opercular plate ; x,
the parasphenoid ; y, the bone which gives attachment to the scapular arch ; Or, the
orbit ; An, the auditory chamber ; N, the nasal sac.

completely ossified skulls of the Cyprinoids and Siluroids. And
again, just as among the preceding groups we found that the
Chimoeroids differed widely from the rest in having the sub-ocular
process, or arch of the skull, to which the mandible is attached,
formed of one piece of cartilage, which is continuous with, and
immoveable upon, the skull ; so, in this series, Lepidosiren is at
once distinguished from all the rest by a similar character.

The skull of the Mudfish (Fig. 84) is composed of a frame-
work of cartilage, which sends down a broad triangular process,
on each side, to articulate with the mandible, and expands,
posteriorly and laterally, into chambers for the auditory organs.
Between these, the roof and the floor of the skull are both con-
stituted by cartilage ; but anterior to them, as far as the ex-
tremity of the parasphenoid (.*;), this tissue becomes very thin or
disappears (Fig. 85). In front of the anterior end of the para-
sphenoid, it makes its appearance again on both the roof ami
the floor of the cranial cavity, beyond which it is continued
as a thin lamella to the end of the snout. A fibrous septum
with a free concave, posterior margin, divides this region of the
cranium into two lateral chambers, one for each olfactory lobe.

THE SKULLS OF FISHES AND AMPHIBIA.

209

Behind the auditory chambers the cartilage is almost ex-
cluded from the walls of the skull by two lateral ossifications
of its substance — the ex-occipitals (E.G.). As in the Amphibia,
there is no ossified supra- or basi-occipital. The rest of the lateral
parietes of the skull would be devoid of bony walls were it not
that the parasphenoid (x) and the great bone (A), which roofs
in the whole length of the skull, and answers to the frontals
and parietals, send upwards and downwards, respectively, lateral
processes, which unite together, and so replace the alisphenoid
(Fig. 85). The ethmo- vomerine cartilage {Eth. Vo.) bears, supe-
riorly, the nasal bones (C), and inferiorly it carries teeth (E).
A long flat bone, pointed posteriorly (B, B), is attached to the
hinder edge of the nasals, and roofs over the orbit and temporal
fossa.

Fig. 85.

Fig. 85. — Longitudinal and vertical section of the Skull of Lepidosiren. The cartilage is
dotted ; the membranous and bony constituents are shaded with lines. A, B, C, D,
as in the preceding figure ; x, x, the parasphenoid ; I, 2, the first and second ver-
tebral arches ; Ch, the notochord ; Au, the situation of the auditory organ.

The notochord, which forms the chief axis of the spinal
column of this fish, is continued into the base of the skull, and
ends in a point about the level of the exit of the trigeminal
nerves (V). There is neither basi-occipital nor basi-sphenoid,
and the presphenoid is represented only by the cartilaginous
floor at (P./S"). The pterygopalatine apparatus is represented,
on each side, by the great dentigerous curved plate (D), which
is applied to the inner surface of the cartilaginous sub-ocular
process, abuts against the parasphenoid by its inner edge, and
descends to the inner side of the articular condyle for the
mandible (a). The hyoidean arch {Hy) is attached to the
middle of the posterior and inferior edge of the sub-ocular

210 ON THE STRUCTURE OF THE SKULL.

cartilage, to the posterior part of the outer surface of which is
applied the bone (F, Fig. 84) with which the opercular bone
(Op) is moveably united by ligament.

The bone (F) has, like most of the bones of the Lepidosiren,
a green colour. Through the greater part of its length it is so
easily separated from the cartilage that it is clearly a membrane
bone. Towards the condyle, however, it adheres firmly to,
though, on the application of a certain force, it springs away
from, a nodule of whitish bone, which lies in the very substance
of the articular end of the cartilage, and repeats its pulley-like
form. I suspect that this nodule, which represents the os
quadratum, is primitively distinct from the bone (F). The
latter, under these circumstances, would have much analogy
with the pre-operculum of osseous fishes, and Op would corre-
spond with the sub- or inter-operculum.

All other fishes, comprising such Ganoidei as have not been
already mentioned, and the Teleostei, have, so far as is at present
known, the palato-quadrate arch primitively distinct from the
hyomandibular suspensor ; the latter is, primitively, moveable
upon the skull ; and, in the walls of the cranium, the pro-otic
bones, at least, are ossified as well as the ex-occipitals ; that is to
say, they are constructed essentially upon the plan of the Pike.
The modifications they exhibit in detail are almost infinite, but
a few of the most important may be enumerated : —

1. The cartilaginous cranium persists throughout life in such
fishes as the Pike and the Salmon ; in very many, as the Perch
and the Carp, it disappears almost entirely.

2. In most fishes the basis cranii is compressed from side to
side in the orbital region, and vertically enlarged, so as to form
an inter-orbital septum, which, as it were, encroaches upon the
cranial cavity and narrows it anteriorly. But in others — such
as the Cyprinoids and the Siluroids — no inter-orbital septum is
developed, the basis cranii remaining flat, and the cranial cavity
of nearly equal size throughout.

3. The last-mentioned fishes have the cranial walls com-
pletely occupied by bone, distinct ossifications representing the
alisphenoids and orbito-sjmenoids.

THE SKULLS OF FISHES AND AMPHIBIA. 211

4. The opisthotic bone, occasionally absent as a distinct
ossification, is very small in some fishes, such as the Perch (where
it is Cuvier's "rocher" or "petrosal"), but becomes very well
developed in such genera as Ephippus, and attains an immense
size in the Gadidee.

5. The canal for the orbital muscles is absent in many fishes,
such as the Cod tribe.

6. The most remarkable modification of the fish's cranium
proper, howTever, is the want of symmetry produced in the flat
fishes, or Pleuronectidse, by a sort of twist, which affects the
anterior and upper, but not the hinder and inferior, part of the
skull. Thus, if the skull of a Turbot be examined, the supra-
occipital wrill be found in its ordinary place ; while the epiotics
and squamosals are symmetrically disposed on each side of it, so
that the skull, viewed from behind, is like that of any other
ordinary osseous fish. The basi-occipital, parasphenoid, and
vomer are likewise arranged, as usual, along the median basal
axis of the skull. The pro-otics and post-frontals are also nearly
symmetrical, but the alisphenoids are thrown over to the left
side, so that the anterior aperture of the cranial cavity, be-
tween the alisphenoids, lies no longer immediately over the
parasphenoid, but to the left of it. The left frontal sends down
a long curved process, which joins with one from the prefrontal
of the same side, and the two eyes come to lie in the secondary
orbit, developed between the curved bony boundary thus formed
and the median frontal crest.

7. An addition takes place to the posterior extremity of the
skull, in many fishes, by the anchylosis with it, and with one
another, of a variable number of vertebra?. Cartilaginous verte-
bra), as I have already pointed out, coalesce with the cartila-
ginous skull in both Accipenser and Spatularia, and two or three
bony vertebra? are anchylosed with the osseous skull in Lejndos-
teus and Pohjpterus. Whether a similar addition takes place in
the other living ganoid, Amia, or not, I cannot say. In many
Siluroids a great number of vertebra? become thus anchylosed
with one another and with the skull.

8. In both Siluroids and Ganoids, again, an addition to the
roof of the skull is effected bv the coalescence therewith of the

p 2

212 ON THE STRUCTURE OF THE SKULL.

suprascapular bones, as well as, in some cases, of dorsal dermal
bones.

9. But certain of the most striking modifications of the physi-
ognomy of osseous fishes are the result of the prolongation of the
region in front of the orbit, which may be effected in two very
different ways. For example, it is chiefly the elongation of the
preinaxillae and mandible which gives rise to the remarkable
beak of the " sword-fish ' (Xiphias) ; while, in Fistularia, the
premaxillaB and mandible remain very short, but are thrust out
to a great distance from the orbit, by the production of the nasal
and vomerine regions, on the one hand, and of the bones of the
suspensorium on the other.

10. In such fishes as Syngnathus and Fistularia, a line join-
ing the articular socket of the hvomandibular with the condvle
of the os quadratum, makes a very acute angle with the base of
the skull. In most fishes this angle is more or less acute ; but
in Polypterus, and still more in Muroena, it becomes a right or
an obtuse angle, the corner of the gape being thus thrown
behind the eye, instead of being, as in most bony fishes, in front
of it. We shall find a similar rotation of the distal end of the
suspensorium to take place in the series of the Amphibia, and
in the passage from the tadpole to the adult state of the
highest of these animals.

11. The connection of the palato-quadrate arcade with the
hvomandibular and symplectic suspensor varies, from the firm
sutural union which is observed in the Pike and most osseous
fishes, to a bond which is hardly closer than that which obtains
in the Plagiostomes and Sturgeons, in Polypterus. In Lepidos-
teus, except for the inter- and pre-operculum, the tie between
the symplectic and the palato-quadrate bones would be very
loose, the palato-quadrate arcade and the suspensor being, as it
were, naturally dissected from one another.

In some Pledognathi and Siluroids, on the other hand, all
these parts become firmly anchylosed together, and with the side
walls of the cranium.

12. Finally, the multiplication of the bony constituents of
the maxilla and the mandible in Lepidosteus — the conversion of
the maxilla into a mere support for a tentacle in many Siluroids

THE SKULLS OF AMrHIBIA. 213

— the absence of branchiostegal rays, and the presence of two
"jugular' plates between the mandibular rami in Pohjpterus,
must not be overlooked even in tliis brief enumeration of a few
of the most salient modifications of the skulls of osseous fishes.

THE SKULLS OF AMPHIBIA.

In cranial structure, as in all the other more important
features of their organization, the Amphibia are closely allied to
Fishes, and widely separated from the abranchiate Vertebrata.

As in Fishes, a single median membrane bone, or parasphe-
noidj is developed under the base of the skull, while no such
median bone is found in the higher Vertebrata. Like Lepido-
siren, the Amphibia have no ossified basi-occipital or supra-
occipital, whereas all the abranchiate Vertebrata possess these
bones.

Again, like Lepidosiren and many other Fishes, the Amphibia
have no true basi-sphenoid, developed in the cartilage of the
basis cranii ; while all the abranchiate Vertebrata have that bone
well developed.

The hyoidean apparatus is, in Amphibia, as in Fishes,* con-
nected with a suspensorium common to it and the mandibular
apparatus. In all the higher Vertebrata the hyoidean apparatus,
if it is attached directly to the skull at all, is united therewith
separately and distinctly.

In all Amphibia which have ossified ex-occipitals, a condyle
is developed on each, for articulation with the first vertebra of the
spinal column ; and the basi-occipital, remaining unossified,
takes no share in the formation of these condyles. In all the
higher Vertebrata, on the other hand, the bony basi-occipital
takes a greater or less share in the formation of the occipital
condyle, or condyles.

The skull of Amphibia resembles that of the Chimseroids and
Lejnclosiren, and differs from that of Teleostean, Ganoid, and
Plagiostome fishes, in the absence of any natural division between
the palato-quadrate and suspensorial cartilages.

* According to Stannius, however, the hyoidean arch is attached directly and
independently to the skull in many Eays. See that author's admirable " Handbuch
der Anatomie der Wirbelthiere," Erster Buch, p. 46.

214

ON THE STRUCTURE OF THE SKULL.

Like the Carp and the Siluroids, the Amphibia are devoid of
any inter-orbital septum, the cranial cavity remaining of tolerably
even size from the occipital foramen to its anterior termination.

In the Frog (Fig. 86) the skull is roofed in by two long flat
membrane bones {Pa, Fr), which correspond with the parietals
and frontals, and, in fact, each originate in two distinct centres,

Fig. 86. — Skull of Rana esculenta. Seen A, from above ; E, from below ; C, from the
side (after Duges). — x, the parasphenoid ; y, the girdle bone ; z, the " temporo-
mastoid " of Dug&s.

one in front and one behind. In front of these are two other
membrane bones (Na, Na), which have been variously inter-
preted, but which probably answer to the nasals. On the base
of the skull is the long single parasphenoid (#), the hinder part
of which is produced into two broad lateral processes, which
underlie the auditory capsules.

AY hen these membrane bones have been stripped off, a sub-
jacent cartilaginous cranium becomes apparent, produced behind
into two lateral enlargements, or tuberosities, for the auditory
organs, and having certain fontanelles or membranous spaces in
its upper wall (Fig. 87). In the substance of this cartilaginous
cranium, posteriorly, are two ossifications, one on each side of

THE SKULLS OF AMPHIBIA. 215

the occipital foramen, which nearly meet in the middle line
above and below. These, the ex-occipitals, bear the condyles
for articulation with the atlas, and partly shelter the posterior
portion of the auditory organ. The front and upper wall of
each auditory tuberosity is also largely ossified, the resulting
bone protecting the anterior part of the organ of hearing, and
being perforated, or notched, for the transmission of the third
division of the trigeminal. This therefore is, without a question,
the homologue of the pro-otic bone of the fish and of Man.

A fifth ossification of the cartilage is the very singular
bone (i/) which Cuvier termed the os en ceinture, or " girdle bone,"
from its encircling the anterior part of the cranial cavity. This
bone has somewhat the form of a dice-box, with one end divided
by a longitudinal partition. The latter — the front part of the bone
— extends into the prefrontal processes in some frogs, protects
the hinder ends of the olfactory sacs, and is perforated by the
nasal division of the fifth. The median partition therefore must
answer to some extent to the ethmoidal septum, while the lateral
parts of the anterior division of the bone correspond with the
prefrontals. On the other hand, the hinder division of the bone
is an ossification of each wall of the cranium, in front of the exit
of the optic nerves ; so that I conceive this part of the bone
can only answer to the orbito-sphenoids, united above and below.
Upon this view of its nature, the girdle bone answers to at least
five bones, viz., the ethmoid, prefrontals, and orbito-sphenoids.

No alisphenoid is developed in any Amphibian. There is
no separate opisthotic in the adult state, and I am not fully
satisfied as to the existence of any distinct epiotic, though such
a bone has been affirmed to exist (under the name of "mastoid")
in the axolotl and the Menobranchus*

The anterior part of the ex-occipital, in front of the foramen
for the eighth nerve, which perforates that bone, probably re-
presents the opisthotic, as between it and the posterior external
margin of the pro- otic is placed the fenestra ovalis, a structure
not met with in the class Pisces.

The facial bones are, for the most part, readily determinable ;

* Mr. Parker informs me that the comuio:i Toad has a thin bony crest answer-
ing to the epiotic.

216

ON THE STRUCTURE OF THE SKULL.

thus there can be no doubt about the preniaxillae (Pmx), the
maxilla (Mx), and the two large-toothed vomers (Vo). The
position of the posterior nares between the last-named bones and
the bones (PI) taken in connection with the relations of the
latter to the prefrontal region of the skull, sufficiently defines
the palatine character of PI; while Pt, connected with the
palatines on the one hand, and terminating on the inner side of
the mandibular suspensorium on the other, corresponds as dis-
tinctly with the pterygoids of the higher Vertebrata.

Fur. 87.

Fig. 87. — The Cartilaginous Cranium of Rana esculenta. A, from above ; B, from below

(after Duges). — y, the girdle bone.

The bone (QJ), which connects the end of the maxilla with
the outer side of the suspensorium, appears to correspond with
the quadrato-jugal of abranchiate Vertebrata.

It is more difficult to determine the nature of the bone z,
the " temporo-mastoid " of Duges, which is a long, hammer-
shaped membrane bone, extending from the skull to the articular
surface for the lower jaw, and sending a long process forwards
over the temporal region.

When this bone, the pterygoids, palatine, quadrato-jugal, &c,
have been removed, the cartilaginous cranium of the Frog (Fig.
87) is seen to give off from the outer ends of the prefrontal region
and the auditory protuberances, two prolongations, the anterior
of which curves round the orbit, and eventually unites with the
posterior in the cartilaginous process which articulates below
with the mandible, and unites by its upper or cranial end with
the suspensor of the hyoidcan apparatus (St., Fig. 86).

THE SKULLS OF AMPHIBIA. 217

This arch clearly answers to the sub-ocular arch of the
Lampreys and to the sub-ocular cartilage of the Chimaeroids
and Lejndosiren, and corresponds with the palato-quadrate,
hyomandibular, and symplectic cartilages of the embryonic
osseous fish taken together. The distal end of this cartila-
ginous pedicle commonly presents a larger or smaller ossification
of its substance, which represents the quadrate bone. Now,
the problematical bone (z) lies on the outer side of the pedicle,
and I was at one time inclined to think that it represented the
hyomandibular bone of osseous fish — being largely led to that
impression by the great size of the hyomandibular and the
comparative minuteness of the quadrate in the Conger and the
Murcenoid fishes. But the hyomandibular is an ossification
in the cartilage of the suspensorium, not a membrane bone.
The bone has been compared with the tympanic, but the
tympanic membrane has a special and distinct supporting ring
in the Frogs. It has been identified, again, with the squamosal,
but it lies too far down on the outer side of the pedicle for that
bone. Tracing the changes of form in this bone (which is very
constant in the Amphibia) downwards to the Menobranchus
and Siren, its resemblance in these perennibranchiates to the
bone (F) of Lejndosiren becomes very striking ; and I am dis-
posed to identify it with that bone, which, as I have stated
above, has much resemblance to the pre-operculum of osseous
fishes.

The mandible of Amphibia is commonly composed of three
pieces — a dentary, an angular, and an articular. The latter,
always continuous with Meckel's cartilage, may itself remain
persistently cartilaginous.

The skull of the tadpole, before ossification has commenced,
presents a cartilaginous base, in which the notochord terminates
in a point, immediately behind the pituitary fossa. At the sides,
the basal cartilage expands into two oval auditory capsules, and
in front passes into the trabecular cranii, which embrace the
membranous floor of the pituitary fossa, and reunite in the
broad ethmo-vomerine cartilage. The .apex of a sub-ocular arch,
connected, behind, with the auditory region of the basis cranii,
and, in front, with the prefrontal region, furnishes an articular

218 ON THE STRUCTURE OF THE SKULL.

surface to the axial, or " Meckel's cartilage," of the mandible.
In the young Tadpole, a line drawn from the mandibular articu-
lation to the auditory capsule makes an acute angle with the
basis cranii ; but, as age advances, the angle becomes more and
more open, until, in the adult Frog, it is obtuse (Fig. 86), the
articular surface for the mandible having passed far behind the
auditory capsule. Of course the width of the gape increases
pari passu with this rotation of the mandibular suspensor.

A survey of the series of the Amphibia from the perenni-
branchiates upwards, shows, in a persistent form, those inclina-
tions of the suspensor which are transitory in the Frog. Thus
in the perennibranchiate Siren, Siredon, Proteus, and Meno-
branchus, the angle is acute ; in the Salamander and Salainan-
droid Menopoma, it is nearly a right angle ; while, in the Frogs
and Toads, and the ancient Labyrinthodonts, the angle is
obtuse.

In the lower Amphibia there is no girdle bone, the orbito-
sphenoid and the prefrontals being usually represented by
distinct bones. The frontals are distinct from the parietals,
and the maxillary and pterygo-palatine arcades become im-
perfect.

Some of the Frogs and the Coecilise — the snake-like apodal
Amphibia — have the cranial bones expanded and anchylosed
into a sort of shield, presenting apertures only for the orbits
and the nostrils ; a process which is carried still farther, by the
addition of bones not known to existing Amphibia, in the extinct
salamandroid members of the class, called Labyrinthodonts.

219

LECTURE XII.

ON THE STRUCTURE OF THE SKULL.

THE SKULLS OF REPTILIA AND AVES.

The skulls of those abranchiate Vertebrata which do not
suckle their young, and are oviparous, or ovo-viviparous, pre-
sent certain peculiarities of construction in which they all agree
with one another, and differ from the branchiate Vertebrata on
the one hand, and from the Mammalia on the other.

Thus, the basi-occipital and the basi-sphenoid are always well
developed, and the former furnishes a large part of the occipital
condyle, which is single and central.

There is no parasphenoid, or median membrane bone, under-
lying the base of the skull.

The lower jaw, each ramus of which is composed of several

pieces, articulates with the quadrate bone, as in the branchiate

Vertebrata ; but the quadrate bone articulates directly with the

cranial wall, and is not . separated from it by any structure

representing the hyomanclibular bone.

It may probably be added that the basi-sphenoid is formed by
the union of three ossifications of cartilage — one supero-median
and two infero-lateral (the basi-temporals of Mr. Parker) ; but
further research is required before this generalization can be
regarded as firmly established.*

* The caution expressed in the text seems to be no longer necessary, as my
friend Mr. Parker, who possesses a remarkably extensive knowledge of the details
of the structure and development of the vertebrate cranium, informs me that he has
now found " the median basi-sphenoid and the symmetrical basi-temporals in
Ophidians, Anguians, Scincoids, Iguanians, Geckos, Chamaeleons, Cyclodonts, Lacer-

220 OX THE STRUCTURE OF THE SKULL.

The combination of peculiarities just mentioned at once cha-
racterises the skulls of Birds and Beptiles, and distinguishes
them from all others.

In all these animals, the basi-occipital bone gives attach-
ment to a pair of ex-occipitals, which articulate, above, with
a distinct supra-occipital. The homology of these bones with
those which have received similar names in the Man's and in the
Pike's skull is not doubted ; and, indeed, their relations to one
another, and to the exits of the eighth pair, are so similar as to
allow of no discussion on this point.

Furthermore, the skulls of all Reptiles and Birds are roofed
in by membrane bones, the correspondence of which with the
parietals and frontals of Man is universally admitted ; and, in
all, there is a single or a double vomer, clearly identifiable with
that of the Man and that of the Fish. So, again, there is no
doubt about the homology of the preruaxillae and the maxillae,
the palatine and the pterygoid bones with the parts so named
in Man. Nor is it questioned that the mandible and the hyoi-
dean arches, in a general way, correspond with his. But there
has been, and is, very great divergence of opinion as to the true
nature of certain bones in the side walls of the skull, and of
some of those which enter into the composition of the maxillary
apparatus. I shall address myself chiefly to the discussion of
these debatable ossifications.

The bone of most importance among these (the misinterpre-
tation of which must needs, indeed, completely vitiate and
render worthless any theory of the vertebrate skull) is that
which lies in the side wrall of the cranium, in front of the ex-
occipital ; while it is connected below with the basi-sphenoid,
and above with the supra-occipital and parietal. In all Birds

tians, Monitors, Chelonians, Crocodiles, and in all kinds of Birds." Mr. Parker
agrees with my suggestion (supra, p. 170), that the basi-teraporals of the Sauropskla
(or Birds and Reptiles) are the homologues of the lingulx sphenoldales of Man. He
has found similar bones in numerous Mammals, and they are of especially large size
in the Mole and in the Shrew. He informs me that the Sheep has no bony centre
for the basi-sphenoid, the alisphenoids meeting in the middle line. Nevertheless
its liucjidx are well developed at the commencement of the last third of intra-
uterine life.

fac

Eth

SO.

i.ic

53 <u .C

— ~ .
E- S o
rt ° c

o ■"
w ^ 53

-9 ** &,

s o 2J

2 "•£ «

IS O -u

■" ,S O

P .22 a-

a,° *

a. _S 2

5 &'
o>

O in .
^* C3 5P

o ° *

r- — *
" O S

^^ o .

£3 ° 3
"^— ' a> '£

-— <H .

00 CO --

J «*- 2 $

^ ° ~<2

S — a;

° * r« .§

-^ c_ .-si

2° 7-»

BO _S

~ r- SJ «3

3 a;
EC cfl

y - g *

1-1 .° •-"
,— -^ -^

^ co to

•^ #0 01

^""h -IT 3

m — <S

!~ -*-» ;-. __
- T 41 —

« t> -*J —

y o ^ c

— _ i — '
f ~ co
• "- ." "*-'

3 c •
a> a> g S

-G 43 « "
^I C4-1 rt

0 o -3 -5

o © £ .3

"-J3 '5 ~.H

o a> a> r-

co a - id

rt ■» - — '

c — 3 £

^1 ■*• .£ .2

:S I § g
c -.0 2 ^

rs „ _ «

■ I §^

■S K S c

^-2-3.2

co o
00 ^_-

^ §

222 ON THE STRUCTURE OF THE SKULL.

and Reptiles the relations of this bone are essentially such as are
shown in the accompanying figures of sections of the skulls of
an Ostrich, a Crocodile, and a Turtle (Fig. 88, A, B, C). In all
these it will be observed, that the aperture for the exit of the
third division of the trigeminal ( V) lies in front of a bone, which
is notched, or perforated, by apertures for the portio dura and
portio mollis, and that the anterior part of the organ of hearing-
is lodged within this bone. Furthermore, an external view of
this region of the skull (Fig. 89, A and B) shows that the bone
in question contributes, in each case, the anterior half of the
boundary of the fenestra ovalis. In other words, the bone in
question has every essential relation of that ossification which,
in Man and in the Pike, I have termed pro-otic*

The other elements of the periotic capsule are not far to
seek, In the Turtle one of them retains its independence
throughout life, and occupies a considerable space on the exterior
of the skull, though, internally, only a small strip of it is seen in
front of the foramen for the eighth pair (Fig. 88, C). This bone
furnishes the posterior half of the frame for the fenestra ovalis,
with so much of that of the fenestra rotunda as is osseous, and it
lodges the posterior and outer part of the auditory organ. It
answers precisely, therefore, to the opisthotic.\

The corresponding ossification in most other Reptiles and in
Birds early coalesces with the ex-occipital.

The third periotic bone, the epiotic, does not remain distinct
throughout life in any Reptile or Bird, and its place appears to
be taken by a triangular process of the supra-occipital, which

* This is the bone called by Cuvier " rocher," and regarded by him and by
most of the German anatomists as the homologue of the p>ars petrosa of the human
temporal bone. I took the same view myself when I delivered the Croonian
Lecture in 1858, and I do not now substantially depart from it. For that part of
the pars petrosa which is most obvious and visible in the interior of the skull is its
pro-otic portion ; and so long as the complex nature of the pars petrosa was un-
known, the identification of the bone Pr.O in the Bird and Reptile with the petrosal
of the Mammal was the nearest approximation that could be made to the truth.
Cuvier's identification would have been absolutely correct if he had termed the
ornithic and reptilian bone not "petrosal," but "anterior part of the petrosal."

t Cuvier termed this bone the "occipital externe." Hallmann regarded it as the
equivalent of the " mastoid," and I followed him in my Croonian Lecture. In the
absence of a full knowledge of the development of the human jjars petrosa, it was
difficult, if not impossible, to see one's way to any better conclusion.

THE SKULLS OF EEPTILIA AND AVES.

223

shelters the summits of the vertical semicircular canals. l>ut
the study of development has shown that this part of the supra-
occipital is, in many, if not all, Reptiles and Birds, developed from

Fig. 89.
A

Fig. 89. — External view of the auditory region of the skull in (A) a Crocodile (C. biporcatus),
and (B) a Turtle {Chelone midas). The walls of the tympanic cavity have been cut
away in each case so far as is necessary to show the auditory fenestra ; and, in the
Turtle's skull, the semicircular canals are also partially displayed. In the Crocodile's
skull (A) F.O. is the fenestra ovalis, separated by the cochlear process of the opis-
thotic, c, from the fenestra rotunda ; Chi is the hook formed by the curved process
(6) of the opisthotic, which supports the cochlea externally. The lower end of the

224 ON THE STRUCTURE OF THE SKULL.

a separate centre, which subsequently coalesces with the supra-
occipital ; so that, just as the opisthotic in these animals ordi-
narily coalesces with the ex-occipital, the epiotic anchyloses with
the supra-occipital.

In many Reptiles, though two of the three periotic bones
coalesce with their neighbours, the suture between the three
persists on the inner surface of the skull, and is always shaped
like a Y (Fig. 88, B) ; the stem of the Y answering to that part
of the suture which separates the pro-otic from the opisthotic
ossifications, while the diverging branches of the Y correspond
with the suture between the opiotic and pro-otic in front, and
that between the epiotic and opisthotic behind. In the Turtle
an interspace filled with cartilage takes the place of the stem of
the Y (Fig. 88, C).

In the adult Crocodile the epiotic is united with the supra-
occipital, and the opisthotic with the ex-occipital ; but that
process of the opisthotic (c, Fig. 89, A) which separates the
fenestra ovalis from the fenestra rotunda (the anterior and inner
edge of which, only, is completed by bone) where it meets the
pro-otic below and anteriorly (at d, Fig. 89, A), sends down-
wards and backwards a process, which curves round the cochlea,
and, expanding to a broad plate, adjusts itself by harmonia (at b)
to the outer and lower edge of the opisthotic, and to part of the
posterior edge of the pro-otic. The anterior and inferior angle
of the broad plate is thicker than the rest, and is seen in the
interior of the dry skull, at the bottom of the stem of the Y-
shaped suture (*, Fig. 88, B). If, as has been remarked, this
part of the curved cochlear plate of the opisthotic be pressed

cochlea rests in the fossa a, formed by the basi-sphenoid and basi-occipital. The upper
end, bounded externally only by cartilage, has disappeared in the dry skull and, with
it, the outer lip of' the fenestra rotunda, the plane of which is horizontal, and nearly on
the level of the dotted line leading from Op.O in the figure, d is a small process of
the pi'o-otic, against which the bend of the curved cochlear process (6) rests. The
dotted line from b indicates the position of the suture betweeu the hinder end of that
process and the remainder of the opisthotic bone. Ca, the carotid canal ; Eu, the
upper opening of the posterior of the two canals by which each tympanum communi-
cates with the common Eustachian tube. The narrow anterior tympanic canal opens
just in front of Ca, the cleft-like aperture being traversed by the dotted line from </.
In the Turtle's skull (B) Op.O is a distinct bone from E.O, and sends down a process
between /.o., the fenestra ovalis, and/.r., the fenestra rotunda, which terminates in no
recurrent hook, but otherwise corresponds exactly with the cochlear process (c) in the
Crocodile.

THE SKULLS OF REPTILIA AND AVES.

225

upon with a point, it gives way ; but this is not because it is
merely suturally connected with the periotic bones, as has been
supposed, but because the lamina of bone by which the cochlear
plate is fixed to the opisthotic is very thin and elastic.

Among the many singular speculations which the historian
of the theory of the skull will have to record, perhaps the
strangest is that which identifies this cochlear loop, imagined
to be a distinct bone, with the entire " petrosal ' bone of the
Mammalia.

In Birds, the three periotic bones anchylose with one another,
as well as with the adjacent supra-occipital and ex-occipital, so
completely, that even the Y-shaped suture becomes obliterated
(Fig. S8, A).

The determination of the homologues of the periotic bones
in the skulls of Birds and Reptiles, and more especially of the
pro-otic, is not only a matter of first-rate importance in itself,

Fig. 90.

WL2F E(l0 so

Pmx

PrOBS Pn j
OfiO

Fig. 90. — A, a vertical and longitudinal section of the skull of a Lizard (Cyclodus). B, a
similar section of the skull of a Snake {Python), f, membranous space in the roof
of the Lizard's skull ; * *, ossifications in the internasal cartilage ; Co, the columella
of the Lizard ; Cm, the bone also called " columella," which corresponds with the
stapes.

226 ON THE STRUCTURE OF THE SKULL.

but it involves that of certain other bones of the side walls of the
skull.

In the Clielonia, and in many Lizards, the lateral walls of
the cranium, between the pro-otics and the prefrontals, are
entirely occupied by cartilage, or by membrane. In the dry
skull of the Turtle (Fig. 88, C) it is true that there is. an ap-
parent bony wall in front of the pro-otic, but this is only a
process sent down from the parietal, which becomes connected
with the pterygoid, and with a small, distinct lamella of bone.

In Lizards a distinct, rod-like bone (Co, Fig. 90, A, and
Fig. 91), occupies a corresponding position, articulating above
with the parietal, and below with the pterygoid, and receives
the name of the columella.

Fig. 91.

BS^Pt

Fig. 91. — The inter-orbital septum of a Lizard (Iguana), B S, the anterior prolongation
or beak of the basi-sphenoid ; a, the inferior median ossification; 6, the superior paired
ossification of the left side of the inter-orbital septum ; Co, the columella.

In both Chelonians and Lizards the basi-cranial axis is
laterally compressed in the presphenoidal region, and is con-
verted into a vertical inter-orbital septum, as in the Pike. In
the Chelonians, neither the septum, nor the membranous, or
cartilaginous, alisphenoidal and orbito-sphenoidal regions con-
nected with it, present any ossifications, but, in many Lizards,
delicate laminae of bone are developed in this region. In the
Iguana tuberculata, for example (Fig. 91), the inter-orbital
septum is supported below by the prolonged beak of the basi-
sphenoid. Above this, it presents a long median jn-esphenoidal
ossification (a) forked posteriorly. The forks are connected
with two bones, one on each side (b), which appear to represent
orbito-sphenoids.

THE SKULLS OF BEPT1LIA AND AYES. 227

The Crocodiles, on the other hand, possess a large and
distinct lateral ossification in front of each pro-otic (A S, Fig.
88, B). This ossification bounds the foramen for the third
division of the fifth nerve in front, and unites with the basi-
sphenoid below and with the parietal above, so far agreeing
with the alisphenoid. Since it extends so much further forward
than the alisphenoid ordinarily does, Cuvier has suggested
that it probably represents both the ali- and the orbito-
sphenoids ; but Stannius has pointed out the existence of two
small ossifications close to the optic foramina, with an in-
significant descending median stem at their bases. The former
he considers to be the orbito-sphenoicls and the latter the
presphenoid.

In these Reptiles, in the Lacertilia and in the CJiehnia, the
basis cranii, as has been already stated, is modified anteriorly
into an inter-orbital septum, as in the Pike ; but in the Ophidia,
the Cyprinoid, or Batrachian, type of skull reappears, and the
cavity of the cranium is continued without any sudden narrow-
ing, from above downwards, from its posterior to its anterior
end. In the Ophidian skull (Fig. 90, B) the side walls of the
cranium, in front of the pro-otics, are completely closed in
by bones, which might readily be taken for alisphenoids and
orbito-sphenoids ; but, according to Rathke, they are merely
downward growths of the parietals and frontals, and therefore
can have nothing to do with the true lateral cranial elements.

The anterior part of the basis cranii in Birds is always
vertically elongated into an inter-orbital septum, as in the Croco
dih'a, Lacertilia, and Chelonia. In the Ostrich (Fig. 88, A) the
presphenoid is completely ossified, but, in other members of
the class, the nature and extent of the presphenoidal ossifications
may vary greatly. The alisphenoid is always well ossified, and
occupies its characteristic position in front of the pro-otic and of
the exit of the third division of the trigeminal nerve (Fig.
88, A). The orbito-sphenoids, on the other hand, may or
may not be represented by bone. In the Ostrich they are
present, and are continuous with the presphenoid.

Reptiles possess prefrontal and post -frontal bones, which
usually remain distinct throughout life, and are admitted to

Q 2

228

ON THE STRUCTURE OF THE SKULL.

be homologous with those of Fishes, and therefore the line of
argument which identified the prefrontals of the Pike with
the lateral masses of the ethmoid in the Man is equally
applicable to the same bones in Reptiles. In Birds, the post-
frontal s have only a doubtful and exceptional distinctness, and
in them the true prefrontals seem early to coalesce with the
ethmoid. The last-mentioned cranial element is usually ossified
and appears upon the upper surface of the skull, in Birds ;
while, in Reptilia, it almost always remains cartilaginous. In
the extinct Dicynodon, however, it and the presphenoidal region
were completely ossified.

In Birds, in consequence of the prolongation of the snout

Fig. 92.

JPrf

S.Or

jptf PrO Fa FO S%

Qu.

Tr 7/

Fig. 92. — Lateral views of the skulls of (A) an Ostrich, (B) a Crocodile, and (C) a Python,
without the mandible. In the Python's skull the maxilla has also been removed. Tl,
turbinal bone of the Ophidian.

THE SKULLS OF REPTILIA AND AVES. 229

into a beak, the internasal part of the basi-facial axis acquires
a considerable size, and becomes the subject of a great variety
of ossifications, which, in many Birds, are so arranged as to
allow the anterior part of the cranio-facial axis to be moveable on
its posterior part. In many Lizards, on the other hand, the an-
terior part of the cranium is rendered moveable upon the pos-
terior in another way. The cranio-facial axis in front of the basi-
sphenoid is cartilaginous, and consequently slightly flexible,
wThile the roof of the skull between the parietals, the supra-
occipital and the periotic bones is merely membranous (f, Fig.
90, A) ; hence, the front part of the skull is capable of being
slightly raised or depressed, in a vertical plane, upon the
posterior part.

Next to the pro-otic, the squamosal and the quadrate bones
of Birds and Reptiles have been the subject of the greatest
amount of controversy among morphologists.

The bone which was originally called " os quadratum '
is that moveable facial bone of the Bird (Qu, Fig. 92, A) which
is articulated, above, with the outer side of the periotic capsule,
and especially with the pro-otic bone, and below with the
os articulare of the mandible, while, internally and anteriorly, it
is connected with the pterygoid. In the Crocodilia (Fig. 92, B)
and Chelonia, a bone, admitted by all to be the homologue of
this, is attached immoveably in the same region : in most
Lacertilia (Fig. 93) it is moveable, and remains connected
with the produced extremity of the pro-otic bone ; but, in most
Ojyhidia (Fig. 92, C) its proximal end is thrust out from the
skull upon the extremity of another bone. However, its
homology with the quadrate of the Bird is not affected by this
circumstance.

With what bone in the human skull does this correspond ?
Cuvier identified it with the tympanic of Man, and his interpre-
tation has been generally accepted ; but the tympanic is always
a membrane bone, whereas this is always a cartilage bone. The
tympanic directly supports the tympanic membrane, while this
bone sometimes gives no direct attachment to the tympanic
membrane at all. The tympanic of Mammals again becomes
smallest in those Mammalia which most nearly approach Birds

230

ON THE STRUCTUKE OF THE SKULL.

and Reptiles, and is never known to articulate, by a moveable
joint, with the malleus, which, as we have seen, is the repre-
sentative of the os articular e of the mandible of the lower
Vertebrata.

It is impossible, therefore, that the quadrate bone should be
the homologue of the tympanic of Mammalia. On the other
hand, it corresponds altogether with the quadrate bone of
Fishes, which is united in like manner with the pterygoid
arcade, and is similarly connected by a moveable joint with the
articular piece of the mandible ; and this quadrate bone of
Fishes is, I have endeavoured to show, the homologue of the
incus of the Mammalia. I make no question that, as Reichert
long ago asserted, the Bird's os quadratum and, therefore, that
of the Reptile, is the equivalent of the Mammalian incus.

It is difficult to understand how any doubt can be enter-
tained as to the bone which is the homologue of the Mammalian
squamosal in Birds. Lying above the tympanic cavity, between
the parietal, frontal, and periotic bones, is a membrane bone
(Sq, Fig. 92, A) which corresponds with the Mammalian squa-
mosal, and with no other bone in the Mammalian skull.

But if this be the Bird's squamosal, there is no difficulty in
determining that of any Reptile, the Crocodilia, Lacertilia,
Chelonia, and Ojjhidia all presenting a bone in a similar
position. It is this bone which, in most Ophidia (Fig. 92, C),

Fig. 93.

Fig. 93. — The skull of a Lizard (Cyclodus).—D D, Dentary piece of the lower jaw ;

Q ', Os quadratum ; Sq, Squamosal.

carries the quadratum as on a lever ; but, as Rathke has well
shown, the final position of the quadratum is a result of develop-

THE SKULLS OF REPTILIA AND AYF.S.

231

mental modification, the proximal end of that bone being
originally in Ophidia, as in other Reptiles, applied to the
periotic capsule.

The palatomaxillary apparatus presents a considerable
diversity of structure in Keptiles and Birds. In all Birds, and in
most Reptiles, the pterygoid and the quadrate bones are more or
less closely connected, but in the Crocodiles and Chamseleons
they are separated. In Crocodiles and CheIo?ua, and in the
extinct Plesiosauria, the quadrate bone is immoveably united
with the skull, and the other facial bones are firmly and fixedly
united with one another and with the cranium. In most Birds
and Lacertilia, on the contrary, the quadrate bone is moveably

Fig. 94.

Fig. 94. — Views of one half of the palatine surface of the skull of, A, a Lizard (Cyclodus),
and B, a Turtle (Chelone midas). N\ the posterior nasal aperture.

articulated with the skull, and its motion may be communicated
by the pterygoid, the quadrato-jugal and the jugal bones to the
fore-part of the face. This mobility reaches its maximum, on
the one hand, in such birds as the Parrots, in which the beak
and fore-part of the basi-facial axis are united by a sort of hinge

232 ON THE STRUCTURE OF THE SKULL.

with the rest of the skull ; and, on the other, in the Serpents,
in which, as has been already stated, the quadrate bone is
shifted to the end of the squamosal, and the palatine, pterygoid,
and maxillary bones are bound only by ligaments to the skull,
so that the utmost possible amount of play is allowed to the
bones which surround the mouth.

In many Reptiles a bone makes its appearance which cannot,
at present, be identified with any bone of Fishes or of Mam-
mals. This is the transverse bone of Cuvier (2V), which unites
together the maxilla with the palatine and the pterygoid.

Eemarkable differences are noticeable in the degree to which
the premaxilla is developed in the various orders of Reptiles
and in Birds. In the Snakes it is very small, or rudimentary ;
in the Lacertilia, Chelonia, and Crocodilia it has a moderate
size ; while in the extinct Ichthyosauria, and still more in Birds,
the premaxilla attains vast dimensions, completely surpassing
the maxillary element, which in Birds is reduced to a mere bar
of bone, connected by similar slender rods, which represent the
jugal and quadrato-jugal, with the outer part of the distal end of
the quadrate bone.

In the Ophidia, most Lacertilia, and Birds, the nasal sacs open
below and behind into the cavity of the mouth, by apertures
placed between the vomer and palatine bones, which correspond
with what I have termed the " median nares " in Man ; or there
is, at most, an indication of a separation between the oral cavity
and the nasal passage, produced by the sending downwards and
inwards of a process by the maxillary and palatine bones on
each side. But, in the Crocodilia (Fig. 95, B), not only the
maxillary and palatine bones, as in Man, but the pterygoid bones,
in addition, send such prolongations downwards and inwards ;
and these, meeting in the median line, shut off from the cavity
of the mouth a nasal passage, which opens into the fauces by the
posterior nares (N[, Fig. 95). The arrangement of the palatine
bones is such that, in most Crocodilia, the vomers are com-
pletely excluded from the roof of the mouth.

When a tympanic cavity exists in the branchiate Vertehrata,
it is little more than a diverticulum of the buccal cavity, con-
nected by so wide an aperture with the latter, that an Eustachian

THE SKULLS OF REPT1LTA AND AYES.

233

tube can hardly be said to exist. In the Ophidia and in certain
Laeertilia the tympanic cavities and Eustachian tubes are alto-
gether absent, and, even in the higher Laeertilia, the tympanum

can hardly be said to have definite bony walls. In the Chelonia,

Fig. 95.

7Jmx

JPm/x

die

Fig. 95. — Views of one half of the palatine surface of the skull in (A) an Ostrich, (B) a
Crocodile, (C) a Python. N, the posterior nasal aperture (median nares of Man) in
the Bird. The dotted line traverses the posterior nasal aperture, situated between
the palatine, the vomer, and the maxillary. The corresponding opening is placed be-
tween Vo and PI in the Snake. N1, the posterior nasal aperture, or posterior nares,
of the Crocodile.

on the other hand, the opisthotic and the pro-otic bones are
produced outwards so as to form the anterior and posterior
boundaries of a cavity, the antivestibulam Bojani — -which is
bounded externally by the great quadrate bone. The latter is
funnel-shaped, and deeply notched posteriorly and inferiorly.
The tympanic membrane is fixed to the margins of the funnel,
and the so-called " columella" which answers to the stapes, is
fastened by one end to this tympanic membrane, and tra-
versing the notch and entering the antivestibulum, passes to its
other insertion into the membrane of the fenestra ovalis. The

234

ON THE STRUCTURE OF THE SKULL.

Eustachian tubes have separate openings into the pharyngeal
cavity, and curve upwards and backwards from the latter round
the inferior and posterior edges of the quadrate bones to open
into the tympana.

In Birds the tympanic cavity is roofed over by the squa-
mosal, while a more or less complete floor is furnished to it by
the basi-sphenoicl, and a back wall by the produced ex-occipital
(and opisthotic ?). It may be completed in front by fibro-
cartilage or even by bone, and the memhrana tympani is fastened
to the outer margin of these boundaries of the tympanum and
not to the quadrate bone.

The Eustachian passages ordinarily traverse the basi-sphenoid,
and when they reach the base of the skull unite into a single,
cartilaginous, common Eustachian tube, which opens in the
middle line, on the roof of the mouth.

Fig. 96.

Fig. (Jfi. — Vertical and transverse sections of the left tympanic cavity of Crocodilus bipor-
catus. A, posterior, B, anterior segment; a, bristle passed into the small lateral
Eustachian passage leading from b, the posterior tympanic passage, which opens into
c, the common Eustachian passage; d, a bristle thrust into the air-passage which
traverses the supra-occipital ;/, bristle passed into the anterior tympanic passage;
<<i,Ca', carotid canal; Ct, fossa for the extremity of the cochlea; Tut, inner division
nt' the tympanic cavity.

THE SKULLS OF REPTILIA AND AVES. 235

Iii the GrocodiUa the tympanic cavities and Eustachian
passages arc still more remarkably disposed.*

The tympanic cavity of Crocodilus bvporcatus (Fig. 89, A ;
Fig. 96) is distinguishable into an inner and an outer part.
The latter is bounded by the squamosal bone, above and behind,
by the quadrate bone, below and in front. Into the former the
supra-occipital enters, above; the quadrate, and, to a slight
extent, the basi-occipital and basi-sphenoid below. To the
posterior wall of the inner division, that outward and backward
prolongation of the ex-occipital, which answers to the opisthotic
of the turtle, contributes, while the front wall is formed partly
by the quadrate and partly by the pro-otic bones.

Externally the tympanum opens by the external auditory
meatus — its internal wall is formed chiefly by the pro-otic,
opisthotic, and epiotic. The two latter are, as I have already
stated, anchylosed, respectively, with the ex-occipital and the
supra-occipital.

Each tympanic cavity communicates with its fellow of the
opposite side, superiorly, by a wide passage, which perforates
the supra-occipital bone and has a secondary diverticulum
traversing the ex-occipital. Below, the tympana communicate
with one another indirectly, by means of the common median
Eustachian tube, the aperture of which, formed, half by the basi-
sphenoid and half by the basi-occipital, is seen on the base of the
skull behind the posterior nares (c). Each tympanum communi-
cates with the common Eustachian tube by two passages : one,
wide, from the posterior and inferior part of the tympanum (b) ;
and one, very narrow, from its anterior and inferior region (/).

The two exits are separated by that part of the floor of the
tympanum which is formed by the basi-sphenoid and basi-
occipital. This presents,, behind, a hemispherical fossa (CI) into
which both the basi-sphenoid and basi-occipital enter, and, in
front, a round aperture with raised edges, situated altogether
in the basi-sphenoid (Ca). The fossa lodges the distal blind end
of the cochlea. The aperture leads into a canal, which, passing

* These were first carefully described by Professor Owen in a memoir published
in the Philosophical Transactions for 1850. Windischmann, " De Penitiori auris in
Amphibiis Structural " (1831,, has given but a very imperfect account of them.

236 ON THE STRUCTURE OF THE SKULL.

downwards and forwards in the basi-sphenoid, opens into the
pituitary fossa, and lodges the internal carotid.

The upper aperture of the anterior, narrow, passage from
the tympanum (/) is situated in front of the lip of the carotid
canal, and, at first, lies between the basi-sphenoid and the pro-
otic ; but, soon turning inwards, it enters the basi-sphenoid,
and passes beneath the carotid canal, to open into a much wider
median channel. The latter ends blindly in front and above,
behind and below the pituitary fossa ; but, inferiorly, it traverses
the substance of the basi-sphenoid, to open into the upper and
front part of the common Eustachian tube (Fig. 96, B).

The posterior, wide, passage (b) leads downwards and inwards
through the substance of the basi-occipital, and the two passages
of the opposite tympana unite to form a short median canal,
which opens, on the front face of the basi-occipital, into the
common Eustachian tube (Fig. 96, A).

The posterior tympanic passage has, however, another means
of communication with the exterior ; for, just before it joins
with its fellow, it gives off, downwards, a narrow canal, which
traverses the basi-occipital, and opens on its inferior face to the
outer side of, and a little behind, the aperture of the common
Eustachian tube (a, Fig. 96, A). There might, then, be said to
be three Eustachian tubes in the Crocodile, — two small and
lateral, one for each tympanum, and one large and median,
common to both tympana.

Where the posterior tympanic passage passes into the tym-
panum, the ex-occipital presents a round aperture wTith raised
edges, which is the anterior termination of the posterior division
of the canal for the internal carotid (C\ Fig. 96, B). In the in-
terval between this aperture of entrance and that of exit already
described, the internal carotid is unprotected by bone, and is
closely adherent to the outer surface of the cochlea ; which, held
by the cochlear hook already described, rests inferiorly upon the
fossa afforded by the basi-sphenoid and basi-occipital (CI).

The posterior wall of the tympanum also exhibits, internally,
the aperture by which the eighth pair of nerves reaches the
exterior ; externally, those by which the portio dura leaves, and
the externa] carotid enters; superiorly, between the supra-

THE SKULLS OF REPTILIA AND AVES. 237

occipital and the squamosal, is a cleft which leads to the occipital
surface in the drv skull.

The early stages of the development of the skull of a Bird
have already been described. The process of formation of the
Reptilian skull has been admirably worked out by Rathke in the
Common Snake, Coluber natrix, and I conclude this Lecture by
an abstract of his researches on this subject.*

The differences between the basis of the skull and the verte-
bral column in the earliest embryonic condition are, —

1 . That round that part of the notochord which belongs to the
head, more of the blastema, that is to be applied, in the spinal
column, to the formation of the vertebrao and their different
ligaments, is aggregated than around the rest of its extent, and —

2. That this mass grows out beyond the notochord to form
the cranial trabecule.

The lateral trabecule, at their first appearance, formed two
narrow and not very thick bands, which consisted of the same
gelatinous substance as that which constituted the whole invest-
ment of the notochord, and were not sharply defined from the
substance which lay between them and at their sides, but seemed
only to be two thickened and somewhat more solid, or denser,
parts of that half of the basis of the cranium, which lies under
the anterior cerebral vesicle.

Posteriorly, at their origin, they were separated by only a
small interval, equivalent to the breadth of the median trabecula,
and thence swept in an arch to about the middle of their length,
separating as they passed forwards ; afterwards they converged,
so that, at their extremities, they were separated by a very
small space, or even came into contact. Altogether they formed,
as it were, two horns, into which the investing mass of the noto-
chord Avas continued forwards. The elongated space between
them, moderately wide in the middle, was occupied by a layer
of softer formative substance, which was very thin posteriorly,
but somewhat thicker anteriorly. Upon this layer rested the
infundibulum ; and in front of it, partly on this layer, partly on
the trabecule, that division of the brain whence the optic nerves

* Entwickeluugsgescliiclite der Natter, 1839.

238 ON THE STRUCTURE OF THE SKULL.

proceed ; and, further forwards, the hemispheres of the cerebrum.
Anteriorly, both trabeculae reached as far as the anterior end
of the head, and here bent slightly upwards, so that they projected
a little into the frontal wall of the head, their ends lying in
front of the cerebrum. Almost at the end of each horn, how-
ever, I saw a small process, its immediate prolongation, pass
outwards and form, as it were, the nucleus for a small lateral
projection of the nasal process of the frontal wall.

The middle trabecula grows, with the brain, further and
further into the cranial cavity ; and as the dura mater begins to
be now distinguishable, it becomes more readily obvious than
before, that the middle trabecula raises up a transverse fold of
it, which traverses the cranial cavity transversely.* The fold
itself passes laterally into the cranial wall ; it is highest in the
middle, where it encloses the median trabecula, and becomes
lower externally, where it forms, as it were, a short ala pro-
ceeding from the trabecula. With increasing elongation, the
trabecula becomes broader and broader towards its free end, and,
for a short time, its thickness increases. After this, however, it
gradually becomes thinner, without any change in its tissue, till,
at the end of the second period, it is only a thin lamella, and
after a short time (in the third period) entirely disappears.

In Mammals, Birds, and Lizards, that is, in those animals in
general in which the middle cerebral vesicle is very strongly
bent up and forms a protuberance, while the base of the brain
exhibits a deep fold between the infundibulum and the posterior
cerebral vesicle, a similar part to this median trabecula of the
skull is found.

In these animals, also, at a certain very early period of
embryonic life, it elevates a fold of the dura mater which passes
from one future petrous bone to the other, and after a certain
time projects strongly into the cranial cavity. Somewhat later,
however, it diminishes in height and thickness, as I have
especially observed in embryos of the Pig and Fowl, until at last
it disappears entirely in these higher animals also, the two layers
of the fold which it had raised up coming into contact. When

* What Ratlike terms the "middle trabecula," appears to be only very indis-
tinetly developed in Fishes and Amphibia.

THE SKULLS OF REPTTLIA AND AVES. 239

this has happened, the fold diminishes in height and eventually
vanishes, almost completely.

The two lateral trabecule, which in the Snake help to form
the anterior half of the basis of the skull, attain a greater
solidity in the second period, acquire a greater distinctness from
the surrounding parts, and assume a more determinate form,
becoming, in fact, filiform ; so that the further forward, the
thinner they appear. They increase only very little in thick-
ness, but far more in length, during the growth of the head.
Altogether anteriorly, they coalesce with one another, forming a
part which lies between the two olfactory organs and constitutes
a septum. As soon as these organs increase markedly in size,
this part is moderately elongated and thickened, without, how-
ever, becoming so dense as the hinder, longer part of the tra-
becule. The prolongations into the lateral projections of the
nasal processes, which nowr proceed from the coalesced part in
question, also become but little denser in texture for the present,
though they elongate considerably.

The lateral parts and the upper wall of the cranium, with
the exception of the auditory capsules, or of the subsequent
bony labyrinth, remain merely membranous up to the end of
the second period ; consisting, in fact, only of the cutaneous
covering, the dura mater, and a little interposed blastema, which
is hardly perceptible in the upper part, but increases in the
lateral walls, towards the base of the skull.

The notochord reaches, in very young embryos of the
Snake, to between the auditory capsules ; and further than this
point it can be traced neither in the Snake nor in other Verte-
hrata, at any period of life, as manifold investigations, conducted
with especial reference to this point, have convinced me.

At the beginning of the third period, the basal plate
chondrifies, at first leaving the space beneath the middle of the
cerebellum membranous ; but this also eventually chondrifies,
and is distinguished from the rest of the skull only by its
thinness.

Lateral processes grow out from the basal cartilage just in
front of the occipital foramen, and eventually almost meet
above. They are the ex-occipitals.

240 ON THE STRUCTURE OF THE SKULL.

The tsvo lateral trabecular — parts which I have also seen in
Frogs, Lizards, Birds, and Mammals — chondrify at the beginning
of the third period. At first, they pass, distinct from one
another throughout their whole length, as far as the frontal
wall, on entering which they come into contact ; they are more
separate posteriorly than anteriorly, and they present, in their
relative position and form, some similarity with the sides of a
lyre. But as the eyes increase, become rounder, and project,
opposite the middle of the trabecular, downwards towards the
oral cavity, the latter are more and more pressed together, so
that, even in the third period, they come to be almost parallel
for the greater part of their length. Anteriorly, where
they w^ere already, at an earlier period, nearest to one another,
they are also pressed together by the olfactory organs (which
have developed at their sides to a considerable size), to such a
degree, that they come into contact for a great distance and then
completely coalesce ; they are now most remote posteriorly,
where the pituitary body has passed between them,* so that
they seem still to embrace it. Anteriorly, between the most
anterior regions of the two nasal cavities, they diverge from their
coalesced part, as two very short, thin, processes or cornua,
directed upwards, and simply bent outwards.

It has been seen above that the median trabecula does not
chondrify, but eventually disappears ; in its place, a truly
cartilaginous short thick band grows into the fold of dura mater
from the cartilaginous basal plate.

Where the pituitary gland lies, there remains between the
lateral trabecular of the skull a considerable gap, wdrich is only
closed by the mucous membrane of the mouth and the dura
mater. But there arises in front of this gap, between the two
trabecular, as far as the point where they have already coalesced,
a very narrow, moderately thick, and anteriorly pointed streak
of blastema, which, shortly before the end of the third period,
acquires a cartilaginous character and subsequently becomes the
body of the presphenoid.

* The pituitary body, however, as Rathke has since admitted, does not pass
between the trabecular, and is developed in quite a different manner from that
supposed in the memoir on Coluber.

THE SKULLS OF REPTJLIA AND AVES. 241

Altogether anteriorly, however, where the two trabeculae
have coalesced, there grows out of this part, from the two cornua
in which it ends, a pair of very delicate cartilaginous plates.
At the end of the third period both plates acquire a not incon-
siderable size, take the form of two irregularly-formed triangles,
and are moderately convex above, concave below, so as to be,
on the whole, shell-shaped. The nasal bones are developed
upon these, while below them are the nasal cavities, and the
nasal glands with their bony capsules.

The alaa, or lateral parts, of the two sphenoids do not grow,
like the lateral parts of the occipital bone, out of the basis cranii,
the foundation of which is formed by the cephalic part of the
chorda, but are formed separately from it, although close to it,
in the, until then, membranous part of the walls of the cranium.

The alae of the presphenoid (orbito-sphenoids), which are
observable not very long before the termination of the third
period, appear as two truly cartilaginous (though they never
redden), irregular, oblong, plates of moderate thickness ; lie in
front of the optic foramina, at the sides of the lateral trabecule
of the skull; "ascend from them upwards and outwards, and are
somewhat convex on the side turned to the brain, somewhat
concave on the other. The alae magnse (alisphenoids) are
perceptible a little earlier than these. They are formed
between the eye and the ear, and also originally consist of a
colourless cartilaginous substance : they appear, at the end of
the third period, as irregular four-sided plates ; lie at both sides
of the anterior half of the investing plate of the chorda ; ascend
less abruptly than the alae orbitales, and are externally convex,
internallv concave.

The upper posterior angle of each elongates, very early,
into a process, which grows for a certain distance backwards,
along the upper edge of the auditory capsule, and applies itself
closely thereto.

The auditory capsules, or the future petrous bones, chondrify,
as it would appear, the earliest of all parts of the skull : the
fenestra ovalis arises in them by resorption.

The ossification of the Snake's skull commences in the
basi-occipital, or, at any rate, this is one of the first parts to

242 ON THE STRUCTURE OF THE SKULL.

ossify. At a little distance from the occipital foramen, there arises
a very small semilunar bony plate, the concave edge or excava-
tion of which is directed forwards ; thereupon, the bony substance
shoots from this edge further and further forwards, until at
length the bony plate has the form of the ace of hearts. Its
base borders the fontanelle in the base of the skull, which lies
under the anterior half of the third cerebral vesicle, while its
point is contiguous to the occipital foramen ; for the most part
it is very thin, and only its axis (and next to this its whole
posterior margin) is distinguished by a greater thickness. The
cephalic part of the notochord can be recognised in the axis of
this bony plate up to the following period. It passes from the
posterior to the anterior end of the bony plate, where it is lost,
and is so invested by the osseous substance of the plate, that a
smaller portion of the latter lies on the upper side of the notochord,
a larger portion beneath it. On this account it forms, on the
upper side of the plate, a longitudinal ridge, which subsequently
becomes imperceptible by the aggregation of matter at the
sides. On one occasion, however, I saw, in an embryo which had
almost reached its full term, a similarly formed and sized bony
cone, which, through almost its entire length, appeared merely
to lie on the body of the basi-occipital, since it had only coalesced
with it below.

The nucleus and sheath of the cephalic part of the notochord
become gradually broken up and the last trace of them eradicated,
as the ossification of the basi-occipital proceeds, like the nucleus
and sheath of the rest of the notochord wherever a vertebral body
is developed.*

The articular condyle is not yet formed. The ex-occipitals
ossify through their whole length and breadth.

The body of the basi-sphenoid is formed between the above-
mentioned posterior fontanelle of the basis cranii and the
pituitary space, therefore far from the cephalic part of the noto-
chord. It ossifies by two lateral centres, each of which forms

* In the Stickleback it has appeared to me that the wall of the anterior conical
termination of the notochord in the basis cranii becomes ossified, or, at any rate,
invested by an inseparable sheath of bony matter, just in the same way as the
" urostyle" is developed in the tail.

THE SKULLS OF REPTILIA AND AVES. 243

a ring round the carotid canal.* The alisphenoids ossify in their
whole length and breadth ; the orbito-sphenoid only slightly, and
the presphenoid not at all. The premaxillary bone arises as an
azyffos triangular cartilage between the cornua of the anterior
ethmo-vomerine plate. It ossifies from a single centre.

The auditory capsule, or the future petrosal [— periotic]
bone, may, even at the end of this period, be readily separated
from the other part of the cranial wall, and still consists, for the
most part, of cartilage. On the other hand, the triangular form,
which it had before, is not inconsiderably altered, since it greatly
elongates forwards, and thus, as it were, thrusts its anterior
angle further and further forwards, and becomes more unequal-
sided. At the lower edge, or the longer side of it, about
opposite to the upper angle, at the beginning of this (third)
period, or indeed somewhat earlier, a diverticulum of the
auditory capsule begins to be formed (the rudimentary cochlea),
and develops into a moderately long, blunt, and hollow appen-
dage, the end of which is directed downwards, inwards and back-
wards, and also consists of cartilage. Close above, and somewhat
behind this ajDpendage, however, there appears, at about the
same time, a small rounded depression, in which the upper end
of the auditory ossicle eventually rests ; and, somewhat later, an
opening appears in this depression wThich corresponds with the
fenestra rotunda of man. Very much later, namely, towards the
end of this period, the auditory capsule begins to ossify. Ossifi-
cation commences in a thin and moderately long, hook-like
process, which is sent forwards and inwards from the lower
hollow diverticulum of the cartilage,, and unites with the basi-
sphenoid. From this point it passes upwards and backwards,
and, for the present, extends so far that, at the end of this
period, besides that process, the diverticulum in question, and
about the anterior third of the auditory capsule itself, are-
ossified, t Later than at the point indicated, an ossific centre
appears at the posterior edge of the auditory capsule, where it
abuts against the supra- and ex-occipitals, but extends from
hence by no means so far forward as to meet that from the

* These are the " basi-temporals " of Mr. Parker.
f This is the pro-otic ossification.

244 ON THE STRUCTURE OF THE SKULL.

other point.* The middle, larger part of the auditory capsule,
therefore, for the present, remains cartilaginous.

In the beginning of the fourth period, a third ossific centre f
arises in the upper angle of the capsule, whereupon all three
grow towards one another. But the mode of enlargement and
coalescence of these bony nuclei is very remarkable. They do
not unite with one another in such a manner as to form a con-
tinuous bony capsule for the membranous part of the labyrinth,
but are permanently separated by cartilagino-membranous
and very narrow symphyses. On the other hand, one [the
epiotic] coalesces, in the most intimate manner, with that edge
of the supra-occipital which is nearest to it ; so that, even in the
more advanced embryos, this bone and it form a moderately
long oblong plate, each end of which constitutes a small, tole-
rably deep, and irregularly-formed shell, containing a part of
the anterior, or upper, semicircular canal. The second bony
centre [the opisthotic] becomes anchylosed with the anterior edge
of the lateral part of the occipital bone, and also forms a small,
irregularly-shaped, but longish scale, which contains the deeper,
or lower part, of the posterior crus of that semicircular canal,
and besides this, the lower sac, or representative of the cochlea.
The remaining bony mass developed in the auditory cartilage
[the pro-otic], however, includes the greater part of the mem-
branous portion of the labyrinth, and is the largest. The same
phenomenon, viz. that the petrosal bone breaks up, as it were,
into three pieces, of which two coalesce with the occipital
bone, occurs also in Lacerta agiJis, and probably takes place in
like manner, if we mav conclude from the later condition of the
petrous bone to the earlier, in Crocodilia and Chelonia.

It has been seen that subsequent observers have fully justi-
fied the conviction here expressed by Rathke.

* The opisthotic ossification. f The epiotic ossification.

245

LECTURE XIII.

ON THE STRUCTURE OF THE SKULL.

THE SKULLS OF MAMMALIA.

We have met with no important difficulties in the way of
identifying the bones of the Bird's skull with those found in the
skulls of the Pieptilia and still lower Vertebrata ; and hence,
if the cranium of a Mammal be compared with that of a Bird,
the bones which correspond in the two will obviously be homo-
logous throughout the series.

The accompanying figure represents a longitudinal and

Fig. 97.

Fio-. 97. — Longitudinal and vertical section of the cranial cavity of a Beaver.

vertical section of the skull of a Beaver {Castor fiber), drawn of
the same absolute length as the section of the Ostrich's skull

246 ON THE STRUCTURE OF THE SKULL.

(Fig. 88, A), and exhibiting a corresponding extent of the cranium.
The three segments of the basi-cranial axis are at once recog-
nisable and identifiable with the basi-occipital, basi-sphenoid, and
presphenoid of the Bird ; but the basi-sphenoid is truncated at its
anterior end, as in the Crocodile, not produced into a long beak,
as in Birds and many Lizards. The ex-occipital and supra-occi-
pital bones, again, have all the relations of those of the Ostrich,
and are universally admitted to be the homolognes of the latter.

In the Ostrich, as we have seen (Fig. S8, A), there lies in front
of the ex-occipitals a large bony mass, composed of the confluent
opisthotic, epiotic, and pro-otic bones. The inner face of the
single periotic bone thus formed is divided into two surfaces, one
anterior and one posterior, by a ridge which runs somewhat
obliquely from above downwards and forwards. The anterior
surface is concave, looks somewhat forwards, articulates in front
with the alisphenoid, and contains no part of the organ of hear-
ing ; the third division of the trigeminal nerve passes out in
front of it. The posterior surface presents, inferiorly, the aper-
tures for the jportio dura and the jportio mollis ; superiorly and in
front, a fossa arched over by the anterior vertical semicircular
canal ; while, superiorly and behind, it contains the posterior
vertical semicircular canal. Between the posterior edge of this
division of the bone and the ex-occipital the eighth pair of
nerves leaves the skull.

In the Beaver (Fig. 97), there is a single mass of bone not
dissimilar in form and proportional size, which has always been
admitted to be the homologue of the pars jpetrosa and jmrs mas-
toidea of Man, and the general relations and characters of which
may be described in exactly the same terms. The inner face is
divided into two surfaces, one anterior and one posterior, by a
ridge which runs, somewhat obliquely, from above downwards and
forwards. The anterior surface is concave, and looks slightly
forwards ; it articulates in front with a bone which, as all agree,
corresponds with the alisphenoid of Man, and lies behind the
exit of the third division of the trigeminal. The posterior divi-
sion presents, inferiorly, the apertures for the portio dura and
portio mollis ; superiorly and in front, a fossa arched over by the
anterior vertical semicircular canal ; while superiorly and behind

THE SKULLS OF MAMMALIA. 217

it contains the posterior vertical semicircular canal. Between
the hinder edge of this division and the ex-occipital, the eighth
pair of nerves leaves the skull.

The inferior, or internal, edge of the periotic bone in the Bird,
and that of the pars petrosa in the Beaver, comes into relation
with the basi-occipital and basi-sphenoid ; externally, each
exhibits the fenestra ovalis and rotunda, and is related, above,
to the squamosal.

In fact, the only noteworthy differences between the ornithic
periotic, and the Mammalian pars petrosa and mastoidea, are that
the former becomes anchylosed with the adjacent bones of the
cranial wall, while the latter remain separate from them ; and
that, while the periotic articulates, above, with the parietal in
the Bird, the corresponding ossification in the Mammal is
separated from that bone by the squamosal.

On the former distinction it would of course be absurd to lay
any weight ; as regards the latter, it is deprived of all signi-
ficance by the circumstances that in some Birds — as, e.g., the
common Fowl — the squamosal interposes between the periotic
and the parietal in the wall of the skull ; and that in some Mam-
mals— as, e.g., the Sheep — the squamosal is completely excluded
from the skull, and the pars petrosa unites with the parietal.

The simple anatomical comparison of the parts appears, then,
to be amply sufficient to demonstrate, that the pars petrosa and
mastoidea of the Beaver correspond in every essential respect
with the periotic mass of the Bird, and therefore with the pro-otic,
opisthotic, and epiotic bones of Keptiles and Fishes. On the
other hand, no one has ever doubted that the petrosal and mas-
toid of the Beaver answer to the petrosal and mastoid of Man ; and
therefore we are led by the comparison of adult structure, merely,
to exactly the same conclusion as that at which we arrived by the
study of development, to wit, that the pars petrosa and pars mas-
toidea of Man answer to the periotic bones of the lowrer Vertebrates.

In front of the periotic, the side wrall of the cranium is formed
by an alisphenoid, anchylosed below with the basi-sphenoid ; and,
still more anteriorly, by a large orbito-sphenoid, united inferior! y
with the presphenoid, which is distinct from the basi-sphenoid
behind, and from the ethmoid in front.

248

ON THE STRUCTURE OF THE SKULL.

In the roof of the skull (Fig. 98) a large inter-parietal, SO1,
which corresponds with the upper part of the squama occipitis
of Man, occupies an interval left, posteriorly, between the two

Fig. 98.

Fie. 98. — Side view of the skull of a Beaver. Ty, tympanic bone ; M, pars mastoidea ;
Pm, the downward process of the ex-occipital, called " paramastoid."

parietals ; otherwise, the bones correspond with those found in the
roof of the skull of the Bird. The ethmoid, the vomer, the nasal
bones, the prernaxillae, maxillse, lachrymal, jugal, and squamosal
bones, the palatines, and the pterygoids of the Boclent, present
no difficulties to the student acquainted with the structure of the
Bird's and Keptile's skull ; but he will miss the pre-frontals, the
post-frontals, the quadrato-jugal, the transverse, and the quad-
rate bones, together with all the pieces of the lower jaw, save
the dentary.

The post-frontals, the quadrato-jugal, the transverse, and
four out of the five missing pieces of each ramus of the lower jaw,
appear to have no representatives in the Mammalian skull.

The pre-frontals, on the other hand, are represented by the
so-called " lateral masses of the ethmoid," with their develop-
ments, the superior and middle ethmoidal turbinal bones, wdiich
answer precisely to those of Man. A third turbinal, developed
from the primitive cartilaginous wall of the olfactory chamber,
eventually becomes united with the maxilla, and answers to the
inferior or maxillary turbinal of Man ; while, in the Beaver, there
is a fourth turbinal, connected with the superior turbinal and with
the nasal bones, which may be termed the " nasal turbinal."

How far these well-defined turbinals of the Mammal answer
to the cartilaginous and osseous turbinals of Birds and Keptiles,
is a question wThich has yet to be elucidated.

THE SKULLS OF MAMMALIA.

249

Fig. 99.

I have already endeavoured to show that the quadrate and
articular bones of the oviparous Vertebrata are represented by
the incus and malleus of Man, and consequently by the corre-
sponding bones in all Mammalia ; and that, as a consequence of
the appropriation of two bones of the mandibular series to the
uses of the organ of hearing, the dentary bone develops its
own condyle, and articulates with the squamosal.

Another bone which appears to have no distinct representa-
tive in most oviparous Vertebrates* is very conspicuous in most
Mammals, and far more so in the Beaver than in Man. This
is the tympanic element, and it will be useful to study with
especial attention the characters of this bone, its relations to the
periotic, and the manner in which both are connected with the
other bones of the skull.

In a transverse section of the conjoined tympanic and periotic
bones, taken through the canal which is common to the anterior
and posterior vertical semicircular
canals (Fig. 99), the periotic mass
is seen to be prolonged, external to
and below the horizontal semicircular
canal and that for the passage of the
portio dura, into a stout " mastoid
process " (If), which appears upon the
outer surface of the skull, between the
ex-occipital, the squamosal, and the
tympanic, as a production downwards
and outwards of the "pars mastoidea"
which is doubtless, as in Man, com-
posed partly of the pro-otic and partly
of the epiotic and opisthotic bones.

The tympanic bone is produced, externally, into a spout-like
tube, directed forwards and upwards, which is the external audi-
tory meatus (Au, Fig. 98) ; below and internally the tube dilates
into a thin walled hemispherical bulla (b, c, Fig. 99), open su-
periorly, and produced in front and anteriorly into a perforated
process, which contains the osseous part of the Eustachian tube.

* I learn from Mr. Parker that all Birds above the Strutliionidse have a more or
less perfect chain of tympanic bones, of which there are six in Corvus corone.

Ficr. 99. — A vertical section of the
conjoined tympanic and periotic
bones of the Beaver (Castor fiber) .
a, external auditory meatus ; b,
groove for the tympanic mem-
brane ; c, the inner lip of the
tympanic ; Eu, Eustachian tube ;
CI, cochlea ; M, pars mastoidea.

250 ON THE STRUCTURE OF THE SKULL.

The outer lip of the bulla and the auditory meatus are anchylosed
with that region of the pro-otic which corresponds with the legmen
tympani in Man. The inner lip of the tympanic bulla is, as is the
case with the corresponding edge of the tympanic bone of Man,
applied against the opisthotic, but it does not anchylose with this
bone in the Beaver ; at any rate, for the greater part of its extent.
Consequently, a very narrow cleft or fissure, leading into the tym-
panum, is opened up, if the inner lip of the bulla is gently prized
away from the periotic mass in this region. I shall term this
the " tympano-periotic fissure." The great difference between the
tympanic bone of Man and that of the Beaver arises from the
circumstance, that, in Man, by far the greater part of the bone
is occupied by the external auditory meatus ; the interval be-
tween the groove for the attachment of the tympanic membrane
and the inner edge of the tympanic bone — which forms the floor
of the tympanum — being quite insignificant, except in the region
of the Eustachian tube. In the Beaver, on the other hand, this
part of the tympanic bone is greatly enlarged, and constitutes
more than the inner half of the bulla tympani.

The tympanic bone and the periotic being thus anchylosed
together externally (though only coadjusted internally), form one
bone in the adult Beaver. But this " tympano-periotic bone " is
not anchylosed with any of the adjacent bones, even the squa-
mosal remaining perfectly distinct. Nor, indeed, is it fixed to
them by very firmly interlocking sutures, so that in the dry skull
it may be pushed out without difficulty. It is held in place, in
fact, only by the descending post-auditory process of the squa-
mosal (answering to the posterior part of the mar go tympanicus),
which curves behind the external auditory passage ; and by the
fitting in of the " pars mastoidea " between the ex-occipital and
supra-occipital.

Of the vast multitude of modifications undergone by the
Mammalian skull, I select for comment, in this place, only a few
of the most important, such as, lstly, those which are the result
of unusual forms or combinations of bones in skulls not other-
wise abnormal. 2ndly. Those which are exhibited by the skulls
of the higher Mammals as compared with the lower, ordly.

THE SKULLS OF MAMMALIA. 251

Those which are presented bv what niav be termed aberrant
Mammalian skulls, e.g., the crania of the Monotremata and
Proboscidia, and of the aquatic Mammalia — the Sirenia, Phocidast
and Cetacea.

I am not aware that there is any example among the Mam-
malia of the bones of the roof, or lateral walls, of the two pos-
terior segments of the skull taking a share in the formation of the
floor of the cranial cavity. On the other hand, a careful study
of development will probably show that it is no uncommon
circumstance for the orbito -sphenoids to unite together in the
middle line, so as to exclude the presphenoid from the cranial
floor, or even to supply its place entirely.

A still more remarkable deviation from the typical arrange-
ment than this occurs in certain Mammals, and has been thus
noted by Cuvier (Lecons ii., p. 319) : — "La lame cribleuse de
l'ethmoide dans tous les Maids, dans les Loris, et les Galagos,
vient toucher comme dans l'homme, au sphenoide anterieur;
tandisque, dans les Singes, elle en reste eloignee en arriere par le
rapprochement des deux cotes du frontal."

I find the union of the frontals to which Cuvier refers in this
passage to take place in Cynocephalus, Macacus, Cercopithecus,
and Semnojnthecus. The frontals, however, do not really separate
the presphenoid and ethmoid, but only form, above the junction
of these two bones, the front part of a thick osseous bridge, the
hinder part of which is contributed by the orbito-sphenoids.

The Gorilla agrees with the Monkeys and Baboons in these
respects. Thus, in the adult male Gorilla in the Museum of
the Koyal College of Surgeons, the distance from the anterior
boundary of the sella turcica to the anterior end of the cribriform
plate of the ethmoid is 14 in. Of this extent of the base of the
skull, 0*35 in. is occupied by the conjoined orbito-sphenoids,
042 in. by the coalesced frontals, and 0*6 in. by the lamina
peiyendicularis of the ethmoid. But, in a vertical section, the
ethmoid is seen to extend back under the basi-cranial processes
of the frontals (which are not more than one-fifth of an inch
thick) as far as the suture between the orbito-sphenoids and
these processes, which end anteriorly in a free rounded, trans-

252 ON THE STRUCTURE OF THE SKULL.

versely concave ridge, constituting the posterior boundary of
the olfactory fossa. Laterally, the basi-cranial processes of the
frontals arch downwards for a short distance and unite with the
lateral masses of the ethmoid.

In the Gorilla, the frontal bridge is much smaller than in
the lower Catarhines. The Chimpanzee approaches Man still
more nearly ; a triangular process of the presphenoid inter-
posing itself between the frontals and joining the ethmoid.
Sometimes, however, very small processes of the frontals just
unite over this junction. In the Orang, the frontals are widely
separated, as in Man.

The epiotic, pro-otic, and opisthotic bones are always an-
chylosed into a single periotic bone in the Mammalia ; but they
unite with the other elements of the temporal bone, and with the
adjacent cranial bones, in very various modes, and the tympanic
cavity presents very different boundary walls in different Mammals.

In the Beaver, as we have seen, the tympanic and periotic
bones are anchylosed into a single " tym'pano-'periotic" which
remains unanchylosed with the squamosal, and is easily de-
tached. In the Sirenia and in Cetacea (sooner or later) the
same anchylosis takes place, but the tympano-periotic is still
less firmly fixed in its place, and, in some Cetacea, does not
appear at all in the interior of the skull, in consequence of the
growth of the adjacent bones towards one another over it.

The tympano-periotic of the Rhinoceros, Horse, and Sheep,
long remains unanchylosed to the surrounding bones, but is so
wedged in between them as to be practically fixed within the
walls of the skull.

In Echidna and in Orycterojms the periotic, the squamosal,
and the tympanic remain perfectly distinct for a long time, if
not throughout life.

The squamosal and tympanic of the Pig anchylose into a
single " ' squamoso-tympanic" which is firmly fixed to the ad-
jacent bones ; but the periotic remains free, and consequently
readily falls not out of, but into the skull.

In the nine-banded Armadillo (Praopus) it is the periotic
and squamosal which are anchylosed, the tympanic remaining

THE SKULLS OF MAMMALIA. 253

rudimentary and free ; and the Opossums and the Tapir exhibit
a similar arrangement.

Other Mammalia, such as the Camivora and Primates, have
the squamosal, tympanic, and periotic all anchylosed together
into one " temporal bone."

Even in one and the same order the constitution of the
tympanic cavity exhibits the most remarkable differences. To
take the Edentata as an example : —

In the Oryderopus the walls of the tympanic cavity have a
wonderfully reptilian arrangement ; the periotic is very large in
proportion to the other bones of the skull, and its plane presents
comparatively little inclination, so that its exterior face looks
more outwards than downwards. A large part of its posterior and
outer face is seen, as a pars mastoidea, upon the exterior of the
skull, between the supra-occipital, the ex-occipital, and the squa-
mosal, but there is no distinct " mastoid process ;" below, the peri-
otic comes into contact with the basi-occipital and basi-sphenoid ;
in front, with the alisphenoid. The latter bone is strongly convex
outwards, so as to present a posterior, as well as an external, face ;
the posterior face forms the front wall of the tympanum, and ex-
hibits a somewhat deep excavation, or alisphenoidal air-cell.

The squamosal, a very large bone, is divided by a well-
marked ridge into an upper face, which constitutes part of the
roof, and an outer face, which forms a portion of the lateral
wall, of the skull. The latter enters into the outer and upper
wall of the tympanum ; the former, very thin, constitutes the
roof of that cavity, abutting internally upon the supra-occipital
and parietal. The Fallopian canal is open for the greater part
of its extent, and a hook-like osseous process, which overhangs
its outer and posterior part, gives attachment to the hyoid.

The tympanic is a strong hoop of bone, incomplete above,,
and much shorter anteriorly than posteriorly. By its expanded
anterior end it articulates by an interlocking suture with the
squamosal. The thin posterior end is free.

In Myrmecophaga tetradactyla (and essentially the same
arrangement obtains in the great Ant-eater), the squamosal, as
in Oryderopus, enters largely into the wall of the cranial cavity ;
but the tympanic, which is large and bullate, is anchylosed with

254 ON THE STRUCTURE OF THE SKULL.

it. The tympanic, however, forms only the outer part of the
posterior wall of the tympanum, the inner and posterior walls
of that cavity being furnished by a downward process of the basi-
occipital, while its inner and anterior wall is formed partly by
the pterygoid and partly by the alisphenoid. These two bones
enclose a great air-cell, which communicates freely with the
tympanic cavity behind. In front, it is closed by a thin bony
partition, which separates it from a second large air-chamber,
enclosed, partly by the alisphenoid and pterygoid, and partly by
the palatine.

In the genus Mams there is a large bulla, formed altogether
by the tympanic, which, in moderately young skulls, at any
rate, is not anchylosed with the adjacent bones.

The squamosal is an immense bone, extending from the ex-
occipital to the orbito-sphenoid, and entering into the lateral
walls of the skull for that extent. Its posterior part, dilated
and convex outwards, contains a large air-cell, which opens into
the roof of the tympanum by a wide aperture. The plane of the
periotic is nearly horizontal. It is a relatively small bone, and
only a small part of it appears on the base of the skull, behind
the tympanic bulla, the squamosal completely hiding it externally.

Of the Armadillos, some, like Eujyhr actus, have a tympanic
bulla of the ordinary construction, with, occasionally, a very
long external auditory meatus ; while others, such as the nine-
banded Armadillos (Praojms of Burmeister), have a mere hoop
of bone open above, almost as rudimentary as that of Echidna.

Or, if we turn to the perissodactyle TJngulata : —

In the Rhinoceros, the periotic and tympanic early anchylose
together, but remain distinct from the surrounding bones, the
compound tyinpano-periotic being only wedged in between the
squamosal, ex-occipital, and other adjacent cranial bones, in
such a manner that it cannot fall out. The "pars mastoidea'''
is completely hidden, externally, by the union of the squamosal
and the paramastoid process of the ex-occipital over it. The
region itself, however, is very well developed, and is continuous,
internally and below, with a very strong, conical, somewhat
curved, styloid process, to the flattened, free base of which the
hyoidean apparatus is attached.

THE SKULLS OF MAMMALIA. 255

The tympanic element is very singularly formed. It lias
the shape of a very irregular hoop, open above and behind, and
much thicker at its anterior superior than at its posterior su-
perior end. The former, irregular and prismatic, is anchylosed
with the periotic, just behind and above the auditory labyrinth ;
it then splits into two divisions, an anterior and inner and a
posterior and outer. The anterior, acquiring a thick and
spongy texture, curves round to form the front part of the wall
of the tympanum, and then ends in a free, backwardly-directed
apex, without becoming in any way connected with the periotic,
or with the posterior division. The latter, much thinner and
denser, curves downwards and backwards in the same way, and
also remains perfectly free, but its hinder end is prolonged into
a flat process, which bends for a short way round the base of
the styloid process. The outer wall of the tympanum is there-
fore very incomplete in the dry skull, opening forwards and
downwards, first, by the fissure between the anterior branch of
the tympanic and the periotic ; and, secondly, by the cleft
between the two divisions of the tympanic.

Posteriorly, there is a large irregular aperture between the
hinder end of the anterior branch of the tympanic and the peri-
otic. Externally, there is no bony auditory meatus — or rather
the merest rudiment of one.

The Horse presents a very different structure. There is a
tympano-periotic bone which is wedged in between the squa-
mosal and adjacent bones, and not anchylosed therewith ; but
the pars mastoidea appears largely on the outside of the skull
between the post-auditory process of the squamosal and the
paramastoid, and the tympanic element consists of a complete
hulla, with a long external auditory meatus. The styloid pro-
cess is almost completely infolded by a vaginal process furnished
by the auditory meatus, and the tympanic is altogether anchy-
losed to the periotic, posteriorly.

No Tapir's skull which I have examined has presented any
trace of an ossified tympanic bone.*

In the Horse, most Primates, Carnivores and Rodents, the

* According to Cuvier, " L'os de la caisse ne paroit jamais bien se souder avec
les os voisins et tombc aisement, comme dans l'herisson, le sarigue," &c.

256 ON THE STRUCTURE OF THE SKULL.

tyrnpano-periotic fissure is closed, either by the close apposition,
or by the actual anchylosis, of the inner lip of the tympanic to
the periotic.

But, in the Sheep and Pig, this fissure is replaced by a wide
elongated aperture, the inner edge of the tympanic bulla being
rolled in like a scroll. In the Seals and Cetacea the scroll-like
form of the immensely thick tympanic bulla becomes still more
marked, and the tympano-periotic fissure wider ; while the latter
is converted into a great gap in the floor of the tympanum in
Orycteropus and in the Sirenia, tie tympanic being reduced to
a mere thick hoop.

In many Marswpialia the alisphenoid dilates posteriorly and
inferiorly into a funnel-shaped, thin-walled, bony chamber,
whicli closes the tympanic cavity anteriorly, uniting by its edges
with the tympanic bone. In certain Insectivora, such as the
Hedgehog and Tenrec, the tympanic cavity is partly walled in
by a process of the basi-sphenoid.

In Hyrax, and in many Marswpialia and Rodents, the jugal
enters into the composition of the glenoid facet for the lower
jaw. In the Marsupials the alisphenoid may also contribute
towards the formation of this articular surface. In almost all
Marsupials the angle of the mandible is continued inwards into
a horizontal plate of bone. This " inflexion of the angle of the
jaw" is peculiar to these Mammals.

The palatine and pterygoid bones present very considerable
differences in their connections among Mammalia.

Thus, in the Ornithorhynchus, in the larger Myrmecophagae,
and in some Cetacea, the pterygoids unite in the middle line
below, so as to prolong tne bony palate beyond the palatines, as
in the Crocodiles. In the Marsupials, on the other hand, the
bony palate, formed only by the maxilla? and palatines, is often
defectively ossified, so that large open spaces are left therein
on the dry skull.

In order to understand the changes which the normal type
of skull undergoes in the Mammalian series, it is necessary to
define a few lines and planes by the help of certain well-marked
organic fixed points.

THE SK ULLS OF !\I A.MMALIA. 257

A line drawn from the hinder extremity of the basi-occipita]
to the uppermost part of the junction between the presphenoid
and the ethmoid, may be called the line of the axis of the basis
cranii, or the " basi-cranial liner

A second line, drawn from the premaxilla to the basis cranii
through the junction of the vomer with the ethmoid, traverses
the axis of the facial part of the skull, and may be termed the
line of the axis of the basis faciei, or " basi-facial liner This
line, if produced upwards and backwards, will cut the foregoing
so as to form an angle open downwards, which I shall term the
" craniofacial angle."

A third line, drawn from the end of the basi-occipital bone
to the posterior edge of the supra-occipital in the median line,
will give the general direction of the plane of the occipital
foramen, or the occipital plane. The angle it forms with the
basi-cranial line is the " occipital angle"

A fourth line, drawn from the torcular Herophili, or junction
of the lateral and longitudinal sinuses, through the middle of a
plane joining the tentorial edges of the pro-otic bones, will give
the general direction of the tentorium, or, in other words, of the
demarcation between cerebrum and cerebellum.* This line,
therefore, may be taken to indicate the " tentorial plane." The
angle it forms with the basi-cranial line is the " tentorial angle."

A fifth line, drawn through the median junctions of the crib-
riform plate of the ethmoid, with the frontal above and ante-
riorly, and with the presphenoid below and posteriorly, will give,
in the same general way, the u olfactory plane." The angle it
forms with the basi-cranial line is the " olfactory angle."

Lastly, the longest an tero -posterior measurement of the
cavity which lodges the cerebrum will give the " cerebral length."

Having defined these lines and planes, the following general
rules may be laid clown : —

1. The lower Mammalia have the basi-cranial line longer in
proportion to the cerebral length than the higher. Taking the
length of the basi-cranial line as 100, 1 have observed the cerebral

* Of course no straight line can give this boundary with exactness, as the co-
adapted surfaces of the cerebrum and cerebellum, and consequently of the interposed
tentorium, are curved in all directions.

258 ON THE STRUCTURE OF THE SKULL.

length to be, in a well-developed European skull, 266 ; in a
Negro, 236 ; in an adult female Chimpanzee, 180 ; in an adult
male Gorilla, 170; in a Baboon, 144; in a Lemur, 119; in a
Dog, 87 ; in a Beaver, 70 ; in a Thylacinus, 60 ; in an Opossum,
93; in Echidna, 100.

2. In the lower Mammalia the olfactory, tentorial, and occi-
pital angles nearly approach right angles ; or, in other words, the
corresponding planes are nearly vertical, while they become more
and more obtuse in the higher Mammals, until, in Man, these
planes are nearly horizontal, in the ordinary position of the skull.

3. In the lower Mammalia (Fig. 100) the cranio-facial angle
is so open as to reach 150° or more, but, in the higher Mammalia,
it becomes smaller and smaller, until, in Man, it may be as little
as 90°.

4. In many of the lower Mammalia, a sudden narrowing of
the front part of the cranial cavity indicates the boundary
between the chamber which lodges the cerebral hemispheres and
that which contains the olfactory lobes of the brain (Fig. 100),
and the latter cavity forms a large and distinct olfactory fossa.
In the higher Mammals this cavity becomes absolutely and
relatively smaller, until in Man it is so shallow and insignificant
as to be hardly noticeable.

5. In many lowrer Mammalia the olfactory fossa is altogether
in front of the cerebral cavity, and the cerebellar fossa is alto-
gether behind it, the three being separated by marked con-
strictions (Fig. 100).

Fig. 100.

Pirtx

Fig. 100.— Longitudinal and vertical section of the skull of Tlvjltcinus cynoceph'ih**.

In the higher Mammals, on the other hand, the excessive
development of the cerebral hemispheres causes the cerebral
chamber to overlap the olfactory fossa in front and the cerebellar

THE SKULLS OF MAMMALIA. 259

fossa behind ; so that these come to be placed respectively
under, instead of in front of and behind, the cerebral chamber.

Thus it may be said, that in passing from the lower to the higher
forms of Mammalian skull, Ave find the cavity for the cerebrum
enlarging in proportion to the basi-cranial axis, and thrusting
the olfactory plane downwards and forwards, the tentorial and
occipital planes downwards and backwards, in such a manner
that these may be said to rotate on the ends of the basi-cranial
axis ; at the same time, the basi-facial line rotates on the basi-
cranial line, being more and more bent downwards and backwards.

It must be clearly understood that I by no means intend to
suggest that all Mammalian skulls can be arranged in a series,
the lower members of which shall be distinguished from the
higher by always exhibiting smaller olfactory and occipital angles,
larger cranio-facial angles, less proportional cerebral lengths, &c.
On the contrary, the various angles and measurements show a
considerable range of irrelative variation ; as, for example, in
the Cetacea, a relatively large cerebral length is associated with
small occipital and olfactory angles, and a very large cranio-
facial angle; in the Edentata and Monotremata a somewhat-
large olfactory angle is associated with a small tentorial and
occipital angle ; and in the Opossum and Echidna the cerebral
length is anomalously great. All that can be said is, that the
crania of the higher orders of Mammals, as a whole, are dis-
tinguished from those of the lower orders by the characters
I have mentioned.

The skull of Echidna (Figs. 101 and 102) may be taken as an
example of the " aberrant " monotreme type of skull. It is com-
posed of a pyriform cranium proper, and a produced, beak-like
maxillary portion. The lower jaw is remarkable for its length
and slenderness, and the very small vertical height of its
rami (Fig. 101).

The basi-occipital (B.O.) is very wide, and so much depressed
as to be quite a thin lamella of bone ; it contributes, to a small
extent, to each occipital condyle, which, like the ex-occipital
bone itself, is very large. The ex-occipitals are connected above
by a wide supra-occipital, which extends so far upon the roof of

s 2

260

ON THE STRUCTURE OF THE SKULL.

the skull that the lambdoidal suture is not very distant from its
summit. The plane of the roof of the skull slopes upwards and
forwards, from the occipital foramen to this point.

Ficr. 101

3/je

Fig. 101,

-The skull of a young Echidna viewed from without, and in longitudinal and

vertical section.

The large parietals, anchylosed together in the middle line,
form all but a very small portion of the rest of the roof of the
skull, and are succeeded by the small frontals. These are met
in the middle line, inferiorly, by the lamina perpendicularis of
the ethmoid, which separates one olfactory chamber from the
other, and are united by sutures, anteriorly, with the long nasals.
These stop short of the anterior nasal aperture, being excluded
therefrom by the premaxillaries.

In the base of the skull the basi-sphenoid, presphenoid, and
ethmoid are anchylosed together. The basi-sphenoid is a wide,
flattened bone, somewhat deflexed at the sides. Its long, thin,
postero-lateral margins articulate externally with the broad, flat
bones (Pt) which contribute above to form the floor of the cranial
cavity by filling up a vacuity which would otherwise exist between
the basi-sphenoid, periotic, and alisphenoid. The thick posterior
and external edges of these bones are excavated by a deep groove,
which forms the front wall of the tympanum and of the Eusta-
chian tubes. The palatine bones are completely anchylosed
with the sphenoid, and pass abruptly inwards from the outer
edges of that bone (Fig. 102). The anterior and internal edges of

mi: skulks of mammalia.

261

the bones (Pt)< which obviously represent the pterygoids, articu-
late with them as well as with the basi-sphenoid. The anterior
and external edges of the pterygoids are united with an antero-
lateral prolongation of the pro-otic part of the periotic; and,
rather above the cleft between the latter and the pterygoid, is
fixed the large process of the malleus (m, Fig. 102), to which the
tympanic ring closely adheres.

The periotic bone is remarkable for the lamellar prolongations
which it sends forwards from its pro-otic, epiotic, and opisthotic
regions, beyond the space required for the auditory organ, and
which enter more largely into the side walls of the skull than
any of its ordinary constituents. The periotic contributes
towards the floor of the skull by a triangular process, which it
sends in between the basi-occipital and the basi-sphenoid. Pos-
teriorly, it articulates largely with the ex-occipital, the foramen
for the eighth pair being situated between it
and the latter. By its wide superior prolonga-
tion it unites behind with the ex-occipital, pos-
teriorly and superiorly with the supra-occipital ;
anteriorly and superiorly, first with the parietal,
and then with a large bone (OS, Fig. 101) which
stretches outwards, upwards, and backwards from
the presphenoid and ethmoid, articulating partly
with the frontal, and more extensively with a&j
the parietal. Except in its unusual articula-
tion with the periotic, this bone corresponds Vo-
with the orbito-sphenoid. Between the superior pi-
prolongation of the periotic, and its thin and
imperfectly-ossified anterior and inferior pro- Bs\
longation, there is an interspace filled up by
the squamosal. The lower edge of this pro-
longation articulates with the pterygoid, and,
in front of this, forms the upper boundary of

the foramen for the third division of the fifth Fig-. 102- — Under
nerve. Between its front edge and a small
process, sent up by the palatine towards the
orbito-sphenoid, is a small plate of bone, which alone seems to
represent the alisphenoid.

half of the skull of
Echidna.

262 OX THE STRUCT UK E OF THE SKULL.

The premaxilke enter largely into the composition of both
the upper and under regions of the snout. As has been already
stated, they unite in front of the nasal bones, so as to exclude
the latter from the anterior nares, as is the case in some Croco-
dilia. The maxillary bones send horizontally inwards a broad
and long palatine process. This, like the corresponding process
of the palatine bone, is separated from its fellow in the middle
line, for some distance, by the vomer. On the left side of the
specimen from which this description is taken there is a distinct
large triangular lachrymal (Fig. 101) ; it is imperforate, and
situated altogether upon the side of the face. An oblique suture
extends downwards and forwards from that which separates this
lachrymal, inferiorly, from the adjacent bones, and seems to mark
off the jugal from the maxillary bone. On the right side neither
this suture exists, nor oxiy indication of a distinct lachrymal.

The essential characters of the Proboscidean cranium are
best displayed in the foetal Elephant, as the sutures become obli-
terated, and the true form of the skull is disguised by the
enormous development of the air-chambers between the tables
of the skull, in the adult.

Fig. 103 represents the longitudinally and vertically bisected
skull of such an Elephant. The whole basi-cranial axis is
slightly concave upwards. The basi-occipital and the basi-
sphenoid, the presphenoid, and the ethmoid are already so com-
pletely anchylosed that the traces of their primitive distinctness
have almost disappeared. On the other hand, the presphenoid
and the basi-sphenoid are widely separated by the remains of a
synchondrosis. The occipital angle is about 90°, the olfactory
angle 160° to 170°.

The frontals enter as much into the front wall as into the
roof of the skull, and extend largely down upon its sides. An-
teriorly and externally they are prolonged into great arched
supra-orbital processes, which form the roofs of the orbits.

The parietals are narrower in the middle line of the vertex
than anywhere else, being encroached upon by the frontals,
anteriorly, and by the supra-occipital behind. Infero-laterally,
the parietals widen out very much and extend far down into tin-

THE SKULLS OF MAMMALIA.

2(33

temporal fossae, where they unite, in front, with the apices of the
tolerably large orbito-sphenoids, and behind, with the periotic
and supra-occipital. Below the inferior margin of the parietals
the squamosals appear largely in the lateral wall of the skull.

The alisphenoids are very small, and are directed horizon-
tally outwards. The foramen for the exit of the third division
of the trigeminal is between the hinder margin of the bone and
the periotic.

The latter bone has a considerable proportional size, and is
devoid of any cerebellar fossa.

Fig. 103.

JPet

Fig. 103.

no -2W

-Longitudinally and vertically bisected cranium of a foetal Elephant (Elephas

Indians).

On the exterior of the skull the squamosal joins the ex-occi-
pital, so that no "pars mastoidea" appears upon the surface.
The post-glenoidal and post-auditory processes of the squamosal
are very large, and bend towards one another inferiorly, so as to
meet (in the adult skull) and form a spurious external auditory
meatus.

But besides this, there is a true external auditory meatus
which is, as usual, an outgrowth from the tympanic. The latter
bone is very large and bullate. It is grooved anteriorly by the
carotid, and the short styloid process appears between it, the
squamosal, and the ex-occipital.

The tympanic and the periotic are anchylosed together and
wedged into the space left between the ex-occipital, squamosal,
parietal, alisphenoid, and the basi-cranial axis.

The very short nasal bones (absent in the specimen figured)

264

ON THE STRUCTURE OF THE SKULL.

are adjusted by a broad posterior face to the frontals at Na*.
The large premaxillaries ascend along the sides of the anterior
nasal aperture to the nasal bones, but are almost excluded from
the palate, inferiorly, by the maxillaries ; their alveolar portion,
however, is very large and long, and this circumstance, together
with the shortness of the nasal bones, throws the anterior nares,
in the dry skull, almost to the top of the head. As the palatine
processes of the maxillaries and palatines are, at the same time,
relatively short, the posterior nares are situated but little behind
the anterior nares, and thus the axis of the nasal passage forms
a large angle with the basi-cranial axis. The lachrymal is a very
small, though distinct, bone.

In the foetal Elephant here described the space between the
two tables of the skull is moderate, and is filled with a spongy
diploe ; but, with advancing age, the interspace between the
tables in the frontal, parietal, and supra-occipital increases until
it equals or exceeds the depth of the cranial cavity, and the
diploe is replaced by vertical plates and pillars of bone, between
which air-cavities extend back from the frontal sinuses and
nasal passages. The skull of the Elephant resembles that of
the Pig in many of its most important and characteristic features,
and, through the Pig, its affinities are traceable to the other
Ungulata. Of these, the skull of the Tapir resembles it most in
some respects, such as the shortness of the nasal bones and of
the palate ; the consequent large angle wrhich the axis of the
nasal passages makes with the basi-cranial axis ; and the pro-
longation downwards and forwards of the frontal bones.

Fig. 104.

\Pa

'HO

Fig. 104. — Side view of the skull of a Calf. — prnd, the paramastoid process of the

ex-oecipital.

THE SKULLS OF MAMMALIA. *265

On the other hand, some Ruminants carry to an extreme the
development of the frontal into a great supra-orbital arch, its
extension backwards in the middle line, and the concomitant
expansion of the supra-occipital forwards ; so that the parietals
of the Ox, for example, are reduced to a comparatively narrow
band in the middle line, while they expand widely in the
temporal fossae (Fig. 104).

The crania of the purely aquatic Mammals, such as the
typical Seals, the Sirenia and the Cetacea, exhibit a certain
similarity of character in the midst of very wide and important
differences.

The basi-cranial axis is either flat or slightly curved upwards
at its anterior and posterior extremities. The olfactory and
occipital planes are vertical, or nearly so. The squama occijritis,
alone, or united with large inter-parietal elements, extends upon
the vertex of the skull between the parietals, and approaches, or
even reaches, the frontals, so that the parietals are very much
shorter antero-posteriorly than at the sides and below.

The frontals take but a small share in the formation of the
roof of the cranial cavity ; the nasals are relatively short, the
anterior nasal aperture relatively large, and the posterior often
situated far forwards. The prefontals, or lateral masses of the
ethmoid, are small or rudimentary. The tympanic and periotic
are always anchylosed together, and, whether connected or not
with the squamosal, are more easily detachable from the skull
than usual.

The Seals are extreme aquatic modifications of the carni-
vorous type of cranial structure ; the Sirenia, of the ungulate
type. The Cetacea present resemblances to both.

In the common Seal (PJwea vitulina) (Fig. 105) the cranial
cavity is exceedingly broad and spacious, and the cerebral
extends far further back over the cerebellar chamber, and is
much larger in proportion to it, than is usual in Carnivora.
There is a strong bony tentorium, and an osseous falx is more
or less developed. The basi-cranial axis, very thin and broad, is
curved, so as to be concave from before backwards. The syn-
chondrosis between the presphenoid and basi-sphenoid persists.

266

ON THE STRUCTURE OF THE SKULL.

The superior and middle turbinal bones are greatly flattened
from side to side, and unite below and internally with the lamina

Fig. 105.

Fig. 105. — Longitudinal and vertical section of the skull of a Seal (Phoca vitulina).

The premaxilla is absent.

perpendicular is, or proper ethmoid, so that all direct communi-
cation with the superior and middle meatuses of the nose is shut
off below. The inferior turbinal, on the other hand, is exceed-
ingly large and complex in its structure. The orbito-sphenoids
are large and, ascending upon the front wall of the skull, unite
anteriorly behind and below the cribriform plate, so as to hide
nearly the half of the ethmoid when the base of the skull is
regarded from above. The presphenoid is relatively small.

Less than half the length of the frontal bones enters into
the upper wall of the cranial cavity, the rest being devoted to
the roof of the nasal chambers. This part of the frontals is very
much narrower than the other, and is bent down at the sides, so
as to form two broad thin plates, which wall in the superior and
middle spongy bones, articulate below with the vomer and with
the palatine, and take the place of the os planum.

The lower edge of the parietal unites with the front part of
the alisphenoid and with the ex-occipital, leaving a great infero-
lateral space, which is filled up in front and above by the
squamosal, and behind and below by the periotic. The squa-
mosal is relatively a small bone, but the periotic and the
tympanic, which are anchylosed with it, are very large. A
swollen pars mastoidea appears on the exterior of the skull, and
is hollowed internally by a cavity which opens into the cranium,

THE SKULLS OF MAMMALIA. 267

and extends under the anterior and posterior vertical semi-
circular canals.

The tympanic forms a very thick bulla, prolonged externally
into an auditory meatus. It is firmly anchylosed with the pro-
otic regions of the periotic and with the squamosal, but for the
rest of its extent it is only applied to, and not anchylosed with,
the periotic. It is pierced by the carotid canal.

The anchylosed squamosal, periotic, and tympanic are very
easily detached from the walls of the skull, as is the premaxilla
from the upper jaw.

The skull of the Dugong (Halicore, Fig. 106) presents the
peculiarities of the cranial conformation of Mammals of the
order Sirenia in a very marked form. The basi-cranial axis is
almost flat above, but very thick. The suture between the
basi-occipital and the basi-sphenoid persists, but that between
the basi-sphenoid and the presphenoid is completely obliterated,
as is that between the presphenoid and the ethmoid, which last

Fig. 106.

Pmx

Fig. 106. — Longitudinal and vertical section of the skull of a Dugong (Halicore Indicus).

has the form of a stout bony plate, with an almost vertical
posterior edge, or crista galli. The upper median part of the
frontals is very narrow from before backwards, so that they cover
not more than the posterior half of the upper edge of the
ethmoid, and appear but very little on the roof of the cranial
cavity ; laterally and below, they are much expanded, and
produced forwards and outwards. The greater part of the roof
of the skull is furnished by the parietals, the longest antero-

268 ON THE STRUCTURE OF THE SKULL.

posterior diameter of which bones is in the middle line, as they
are not separated, posteriorly, by the supra-occipital, or, ante-
riorly, by the frontals.

The orbito-sphenoids are large, and enter into the com-
position of the front wall of the skull. The alisphenoids are
also large, and contribute to the formation of the side walls, as
well as of the base, of the skull.

The squamosal appears in the interior of the cranium between
the parietal, supra-occipital, and periotic, with which last it is
not anchylosed.

The periotic, a large and dense ossification, has a very pecu-
liar form, being divided into an inner portion, corresponding
with the pars petrosa, and an outer thick mass which answers to
the teg men tympani and pars mastoidea.

The tympanic is a mere ring of bone, open above, and having
a thicker anterior than posterior cms. It is by the former that
it is more especially attached to the periotic, though the hinder
thinner eras also becomes anchylosed with that bone.

The squamosal unites behind and below with the ex-occipital,
but leaves a space, superiorly, in which the pars mastoidea
appears on the exterior of the skull. The malar process of the
squamosal is exceedingly thick, and extends far forwards as well
as transversely outwards. The frontals send very large processes
downwards and forwards, as in the Elephant and Tapir, which
are not only met by the maxilla?, as in the latter animal, but
also meet, and indeed are covered by, the nasal processes of the
preruaxilla?. The lachrymals are large, but imperforate. The
jugals, thick and curved, are connected with them.

The very small nasal bones are fixed by the greater part of
their under surfaces to the anterior half of the ethmoid, beyond
which they project but little, so that almost the whole of the
vast anterior nasal aperture is, in the skeleton, uncovered.
The premaxillse are enormous, and constitute a large propor-
tion of the lateral margins of the upper jaw as well as the
whole of its anterior region. Their ascending, or nasal, processes
are produced forwards instead of downwards, so that the point
which corresponds with the spina nasalis anterior in Man is
nearly on a level with the top of the head. The alveolar

THE SKULLS OF MAMMALIA. 269

process is even more largely developed, to contain the incisor

tusks of the animal.

The maxillae, also large and prolonged forwards, have very
thick and long palatine processes, separated by a wide incisive
foramen from the premaxilla. The palatine process of the
palatine is also very thick, but it is shorter than deep, so that
the posterior nares, which open behind it, are placed vertically
under the hinder part of the anterior nares, in the dry skull.
The vomer, thick and stout behind, thin and ridge-like in
front and above, embraces the lower edge of the ethmoid, and
is suturally united to both the palatines and the maxillaries.

The skulls of the Sirenia have resemblances on the one side
with those of the ungulate Mammals and Proboscidea ; on the
other, with those of the Cetaeea, but yet differ in many and
most important respects from all.

. The skulls of the Cetaeea present more singular modifications
than those of any other Mammalia. In all these animals, the basi-
cranial axis is concave superiorly, and the primitive separation
between the basi-sphenoid and presphenoid persists for a long time.

The vomer is very long, and extends backwards on the base
of the skull at least as far as the basi-sphenoid, and sometimes
covers the whole length of that bone.

The ethmoid has its posterior edge perpendicular, or nearly
so, to the basi -cranial axis, and the foramina for the exit of the
olfactory nerve are small or obliterated.

The frontals enter but very little into the roof of the skull,
largely into its anterior and lateral walls. They are prolonged
outwards and forwards into the long and broad supra-orbital
processes, which are concave interiorly, where they form the
roof of the orbital cavity.

The parietals hardly appear at all, externally, upon the top
of the skull, their median parts being obscured or interrupted
by the inter-parietal and supra-occipital. They occupy a large
space, however, in the temporal fossae.

The ex-occipitals and supra-occipitals are enormous. The
latter, usually increased by coalescence with the large inter-
parietal, extend up to, or beyond, the vertex to meet the

270

ON THE STEUCTUEE OF THE SKULL.

frontals. The orbi to-sphenoid and alisphenoid vary in size.
The squamosal is large, and is firmly fixed to the side of the
skull, forming part of the wall of the cranial cavity. The
periotic, usually anchylosed into one bone with the bullate
tympanic, sometimes enters largely into the wall of the cranium,
sometimes is almost altogether excluded therefrom by the
parietal, alisphenoid, and other adjacent bones, which send
prolongations over it.

The maxillary apparatus is greatly elongated, so as to form

"mx-

U7V-

rtc

Z*r

Fig. 107.— Upper, under, and side views of the skull of a foetal Whalebone Whale {Ealceni
australis). The jugal bones are absent. In the under view the palatine bono is acci-
dentally marked Pt instead of PI.

THE SKULLS OF MAMMALIA. 271

a kind of beak. The premaxillae enter into the upper and
inner part of the whole length of this maxillary beak, bnt
contribute little or nothing to its palatine surface and lateral
boundaries, which are formed mainly by the maxillae. The
latter bones are always prolonged over, or in front of, the
supra-orbital processes of the frontals.

The imperforate lachrymal is small, and sometimes coalesces
with the jugal.

The nasal bones are always short, sometimes rudimentary ;
and the palatine bones are so disposed that the posterior nares
are situated almost vertically under the anterior nares.

The squamosal bones are produced outwards, and the pro-
cesses thus formed approach, or come into contact with, the
posterior part of the supra-orbital processes of the frontals,
which they separate from the jugal. Inferiorly, these processes
support the glenoidal facets for the condyle of the lower jaw.

The sides of the broad basi-occipital are always prolonged
downwards into free plates, which are concave outwards. These
plates join the pterygoids in front, and the ex-occipitals behind,
and so constitute the inner and posterior walls of an auditory
chamber, the anterior and outer boundaries of which are fur-
nished by the alisphenoid and the squamosal. In this chamber
the tympano-periotic is lodged, sometimes quite loosely, at
others fixed firmly in by interlocking sutures.

In the Balsenoidea, or " Whalebone Whales," the symmetry of
the skull is undisturbed, though there may be a slight inequality
of the maxillae. The skull of the foetal Balmna atistralis, repre-
sented in Fig. 107, is perfectly symmetrical. Each lateral edge
of the broad and flat basi-occipital is prolonged downwards and
outwards into a broad process, concave outwards and convex
inwards, the inferior edge of which is free, while the hinder edge
unites with the ex-occipital, and the front edge with the ptery-
goid, to form the inner wall of the funnel-shaped chamber which
lodges the tympano-periotic bone.

In front, this chamber is bounded by the pterygoid and the
squamosal, and between and above them, for a small space, by
the alisphenoid ; behind, it is constituted almost entirely by the
ex-occipital, while, externally and above, it is bounded and roofed

272 on the structure of the skull.

in by the squamosal. Between these bones there is left, at the
apex of the chamber, a considerable irregular aperture, which
communicates with the cranial cavity.

The anterior and outer part of the under-surface of the
squamosal is produced downwards into a great trihedral pillar,
the obliquely truncated inferior face of which bears the articular
surface for the mandible (GI, Fig. 107). Behind this the squa-
mosal presents a comparatively low wedge-shaped ridge (a,
Fig. 108), between which and the "trihedral pillar" is a groove;
Fio. 108 while behind it, or between it and

the ex-occipital, there is a deeper
and wider transverse channel.

The periotic bone is irregu-
larly triangular ; the apex of the
triangle, turned inwards and for-
wards, is thick and rounded, the
anterior, posterior, and outer edges
being thinner and more or less
O/iO irregular. The upper smooth and
^\ concavo-convex surface of the

Fig. 108.— Enlarged view of the chamber periotic adjusts itself to the Under-

which lodges the left tympano-periotic c r> ,i -i -i .,

bone of the fetal Baimia i austral, is.- surface ol the squamosal, where it
a, the « wedge-shaped ridge " of the forms the roof of the funnel-shaped

squamosal ; C, the aperture which . rT11 _ 1 . .

leads into the interior of the skull; Cavity. Hie apex Ol the penotlC,

f.o, fenestra rotunda. however, projects beyond this, and

incompletely divides the irregular aperture above mentioned
(b, Fig. 108) into an anterior division, which corresponds with
the foramen ovale and foramen lacerum medium, and a poste-
rior which answers to a foramen lacerum poster 'his.

The under-surface of the periotic, much more irregular, is
divisible into three regions : an outer anterior ; an outer pos-
terior ; an internal. The first and second are separated by a
deep triangular notch in the outer margin of the bone, into
which the inner end of the wedge-shaped ridge of the squamosal
is received. The first, broad and short (PrO), presents a rough
surface in front, with which the tympanic articulates, and
eventually anchyloses; and behind, a concave surface, which,
entering into the roof of the tympanic cavity, answers to the

TIIK SKULLS OF MAMMALIA.

273

tegmen tijmpani. The second, narrower, elongated, and prismatic,
fits into the transverse channel behind the wedge-shaped pro-
cess {a). It corresponds with the pars mastoidea, and its rough
outer extremity appears on the exterior of the skull, between
the squamosal and ex-occipital.

The internal division, convex and rounded below, is formed
by the pro-otic and opisthotic, and presents a large promontory
with the fenestra rotunda (f.o) on its posterior surface, while the
fenestra ovalis and Fallopian canal are visible upon its exterior.
The tympanic bone (Fig. 109, Ty) is large, and scroll-like in
form, very thick internally and below, and thin above and
externally, where it presents the aperture of the external audi-
tory meatus. It is by this thin upper and outer edge only, that
it eventually anchyloses with the tegmen tympani and pars
mastoidea, and hence, as its substance is very dense and brittle,
readily breaks off.

In the adult Whale the tegmen tympani and pars mastoidea

Fig. 109.

V/7

Fig. 109. — "Ear bones" of the adult Bakena anstralis. The upper figure gives the view

from within ; the lower, from without.

274 OX THE STRUCTURE OF THE SKULL.

become greatly elongated and very rugged, the tympanic also
acquiring a very large size (Fig. 109).

The vomer is a very long and large bone, deeply grooved
above for the ethmoidal cartilage, which extends downwards
and forwards between the premaxillae and the maxillae to near
the anterior end of the snout. Its expanded upper and pos-
terior end unites with the basi-sphenoid in the middle line, and
with the pterygoid laterally. In front of the basi-sphenoid it
embraces, not a distinct presphenoid (as in Pterobalsena, accord-
ing to Eschricht), but the inferior surfaces of the orbito-
sphenoids, which are very thick ; and, being applied together
by their flat median faces, apparently replace the proper pre-
sphenoid.

Both these bones and the alisphenoids are small, and almost
confined to the base of the skull.

The supra-occipital and inter-parietal are united together,
and completely overlap and hide the parietals in the roof of the
skull. The separate frontals only enter into the anterior wall
of the skull, and between them and the orbito-sphenoids an oval
aperture is left, doubtless diminished in the recent state by the
ethmoidal cartilage. Laterally, the frontals are prolonged
outwards and backwards into two great supra-orbital processes,
which nearly meet the zygomatic processes of the squamosal.
The short jugal bones, absent in the specimen figured, extend in
the Balsenoidea from the zygomatic process to the anterior and
external angles of the supra-orbital prolongations, and are dis-
tinct from the lachrvmals.

The pterygoids are completely separated by the palatines
(Fig. 107). In front of the latter the maxillae almost wholly ex-
clude the premaxillaries from the palate, while they send great
processes obliquely outwards and backwards, in front of the
supra-orbital prolongations of the frontal. The long premaxillae,
on the other hand, pass upwards and backwards on each side of
the elongated and symmetrical nasals to meet the frontals, and
exclude the maxillae altogether from the anterior nares.

The rami of the lowrer jaw are very narrow, and so much
arched outwards as to be able to enclose the baleen plates
attached to the upper jaw when the mouth is shut.

THE SKULLS OF MAMMALIA.

275

Eschricht has described, with much care, the changes which
the skulls of the BalsenMea undergo in passing from the foetal
to the adult condition, justly remarking that the skull of even
a large fcetus is more different from that of the adult, than the
skulls of distinct species of the same genus of Whales are from
one another.

The growth of the walls of the cranial cavity relatively to
that of the external prolongations of the cranial bones and to

Fig. 110.

Pin x

Fjg# no. — A, upper, B, under, and C, side views of the skull ct* a foetal Cachalot in the
Museum of the Royal College of Surgeons. The nasal bones are not represented in
Fig. a. — N' left, N" right, nostril. The hinder extremity of the jugal, Ju, has fallen
down from its natural connection with the zygomatic process of the squamosal.

T 2

276 ON THE STRUCTURE OF THE SKULL.

that of tlie jaws, is soon arrested, and in addition the position
and relations of some of the cranial bones become altered. In
the smallest foetuses of the lesser Fin-back (Pterobalmna minor),
for example, the parietal region is occupied by the inter-
parietal bone and the great fontanelle which lies in front of it.
In larger foetuses the fontanelle becomes closed by the progressive
backward growth of the frontals, but the extension of the bones
does not cease with their contact. The parietals grow over the
inter-parietal and spread over it until they meet in the middle
line. Hence the interparietal is eventually visible only in the
interior of the skull. Anteriorly, the parietals grow over the
frontals almost to the same level as the nasals, and thus conceal
the share which the frontals take in the formation of the roof of
the skull. But, at the same time, the supra-occipital extends
from behind over the parietals ; so £hat, at length, in that region
which, in the youngest foetus, was covered only by the inter-
parietal, three bones — the inter-parietal, parietal, and supra-
occipital — are superimposed.

The skulls of the other great division of the Cetacea, the
Delphinoidea — or Dolphins, Porpoises, and Cachalots — are almost
all distinguished by their very marked asymmetry.

In the Cachalot, or spermaceti Whale (Physeter), for example,
the right premaxilla is much longer than the left, extending far
back upon the right frontal, while the left does not reach the
left frontal ; the left nostril, on the other hand, is much more
spacious than the right (Fig. 110, A). On the base of the skull
(Fig. 110, B) the pterygoid bones unite in the middle line and
prolong the palate, as in Myrmecopliaga and Omitliorliynchus.
AVhen they and the palatine bones are removed, the axis of the
lower part of the ethmoid is seen to continue that of the basi-
cranial bones, which are, as usual, quite symmetrical. Supe-
riorly, however, the ethmoidal plate is twisted over to the left
side, and deeply grooved on the right side to form the inner
wall of the small right nostril.

The vomer, which embraces the ethmoid and the presphenoid
below, is also asymmetrical posteriorly, presenting a long and
shallow lateral excavation, on the left side, and a short and deep
one on the right. The maxilla? are correspondingly unsym-

THE SKULLS OF MAMMALIA.

211

metrical in the region of the nasal aperture, but elsewhere they
are pretty nearly symmetrical. But it is the nasal bones which
exhibit the greatest distortion, the left and right being very
unequal in size and dissimilar in form.

The jugal and the lachrymal commonly become anchylosed.

The basi-occipital, as in the Balwnoidea, gives off a lateral
downward process, which unites, behind (Fig. Ill), with an out-

Pmi

Fig. 111. — Longitudinal and vertical section of the skull of the White Whale or Beluga
(Delphi napter us). * marks a distinct bony element interposed between the alisphenoid,
the parietal, and basi-cranial axis ; a, the ossified falx.

ward prolongation of the ex-occipital, and, in front, with the
pterygoid, to constitute the inner wall of a deep chamber for the
tympano-periotic. But the roof of this chamber is chiefly
formed by the very large alisphenoid, which extends outwards
to unite with the frontals, parietals, and squamosals, and back-
wards to the ex-occipitals. The aperture which is left between
the hinder edge of the alisphenoid, the ex-occipital, basi-occipital,
and basi-sphenoid, is exceedingly small, so that the tympano-
periotic is still more shut out from the cranial cavity than in
Balsena. In Hijperoodon and Orca the aperture is still further
reduced ; but this peculiarity cannot be said to be a distinctive
character of the Delphinoid skull, as in Platanista the aperture
is large, and the periotic appears in the interior of the cranial
cavity in the ordinary way.

278

LECTURE XIV.

ON THE STRUCTURE OF THE VERTEBRATE SKULL.

THE THEORY OF THE VERTEBRATE SKULL.

In the preceding Lectures I have, as far as possible, confined
myself to a statement of matters of fact, and to the conclusions
which immediately flow from the application of a very simple
method of interpretation to the facts. That method of interpre-
tation is based upon the principle that, in any two skulls, those
parts which are identical in their principal relations in the adult
state, and in the mode in which they reach this state (or in
their development) are corresponding, or homologous, parts, and
need to be denominated by the same terms.

By the application of this method it has been possible to
demonstrate the existence of a fundamental unity of organiza-
tion in all vertebrate skulls ; and, furthermore, to prove that
all bony skulls, however much they may differ in appearance,
are organized upon a common plan, no important bone exist-
ing in the highest vertebrate skull which is not recognisable in
the lowest completely ossified cranium.

The enunciation of these results alone is a " Theory of the
Skull," but it is by no means what is commonly understood as
the theory of the skull.

For it will be observed that the statement just put forth
confines itself to a simple generalization of the observed facts
of cranial structure, and would be perfectly complete were the
skull a self-subsi stent structure, devoid of any connection with a
trunk. On the other hand, that doctrine to which the title of
" The Theory of the Skull " is ordinarily applied, embraces not

THE THEORY OF THE VERTEBRATE SKULL. 279

only such a generalized statement of the facts of cranial structure
as this, but adds a hypothesis respecting the relations of the
skull to the spinal column. It assumes that the bony cranium (the
cartilaginous and membranous states of the cranium it usually,
ignores) is composed of elements homologous with those which
enter into the structure of the spiual column ; that, in fact, it
consists of modified vertebrae. And it is commonly conceived
that it is the doctrine of the unity of structure of the skull and
of the vertebral column, rather than the demonstration of the
unity of organization of skulls, which is one of the chief glories
of morphology.

The assumption that every skull repeats the organization of
the trunk and consists of a certain number of modified vertebrae,
evidently implies a belief in the unity of organization of skulls ;
but it is to be carefully noted that the converse proposition does
not hold good ; for it is quite possible to hold that all skulls are
modifications of one fundamental plan, while wholly disbelieving
that plan to be similar to the plan of a vertebral column.

Looking broadly at the history of the theory of the skull
(using the phrase in its widest sense), I note three great lines
of inquiry which have brought that theory into its present
condition,— -the first originated by Oken and Goethe ; the
second, not originated, perhaps, but chiefly fostered and deve-
loped by Geoffroy St. Hilaire and Cuvier ; the third, originated,
and almost exclusively worked out, by Keichert, Rathke, and
their followers among the embryologists of Germany and
England.

I. I have united the names of Goethe and of Oken as the
originators of the hypothesis of the vertebral structure of the
skull, as a matter of equity, and to aid in redeeming a great
name from undeserved obloquy ; though, in strict technical
justice, the claim of the one to priority lapsed through lack of
publication.

Goethe combined with a fervid creative genius, which has
placed him on a level with the greatest poets of all ages, so
much of observational acuteness and of intellectual precision
as might have sufficed for the equipment of a well-reputed

280 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

man of science. From his youth up, passionately devoted to
the natural sciences, more especially to botany and to osteology ;
and induced by the habit of his mind to search for the general
truths which give life to the dry bones of detail, Goethe had
been led to drink deeply of the spirit of morphology, during
his study of the metamorphosis of plants and his successful
search after the premaxillary bones of man, imagined, before
his time, to be wanting. With a mind thus prepared, it was
no wonder that, as Goethe writes, the notion of the vertebral
composition of the skull had early dawned upon him : —

" The three hindermost parts I knew before, but it was only
in 1791, on picking up an old and broken sheep's skull amidst
the sandy dunes of the Jewish cemetery in Venice, that I per-
ceived the facial bones also to be made up of great vertebras ;
and observing, as I clearly did, the gradual passage from the
first pterygoid bone to the ethmoid bone and to the spongy
bones, the whole became plain."

Not improbably deterred, however, by the many difficulties
which must have presented themselves to him, in attempting to
carry out these views with due scientific sobriety, Goethe kept
them to himself, or shared them only with his immediate
friends, for thirty years ; the passage cited, in which they are
first mentioned, bearing the date of 1820.

But, in 1807, Lorenz Oken independently originated and,
what is more to the point, published, those views of the verte-
bral composition of the skull which have since attained such
world-wide celebrity ; so that the great poet's silent partner-
ship in the affair would be hardly worth mentioning were it not
that his reticence has been made the ground of severe attacks
upon his honour and veracity. It has been suggested that
Goethe, iull of years and of honours, thought it worth while to
attempt to steal from the young Professor of Jena the fame that
had accrued to him. And upon the infamy of such petty
larceny the poet's latest accuser has heaped the insinuation
that the author of "Faust ' and of "Meister" was so stupid a
plagiarist, as to copy, not only Oken's views, but his account of
the manner in which he came by them.

" Vaguely and strangely, however, as Oken had blended the

THE THEORY OF THE VERTEBRATE SKULL. 281

idea with his a priori conception of the nature of the head, the
chance of appropriating it seems to have overcome the moral
sense — the least developed element in the spiritual nature — of
Goethe, unless the poet deceived himself." *

" The circumstances under which the poet, in 1820, narrates
having become inspired with the original idea are suspiciously
analogous to those described by Oken in 1807, as producing
the same effect on his mind."t

It would be difficult to couch an offensive accusation in
stronger phraseology than this ; but, by a singular chance, the
scientific morality of its object has recently been fully vin-
dicated. Goethe, when in Italy, kept up a correspondence
with the family of his friend Herder. His letters have been
published, and in one addressed to Madame Herder, and dated
May 4, 1790, this passage occurs : —

" By the oddest, happy chance, my servant picked up a bit
of an animal's skull in the Jews' Cemetery at Venice, and, by
way of a joke, held it out to me as if he were offering me a
Jew's skull. I have made a great step in the explanation of
the formation of animals."

Can it be doubted that this " great step " is exactly that
vertebral theory of which Goethe says, writing in 1820, he had
as clear a view " thirty years ago ?" It is to be hoped that
this evidence, which Professor Virchow has so strikingly put
forward, will henceforward silence even the most virulent of
Goethe's detractors, although a careful perusal of the arguments
used by Mr. Lewes, in his " Life of Goethe," might have already
sufficed those who were open to conviction.

The idea, which dropped still-born from Goethe's mind,
was, as I have said, conceived afresh by Oken, and came
vigorously into the wTorld in that remarkable discourse (occupy-
ing in print about fourteen quarto pages) with which he
inaugurated his professorial labours at Jena.

It is hard to form a just judgment of this singular man ; and,
I must confess, I never read his works without thinking of the

* •• Encyclopaedia Britaimica," eighth edition, vol. xvi., p. 501 ; article,
"Oken." t Ibid., p. 501.

282 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

epithet of " inspired idiot," applied to our own Goldsmith : so
strange is the mixture of insight and knowledge with what, to
my apprehension, is mere " sound and fury, signifying nothing."
But the " Programm ' contains far more of the former and
less of the latter ingredient than is usually noticeable in Oken's
lucubrations, and it appears to me to be, at the present moment,
by far the best specimen extant of the style of speculation about
the skull, characteristic of the school which Oken originated.
Indeed, if for the term " cranial vertebra?," " cranial segments ':
be substituted, I do not know that the plan of composition of the
osseous brain-case can be better described than in the language
which I shall now quote.*

The " Programm " opens thus : —

" A vesicle ossifies, and it is a vertebra. A vesicle elongates
into a tube, becomes jointed, ossifies, and it is a vertebral
column. The tube gives off (according to laws) blind lateral
canals ; they ossify, and it is a trunk skeleton. This skeleton
repeats itself at the two poles, each pole repeats itself in the
other, and they are head and pelvis. The skeleton is only a
developed (aufgetvachsenes), ramified, repeated, vertebra ; and a
vertebra is the preformed germ of the skeleton. The entire
man is only a vertebra.

"i.

"Take a lamb's skull, separate from it those bones which
are considered to be facial, and those bones of the cerebral cap-
sule which take no share in the base, such as the frontal bones,
the parietal bones, the ethmoid and the temporal bone, and
there remains a bony column, which every anatomist will at
once recognise to be three bodies of some sort of vertebra?, with
their lateral processes and foramina. Replace the bones of the
cerebral capsule, with the exception of the temporal bones
(for the cavity is closed without these), and you have a vertebral
column, which is distinguished from the true one only by its
expanded spinal canal. As the brain is the spinal marrow
more voluminously developed [in relation] to more powerful
organs, so the brain-case is a more voluminous spinal column.

* " Ueber die Bedentnng der Schadelknochen. Ein Programm beim Antritt
der Professur an der Gesammt-Universitat bei Jena.'' Von Dr. Oken. Jena. 1807.

THE THEORY OF THE VERTEBRATE SKULL. 283

" If there are three vertebral bodies in the brain-ease, there
must be as many vertebral arches. These are to be sought
out and demonstrated.

" You see the sphenoid separated into two vertebrae : through
the first one pass the optic nerves, through the hinder the
nerves of the jaws (par trigeminum). I term the former the
Eye vertebra, the latter the Jaiv vertebra. Against this last
abuts the basilar process of the occipital bone with the rpetrous
bone. The two form one whole. As the optic nerve traverses
the Eye vertebra, and the jaw nerve the Jaw vertebra, so the
hindermost vertebra is related to the auditory nerve. I there-
fore term it the Ear vertebra. Again, this is the first cephalic ver-
tebra ; the precedent, the second ; and the eye vertebra, the third.

" It has given me unspeakable trouble to make out whether
the petrous bone belongs to the first or to the second cephalic
vertebra. Before I had taken into account the relations of
the nerves, vessels, and muscles, my decision was based only
upon the structure of the skulls of Birds, Lizards, and Chelonia ;
but now I have fortified it by a multitude of concurrent argu-
ments, of which I will state only a few in this place.

" You will have observed, in fact, that each of the two anterior
vertebrae has appropriated a sense. (As the jaws end in the
lips, I reckon them also among the [organs of] sense, and I
shall demonstrate that they are so, and how they are so.) Now,
if the petrous bones belonged to the jaw vertebra, one vertebra
would give off nerves to the sensory organs, while the first
vertebra would be sent empty away. True, it transmits nerves
to the tongue, but these are variable ; and it will be shown in
the sequel that neither tongue nor nose have, or can have, a
proper vertebra. Lastly, in Lizards, the auditory apparatus lies
distinctly in the occipital bone.

" The cephalic vertebras are, therefore, sensory vertebras, and
only exist in correspondence with the [cephalic] senses. (The
tongue and the nose are trunk senses, of which presently.)
Vertebral divisions and cephalic sensory nerves go parallel
with one another. Bones are the earthly, hardened nervous
system ; nerves are the spiritual, soft, osseous system — con-
tinens and contentum.

284 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

"Between the sphenoid and occipital bones, between the
sphenoid and petrous bones, between the parietals (the tem-
poral bones are away) and the occipital bone, draw a line, and
you have marked off the first vertebra. Draw another line
between the two sphenoids, or, in Man, in front of the ptery-
goid processes ; laterally, through the fissura orbitalis in front
of the aide magnse ; lastly, between the frontals and the parietals,
and you have the second vertebra separated from the last.

" 1. Now, take the ear vertebra of a foetus of any Mammal
or of a Man ; place beside it an incompletely-developed dorsal
vertebra, or the third cervical vertebra of a Crocodile, and
compare the parts of which the two are composed — their forms,
their contents, and the exits of the nerves.

" According to Albinus and all anatomists, each vertebra
of a fcetus consists of three separate pieces — the body and the
two arches, which together form the spinous, transverse, and
oblique processes. You have the same in the occipital bone,
only more distinct and separate. The pars basilaris is a
corpus vertebrae still more separated from the condyloid parts,
which form the lateral regions ; these are again separated
from the pars occipitalis, which forms the spinous process. In
fact, this part itself is often split again, like the spinous pro-
cesses in spina bifida. The occipital bone, therefore, is decom-
posable, according to the mode of its origin, into five pieces,
since the lateral, or articular, and the spinous parts appear as
independent developments ; as is found also in actual vertebra?,
which consist of five pieces, and in the third cervical vertebra of
the Crocodile. Finally, I need take no further pains to prove
that the occipital foramen is the lower aperture of a vertebral
canal ; that the foramen lacerum is an inter-vertebral foramen,
and the occipital protuberance is a spinous process ; that, there-
fore, the occipital bone, in respect of form, as of function (since
it encloses the cerebellum, as a continuation of the spinal
marrow), is in every sense a true vertebra, since the mere
naming of these parts is enough to cause their recognition as
such.

" You will think T have forgotten the petrous bone. No !
It seems not to belong to the vertebra? as such, but to be the

THE THEORY OF THE VERTEBRATE SKULL. 285

sensory organ in which the vertebral — the auditory — nerve
loses itself, and, therefore, to be an organ as completely sepa-
rated from vertebral production as any other viscus, or as the
ball of the eye ; the deception lies only in this, that it is the
essence of this organ to be ossified, as it is that of the eye to be
crystalline.

" The mastoid process is, in animals, and also in the human
foetus, a proper bone, in which the styloid process lies. It is
plainly inserted into the first vertebra, but it receives its signi-
fication from the tongue.

" 2. Having entered so fully into the discussion of the first
cephalic vertebra, I might, except for clearness' sake, spare you
any delay over the second. But I will also demonstrate in this
how completely the brain-case is formed according to the idea of
a vertebra, and has eyen been partially produced as such.

" In every skull of a foetus you may find the alee orbitcdes
of the sphenoid separate from it. They belong to the third
vertebra. But, in the half-developed foetus, the great wings
and the pterygoid processes are also separate from the body of
the sphenoid. The last-mentioned processes are foreign to the
sphenoid, and only coalesced with it ; they belong to quite
another formation, and very probably have the same significa-
tion as the os omoicleum of the Bird's head, as Cuvier has already
indicated. I shall return to them.

" There remain, therefore, for the posterior sphenoid, or the
jaw vertebra, three portions of bone — the body and the great
wings, or the lateral and oblique processes of the vertebra.
The spinous processes are formed by the two parietals, which,
in many animals, coalesce so as to leave no suture, but are yet
originally two. It is to be remarked that, in the Sheep, this
vertebra is closed by the bones in question, without the inter-
mediation of the temporal, which also does not belong to the
vertebral group. The same occurs in the Chelonian, the
Crocodile, &c.

" 3. Whoso has recognised the second vertebra, as such, need
only look at the third, especially in Kuminants, to discover
quite the same structure. The anterior sphenoid with the
aide orbitcdes represents the body, together with the lateral pro-

286 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

cesses ; the two frontals form the spinous process, together with
its lateral parts.

" The sphenoid is separated into two vertebrae, not merely in
the human foetus and in Ruminants, but also in the Apes, in
Bradyjms tridadylus, Dasypus novemcinctus, Dog, Wolf, Bear,
Otter, Rodents, and probably in all Mammals, if examined in a
sufficiently young state. The law is therefore universal.

" The inter- vertebral foramina are very well marked between
these vertebrae. A deviation seems to exist, on account of the
foramina which lie in front of the first cephalic vertebra, namely,
the foramen caroticum and lacerum, concerning which I must
leave it undecided, whether they are originally two, or only one
which has become separated. On this point, evidence enough is
to be found among animals. The organ of hearing has here
interposed itself. On the other hand, it is characteristic of the
cephalic vertebrae that their sides are perforated by nerves, —
by the optic nerve, the jaw nerves, and the hypoglossus, if we
reckon the auditory and facial nerves as inter-vertebral nerves :
a circumstance which demands further inquiry.

" So much of the cephalic vertebral column. I might have
been able to treat more fully and thoroughly of it, and to have
indicated the nerves, veins, and muscles, which in the head cor-
respond to those of the trunk, and the like for the bones ; but
in a programme one must be content with merely putting forth

one's view of a question.

"II.

" If the cerebral capsule is the repetition of the spinal
column, only more expanded and organized (I speak as an
anatomist), the head must repeat the outgrowths of the spinal
column, the thorax, the pelvis, and the limbs ; and, indeed,
thereby must it attain completeness.

" By this union of the representatives of all the bones of the
trunk arises the wonderful, but yet analvsable, mixture and
intercurrence of formations which appear as the facial bones.
The spinal column becomes the brain-case ; the walls of the
trunk, with the extremities, become the face."

In developing this idea, Oken arrives at the conclusion that
the nasal cavity is the thorax of the head, and the oral cavity

THE THEORY OF THE VERTEBRATE SKULL. 287

the abdomen of the head. The squamosal is the conjoined
scapula and ilium of the head ; the pterygoid, the clavicle ; the
hyoidean apparatus, the other pelvic bones. The jugal arch
represents the humerus, radius, and ulna ; the maxilla, the
hand; the premaxilla, the thumb; the teeth, the fingers. The
lower jaw represents the legs of the head ; the teeth the toes ;
and, of all imaginable hypotheses, the styloid processes are the
sacrum of the head !

Reasons, worthy of the name, for these identifications are
not to be found in the " Programm." Oken, having assumed
once for all, that, as the brain-case repeats the spinal canal, the
facial bones must repeat the other appendages of a vertebral
column and the limbs, seems to have troubled himself no fur-
ther about demonstration. What a bone should be, in order to
fit plausibly into his scheme, that it was at once settled to be —
an appeal to the " idea " dispersing all doubts.

A few years later Oken modified his original conception so
far as to regard the nasal apparatus as a fourth vertebra.

Whatever may be thought about the more speculative pas-
sages of the extract above cited from Oken's work, and of his
a priori conception of what a skull must be, it contains ample
evidence that he did, a posteriori and inductively, demonstrate
the segmented character of the bony brain-case; and had
nothing more ever been written on the subject, this great truth
would have remained as a splendid contribution to morphology.
But Oken greatly amplified the observational basis of his own
doctrine ; Spix took it up, in a modified form, and worked it out,
in his own way, through the series of the Vertebrata in his great
illustrated " Cephalogenesis," published in 1815 ; Bojanus did
the like in the pages of the " Isis," and in the "Parergon ' of
his splendid monograph, the "Anatome Testudinis ;" and,
finally, C. G. Carus developed the doctrine, as far as it conld
well go, both a priori and a posteriori, in his " Urtheilen des
Knochen unci Schalen-Gerustes," published in 1828 ; in which,
under the names of " Grund-form ' and " Schema," we have,
among other things, " archetypal ,: diagrams of the Vertebrata
generally, and of each vertebrate class.

Under these circumstances, the following passage, extracted

288 ON THE STRUCTURE OF THE VERTEBKATE SKULL.

from the article in the "Encyclopaedia Britannica" already
cited (supra, p. 281), may not improbably excite in other minds
as much astonishment as it has in mine : —

" As to the question of the superiority of the deductive over
the inductive method of philosophy, as illustrated by the writings
of Oken, his bold axiom that heat is but a mode of motion of
light, and the idea broached in his essay on * Generation '
(1805), viz., that ' all the parts of higher animals are made up
of an aggregate of Infusoria, or aggregated globular Monads/
are both of the same order as his proposition of the head being
a repetition of the trunk, with its vertebra? and limbs. Science
would have profited no more from the one idea without the
subsequent experimental discoveries of Oersted and Faraday, or
by the other, without the microscopical observations of Brown,
Schleiden, and Schwann, than from the third notion, without
the inductive demonstration of the segmental constitution of the
skull by Owen. It is questionable, indeed, whether in either
case the discoverers of the true theories were excited to their
labours, or in any way influenced, by the a priori guesses of
Oken ; more probable is it that the requisite researches and
genuine deductions therefrom were the results of the correlated
fitness of the stage of the science, and the gifts of its true culti-
vators at such particular stage." — P. 502.

Thus does the moralist upon Goethe's supposed delin-
quencies think it just to depreciate the merits of Oken, and
exalt his own, in the year 1858. But if he himself had not
been " in any way influenced " by Oken, and if the " Pro-
gramm " is a mere mass of " a priori guesses," how comes it
that only three years before Mr. Owen could write thus?*

" Oken, ce genie profond et penetrant, fut le premier qui
entrevit la verite, guide par l'heureuse idee de l'arrangement
des os craniens en segments, comme ceux du rachis, appeles
vertebres.'

And, after sundry extracts frorn Oken's " Programm," could
continue : —

"Ceci servira pour exemple d'un examen scrupuleux des

* H

Principes d'Osteologie comparee, ou Recherches sur l'Archetype et lcs
Homologies du Squelette vertebre." — P. 155. 1855.

THE THEORY OF THE VERTEBRATE SKULL. 289

faits, (Tune appreciation philosophique de leurs relations et
analogies, en un mot de l'esprit dans lequel Oken determine les
relations vertebrales des os du crane." — P. 158.

And again : —

" Quand on couimenca a apprecier la verite de la generali-
sation d'Oken, on se rappela, comnie c'est l'liabitude, que

qnelqu'nn avait en tin id6e a pen pres semblable

Mais toutes ces anticipations ne sauraient enlever a Oken le
merite de la premiere proposition definie d'nne theorie." —
P. 161.

The space at present occupied by the proclamation of the
weakness of the " moral sense " of Goethe may not unfitly be
taken up, in the next edition of the " Encyclopaedia Britannica,"
by the extrication of the author of the article " Oken," from the
singular dilemma in which these citations place him.

The fact is, that, so far from not having been " in any way
influenced" by Oken, Professor Owen's own contributions to
this question are the merest Okenism, remanie. In the work
I have cited, not a single fact, nor a single argument, can be
found by which the doctrine of the segmentation of the skull
is placed on a firmer foundation than that built by Oken.
Two novel speculations are indeed brought forward, the one
of which confuses the petrosal (in the Cuvierian sense) of the
lower Vertehrata with the homologue of the alisphenoid of Man,
and, consequently, would, if adopted, throw the whole subject
into hopeless chaos ; while the other — the supposition that the
fore limb is an appendage of the head — can only be explained
by that entire want of any acquaintance with, or appreciation of
the value of, embryology which all the writings of the same
author display.

II. The great works of Spix and Bojanus contain, apart
from the theory which they attempted to establish, abundant
evidence of the unity of composition of the bony skull, but it
was Geoffroy St. Hilaire and, more especially, Cuvier, who
demonstrated that unity of organization, apart from all hypo-
theses, most thoroughly and completely. The fresher one's
study of the writings of the wilder Okenians — the more one has

290 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

become weary of wading through empty speculations upon
" connation ' and "coalescence," "irrelative repetition' and
"transposition," the Dei ex machind who are called in to solve
every difficulty — the more heartily does one sympathise with
the sarcastic vigour with which Cuvier annihilates the products
of their exuberant fancy in the notes to the " Ossemens Fos-
siles," and the "Histoire Naturelle cles Poissons." Nor is it
possible to peruse without admiration the sagacious reason-
ings by which he was led to determinations which, in the
majority of cases, have been accepted by those who have fol-
lowed him.

Meckel, Kostlin, in his elaborate and valuable special work
on the Vertebrate Skull, and Hallmann, in his excellent essay
on the Temporal Bone, have built on Cuvier's foundations,
applying further and, in some cases, bettering, his determinations
of the homologues of particular bones. No one can study these
works carefully and retain a doubt that osseous skulls are con-
structed upon a uniform plan, though he may, with Cuvier, give
but a hesitating and grudging assent to the notion that it is, in
some sense, a modified vertebral column.

III. That criterion of the truth or falsehood of the vertebral
theory of the skull, for which the Okenians do not think it
necessary to look, and which Cuvier seems to have sought in
vain, has been furnished by the investigations of the embry-
ologists from the year 1837 to the present time.

The first step was the discovery of the visceral arches by
Beichert ; the second, the demonstration of the mode of develop-
ment of the skull, in all classes of the Vertebrata, by the remark-
able researches of Batlike, contained in the " Yierter Bericht
uber das Naturwissenschaftliche Seminar bei der Universitat zu
Konigsberg," which was published in 1839. I will quote Bathke's
statement of his conclusions at length, so that we may have the
means of fairly comparing his mode of going to work with that
of Oken : —

" The following results, among others, are deducible from the
observations which have been detailed : —

" (1 .) At the earliest period of foetal life the notochord ex-

THE THEORY OF THE VERTEBRATE SKULL. 291

tends backwards, as far as the end of the body ; forwards, only
to the interspace between the auditory capsules.

"(2.) The gelatinous investing mass, which, at first, seems
only to constitute a band to the right and to the left of the
notochord, forms around it, in the further course of development,
a sheath, which ends in a point posteriorly. Anteriorly, it sends
out two processes which underlie the lateral parts of the skull,
but very soon coalesce for a longer or shorter distance. Pos-
teriorly, the sheath* projects but little beyond the notochord ;
but, anteriorly, for a considerable distance, as far as the infun-
dibulum. It sends upwards two plates, which embrace the
future central parts of the nervous system laterally, probably
throughout their entire length.

" (3.) The investing mass of the notochord is the material
out of Avhich the vertebral column and a great part of the skull,
though not the whole skull, are developed.

" (4.) The most essential part of a vertebra is its body.
With the exception of a few cartilaginous fishes, the cartilaginous
foundation of that body (the notochord having disappeared ear-
lier or later), has the form of either a ring, or a half ring ; or,
as is the case among the Mammalia, forms a solid mass, having
the form of the segment of a cylinder. Subordinate parts of
the vertebra are the vertebral arches and transverse processes,
together with the ribs, which all, at the time they take on a car-
tilaginous character, appear as rays of the body, though some-
times they are not developed at all. Only in rare cases (Petro-
myzon) are vertebral arches developed without vertebral bodies ;
that part of the investing mass of the notochord which is, in
other cases, applied to the formation of such bodies, acquiring
only a membranous consistency.

" (5.) From that part of the investing mass of the cephalic
part of the notochord, which consists of the anterior part of the
sheath of the notochord and its anterior paired processes, are
developed the basi-occipital, the basi-sphenoid, and the ethmoid,
so that the ethmoid is the most anterior of the parts of the
skeleton which take their origin from the investing mass of the
notochord. The basi-occipital is formed in that part of this

* Perhaps with rare exceptions, as in Fistularia tabaccaria.

v 2

292 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

mass which surrounds the cephalic part of the notochord like a
sheath ; the basi-sphenoid, in that part of it which lies between
the paired processes (the trabecule) and the anterior end of the
notochord ; and the ethmoid (more particularly its body, or pars
perpendicularis), in the anterior coalesced part of those two pro-
cesses. The body of the presphenoid, on the other hand, is
formed below the processes in question, rarely between them.

" (6.) The parts of the skull just mentioned, however, do not
ossify in all Vertebrata with an osseous skeleton, but one, or
several, of them sometimes remain cartilaginous, and then grow
relatively far less than the others, so that they seem to be pushed
aside and suppressed by the neighbouring bones. This holds
good especially of the basi-occipital of the Batrachia, and of the
basi-sphenoid of these animals and of osseous Fishes.

lt (7.) The basi-occipital (or, at least, the substance out of
which it will become developed) constitutes, originally, like the
body of a vertebra, a sheath round a part of the notochord, and
the ex-occipitals appear, whilst they chondrify, as outgrowths
from the basi-occipital part ; just as the arches of a vertebra,
when this is normally developed, appear as outgrowths from its
already chondrified body. For the rest, however, the normal
development of the occipital bone is quite similar to that of a
vertebra, and it therefore may with perfect justice be held to be
a cephalic vertebra.* The squama occipitis, which occurs in
many, but not in all Vertebrata, and which is not always placed
between, but sometimes lies in front of the ex-occipitals, presents
no difficulty in the way of this interpretation ; it is an accessory
structure, a so-called intercalary bone, the presence of which
depends upon the excessive development of the brain.

" (8.) The two rings, on the other hand, which are formed
by the two sphenoids, with the parietals and frontals as their in-
tercalary bones, are no longer constructed upon quite the same
type as the vertebra. That the alisphenoids and orbito-sphenoids.
when they are already chondrified, do not appear to take the
form of outgrowths of their centres, but are united with them by
membrane, need not, perhaps, be taken very much into account,

* The Foramina eondyloidea, which occur in the ex-occipitals of many Verte-
brata, remind one of the holes of the vertebral arches of the Sharks.

THE THEORY OF THE VERTEBRATE SKULL. 293

since, in the Lampreys, the arches of the vertebrae arise inde-
pendently within the lamella?, which the investing mass of the
notochord has sent out to embrace the central parts of the
nervous system. Still less weight can be attached to the cir-
cumstance that not unusually, even when both sphenoidal centra
are present, only one pair of the corresponding ala? appears ;
while, in other cases, two pair of aire and only one central part
are present, since the caudal vertebra? of Mammalia usually
exhibit no traces of arches, and the Lampreys have such arches
without centra. On the other hand, the circumstance is impor-
tant that the basi-sphenoid, although it arises within the invest-
ing mass of the notochord, is not developed around this (as, so
far as our present observations go, even the most posterior
caudal vertebra? are), but in front of it, in a process of the
investing mass ; and that the body of the presphenoid is no
longer developed, even in a part of this mass (except in a few
Mammalia), but arises quite independently of it. Hence, the
two sphenoids no longer agree perfectly with vertebra? in their
development — the anterior diverging more widely from the ver-
tebral type than the posterior.

" (9.) Yet the two sphenoids, like the proper vertebra?, still
embrace segments of the nervous tube (such as is formed by the
brain and spinal marrow, at any rate in the early stages of
development), and they constitute, as the vertebra? at first nor-
mally do, open rings, or rather segments of rings, round that
tube. The ethmoid, however, at no time surrounds a segment
of the nervous tube in question ; but, in a few animals only, im-
perfectly includes, by its hinder part, two anterior prolongations
of that tube, wThence the olfactory nerves arise. Its mode of de-
velopment, and its ultimate form likewise, are of such a character
that it no longer offers any special resemblance to a typically-
formed vertebra. Nevertheless, considering that it arises from
*a part of the prolonged investing mass of the notochord — viz.,
from the anterior, early-coalescing parts of the two trdbeculse —
and that its body (the pars perjoendicularis) presents even a cer-
tain resemblance to the last caudal vertebra? of many Birds and
osseous Fishes, it may well be considered to be a modified vertebra.
We may look at it, in short, as the representative of only the

294 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

body of a vertebra — such as normally each caudal vertebra of a
Mammal is ; and that from this, for the purpose of investing the
olfactory apparatuses, which are developed at its sides, lamellar
processes grow out, which are altogether peculiar to it. In any
case, however, the ethmoid may be regarded as the anterior end
of the vertebral column.

" (10.) From what has been stated, it appears that the four
different groups of bones — the occipital, with its intercalary
bone, the squama ; the basi-sphenoid, with its intercalary bone,
the parietals ; the presphenoid, with its intercalary bones, the
frontals; and the ethmoid, together with its outgrowths, the
spongy bones and the cribriform plate — exhibit in their succes-
sive order from behind forwards, a greater and greater deviation
from the plan according to which ordinary vertebras are deve-
loped, so that the occipital bone is most like a vertebra, while
the ethmoid is least like one.

"(11.) Among the bones of the face, the preinaxillas, the
nasal bones, and the vomer are developed altogether indepen-
dently of the investing mass of the notochord ; and they never
coalesce with parts of the skeleton, which are immediately
derived from the latter. On this account, alone, they cannot be
regarded as vertebras, or parts of vertebras. Furthermore, they
at no time enclose, or help to enclose, a segment of the central
nervous system. The nasal bones and the vomer are, properly
speaking, ' splint-bones ' (Belegungsknochen) for the ethmoid, such
as occur in the vertebras of no animal ; and the premaxillas are
applied, although in a different plane, to the one end of the ver-
tebral column, as, in Fishes, the median rays of the anal fin are
applied to the other end of it.* Furthermore, the palatine bones
are developed, together with the pterygoids, in lateral processes,
or rays, which have grown out from the middle part of the base
of the brain-capsule, and which, as regards their original form,
disposition, and connections, resemble the ribs, and may be"
regarded as a pair of ribs united with the brain-case. In Mam-
malia the two mallei are developed in these two rays, and

* The study of the development of the skull necessitates the assumption that
Sturgeons, Sharks, and Rays have no premaxillpe, and that their skulls end ante-
riorly with the ethmoid cartilage.

THE THEORY OF THE VERTEBRATE SKULL. 295

perhaps the quadrate bones of many other Vertebrata in a part
of them. Around them, however, is developed, in animals
provided with an osseous skeleton, a coating of bony plates,
which becomes metamorphosed into the lower jaw.

" At the outer side of those parts, moreover, in which the
pterygoid and palatine bones arise — or, in other words, along-
side the processes of the ' rays ' — a substance arises, whence the
upper maxilla and the malar bone are developed.

" The upper maxilla and malar bone therefore might be
regarded, like the lower maxilla, as splint bones or rib-like
bones (which, however, do not occur in connection with true
ribs), but not as parts of the vertebra itself.* The lachrymal
bone, lastly, only fills up a gap between other bones of the face,
and therefore, if analogies must be discovered, can only be
regarded as an intercalary bone.

" (12.) The auditory capsules and the petrosal bones, which
are developed out of them in many animals, may, in respect
of their place and origin, be most fittingly compared with those
intercalary bones which occur in Sharks and Sturgeons, between
the arches of the vertebra? ; but, in respect of their form, take a
different course from these. And since those intercalary pieces
can hardly be considered to be parts of vertebrae, the auditory
capsules cannot be regarded as such."

Vogt and Agassiz, resting upon embryological observations
which entirely confirmed those of Rathke, carry out the argu-
ment suggested by the latter more rigorously.

" It has therefore become my distinct persuasion (says Vogt)
that the occipital vertebra is indeed a true vertebra, but that
everything which lies before it is not fashioned upon the verte-
brate type at all, and that all efforts to interpret it in such
a way are vain ; that therefore, if we except that vertebra
(occipital) which ends the spinal column anteriorly, there are no
cranial vertebrae at all." f

* In the Chelonia and a few Mammalia bony elements occur, which cover the
ribs and, in the first-mentioned animals, even become united with the ribs ; they are
developed, however, in the integument, and belong to the integumentary skeleton,
and not to the nervous skeleten, so that they need not be considered here.

f " Entwickelungs-Geschichte dor GeburtsheliVr Krote."— P. 100. 1842.

296 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

But the further investigations of embryologists have demon-
strated that the occipital segment of the skull is, developmentally,
as different from a vertebra as all the rest, seeing that, as
Kemak has more fully proved than any other observer, the
segmentation into "urwirbel," or proto-vertebrse, which is
characteristic of the vertebral column, stops at the occipital
margin of the skull — the base of which, before ossification,
presents no trace of that segmentation which occurs throughout
the vertebral column. By this third great step the vertebral
hypothesis of the skull seems to me to be altogether abolished ;
even though Professor Goodsir, whose thorough acquaintance
with embryology gives his opinions on these subjects great
weight, has endeavoured, in his learned aud ingenious essays,
to combine the facts of development with that hypothesis.

IV. A fourth line of investigation, not bearing so directly
upon the vertebral hypothesis, but still of great moment, was
opened up by the observations of Arendt on the persistent
cartilaginous cranium of the Pike,* and by the subsequent
investigations of Von Bar, of Duges, of Reichert, of Agassiz, of
Jacobson, Sharpey, Spondli, and Kolliker, and all the discussions
which have taken place on the " primordial cranium " question.
The problems attempted to be solved by these inquiries are —
Is there a clear line of demarcation between membrane bones
and cartilage bones? Are certain bones always developed
primarily from cartilage, while certain others as constantly
originate in membrane ? And further, if a membrane bone is
found in the position ordinarily occupied by a cartilage bone, is
it to be regarded merely as the analogue, and not as the
homologue, of the latter ? In other words, is histological de-
velopment as complete a test of homology as morphological
development ?

At present the course of investigation appears to me to tend
towards giving an affirmative answer to these questions ; but
much and careful observation is yet needed.

* " De Capitis Ossei Esocis Lucii Structura Singulari." 1822. Nesbitt, how-
ever, appears to have been the first to direct attention to the difference bit ween
membrane bones and cartilage bones.

THE THEORY OF THE VERTEBRATE SKULL. 297

Having concluded this rapid historical sketch of the gradual
growth of the true theory of the skull, it may be well if I state,
in a brief summary, what I conceive to be the present condition
of our knowledge respecting its structure and development : —

1. All crania result from the modification of the anterior
part of that " primitive groove ' of the embryo, the posterior
part of which gives rise to the vertebral column ; and, at the
very first, there is no discernible difference between that part of
the groove which will give rise to the vertebral column, and that
from which the skull will be produced.

2. The first changes which take place, in both the cranial
and the spinal regions of the primitive groove, are also pre-
cisely similar, the dorsal laminae growing up and uniting
together in the middle line, so as to enclose a cavity which is,
on the one hand, the primordial brain-case, and, on the other,
the primordial spinal canal. So far, a unity of organization may
be predicated of both brain-case and spinal canal ; but the brain-
case is not yet a skull, nor the spinal canal a vertebral column.

3. Beyond this point, the course of development of the
cranial region differs absolutely from that of the spinal region.
In the latter, that histological differentiation takes place which
results in the formation of the proto-vertebrae, while in the
skull no such process occurs. Again, the notochord extends
throughout the whole length of the spinal column ; while, as
soon as the skull is distinguishable, as such, the notochord
ceases to extend beyond the middle of its floor, stopping im-
mediately behind that part which lodges the pituitary fossa.*

4. Furthermore, when chondrification takes place in the
spinal column, separate masses of cartilage are developed in
each proto-vertebra ; but, when chondrification commences
in the base of the skull, it gives rise to a continuous body of
cartilage, which never exhibits any trace of transverse division,
or segmentation ; but is always divided under the pituitary body
into two longitudinally-arranged crura, the " trabecular cranii"

5. Hence it follows that, though the primordial brain-case
and the primordial spinal canal are identical in general j^lan of
construction, the two begin to diverge as soon as the one puts

* Amphioxus forms an exception, probably only apparent, to this generalization.

298 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

on the special characters of a skull, and the other those of a ver-
tebral column ; the latter taking one road, while the skull takes
another. The skull is no more a modified vertebral column
than the vertebral column is a modified skull ; but the two are
essentially separate and distinct modifications of one and the
same structure, the primitive groove.

6. The skull, having assumed its special and distinctive
characters, may pass through three successive states — the mem-
branous, the cartilaginous, and the osseous — in the course of
its development ; and the order in which these states succeed
one another is always the same, so that the osseous skull has
a cartilaginous, and the cartilaginous, a membranous, pre-
decessor. Nor does any one of these states ever completely
obliterate its predecessor; more or less cartilage and mem-
brane entering into the composition of the most completely
ossified skull, and more or less membrane being discoverable in
the most completely chondrified skull.

7. The adult skull may, however, have got no further than
one of these states. In the Ampliioxus, the skull (if skull it
can be called) is membranous. In many Fishes, as we have
seen, it is cartilaginous, with, at most, a superficial conversion
into bone. In the rest of the Vertebrata definite bones are
added, to the more or less complete exclusion of the cartilaginous
cranium.

8. When definite cranial bones are developed, they arise in
one of two ways, either in the substance of the cartilaginous
cranium, as "cartilage bones," or in the perichondrium, or
remains of the membranous cranium, as " membrane bones."
It is highly probable that, throughout the vertebrate series,
certain bones are always, in origin, cartilage bones, while certain
others are alwavs. in origin, membrane bones.

9. With the exception of Amjihioxus, three sets of sensory
organs — olfactory, optic, and auditory — are evolved in the
walls of the skull of every vertebrate animal, and they are
disposed, from before backwards, in the order in which they are
named. All these sensory organs are originally developed in
connection with involutions of the integument, which, in the
case of the olfactory organ, remain open, but, in that of the eye

THE THEORY OF THE VERTEBRATE SKULL. 299

and ear, become shut. Each sensory apparatus is, throughout
the Vertebrate series, related to the same nerves : the olfactory
being supplied by the first pair ; the optic, by the second ; the
auditory, by the portio mollis of the seventh ; while the fifth
pair leaves the skull in front of the auditory capsule, and the
eighth pair behind it. These relations of the cranial nerves to
the sensory organs, and consequently to the cranial walls, are
established antecedently to chondrification, and a fortiori to
ossification ; so that the cranial nerves and the sensory organs
serve as fixed points by which the nature of the various ossifica-
tions can be determined.

10. By the help of these landmarks, chiefly, it has been
possible to identify the bones known as basi-occipital, ex-
occipitals, supra-occipital ; basi-sphenoid, alisphenoids, parietals ;
presphenoid, orbito-sphenoids, frontals ; or, in other words, the
constituents of the walls of the brain-case, throughout the whole
series — from the Pike to Man. And it is found that these
bones, when they all occur together, are so disposed as to form
three, originally distinct, segments.

11. Eecourse to long-established, but frequently-forgotten
facts in the history of the development of the so-called "pars
petrosa" and "pars mastoidea," or periotic bone, of the human
skull, has shown that these parts ossify from three centres,
which have hitherto received no names, and which I have
termed the " pro-otic," " opisthotic," and " epiotic ': bones. It
has been one of the principal objects I have had in view to
prove, by paying careful attention to the relations of these
osseous elements, on the one hand to the nerves, and on the
other to the parts of the auditory organ which they enclose,
that they are very generally represented, sometimes in a dis-
tinct form, and sometimes coalesced with one another, or with
other bones, throughout the series of skulls provided with car-
tilage bones; and that the pro-otic, especially, is one of the
most constant and easily-identifiable bones throughout the series
of vertebrate skulls.

12. The eye is not invested by any cartilaginous or osseous
elements of the cranial wall ; but the olfactorv sacs become
more or less enclosed in a capsule, formed partly by a median

300 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

cartilaginous plate, which results from the coalescence and out-
growth, beyond the boundaries of the brain-case, of the trabecular
cranii ; partly, by outgrowths from the superior and inferior
edges of that plate ; and partly, by a prolongation outwards
of the front part of the outer wall of the brain-case, into an
antorbital process, between the orbit and the nasal sac, on each
side. Cartilage bone developed in the septum gives rise to the
ethmoid ; in the antorbital processes, to the prefrontals ; in the
superior, or inferior, lateral prolongations of the side wTalls, to
the turbinal bones. Membrane bones developed upon the roof
of these olfactory capsules give rise to nasals ; beneath the
septum, to vomers.

13. The ethmoid and its dependencies are developed within
the median " fronto-nasal ' process, which grows out from the
front wall of the embryonic skull, between the rudimentary
nasal sacs ; and the inferior, broad, free edge of which bounds
the mouth. It is in this free edge that the preniaxillae are
developed, and they are, at first, perfectly distinct from the
maxillae. The latter, together with the palatine and pterygoid
bones, are formed within the maxillary processes, which bound
the sides of the primitive oral cavity, and run, parallel with one
another, along the base of the fore-part of the embryonic cranium,
uniting, behind, with the first visceral arch, but being, at first,
completely separated, anteriorly, from the fronto-nasal process.
Clearly therefore, if the premaxillse and maxillae, &c, are to be
regarded as constituents of inferior arches of the skull, they are
not parts of one arch, but of, at least, two distinct arches.

14. Of the first and second visceral arches, which He imme-
diately behind the mouth, the former, which gives rise to the
mandible and quadrate bone, passes into the skull under the
front part of the auditory capsule ; while the root of the latter,
in which a greater or smaller part of the hyoidean apparatus is
developed, underlies the hinder part of that capsule. It is
therefore impossible that the mandibular and hyoidean arches
should be dependencies of any other parts of the skull than
those which lie immediately in front of, or behind, the auditory
capsules ; and in the completely ossified skull we never, as a
matter of fact, meet with these arches in any other position.

THE THEORY OF THE VERTEBRATE SKULL. 30 L

15. There is not a shadow of evidence that the mandibular
and hyoidean arches suffer any shifting of position from before
backwards, in the course of their development; but the extre-
mities of those arches which are attached to the skull undergo
very singular metamorphoses, the effect of which is, that the
dentary part of the mandible is brought into closer connection
with the skull the higher we ascend in the Vertebrate series.
Thus, in the Fish it is separated from the skull by the hyoinan-
dibular, quadrate, and articular bones ; in the Reptile by the
quadrate and articular ; while in the Mammal the quadrate
and the articular are metamorphosed into the incus and the
malleus, and the dentary comes close to the skull, articulating
with the squamosal.

These are, I believe, the most important facts regarding the
structure and development of the skull, which may now be
regarded as well established. If we inquire how they bear
upon theories of the skull, it will be obvious that they place the
doctrine of the unity of organization of the vertebrate skull
upon a perfectly sure and stable footing, while they appear to
me, as clearly, to negative the hypothesis that the skull is, in
any sense, a modification of vertebrae.

But though the skull has not a vertebral structure, and in
its membranous and cartilaginous states is not even segmented,
it assumes a very definite segmentation in its completely ossified
state.

In every well-ossified cranium there is, assuredly, an occipital
segment (" Ear Vertebra " of Oken), formed by the basi-occipital,
ex-occipitals, and supra-occipital ; a parietal segment (" Jaw
Vertebra " of Oken), constituted by the basi-sphenoid, alisphe-
noid, and parietals ; a frontal segment (" Eye Vertebra " of
Oken), composed of the presphenoid, orbito-sphenoids, and
frontals ; and a nasal segment (" Nasal Vertebra " of Oken),
formed by the ethmoid, prefrontals, turbinals, nasals, and
vomer.

Leave out the hypothetical considerations that these seg-
ments are equivalent to one another, and that they are homo-
logous with vertebra?, and Oken's expression of the broad facts

302 ON THE STRUCTURE OF THE VERTEBRATE SKULL.

of the structure of the completely ossified brain-case is, I believe,
the best that has yet been given. Nay, we may go further with
him, and look on the periotic bones as no part of the proper
cranial wall, but as special developments within the otic capsule.
But here we must stop, for neither anatomy nor development
are reconcilable with the notions of the Okenian school respect-
ing the limbs of the head. Carus suggested, from the Okenian
point of view, that the premaxillse and maxillae must be cephalic
ribs, and not cephalic limbs ; but Eathke was the first to
demonstrate that the inferior arches of the skull must be con-
sidered, if they are homologous with anything in the trunk, to
partake of the nature of ribs rather than of that of limbs. But
the confusion between analogy and affinity has led to such
grave errors in the interpretation of the upper arches of the
skull, that we must be upon our guard against running into
similar mistakes with respect to the lower arches.-

It is easy enough to enumerate four inferior arches to the
skull, just as there are four superior arches — the premaxillae
forming the first of these arches ; the palato-pterygoid and
maxillary apparatus, the second ; the mandible, with its sus-
pensorium, the third ; the hyoiclean arch, the fourth : and it
might be plausibly enough reiDiesented that the first of these is
united with the nasal segment of the skull, the second with the
frontal segment ; while the third and fourth, being connected
respectively with the anterior and the posterior parts of the
periotic capsule, might be fairly considered to belong to the
parietal and occipital segments.

But do they really belong to those segments ? and if so,
why do they not remain attached to them ? What relation
have the branchial arches to the skull, again ? It is hard to
see in what morphological character the first branchial arch of
a fish differs from its hyoidean arch ; and if so, is it an arch of
the skull, or an arch of the vertebral column ? What, further-
more, are the original connections of the palato-pterygoid arch ?
Does it grow out of the mandibular arch from behind forwards,
as Bathke seems to think ; or has it, primitively, that connection
with the prefrontal region which is so constant a character of
the palatine bone ?

THE THEORY OF THE VERTEBRATE SKULL. 303

These questions must be answered before the theory of the
lower arches of the skull can be placed upon as satisfactory a
footing as that of the upper arches ; and they can be answered
only by the embryologist, who may be encouraged to the
difficult task by reflecting on what he has done already ; though
keeping in view the adage of the Roman, and

" Nil actum reputans si quid supercsset agendum."

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Lectures on the elements of comparative anatomy"

See other formats