INAS GCr.Fo POG ANS, fe Ce AY
ww (d\ ss Ye vy * a \ \ Aa ¢ x a,
ANSE any) Ai, \\ 4 We t/t f 7s DS

ve hviae vty tt itr uy we a ae “Ss LUN KA [ha) @

;
ai

| ~ i
}
i |

oa, Nees:
Siiehaie mit
Sage

a Le

AM UG: Hg op, fee Sit tft CL 7e.
eS -

ALS TE aN: |

J
We
/
&

A ie ey Ve
BCU TES sel GH

CABINET CYCLOPADIA.

= : GONDON 2 32 oe
Printed'by A. SpPoTTISwOODE, =~
New Street-Square. ,

1 Sent

THE

CABINET CYCLOPADIA.

CONDUCTED BY THE

REV. DIONYSIUS LARDNER, LL.D. F.R.S. L. & E.

MLR.ILA. F.R.A.S. F.L.S. F.Z.S. Hon. F.C.P.S &e, &e,

ASSISTED BY

EMINENT LITERARY AND SCIENTIFIC MEN.
Patural Historp.

ON. THE
NATURAL HISTORY AND CLASSIFICATION

OF

FISHES, AMPHIBIANS, AND REPTILES.

BY

WILLIAM SWAINSON, A.C.G.

F.R.S. LS. ETC.”

VOL. I.

LONDON:

. PRINTED FOR
LONGMAN, ORME, BROWN, GREEN, & LONGMANS,
PATERNOSTER-ROW 3

AND JOHN TAYLOR,
UPPER GOWER STREET,

1838.

] “THE EARTH IS FULL OF THY RICHES—SO IS THE GREAT AND
a WIDE SEA, WHEREIN ARE CREEPING THINGS INNUMERABLE, BOTH
s SMALL AND GREAT BEASTS. “4
— “THESE WAIT ALL UPON aga THAT ‘THOU ‘MAYEST GIVE
TEEM THEIR FOOD IN DUE SPagON.

vere
toe

i
he

OY eos
rs ee SO

INSTORY «)/*

——

MONOCARIDIAN ANIMALS,
aes Bes eet le ees 22 oF

- YU D ? ZG; pe > pp

A. “ S y Yo Zz

ag WLLL CAGE LAL JO 2 YE, oe lL REL 29g, CYC.
' 4 Z ee 44

CONTENTS.

PART I.

ON THE NATURE AND RELATION OF MONOCARDIAN
ANIMALS ; AND MORE ESPECIALLY FISHES.

CHAPTER I.
Preliminary Remarks. — On the general Nature and ‘Relations of Mono-
cardian Animals : - s = Page 1
CHAP. IL .
On Fishes in general 2 - = enc
CHAP. III.
A Sketch of the History and Bibliography of Ichthyology, with some Re-
marks on collecting and preserving Fishes - - ~ 54
CHAP. IV.
On the Systematic Arrangement of Fishes - - ES Te
CHAP. V.
On the Natural Arrangement of Fishes, the Primary Types of Form, and
the Analogies they present to other Classes of Animals - 4 Ys)
CHAP. VI.
On the Order Cartilagines, or Cartilaginous Fishes - - 118
CHAP. VII.

On the Plectognathes, or Cheloniform Order < = - 189

} CHAP. vu
As ton the eee or ‘Anguilliform Order -

‘CHAP. ae

CHAP. X.
_ On the Pleuronectidz, or Flat-fish ; and the Gadide, or Cods

CHAP. XI.-

i ah On the Families of Siluride — the Catefish, or Silures ; and ‘the Cobitide,
ae or Loaches 2 oF = a, = eee
as t 4 ; /
fh
aie
; Ky
ny s
me a
at ? N .
of . 3
1 { ; Ah

ON THE

NATURAL HISTORY

AND

CLASSIFICATION

OF

MONOCARDIAN ANIMALS.

PARLT

ON THE NATURE AND RELATIONS OF MONOCARDIAN
ANIMALS: AND MORE ESPECIALLY OF FISHES,

CHAPTER I.

PRELIMINARY REMARKS. — ON THE GENERAL NATURE AND
RELATIONS OF MONOCARDIAN ANIMALS.

@.) We commenced our review of the animal king-
dom with the proposition that it presented certain primary
types of form, which, under a diversity of modifica-
tions, pervaded all the classes, and all the divisions, of
this part of creation. The assertion was altogether
novel, and somewhat startling, yet it was neither hastily
nor presumptuously made. Many years had been de-
voted to those researches necessary to ascertain its gene-
ral correctness, before we ventured to announce this
theory, four years ago, in our “ Preliminary Discourse.”’
Ample time has thus been afforded for its refutation ;
but so far from this having been accomplished, it has
not even been attempted. We may fairly conclude,
therefore, that we are building upon no verv erroneous
VOL. I. B

2 CLASSIFICATION OF MONOCARDIAN ANIMALS.

foundation; we may presume, in short, that if the facts
we have brought forward were capable of otherinferences,
and other combinations, those we have advanced would
long ago have been overthrown ; for mere individual opi-
nion has nothing to do with the question at issue. We
almost regret, indeed, that this has not been attempted
by some naturalist with talents.and knowledge equal to
the task: such a discussion might have elicited many
truths, and have led to many useful explanations. But
the fact of the matter seems to be, that all those among
our own countrymen whose works have placed them
as the most eminent in the different branches of zoology
they respectively cultivate,—all these, we say, with
hardly a solitary exception*, have expressed their belief
in one or other of the propositions formerly stated.
That the old empiric mode of study is daily giving way
to the inductive or philosophical, is a fact which will at
once be placed beyond doubt, when we mention the
names of MacLeay, Kirby, Horsfield, Westwood, Ste-
vens, and Waterhouse, in entomology; Vigors, Sir
William Jardine, Selby, and Gould, in ornithology ; and
John E. Gray, in general zoology. Now here we actually
have the names of nearly all the most experienced
naturalists and best known authors in the kingdom, who
are thus, from their extensive knowledge of details, the
only competent judges. Each of these have adopted,
either wholly or partially, the theory of the circularity,
the parallelism, or the symbolical relationship of natural
groups. If the weight of authority, therefore, was to
become the test of truth, the Quinarians may well
exult in their strength. But this is not all,—the spirit
has spread far and wide: we could name a long list of
students, some indeed alréady masters, both in England
and our colonies, who have caught the spirit of induc.
tive zoology, and are now pursuing it with an ardour
and a success that will soon render them worthy to fill
the seats of those among us who may drop, full of years

* T believe I should bring into this list our admirable entomologist Mr.
Curtis; but I know not exactly where his opinions have been expressed.

; see

PRELIMINARY OBSERVATIONS. 8

and of honour, into the silent grave. The day, in short,
has gone by when mere opinions, unsupported by argu-
ments, will have any effect among those whom they are
intended to influence ; or when new systems, built on an
imperfect acquaintance with only one division of zoology,
will be at all regarded by those who can alone give them
notoriety ; for higher naturalists have long dismissed the
idea of studying nature under such narrow and purblind
views. Our firm belief indeed is, that as these systems
of late have emanated only from students, their very
authors will throw them aside when greater experience
shows how artificial and futile they really are.

(2.) In the two most perfectly organised classes of ani-
mals, quadrupeds and birds, we have endeavoured to
show the prevalence of these primary forms, and the
harmony that results from tracing their modifications.
We are now to make a similar effort in respect to the
remaining vertebrated classes. Our investigation, how-
ever, of the natural arrangement of these animals must
be conducted, in part, on a different plan to that we
have pursued in ornithology. We must occasionally
adopt the synthetical rather than the analytical process
of investigation ; or, in other words, we must presume
that our propositions, in the abstract, are correct, and
that we have only to extend them to another class: we
do this, not from choice, but from necessity. In the
first place, the state of ichthyological science, to which
the greater part of our two volumes will be devoted,
however rapidly it has advanced in a knowledge of
groups and species, is, and long must be, from the very
nature of the animals upon which it treats, considerably
behind ornithology. Inhabiting an element whose re-
cesses cannot be explored by man, and with a peculiarity
of structure and of colouring which renders their bodies
very difficult to preserve, the natural history (properly
so called) of fishes, when compared with that of ter-
restrial animals, will ever remain little more than a col-
lection of a few superficial anecdotes ; while, from the
difficulty of their preservation and the unattractive

B 2

4. CLASSIFICATION OF MONOCARDIAN ANIMALS.

appearance they then exhibit, few will study, and still
fewer will collect them. Hence the ichthyologist has
much greater difficulties to contend with, in regard to
materials, than he would experience in any other division
of the Vertebrata, while he finds himself totally at a loss
for that information on their natural habits, “ their lives,
and their loves,” which gives such a charm to the his-
tory of other animals, and excites such a popular interest
with the generality of readers. But to these difficulties
lying in the way of nearly all who write upon ichthyo-
logy, must be added others, more particularly applica-
ble to our present undertaking. So little has been done
towards a natural classification of fishes, more especially,
that to attempt those rigorous definitions we have ven-
tured upon in the class of birds, would be altogether im-
possible. The synthetic mode of investigating our sub-
ject is, therefore, that which we shall in many instances
adopt. We shall set out, it is true, with the impression
that the same general laws which regulate the forms of
quadrupeds and birds will be equally apparent in mono-
cardian animals. But this belief is not to be received
as true, upon trust; it is not to be unsupported by facts,
or to remain as a mere assertion. We shall not, indeed,
begin with analysing the smaller groups, and then gra-
dually proceed to higher assemblages; for this is the
analytical method of investigation—the very reverse of
the synthetic: we shall, on the contrary, take a com-
prehensive view of those large assemblages, or groups,
which nearly all our predecessors have agreed to keep
distinct, however they may have differed in their sub-
ordinate details, or in the series wherein they have
placed them. These we shall endeavour to define by their
most prominent characteristics, and combine in such a
way as that no palpable violation of nature should be
committed. We shall then proceed to the results: it
will then be seen how far this arrangement is in har-
mony with our previous disposition of the other Verte-
trata, and how far it is supported by the analogies or
resemblances that may be traced between them and the

PLAN OF THE TREATISE. 5

primary types, which we set out with supposing to
exist.

(3.) Our introductory notices will contain, as formerly,
a rapid sketch of the chief peculiarities of these classes,
more especially in regard to their external anatomy ; not
only as being that part of their structure most essential
for determining the species, genera, and families, but
also because it can be best understood and comprehended
by all. Our own classification, in fact, is as strictly
founded on anatomical organisation—even more so—
than those of any of our predecessors; with this differ-
ence only—that we have selected the more determinate
characters for the definition of our groups, rather than
resting them solely upon one or two, It. is a law of
nature, that the internal and external anatomy are mu-
tual indexes to each other; and, therefore, to give a
preference to internal characters (manifested only to the
view by skilful dissection) over such as are apparent
externally, has ever appeared to us not only objection-
able, but absolutely unnecessary. The study of zoology,
from its vast increasing extent and consequent difficulty,
stands in need of every help and of every facility for its
acquirement. Comparative anatomy, indeed, is of the
highest importance in determining questions which
could otherwise not be solved; but among vertebrated
animals, at least, this study seems to have been pushed
much too far; and if ichthyological groups are to be
regulated by the bones of the head and the armature of
the mouth, we may, with equal propriety, draw up an
ornithological system from the structure of the wind-
pipe, the form of the sternum, or the number of the ver-
tebre.* Were it possible to frame such systems — which
it manifestly is not—very many facts, of peculiar interest
to the mere comparative anatomist, would unquestionably
result. But the question arises, of what practical use
would they become? The great mass of mankind look

* Since this was written, a system, much on this plan, bas actually been
but forth in one of our periodicals: the next month will probably bring out

another, founded on the structure of the gizzard, or the bones of the cra-
nium, to add to the ninety and nine that have already died natural deaths.

B 3

6 CLASSIFICATION OF MONOCARDIAN ANIMALS.

to scientific men for placing the different branches of
human knowledge before them in the most easy and
comprehensible form consistent with sound philosophy ;
and however highly they may estimate the profundity
of those who expatiate on the intricacies of their art,
they will most assuredly follow and admire such writers
only as choose an opposite course ; and by the simplicity
of their instruction, and the facility with which their re-
searches may be verified, hold out attractions to those who
desire to see science disencumbered of all unnecessary
mystery, abstruse technicalities, or empirical assertions.

(4.) In prosecuting our labours upon these principles,
we shall, in the first place, inquire into the station occu-
pied by the monocardian animals in the zoological circle;
and then, taking each of the classes of fish, amphibia, and
reptiles, separately, condense the most remarkable and
essential facts relative to their organic structure, both
internal and external. Of these three classes, Icatuyo-
LoGy, or that which treats of fishes, will claim our first
and chief attention, not only as being by far the most
numerous and interesting, but also because it is that
with which we are most conversant. Ichthyology, in
fact, engaged our attention long before ornithology ; and
no opportunity has been lost, during a period of twenty-
three years, of making drawings and descriptions from
living specimens in all those foreign countries we have
visited. Many years’ residence in Sicily and other parts
of the Mediterranean will enable us to give much inform-
ation, hitherto unpublished, on the rarer fishes of
those coasts, sufficient, at least, to show how imperfectly
they have as yet been made known. Our information on
the reptiles and Amphibia is more confined; but as the
determination of the natural groups, and not the species,
is our chief object, this circumstance becomes of less
consequence. In this we have derived much assistance
from the labours of our friends MM. Gray and Bell,
as well as from the numerous and valuable continental
works published of late years on these animals.

(5.) Fisus, along with frogs and reptiles, constitute

RANK AMONG THE VERTEBRATA. ff

that great and primary division of vertebrated animals
which are distinguished by their cold blood, in oppo-
sition to the two classes of quadrupeds and birds, which
‘have their blood warm. In all cold-blooded Vertebrata,
the body is either naked—that is, merely covered by a
skin more or less thick—or it is protected by osseous
pieces or plates: in some, these plates are excessively
hard, and are joined together at their edges, as in tor-
toises, and in some few of the aberrant fishes ; but in
the majority, both of the fishes and reptiles, the plates
assume that form denominated scales, the outer edge of
one reposing upon the base of the next.

(6.) The rank of the monocardian classes in the
circle of the Vertebrata has already been touched upon.*
All naturalists, both ancient and modern, agree in con-
sidering them—what, indeed, is self-evident— as the
most imperfect or least organised of vertebrated animals ;
from the types of which, as seen in quadrupeds and birds,
they are at once distinguished by their cold blood,—a
character which 1s perfectly absolute, inasmuch as no
exception has been yet discovered: for no quadruped or
bird, now in existence, has any other than warm blood.
M. de Blainville, we believe, was the first naturalist who
absolutely arranged the Amphibia, or frogs, as a distinct
class from the true reptiles. And although this im-
provement on the old method has not been adopted in
the Réegne Animal, it has generally been followed by
subsequent naturalists. Indeed, the very circumstance
of the amphibians, as Cuvier himself says, passing from
the form of a fish respiring with gills, to that of a rep-
tile respiring by lungs, is quite sufficient to separate
them both from fish and serpents ; since this very struc-
ture points them out as forming a link by which the
two are connected. The scientific world, however, have
long made up their minds on this question ; and we thus
find the aberrant division of the vertebrated animals
resolvable into three others, namely, 1. Pisces, or fishes ;
2. Amphibia, or frogs ; and, 3. Reptilia, or serpents.

* Classif, of Quadrupeds, p. 45. Classif. of Animals, p 204,
B 4

8 CLASSIFICATION OF MONOCARDIAN ANIMALS.

(7.) The relations which these animals bear to
quadrupeds and birds may next be glanced at. Com-
mencing with fish, we find that the dolphins, porpoises,
and the other aquatic Mammalia without feet, were
always regarded by the ancients as true fishes ; and even
Artedi, the great renovator of ichthyology in the eighteenth
century, viewed them in this light. The passage, there-
fore, from quadrupeds to fish is absolutely perfect ; and
the affinity of the sharks to the dolphins shows that this
passage takes place among the cartilaginous fishes ; of
which Cuvier remarks, that they also evince an affinity
to the Reptilia. Fishes are remarkable, among their
other peculiarities, for being destitute of feet; these
members being replaced or represented by two sets of
fins ; the pectorals representing the anterior feet of four-
footed beasts, while the ventral fins equally represent’
the hinder feet. But among the least perfect or aberrant
groups of this class we find these fins so constructed,
that they are placed on a jointed peduncle, so that they
have nearly as strong a resemblance to the foot of a frog,
or that of a swimming bird, as toa fin ( fig. 1. @) ; nor
is this in appearance only ; for it has been frequently
asserted by those who have seen the Indian Chironectes,
or frog-fishes, alive, that those singular animals crawl
about by means of these foot-fins, and that they are
so far amphibious as to live comfortably two or three days
out of water. Their thick grotesque shape, naked and
tuberculated body, and their whole general aspect, give
them, in short, much more the appearance of frogs than
of fishes, —an assertion to which the most unscientific
of our readers will acquiesce upon looking to the annexed
cut of the Malthe nasuta Cuv. (fig. 1.), accurately drawn
from a specimen we procured on the Brazilian coasts: (ais
the pectoral fin.) Nor is this a solitary affinity between
the amphibians and the fishes; the whole of Cuvier’s
genus Chironectes, which is evidently a natural family,
abounds with similarly formed animals, where the gene-
ral aspect and characters of true fishes are so much
changed as to assume the appearance of frogs. Quitting

RANK AMONG THE VERTEBRATA. | 9

the fishes by these compound animals, we enter among
the batrachians, or Amphibia ; composed of the frogs, toads,
sirens, salamanders, efts, and a few other lizard-formed

animals, distinguished from all other Vertebrata by the
heart having but oné auricle, the body naked, and the
whole animal undergoing metamorphosis before it reaches
maturity. All these are furnished with either two or
four feet ; but sometimes these members are so srnall,
that they appear more as rudimentary appendages ; while,
in their eel-shaped bodies, they so much resemble
many of the apodal fishes, that it may hereafter become
a question whether the true passage between the classes
is not effected by the eels in one, and the sirens in the
other. So closely do the salamanders, again, resemble
the lizards, that none but professed naturalists can tell
their difference ; so that the classes Amphibia and Rep-
tilia are thus inseparably linked. . The connection of
the saurian reptiles, or lizards, to the ophidians, or
serpents, need not here be insisted upon. The passage
from these latter to the gigantic Ichthyosauri is again
rendered easy by the Plesiosaurus, where the head and
neck of a serpent seems engrafted, as it were, on the
body of an Ichthyosaurus. Lastly, it is quite evident
that the flying lizards, or Pterodactyli, belong to the
same great group, and to the same era as the aquatic
monsters of a former world just mentioned ; and it is
equally certain, that of all the reptiles yet discovered,
these make the nearest approach to birds: the fore-
feet, in fact, were dilated into wings, like those of a

10 CLASSIFICATION OF MONOCARDIAN ANIMALS.

bat, while the hinder ones were clearly intended for
walking: the jaws are enormously prolonged, analogous
in their length to those of a woodcock: the whole
structure, in short, is such an extraordinary compound
of a reptile and a bird, that no doubt can remain on
the affinity between the two classes ; for although the
passage is not marked by existing animals so clearly as
that between quadrupeds and fishes, it is quite evident
that the Pterodactyli are more allied to birds than to any
other vertebrated animals out of the class of reptiles.

(8.) By thus tracing the natural series of the verte-
brated animals according to their affinities, we find they
form one great circle. Commencing with quadrupeds, we
pass on to fishes ; to these succeed the amphibians and
the reptiles: these latter are followed by birds ; and
birds, as already explained, bring us back again, by a
different route, to quadrupeds.

(9.) We are now to investigate, however, the truth of
another proposition formerly stated regarding natural
groups ; which was, that the aberrant divisions of every
eircle formed a distinct circle by themselves, quite in-
dependent of their union with the two typical circles.
Now, the: aberrant divisions of the Vertebrata are the
fishes, amphibians, and reptiles. Two questions there-
fore arise: first, Is there not a greater similarity between
these three, than there is between them and quadrupeds
and birds? and secondly, Is this similarity so strong as
to favour the belief that they actually do unite into a
circle of their own? The first question must, of course,
be answered in the affirmative; for although an ordinary
observer might easily mistake an eel for a serpent, a
salamander for a lizard, a young frog for a fish, or
even a Chironectes for a frog, no one is likely to confound
any one of these animals with a quadruped or a bird.
As to the second question, we have the opinion of
Cuvier, —an opinion adopted by others, that many of the
cartilaginous fishes evince a decided affinity to the
reptiles; and this is the very point where the two
extremes of the monocardian animals would meet, if —

PECULIAR CHARACTERS OF FISHES. 11

they really formed a circular group by themselves.
Again, it is notorious that some of the eels of the
genera Murena, Ophisurus, &c. have so completely
the aspect of the water serpents, that it is only upon
the naturalist examining them, that their different
classes are detected.* Upon the whole, therefore, we
must admit that every thing yet known regarding the
classes in question strengthens this belief, and adds
another instance to what we have seen among birds,
that the “ primary divisions of every group are three
actually, but five apparently.” |

CHAP, II.

ON FISHES IN GENERAL,

(10.) FisHes constitute by far the most numerous class
of vertebrated animals, whether we regard the number of
individuals, or the variety of their forms. When we
consider that more than two thirds of the globe is
covered by water, —that element peculiarly appropriated
for their habitation,—we shall not be surprised at this
superiority of numbers. On the contrary, we may
fairly suppose that not more than one half of the
species really existing have yet been made known.

(11.) The peculiarities in the inward form of fishes,
by which they are distinguished from ail other animals,
need not be enlarged upon ; yet, as many of them, like
the eels, assume the form of serpents, and others re-
semble the young of the amphibian frogs, it is neces-
sary to characterise them as aquatic vertebrated animals,
breathing by means of internal gills, and undergoing

* The museum of the Zoological Society contains many striking illus-
trations of this fact.

12 CLASSIFICATION OF FISHES.

no metamorphosis. These gills, or branchia, as every
one knows, are composed of certain semicircular arches,
fringed, as it were, with thin fleshy processes, resembling
little leaves, or lamine, having innumerable blood-ves-
sels: the water taken in by the mouth, again escapes
between the openings of the branchia, which are pro-
tected externally by certain bony plates united together,
yet generally movable, which are called the opercule,
or gill covers: this apparatus for breathing is variously
modified, but never lost, so that it is the primary cha-
racter by which fishes are at once distinguished from
reptiles and amphibians.

(12.) The operculum, or gill cover, just mentioned,
is articulated on the os tympani, and is moved upon a
piece called the pre-operculum: it is composed of three
bony plates, termed the operculum, the sub-operculum,
and the inter-operculum; the modifications of which
are sometimes of much use in determining natural
affinities. In many groups, however (as in the eels),
the operculum is so entirely covered by the common
skin as not to be visible but upon dissection; and
among the cartilaginous and some other fishes, the oper-
culum is entirely wanting.

(13.) The sxenxron of all fishes, except such as
are lowest in the series, present a vertebrated column,
and other internal bones; but the structure of these
bones is very different, and, as may be expected, indi-
cate the primary divisions of the whole class. In the
most perfectly formed fishes, the bones are completely
osseous, and generally of great hardness: in another
large division, they are fibro-cartilaginous— that is to
say, the base or heart of the bones is of gristle, or is
cartilaginous, mixed only with fibres or layers of phos-
phate of lime, so that the texture is never so hard as in
the osseous groups just mentioned: some of these semi-
cartilaginous genera, indeed, have their bones quite
soft, and thus lead to the third group, or truly car-
tilaginous fishes, which, like the sharks and skates,
have their skeleton composed of gristle or cartilage

INTERNAL ANATOMY OF FISHES. 13

only. In no fishes, however, is there any medullary
canal. The more perfect groups or orders have ribs,
but these disappear in many of the fibro-cartilaginous
genera ; and finally, in such as pass into the annulose
animals or insects, the whole skeleton is soft and mem-
branaceous: first the fins, then the eyes, and, finally,
the vertebrated column itself, almost disappear, so that
we have the mere external a of a@ worm, provided
with a mouth.

(14.) As fish are destined to inhabit an element
where motion is much more essential to them than
either to quadrupeds or birds, their Omnipotent Creator
has given them greater powers for sustaining this motion
than are possessed by any other animals in creation.
Their body, in fact, is surrounded by fins; and their
tail (the fin of which acts as a rudder) is generally
as thick, and often much longer than the body itself.
These are the only members adapted for motion pos-
sessed by fishes; but their construction, number, and
position, are varied in almost an infinity of ways,
and thus contribute some of the most obvious and na-
tural characters for determining the different families
and genera. As the formation of the fins comes under
the head of external anatomy, we shall subsequently
treat of these members more at large.

(15.) The arr-Biapper is situated immediately under
the spine: by being compressed or dilated, it influences
the specific gravity of the fish, and assists it in rising
or descending in the water, This vessel, however, is
very partially possessed ; and even its presence or ab-
sence may be detected in genera, and even species,
which are closely and internally allied, so that it be-
comes of no value whatever as a character for desig-
nating groups.

(16.) The mourn is sometimes provided with very
strong teeth, and sometimes entirely without ; and this
remarkable variation takes place in genera close to each
other, and even, according to some of Cuvier’s groups.
in species of the same sub-genus. The anatomical

14 CLASSIFICATION OF FISHES.

structure of the mouth may be thus described : — The
inter-maxillary bone, in most fishes, forms the edge of
the upper jaw, behind which is the os Jabiale, or max-
illary ; a palatine arch, composed of the palatine, the
two pterygonian processes, a jugal bone, a tympanic and
squamose bone, constitutes, as in birds and reptiles, a
sort of interior jaw, and supplies, behind, an articula-
tion to the lower jaw, which has, in general, two bones
on each side. In such fish as have teeth, these pro-
cesses are varied in innumerable ways: they are found,
for instance, on the inter-maxillaries, the maxillaries,
the lower jaw, the vomer, the palatines, the tongue, the
arches of the gills, and even on certain bones, behind
the latter, called by Cuvier ossa pharyngis.

(17.) A few other anatomical characters may be
briefly noticed. The nostrils are situated between the
eye and the end of the muzzle or upper jaw, and are
usually double, that is, opening by two perforations on
each side. The eyes are usually rather large for the
size of the body ; but in some types they become very
small, in which case they are always situated on the top
of the head, and are then termed vertical ; the cornea
is very flat, the aqueous humour small in quantity,
while the crystalline is nearly globular, and very hard.
The tongue is small, hard, and bony; so that the taste
enjoyed by fishes, must be very trifling.* The stomach
and intestines present nothing essentially peculiar: in
the generality of fishes, the pancreas is represented
either by ceca of a peculiar tissue, situate round the
pyloris, or by this tissue itself, at the commencement of
the intestines: the kidneys are placed on the sides of
the spine ; but the bladder, contrary to what is seen in
quadrupeds, opens behind the anal and the generative
organs. The majority of fishes are oviparous ; but the
cartilaginous order, and a few others representing them,

* This sense, indeed, is rendered almost unnecessary, for the great ma-
jority of fishes swallow their food whole. This is one ef the great charac-
teristics of the fissirostral type of birds; and as the fishes represent the

same type in the circle of the Vertebrata, we are accordingly prepared to
expect such coincidence.

EXTERNAL ANATOMY OF FISHES. — FINS. 15

are viviparous, the young being protruded through a
very short canal. |

(18.) On the external anatomy of fishes, and of
their natural history, we shall be less concise. Next to
the structure of the bones, the fins claim our greatest
attention ; since itis anacknowledged fact, that the organs
of locomotion are those which have furnished the best
characters, above all others, for distinguishing the various
subdivisions, not only of vertebrated, but of annulose
animals. We shall first describe their number and
position, and then point out several interesting con-
clusions resulting therefrom.

(19.) There are rive sorts of Fins possessed by the
typical groups ; which are named pectoral, ventral, anal,
dorsal, and caudal : the two first of these are in pairs, and
are the most important, inasmuch as they represent those
members in the higher organised Vertebrata, that are
called legs and wings. The pectoral fins, in fact, are
only the anterior feet of quadrupeds, and the wings of
birds, presented under a new and strikingly different
form: the three other fins are single, or, in other words,
they are not in symmetrical pairs. ach of these will
require a separate consideration, more especially as they
have hitherto been regarded with little attention.

(20.) The pectorals are, obviously, the most im-
portant to fishes in general; because we find them in
groups, where several of the other fins are wanting, and
it is only among the lampreys, and a very few genera, so
low in the scale as to form a passage to the worms, that
they disappear. In the majority of fishes they are of the
same moderate size as the ventrals, but in particular fa-
milies they become much more developed: they arealways
composed of flexible*, and, generally, branched rays,
so as to yield to every stroke on the water made by fishes
in the act of swimming. When the shape is pointed
or triangular, the first ray is either very strong or spinous.
This spine, in the silure family, is not only remarkably

* The only exception we are aware of at this moment, is a small species of
blenny, the B. variabilis of Rafinesque, whose pectoral rays are all spinous.

16 CLASSIFICATION OF FISHES.

thick, but is generally barbed on one or both sides, so
that it becomes a formidable weapon of defence. The
great importance of the pectoral fin to the rapid motion
of fishes is still further manifested by the fact that, in
all such groups as are peculiarly rapid, the pectoral
fin is pointed, or rather triangular. The flying fish, the
tunny family, the rays, are familiar examples of this
form in its highest state of development; while we
find the same, in a less degree, among the Spari, the
herrings (Clupeide), the typical cod-fish, Gadiade,
and many others. These, in fact, are nearly all pelagic
fishes, performing, like fissirostral birds, either annual
migrations, or living almost entirely in the open sea
except at the breeding season. In such families, on the
contrary, as live in rivers and lakes, or only in shallow
rocky shores, the pectoral fins are always round. The
whole of the apodal or anguilliform order, in which are
the eels, the lampreys, and the suckers, together with
the blennies, gobies, the rocklings (Motel/a), and nu-
merous other families, are of this description. Even the
Triglide, or gurnards, and their allies, although their
pectorals are of an extraordinary size, yet, with the ex-
ception of those of the fissirostral types, they are always
round ; and it is well known that these fishes keep near
to the shore, and live near the ground. There are some
singular modifications of the rounded pectoral, which
deserve particular notice ; for they are either, 1. partially
cleft; 2. digitated ; or, 3. very broad at their base, and
extended under the throat. Examples of the first
are seen in the genera Lepidopus and Cheilodactylus,
and in a very few others, where the middle rays are
shortest, so that the fin appears lobed in the middle ;
but in the last named genus, some of the lower rays, or
those nearest the belly, are much longer than the others:
the fin has thus an appearance of being injured. The
pectoral of Cephalocanthus is represented (Cuv. pl. 73.
77.)* as being rounded, but divided in the middle,
without any diminution in the length of the rays. In

* J have not had an opportunity of examining this rare fish.

PECTORAL FINS. 17

Cirrhites, again, the last five rays are not only spinous,
but much thickened and prolonged beyond the mem-
brane ; a structure which excites a strong suspicion of
this genus being analagous, in its own circle, to Cheilo-
dactylus ; and this seems the intermediate state of deve-
lopment leading to the next. 2. Digitated pectorals
are exclusively confined to the typical Canthileptes, or
gernards, and spine cheeks (Scorpenide). Among the
first, and particularly in the genus T'rigla, there are
three detached, finger-like processes, unconnected by
any external membrane, which are situated just before
the lower base of the pectorals, and which almost appear
to be detached rays of this fin, much thickened, and
somewhat removed from the connected rays. In the
Mediterranean and Atlantic Dactylopteri, these ex-
traneous rays are united by a membrane, so as to form
a spurious or supplementary pectoral ; while in those
from India, of which there now appears to be several
species*, this supplementary fin is united to the true
pectoral so as to form but one. The pectorals of nearly
all the remaining families of this tribe have the lower
rays, or those nearest the throat, thickened ; much in
the same way as in Cirrhites, but with this difference,
that the fins are so broad at their base, on account of
the number of rays, that they are often carried half-way
under the throat; a character so very peculiar, that we
look upon it as a distinct modification. It seems pro-
bable that this unusual strength is given to such fish as
have very large heads, for the purpose of additional
support ; for it will be observed that the heads of all
these genera are not only large, but particularly heavy,
on account of the bony armature with which this part
is covered ; and it may be further remarked, that it is
among small headed fishes we find the most delicate
pectorals. In truth, however, nothing can be affirmed
with any degree of confidence on the reasons of these
variations from the ordinary structure. We cannot,
as in land animals, watch the habits and explore the

* See the Appendix, wherein these are described.
VOL. I. Cc

is CLASSIFICATION OF FISHES.

haunts of these marine creatures: the most we can do,
is to found our conjectures, on such matters, upon ana-
logical reasoning.

(21.) The pectoral fins, as being the wings of fishes,
are consequently found in the highest state of develop-
ment in such families as represent the fissirostral and
natatorial birds, whose powers of flight are so superior te
others. Hence we find that in the two chief families
of the cartilaginous order, namely, the sharks and rays,
these fins are universally very large, and in the latter they
are so much developed as to occupy more than one half
the surface of the body ; they appear, in fact, to be sur-
rounded and enveloped in their enormous pectorals,
“which, being generally angulated or pointed, must give
to these rapacious monsters a swiftness of swimming
analogous to that possessed by their representatives, the
swallows, in flying. The pectorals of the sharks,
although not proportionably large, still exceed all the
other fins in size; and thus render them such rapid
swimmers. The actual volatile powers of the flying-
fish is, likewise, entirely owing to the enormous size of
their pectorals; but there is nothing peculiar in their
shape or construction, since they merely have the form
and structure of an ordinary pointed fin, only excessively
enlarged. Now, as we find these fins are very complete
in fishes which are constantly moving about in their
watery element, as birds do in the air, so, among such
as are more stationary, and swim but little, the pectorals
are proportionably small: this is particularly observable
in the family of the Pleuronectide, or flat fish, whose
whole structure is adapted for laying flat upon the
bottom of the sea, and there waiting for their prey in
ambuscade. These fishes, in proportion to their size,
have the smallest pectorals in the whole class ; while
the flying fish, which habitually live only in the wide
ocean, and are perpetually traversing it, have the
largest. The Lophide, or fish-frogs, again, may almost
be said to have no real pectoral fins, inasmuch as these
members are so formed as to perform the office of feet,

ee 8

VENTRAL FINS. 19

with which they doubtless crawl on the bottom of the
sea, just as they are known to do when placed upon
land. The pectorals of the Malthe nasuta, as before
remarked (fig. 1. a), are rather paddles, or cartilaginous
lobes, than real fins ; the rays are numerous, but so close
together, and the membrane which connects them so
tough and inextensible, that we feel fully persuaded they
are more used for walking than for swimming.

(22.) Having just mentioned the processes of the
pectoral fins in the T'riglide, we may here notice those
of the genus Polynemus, which are strikingly analogous
to, although very different in structure from, the digitated
processes of the former. The general form of these
fishes bears much resemblance to the grey mullet,
while their serrated gill-covers show a relation to the
percoid families ; from both of these, however, they
are too distinct to be classed as a subordinate group ;
while the fact of these and the Tvriglide being the only
genera possessing pectoral processes, has induced us to
class them as the representatives of each other. In
Polynemus, these processes assume the form of slender,
setaceous, and articulated rays, varying in different spe-
cies from four to ten on each side, where they are inserted
a little in advance of the pectoral, and are sometimes so
long, as in P. paradiseus, as to exceed the entire length
of the whole fish. Although this and several other spe-
cies are by no means uncommon in India, the use of
these processes remains to this day entirely unknown.

(23.) The venrrat fins rank next to the pectoral, as
representing the hinder feet of four-footed animals, and
the legs of birds. That they are less necessary, however,
to the swimming motion of fishes, than either the dorsal
or caudal, may be presumed from the fact, that in the en-
tire order of Apodes, or eels, these fins are totally wanting:
they are the smallest in size of all the others, but by no
means always so. In general they are less than the pecto-
rals, often of the same size, and very rarely, as in Gym-
netrus and Zeus, considerably larger. Much diversity is
observable in their situation and form: like the pectorals,

© 2

20 CLASSIFICATION OF FISHES.

the rays are always soft; but those in the spine-rayed
fishes are strengthened by an external spine, which of
course is never branched. Linneus employed the situ-
ation of the ventral fins to construct some of his primary
divisions; classing together those which had the ventral
placed before the pectoral, those in which it was imme-
diately beneath, and those where it was placed behind.
This plan did very well in the infancy of our science;
but it was soon discovered that this artificial arrange-
ment separated the most natural and connected genera
into different orders, and that even, if rigorously acted
upon, individual species would be similarly dissevered.
M. Cuvier has therefore, with much propriety, rejected
these divisions, and yet not so thoroughly but that some
of his great groups are formed nearly on the very same
artificial principles as those of Linneus.* The situation
of these fins, however, is by no means unimportant,
when used for subordinate characters: in some, as in
Pieracles and Uranoscopus, they are placed immediately
under the throat; in others, as the sharks, they are
nearly half way between the pectoral and the caudal ;
while in that extraordinary genus, Polypterus, it is
close upon the base of the caudal fin.

(24.) The shape is no less diversified: in the great
majority of fishes it is symmetrical with the pectoral fin;
both being either round, as in the Labrine, or pointed,
as in the Sparing. Several instances occur, however,
where this uniformity is disturbed: in some of the
Chetodonide, the pectorals are obsoletely rounded (as in
Platax teira Cuv.), but the ventrals are particularly long
and pointed; while in Tauricthys varius (if the figures
of these two singular fishes are correct) the pectorals are
acutely pointed, while the ventrals are decidedly rounded.
We cannot but entertain some suspicions, however, on
the correctness of these figures; and, indeed, the diffi-
culty of making accurate drawings from preserved fish,
whether dried or in spirits, is frequently so great, that

#* Such, for instance, as the divisions of the order Malacopieryges, and
the insertion of Trachinus and Uranoscopus among the perches, because
they have jugular ventrals.

VENTRAL FINS. Q1

some allowance must always be made on this head, even
to the best artists. The more unusual forms of the
ventral fin may now be noticed. Sometimes it is
single, and merely represented by a prickle, as in
Psettus Sebii, and a large number of the cheloniform
fishes (Balistes Linn.) ; more rarely there are two spines,
representing the two anal fins; for although, strictly
speaking, the bony processes in the cheloniform
fishes are not real ventrals, yet, as they perform the
same office, and are placed in the same situation, we
see no reason why they should not be so termed. In
the type of the genus Xiphias, or sword-fish, the ven-
trals are entirely wanting ; but in the sub-genus His-
tiophorus they consist of two slender cirriform filaments,
either of equal lengths, as in H. indicus, or with one
shorter than the other, as in H. pulchellus. The ma-
jority of the Gadiade, or cods, and of the Blennida, or
blennies, show us an equally slender form of ventrals ;
- sometimes with a single worm-like ray, forked towards
the middle, as in the hakes (Physis Cuv.) ; and some-
times with three, four, or five other rays: yet these
latter are generally so diminutive, that they become
merely rudimentary. Five soft branched rays, and
one spined or stronger one in front of the others, is
the usual number seen in the ventrals of ordinary fishes.
The most remarkable modification in the form of this
fin is seen in the sucking or adhesive fishes, of which
there are two distinct groups, both possessing the power
of adhering, by this member, to other substances, but
very different, not only in their general organisation,
but in the structure of those members by which this
property is performed. One of these is the family of
gobies, Gobiade ; the other, that of the Cyclopterine, or
true suckers: in the first, the two ventral fins are
united, so as to form a circular funnel. Upon what oc-
casions, however, this instrument of adhesion in the
gobies is used, remains at present unknown. Montagu
has observed, that in live gobies which he has captured,
and put into vessels of water, no instance occurred
c 3

a2 CLASSIFICATION OF FISHES.

of the fish adhering to the sides, or to the hand.
This perfectly accords with the result of our own
observations upon a great number procured on the
coasts of Sicily, where this genus abounds. Never-
theless there can be no doubt of these fins being
formed for suction ; and the probability seems to be
that they are only used in stormy weather, when the
_ water is violently agitated. After such commotions of
the sea, we have frequently picked up on the beach
many small fishes seldom seen on other occasions ; but,
although the gobies are nearly all very small, and often
delicate, we never remember to have found a single
specimen cast up upon the beach. The true suckers,
however, forming the genus Cyclopterus, possess the
faculty of adhesion in an extraordinary degree. On
the breast and belly are two circular concave disks:
one of these is formed by the extension of the pectoral
fins under the breast; the other by the union of the
ventral fins. The tenacity with which these fishes
adhere, upon being captured, to the first object which
comes in their way, is very remarkable. Their form
is repulsive ; and they fasten themselves so firmly upon
the hand, that, to inexperienced persons, an involuntary
feeling of dread arises in the mind, lest they should
be venomous. If loosened from the hand and placed
in a vessel of water, they immediately swim with a
quick undulating motion, and affix themselves to the
sides. Several species of these fish occur on the
British coast ; and others, quite different, are not un-
common in the Mediterranean. The most extraordinary
ventral fins are seen in some of the Gymmetes, or riband-
fish, where the rays sometimes resemble oars, being
spatulate or broad at the tip. This form is peculiar to the
genus Gymnetrus ; but in that of Trachypterus (Gouan)
the rays are even still longer, and appear to consist
of slender flexible filaments.

(25.) The porsau fin, with the anal and caudal, are
the three members for progression, of which nothing
analogous can be traced among quadrupeds and birds ;

DORSAL FINS. 93

except, indeed, that the caudal fin represents the tail
feathers of the latter, but not the true tail of the
former, which is an actual continuation of the verte-
bre. The dorsal, after the pectoral, seems to be the
most essential for the aquatic economy of fishes, be-
cause there are only a very few instances yet known
where it is entirely wanting, and all these occur in that
order where the fins gradually disappear, and nature
passes into the marine worms. There seems to have
been a notion that the office of the dorsal was to pre-
serve the fish in a perpendicular position; but some
recent experiments does not sanction this idea, and
there is every reason to suppose that this object is
really effected by the pectorals, which, being placed
symmetrically, one on each side, preserve the body in
equilibrium. Besides, it is quite clear, that if this pur-
pose could only be effected by the dorsal, it would
follow that such fish as the Gymnotus brachiurus and
its allies, where this fin is altogether wanting, could not
swim at all. Yet these are compressed fishes, and,
therefore, obviously intended for a perpendicular posi-
tion ; and they all have pectorals. Dorsal fins will
now be viewed as regards their construction, number,
form, and disposition.

(26.) The construction of the dorsal is so far like
the other fins we have been describing, that it is gene-
rally composed of rays, connected, either partially or
entirely, by a membrane: but then the nature of these
rays varies in the different groups ; and in certain fa-
milies, where there are two dorsal fins, the hinder one
is adipose, that is, resembling a thick fleshy lobe,
attached to the back, and covered by the common skin,
in which neither rays nor membrane can be distinguished.
Fins of this description, with but one exception yet dis-
covered, are confined to the soft-rayed fishes. Native
examples occur in the salmon family: while among the
Siluride, or cat-fish, these fins are almost universal.
The Gadiade, or cods, show us the next advance towards
a more organised construction: the fins, indeed, are

c 4

GA. CLASSIFICATION OF FISHES.

composed of rays; but these rays are so very slender,
and are so thickly covered by fat and skin, that in many
instances they cannot be counted without dissection :
the whole of the Gadiade, in fact, have their dorsals
remarkably thick and fleshy, so that the rays by which
they are supported only become distinct towards their
termination. It is in this family, also, that we find a
modification of this fin unexampled among fishes. In
the rocklings, forming the sub-genus Motella, there is,
before the true dorsal, another, which may be termed —
a rudimentary fin: it is composed of a great number of
extremely fine, short, fleshy filaments, resembling cirri,
preceded by one somewhat longer and thicker than the
rest, but all united at their base by a true membrane:
the peculiarity consists in these filaments having the
form of rays, without the least degree of firmness; for
in other respects these fins are formed in the usual
manner, and are situated in a deep groove.

(27.) The spurious fins, or finlets, as they are sometimes
called, seen in mackerel and other allied genera (fig. 2. d),
may be considered as a modification of the true adipose
dorsals in the corresponding or analogous group of the
salmons, among the soft-rayed families. Theymay be con-
sidered as single detached rays, excessively branched from
their insertion on the back, where they are remarkably
thick and fleshy : like the adipose fins before described,
ythe are always situated behind the first dorsal; but
while no fish has yet been discovered with more than
one adipose fin, those which we are now speaking of
are almost always numerous, varying, among the mack-
erel, from four to seven, and even more. The only two
genera yet known, we believe, where these finlets are
placed near to the head, are those of Polypterus and
Plesiops : in both these, indeed, they supply the place of
the true dorsal fin, the remnant of which, so to speak,
only shows itself in a few connected rays, adjoining and
uniting with the caudal.

(28.) The dorsals, as well as all the other fins, among
the cartilaginous fishes, are so thick, from being covered

DORSAL FINS. Day

with the common skin of the body, that their rays are
completely hid, except in such few as are provided with
an anterior spine, which, being obviously employed as
an offensive weapon, is consequently naked, and par-
tially unattached to the other rays. The fins of the
Pleuronectide, or flat fish, are nearly as thick as those
of the cods; but the rays, being spinous, are more
naked at their extremities. Those lovely fish, the
chetodons, have their dorsal fins remarkably thick, and
so covered with compact scales, nearly to their margins,
that their motion would seem to be very limited.
The great majority of thick-finned fishes are found in
the soft-rayed order (Malacopteryges), while those of
an opposite nature are almost confined to the typical
osseous division, or the Acanthopteryges. The dorsal
fins of the great tribe of perches, together with those
of the Spari, Labri, Triglide, Gymnetes, &c., are
thin ; that is to say, the rays, whether slender or strong,
are not in any way covered by the common skin of the
body, but are bare almost to their base, and united by
a thin membrane, sometimes, indeed, beautifully coloured
and opaque, as in Serranus, Perca, Labrus, &c., but
generally sub-transparent, and almost colourless, as in
the whole of the Sparide, Scomberide, Zeide, &c.
(29.) The number of the dorsal fins is variable ;
for although they are all placed upon the same line,
which is invariably the ridge or summit of the back,
they are yet separated, more or less, into divisions; and
these, when perfectly detached one from the other, are
viewed in the light of separate fins, although, strictly
speaking, they should simply be considered as so many
divisions of a single one. Where the intervals are marked
by a secession of a connecting membrane between the
rays, there is no difficulty in determining whether, ac-
cording to the common mode of reckoning, a fish has
two or three dorsal fins: but it frequently happens,
even in the same genus, that in one species the mem-
brane of the last ray of the first dorsal terminates or
adheres to the back ; while, perhaps, in the very next it

26 CLASSIFICATION OF FISHES.

is attached to the base of the anterior ray of the second
dorsal ; so that, in effect, although there are two divi- |
sions, or fins, yet they are, in fact, connected, and, con-
sequently, become one. The genus Gadus, as now
restricted, has obviously three dorsal fins, each separated
by an interval from each other ( fig. 2. a): while in the
genus Blepsias there are also three (0), nearly of the same
form ; and yet, because they are very slightly united by
a membrane, in the manner above described, this genus
is said to have but one dorsal. Now the transition
from two approximating dorsal fins to one, cleft be-
tween the spiny and branched rays, is so gradual, that
it is impossible to define every stage of the progression ;
and much ambiguity will always attend this part of the
definition of the subordinate groups. Nevertheless, as
characters taken from the fins will be eventually found
to be of much more real and practical value than has
hitherto been supposed, we should propose the adoption
of the following terms, as calculated, in some degree,
to express the modifications just mentioned. Where,
for instance, a portion, however small, of the naked
back intervenes between one or more of these divisions,
as in the common cod (a), they may be considered, as at
present, three distinct fins. When the last membrane
of the first dorsal is in any way united to the anterior
ray of the second dorsal, we might consider them as
“two dorsal fins united ;” and when this union is so
close, as that the membrane in question ascends up
the side of the next ray, instead of descending in a
direction to the base, we would then term the dorsal fin
‘single, but emarginate,” deeply, or slightly, as the
case may be. The annexed cuts will more effectually
explain our meaning ; and we may now consider their
other peculiarities.

(30.) The rays of the dorsal are either simple or
branched. Simple rays, again, are of two sorts:
sometimes they are slender and flexible, although with-
out any joints; in which case they are generally ter-
minated by a fleshy or membranaceous filament; and

DORSAL FINS. Bag

9 these are either
isolated, as in
Dactylopterus, or
united to the dor-
sal fin, as in most
fishes. The ge-
neral _ character,
however, of simple
rays is that of
being strong, ri-
gid, and so sharp
as to become spi-
nous: that these
spines are used as
instruments of de-
fence, becomes evi-
dent from the fact of many fishes suddenly raising them
when captured, so as to inflict wounds on the hand
of an incautious person ; and that they also are essential
to the perfect use and efficiency of the fin itself, by
strengthening and supporting the other rays, is also to
be inferred from this fact, —that in all soft-rayed fishes
the first ray of the dorsal, if not simple, as in the
carps, &c., is almost invariably stronger than the others,
—a structure intended to break the resistance of the
water during the swimming of the fish, on the very
same principle that a boat or vessel is furnished with ©
astem. Fishes which swim but little, and in calm
waters, like the eels and a few others, do not possess
this peculiarity ; but in those which belong to the
most perfect division of the osseous fishes (the order
Acanthopteryges), the development of the spiny rays is
at its maximum, and constitute the primary distinction,
even by the confession of Cuvier, of this most natural
group. Sometimes these spines are detached and iso-
lated, when they are always short, and repose in a
groove on the back (as in Naucrates, &c.) ; in which
case, however efficacious they may be for defence, they
can be of no use in swimming. When these spines

fi
ay A AGILE
Ls Y/ Y >
Whe, YS by f2, Z
Yi fff V4
Wi, y
y UY,

Y,
Vd,

Y,

98 CLASSIFICATION OF FISHES.

are so very short as just to appear above the scales, they
have no membrane ; but if longer, a slight one connects
each of them to the back, but not to each other: in
general the point is directed backwards; but in some
few genera, allied to the mackerel, some of these prickles
are directed forwards, and others terminate in two or
three spines, or are bifid or trifid. The most remark-
able instance of these dorsal spined fishes is the genus
Acanthonotus, where there is a row of ten of them, de-
tached, placed both before the dorsal and the anal fins:
more familiar examples are seen in our sticklebacks
(Gasterosteus Linn.), of which the G. spinochia Linn.
has no less than fifteen before the dorsal. The spines in
the first dorsal fins of the acanthopterygeous fishes
are almost always graduated; the first being short,
while the second is intermediate between that and the
third ; which latter (or the fourth) is usually the
longest: in particular groups, however, there is always
some prevalent modification of this fin, which we shall
now notice.

(31.) The shape or form of the dorsals is consider-
ably varied: where there are two or three, those which are
in front are almost always triangular, while the hinder
one is of more equal breadth throughout. In the common
cod (fig. 2. a), the first is acutely triangular, the two
next less so; but in Blepsias, its representative among
the Canthileptes, the posterior of the three connected
fins is broadest in the middle (fig. 2. 6). In Trachinus
and its numerous representatives, the first dorsal is short
and triangular, while the second is long and narrow (c).
In the mackerel family, however, where all the fins are
subfalcated, both the dorsals are consequently of the
same form ; but this comparatively is a very unusual
structure, although it affords an absolute character to
the Scomberide (d). In the sharks, the mullets, and a
few others, where the two dorsals are wide apart, both
of them are triangular. Nearly all the typical Gymnetes
have the dorsal fin highly developed ; it is here also
sometimes particularly broad, with the anterior rays

ANAL FIN. 29

often excessively prolonged, and ending in spatulate or
thread-like filaments. This sudden elongation of the
first two or three rays we shall term falcate ; and it is
particularly observable that this shape occurs in nearly
all genera which represent the tribe of Gymmnetes in
their own circle. Nevertheless, the secondary modi-
fications of this fin are so numerous, that to describe
them all in this place would be tedious and unne-
cessary. Among the eels, the dorsal is always simple
and undivided, narrow, and of equal breadth throughout ;
and this occurs in almost all the representatives of the
apodal order, as Lepidotus, Ammodytes, Cepola, and Ophi-
dium, among the Gymnetes; Blennius, Anarhichas,
&e. in the Gobiade ; Chimera, in the cartilaginous
order ; and Ophiocephalus, among the Macroleptes. In
most of these the dorsal fin unites with the caudal,
as in the eels and other Wurenide ; while in the Blen-
nide, or blennies, there is a small interval between them.
Lastly, we may notice the long fleshy filaments which
in some few genera surmount the spinous rays of the
dorsal fin, and produce a very singular appearance.
These appendages are mostly found among the Zeide,
or sun-fish, of which the common dory of our coasts is
a striking example.

(32.) The ana fin may be termed symmetrical to
the dorsal ; or, at least, its situation on the under part of
the tail is analogous to that of the dorsal on the back.
It must be observed, however, that this fin is always
placed behind the vent, so that the length of the tail,
in many cases, is indicated by the length of the anal
fin. It is subject to very little variation in form, and
still less in construction, for it generally corresponds —
with the hinder part of the dorsal: it is almost always
nearly the same breadth throughout, and without any
particular variation in other respects; all the rays,
except the two or three first, which are more or less
spinous in the most perfect families, are articulated and
branched. The anal fin is most developed in the apodal
order and its representatives, where we have it some-

30 CLASSIFICATION CF FISHES,

times, as in Gymnotus and Chimera, extending nearly
the whole length of the fish. This is very observable
in the sub-family Si/urine, which also represents the
Apodes ; and we again trace a similar development of
the anal fin among the Blennide and the genera Cepola,
Ophidium, &c.

(33.) The caupat fin alone remains to be noticed.
It is, to us, a most unaccountable circumstance, that
every naturalist who has hitherto written upon Ichthyo-
logy, should have followed each other in paying so little
regard to the fins in general, but more especially to this,
which (with the tail itself) is as important to the
motion of a fish as the rudder is to a ship, or as is the
‘tail of a swallow in directing its flight. That such is
the true office of the tail and its fin, among fishes, is
too obvious to require being enforced by argument ; and
yet, while the importance of this member is so fully
- acknowledged in quadrupeds and birds, that it often
furnishes the only decisive generic character, it has
hardly ever been considered in this light in ichthyology ;
and not only whole groups of species, but even of sub-
genera, have of late years been described, where the
tail is hardly ever mentioned, or, if so, only inci-
dentally. Our own impressions on this subject, after a
long and laborious investigation, induce us to consider
that, in a natural arrangement of this class, the form
of the caudal fin is just as important in fishes as that
of the tail in birds; and that it is, consequently, one of
the best characters for the determination of natural
groups or types that can possibly be found. We
view it, in fact, as much more determinate than those
slight modifications of the teeth, upon which so many
of the modern sub-genera have been founded, to the
infinite perplexity of all but the professed anatomist;
and, what is worse, to the cutting up and frittering
away, as we conceive, of natural alliances, subordinate,
in different degrees, to each other. The experienced
ichthyologist, well acquainted with the variation of this
member, will not fail to observe that the swiftest

CAUDAL FIN. 31

swimming fish are all distinguished by a tail more or
less forked ; and that the most sluggish are invariably
characterised by a rounded tail. Now this is precisely
what we find in ornithology, where no instance is upon
record of a rounded tail and wings being given to swift
flying birds, or the reverse. Were we asked to name,
from our own experience, that family of fish whose
swimming was most rapid, we should hesitate between
the fiying fish and their enemies, the different species
of tunnies, by which they are so frequently pursued :
the latter, indeed, would seem to have the superiority,
since they frequently overtake the other, upon which
they are known to feed ; but this superiority lies more,
we apprehend, in their greater size and muscular
strength than in the absolute power of swimming: it
is clear, in fact, that if the strength of the flying fish
did not fail after a long chase, the bonatos or tunnies
could not overtake them, any more than the dog could
outstrip the hare. In both instances the superiority of
speed lies with the pursued, while that of muscular
strength is with the pursuer ; thence the latter qualifi-
cation, in the end, triumphs over the former. Now the
whole of the Scomberide, or tunny family, have the
tail more deeply forked than any other fishes, perhaps,
in the entire class ; for not only are the two lobes deeply
cleft, but in most instances they are actually divided ;
and they are further provided with two additional finlets
on each side, by which the rapidity of motion is
doubtless accelerated: this is further increased, in many
groups of this family, by a prominent fleshy keel
which projects on each side, near the base of the caudal
fin, and parallel to the lateral line: these ridges are ob-
viously intended to cut the water on each side, and they
are only found among those families we have arranged in
the tribe of Macroleptes. Forked tails are only found
among the two great divisions of osseous fishes, and
a few of their representatives ; for those of the sharks,
when they approach this form, are more properly
lobed or emarginate in the middle, the lobes themselves

ere CLASSIFICATION OF FISHES.

being rounded. The Balistide, the most perfect of
the cheloniform fishes, and which represent the spine-
rayed order (Acanthopteryges), consequently present
us with nearly all the modifications of fin observable
among their prototypes ; but in the whole of the remain-
ing families, and the entire order Apodes, where we have
the most sluggish of all fishes, as the Chironectide, the
Lophide, Cyclopteride, Murenida, &c., not a solitary
example occurs of a forked tail, much less of those ad-
ditional helps for speed just noticed, which have been
given to the Scomberide. If we carry our inquiry
into the minor groups or families, we shall find the
same determinate prevalence of one set of characters in
the fins, running through each particular group. Nu-
merous instances of this will be brought before the
reader, for the first time, in the progress of our work.
Having now adduced sufficient reasons, as we imagine,
for the opinions above expressed, we may at once pro-
ceed to notice the different forms observable in this fin.

(34.) The caudal fin presents every modification
between a perfectly lanceolate shape, where the largest
rays are in the centre (fig. 3. a), to that of a deeply
forked one, where
the central rays
are so short as
almost to be-
come obsolete,
giving the tail
an appearance of
being _—_ divided
SS = «36s Into.s« two parts.
LALLA, "The first of these
forms is shown in the genus Cepola, and its repre-
sentatives the Indian gobies; the second runs through
the whole of the mackerel, tunny, sword-fish, and a large
proportion of the Zeide, or dories. Besides these, there
is also a third, peculiar only to two or three genera,
where the tail may be said to be doubly forked; a few
of the central rays being lengthened nearly as much as

d

CAUDAL FIN. 33

the external ones, so that they form a lobe on each of
their sides — one above, the other below. Where such ~
numerous gradations occur, it is impossible to define with
strictness the limits of our definitions ; we may, never-
theless, arrive at some degree of precision, by con-
sidering each of these forms as presenting the following
modifications : — A rounded fin is either lanceolate, oval
round, even, or truncate. On reaching this latter, we
may draw an imaginary line, and enter upon the fork-
tailed division. Truncate fins pass into those which
are slightly crescent-shaped ; they next become lunate,
forked, lobed, and, finally, emarginate; while these latter,
again, pass into rounded fins: the highest development
of each of these is seen in the lanceolate and the forked ;
the other modifications gradually recede from each other.

(35.) We shall now endeavour to define each of these
forms.—1. Lanceolate,so named from the shape bearing a
resemblance to the head of a lance: the longest ray is in
the centre, and stands singly; all the others are in pairs,
diminishing, more or less gradually, in length, until
those that are external become the shortest. Nearly
all the species of Cepola (fig. 3. a) possess this character,
but it is by no means frequent, and is chiefly seen
in that genus, Gobius, and in Sciena pama (Cuv.
pl. 101.).—2. Oblong oval: not quite so long in pro-
portion as the last; the middle is not pointed, and
the shape is that of the smaller end of an egg.
This form may be called a highly developed state of
the next, and is confined to few examples. — 3. Round :
the fin is of moderate size, always shorter than the last,
and the extremity describes the segment of a circle.
This is the most common shape in this division, and
pervades all the flat fish (Plewronectide), a few of the
rocklings (Motella Cuv.), all the gobies (Gobiade), the
genus Syngnathus, &c., none of which are capable of
long sustained swimming: the degree of roundness
varies ; but we still retain the name to all such fins as
have the central rays in any degree longer than those
on their sides. — 4. Even: the majority of the rays are

VOL. I, D

34: CLASSIFICATION OF FISHES.

of equal length, the central ones not exceeding the
others, while the outermost ones are rounded at their
angles only. Many of Cuvier’s genus Servanus have
this fin, also nearly all the sticklebacks (Gasterosteus),
Sciena aquila (Cuv. pl. 100.), Blepsias (1b. 90.), Ura-
noscopus, Priacanthus, Hemitripterus, &c. — 5. Trun-
cate: when the extremity of the fin appears to be
abruptly cut off, so that the external rays are just as
long as those in the middle, and the angles are not
rounded, as in the last. Zeus and Tirachinus may be
cited as the most familiar examples of this form, which
is only distinguished from the next by the marginal ex-
tremity of the fin being in no degree concave or cres-
cent-shaped, or, in other words, not having the central
rays shorter than the external: it must be observed,
however, that fins of this description can only be de-
tected when extended ; for when closed, the margin
generally has the appearance of being slightly concave.

(36.) Forked caudal fins are as much, and even
more, varied than the last. The incipient develop-
ment of this structure is seen in such as have the mar-
ginal extremity slightly concave, as in the majority of
the Triglide, or gurnards, the angles being pointed, and
the interval between them slightly hollowed out, so that
the central rays are shorter than the external ones.
Trachinus radiatus, according to Cuvier (pl. 61.), has
a concave fin, although in the common species of the
Mediterranean it is completely truncate. This is a very
prevalent form, and several examples occur in the sub-
family of the Scienine, as Leiostomus, some Corvine,
&c. — The Junate shape is on the same principle
as the last, but the concavity of the margin is much
deeper, and the two extremities are prolonged, often
(as in Naseus, some of the sub-genera of Acanthurus,
&c.) to an excessive length, in the shape of filaments.
Forked caudals, properly so called, are of two kinds:
in one, the divisions are equal (fig. 4.) ; in the other,
unequal (c). The most typical of the first form, as be-
fore intimated, is universal among the Scomberide, or

CAUDAL FIN. 35

mackerel, where the middle of the fin is cleft to its
base, or very nearly so; and each division is falcate, as
in the tunnies, or somewhat lanceolate, as in the common
mackerel. This form, so prevalent among the genera of
the Microleptes (or that tribe which includes the whole
of the Scomberide, Zeide, &c.), hardly exists in the
pre-eminently typical tribe of Macroleptes. ‘These latter
fishes, on the contrary, have a simply forked caudal ;
that is, the lobes are not attenuated, and the central rays
are nearly equal to half the length of the external ones.
This structure is the most general in fork-tailed fishes,
and is generally constant in natural groups, of which
the Sparine, the true perches, and several others, afford
ample proofs. The most extraordinary development of
a simply forked tail, yet discovered, is to be found in
the Macropodus venustus (Cuv. pl. 197.), where the
length of this fin is nearly equal to that of the body:
and this is the more remarkable, since, in no other genera
of its own circle, is the caudal of this form ; a clear
indication that it is the rasorial sub-genus. Sometimes,
as in Womeus and Hoplostethus, the caudal, although
deeply cleft, has the two divisions rounded ; but this
form is very uncommon. — Unequally forked, is when
one of the divisions of
the fin is larger than
the other: our English
sand-lance (Ammody-
tes) shows this very
well (fig. 4. a) ; and it
is likewise found in all
the flying fish (6), and
the greatest part of the
sharks and sturgeons.
The caudal fins, how-
ever, of these latter
families are altogether
- peculiar: the rays are
by no means symmetri-
cal, so that the upper

36 CLASSIFICATION OF FISHES.

lobe of the tail is not formed, as in ordinary fishes,
by rays, but by the terminal vertebre, round which
the fin is short; while the other, or lower lobe, often
irregular, is alone composed of rays. No other fishes,
yet discovered, possess this sort of caudal fin, nor is
there any thing analogous to it among the osseous or
semi-cartilaginous orders. Another modification of the
forked structure occurs in a very few genera, where
there are two divisions, or rather sinuosities, in the ter-
minal margin, analogous to the double fork seen in one
or two birds of the Caprimulgide. Finally, this struc-
ture blends into the rounded form in such fish as have
the even tail already described, but with the middle rays
very slightly shorter than the outer ; so that the margin
becomes widely notched, or sinuated, as seen in several
of the salmon family, and many others.

(37.) In some genera the caudal fin is either indis-
tinet or obsolete. The first appellation may be given,
when the fin is so united to the dorsal and ventral that
there is no perceptible difference between the rays of
either: such is the case in the greater number of the eels
and congers, in the genus Ophidium, in certain silures,
or cat-fish (Silurid@), and in several other anguilliform
types: in some these three fins form an acute point, as
in Ophidium, Synbranchus, &c.; or a rounded one, as in
the lampreys (Petromyzon), and many of the soles, and
other Pleuronectide. The caudal fin may be also termed
obsolete in most of the Raid@e, where it either assumes
the form of one or two small lobes, or of merely a long
narrow membrane bordering the lower extremity of
the pointed filiform tail so common in this family.
The caudal, however, is completely obsolete in such ge-
nera as Tvichiurus; for in them the body terminates in
a long slender process resembling a filament. In Tri-
chiurus, Chimera, and some Syngnathi, the tail is desti-
tute of either a terminal or lateral fin; and the same is
observed in many of the sting rays (Zvrygline): but
in Gymnotus the under part of the tail is margined by
a continuation of the anal fin which reaches to the

GENERAL OBSERVATIONS ON FINS. OW

point. One or two extraordinary departures from the
ordinary form of this fin may here be noticed, but they
are mostly confined to single genera. In T'rachypterus,
the fin, although large and truncate, is mounted vertically
upon the point of the tail, so as to form an angle with
the line of the body. This structure is altogether unique
among fishes ; for it does not exist in the neighbouring
genera Argycthis Sw. and Nemotherus Raf., whose tails
are situated as in ordinary fishes. The other modifi-
cation belongs to Cuvier’s Serranus pheton (fig. 4. e):
the tail is forked; but from the centre or deepest part
of the cleft springs a long filamentous ray, near three
times the length of the fin itself, —a structure of which,
as yet, we know of no parallel.

(38.) Having now brought before the reader (what
has never hitherto been done) an enumeration of nearly
all the different forms observable in the fins of fishes,
we shall conclude this part of our subject with an
attempt to generalise, in some degree, the facts thus
brought together, in order to show that the results thus
obtained will correspond in some remarkable points
with the locomotive organs of birds. In the first place,
it must be remembered that these organs are more nu-
merous in fishes than in any other vertebrated animals .
this is the necessary consequence of their being the
fissirostral or aquatic type of the vertebrated circle ;
which type, as we formerly explained, invariably possesses,
in this circle, the greatest powers of motion. The
ornithologist is quite aware of this; but it may be as well
to inform the ichthyologist, who may not have studied
that branch of zoology, that the swallow, goatsucker,
tern, albatross, and kite,—the swiftest flying birds that
are known to exist,—are all of them of the fissirostral
structure, whether by affinity or analogy: and thus do
we find this law pervading the class before us, —a class
which may be said to be in perpetual motion; for
although a quadruped can lay down to repose, and a
bird can roost on its legs, it seems difficult to imagine

dp $

38 ' GLASSIFICATION OF FISHES.

how a fish can rest without any motion of its fins*,—
more especially those which habitually live in the open sea.

(39.) We have already shown in what manner the
Fins of fishes represent the organs of motion in birds.
Now, in all these latter, superior powers of flight are
invariably indicated by the great length and pointed
structure of the wing; and this power among birds is at
its maximum when the tail, also, is forked: the common
house swallow shows this in perfection. Now this is pre-
cisely analogous to what we see among fishes: all those
with pointed pectorals swim much faster than those which
have this fin rounded; an inference which does not merely
rest on analogical reasoning, but from the remarkable
fact, that the far greater majority of those fish which
have pointed pectorals habitually live in the open ocean,
or far from the shelter of the shore. We know not,
at this moment, of any freshwater genus, wherein the
pectorals fins are decidedly pointed; while, if we look to
the oceanic families of the Zeide and Scomberide, and
even the majority of the Percide and Chetodonide, we
shall find very few instances of the pectorals being
rounded. But if, in addition to this pointed form, a fish
has the tail also deeply forked, and the pectoral fins
faleated or curved, as are the wings of the goatsuckers
and humming birds, then we have the highest develop-
ment of the powers of swimming possessed by this class.
Hence it is that the mackerel, the sword-fish, and the
tunnies—more especially the latter—are, together with
the flying fish, the most perfect of all swimmers. Every
one who has seen the astonishing rapidity with which
the tunnies will sometimes play about a vessel in the
Atlantic Ocean, when sailing at its utmost speed, will be
perfectly convinced of this: for although their rapidity,
for a time, may not equal that of the rays, it is quite
evident that they have a vast superiority over the latter
in their adaptation for sustaining swimming; the rays,
indeed, being obviously ground-fish, or of those fami-
lies which seek their prey at the bottom of the sea. The

* Except the flat fish, which, of course, lie on the ground.

ANALOGY OF FINS TO WINGS. 39

tunnies, of which the bonitos and albicores of seamen
are only different species, will sometimes, in a stiff gale,
play about a vessel in full sail, with as much ease as if
she was perfectly still—one moment they will be near
the stern, while the next, as if by a single dart, they are
many yards ahead of the bowsprit: this we have re-
peatedly witnessed; and the thought then struck us that
no fish, by any possibility, could move more rapidly. In
comparing, therefore, the functions of the pectoral and
caudal fins of fishes to the wings and tails of birds, we
find they are perfectly analogous, and that their import-
ance, as furnishing generic characters, is equally great.
(40.) It is somewhat remarkable, that although many
instances occur among swift-flying birds where the
wings are pointed and the tail rounded, yet in the class
of fishes, the shape of the pectoral and the caudal fins are
almost always symmetrical ; that is to say, the caudal is
forked in the same proportion as the pectorals are
pointed ; nor does an instance at this moment occur to
us where the pectoral is pointed and the caudal rounded,
or the contrary : hence we may infer that the caudal
fin in fishes is more important in its offices than is the
tail in birds, and this is an additional argument in
favour of the importance we attach to this member.
(41.) A comprehensive view of the coincidences in
the formation of the dorsal and ventral fins in genera
widely distinct in affinity from each other, will lead
the philosophic naturalist to suspect that these may
offer one of the best clues for determining the ana-
logical relations of widely separated groups. This
intricate subject has claimed much of our attention ;
and although, from its nature, we have been obliged to
leave it unfinished, the progress we have made seems
to sanction the following observations: —It would ap-
pear that in every one of the tribes composing the two
erders of osseous fishes (the Acanthopteryges and the
Malacopteryges), thetwo chief divisions are characterised,
the one by having the dorsal fin single, while in the
latter it is double, or at least deeply cleft: in another
D 4

40 CLASSIFICATION OF FISHES.

division the pectorals are larger than usual, and the
ventrals often remarkably developed : in a fourth, these
latter fins are quite the reverse; they are either small,
very imperfect, or altogether wanting, while the dorsal is
long and often very broad: finally, there is a fifth form
where the first dorsal is short and triangular, and the
second long and narrow, as in those two well-known
genera T'rachinus and Uranoscopus. Numerous ex-
amples of the prevalence of these forms, following each
other in a natural series of affinity, may be traced in
the synoptical definitions of the arrangement we have
made of this class; and although the preceding remarks
are more particularly drawn from the two typical orders,
instances are not wanting to show the same tendency
in the more incomplete or aberrant orders.

(42.) We may here explain the terms by which we
propose to designate the different forms of the fins, and of
their relative situation. The true length of a fin should
probably be reckoned from the base to the tip of its
rays, while its breadth would be estimated by the ho-
rizontal space it occupied between one extremity of
the body and the other ; but this terminology, however
abstractedly just, would be in complete opposition to
the terms we apply to the figure of the fish itself, and
might lead therefore to some perplexity. We may
take the eel as an example: we should say that this
fish is very long, and justly so ; but although its dorsal
fin extends to near its entire length, we must describe
this fin, in accordance with the foregoing rule, as very
short, because the length of the rays (not the fin itself)
is really so; while, by the same rule, we must term the
body very narrow, and, the dorsal fin very broad. To
common apprehension, these terms would seem to con-
tradict each other: and, in truth, the subject is beset
with some difficulty. It seems to us, however, that by
looking to the fin itself, instead of its rays, we may
get a greater uniformity of terms than by any other
rule. Thus, we should describe the dorsal fin of the
eel as very long, but very narrow or low ; and that of the

TERMS APPLIED TO FINS. 41

Pteracles trichipterus Cuv., as very long and remarkably
broad.

(34.) The number of rays of which the several fins
are composed, affords one of the best characters for
specific distinction ; for although it has been said that
they vary in individuals of the same species, we must
confess that this opinion has not been verified by our
own observations, — and they have neither been few, nor
partial, nor taken from preserved specimens. We are
more disposed to believe that such differences are more
apparent than real: first, because in many instances,
when the rays are very small and close together, or
very numerous, we have found it almost impossible to
attain perfect accuracy in this respect, except by re-
peating the examination several times, even on fresh
specimens ; and secondly, because the fins of many of
the ground-feeding families are so thick and fleshy that
the number of rays cannot be distinctly counted. The
eels, the Gadiade, the Siluride, and several other
thick-finned families, are familiar instances of this; but
very few will be found among the spine-rayed groups,
where, from the membrane being thin, the rays of the
dorsals, ventrals, anals, and even the pectorals, may be
numbered with accuracy.

(44.) The external covering of the erius, and more
especially their aperture, are of great importance, and
require to be further noticed. The use of the gill-cover,
or operculum, is obviously to protect the gills themselves,
and, at the same time, to admit the egress of the water
taken in by the mouth: in the majority of typical
fishes it is moveable ; for, although composed of bony
plates, these plates are articulated or jointed at their
sutures by a membranaceous skin which acts as a hinge.
Properly speaking, the operculum consists only of three
pieces, which are attached to the cheek-bone, called the
pre-operculum ( fig. 5. a): of these three plates, the upper
is more especially termed the operculum, and it is al-
ways the largest ; the next is the sub-operculum (b) ;
and the third, which is very small and sub-triangular,

49 CLASSIFICATION OF FISHES.

is the inter-operculum (c), because it fills up the interval
between the base of the pre-operculum and the sub-
operculum. These distinctions will be rendered more
intelligible by the annexed figure, which represents the
head of the common tench.
These bones are either
smooth both on their outer
surface and their edges, or
they are armed, in one or
both situations, either with
spines or prickles, or very
fine dentations like the
teeth of a saw; and these modifications are of great
importance in determining generic groups. The hinder
margin of the entire operculum, or gill-cover, is gene-
rally bordered by a thin membranaceous skin, for the
purpose of closing the opening of the gills more ef-
fectually: this skin is a continuation of that which
supports the branchial rays, and these latter commence
at the bottom of the head, adjoining the throat; and the
number of these rays are considered indicative of ge-
neric peculiarities. In many groups which possess gill-
covers, the plates are either immoveable, or are so com-
pletely concealed under the skin that they are not to
be detected except by dissection. In such instances,
the aperture becomes so small as to be analogous to the
spiracles of the cartilaginous tribe; it assumes, in
fact, the appearance of a slit, and is then termed a
spiracle. This character pervades the whole of the
aberrant tribes of our present arrangement, and even
extends to such osseous fishes as represent them in
their own circles. It seems to be a general law, that
those fishes which have the gills highly developed, and
the aperture very large, like the herring and mackerel,
very soon die on being taken from the water ; while
those, on the contrary, as the eel, which breathe by spi-
racles, live for a considerable time on being exposed
to atmospheric air. It is among such that we find

TEETH OF FISHES. 43

those crawling species before alluded to, which volun-
tarily quit the water in search of new habitations.
(45.) The reer of fishes, as before remarked, are
varied in the most surprising manner in regard to their
situation, but less so in their construction ; and these
instruments are far more numerous in this than in any
other class of animals. They are not confined, like those
of quadrupeds and reptiles, to the two jaws, but are often
disposed in all parts of the mouth. The maxillary
teeth are those which are most external, and are placed
on the jaws, properly so called, in quadrupeds; and they
correspond to the cutting edges of the two mandibles
in birds. Parallel to the upper jaws, internally, are
the palatine bones, which often support other teeth:
between these bones, and in the centre of the palate, is
the vomer,—a name given to that bone which forms the
roof of the mouth; and this also is frequently armed
with teeth, even when the jaws and lateral palatine
bones are completely smooth, as in the instance of the
common carp, tench, &c. : the tongue, also, is sometimes
armed with other teeth, as in the pike, &c.: sometimes
all these are so thick and numerous, that they seem like
a dense forest of teeth, capable of crushing the most
minute substance. To describe the different forms of
these teeth would be almost impossible: they are in ge-
neral more or less pointed; in the herbivorous fishes
they are formed for the purpose of pressing; and in
such as feed upon testaceous animals, they are so much
rounded as to be analogous to the molar teeth of quad-
rupeds. In the Siluride, and other genera, they are so
delicate and flexible as to resemble the pile of velvet:
hence we may term such teeth setaceous. In the sharks
they are compressed, and serrated on their sides; while
in the Rays they are round, and placed in the manner of
paving stones or mosaic: such teeth are therefore termed
tessellated. In many of the genera (as Laurida) they
are moveable at their base in an inward direction, to
admit a free passage to what is swallowed. Among the
cheloniform fishes, the absence of true teeth is supplied,

4A CLASSIFICATION OF FISHES.

as in their prototypes, by the sharpness of the jaw-
bones, which are remarkably strong: the analogy, how-
ever, here ceases ; for the jaws of the true chelonian
reptiles are entire; whereas those of the Balistide are
divided in such a manner that they wear the appear-
ance of being like the true and distinct teeth, placed in
a single row, of ordinary fishes. Finally, we perceive
sub-genera, and even species, as they are now classed,
scattered among the greatest number of these toothed
races, which have the jaws entirely smooth; a clear proof,
if any other were wanting, that genera built entirely on
these organs are more likely to be artificial than natural;
indeed, we have only to look to those among the S?-
luride, as they stand in the most recent systems, for a
justification of this opinion. It frequently happens
that in natural groups, like the last, the teeth offer no
variation of the least importance ; while in others they
are scarcely the same in two species, and vary in the
most remarkable manner, even in the same fish, at
different stages of its growth. This is particularly ob-
servable among the salmons, and even in the family of
sharks. The value of a zoological character is well
known to be proportioned according to its prevalence
in groups or individuals, which, in every other cha-
racter they possess, show a clear and unquestionable
affinity. Thus the prevalence of the spiny or of soft
rays in the osseous fishes indicate, with other peculiar-
ities, the two great typical divisions; and thus, from
its prevalence among families and genera, clearly re-
lated, we infer its primary value. But when, in
another group, we observe the teeth vary in almost
every third or fourth species, although their charac-
ters in other respects are precisely the same, it is
quite clear that we must look for some other marks
of discrimination, possessed by ail these individuals,
whereby to preserve in our systems that bond of union
which we see in nature. Among the sturgeons, for
instance, we have some species with teeth, and others
without ; yet there is no other difference. Still more

VALUE OF TEETH AS CHARACTERS. 45

remarkable are the variations in the teeth of M. Cuvier’s
genera Pimelodus and Bagris: he himself observes this;
and yet these two groups are attempted to be charac-
terised by their teeth alone. Numerous other instances
might be named ; so that the only conclusion we can
come to is, that as no organs vary so much among
fishes as the teeth, so do they offer the most uncertain
characters, when taken by themselves, for designating
natural groups. For these reasons, we consider such
characters inferior to those drawn from the fins, the
gills, the eyes, the body, and the scales.

(46.) The taTERAL Line, where it exists, as in the
more typical groups, deserves much attention: the
scales of which it is formed are always of a peculiar
construction, — being perforated in the middle for
the free issue of that mucous substance which is so
prevalent among fish, and which is secreted in certain
glands beneath: these scales are generally of a different
shape from those of the body; and they have been re-
cently employed by our best ichthyologists as additional
aids for discriminating species, which otherwise bear a
close resemblance. Sometimes, as in the family of
Scomberide, the scales of the lateral line are raised and
carinated, so as to present a prominent edge like that
of the sharp ridge of a triangle ; while in others they
assume the form of spines or prickles: then, as to the
direction, it is either straight, arched, broken, or sinu-
ated. In some of the Indian Silurid@ it is double ; and
in many genera it cannot be distinguished.

(47.) Many of the soft-finned fish are provided with
cirri, or barbels, placed round the mouth: these are
soft fleshy processes, and are supposed, with every ap-
pearance of reason, to be employed both as organs of
touch, and also of allurement to their prey. We con-
cur with Mr. Yarrell in believing that all cirrated fish
are ground-feeders, that is, seeking their food close to
the bottom. We may alsc remark, that such genera as
have these appendages very highly developed, as in
nearly all the Si/uride, or cat-fish, they are employed

46 CLASSIFICATION OF FISHES.

to attract others, upon which they prey. The cat-fish,
safely screened from observation in the natural hollows
or holes of the bank, throws out his long cirri, which,
being flexible, may well be.taken for worms by other
smaller fish, which are thus brought within reach of
their true owner. Among the cod-fish they are much
shorter ; and in the tench family they are very slightly
developed. We find them, in a slight degree, in the
cartilaginous genera of Acipenser, Squatina, and Cros-
sorhinus ; but in this latter they assume the form of
short flat processes, so that they may here perform a
different office. Perfectly analogous to these cirri is
the long appendage rising from the nose of the frog-
fishes, composing the genus Lophius of Linneus. It
is probable that these amphibious-looking creatures are
the most imperfect swimmers in the whole class; and
being carnivorous, this inaptitude for pursuing their
prey is made up to them bya very long filament, rising
from the head, and terminated by a flat spoon shaped
enlargement, so as to bear a ludicrous resemblance to a
fishing-line with a bait at the end: the fish lurks in
its hole, and throws out this natural line, and thus at-
tracts its prey. Its vulgar name of fishing-frog is,
therefore, peculiarly expressive ; for it not only angles,
but it is of that type which represents the amphibious
frogs among the aberrant fishes. Having now laid
before the reader the chief characters of structure by
which the different tribes, families, and genera of fish
are distinguished, we may briefly touch upon the senses
they seem to possess, and then enumerate some of the
most interesting points of their natural history or
economy.

(48.) The senses of fishes are much less developed
than those of quadrupeds or birds. Some of these
faculties have been already incidentally mentioned, to
which it is only necessary —in such a rapid view as we
are now taking—to add the following : — The sense of
touch is very “partially developed, for it is difficult to
understand how it is possessed by those families which

SENSES OF FISHES. 47

are destitute of cirri: the lips, indeed, of some few
genera are thick and fleshy ; but analogy would lead us
to believe that this peculiarity had a greater reference
to taste than touch. It seems, however, that nature
compensates her partial denial of this sense by increas-
ing that of sight. The eyes of nearly all the spiny- —
rayed fishes, very few of which are provided with cirri,
are particularly large ; and this circumstance alone would
lead to the conclusion that the faculty of sight is highly
developed in such groups. It may be observed, on the
other hand, that nearly all the soft-rayed genera, that
are provided with cirri, have the eye comparatively
very small: and such is also the case in most of
the ground-fish ; witness the eels, the flat fish, the
sharks, skates, and lophians.* The mackerel, the her-
ring, the Spari, and the dolphins, which are pelagic, or
roaming for the most part in the wide sea, have all
large and brilliant eyes; while a few others, which there
is reason to believe live almost entirely in the profound
depths of the ocean, have eyes even still larger than the
last. This brings us to the sense of smelling, which
there is equal reason to believe is very great; for the
nostrils generally have a double opening on each side,
although both lead to the same canal; while the internal
nerves connected with the nostrils are very large, and
occupy a considerable space.

(49.) Fish are exposed, on all sides, to the approach
of enemies, from whom there is rarely that facility of
shelter afforded in the open sea which is enjoyed
by land animals. A highly developed state, there-
fore, of the organs of sight and smell appears abso-
lutely necessary to them, not only for their own safety,
but also to discover the food, whether animal or ve-
getable, upon which they subsist; with these qualities
the faculty of touch is hardly required, and we conse-
quently find it either very partially or, to appearance,
not at all given. Inductive reasoning, again, teaches us

* We propose this designation for the Lophiade and the Chironectide,
forming the Linnean genus Lophius.

48 CLASSIFICATION OF FISHES.

to infer that the sense of taste is very slightly possessed
by fishes. The structure of their teeth, with few ex-
ceptions, shows that the food is generally swallowed in
an entire state, since it is so found in the stomach; and
very few instances occur of fish having cutting or grind-
ing teeth. Besides, it has been justly observed by Mr.
Yarrell*, that from being obliged unceasingly to open and
close the jaws for the purpose of respiration, fishes can-
not long retain food in the mouth when shut; the sub-
stance, if of small size, must be swallowed quickly. The
structure of the tongue tends to the same conclusion ;
we believe it is in all cases small, hard, and generally
cartilaginous, and consequently incapable of conveying
that exquisite taste of their food enjoyed by all the qua-
drupeds, and a few of the birds.t Fishes have been
supposed destitute of the faculty of hearing, but this
is disproved by many circumstances. It is known as a
well-authenticated fact, that the Chinese, who breed great
numbers of goldfish, call them together, at the time of
feeding, by a whistle; and the same mode of summon-
ing other species by a noise, in aquatic preserves, are
upon record. There are, indeed, no external indications
of ears in any fish, excepting the rays, where there is a
small spiral cavity (placed before the meatus externus,
and covered by the common skin), which may be ana-
logous to the external ear of other animals. The internal
labyrinth, however, is always present, although much
less complicated than in,the more perfect Vertebrata.
(50.) The vitality of fishes may here be adverted to.
There is not sufficient evidence to show us the average
age of the generality of fishes; but some well authen-
ticated facts regarding carp, and some other domes-
ticated fish, tend to prove that the former have reached
to a century. Cartilaginous fishes, from the nature of
their bones, continue to grow all their lives; and as
many of these, particularly the rays, habitually live in
the deep recesses of the ocean, and thus seldom run the

* Yarrell’s British Fishes, i. xvii.
+ Particularly the whole family of Anatéde, or ducks.

VITALITY OF FISHES. AQ

ehance of being captured by man, we may probably
attribute their enormous and almost incredible size to
their great age. Several genera, like the Ophicephali and
eels, are so tenacious of life; that they are well known
- to live under sufferings which, to other animals, would be
the most cruel torments; while others die almost the
minute they are taken out of water. Many fish show
their tenacity of life in other ways: some can not only
exist, but actually breed, in hot springs of various coun-
tries, whose temperatures vary from 80° to 120° Fahy.
But a statement by baron Humboldt, on this subject, is
still more surprising: he mentions, that during his
researches in Tropical America, he found fish thrown
up alive from the bottom of an exploding volcano,.-
along with water so hot as to raise the thermometer
to 210°, being two degrees only below boiling. Con-
sidering this excessive heat, it is, we think, too much to
suppose that the water in which these fish habitually
resided was always of such a temperature. It is a well-
known fact, that springs in the vicinity of volcanoes are
very often considerably heated before an eruption takes
place; and until we are in possession of further evidence
on this point, we believe that such was the case in the
present instance: the internal fires, in all probability, had
greatly heated the water previous to its having been
expelled from its natural basin, before the increased
heat had killed the fishes; a supposition much more
probable, it appears to us, than that fishes would live
and sport in a fluid whose temperature would be suffi-
cient to prepare them for the table. We have already
alluded to the singular faculty possessed by the Ophice-
phali,and some other fish, of crawling upon dry land, and
thus living in an element not their own: it is well known
that the tanks or isolated reservoirs of water in the Kast
Indies are often completely dried up during summer;
and yet, when they become again filled during the rainy
season, fish are also found in them. This singular fact
appears to be accounted for very satisfactorily by Mr. Yar-
rell: the impregnated ova (he observes) of the fish of one
VOL. I. E

<p) ae
LS

50 CLASSIFICATION OF FISHES.

rainy season are left unhatched in the mud through the
dry season, and, from their low state of organisation as
ova, the vitality is preserved till the occurrence and
contact of the rain and the oxygen of the next wet
season, when vivification takes place from their joint
influence. ‘If this solution ot the problem,” continues
our author, “ be the true one, it points at once to what
perhaps may be effected after a few experiments, —
namely, the artificial fecundation of the roe, the drying
of that roe (or of other roe naturally impregnated)
sufficiently to prevent decomposition, and its possible
transportation to, and vivification in, distant countries.”

(51.) Contrasted with these instances of fishes living
in heated water, there are numberless others proving their
vitality even in a frozen state. It is even said, that in
northern latitudes, advantage is taken of this circumstance
to transport eels and perch from one locality to another.
It must not be supposed, however, that this vitality
exists in all species inhabiting the same latitudes; and we

‘ean illustrate this idea by a fact which has unfortunately

come under our personal observation. Upon the breaking
up of the long and severe frost of this winter (1837-8),
we have had the mortification of seeing the dead bodies
of between thirty and forty fine tench floating on the
surface of a pond in the garden, into which three or
four pair had been put four years ago. The pond is of
rain water, with a soft muddy bottom, which has a depth
of from two to four feet, and is fringed with many
aquatic plants. Abundant shelter was thus afforded for
the fish ; and yet there can be no doubt, we think, that
they have all been killed by cold. The people about
the place assert that this mortality would not have hap-
pened, had holes been broke in the ice for the admission
of air: but were this absolutely necessary in all cases, it
would follow that the tench of all such ponds as had
not been opened would have been likewise killed.

(52.) The fecundity of fishes is something so pro-
digious as to stagger the belief of ordinary minds. When
we say that a single female lays hundreds of thousands

GEOGRAPHIC DISTRIBUTION GF FISHES. 51

of eggs in a single season, the statement is not exagge-
rated; and yet the waters are not more densely populated
now than they were in the last generation. The reason
appears to be this: all fish are more or less carnivorous,
and feed not only upon other marine animals, but upon
each other. To supply this latter food in sufficient quan-
tity, as well as to provide against other casualties, Infinite
Wisdom has given to these His creatures a power of re-
production without parallel in the animal creation: were
it not so, the seas would be depopulated of all other in-
habitants, or thousands would perish by the most cruel
of all deaths, starvation : as it is, a momentary pain is all
that can be experienced by a fish which is seized and
swallowed in an instant by a larger one: and although
this is probably the fate of countless millions, little or
no corporeal pain, in the true sense of the term, can be
experienced by a death so instantaneous.

(53.) The natural history, or, in other words, the
habits and economy of this class, in comparison to
that of terrestrial animals, is involvéd in great ob-
scurity, and presents little of that popular interest
attached to the economy of birds and quadrupeds.
Nevertheless, the history of such fish as the salmon,
herring, mackerel, &c., is highly interesting both to the
naturalist and the general reader: they form an im-
portant part of our subsistence ; while great numbers of
men, and large amounts of capital, are engaged in their
capture. We should have regretted that our limited
space would not allow of entering into all these details,
could we not refer our readers to the two interesting
volumes already cited on British Ichthyology.

(54.) The geographic distribution of this class has
been very much neglected; for, with the exception of the
valuable observations of colonel Hamilton Smith, we are
unacquainted with any author who has written upon this
interesting and important subject. Our own observations,
made in different parts of the world, tend to confirm
nearly every circumstance mentioned by the above-named
able and accomplished naturalist. Nevertheless, from

E 2

ae CLASSIFICATION OF FISHES.

many facts that could be mentioned, we believe that the
geographic range of the certain genera and species is
much more definite than has hitherto been supposed.
Several of the Mediterranean species, which are uni-
versally believed to inhabit the seas of Tropical America,
we consider to be truly distinct; and similar differ-
ences may be detected even between the fish of Northern
and of Southern Europe. One great cause of the sup-
position that the same species so frequently inhabits
widely separated shores, is the fact that this class is less
affected by temperature than any other vertebrated ani-
mals ; and it is therefore inferred that the similar species
may exist both in temperate and tropical latitudes: this
may seem to be true; but then the question arises, whether
their peculiar food is also found in the same seas? The
majority of fishes are carnivorous ; and it may be said,
that as small fish are to be found every where, the larger
can prey upon them ; but such is not precisely the fact.
We know that every family, nay, almost every species,
of insectivorous birds, feeds only upon certain genera of
insects ; and all we know, both from fact and analogy,
favours the idea that carnivorous fishes are limited in
their choice of food by similar laws: indeed, this belief
almost amounts to absolute certainty, when we consider
that different tribes are generally found restricted to dif-
ferent depths and descriptions of sub-marine soils. This
fact has been so ably illustrated by colonel Smith, that
it need not be insisted upon in this place. Nov, it is
quite evident that this allotment of particular depths or
localities is an instinct given to them for frequenting
those situations, and those only, where they are sure of
finding their congenial food. We may even suppose
that such as live upon testaceous Mollusca and crusta~
ceous insects are more limited in their range than those
which live upon young fish, because the former animals
are more limited in their distribution than the latter:
again, the soft pelagie Mollusca are more widely dis-
tributed than shells or Crustacea; and, therefore, those fish
which fed upon them would enjoy a greater range than

GEOGRAPHIC DISTRIBUTION OF FISHES. 53
others. That certain families, and even genera, are
strictly limited, so far as we yet know, to the shores or
rivers of particular countries, is unquestionable; and of
this the family of the Si/uride, upon which we shall
subsequently dilate, offers several singular proofs. The
most typical belong only to the equinoectial rivers of
America; while nearly all the sub-typical, that is, of the
Pimelodine, occur in the great rivers of India: we
suspect, even, that such of these latter as have been found
in America will prove to be distinct geographic sub-
genera. The whole family may be considered tropical ;
for the only species yet found in the rivers of Europe is
of a very aberrant form, and is as much related to the
aberrant Gadiade as to the typical Loricarine. ‘The
Gadiade, or cod-fish, again, seem to supply, in cold and
temperate regions, the place of the Siluride: they are
most abundant on the confines of the Arctic seas, and
graduaily diminish as they approach the southern shores
of Kurope, where the species, although many, are almost
all of the aberrant forms ; and yet not one example of
the whole family was observed by us in the Brazilian
seas. As we shall occasionally touch upon this subject
in the succeeding pages, further instances need not be
mentioned in support of our opinion. The fact, we
have no doubt, will ultimately be established, that fishes
are nearly as much limited in their geographic distri-
bution as birds; and that temperature alone has very
little to do in regulating this distribution.

54 CLASSIFICATION OF FISHES.

CHAPS PEI.

A SKETCH OF THE HISTORY AND BIBLIOGRAPHY OF ICHTHYO=
LOGY, WITE SOME REMARKS ON COLLECTING AND PRESERY=
ING FISH.

(55.) A LENGTHENED exposition of the rise and progress
of ichthyology is not suited to the present publication, and
would occupy more space than we could devote to this
department of zoology; but a few general remarks on
this subject cannot well be dispensed with. Like ali
other sciences, its progress has been unequally progres-
sive, according to the degree of attention or of neglect it
has received in different periods.

(56.) The ancients appear to have paid more attention
to this class of animals than any other, and have left us
thenamesof nearly 200 different species, chiefly inhabiting
the shores of the Mediterranean—the majority of which
were then, as now, in request as food for the highest as
well as the lowest ranks. After the revival of learning,
and in the middle of the sixteenth century, ichthyology,
as a science, first began to assume a new birth in the
writings of Belon, Salviani, and, more especially, Ron-
delet, better known under the name of Rondeletius. It
is a most fortunate circumstance that these early writers
bestowed so much labour in determining the names by
which the Mediterranean fishes were known to the an-
cients, which they justly considered of much import-
ance. Immense labour, research, and doubtful disput-
ation have thus been saved to the mederns; while, on the
other hand, had they attempted to describe, in greater
detail, the internal and external structure, the proba-
bility is, considering the age in which they wrote, that
their bocks would have been utterly useless to modern
science. As it is, however, they are actually useful, and
often essential, not only as high authorities for the ne-

HISTORY OF ICHTHYOLOGY. 55

menclature of antiquity, but for the characteristic, al-
though often rude, accuracy of their figures. Those of
Rondeletius, more especially, are in several instances
more faithful than many that are inserted in our modern
publications ; and to this day they continue to be quoted
as authorities by our best writers. The credulity of the
age, indeed, was in some degree shared by these twin-
kling stars of light in the returning dawn of knowledge;
for Rondeletius has left us pictorial representations of
certain cunning fabrications, called the monk-fish, the
bishop-fish, and the sea lion. It is singular that these three
fathers of science flourished at the very same pericod— all
three having published their works between the years
1553 and 1558. They seem, however, to have left no
disciples ; for during more than a century ichthyology
appears to have lain dormant, until, in 1686, it was again
revived by the labours of our illustrious countryman,
Willughby, the patriarch of zoological science in Britain,
in conjunction with his tutor and companion, the learned
and pious Ray. When it is considered that no less than
186 folio copper-plates form the pictorial volume of Wil-
lughby, in an age when natural history had not a twen-
tieth number of the votaries who now profess to be so, we
cannot but feel surprised at what may be called the
““ spirit” of the booksellers of that age, in undertaking
the publication of a work which none of cur modern
bibliopoles would think of venturing upon. This vo-
lume is altogether not only highly curious, but even
valuable. The figures are very unequal, since it seems
to have been intended to comprise a complete collection
of all known fish: hence those found in the volumes of
Rondeletius, Salviani, Marcgrave, &c. are faithfully co-
pied ; but these are interspersed with a large number
of original designs, many of which are drawn and etched
with a degree of accuracy, spirit, and effect, which it
would be even now difficult to surpass.* Ray’s Systematic

* Among these, the reader may refer to tab. E. 2. F.i. 3, 4, 5, 6, &c.
The holibut (fig. 6.), is uncommonly fine, and the flatness of the sole (F.
Jig-7.) is inimitably expressed.
E 4

56 CLASSIFICATION OF FISHES.

Synopsis appeared only in 1713, and contains little that
can be said to have advanced the science.

(57.) But ichthyology was now to assume a form
and order which it had never yet appeared in; for,
in the year 1738, the works of the great Artedi,
the friend and disciple of Linneus, were given to
the world by his no less celebrated master, whose
arrangement of the fishes in the Systema Nature,
no doubt, laid the foundation of that by his scholar.
Having already, in a former volume, expressed cur sen-
timents on the general system of Linneus, we may at
once pass to that of Artedi, whose knowledge of fishes,
and whose views on their natural classification, are un-
doubtedly much more profound and correct than those
exhibited in the Systema Nature. Artedi, in short, must
be considered the true founder of systematic ichthyology:
he has treated the subject both as a philosopher and a
naturalist ; and we presume to think he deserves much
higher honour than some writers of the present day have
been disposed to give him. It is not a little remarkable,
and may be urged as a proof how truly he deserves this
praise, that three out of his five primary divisions have
been adopted by M. Cuvier ; of the other two, one (Pla-
giuri) is composed of the cetaceous Mammalia, and the
other of the Plectognathis (Cuv.). True it is that Artedi,
like all the naturalists of that time, was not aware of this
latter order possessing branchial rays, and consequently
named them Branchiostagi; nevertheless, it is quite clear
that Artedi perceived they formed a natural group, how-
ever he erred in part of their definition, for he united
with them the genus Lophius and Syngnathus, the whole
of which, as will hereafter appear, possess all the cha-
racters of a primary order. We must leave this sub-
ject, however, which more properly belongs to another
part of our volume, and turn to another labourer in the
same vineyard, although in a different department. We
allude to Klein, whose valuable labours on the anatomy
of fishes first opened the view of a new and untrodden
field to future ichthyologists, and laid the foundation of

HISTORY OF ICHTHYOLOGY. ae

all that has since been accomplished. Klein was a most
industrious and even voluminous writer; and though but
little can be said of his ornithological writings, those
which relate to the class before us place him, in our
estimation, among the most eminent writers in this de-
partment of zoology. His chief work is now become so
very scarce *, that we have never seen a complete copy
offered for sale; while the numerous figures it contains,
although perhaps not equal to those of the present day,
will always render the work a standing authority.

No publication of moment appeared during the next
fourteen years, excepting that of Gronovius, whose name
still ranks high both in botany and zoology. Of his
writings we have already spoken. t His latest work on
ichthyology, the only one we possess {, is still of much
value, not only from containing the characters of several
genera first defined by this author, but also for the ex-
cellency of the plates; nearly all the figures, indeed,
are admirable, and most of them, in the artistical spirit
of their execution, are equal to the very best of the
present age. The next author of any considerable note
was Gouan §, whose ichthyological labours were confined
to one volume, in which the genera are described with
all that attention to detail, and in that technical lan-
guage, introduced by Linneus with such incalculable
advantage to science.

(58.) Hitherto, however, ichthyology had been en-
tirely without any work expressly devoted to coloured
representations of fishes: the magnificent volumes of
Catesby, indeed, on the natural history of Carolina,
contained several figures of this class of animals ; yet it

* Jacobe Theodore Klein, Historia Piscium Naturalis, promovende
missus, 1—5. Gedani, 1740—1749. The first part contains six plates; the
second, four; the third, seven; the fourth, sixteen; the fifth, twenty ;
besides a portrait of the author. ——, Mantissa Ichthyologica de Sono et
Auditu Piscium. Lips. 1746. In my copy of this volume the following
note is inserted: -—‘* This is one of the scarcest modern books of its
kind that 1 know of; I desired Dr. Schceffier, of Dantzig, to procure me
a copy, but there was not one to be had in 1772.— A. Y. B.”

f+ Preliminary Discourse, p. 43.

i LT. Gronovius, Zoophylacii Gronoviani, fascic. 1. Lugduni Bata-

yorum, 1763, folio, with thirteen plates of fish,
. 4 Ant. Gouan, Historia Piscium. Strasb. 1770. 1 vol. 4to.

58 CLASSIFICATION OF FISHES.

was not until the year 1785 that the first work of this
sort, expressly devoted to fishes, was commenced by the
celebrated Bloch. He was a Jewish physician settled
in Berlin; and his ichthyology, in twelve folio parts,
contains no less than 452 coloured plates: of these, 216
belong to the first six parts, and comprise nearly all
the European fish ; the other six, more especially de-
voted to the exotic tribes, are now very rare, in conse-
quence of a fire having destroyed the greater portion of
the copies. The figures, however recognisable, in most
instances are very inaccurate both in their drawing and
colouring, particularly those in the latter volumes; so
that they fall short, in every respect, to those of Grono-_
vius and the original plates of Willughby: neverthe-
less, Bloch must always be classed among the highest
ichthyologists: his descriptions are generally very good,
and he refrained from incorporating in his work a great
number of species loosely described. and still worse
figured, in former publications. This judicious plan,
however, was not followed by Schneider, his commen-
tator and continuator, who published two additional
volumes with 110 plates, so late as the year 1801.* The
admirable volume on the anatomy of fish, by Dr. Munro,
was also published in 1785.f It is gratifying to our
national character that the labours of our distinguished
countryman should thus have laid the most permanent
foundation for all that has been subsequently achieved
in this department. The great work on the natural
history of fish, by the count Lacepede {, was the next
publication after that of Bloch upon general Ichthyology.
As it embraced an account of all recorded species, whe-
ther examined by the author himself or known only from
the descriptions of others, it became, in some degree, 2
compilation, as all general systems so constructed must
be ; when, therefore, we make allowance for this, and
for the very little attention that was then paid to eha-
* Schneider, Systema Ichthyologia. Berlin, 2 vols. 8vo. 1801.
+ This is omitted by some oversight in the lists of the Régne Animal.

t Lacepede, Comte de, Hist. Nat. générale et particuliére des Poissons.
Paris, 5 vols. 4to. 1798—18¢3.

" HISTORY OF ICHTHYOLOGY. 59

racters now found to be of much importance, we must
pronounce this the most valuable ichthyological system
that had then appeared. It is not, like others in differ-
ent branches of zoology, a servile copy of the Linnean
divisions, but numerous others are defined for the first
time : and when we look back to what systematic ich-
thyology was before, and what it became by the labours
of Lacepede, no one can in fairness deny but that a
great and important advance in this science had been
effected. No naturalist can hope to achieve more than
this, however great may be his abilities; and we do not,
therefore, understand upon what ground so much cen-
sure has recently been cast upon the works of this dis-
tinguished Frenchman by some of his own countrymen.
Lacepede’s generic names, indeed, are destitute of
euphony ; but this is secondary, and can easily be reme-
died ; and numerous errors may, no doubt, be found in
ate a vast undertaking: but we contend again, that
these errors were feria! and resulted more from the
paucity of his materials, and the inaccuracy of those
who had gone before him, than from any deficiency in
his powers of Mecrinnatan. Such errors might be
pardoned half a century ago, but are totally inexensabile
in the present day. Certain it is, however, that Lace-
pede’s Ichthyology will always be a standard authority,
even for his supposed errors ; and it will be found by
those who have occasion to consult them, that he is by
no means chargeable with several that have been of late
attributed to him. The figures, on the other hand,
although well engraved, are, in general, very deficient
in accuracy; the major part being either copies, or
drawn by artists who were totally ignorant of the sci-
entific details of their subject. It is certain, however,
that the work had a great and almost immediate effect
in awakening attention to this long neglected branch of
zoology. The interval between the respective works of
Bloch and Lacepede comprised a period of near twelve
years, in which, with the exception of a number of
valuable anatomical dissertations, nothing of material

60 G6LASSIFICATION OF FISHES.

impcrtance on the general subject had appeared. We
here except the compilation of Gmelin, which, however
useful it might have been in regard to species, cannot be
said to have permanently advanced the science. But no
sooner had the great reformations effected by Lacepede
become generally diffused, by reprints and translations,
than ichthyology received a new impetus ; whether this,
however, was the true cause, or whether, about this time,
zoology in general began to be more studied, certain it is
that it advanced more rapidly. The clear and compen-
dious tables of M. Dumeril*, which incorporated the
new divisions of Lacepede, placed all the modern im-
provements of artificial classification in the hands of
students ; and although the naturalists of Britain still
adhered to the Linnean system, that of Lacepede was
generally adopted on the Continent. A most valuable
addition to our knowledge of the fishes of India -was
made in 1803 by Dr. Russell ; the descriptions are excel-
jJent, and the figures, although in outline, and executed by
Indian artists, sufficiently good for scientific purposes.
(59.) The year 1810 was remarkable in the annals
of our science for the appearance of two important works
on the ichthyology of the Mediterranean: one was by
M. Rafinesque Schmalz, 2 Su bseauciy professor of natural
history in Lexington, U.S. ; the other, relative chiefly
to the fishes of ‘Nice, was from the pen of M. Risse.
The first of these is of much importance; and, from
particular circumstances, will claim more of our attention
than would at first appear necessary. M. Rafinesque’s
Sicilian works are now become so very scarce (the greater
part of the unsold copies having been lost at sea), that
few naturalists will have the power of consulting them.
His chief ichthyological work is a synopsis of ‘“ New
Genera and Species of Animals and Plants” found by
the author in Sicily; and this was followed by a pamphlet,
entitled “ Indice d’Ittiologia Siciliana.’’ The details of
the new views of M. Rafinesque, in regard to classifica-

* Dumeril, Zoologie Analytigue, 1 vol. 8vo. Paris, 1806.

HISTORY OF iCHTHYOLOGY. 61

tion, are too long to be inserted in this volume, but they
will be occasionally adverted to. The faults that have
been dwelt upon* in these two works are such as all
authors, even M. Cuvier himself, is not exempt from ;
they seem to us, in short, too trivial for the notice of the
historian, and too general to be affixed to any one author -
in particular. We tt eely admit that M. Rafinesque (then
living, as we were, in a remote part of Europe, cut off,
by the late war, from all intercourse with the Continent)
was not well informed upon the current and almost daily
discoveries going on there; and that some few of his
species then supposed new, were really not so: but who
is exempt from such errors, if errors they are ? or how are
such coincidents to be prevented, when naturalists, in
distant places, and unknown to each other, are working
at the same time upon the same subject? On tne other
hand, it must not be conceaied that M. Rafinesque an-
ticipated, by nearly ten years, a very large proportion of
the generic and sub-generic distinctions subsequently
taken up in the Régne Animal, in the first edition of
which it is clear that its learned author was totally
unacquainted with the works above mentioned, or
that he was unconsciously repeating, under new names,
a considerable number of the genera and sub-genera

* Mi. Cuvier observes: ** He has, besides, entered in his catalogue, with-
out examination, all the species given by Lacepede and Linnzus as belong-
ing to the Mediterranean, which has caused him to reckon several which
are purely imaginary ; and this extends even to his genera: thus, his dodon,
taken from Lacepede, is the Raie cephaloplére ; his Macroramphus, taken
from the same source, is the Centriscus. He has greatly multiplied the
genera, and sometimes on slight grounds; so that, without reckoning
those which are not inhabitants of the Mediterranean, there are 139; and
yet, notwithstanding his readiness to make these divisions, he has not
done so in circumstances in which it would be imperatively commanded by
the laws of classification. He leaves, for instance, the anchovy in the
herring genus, and the plaice in that of the sole; while of the single Lin-
hzan genus of Sgualus he has made sixteen.” ‘* These two works are,
nevertheless,” continues M. Cuvier, ‘‘ very worthy of attention, on account
of some original ideas, and of descriptions and figures of the fishes them-
selves, which are to be found nowhere else. The author, also, has paid
attention to the Sicilian names of most of his species.” If Rafinesque made
too many genera, M. Cuvier has nearly doubled them ; and as for the “ laws
of classification,” which imperatively command the formation of these
genera of M. Cuvier, the term is totally misapplied. Genera, like those of
Rafinesque and Cuvier, are mere matters of individual opinion, beeause
they are made without any ulterior reference, and are merely divisions,
with which no laws of artificial classification have any thing to do.

62 CLASSIFICATION OF FISHES.

long before established in the volumes of professor
Rafinesque. It would have been well had these un-
intentional errors been rectified in the second edition,
or in the general ichthyological work of MM. Cuvier
and Valenciennes ; but they are not so ; and naturalists
will judge how far this is consonant with common jus-
tice, or with that law of priority which is the only safe-
guard to the reputation we all covet. The generic cha-
racters of Rafinesque are as simple and intelligible as those
of Linneus, and the derivation of their names strictly
classical and euphonious. In regard to the majority of
those species which have been termed “ imaginary,” or
inaccurately described, our firm conviction is, that
nearly all, eventuaily, will be as fully established as those
of the best known in our systems. We have formed this
opinion not from theory, but from actual observation,
and from having verified, in many instances, the va-
lidity of Rafinesque’s characters.* The truth is, that

* In further justification of the opinions here advanced, it may be proper
for me to state that I had the pleasure of M. Rafinesque’s society, during
the three years of my official residence in Sicily, from 1807 to 1810, and
again in 1812, when we were both at Palermo, prosecuting our botanical
and ichthyological researches together. Circumstances have hitherto pre-
vented me from giving them to the public; but an extensive series of
drawings and descriptions, made from the life, of the Sicilian fishes, not
only confirms the accuracy of M. Rafinesque, in many instances where he
has been charged with error, but affords strong grounds for believing that
one half of the Sicilian species, said to be found also in the Atlantic Ocean,
Britain, &c., are, in reality, quite distinct. M. Rafinesque, unfortunately,
was unable te publish more than a synopsis of his ichthyological dis-
coveries; and his figures, being very slight, are often not calculated to
clear up those doubts which the brevity of his deseriptions sometimes
creates: nevertheless, to one who examines the species on the spot, in a
Fresh state, there are few which may not be identified. M. Cuvier often
asserts that all M. Rafinesque’s species were described from preserved
specimens; but this is an error —they were all taken from the life. We
Doth used to frequent the fish-markets, and we procured all our specimens
there, or from fishermen who were in our employ. I was frequently
urgent with my friend to preserve, at least, such as were the most remarkable
of his new genera, anticipating the incredulity that has sincebeen attached
to them; but this advice, unfortunately, he never adepted. The greater
part of those which J] examined, after being drawn and described, were
thrown away or eaten; a military life not being suited to the formation of
such collections : but many of those species met with near Palermo, were
preserved in spirits, and sent to the British and Zoolegical Museums ;
few, however, of these ate now In existence. One cause, perhaps, of the
errors of M. Cuvier regarding the Mediterranean fishes, may be, that he
had only examined preserved specimens, either distorted by stuffing, or
bleached and shrivelled by alcohol, so that it becomes often difficult to
recognise the most common species. If I have dwelt too long upon this
subject, I hope the benevolent and candid reader will excuse me: it has
eriginated in my desire to do adequate, though tardy, justice te ene whose

HISTORY OF ICHTHYOLOGY. 63

Sicily is perhaps the richest field for the ichthyologist,
of any yet explored in the Mediterranean, in whose
warm and prolific waters, washing the tranquil shores
of so many islands, an immense variety of fish are
constantly found. Besides these two works, more es-
pecially devoted to the ichthyology of Sicily, many
other papers by the same author are scattered in the
periodical publications of Palermo; and he has also
given a most original and valuable account of the fishes
of the great river Ohio. The second volume on
Mediterranean Ichthyology, by M. Risso, just alluded to,
is highly interesting, from an account of several new
species, and a few new genera; but the classification is
that of Lacepede, and the figures too small to be ser
viceable: a second edition, as we find, was subsequently
published ; but this we have not yet seen. The fish of
these shores were subsequently illustrated, in detached
portions and separate essays, by several learned foreigners,
among whom the names of Viviani, Spinola, and Va-
lenciennes, are conspicuous ; while the labours of
Leach and Montagu, in our own country, have been
justly praised. A most perfect and masterly account
of the singular fishes of Egypt has proceeded from
the accomplished pens of the illustrious Geoffroy Saint-
Hilaire, and his talented son Dr. Isidore Geoffroy;
the figures are drawn from the life by the younger
Redouté, but they are by no means good: the expense
of this valuable work renders it inaccessible to the ge.
nerality of purchasers. A decade of Cuban fish, very
fully and perfectly described, came from the pen of
M. Desmarest; but the plates by which it was intended
to be illustrated, we have never seen. The ichthyology

whole life has been devoted to science, and who has been singularly un-
fortunate in his worldly concerns ; who, notwithstanding his eccentricities,
has a kind and benevolent heart; and whose labours have never been
appreciated as I think they deserve. But for this, M. Rafinesque would
not, in advancing life, have to contend with pecuniary difficulties, from
which a small pension from the American government, prover bially ge-
nerous to her scientific sons, would set him free.

64 CLASSIFICATION OF FISHES.

of America, about this time, began tc excite the zealous
attention of several of our transatlantic brethren; and
the various essays and papers by Dr. Mitchell, Le
Sueur*, Harwood, and Rafinesque, have accumulated
such valuable materials, that we trust they may be soon
augmented, and condensed into a general work devoted
to this branch of American zoology.

(60.) But we must not depart from the chrono-
logical order of our rapid survey. ‘The year 1817
saw the publication of the first edition of the “‘ Regne
Animal,” —a work replete with profound anatomical
science, and with many just and admirable improve-
ments in scientific arrangement. Having already spoken
so fully of these celebrated volumes, on a former occa-
sion, we have only to look to its ichthyological portion.
Besides the genera that had previously been named and
defined by Rafinesque, but unknown, and therefore
unacknowledged by M. Cuvier, there are a great number
of others really new; and the whole, being well
digested, give us the most finished and popular system
that had appeared since the days of Lacepede. It must
not be supposed, however, as some have imagined, that
there was any thing sudden or astonishing in the ad-
vance which was thus made. Ichthyology, like all
other branches of natural history, and, indeed, all other
sciences, had been advancing gradually and progressively.
Since the decline of the Linnzan school, the first, and
therefore the most signal, reformation in the genera
was undoubtedly effected by Lacepede: the new groups
pointed out by Rafinesque, materially advanced this

* Tt is scarcely possible to praise too highly the delicate and masterly
delineations which so peculiarly characterise every subject which comes
from the pencil or the graver of Le Sueur, whom I have ever looked upon
as the first zoological artist of the age. His are the only delineations I
have seen, where the delicacy, the accuracy, and the high finish of the
French school are united with the freedom, grace, and decision of the
English style: the ease and ingenuity with which he can comprise large
subjects within a small compass, without the least confusion of the parts,
is seen in many of the exquisite outline plates, drawn and etched by him-
self, in the early volumes of the American Transactions. Science and the
fine arts must ever deplore that the noble work on the Meduse, long con-

emplated by this prince of zoological painters, has never been given to

the world. Surely a sufficient number of subscribers might be found to
protect the author from pecuniary loss ?

HISTORY OF ICHTHYOLOGY. 65

reform ; and M. Cuvier’s system, again, aided by his
high reputation and anatomical skill, made another
signal yet graduated advance towards a knowledge of
the true structure of this difficult class. Like all other
new systems, however, it was some time before this re-
ceived favour or adoption, at least in this country ; and
such will ever be the case when old ideas are to be cast
aside, new ones learned, and prejudices overcome. . The
truth is, that no favour or support can be expected to new
views from old naturalists: we do not like to have our
long-cherished creeds disturbed ; and without, perhaps,
being aware of it, we naturally, and almost inevitably, be-
come strongly prejudiced in favour of what is old and
established. It is, therefore, not so much to the existing
as to the succeeding generation that we must look for a
candid and impartial judgment upon those innovations,
and which are in direct opposition to high authorities and
long-cherished views. And this, perhaps, is for the best.
Throughout nature, that which is most permanent is of
the slowest growth: the oak is only in its vigour, when
the surrounding plantations of poplars and larches are
Withering into decay.

(61.) Additions to ichthyological science now be-
came so numerous, that we must altogether confine our
notices to such as are of leading importance. In this
view we must regard the most valuable account now
extant of the fishes of India, more especially those of the
Ganges, by Dr. Buchanan Hamilton. The descriptions,
which are clear and ample, are interspersed with many
original and interesting observations on affinities and
natural groups; while the figures, much superior to those
of Russell, are very neatly executed. A vast number of
new species are here first described. We have no he-
sitation in considering this work as the most original
and valuable that this country has yet produced ; and it
places its author, now dead, in the foremost ranks of this
science. The different artificial systems of MM. Blain-
ville, Risso, Pallas, Goldfuss, and several others, need not
here be mentioned ; they are not founded upon any ge-

VOL. I. F

66 CLASSIFICATION OF FISHES,

neral considerations, drawn from other classes of the
‘animal kingdom ; and although each makes, in some of
the details, a greater or lesser approach to nature, each
may be also said to have its weak points. Neither have
we space to particularise, in detail, the valuable additions
made to the comparative anatomy, or rather the internal
structure, of fish, by many able and skilful men, who now
began to take up this department of the science ; most
of these essays are in the voluminous and expensive
Transactions of societies, and are therefore not very acces-
sible to the student. This latter obstacle, unfortunately,
is also an impediment to the possession of the numerous
and beautiful figures of fish dispersed in the Zoological
Atlases of the French circumnavigators, and described
by the naturalists who accompanied the different expe-
ditions: many interesting fish are also figured among
the plates taken from the late general Hardwicke’s
Indian drawings, edited by Mr. J. E.Gray; and the
volume on those discovered by Dr. Richardson forms a
valuable addition to our knowledge of the Arctic species.

(62.) There are two important works, however, which
deserve a more particular notice: one of these includes
the numerous and beautiful species discovered by that
enterprising traveller and accomplished zoologist, Dr.
Riippell, on the shores of the Red Sea.* Although,
from being drawn on stone, the execution of some of the
figures appears to he coarse, yet they are the most masterly
and artistical (next to those of Le Sueur) that we have
ever seen: they wear every appearance of having been
drawn and coloured from the fresh subjects with evident
care and exactitude; so that they deserve to be ranked
among the most valuable that have ever been published :
the descriptions are in German, but the specifie cha-
racters are also in Latin. We anxiously look forward
to this unrivalled collection of coloured figures being
augmented, and in the same style, by those new species
discovered during the second expedition of this zealous

* Atlas zu der Reiseim Nordlichen Afrika, von Eduard Riippell — Fische
des Rothen Meers, Frankfurt am Maine, 1828, folio, with 35 plates,

HISTORY OF ICHTHYOLOGY. 67

naturalist to the same regions.* The other work we
allude to is on the fishes of Brazil, discovered by the
late Dr. Spix, and edited by one of the most accom-
plished of living ichthyologists, M. Agassiz: the figures
seem to he accurate}, and are highly finished, —too
much so, indeed, since this circumstance renders the
work very expensive; while the letterpress is particu-
larly ample and elaborate: several plates are devoted to
pictorial representations of the modes of fishing pursued
by the native tribes; and others, to the delineation of
the different form of the scales in various species, a
subject upon which M. Agassiz is known to have be-
stowed great attention. It is to be regretted that so
very few of the discoveries, not merely of new species,
but of singular and hitherto unknown types, contained
in the works just mentioned, should have been incor-
porated in the second edition of the Réegne Animal, of
which the ichthyological volume appeared so late as
the year 1829. M. Cuvier, indeed, has here character-
ised several additional genera, not contained in the first
edition; but they are chiefly, if not entirely, the fruits of
his own observation. These additions, however, form —
but a small! proportion of the discoveries effected in this
Science since 1817 ; so that the last work must be looked
upon more as the result of the learned author’s indi-
vidual researches, than as giving a general exposition of
the present state of ichthyological knowledge. As a
collection of important facts, and of anatomical investi-
gations, it excels all others; and whatever objections
may be raised to the formation of the groups, there can
be but one opinion of its being of great usefulness. The
extensive researches of the author are more conspicuous
in the great work commenced by him in conjunction
with M. Valenciennes; and this will ever remain an

* Since the above was written, Dr. Riippell has kindly forwarded us a
copy of his Second Atlas, entitled “ Neue Wirbelthiere zu cer Fauna
Abyssinien gerdrig, &c.” The fish form one vol. folio, with 33 plates, more
delicately but less vigorously delineated than the other.

+ Although the majority are deficient in grace, and what is called good
érawing,the minute details of the teeth, &c. are particularly weli done.

EF 2

68 . CLASSIFICATION OF FISHES.

honourable monument of their joint labours. The de-
scriptions are generally ample, and the plates are deli-
eately and, for the most part, correctly executed: we
sincerely trust that this valuable work has received no
check in its publication. It is now near twelve months
since the last or twelfth volume was published; and
nearly as many more will be necessary to complete it on
the same plan. Such a work, as a general history of
fish, is the most perfect yet contemplated ; and no library,
whether public or private, ean be called well selected,
without these volumes. Having said thus much, our
further observations will be offered under the head of
those groups to which they are more especially appli-
cable. Before concluding this hasty sketch, we cannot
emit to mention the valuable addition made to our native
ichthyology, by the two volumes upon British fishes, by
Mr. Yarrell* ; they form a most important acquisition
to the British naturalist ; and they doubtless will be the
means of eliciting, in a few years, a vast mass of new
information on these animals. Notwithstanding the
numerous additions thus made to our marine fauna,
the perusal of these volumes has convinced us, that
‘many species require further investigation. The re-
cent discovery of that extraordinary fish by Mr. Couch,
our well known Cornish ichthyologist, which has been
named Amphioxus lanceolatus by Mr. Yarrell, seems to
justify our anticipation of the novelties yet to be found
on the British coast. We have seen the prospectus of a
general work on the fluviatile fish of Europe, with coloured
plates, projected by M. Agazziz, but we know not whether
its publication has commenced.

(63.) A few remarks on the PRESERVATION OF FISH
will probably be useful to many of our readers, parti-
cularly in a volume which is intended:as a compendium
and text book for the ichthyological student. Unfor-
tunately for our museums, no method has yet been
discovered by which the rich and vivid colouring,

* William Yarrell, V.P.Z.S., F.L.S., A History of British Fishes, illus-
trated by nearly 400 wood-cuts. 2 vols. 8vo. London, 1836.

PRESERVATION OF FISHES. 69

so often seen in these creatures when fresh, can be
preserved. Hence it is that so few collectors possess
them ; for as there is nothing pleasing to the eye in the
discoloured body of a fish immersed in spirits, they
will only be preserved as objects of curiosity, or for
purely scientific purposes. There are two processes by
which this object may be accomplished: the one, by
drying the specimen; the other, by immersing it in
alcohol.

(64.) Large fish, having tough skins, as the sharks,
and others covered with bony plates or spines, like the
cheloniform genera, are best preserved in a dry state.
For this purpose, the most simple method is to make a
longitudinal cut from the throat to the vent, sufficiently
long to admit the whole of the flesh and bones to be
removed ; or, when practicable, to allow the fish to be
skinned, leaving the bones of the head entire: the in-
side surface may then be anointed with the arsenicai
soap; and after being filled with sand to its natural
dimensions, and gradually dried, the skin retains its
form: a portion of the sand may then be removed, to
render the specimen lighter, and the cavity filled with
cotton. The incision, of course, must be sewed up in
the first instance ; but if the specimen is re-opened to
substitute any softer material for sand, it can be again
sewed up, as the original holes remain. The cheloni-
form fishes, being small, will not require skinning; and
their mailed plates being hard and compact, the form
will be retained even without any stuffing.

(65.) The most useful, as well as the most simple,
method, however, is to preserve all such fish as are of
a moderate size, in spirits. Wide-mouthed bottles with
ground glass stoppers, such as are seen in apothecaries’
shops, are the best vessels for this purpose: but when
these cannot be procured, old pickle bottles, of green
glass, will do very well: these can generally be pro-
cured abroad and at home; and, if well corked, and the
top afterwards covered with bladder, they will travel,
with ordinary care, over the world. The great object

BS

70 CLASSIFICATION OF FISHES,

is to render them, if possible, air-tight, to prevent the
evaporation of the spirits. The best liquor, perhaps,
that can be used, is spirits of wine ; but this is very
expensive, and cannot always be procured. We believe,
however, that the common rum of the West Indies is
equally efficacious; and, indeed, pure spirits of any
sort will answer the same purpose; but Mr. Yarrell
confirms what we have also experienced, that the com-
mon English gin, as sold in the shops, is so much
adulterated, that it is quite unfit for this purpose ; so that
its only qualities seem those of destroying living men and
dead animals. Next, in regard to the specimens, they
should not be so much crowded as to press upon or
against each other, so as to cause injury ; it would even
be adviseable, where many are put into one bottle, that
a little cotton or tow be inserted between them; or each
may be wrapped in cotton, or even sewed up in a thin
calico bag, befcre being put in the spirit. Where it is
intended to form a large collection for transmission to
Europe, and glass bottles are not to be procured, a small
keg may be used as a substitute; and one end should be
left open until a sufficient number of specimens are
procured to fill it: these may be placed in layers,
alternately, with a thin one of cotton or tow, and the
spirit progressively added, as the filling goes on, taking
care that no greater quantity of the liquor is put in at
one time than is sufficient just to cover the specimens ;
by this preeaution they will be preserved compact, the
liquor will have time to insinuate itself into the bodies,
and the replenishing can proceed gradually. When the
cask is full, the head is to be again fixed, and the su~
tures secured outside by pitch, to prevent leakage.
(66.) When any particular notes are made as to the
colours, habits, or other peculiarities, the most effectual
method of identifying the specimens is, by attaching to
them a small label of thin lead, whereon is stamped a
number, agreeing with that of the catalogue. This
will supersede all necessity for taking notes on the
structure, or of such particulars as can be seen in the

LOCALITIES OF FISHES. 71

preserved specimens. As the colours, however, are
entirely changed by the action of the spirits, notes upon
each, when practicable, should always be made. To
those collectors, however, who are draftsmen, we should
recommend the plan pursued by us abroad. A rough
sketch was made from the fish, and all the tints washed
in, to enable any one to make an accurately finished |
drawing afterwards, provided he possessed the specimen
itself, and the finished outline.

(67.) The readiest way of procuring specimens, to
a person not himself a fisherman, but residing in a
maritime town, is by regularly frequenting the fish
markets, where nearly all the edible species found upon
the neighbouring coast will, at one season or other, be
exposed for sale. Particular people, however, have their
local prejudices in regard to such as are considered not
wholesome ; for iHece the collector should inquire of
the fchernen themselves; or, what is much better, let
him go in their boats, and be present at the drawing
up of their nets: numerous species too small for the
market, or not usually eaten, will thus be procured.
Both these plans we pursued, with the greatest success,
at Palermo, Messina, Pernambuco, Bahia, &c.

(68.) In regard to the localities most likely to pro-
duce abundance of species, it may be stated, almest
as a general rule, that the coasts of islands, widely
separated from continents, are the most productive :
hence it is that the tropical archipelagos of the East and
West Indies are much richer in fish than the coasts of
the neighbouring continents ; and to this, also, we attri-
bute, in a great degree, the peculiar abundance, both in
number and variety, found along the coasts of Sicily and
Malta. The Grecian islands, no doubt, are equally
abundant; yet they have never been explored. But of
all the islands bordering the European geographical
range, we apprehend none offer such a splendid field
for the researches of the ichthyologist as the Madeira
islands on one hand, and the Azores on the other; the
latter, more especially, may be called an unexplored pre-

F 4

7 hes CLASSIFICATION OF FISHES.

serve for new discoveries. Situated at such a long dis-
tance from any continent, this cluster of islands must be
a central rallying point for innumerable species during
the breeding season, and to which they make their way
from every point of the compass.*

CHAPS IV:

ON THE SYSTEMATIC ARRANGEMENT OF FISHES.

(69.) On a former occasion we have explained and
fully discussed the nature of those various arrangements,
methods, or systemst, which are used by naturalists for
making known the objects of their study; but as the
former volumes may not be in the hands of all who
possess this, and as it is desirable that each, as far as
possible, should be complete in itself, we deem it ad-
visable, before entering into the details of this chapter,
briefly to recapitulate some of the most important con-
siderations on this subject,—the more so, as much of
novelty will be found in our views of the natural
arrangement of this class, and it may justly be expected
from us to state the grounds upon which we venture
to bring forward an entirely new arrangement.

(70.) There are two modes by which the various
classes of natural objects may be arranged: one is to
view each class or division as isolated, and to construct
a system upon principles applicable to them, and to them
only ; the other is to view them only as parts of one
vast whole, and to construct our arrangement of them

* We have long had an ardent wish to investigate either Madeira,
or the Western Islands, — the latter a bright, although neglected, cluster
of jewels in the diadem of the young and lovely queen of Portugal; and
we take this opportunity of soliciting information from such of our readers
as may be living there, or have the means of rendering a six months’
residence at some one of these islands agreeable in point of society, and
beneficial to our scientific pursuits.

+ Geography and Classification of Animals, p. 122.

SYSTEMATIC ARRANGEMENT OF FISHES. Wes:

upon principles that are not merely applicable to them,
but to all other portions of the animal or vegetable cre-
ation. Now, if the simple question were put to any
reflecting mind, which of these plans was the most phi-
losophical, or the most likely to exhibit the true series
of nature, no one would hesitate to decide upon the last.
This is only a different method of stating the true
nature of artificial and of natural systems. The former,
indeed, cannot be said to be founded on any general or
fixed principles, extending their influence to other
branches of zoological science; for although, in one
sense, each class may be arranged on a principle, yet
that principle is altogether arbitrary. There may be
principles of ichthyology, of ornithology, and of all the
other classes, but there cannot be principles of zoology,
unless the whole of its divisions present a consistent
uniform harmony in their arrangement. Upon this
vantage ground, therefore, the philosophic naturalist
takes his stand; and while he willingly confesses the
advantages, nay, the absolute necessity, of availing him-
self of the artificial mode of arrangement in little known
groups, he feels fully persuaded that the very first im-
perfect glimpse of the natural system should be seized
and adopted, since its very errors will eventually lead to
truth, and accelerate the discovery of those principles
upon which alone zoology can be rendered a science of
demonstration, at least in the opinion of those who have
given laws for the prosecution of the physical sciences,
of which zoology, vast as it is, forms but a small
part.

Gil.) One of the consequences involved in the
law of representation (or that by which one group of
animals represents another group in a totally different
class) is, that the primary divisions of a class are no
longer arbitrary. We advert to this subject more par-
ticularly in the present volume, because, although we
have adopted, in almost every instance, the higher
groups pointed out by our predecessors, we have not
given to them that rank in the class which some have

74: CLASSIFICATION OF FISHES.

assigned to them. We have already shown, in former
volumes, that the primary groups of birds. represent
those of quadrupeds ; and it therefore follows, that if we
can find certain groups of fish which represent both
these, we arrive, by induction, to the sure conviction
that such groups of fish constitute the primary divisions
of the class. The Plectognathes, for instance, are placed
by M. Cuvier as a part of the osseous fishes, when,
even by his own admission, their skeleton is semi-car-
tilaginous. The group, however, is evidently natural ;
and we accordingly preserve it, giving it only a higher
rank. But this change, however, is not the result of
arbitrary opinion: neither is it because the great fathers
of ichthyology did the same ; for they also were guided
in their decision, not by principle, but opinion. It is
because these fishes, besides the peculiarity of their
skeleton, unquestionably represent one of the grand
divisions of the Vertebrata, as well as one of the
primary orders of quadrupeds, of birds, and of reptiles:
and as there is no other division of fishes which
does the same, the Plectognathes are thus proved to be
one of the chief divisions of the class. The same
remark is applicable to the apodal fishes of authors,
where we find all the species destitute of ventral fins ;
but the skeleton is variable. Why, then, is this an order P
The question is thus answered: Cuvier has shown they
are closely connected, and, in fact, pass into the osseous
fishes ; and he also coincides in the opinion of all our
best zoologists, that they likewise make an equally close
approximation to the Vermes, or worms. Now these
apparently opposite relations could not well be true, if
some of the eels had not the bony skeleton of the more
perfect tribes, while in others it was rudimentary, in
order to mark their proximity to the Vermes. The
skeleton is consequently variable ; but in all other respects
the characters of the apodal fishes are constant.

(72.) Preserving the distinction between artificial
and natural systems elsewhere explained*, we shall at

* Geography and Classification of Animals, p. 125.

ARTEDIS SYSTEM. 75

once proceed to the enumeration of those which have
been the most celebrated ; but the curious reader will
find several others in the elaborate history of ich-
thyology drawn up by MM. Cuvier and Valenciennes.
We shall confine ourselves, on the present occasion, to
those of Artedi, Linneus, Cuvier and Valenciennes,
Bonaparte, and Oken.

(73.) One of the primary divisions in the system
of Artep1 (1738), as before mentioned, is composed
of the Cet@, or aquatic IZammalia. The other four
are characterised as follows : —

Tail perpendicular, fins supported by rays.

Skeleton § With bony § Fins with soft rays. MALACOPTERYGII.
bony branchia. ¢ Fins with spined rays. ACANTHOPTERYGII,
; Branchia destitute of bones. BRANCHIOSTEGI.
Skeleton cartilaginous. CHONDROPTERYGII,

The first order, or the Malacopterygii, are arranged in
six divisions, according to the number and position of
the dorsal fin; while the Acanthopterygii are merely di-
vided into those having the head smooth or rough. The
genera are as follows :—

ORDER J. — MALACOPTERYGII.

Syngnathus. Osmerus. Stromateus.
Cobites. Salmo. Gadus,
Cyprinus, Esox. Anarhichas.
Clupea. Echeneis. Murena.
Argentina. Coryphena, Ophidion.
Exocetus. Amodytes. Anableps.
Coregonus. Pleuronectes. Gymnotus.

OrverR II. —- ACANTHOPTERYGII.

Blennius. Sparus. Scorpena.
Gobius. Scizna. Cottus.
Xiphias. Perca. Zeus.
Scomber. Trachinus, Chetodon.
Mugil. Trigla. Gasterosteus.
Labrus.
Orver III. — BRANCHIOSTEGI.

Balistes. = Cyclopterus. Lophius.
Ostracion.

OrDER IV. — CHONDROPTERYGII.
Petromyzon. To these are added in an rend, Hepatus, Capri-
<Acipenser. Appendix the genera scus, Tanta, Pholis,
Squalus. Silurus, Lepturus, Citharus, Atherina, Li-

Raia. Phycis, Cicla, Sphy- paris, and Chelon.

76 CLASSIFICATION OF FISHES.

(74.) The arrangement of Linnzvs, as given in the
twelfth edition of the Systema Nature, differs but little
from that of Artedi. As an artificial system, it is on a
more simple plan than that of any other. We shall
give the reader, by the following table, a much better
idea of the system of the great Swede than by any other
means; and we shall then offer a few observations on
the general nature of the groups. The whole are dis-
tributed into six orders, founded either on the position
of the ventral fins, or, what is much better, on the struc-—
ture of the gills. The orders are named, I. Apodal;
UT. Jugular; II. Thoracic; 1V. Abdominal; V. Bran-
chiostegious ; and VI. Chondropterigious: the contents of
each being as follows : —

TI, Apopat. Ventral fins none.

1. Murena. 5. Anarhichas. 9. Xiphias.
2. Gymnotus. 6. Ammodytes. 10. Sternoptyx.
3. Gymnothorax. 7. Ophidium. 11. Leptocephalus.
4. Trichiurus. 8. Stomateus.

Il. Jucutar. Gills bony, ventral fins placed before the pectoral.
12. Callionymus. 14. Trachinus. 16. Blennius.
13. Uranoscopus. 15. Gadus.

Til Tuoracic. Gills bony, ventral fins placed directly under the thorax.

17. Cepola. 95. Zeus. 99. Perca.

18. Echineis. 94. Pleuronectes. 350. Gasterosteus.
19. Coryphea. 95. Chetodon. 31. Scomber.
20. Gobius. 26. Sparus. 32. Mullus.

21. Cottus. 27. Labrus. 33. Trigla.

29. Scorpena. | 28. Scizna. |

IV. ApDomMINAL. Gills bony, ventral fins placed on the belly behind the

thorax.
34. Cobites. ; 40. Fistularia. 46. Mormyrus.
35. Amia, | 41. Esox. 47. Exoccetus.
36. Silurus. 42, Elops. 48. Polynemus.
37. Teuthis. 43. Argentina. 49. Clupea.
38. Loricaria, 44. Atherina. 50. Cyprinus.
59. Salmo. | 45. Mugil.

V. BRaNncHiostecious. Gills without bones.

51. Ostracion. 54. Syngnathus. 57. Balistes.
52. Tetrodon. 55. Pegasus. 58. Cyclopterus.
53. Diodon. 56. Centriscus. 59, Lophius.

CUVIER'S SYSTEM. v7

VI. CHONDROPTERIGIOUS, Gills and bones cartilaginous.

60. Acipenser. 62. Squalus. 64, Petromyzon.
61. Chimera. 63. Raia.

The above arrangement is so far natural, that it pre-
serves in a distinct group all the cheloniform fishes
(Plectognathes, Cuv.) whose body is encased in a coat of
mail, or covered with hexagonal scales, and which more
especially differ from true fish in having the branchia
concealed and the operculum fixed. This, which we
have shown to form a primary group, is placed next to
the chondropterigious order, where the skeleton becomes
entirely cartilaginous. The apodal order, had it been
restricted to the eel-like fishes, would have corresponded
in its contents to ours; but there seems no reason what-
ever for placing the sword-fish (Xiphias) next to Ophi-
dium, or Leptocephalus next to Sternoptyx. The three
next orders, of Jugular, Thoracic, and Abdominal,
are excellent as artificial groups, enabling the student,
by attention to the single circumstance of the position
of the ventral fins, to ascertain the nomenclature of his
specimens.

(75.) The system of Cuvier, and of his able coad-
jutor VALENCIENNES, will now be more particularly de-
tailed, as given in the last edition of the Régne Animal.
The primary divisions are two:—the first composed
of what are called true or osseous fishes, having the bones
solid; the second are the Chondropterygii, or cartila-
ginous fishes. In these latter the bones of the lower
jaw are supplied by those of the palate.

(76.) Ossrous, or TRUE FIsHES, are divided by our
author, in the first instance, into two most unequal
assemblages :—1. Those in which the gills, or branchia,
are pectinated ; and, 2. those in which they resemble
a series of small tufts. All true fishes come under the
first of these divisions, excepting the genera Syngnathus
and Pegasus of Linneus, which constitute M. Cuvier’s
order Lophobranchii. The first division of osseous
fishes is again divided into two groups of equal dis-

78 CLASSIFICATION OF FISHES.

parity: the one containing the Plectognathes, or our che-
loniform fishes, answering to the Branchiostegi of Artedi;
these having the maxillary bone and the palatine arch
fixed to the cranium: the whole of the remainder, or
the vast multitude of ordinary fishes wherein the upper
jaw is not fixed, form the osseous division. In this
latter, observes M. Cuvier, “ there remains an immense
number of fishes to which no other character can be
applied than those of the external organs of motion.
After an extensive research, I have found that the least
objectionable of these characters is the one employed by
Artedi and Ray, drawn from the nature of the first rays
of the dorsal and anal fins. Thus the ordinary fishes
are divided into (1.) Manacorreryert, in which all the
rays are soft, with the occasional exception of the first
of the dorsal, or of the pectorals: and (2.) AcanTHoP-
TERYGII, in which the first portion of the dorsal, or of
the first dorsal where there are two, is always supported
by spinous rays, some of which are also found in the
anal, and at least one in the ventral fins.”

(77.) The Malacopterygii, or soft-rayed fishes, ‘ may
be conveniently divided,” observes Cuvier, “‘by a re-
gard to the position of their ventral fins, which are
either situated behind the abdomen, as in the Abdo-
minales ; sometimes placed adjoining the shoulder, as in
the Subbrachiati; or altogether wanting, as in the Apodes
(Linn.). It is impossible, however,” as Cuvier thinks,
“to apply this mode of division to the AcaNTHOPTE-
Rye; and their subdivision in any other way than by
that of natural families is a problem that I have hitherto
vainly endeavoured to solve. Fortunately, many of these
families are possessed of characters nearly as exact as
those that could be given to orders.”

(78.) We shall first concentrate the foregoing out-
lines of Cuvier’s system in the following table, and then
proceed to enumerate more particularly the genera com-
prised in the families.

QD
~

CUVIERS SYSTEM.

“uoxAWOIIIT
*snyenbs
‘osuodiow

‘snyyeusig

"uOpor(y
"saqstpeq

"RUBIN TAL
*sn.aojdopaAy
*saqooUOINd[ gq
‘supex)

*xOSTT

*vadni,)
‘oul]eS
*SNAN[IG
*snutidsg
‘sniqe’y
*suuIqoy)
*sniydo'y
“snyeydootydo@
mM UESLED AN
*1aquUosg
‘uopoJwyy
‘sniedg
"BUBIOG
“e[SILT,
SOTINIAL

UUVT Rdg

DLIUIE)
qwoadh J,

*“sautoysopo sg
*SOULOJSOLSE]
"SUDIUOLINS

‘SUHSIY SOONISVIIMLUVD ‘TI

‘sayouvaqoydoT

*soquopoumsy
. *SOULLIPOLIIIS

*soprlougdmnyAl
"Sa OQoosiq
*saqoouoInoe| J
‘soplopry

‘saplolony |

*soptoodnyy
‘sOplOUOMRS
‘SOplolmnytg
‘soplroultdag
“‘soploliqu’y
*SOPLOTQOX)

*saplorydo'y
‘sonbiyquiAqey sargour.tcg

‘sosnyAL
*SOplLotaquLoag
“sopPLOUOPOJI YO
‘saploandg
“SOPLOUdLOG
*BYVOLO]T voongy
“SOTLOTAL
*souloUuA[Og
*SOPION19,]

*SOULONT
YOU

‘sapodp °S

t "TDLYIUDAQ-QNsy °G'

5

{ “qpUurmopagy ‘|

|
J
\
| = = SADAUTDLAOWLINVOW

‘SHHSI,J SAOASSE 'T

‘SUDAUALAOOVIVIA f

5

TIMYS
dY} 03 poxy
Mel gaddq

"29.
Met saoddq

*syjny Ul PIYOUPIG, °S

*pozyeuryaod
S][Is 10 SeiqouRrg °*T

80

CLASSIFICATION OF FISHES.

Division I. — Osszous FisHEs.
* Pectinibranchia.

A. The upper jaw free.

I. ACANTHOPTERYGES.
1. Family. PERCOIDEs.
* With two dorsal fins ; no canines.

Perca Linn.
Labrax Cuv.
Lates Cuv.
Centropomus Lac.
Grammistes Cuv.
Aspro Cuv.
Huro.
Etelis.
Niphon.
Enoplosus Lac.
Diploprion Kwhl.
Apogon Lac.
Cheilodipterus Lac.
Pomotomus Risso.
Ambassis Comm.
Lucioperca Cuv.

* * With one dorsal fin and canines.

Serranus.
Anthias Bloch.
Merra.

Plectropoma.

Diacope Cuv.

Mesoprion Cwv.

Acerina Cuv.

Rypticus Cuv.

Polyprion Cuv.

Centropristis Cuv.

Gristes Cuv.

Cirrhites Comm.

Chironemus Cuv,

Pomotis Cuv.

Centrarchis Czv.

Priacanthus Cuwv.

Dules Cuv.

Therapon Cwv.
Datnia.

Pelotes Cuv.

Helotis Cuv.

*** With two dorsal fins, and
less than six branchial rays.
Trichodon Stedler.
Sillago Cuv.

**** More than seven branchial
rays.
Holocentrum Bloch.
Myripristis Cuv.
Beryx Cuv.
Trachichtys Shaw.

xxx * With jugular ventrals.
Trachinus Linz,
Percis Bloch.

Pinguipes Cuv.
Percophis Cuv.
Uranoscopus Lznn.

* **** * Ventral fins behind the

pectoral.

Polynemus Linn.
Sphyrena Artedi.
Paralepis Cuv.
Mullus Linn.

Q. Family. TRIGLIDE= (Sw.).

Trigla Linn.
Prionotus Lac.
Peristedion Lac.
Dactylopterus Lac.

Cephalocanthes Lac.

Cottus Linn.
Aspidophorus.

Hemitripterus Cuv.

Hemilepidotus Cuv.

Platycephalus Bloch.

Scorpena Lznn.
Teenianotus Cuv.
Sebastes Cuv.

Pterois Czuv.

Blepsias Cuv.

Apistes Cuv.

Agriopus Cuv,

Pelor Cuv.

Synancea Bloch.

Monocentris Bloch.

Gasterosteus Linn.

Oreosoma Cuv.

3. Family. ScrENoIDES (Cuv.).

Sciena Lin.
Otolithus Cuv.
Ancylodon Cuv.
Corvina Cuv.
Johnius Bloch.
Umbrina Cuwv.
Lonchurus Bloch.
Pogonias Lac.

Eques Biock.

** Dorsal fin one.

Hemulon Cuv.
Prestipoma Cuv.
Diagramma Cuuv.
Lobotes Cuv.
Cheilodactylus Lac.
Scolopsides Cuv.
Micropterus Lac.
Amphiprion Bloch.
Premnas Cuv.
Pomocentrius Lac.
Dascyllus Cuv.

2)

a

CUVIER'S SYSTEM. 81

Glyphisodon Lac.
Heliasus.

4, Family. SPAROIDES (Cwv.).
Sargus Cuv.
Chrysophris Cuv.
Pagrus Cuv.
Pagellus Cuv.
Dentex Cuv.
Cantharus Cuv.
Boops Cuv.
Oblada Cuv.

5. Family.
Mena Cuv.
Smaris Cuv.
Cesio Lac.
Gerres Cuv.

MENIDES (Cwv.).

Family. SQuUAMIPENNES (Che-
todon Linn.).
Chetodon Linn.
Chelmon Cwv.
Heniochus Cuv.
Ephippus Cuv.
Taurichtes Cuv.
Holocanthus Lac.
Pomocanthus Lac.
Platax Cuv.
Psettus Comm.
Pimelepterus Lac.
Dipterodon Cuv.
Brama Bloch.
Pempheris Cuv.
Toxotes Cuv.

Family. ScoMBEROIDES (Cuv.).
Scomber Linn.
Thynnus Cuv,
Orcynus Cuv.
Auxis Cuv.
Sarda Cuv.
Cybium Cuv.
Thyrsites Cuv.
Gempylus Cuv.
Xiphias Linn.
‘Tetrapterus Raf.
Makaira Lac.
Histiophorus Lac.
Centronotus Lac.
Naucrates faf.
Elacates Cuv.
Lichia Cuv.
Trachinotus Lac.
Rynchobdella loch.
Macrognathus Lac.
Mastacembelus Grom.
Notocanthus Bloch.
Seriola Cuv,
Nomeus Cuv.
Temnodon Cuuv.
Caranx Cuv.
Citula.
Vomer Cuv.
Olistus Cuv.

VOL. I.

Scyris Cuv.
Blepharis Cuv.
Gallus Cuv.
Argyreosus Cuv.
Zeus Linn.
Capros Cuw.
Lampris Retzius.
Equula Cuv.
Mene Lae.
Stromateus Linn.
Pempla Cuv.
Peprilus Cuv.
Luvarus af.
Seserinus Cuv.
Kurtus Bloch.
Coryphena Linn.
Caranxomorus Zac.
Centrolophus Lae.
Astrodermus Zon,
Pteracles Gron.

8. Family. TN10iDEs (Cuwv.).

t Mouth lengthened; teeth strong.

Lepidopus Gouan.f
Trichiurus Linn.
Gymnetrus Bloch.
Stylephorus Shaw.

** Mouth short.
Cepola Linn.
Lophotes Giorna.

9. Family. THrEuriprs (Cwv.).

Siganus Forsk.
Acanthurus Lac.
Prionurus Lac.
Naseus Comm.
Axinurus Cuv.
Priodon Cuv.

10. Family. PHARYNGIENS LABY-=
RYNTHIFORMES.

Anabas Cuv.
Polyacanthus Kuhl.
Macropodus Lac.
Helostoma Kzhi.
Osphromenus Comm.
Trichopodus “ac.
Spirobranchus Cuv.
Ophicephalus &loch. .

11. Family. Mueiboipes.

Mugil Linn.
Tetragonurus Risso.
Atherina Linn.

12. Family. GoproipEs.

Blennius Linn.
Myxodes Cuv.
Pholis Cuv.
Salarias Cuv.
Clinus Cuv.
Cirrhribaba Cuv.

t
t

Gunellus Cuv.
Opistognathus Cuz.
Zoarcus Cuv.
Anarhichas Arfedz.
Gobius Linn.
Gobioides Lac.
Tenioides Lac.
Periophthalmus Sch.
Eleotris Gron.
Callionymus Linn. *
Trichonotus Sek.
Comepheorus Lae.
Platypterus Kuhl.
Chirus Sfeller.

13. Family. PEDICULATZ.

Lophius Lznn.
Chironectes Cuv.
Malthe Cuv.

Batrachus Cuv.

14. Family. LaBroipes (Cuv.).

Labrus Linn.
Labrus.
Cheilinus Zac.
Lachnolaimus Cuv
Julis Cur.
Anampsis Cuv.
Crenilabrus Cuv.
. Coricus Cuv.
Epibulus Cuv.
Clepticus Cuv.
Gomphosus Cuv.
Xirichythys Cuz.
Chremis Cuv.
Cychla Block.
Plesiops Cuv.
Malacanthus Cuv.
Searus Linn.
Calliodon Cuv.
Odax Cuv.

15. Family. FIsTULARID= (Sch.).

Fistularia Linn.
Aulostomus Lac.

Centriscus Linn.
Amphisile Klein.

II. MALACOPTERYGII.

Order 1. Malacopterygiz abdomi-
nales.
1. Family. CYPRINIDZ.
Cyprinus Linn.
Cyprinus Cuv.
Barbus Cuuv.
Gobio Cuv-
Tinca Cuv.
Cirrhinus Cuz.
Abramis Cuz.
Labeo Cuv. ~
Catastomus Le Sueur.

CLASSIFICATION OF FISHES.

i

Leuciscus Klein.

Gonorynchus Gyron.
Cobites Linn.

Anableps Artedi.

Peecilia Sch.

Lebias Cuv.

Fundulus Lac.

Molinesia Le Sueur.

Cyprinodon Lae.

2. Family. Esocrs (Cuv.).

-Esox Linn.

Esox.
Galaxias Cuv.
Alepocephalus R7sso.
Microstoma Cuv.
Stomias Cuv.
Chauliodus Sch.
Salanx Cuu.
Belone Cuv.
Sairis Raf.
Hemiramphus Cuwv.
Exsocetus Linn.
Mormyrus Linn.

5. Family. Srmuriwz (Cuv.).

Silurus Linn.
Schilbe Czz.
Mystus 4drtedi.
Pimelodus Lac.
Bagrus Cuv.
Synodontis Cuv.
Ageniosus Lac.
Doras Lac.
Heterobranchus Geoff:
Clarias Gron.
Plotosus Lac.
Callichthys Linn.
Malapterurus Lac.
Aspredo Linn.
Loricaria Linn.
Hypostomus Lae.

4, Family. SaLMonibes (Cuv.).

Salmo Linn.
Osmerus Artedi.
Mallotus Cuv.
Thymallus Cuv.
Coregonus Cuv.
Argentina Cuu.
Curimata Cuv.
Anostomus Cuv.
Gasteropelecus Bloch.
Piabucus Cuv.
Serrasalmo Lac.
Tetragonopterus Artedi.
Chalceus Cuv.
Myletes Cuv.
Hydrocyon Cuz.
Citharinus Cuv.
Saurus Czv.
Scopelus Cuv.
Aulopus Cuv.

Sternoptyx Herm.

CUVIER S SYSTEM.

5. Family. CLuPez.

Clupea Linn.

Clupea.

Alosa Cuv.
Chateessus Cuv.
Gnathobolus Sch.
Pristigaster Cuv.
Notopterus Lac.
Engraulis Cww.

Thryssa.
Megalops Lac.
Hlops Linn.
Butirinus Comm.
Chirocentrus Cuv.
Hyodon Le Sueur.
Erythrinus Gron.
Amia Linn.

Sudis Cuv.
“Osteoglossum Vana.
Lepisosteus Lac.
Polypterus Geoff

Order 2. Ventral fin beneath the

pectoral.
6. Family. GaADITES.

Gadus Linn.
Morrhua Cuv.
Merlangus Cuv.
Merlucius Cuv.
Lota Cuv.
Motella Cuv.
Brosmius Cuv.
Brotula Cuv.
Phycis Artedi.
Raniceps Cuv.

Lepidoleprus fésso.

7. Family. PLEURONECTIDA.

Pleuronectes Linn.
Platessa Cuv.
Hippoglossus Cuwv.
Rhombus Czv.

Solea Cuv.
Monochirus Cuv.
Achirus Lac.
Plagusia Cuv.

8. Family. Discogo.t.

Lepidogaster Gouan.
Gobiesox Lac.

Cyclopterus Linn.
Lumpus Cuv.
Liparis Artedi.

Echeneis Linn.

Order 3. Ventral fing wanting.
9, Family. ANGUILLIFORMES.

Murena Linn.
Anguilla Thunb. ,
Murena.
Ophisurus Lac.
Gymnothorax Bloch.
Sphagebranchus Bloch.
Apterichtes Dum.
Monopterus Comm.
Synbranchus Bloch.
Alabes.
Ophiognathus Harwood.
Gymnotus Linn.
Carapus Cwv.
Sternarchus Sch.
Gymnarchus Cuv.
Leptocephalus Pennané.
Ophidium Linn.
Fierasfer Cuv.
Ammodytes Linn.

Order 4. Lophobranches.
10. Family. SYNGNATHIDZ.

Syngnathus Linn.
Hippocampus -Cuv.
Solenostomus.

Pegasus Linn.

Order 5. Plectognathes.

11. Family. GyMNODONTEs.

Diodon Linn.
Tetraodon Linn.
Cephalus Sch.
Triodon Cuv.

‘12, Family. ScHERODERMEs,

Balistes Linn.
Monocanthus Czuv,
Aluterus Cuv.
Triacanthus Cuv,

Ostracion Linn.

Division I]. —Cartinaginous Fisnes.

Order 1. Sturzones.

Acipenser Linn.

Spatularia Shaw.

Chimera Linn.
Callorhynchus,

Order 2. Chondropterugii.
1, Family. Sracurr,

Squalus Linn.

Scyllium Cz.
Carcharias Raf.
Lamna Cuv. -
Galeus Cuv.

84 CLASSIFICATION OF FISHES.

Raia Linn.
Trygon Aniig. ,
Anacanthus ELhren.
Myliobatis Dum.
Rhinoptera Kuhl.
Cephaloptera Dum.

Mustelus Czv.

Notidanus Cuv.

Selache Cuz.
Cestracion Cuv.
Spinax Cuv. }
Centrina Cuz.
Scymnus Cuwv.

Zygena Antigq.

Squatina Dum.

Pristis Lath.

Raia Linn.
Rhinobatus Schn. |
Rhina Schn._ |
Torpedo Antig. |

2. Family. Sucroris.

Petromyzon Linn.
Myxine Linn.

Heptatremus Dum.

Gastrobranchus Bloch.
Ammoceetes Dum.

(79.) Having already offered a few general remarks on
the foundations of this system, we shall only advert,
in this place, to some objections regarding the nomen-
clature of certain groups. M. Cuvier, in making his
divisions of the Linnean genera, generally places the
original name for designating the group ; but in several
instances he gives to every one of his divisions a new
name; so that, although it seems at first as if the Lin-
nean denomination was preserved, it is, in fact, com-
pletely done away with, and only remains an indication
of a genus not adopted. One instance of this will suf_
fice to explain our meaning. The well known genus
Gadus is divided, very properly, into several others,
but is not retained or restricted to any one ; so that, if we
adopt ali M.Cuvier’s new generic names, we must totally
reject, as such, the genus Gadus: no such group, conse-
quently, is to be found in the Régne Animal. As this,
we presume, never could have been intended by the
illustrious author, we have retained this and other
original names to that division of a Linnean genus
which seems to us the most typical. M. Cuvier’s generic
names, in general, are well and harmoniously com-.
pounded; but many have no ciaim on the score of
priority ; anda few others, as Vomer, Saurus, Barbus,
&e., are founded on principles which he himself has
rejected in all other instances: these blemishes have
therefore been corrected, and that name adopted which
has the priority.*

* Another practice has recently been introduced by one or two foreign
naturalists of some eminence, who do not appear to be aware of the con-
sequences to which it leads; we think it, however, almost as objectionable

———

BONAPARTE’S SYSTEM. 85

~ (80.) The prince of Musignano’s arrangement of this
class is the most recent.* As we think it contains some
decided improvements upon M. Cuvier’s, we shall lay
the following abstract of it before the reader, particu-
larly as we know that it is the result of no inconsider-
able share of knowledge and of attention to these animals
in their living state. It is not so much in the primary
divisions (which, like those of M. Cuvier, are entirely
arbitrary), as in the series in which some of the genera
are placed, that we conceive these improvements will be
found. ‘The number of species which the noble author
believes to be comprised in each genus is added.

I. Order. — ACANTHOPTERYGII.

: Diacope Cuv. India 3

1. Family. PErcwz. Mesoprion Cwv. Tropics 48

1. Percini. Acerina Cuv. Europ. rivers 3

; Polyprion Cwv. Warm seas 1

Perea . Temperate rivers 11 Pentaceros Cuv. Africa i
Labrax Temp. seas 7 Centropristis Cuv. Warm seas 10
Lates Africa, India 3 Grystes + Am. rivers 2
Centropomus America 1 Aprion Cuv. i)
Lucioperca Cuv. Black Sea 4 Rypticus Cuv. Am. 2
Huro Cuv. Lake Huron 1 Apsilus Cuv. Atlantic 1
Etelis Cuv. Europe 1 Cirrhites Comm. India 6
Niphon Cuwv. Java 1 Chironemus Cuv. Australia 1
Enoplosus Lac. Australia 1 Pomotis Cuv. Am. rivers 8
Diploprion Kuhl Java 1 Centrarchus Czv. Ditto 7
Apogon Lac. Warm seas 22 Bryttus Cuv. Ditto 3
Cheilodipterus Zac. India 3 Priacanthus Cuv. Atlantic 15
Pomatomus Risso Medit. 1 Dulichethys t Bon. Warm seas 11
Ambassis Comm. 12 Therapon - Cuv. Red Sea 10
Priopis Kuhl Java 1 Datnia \ Cuwv. India 3
Aspro Cuv. Europ. rivers 2 Pelates Cuv: Ditto 3
'Grammistes Cuv. India 2 Helotes Cuv. Australia 3
Anthias Bon. Ind., Am., Eur. 7 Nandus Cuv. Ind. rivers 1
Serranus Cuv. All seas 22 Trichodon Cuv. Arctic 1
Merrus Cuv. Ditto 98 Sillago Cuv. India 7
Plectropoma Cuy. Ind.,Am. 14 Rhynchithys Cuv. Ditto 1

as the former, although on a different ground. Ifa genus is to be divided,
the divider not only affixes his own name as founder of the new group,
but he does the same to the original one; so that, in fact, the merit of the
original founder of the genus is completely cancelled, and the generic name,
although retained, seems as if it originated solely in him who divides it.
If this is once allowed, there is no calculating the confusion, not to say the
injustice, that will follow: the fame or reputation of no one, who has de-
fined and named a group, is safe; since it may be cancelled by the very
first who thinks it necessary to divide it. On this principle, the genus
Gadus, in our synopsis, would be ours, not Linneus’s ; and Teuih7s would
be recorded in our systems, not as an effective genus made by Linneus,
but by Bonaparte. ;

* Saggio di Una Distribuzione Metodica degli Animali Vertebrati; di
C. L. Bonaparte, Principe di Musignano, Rema, 1831.

+ Micropterus Lac. £ Dules Cuv. § Coius Buchan.

G3

86

CLASSIFICATION
Heclocentrum Block Warmseas .
Myripristis Cuv. Ditto 11
Holocentrum C. Warmseas 19
Beryx Cuv. Australia 1
Trachicthys Shaw Ditto 1
2. Trachinini.
Trachinus Linn. Medit. 4
Percis Bloch India 12
Aphritis Cuv. Atlantic 1
Pinguipes Cuv. Brazil 1
Percophis Cuv. Ditto 1
Bovichthus Cuv. Chili 1
Uranoscopus Lizn. General 13
3. Polynemini.
Polynemus Gyon. Warm seas 15
Aplodactylus Cuv. Chili 1
2. Family. SPHYRNIDZ.
Sphyrena Lac. General 11
Paralepis Risso Medit. 4
3. Family. MuvLLiIp#.
Mullus Linn.
Mullus Cuv. Europe 2
Upeneus Cuv. Warm seas 40
4, Family. TRiGLmID2£.
1. Triglinz.
Trigla Lznn. General 15
Prionotes America 4
Peristidion Lac. Medit. 1
Dactylopterus Warm seas 2
Cephalacanthus Ditto 1
2. Cottint.
Cottus Linn. — Atlantic 19
Aspidophorus Cuv. Ditto 9
Platycephalus Block; India 21
Hoplichthys Cuv. Japan 1
Bembras Cuv. Ditto 1
Hemitripterus Cuv. Atlantic 1
’ 3. Scorpenini. —
Hemilepidotus Cuv. Atlantic 1
Scorpena Linn. General 19
Sebastes Cuv. Europe 10
Pterois Cuv. India 7
Tenianotis Lac. 1
Blepsias Cuv. Pacific 2
Agriopus Cuv. Atlantic 3
Apistus Cuuv. India 15
Minous Cuv. Ditto 2

OF FISHES.

Pelor Cuv. India
Synanceia Bloch Ditto

4. Gasterosteinz.

Monocentris Cuv. Japan
Hoplostethus Cuv. Medit.
Gasterosteus Linn. Arctic
Oreosoma Cuv. Atlantic

5. Family. ScuNInNI.
Sciena Linn.

Ctolithus Cuv. Ditto

Ancylodon Cuv. Ditto
Corvina Cuv. Ditto
Johnius Bloch Ditto
Leiostomus Lac. Atlantic
Larimus Cuv. Ditto
Nebris Cuv. Ditto
Lepipterus Cuv. Ditto
Boridia Cuv. Ditto
Conodon Cuv. Ditto
Eleginus Cuv. Ditto
Eques Bloch Ditto
Umbrina Cuv. India, &c.
Lonchurus Bloch
Pogonathus Bon.* i
Micropogonias Ben. Atlantic
Hemulon Cuv.
Pristipoma Cuwv. India
Diagramma Cuv. Ditto
Lobotes Cuv. Ditto
Scalopsides Cuv. Ditto
Cheilodactylus Cuo. Ditto
Latilus Cuv. Ditto
Macquaria Cuv. Australia

2. Pomocentrini.

Amphiprion Bloch India
Premnas Cuv. Ditto
Pomacentrus Cuv. Ditto
Dascyllus Cuv. India

Glyphisodon Lac. Atlantic
Etroplus Cuv. India
Heliases Cuv. Am., Ind.

6. Family. SpPARIDZ.

1. Sparini.
Sargus Klein Warm seas
Charax Risso Medit.
Sparus Linn., Bon. Ditto
Pagrus Cuv. Warm seas
Pagellus Cuv. Ditto

2. Denticini.
Dentex Cuv. Warm seas
Pentapus Cuv. India

Warm lat. °

1
i
6
i

eo ed
©9 EO ED CO GO ee a et bt 1D BD G) ST LO Go 09

nD G9

ay
rtd O10 BOO

—\
to

14

1
92
12
10

Q7
8

* Pogonias of Cuvier, &c.; but this name cannot be retained, having

long been used in ornithology. ;

BONAPARTE'S SYSTEM.

3. Lethrininz.

Lethrinus Cuv. Warm seas

4. Cantharini.

44,

Cantharus Cuv. Warm seas 12
5. Obladini.
Box Cuv. Warm seas 4
Oblada Czv. Med., Aust. 2
Cantharus Cuv. Medit. 1
Crenidens Cuv. Red Sea 1
7. Family. M#nipm.
1. Menini.
Mena Cuv. - Medit. 4
Smaris Cuv. Atlantic 10
2. Cesionini.
Cesio Comm. India 9
Gerres Cuv. Pacif., Atlant. 18
Aphareus Cuv. India 2
8. Family. CH£TODONTINI.
Chetodon Linn. Torrid seas 61
Chelmon Cwv. India 2
Heniochus Cuv. Ditto 5
Zanclus Cuv. Ditto 2
Ephippus Cuv. Am., India 4
Drepanis Cuv. India 2
Scatophagus Cuy. Ditto 5
Taurichthys Cuv. Ditto 5
Holocanthus Lac. Ditto 23
Pomacanthus Cuv. Am. 6
Platax Cuv. India 14
Psettus Comm. India 3
2. Pimeleptini.
Pimelepterus Lac. Pacifie 1
Dipterodon Cuz. Cape
Scorpis Cuwv. Australia
Brama Bloch India
Pempheris Cuv. Pacific
Toxotes Cuv. India
9. Family. ScomBpripz&.
1. Scombrine.
Scomber Linn. General 12
Thynnus Cuv. Ditto il
Auxis Cuv. Ditto 3
Pelamis Cuv. Warmseas 2
Cybium Cuv. India 16
Thyrsites Cuv. Warm seas 3
Gempylus Cuv. Atlantic 4

87
2. Trichiurini.
Lepidopus Gowan Atlantic 1
Trichiurus Linn. India, &c. 3
3. Xiphiadini.

Xiphias Cuv. Medit. 1

Histiophorus Zac. Warm seas 3
Tetrapterus Lac. Ditto 2
Makaira Lac. Atlantic 1

4. Centronotint.

Naucrates Raf. Warm seas 4

Elacates Cuv. Ditto 5

Centronotus Lac.

Lichia Cuv. Medit. 4

Chorinemus Cuv. Pacific 16

Trachinotus Cuv. India 23

Apolectus Cuv. Ditto 1

Macrognathus : Lac.

Rhynchobdella Cuv. Asia 1

Mastacembelus Cuv. Ditto 8

Notocanthus Cuv. Atlantic if
5. Carancini.

Caranx Lac. Universal 16
Carangus Cuy. India, &c. 20
Citula Cuv. Medit. 4

6. Vomerinz.

Seriola Cuv. Universal 7

Nomeus Cuv. America 1

Temnodon Cuv. Pacific 2

Olistus Cuv. India 1

Scyris Cuv. Egypt 2

Blepharis Cuv. America 4

Alectris Raf.* India, Am. 4

Argyneosus Lac. America 6

Vomer Cuv. Ditto 10

7. Zeini.

Zeus Linn. Medit., Atlant. 2

Capros Lac. Medit. 1

Lampris. Retz + Ditto IL

Equula Cuv. India 15

Mene Lac. Ditto 1

8. Coryphenini.

Stomateus Linn. Medit., Atl.
Peprilus Czv. America
Luvarus Raf. Atlantic
Seserinus Cu, Medit.
Kurtus Bloch India
Coryphena Linn, Atlantic

Caranxomorus Lac. Atlan.

Centrolophus Lac. Ditto
Pteraclis Gron. t America

1

DS 09 4 rt ot

1

mp 09

* Gallus Lacepede.

} Chrysotosus Lacepede.

t Pteridivm of Scopoli, and Oligopodus of Lacepede.
G 4

88 CLASSIFICATION OF FISHES.

10. Family. CEpoLipz.

Gymnetrus B/. India, Med., &c. 9
Stylephorus Skaw Trop. Am. 1
Cepola Zinn. Medit., Pacif. 3
Lophotes Gzorna Medit. 1
11. Family. TEuraHipiIpz.
Siganus Fors India 20

Teuthis Linn. Warm seas 25
Acanthurus Zac.
Scopas Bon.*
Ctenodon Bon.

Prionurus Lac. Warm seas 2

Naseus Cov2. Ditto 11
Axinurus Cuv. India 1
Priodontichtys Bon. + Ditto 1

12. Family. OPpxHiocEPHALIDZ.
1. Anabatini.
Anabas Cuv. Fresh waters, Asia

Helostoma Kuhl Ditto
Polycanthus Kuki Ditto
Colisa Cuv. Ganges

Macropodus Lac.
Osphromenus Com. Ind. rivers
Spirobranchus Cuv. Af. rivers

bet C9 09 (© CO eS

2. Ophiocephatini.
Ophiocephalus #2. Ind. rivers 20

13. Family. Mueims. _
1. Mugiilini.
Mugil Zinn. General 30

2. Tetragonurini.
Tetragonurus Risso Medit. 1
3. Atherinini.

Atherina Linn.
Aphia sso

Medit. 20
Ditto. 1

14. Family. Gopipz.

1, Blennini.

Blennius Zinz.
Pholis Artedz

General 25

Tripterygion Medit. 1
Myxodes Cuv. India 5
Salarias Cuv. Ditto 9
Clinus Cuv. Medit. 16
Cirrhibarbus Cuv. India 1
Murenoides Lac.~ Atlantic 38
Opistognathus Cuv. India 1
Zoarces Cuv. Medit. 5
Anarhichas Artedé Atlantic 3

2. Gobini.

Gobius Linn.
Gobioides Lac.

General 50
India 6

Tenioides Lac. Ditto 1
Periopthalmus Sch. Ditto 5
Eleotris Gron.t General 10
3. Callionymini.
Callionymus Cuv. Medit. 18
Trichonotus Sch. India 4
Comephorus Lac, Baikal 1
Platypterus Kuh India 2
Chirus Szedler Kamtch. 7

15. Family. Lopxipz.
Lophius Zinn. Med., Atlant. 4

Antennarius 4. Tropics 16
Malthe Cuv. * Ditto 8
Batrachus Bloch Pacific 12
16. Family. LaBrips.
1. Labrini.
Labrus Linz. General 40

Crenilabrus Cuv. Warm seas 90

Cheilinus Zac. India 12
Lachnolaimus Cuv. Am. 4
gulus Cuv. Warm seas 40
Anampses Cuv. India 2
Coricus Cuv. Medit. 3
Epibolus Cuv. India 1
Clepticus Cuv. W. Indies 1
Elops Co. || India 5

Xirichthys Cuv. Warm seas 12

9. Chromidine.

Malacanthus Cuv. India 3
Chromis Cuv. Nile 10
Cychla Bloch India 16
Plesiops Cuv. Ditto 4

3. Scarini.
Scarus Linn. Warm seas 29

Calliodon Cuv. India 7
Odax Cuv. Ditto 4

17. Family. FisTULARINI. —

1. Fistularinz.
Fistularia Linz. Warm seas 5

Aulostomus Lac. India 1
2. Centriscini.
Centriscus Linn. Medit.
Amphisile Klein. India 1
8

* * Scopus is already used in ornithology.

+ Priodon of Cuvier.
§ Chironectes Cuv.

~ Prochilus Cuv.
|| Gomphosus Lac.

b]
BONAPARTE S SYSTEM. 89

II. Order. — MALAcOPTERYGII.

18. Family. CyPRinipm&,
1. Cyprinine.

Fresh waters

Cyprinus Linn.
General 15

Cyprinus Cuv.
Barbus Cuv. Ditto 24
Gobio Cuv. Europe, Asia 6
Tinca Cuv. Eur., As., Af. 4
Cirrhinus Cuv. India 4
Abramis Cw. Eur., Asia 10
Labeo Cuv. Af., Am., As. 7
Catostomus Le Sueur Am. 20
Leuciscus Klein General 151

Chela Hamilton India 3
Gonorhynchus Gyon. Africa 1
Cobites Linn. EB. Asia 16

2. Anableptini.
Anableps Bloch Am. tivers 1

8. Peoecilini.
Peecilia Sch. Am, rivers 6
TLebias Cuv. Med., Afr. 6
Fundulus Lac. America 5
Molinesia Le Sueur Ditto 3
Cyprinodon Lac. Eur., Am. 4

19. Family. Esocipz.
1. Esocinz.

Esox Linn. Eur., Am. 4
Galaxias Cuv. America 2
Alepocephalus Fisso Medit. 1
Microstoma Cuv. Ditto 1
Stomias Cuv. Ditto 2
Chauliodus Sch. Ditto 1
Salanx Cuv. Atlantic 1
Ramphistoma haf. General 15
Scombrisox Lac. Medit. 3
Hemiramphus Cuv. Tropics 14
2. Exocetini.
Exocetus Linn. Tropics, &c. 12

3. Mormyrini.
Mormyrus Linn. Af. rivers 16

20. Family. SiLuripZ.
1. Silurini.

Silurus Linn. Trop. rivers
1 Silurus Artedz Europe 9
Schilbe Cuv. Nile 5
Mystus Artedi* Am. riv. 77
Pimelodus Lac.

Bagrus Nile, India 24

Sorubium Spéx ‘America 7
Hypopthalmus Spix Ditto 2
Pimelodus Cuv. Ditto 40
Synodontis Nile 3
Argeniosus Lac. <Ganges 3
Heterobranchus Geoff:
Clarias Gron. India 5
Heterobranchus Geof; Do. 2
Plotosus Lac. Asia 2
Platystacus Bloch Ditto 2
Plotosus Lae. Ditto 2
Callichthys Linn. Ditto 2
Malapterurus Lac. Africa 2
2. Loracarini.
Aspredo Linn. America 5
Loricaria Linn. Ditto
Loricaria Ditto 4
Hypostomus Lac. Ditto 2

la Family. SALMONIDA,
1. Salmonini.

Salmo ZLznz. General 50
1

Osmerus Artedi Eur., Am.

Mallatus Cuv. Atlantic 1
Thymallus Czv. Eur., Am. 3
Coregonus Cuv. Ditto 15
Argentina Linn. Medit. 1
Curimatus Cuv. America 10
Anastomus Cwuv. Ditto

Gasteropelicus Bloch India 1
Characinus Artedi, America 10
Serrasalmo Lac. Ditto 5
Tetragonopterus Art. Ditto 3
Chalceus Cuv. Ditto 3
Myletes Cwv. Am., Nile 6

2. Aulopodinz.

Hydrocyon Cuv. Trop. rivers 9
Citharinus Cuv. Nile 3

Saurus Cuv.t General 20
Scopilus Cuv. Medit. 3
Aulopus Cuv. Ditto 1
Sternoptyx Ditto 2

22. Family. CiLUPEIDm.
1. Clupeini.
Clupea Linn.

Clupea General 12
Alosa Cuv. Ditto 20
Chetoessus Cuv. Ind., Am. 8
Pomolobus Raf. Ohio 2
Dorosoma Raf. Ohio 1
Notemigonus Raf. Ditto 2
Odontognathus Lac. America 1

* Doras Lac. + Laurida Aristotle, and including Harpodon Le Sueur.

90

Pristogaster Cuv. Atlantic 4
Notopterus Lac. Asia 1
Engraulis Cuv.

Engraulis General 12

Thryssa India 4

Alpismaris Medit. 2
Megalops Lac. Am., Asia 2
Elops Linn. Ind., Am. 4
Butirinus Comz. Ditto 5
Chirocentrus Cuv. India 1
Hyodon Le Sueur Am. 2

2. Amini.

Erythricthys Bon.* Warmrivers 6
Amia Linn. America 1
Sudis Cuv. Amer., Afr. 3
Ostecglossum Vand. Brazil 1
Lepisosteus Lac. America 7.
Polypterus Geoff Nile 2

(Tribe 2. Subbrachianii.)

23. Family. Gapbip®.
1. Gadini.

Gadus Linn.

Morrhua Cuv. Atlantic, &c. 12

Merlangus Cuwv. Ditto 4

Merlucius Cuz. Ditto 3
Lota Cuz. Medit. 5
Motella Cuv. Atlantic 5
Brosmius Cuv, Ditto 2
Brotula Cuz. W. Indies 1
Physis Artedi Medit., &c. 4
Raniceps Cuv. Atlantic 2

2. Macrourini.

Macrourus Bloch Medit. 3

24. Family. PLEURONECTID#.
Pleuronectus, Linz.
Platessa Cuv.
Hippoglossus Cuz.
India, Europe 10

Atlantic 10

Bothus Raf. General 20
Solea Cuv.
Solea Cuz. Ditto 20
Monochir Cuz. Ditto 7
Achirus Zac. India 4
Plagusia Br. Ditto 6

CLASSIFICATION OF FISHES,

ne

|
|

25. Family. CycLopreripx. ~

Lepadogaster Gouan General 11 ©
Gobiesox Lac. Med., Atl. 4

Cyclopterus Linn. Atlantic 8
Liparis Artedi Ditto 4

26. Family. ECHENEIDIOR.
Echeneis Linn. 4

{Tribe 3. Apodes.)
Q7. Family,

Ophidium Linn.
Fierasfer Cun.
Ammodytes Linn.
Leptocephalus Gron.

OPpHIDID#.

Medit., &c.
Ditto
Ditto
Ditto

OM 9 tS Or

MuRENID2.
1. Gymnotini.

Eremophilus Humb. Am. rivers

Gymnarchus Cuv. Nile

Gymnotus Linz. America -
Carapus Cuv. Ditto
Apternarchus Schn. Ditto

28. Family.

tO G1 tO A

92. Murenini.

Saccopharynx Mitcheli+ Amer. 2
Murena Antig. General 20
Anguilla dnzigq. Ditto 6
Conger Cuv. Ditto 10
Ophisurus Lac. Ditto 12

8. Apterichthini.

Sphagebranchus Bloch | India 6
Apterichthys Dum. Medit. 2
Synbranchus Bloch India 5
-Alabes Cuv. India 1

99, Family. SYNGNATHIDE.
y

Syngnathus Linz. General 25

Typhle Raf.

Sephostoma Raf.

Nerophis Raf.
Hippocampus Cuv.
Solenostomus Lac.

Pegasus Linz.

Warmseas 12
India 1
India 5

Ill. Order.— PiecroegnatTat.

30. Family. TETRAODONTIDE.

Diodon Linn. Warm seas 20
Tetraodon Linn. Ditto 30
Cephalus Shaw Ditto 7
Triodon Cuv. India 1

* Erythrinus Gron.

31. Family. Batistipz.

Balistes Linz.
Balistes Linn. Warm seas 32

Balistopus Tvles. Ditto 1
Monacanthus Cuv. Ditto 20
Aluterus Cuv. India, &c. 10
Triacanthus Cuv. India 1
Ostracion Linn. Tropics 26

+ Ophiognathus Harwood.

BONAPARTE S SYSTEM. O1

IV. Order. — CarrinacinEl.

32. Family. ACIPENSERIDZE. Centrina Cuw. Medit., &c. 3
4 Scymnus Cwv. General 7
Acjpenser Linn, General 2 | Zyganadneig, Cuv, india 4
y : i Squatina Dum. Med., Am. 6

33. Family. CHIMERIDE.

: 36. Family. Ramwaz,
Chimera Zinn. Med., Arctic 1 2 y

; Pristis Lath. Warm seas 7

Callorhynchus Gron. Pacific 1 Hinenatae Son Ditto u

3 : Rhina Sch. itto 4

34, Family. SQuaLipm. Torpedo Dum. Ditto 11
Scyllium Cuv. General 15 Leiobatus Blazin.
Pristiurus Bon. Medit. 1 Dasybatus Blain.

Squalus Linn. Trygon Antiq. Ditto 20
Carcharias Raf.,Cuv. General 20 Anacanthus EHhrenb. Redsea 3
Alopius Raf. Medit. 1 Myliobates Dum. Warm seas 11
Rhineodon Smith Atlantic Rhinoptera Kuhl Ditto 4
Somniosus Le Sueur Ditto Cephaloptera Dum. Ditto 3

Lamna Cuv.

1

1

Galeus Raf., C ‘

aleus faf., Cuv. 4 2 : -

WWactolas Cav. ? 3 36. Family. PETROMYZONID&.
4

Notidanus Cuv. Gastrobranchus Bloch Atlantic 2
Hexanchus Raf. Petromyzon Linn. General 6
Heptranchus Aaf. Myxine Linn. india 1

Selache Cuv. Atlantic, &c. 2 Ammocetus Dum. Europe 2

Cestracion Cuv. Australia 1

Spinax Cuv. Medit., &c. 5 Total number of the species 3586

(81.) We are not, in general, favourable to these ex-
positions of methods which we do not adopt ; and we have,
therefore, somewhat abridged the foregoing by omitting
the divisions of the sections, orders, &c. ; but the fami-
lies, genera, &c. are all included, so that the reader will
at once perceive in what way our own series differs from
both this and M. Cuvier’s. We regret, however, that
our space will not allow us to insert a similar expo-
sition of the arrangement of professor Rafinesque, be-
cause, although artificial, there is much to admire in it,
and he was the first to commence that general breaking
up of the Linnean genera into minor groups, which
Cuvier and his disciples subsequently followed. We
shall, however, in the course of this work, introduce
several of the genera and sub-genera proposed by this
able and zealous zoologist, and shall substitute his names
for those of other writers, whenever they have a prior
claim, and whenever his groups can be sufficiently made
out.

(82.) Of naTurAL sysrEms of ichthyology, or those

-

92 CLASSIFICATION OF FISHES.

which are framed with a reference to certain general
laws of creation, real or supposed*, we are only ac-
quainted with those that have at different periods been
proposed by M. Oken, one of the most celebrated among
those metaphysical naturalists who have arisen, of late
years, in Germany. That we may not be thought to
undervalue the labours of those whose aim, like cur
own, is to “ establish resemblances and explain ana-
logies,” we shall here enumerate these systems, which
M. Oken, at different periods, has successively drawn up.

M. Oken’s first system is founded according to an

‘idea he entertained of the predominance which water has:

on the different parts of the body; he accordingly con-
ceives that all fish should be arranged under the follow-
ing orders : —

T. Poissons VentRiers. Bony fish, without scales.

en: THORACIERS. with scales.
IIl. Mempriers. The genera Fistularia, Pe-
gasus, Diodon, &e.
IV. TETIERS, Petromyzon, Squalus, and Raia,
Linn.

In the second, “published five years aftert, this idea
is abandoned for another, by which M. Oken believes
he ean arrange the whole class so as to represent what
he thinks to be the seven primary divisions of the animal
kingdom. A general idea of this system will be ob-
tained by the following enumeration of its chief divi-
sions. He first divides the whole into two great
groups— Osseous and Cartilaginous fishes: under the
first he brings in six of his orders, leaving only the
- seventh in the last. These seven orders are thus desig-
nated : —

* See definitions of natural and artificial systems, Classif. of Animals.
+ Cuvier, Hist. Nat. des Poissons, tom. i. p. 258.

*
a
#
q
&
a
5

OKEN’S SYSTEM. 93

T. Poissons Zooruytes, as the eels, Anguilla, &c.

IJ. —— Vers, Gadus, Blennius, Scomber, &e.
1686 INSECTEs, Labrus, Sciénes.
IV. —— Poissons, Mugil, Cyprinus.
VW ee TET, Cobites, Silurus, Salmo, E'sox.
i Oiseau Callionyus, Gobius, Chetodon,
Pleuronectes. LY
Vit. —— Mamoravx, Acipenser, Lophius, Diodon,

Raia, Squalus.

The families, or “ sub-orders” as they are called,

placed under each of these “ orders,” will be best un-
derstood by the following table. They amount to four
in each ; and these, again, have each four “ genera.”
We do not, however, enumerate the whole of the latter.

met

mor

Order I.—PoIssoNS ZOOPHYTES.

Sub-orders:— . | 3. Cultriformes, includ-| 4. Cepola.
. Murena. ing Trichiurus, and
. Anguilla. Leptocephalus.
Order II.—Poissons VERS,
Sub-orders : — 2. ** Kleques.”’ 4, Gasterosteus, Cen-
. Lotes, including | 3. Scomber. tronotus, &c.
Blennius, Phycis,
Gadius, &c.
Order III.—Poissons INSECTEs.
Sub orders :— 3. Labroides, including | 4. Dorades, as Mullus,
. Perches. Labrus, Sparus, Scarus, &c.
. Gymnocephalus, An- Ophiocephalus, &c.
thias, &c.
Order IV.—Polssons Poissons.
Sub-orders : — 2 Dactyles. 4, Cyprinus, including
. Mugiloides, as Mu- | 3. Clupea (Zznn.). also Atherina, Ar-
gil, Exocetus, &c. gentina, &c.

Order V.—Potssons REPTILZs.

Sub-orders : — 9. Silurus (Linn.). 14. Esox, Hlops, &c. _
. Cobites, Anableps, | 3. Salmo, including |
&e. Serrasalmo,

Order VI.—Polssons CIsEAUX.

Sub-orders : — 3. Pleuronectes, Zeus, |4. Centriscus, Fistu- '

. Callionymus, Urano- Chetodon, Stoma- laria, Stylephorus,

. Gobius, Cottus,

scopus, &c. | teus.

Scorpina, Trigla.

94 CLASSIFICATION OF FISHES.

Order VII.—Poissons Mawaux.

Sub-orders : — I “© Morques,”’ includ- | 3. ‘* Chirques,’? as the J
1, Myxine, Petromy- ing Cyclopterus, Ra- genera Acipenser, :
zon, Syngnathus, listes, and the rest Ziphias, &c.
and Pegasus. of the branchioste- | 4. Squalus, Raia, Chi-
gous fishes, mera, and Lophius. ©

But, as our author soon after discovered that there
were not seven primary divisions in the animal king-
dom, he abandoned his second system, and formed an-
other, in which the number four should predominate.
In his third arrangement, therefore, M. Oken makes
four orders ; four sub-orders, supposed to represent the
orders ; and each of these sub-orders is again composed
of four genera. The result of all this will be sufficiently
seen in the following table :—

Order I.—PoIssoNns PolssoNs.
This order includes, among others, the genera—

Murena. Trichiurus. Gymnotus.
Gymnotus. Leptocephalus. Anarrhicas.
Ophidium. Cepola. Xiphias.
Ammodytes.

Order II.— Poissons REPTILES.
Composed chiefly of the genera —

Gadus. Cottus. Stomateus. ;
Echineis. Gobius. | Cobites.
Gasterosteus. Cyclopterus. Silurus.

Scomber. Pleuronectes. i Salmo.
Callionymus. | Zeus. | Esox.
Uranoscopus. Chetodon.

Order Ili.—Poissons OISEAUX.

Scorpena. | Perca. 1 Mugil.
Trigla. ; Mullus. : | Clupea.
Polynemus. ~ Labrus. ; Atherina.
Exoceetus. . Sparus. Argentina.
Scizna. Coryphena.

Order IV.—Potssons MamMaux.

Centriscus. Tetraodon. Lophius.
Fistularia. Pegassus. Myxene.
Stylephorus, Acipenser. Petromyzon.
Sygnathus. Spatularia. Raia.
Mormyrus. Chimera. Squalus.
Balistes.

Finally, our author, abandoning four as the regulat-
ing number of his groups, adopts that of five, probably

OKEN’S SYSTEM. 95

from his illustrious countryman Fries. He seems to
imagine, however, that because the number five holds
good in the primary divisions of structure in the animal
and vegetable kingdoms, it should equally do so in
every thing; and he accordingly constructs a fourth
system, founded on the organs of sense in these obscurely
known creatures. This fourth system, M. Cuvier says,
was published in Paris in 1822 ; but as we have it not
at present to refer to, we shall transcribe the following
table, from that given in the Hist. Ee des Poissons,
tom.i. p. 234.

Order I.—PoissoNs GERMIERS.

Apterichte.* Lophius. Pleuronectes.
Sphagebranchus. Gymnetus. Echineis.
Synbranchus. Régalec. Platycephalus.
Murena. Cepola. Macroure.
Anguilla. Trachypterus. Phycis.
Gymnotus. Gymnogaster,. Gadus.
Ophidium. Stylephorus. Centronatus.
Leptocephalus. Lépidope. Blennius.
Ammodytes. Trichiurus, Anarhichas.
Order IJ.—PolIssons SEXIERS.
Gobius. Chetodon. Otolithe.
Periopthalme. Stromateus. Sciena.
£iléotris. | Eques. Perca.
Coméphore. Vomer. | Cichla.
Trichionate. Zeus. Serran.
Callionymus. Coryphena. Dentex.
Trichiurus. Rhinchobdella. Labrus.
Trigla. | Gasterosteus. Scarus.
Lépisacanthe. | Scomber. Sparus.
Order I1].—Polssons ENTRAILLIERS.
Cobites. ) Atherina. Salmo
Anableps, Sphyrene. Mullus
Pecilie. Polyptere. Mugil
Pimelodus. Erythrinus, Clupea
RG HT. Lepisostee. Elops.
Silurus. Esox. Exocetus.
Doras. Sternoptyx. Gonorynchus.
Heterobranchus. Gasteropelicus. Cyprinus.
Cataphractus.

* The generic names printed in Italics are vernacular, and not used in
this volume, What these French names mean, M. Cuvier has not ex-

plained.

96

CLASSIFICATION OF FISHES.

Order 1V.—Potssons CARNIERS.

Lepadogaster. | Syngnathus. Ostracion.
Cycloptera. Solénostome. Tetraodon.
Uranoscopus. Pegasus. Diodon.
Cottus. Fistularia. Orthagoriscus.
Batrachus. Aulostoma. Platystacus.
Tenianotus. | Centriscus. Loricaria.
Synancée. Amphisile. Lepidoleprus.
Scorpinus. Mormyrus. Polyodon.
Malthée. Balistes. Acipenser.
Antennatre. Triacanthus. Xiphias.
Lophius.

Order V.—Poissons SENSIERS.
Murena. Petromyzon. Squalus,
Chimera. Raia.

(83.) The first circumstance that strikes the na-
turalist on inspecting these systems, is the different
plans upon which they are constructed, and the separa-
tion they effect, more or less, between groups which all
other naturalists agree in thinking are closely and inti-
mately united. Thus the genus Doras of Lacepede is
so closely connected to that of Loricaria, that it is almost
impossible to determine where one ends and the other
begins; and yet in the last table of these systems we find
they are placed in two different orders. On the other
hand, the genera Acipenser and Xiphias are arranged
close to each other, without possessing, so far as we can
discover, any one indication of affinity. The merits of
every natural system can alone be judged of when the
principles it sets cut upon are worked out in detail:
this done, the materials are before us for forming a cor-
rect judgment, whether the series appears to be that of
nature or of man. We quite agree with M. Oken, in
thinking that the primary orders of fish represent those
of vertebrated animals; and every allowance should be
made for the imperfect labours of all who endeavour to
establish this most important law. But we must con-
fess our inability to make out what are M. Oken’s views
on this subject ; and not being able to comprehend, we
have not adopted them.

(84.) And now, having thus far proceeded in what
relates to ichthyology in general, we must attempt to
establish, in some degree, those primary laws of the

CONCLUDING REMARKS. 97

natural system we have ventured to announce. If we were
to be guided by the high authority of deservedly great
names, rather than by our own impressions of what are
the true affinities of nature, we should be equally autho-
rised and encouraged in making this attempt. The

is an authority now reigning over this department of
zoology, as omnipotent, perhaps, as that which Linneus
once exercised over all branches of natural history ; —
a zoologist whose superior genius every one must
acknowledge, and whose materials for study and re-
fiection, during a long and brilliant career, were almost
boundless. We have laboured for the last fifteen years
to dispel the illusive idea, that natural affinities could
be expressed by a simple series; and that all such ex-
hibitions of nature, however useful, were merely arti-
ficial combinations. Now if those few who still doubt
on this subject, required such an authority as we have
intimated to decide their wavering opinion, such a one
exists, and will be found in the learned author of the
system we have just surveyed, — the illustrious Cuvier.
This extraordinary man, as if to bequeath to us the
result of all his varied and profound experience, thus
concludes his preliminary observations upon fishes
in general, and they deserve from all the most profound
attention. In speaking of the cartilaginous order, he
thus expresses himself * : — ‘‘ It is chiefly in these that
the futility of classing henioed in a single series is visible;

several of its genera, the rays and the sharks among
others, are considerably above common fish, by the com-
plicated nature of their organs of sense and of generation ;
these latter being more developed, in some respects,
than those even of birds: yet other genera, which are
- approximated by evident transitions, such as the lam-
preys and Ammocetes, become so simplified, that they
have been regarded as forming a passage to articulated
worms ; for the latter certainly do not possess a
skeleton, and their muscular apparatus is attached to
membranous and tendinous supports.’ — “ Let it,

* Rég Anim. Griff. Cuv, p. 22.
VOL. I. H

98 CLASSIFICATION OF FISHES.

therefore, never be supposed, that because one genus, or
one family, is placed before another, we consider it
more perfect or superior to the others in the system of
beings ; — he alone could build up such a pretension,
who would attempt to place animal nature on a single
line. Such a project we have long since renounced,
as one of the most false that can be entertained in
natural history.” — ‘* True system,” he again observes,
“ sees each being in the middle of all the others, and
shows all the radiations that link it, more or less inti-
mately, in the vast web of organic nature; and thus
alone we acquire enlarged ideas, worthy of that nature
and of nature's God ; but ten or twenty of these radi-
ations will be often insufficient to express these mul_
tifarious relations.” Nothing can be more in unison
with all that has been urged on the “ multifarious re-
lations” of natural objects than this ; and no authority
can bring more weight to the opinion than this of
Cuvier’s. True it is, that this conclusion was arrived at
by the celebrated Lamarck more than twenty years ago,
and that it has long been acted upon by a few of the
greatest naturalists now living. Nevertheless, the tardy
admission of M. Cuvier to the impossibility of naturally
arranging objects in a single series, is even more valu-
able than if it had come sooner: the very delay shows
us that, in truly great minds, truth will finally triumph
over early imbibed prejudice, and, although not acted
upon, it will yet be acknowledged. If, therefore, we
make some attempt, in the following pages, to explain
and reconcile these “‘multifarious relations,” and abandon
altogether the trammels of an artificial system, the very
essence of which is to place fishes in a single series,
we do nothing more than follow up the theoretical idea
of Cuvier ;— a course, however, which imposes the
absolute necessity of abandoning all those parts of his
arrangement which interfere with the exposition of those
‘‘ multifarious relations” he speaks of, yet makes no ef-
fort to explain on any general principles. To attempt
to do this, however, in ail the groups, would be mani-

ON THE NATURAL ORDERS, 99

festly impossible ; and yet, on the other hand, if only
one, however small, can be sufficiently analysed to
establish what has been advanced on natural arrange-
ment, philosophy teaches us to conclude that similar
results would attend the analysis of all others. No
other conclusion, in short, can be arrived at, whether
by inductive philosophy or common sense.

CHAP. V.

ON THE NATURAL ARRANGEMENT OF FISHES, THE PRIMARY TYPES
OF FORM, AND THE ANALOGIES THEY PRESENT TO OTHER
CLASSES OF ANIMALS.

(85.) Ir is manifest to every naturalist, that the
most perfectly organised groups, in the great class before
us, are composed, as M. Cuvier has truly said, of the
osseous fishes, or those whose skeletons are of solid bone.
This being their most characteristic mark, it follows,
that although osseous fishes (less perfectly organised in
every other respect) may be found in other orders
which approach these, yet, that none with a carti-
laginous skeleton can naturally belong to this most
typical division. Now this great assemblage, like those
of all others equally typical in the animal kingdom,
resolves itself into two groups — the one composed of
such as have the rays of the dorsal fins more or less
spinous, the other of such as have them soft or articu-
lated. These groups were long ago perceived and
defined by the old ichthyologists ; and if any authority
were necessary to sanction our belief that they are truly
natural, we cannot cite a higher than Cuvier. ‘The
osseous skeleton, however, although the paramount, is
not the only character possessed in common by these two
groups. The ventral fins, which are analogous to the
H 2

100 -CLASSIFICATION OF FISHES.

feet of birds and quadrupeds, are almost always present ;
and the gill-covers are not only moveable, but the
branchial aperture is fully developed — in other words,
it does not assume the form of a simple slit or spiracle,
as in the eels and rays. Here, then, we find the three
chief characters of osseous fishes; the first absolute,
the two next less so: and it may safely be asserted,
that every fish which possesses two out of these dis-
tinctions, finds its natural place in the spiny or the
soft-rayed divisions. These we regard, like our prede-
cessors, as the two most typical orders of the whole
class. We shall now enumerate their characters some-
what more in detail.

(86.) The AcanrHopreryess have the anterior rays
of the dorsal fin simple, rigid, and acute ; the remainder
being branched and articulated ; or, if there are two dorsals,
the first is entirely composed of spinous rays. We are
now, as in the following definitions, speaking of the
pre-eminently typical examples ; the exceptions will be
noticed afterwards. The anal fin is also usually furnished
with both sorts of rays, and the membrane is never
fleshy. The branchial aperture is large; the bones of
the operculum fully developed, and frequently spinous
or serrated ; the eyes large and lateral ; the body ovate
or oblong; the ventral fins placed near the pectoral ;
the scales hard and shining, ornamented with beautiful
colours, or richly silvered. They are almost all marine
fish, and are more constructed for long continued motion.
The aberrant families of this immense order, which in-
cludes more than one half of all the fish yet discovered,
presents us with several deviations. Some of the blen-
nies are viviparous ; and the simple rays of their dorsal
fins are sometimes soft; so also are those of the Ophi-
cephali. In the Gymetres, the ventral fins are occa-
sionally wanting ; but the branchial aperture is large:
the fins are fleshy in the blennies, and scaly in the
chetodons.

(87.) The Matacorrerycess, or soft-rayed order, is
less numerous than the last, and are so much diversified,

ON THE NATURAL ORDERS. 101

that it is somewhat difficult to find any more certain
indication than that presented by their fins: the ventral
fin, however, is always present ; and the branchial aper-
ture, with one or two exceptions, is unconfined. We
thus get three characters; one of which separates this
order from the last, another detaches it from the next,
and the osseous skeleton from all other divisions.
Their organisation, as fishes, appears less perfect than
the more typical group ; for it is among these we find
all the ground fish, — those which are restricted to fresh
waters, and such as lie in wait for their prey. In this
order, also, we have a small group of viviparous fish,
analogous to the blennies in the last. The salmons,
pike, herrings, cods, carps, and flat fish, have been
justly included in this order, which, in regard to the
subsistence it furnishes to man, becomes the most im-
portant of all others.

- (88.) In the next order, the typical structure begins
to disappear, and is finally lost. The skeleton, in some,
is still osseous ; but in many others is sub-cartilaginous ;
and even finally becomes membranaceous: the fins, which
represent the feet, entirely disappear: the branchial
aperture assumes the form of a slit, and is termed a spi-
racle : the shape is long, and like that of a serpent: the
dorsal, anal, and caudal fins, when all are present, are
generally united : the body is slimy and naked; or the
scales are very minute, and imbedded in the cuticle.
The reader cannot fail to recognise, in this description,
the essential characters of the eels, lampreys, and other
similarly formed families, which have as much the
outward aspect of serpents as of fish. To this order
we retain the original name of Aroprs bestowed upon
it by Linneus,

(89.) Having entirely quitted the osseous structure
of bone in the last tribe, we next come to such families
as have their skeleton fibro-cartilaginous: these, also,
breathe by a spiracle ; the operculum being either obso-
lete, or entirely: concealed beneath the common skin.
They differ, however, materially from the last, by pos-

H 3

102 CLASSIFICATION OF FISHES.

sessing ventral fins, and by the following additional
peculiarities : —the body is thick, very short, heavy, and
often, as it were, deformed ; the ventral fins are placed
upon a peduncle, so that they may be used, in some de-
gree, as feet, enabling the animal to crawl on the
ground ; the eyes are small, and placed nearly vertically ;
the mouth opens in the same direction, and has the
under jaw longest: in the most typical family, the
body is soft ; but in the sub-typical, it is either covered
with osseous plates soldered together, or with acute
prickles: the ribs are almost always wanting; and
they are the only fishes which have the anatomical cha-
racter of the maxillary bone and the palatine, arch in-
serted in the cranium. Adopting Cuvier’s name for
these fish, rather than that of Linneus, which was
founded in error, we term this order the Puizcro-
GNATHEs, or cheloniform fishes.

(90.) The fifth and last primary group consists
of those truly cartilagious families which have the fins
and mouth of ordinary fishes, but who breathe by one
or more spiracles: the mouth is placed beneath the
snout, which is very broad and projecting ; the major
part are viviparous ; and the body is smooth, or, at
least, destitute of true scales. The sharks and rays are
the best known, and are the most typical of these fish,
which, as indicating their typical character, we propose
to call the CarTinacInEs.

(91.) That there is every reason to believe these
primary divisions of the class are founded in nature,
will be apparent from their accordance to the divisions
of the same rank that have been generally adopted by
the most eminent zoologists. Without attempting, in
our present rapid course, to skow in what manner they
blend.into each other and form one great circle, we
shall at once proceed to compare them in the order in
which they have been noticed, with other groups better
authenticated, or rather, we should say, more familiarly
‘known to naturalists. If we are successful in‘ this
effort to establish a uniformity of analogical relations

THE PRIMARY TYPES OF FORM. 103

between each and all of such as we may select for this
purpose, the circular affinities of the whole will be
_ sufficiently established by anology ; whether we are ac-
guainted, or not, with the precise links that connect the
several portions. Our main object, however, is to
adduce further proofs of the proposition contained in
our early volume of this series, namely, that all animals
could be referred to certain primary types of form. It will
therefore be advisable, in this place, briefly to recapitu-
late what was then said, that the naturalist may judge
how far the characters there given accord with those by
which we have defined the primary types of fishes.
(92.) In the first place, we have said that the most
perfectly typical individuals of every natural group
are those which exhibit the highest development of
those characters by which the group, as a whole, is
distinguished ; or, in other words, “ they are endowed
with the greatest number of perfections, and capable of
performing to the greatest extent the functions which
peculiarly characterise their respective circles.’ This
pre-eminent perfection shows itself, also, in nearly all
such types as are of this primary rank. ‘ This is
apparent in the order Quadrumana among beasts, and
in that of Jnsessores among birds ;” both of which are
the most perfect, and by far the largest, groups in their
respective circles. Among the Annulosa, again, the
Ptilota, or winged insects, are probably ten times more
numerous than all other annulose groups put together.
In tracing this peculiarity in the typical groups of
lower divisions, we find it also very prevalent ; and even .
in looking to sub-families, or even genera, we find that
the genus Picus, Sylvicola, Sylvia, among birds, and
that of the restricted sub-genus Scarabeus (Macl.)
among insects, are all remarkably abundant in indi-
viduals, when compared with the remaining contents of
their respective circles.”’ Every ichthyologist will per-
ceive that the foregoing observations are as applicable
to the order of AcanrnoprErycEs among fish, as if
they had been expressly written to distinguish them
H 4

Bae ALS
- y

from all the others. We can therefore have no he.
sitation in admitting the conclusion which Cuvier ar-
rived at by analysis, — that this immense group contains
the most perfect fishes in existence.

(93.) ‘“ Sub-typical groups, as the name implies, are
a degree lower in organisation than the last, and thus
exhibit an intermediate character between typical and
aberrant divisions.” This, also, is precisely the nature
of the Malacopter ‘yges, Or soft-rayed fishes: they only
yield to the last in the perfection of their structure.
“The numerical contents of sub-typical groups are
almost universally less than in those which are typical.”
The truth of this remark is exemplified in the present
instance: the number of the soft-finned osseous fishes
is probably more than two thirds less than that of the
typical group, to which they are evidently inferior in
their general structure and in their power of swimming.

(94.) “ The Nararorra or Aquatic type of nature,
as seen in quadrupeds, birds, and reptiles, are more espe-
cially inhabitants of the waters. They possess many and
striking peculiarities, modified, indeed, in the most asto-
nishing manner, but more conspicuous, perhaps, through-
out all natural groups, than any of those belonging to
other types.” They are chiefly remarkable for their enor-
mous bulk, the disproportionate size of their head, and
the absence or very slight development of their feet.
These aquatic characters are exemplified in the Radiata
in the animal circle; in the class of fishes among Ver-
tebrata ; in the Cete, or whales, among the Mammalia ;
and in the Watatores among birds. ‘“‘ As we approach
the more perfectly organised animals, we see the deve-
lopment of another singular feature — namely > a very
‘large, thick, and obtuse ‘head, furnished with jaws ge-
nerally capable of great expansion, and terminated by
a blunt or truncated muzzle. As fishes constitute the
pre-eminent natatorial type of vertebrated animals, so we
find that such groups as represent them in other circles
of the Vertebrata have the feet transformed, as it were,
into fins. How beautifully is this exemplified in’

104 CLASSIFICATION OF FISHES.

THE PRIMARY TYPES OF FORM. 105

whales (forming the natatorial order of the Mam-
malia) ; the swimming order of birds ; and the Sauri, or
aquatic reptiles! As to the economy of aquatic types,
we have already premised that they are almost entirely
carnivorous. In those that belong to quadrupeds and
birds, the food is seized by the mouth alone ; the feet
being slightly, and often not at all, developed: and all
such as do not wander in search of their prey, dart
upon it from a fixed station.” This is the substance
of what was formerly advanced regarding the aquatic
types of all animals, and we are now to determine
whether the cartilaginous order of fishes does not accord
with this theoretical description. Independent of the
nature of their bones, they can be immediately recog-
nised from all other fish by the muzzle being so
enlarged and produced beyond the jaws as to alter the
position of the mouth, which is actually placed beneath
the head — not, as in all other fishes, at its termination.
The sharks, no less than the rays, are the most gigantic
monsters among fish ; and that they are eminently car-
nivorous is unfortunately too true, since the first are
declared enemies to the human race. The great size
of the head observed in the aquatic Mammalia is not
equally conspicuous in the same type among fish,
although none have their head larger in proportion to
their body than these; and such is the peculiar shape
of the ray, that they seem, like their prototypes the
Crustacea, to have the head cenfounded with the
thorax and body, so as to give the impression that all
three parts were united in order to form an enormous
head. The fishes of this family, which we place at the
head of the Cartilagines, seem also to possess the habit
of natatorial birds, in lying in wait for their prey,
and darting upon it from a fixed station ; while their
viviparous nature is at once explained, when we re-
collect that these creatures effect the passage between
fishes and aquatic Mammalia. There can be no doubt,
in short, that in' the CArTimnAGines we have an exem-
plification of the natatorial type.

106 CLASSIFICATION OF FISHES.

(95.) The type which succeeds the last is the most
aberrant division of every circular group. On a former
occasion we have stated that one of its most prevalent
characters is that of having the mouth very small, or
otherwise but slightly developed ; and because: all suck-
ing animals seem to belong to this type, we formerly
called it the suctorial: but such a function, in the present
class, has not been clearly made out; and, as we have
already shown it is represented among reptiles by the
tortoises, we shall designate it by the same name there
employed, and here, also, call it the cheloniform type.
This, as was formerly mentioned, is the same as the
grallatorial type among birds, the gliriform among
quadrupeds, the onisciform or vermiform among in-
sects. The most prevalent distinctions of this type,
besides the smallness of the mouth, and the absence of
true teeth, may be thus concisely stated and illustrated.
1. The general structure is always more dissimilar than
any other from the pre-eminent type; they are, con-
sequently, the most imperfectly developed of their own
circle. 2. The jaws, or muzzle, or mandibles; are often
turned upwards, the lower being longer than the upper:
this we see in the Brazilian racoons (Vasua); while
the avosets, and other grallatorial types, present the
same unusual character; and these are the smallest
mouthed birds in creation. 3. The eyes are always
particularly small, as in the mole, and other gliriform
quadrupeds; and in the Trochilide, Tringide, and other
grallatorial birds: sometimes, indeed, in the aberrant
Vertebrata, they are even wanting, as in Mywine, among
fish, and nearly soin Cecilia in the class of reptiles: the
situation of the eyes, in all these groups, is likewise very
peculiar ; they are placed at a distance from the mouth,
and very far back upon the head, towards the crown,
and thus approximate. This is very observable among
the tenuirostral and grallatorial types of birds; and
we find the same in the genera Chironectes, Uranoscopus,
and similarly formed fish, of which numerous examples
may be cited. But perhaps there is no character of this

THE PRIMARY TYPES OF FORM. 107

type more widely diffused among nearly all the classes
of the animal kingdom, than that of the body being
mailed, or protected, as in the chelonian reptilés, by
bony plates, either united or articulated at their su.
tures, or lying over each other in the manner of scales.
We have already cited numerous instances of this struc-
ture in the animal kingdom ; nor is it more conspicuous
in the chelonian reptiles than in the cheloniform fishes :
the family of the Balistide, in short, is as complete
a prototype of the tortoises, the hedgehogs, the scaly
anteaters, the porcupines, and other spined gliri-
form quadrupeds, as it is possible to conceive. Our
surprise is that such resemblances should exist where
the nature of the animals are so different. Again, the
smallest and most imperfectly formed mouths, destitute
of true teeth, are to be found among the PLecroena-
THEs, or cheloniform fishes, which thus became the most
aberrant type in the great circle of Pisces.

(96.) There is still a fifth primary form in the animal
kingdom, which has been designated the Rasorial type in
ornithology, and the Unguiculate among quadrupeds.
The characters by which this form may be recognised,
among the animals just named, have been already so
fully explained, that they need only te be touched upon
in this place. In the more organised or warm-blooded
Vertebrata, great strength of foot, the faculty of climb-
ing, with a facility and aptitude for domestication, are
among the most prominent peculiarities observable in
this type ; but none of these can be expected in fish.
This is the type, however, which is so remarkable for
the great development of the tail ; for, if we went through
the whole class of birds, and selected those, beginning
with the peacock, wherein the tail was most conspicu-
ous, either for its size, its length, its singularity, or for
the beauty of its colours, we should unknowingly fix
upon those birds which analysis has demonstrated to be
rasorial types. The same results would attend a similar
selection of quadrupeds, and of winged insects. All
these, collectively, furnish many hundred proofs by

108 CLASSIFICATION OF FISHES.

which the uniformity of this structure is preserved.
We can now add to these proofs others, equally strong,
presented by the reptiles and the fish. A great and
peculiar development of the tail pervades the whole of
the order Apopss, and of all other groups by which it
is represented ; so that, by designating this type, when
speaking of fishes, as the anguilliform, instead of the
rasorial, the reader will immediately be reminded of
the eel-shaped form, which is its chief characteristic.
By the tail, we do not, of course, mean the caudal fin ;
for that, in the fishes we are now speaking of, is
usually very small ; and, among several, it is sometimes
wanting. The true tail of a fish, strictly speaking,
commences with the termination of the stomach ;
the length of the latter being manifested, externally, by
the situation of the vent. The abdomen of ‘the eels
is so unusually short, as not to equal one fourth
the length of the tail; and this structure is just as
prevalent in groups which represent the apodal order
as in the order itself. Thus, although there seems but
one character of the rasorial type of birds to be traced
also in that of fishes, yet it is the principal one, and it
is so universally prevalent, as to render the presence of
others unnecessary to detect the analogy. The only
instance yet ascertained of the scansorial power being
possessed by fish, is that of the Perca scandens, which
is said to climb banks and aquatic plants by using its
pectoral spines as feet.

(97.) We shall now state a few of the modifica-
tions under which the anguilliform type appears in
such groups as represent, without belonging to, the
apodal order; all being distinguished, as just observed,
by having the abdomen much shorter than the tail. In
the eels, the body is cylindrical; but in many other
analogous families it is compressed, and that to such an
extent, as to have given rise to Cuvier’s expressive name _
of riband-fish. The genera Cepola, Leptocephalus, Ophi-
dium, &c. are good illustrations of this structure ; not to
mention such extraordinary forms as the Gymnocephalt

THE PRIMARY TYPES OF FORM. 109

of Bloch. The three fins of the tail, that is, the
hinder dorsal, the caudal, and the anal, if not united, as
in Cepola, Ophidium, Plotosus, Anarhichas, &¢. are
only separated by a small interval, as in Physis, Mer-
langus, Blennius, &c.; or the ventral fin only is exces-
sively long, as in all the genera and sub-genera cf the
anguilliform division of the Siluride@. In other genera,
the dorsal and caudal fins are obsolete; but the anal ex-
tends the entire length of the tail, which terminates in a
point. So far as our analysis has extended, it seems
that all these are but modifications of the anguilliform
structure. The ventral fins, which are universally
absent among the true Apodes, are sometimes wanting,
also, in their representatives, as in Ophidium, Anarhi-
chas, Ammodytes, &c.: usually, however, their slight de-
velopment marks the type we are now speaking of ; thus,
in the two families of the Blennide and the Gadide,
the typical genera have their fins composed only of two
rays, or, when the others are present, they merely exist
in a rudimentary state. The scales, again, frequently
present a peculiar character: when present, they are
very small, often scarcely perceptible, and appear to be
inserted, as in the eels, beneath the cuticle: this is
seen in most of the Gadide ; while in other anguilli-
form types, like the Blennide, the body is slimy and
naked, either covered with an opaque skin, or semi-
transparent. The snout is always short and obtuse,
the mouth not extensible, and the teeth either very
small or none. Nearly the only mailed genus that
possesses the anguilliform shape is Polypterus; and
this, as we suspect, may probably belong to the order
Plectognathes.

(98.) Having now stated some of the most preva-
lent analogies between the primary types and divisions
of fishes, and those of the warm-blooded Vertebrata,
we may exhibit the results in a more compact form by
placing these groups in three columns ; and it will then
be more distinctly seen in what way each is related
to the other by analogy.

110 CLASSIFICATION OF FISHES.
} u : : Orders of Quad-
Orders of Fishes. Analogies. Orders of Birds. rupeds.
ACANTHOPTERYGES, ‘Typical. INSESSORES, QUADRUMANA,'
MaALACOPTERYGES. Sub-typical RAPTORES. FER.
s Snout broad, de-2.
CARTILAGINES. f pressed. i NATATORES. CET.
Eyes small, placed
PLECTOGNATHES. far back towards GRALLATORES. GLIRES.
the crown.
APODES. Tail very long. RASORES. UNGULATA.

We were at first perplexed to discover how it was that
the Malacopteryges, by being the sub-typical order, should
represent the Raptores and the Fer@; because these fishes,
so far from being pre-eminently carnivorous, comprehend
the greater part oe such as habitually feed upon vegeta~
bles; nor can this apparent contradiction be explained so
readily as we could wish, unless by looking to the nature
of the whole group. Now, the class of Pisces is that
aberrant division of the Vertebrata which represents the
aquatic or fissirostral type of vertebrated animals: this
type, therefore, being eminently carnivorous, the ani-
mals which represent it, in its greatest perfection, must
equally be so: and thus we have an additional verifi-
cation of M. Cuvier’s opinion, that the Acanthopteryges
are the most perfect of fishes; while the Malacop-
teryges, which are next in affinity, become the next in
analogy, and are, therefore, the sub-typical. This view
of the question is confirmed on looking to the analogies
of other aberrant circles. If we take, for instance, the
scansorial birds, which form an aberrant tribe in the
circle of the Jnsessores, just as does the class of fish
in that of the Vertebrata, we find the analogies reversed
precisely in the same way. Of the two typical families,
the woodpeckers are the most carnivorous, although
they are the pre-eminent type; while the parrots,
which are the sub-typical, are entirely frugivorous.
Those naturalists, who may be interested in this ques-
tion, will remember how often we have adverted to it on
former cecasions; and we only again touch upon it here,
to show that, however contradictory our second analogy
in the foregoing table may at first appear, it is not dif-

REMARKS ON ANALOGIES. 111

ficult to be explained in no unsatisfactory manner. The
other three analogies, having already been enlarged upon,
require no further elucidation, but may be left to speak
for themselves.

(99.) Before proceeding further in this inquiry,
we shall here irtroduce a few observations upon the
nature of analogies in general, which have only been
glanced at in our former volumes; the more so, be-
cause, upon further reflection, some considerations have
arisen which seem to us of much importance. It has
not been—although it may be—objected to these tables
of analogies, that the resemblance between two groups,
supposed to represent each other, is usually confined
to two, and often to one, analogical character only;
while, in all other points of structure, there is a marked
dissimilarity. ‘This objection, upon a first view, seems
not easily surmounted, because it may be further urged
— If these two groups really represent each other, why
are they not more alike? Why are we so frequently
obliged to labour and search for the purpose of finding a
single point of resemblance, which, after all, is sometimes
so trivial, and depends on a modification of structure
so secondary, that no great importance can be attached
to it? To this we should reply, that the importance of
a character is by no means to be measured by mere in-
dividual or preconceived opinions, but by its constancy
in certain groups, whereby affinities or analogies may be
detected. And in answer tothe main objection, we main-
tain that this paucity of mutual or common characters,
so far from being a stumbling-block, is both inevitable
and essential to our theory. Did two analogous groups
present such strong resemblances, in most of their cha-
racters, that every one would immediately confess the.
likeness, there would not be a hundredth part of that
variety in nature which actually exists. This will be
apparent to the reader, when he remembers, that, on the
principle of universal representation which we now
assume, every group shows an analogy, direct or in-

112 CLASSIFICATION OF FISHES;

direct, near or remote, to many hundred others: it
is, therefore, absolutely impossible for all these simili-
tudes to be so clear as to silence gainsayers, or even to
strike, at first sight, the more experienced naturalist,
who often can only estimate the value of the analogy
between two remote groups, by tracing these analogies
through a series of intervening forms. The innume-
rable modifications of the same structure which we see in
nature, accomplishes two objects: they excite our won-
der and admiration of the Infinite Mind whose fiat has
produced them; and they enable us, though often dimly,
to trace, in one or two characters, a symbolical relation-
ship between a great number of groups, quite different
in all other respects. But, perhaps, an example will best
explain our meaning. No analogies can well be stronger
than those between the chelonian reptiles and the che.
loniform fishes, forming our present order Plectognathes:
but then, if a// the fishes in this latter group were cased,
in the same way, in hard plates—if they al] had very
small mouths, the sharp and crenated jaws performing
the office of teeth—if they a// were eminently aquatic—
and, lastly, if al] their pectoral fins were formed as in
ordinary fishes—what possible characters would be left
by which to indicate their analogy also to the Amphibia,
or frogs, which are as truly and confessedly analogous
to the tortoises, as the tortoises are to the cheloniform
fishes? No such resemblances, that we know of, would
remain, except their imperfect skeleton; or none, at
least, which would strike an ordinary observer; and we
should thus haveno apparent mark by which to conjecture
the relationship. But Nature has provided against this:
has created such a diversity in the order Plectognathes,
that, while one division immediately reminds us of the
chelonian reptiles, another is an equally strong repre-
sentation of the amphibious frogs. The Lophius picta
of Shaw (fig. 6.) will convince the student we are not
prone to exaggerate resemblances. We have only to
point to the Chironectide in proof of this latter relation:

-ANALOGIES OF THE ORDERS. 113

and thus, by the paucity of ber analogical characters,
; relative to one

group only, she
is enabled, as it
were, to disperse
the rest over a
number of others,
but of which, each
—as the inevi-
table consequence
of this. rule —

can possess only ene or two.

(100.) The two comparisons which we shall now
institute, illustrate, and will tend to confirm, the above
remarks: the first will be between the primary types,
or orders of fishes, and those of the entire circle of the
Annulosa; the second will be between the fishes and
the primary groups of the reptiles.

Primary Divisions Primary Divisions

3 logies.
of Fishes. Analogies of Vertebrata.
ACANTHOPTERYGES. ? The most highlyorganised groups { QUADRUPEDs.
MMALACOPTERYGES, in their respective circles. BIRDS.
CARTILAGINES. Mostly viviparous, REPTILES.

§ Semi-aquatic. No true teeth, ae ARTE

PLECTOGNATHES, ? scales.

APGDES. Posterior limbsor finssmallornone, -FISHEs.

Whether the two first groups in each of these
eolumns present any absolute points of resemblance in
their structure, we know not; but certain it is, that. the
osseous fishes, as no less an authority than Cuvier main-
tains, are the most perfect in their own class, just as
the warm-blooded Vertebrata are in the cpposite column.
We have already endeavoured to account for the rever-
sion, as it seems, of the analogies in the two typical
divisions of this class; for, were it not so, it might
almost be thought that, as the organs of locomotion are
most developed in birds, and pelagic or acanthopterous
fishes, they would be analogous, as in this respect they
certainly are: while the ground fishes, or Afalacop-

VOL, I. 1

iTt4 _ CLASSIFICATION OF FISHES.

teryges, and quadrupeds, where these powers are evi-
dently diminished, would stand opposite to each other.
Be this, however, as it may, we had better, perhaps, for
the present, leave these groups as they now stand, and
proceed to the two next ; that is, the cartilaginous fishes
and the reptiles. Between these two there is no ana-
logy, however remote, to be discovered in their external
shape ; and yet, independent of the mode of their pro-
duction, this is the strongest point in our present pro-
position, because it rests upon an authority which no
one would be disposed to question. M. Cuvier com-
mences his remarks upon the Chondropteryges, by ob-
serving that “many of the genera approximate to the
reptiles in the conformation of the ear and of the
genital organs ;’ and one of our best ichthyologists has
expressed a similar opinion.* Having before adverted
to the analogy of the typical Plectognathes and the
amphibious frogs, we may pass on to that by which the
apodal order remains to represent the whole class of
fishes. Now this may be inferred, if not substantiated,
in two ways, negatively or positively: first, it might be
safely concluded, that if the four previous analogies are
correct, then there can be no doubt about this last, see-
ing that it embraces the only two groups which yet
remain; but we do not rest altogether upon this de-
duction. The whole class of fishes are remarkable for
the smallness of their posterior members, which, in
them, are fins: these are almost universally of a much
less size than their dorsals, pectorals, ventrals, or caudals.
Now, this characteristic is more conspicuous in the
Apodes, or anguilliform type of fishes, than in any
other; because, among them, the ventral fins are alto-
gether wanting. If we wished to trace this character
through other orders of animals, we need only look to
the aquatie division of the Mammalia, and to the nata-
torial order of birds; both of which have the most
imperfect feet of their respective classes: the corre-

* Yarrell’s Brit. Fishes, vol. i. p. 40.

ANALOGIES OF THE ORDERS. 115

sponding type by which all and every of these groups are
represented among the reptiles, being the saurian or
natatorial order.

(101.) Although we have hitherto invariably refrained
from employing, as instruments for reasoning, the con-
tents of circular groups which have not been previously
laid before the reader in detail, and in some degree
demonstrated, yet, as the class of reptiles is contained
in this treatise, and will follow that of the fishes, we
shall here, in some measure, anticipate the results of
their investigation, by naming the orders into which, as
we believe, they are first divided; and this we do for
the purpose of showing their relation to those of the
present class, each being arranged in two distinct

columns.

; : : Orders of .
Orders of Fishes. Analogies. Reptiles.

ACANTHOPTERYGES, (eae most highly organised of § Lacrrrss.

MALACOPTERYGES. their respective classes. OPHIDEs,
CARTILAGINES. ; as een hea ee. road, ee SAURES.
PLECTOGNATHES. ee hic’ ake Hee CHELONIDES.
Moun paetor extremities imperfect ae Chana:

Until very lately we have always been impressed
with the idea that the ophidian reptiles, or serpents,
were the pre-eminent types of the reptiles; because
their form is that which seems to be most prevalent
in other animals which represent that class; yet, as the
pre-eminent type is found invariably to be that which
is most highly organised, so it would seem to follow
that this rank belongs to the lizards (Lacertes) rather
than to the serpents. This theoretical conclusion is borne
out by the above table, where we find the acanthopte-
rous fish and the lacertine reptiles standing opposite ;
each being the most highly organised of their own class.
The affinity between the lizards and serpents is equally
close as that between the two typical orders of fishes ;
and both are sub-typical. The relationship between the
cartilaginous fish and the saurian or aquatic reptiles

Lee

116 “ CLASSIFICATION OF FISHES.

(which includes the crocodiles and most of the extinct
fossil genera) is very striking: both are the most gigan-
tic and ferocious inhabitants of the water; and, like all
types pre-eminently ‘aquatic, they have the head large,
the muzzle long and generaliy broad, the mouth large,
and armed with formidable teeth. The close resemblance,
again, between the tortoises and the mailed Plectogna-
thes require no additional evidence in support of their
perfect and beautiful analogy. Lastly, we have apodal
fishes, standing opposite to that most singular group of
reptiles represented by the chameleon. We shall not
here anticipate the reasons subsequently given for
placing these scansorial lizards as the representatives of
a distinct order; but we may here call the attention of
the naturalist to the following resemblances existing
between these two groups. The locomotive mem-
bers of the chameleon assume, indeed, the form of feet,
and not cf fins; but then they are the least organised
feet of all the lizards, and are formed completely on the
scansorial model; the toes being in pairs, of which two
are placed forward and two backward: the tail, again,
as if to make up for this deficiency, is highly developed,
not so much in its length, as in the faculty it possesses
of being prehensile, so that it can be used, like that of
scansorial birds, as a hinder foot or support. Now, the
structure of the apodal fishes is singularly analogous to
all this: the fins which represent the feet are entirely
wanting; while, at the same time, they have invariably
the longest tails. The apodal order passes into that of
the Acanthopteryges; and they are as closely united as
the chameleons are to the Lacertes, or lizards.

(102.) To pursue these details further appears un-
necessary. If we have been successful in determining
the primary types of the class now under consideration ;
and if they truly represent, as here stated, the corre-
sponding types in the other vertebrated animals; it fol-
lows that, through these latter, they represent all others
contained in our preceding volumes. ‘These compari-
sons will amply repay the labour of those naturalists

ANALOGIES OF THE ORDERS. Lig

who feel, with ourselves, the inexpressible pleasure of
tracing resemblances under innumerable disguises, as if
they were employed to conceal the simplicity of a few
general laws, by which all the variations in the animal
world are regulated.

(103.) It now only remains to bring before the eye,
at one glance, all the groups we have touched upon ;
the affinities being expressed perpendicularly, and the
analogies horizontally.

Circle of the Class _ Circle of Circle of Circle of - Circle of
of PiscEs. VERTEBRATA. REPTILIA. — Birps. MamMMALIA.
1. Acanthopteryges. QUADRUPEDS. Lacertes. Insessores. Quadrumana.
2. Malacopieryges. Birps. Ophides. fiaptores. Fere.
3. Cartilagines. REPTILES. Saures. Natatores. Cetacea.
4, Plectognathes. AMPHIBIANS. Chelonides. Grallatores. Giires.
5. Apedes. FISH. Chemelides. Rasores. Ungulata.

One advantage attending this recapitulation, is the
facility it gives of embracing, at a single -glance,
the different degrees of analogy of the whole Vertebrata:
the sharks, for instance, are thus shown to be repre-
sentatives of the natatorial birds ; an analogy which, if
simply stated as an isolated proposition, would certainly
appear fanciful and altogether improbable; and yet,
when traced through the medium of the aquatic Mam-
malia, or Cetacea, and then through the Ichthyosauri,
and other aquatic reptiles, is at once brought home to
the conviction of every unprejudiced mind, even without
the high authority of Cuvier. Here, then, we may
close our general introduction, and proceed at once into
as many details of the several orders as the nature of
eur work will permit.

118 CLASSIFICATION OF FISHES.

CHAP. VI-

ON THE ORDER CARTILAGINES, OR CARTILAGINOUS FISHES,

(104.) Tue cartilaginous fishes, at the head of which
stand the sharks and rays, are well known to be the
largest and the most formidable of the whole class. The
peculiar structure of their skeleton, which gives rise to
their name, admits of these animals continuing to grow
as long as they live ; the consequence of which is, that
as they inhabit the wide ocean, and have few enemies,
they are sometimes met with of such an enormous size,
that their weight and dimensions are almost incredible.
Besides these two families, numerous both in minor
divisions and species, we include the sturgeons, the
spoon-fish (Spatularia of Shaw), and those extraordi-
nary fish, the Chimerine, or sea-monsters.

(105.) The distinguishing anatomical characters of
this order consist in the skeleton or bones being en-
tirely cartilaginous ; that is to say, it is not formed of
osseous fibres, but the calcareous matter is deposited in
small grains, and not by filaments: hence it is that
there are no sutures in their skull, which is always
composed of a single piece ; the usual divisions, how-
ever, of the cranium of ordinary fishes may, in these,
be readily distinguished by the angles, hollows, and
other inequalities on the surface of the cranium. It is
remarkable, also, that the moveable articulations in the
other orders are here not at all apparent. As an in-
stance of this, it may be mentioned, that a part of the
vertebre of certain rays (Raia) are united into a single
body ; while, in other instances, some of the articula-
tions of the bones of the face, according to Cuvier, dis-
appear. The most apparent anatomical characters of
this class is, to want the maxillary and inter-maxillary

CARTILAGINES. — GENERAL CHARACTERS. 119

bones, or, rather, only to have them in an incipient
state, concealed under the skin, while their functions
are performed by the bones analogous to the palatine
arches. The gelatinous substance, which, in other fish,
fills the interstices of the vertebre, and communicates
only from one to the other by a small hole, forms,
in many of these fish, a cord, which threads the whole
body of the vertebre, with scarcely any variation in its
diameter. *

(106.) The connection of this order of fishes to the rep-
tiles, properly so called, is effected by means of the E/anio-
sauri, or the fossil genera of Ichthyosaurus, Plesiosaurus,
and other swimming lizards of gigantic dimensions,
now extinct. M. Cuvier, without being aware of the
full value of his observation, confirms our theory in this
point, when he declares that “ these cartilaginous fishes
approach the reptiles by the conformation of their ear
and of their generative organs ;’’ while, on the other
hand, to prove their affinity to the cetaceous quadrupeds,
it has been well observed that these latter “‘ lead us, by a
very distinct and natural transition,” to fish. “ The vivi-
parous sharks, such as the basking shark (Selache maa-
ima Cuv.), with their ear more perfectly organised than
that of other fishes, and their body destitute of scales,
the particular disposition of their fins, and their closed
branchiz, all indicate at what place we are to enter
among the fishes upon leaving the cetaceous quadru-
peds.’+ It is curious to see, by the above opinions,
how perfectly these two naturalists really agree, at the
very time when, from a partial consideration only of
their theories, they would appear as opposing the views
of each pte: both may, indeed, be said to be in part
right. M.Cuvier, by depending entirely on his con-
summate knowledge of comparative anatomy, came to
the determination of placing the class of fish imme- |
diately after that of reptiles: while Mr. MacLeay, fol-
lowing the simple circle of affinity in the Vertebrata,

= Reg. An. 2d ed. tom. ii. p. 376. + Hor. Ent. p. 272.
I 4

120 CLASSIFICATION OF FISHES.

places the birds after the reptiles, the quadrupeds after
the birds, and the fishes after the quadrupeds, without
having the least suspicion that, although this series was
natural, it possessed another property, by which the
amphibians, the reptiles, and the fish, formed a primary
cirele of their own ; and thus reduced the three aberrant
divisions into one. The cartilaginous fishes, in short, unite
the aberrant divisions of the vertebrated circle into one ;
while, at the same time, they open a passage to the
quadrupeds by means of the whales, the dolphins, and
the porpoises. If the student wishes to comprehend
this double affinity, let him compare the figures of the
Ichthyosaurus with that of a shark, and he will be
immediately convinced that no reptile so much resembles
a fish as does the fchthyosaurus: again, if he looks to
the porpoise, its resemblance to the cartilaginous fish is
so peculiarly striking, that he will be not at all sur-
prised at the older naturalists placing them in the same
class.

(107.) The views we have taken of the cartilaginous
order in other respects are so different from those of M.
Cuvier, that we deem it necessary, in this place, to explain
our reasons. Although the arrangement of this order in
the Régne Animal is confessedly artificial; it is liable te
much fewer objections than usually attend such methods,
because the two typical divisions (the sharks and the
rays) are so peculiarly marked, that upon this point
there never had been the least difference of opinion.
The only objections, therefore, that may be made to his
remaining series, regard the aberrant groups: It is
quite evident, that if all fishes whose bones are car-
tilaginous are to be placed in this order, the genera
Leptocephalus, Lophius, Cyclopterus, and several others,
have as great a claim to be associated with the sharks
and rays as Petromyzon; while, if we extend the
order to such as have the branchia so hid, that they only
present an external slit, the order must be enlarged
so as to include the eels and several cognate genera.
Both these principles appear equally objectionable ; the

°

CARTILAGINES. — GENERAL CHARACTERS. 121

more especially as we should then cast aside all regard
for outward form, by which Nature, as it were, stamps
the most obvious and tangible affinities of her own
groups. The lampreys, indeed, have a second cha-
racter in common with the sharks and rays ; which is,
in having more branchial apertures than any of the
other eel-like fishes of the order Apodes: but when
We see, even in the same genus of sharks, that the
number of these orifices is by no means constant, and
that in the sturgeons and the chimeras, regarded by all
writers as true Cartilagines, these orifices are only one
on each side, as in the Murenide, it becomes obvious
that number alone is but an inferior character, ‘and
cannot be considered as a primary distinction even of
a genus, much less of an order. These considerations
are sufficient to excite very strong doubts on the pro-
priety of placing Petromyzon in the present order. If
we look again to the relations of these two groups,
this opinion receives additional strength. The affinity
which the cartilaginous fishes bear to the aquatic order
_of quadrupeds — that is, to the whales and the porpoises
—is too well known and acknowledged to be here de-
tailed ; while that between the lampreys and the red-
blooded worms is no less evident: both these affinities,
indeed, have been acknowledged by Cuvier; and it
therefore follows as an inevitable consequence, that these
two groups of fishes must be kept distinct,—the car-
tilaginous being placed nearest to the Mammalia, while
the lampreys are arranged so as to form a passage to
the Annulosa, by means of the Annelides, or red-blooded
worms. Cuvier, indeed, well observes that the “‘lampreys
have a skeleton so defective, and such simplicity of
organisation, that we might almost arrange them with
the worms:” they are, in short, if not “ the most im-
perfect of all vertebrated animals,” at least the most
imperfect of the entire class of Pisces. Excluding,
therefore, the Cyclostomi Cuv. from this order, we find
that the remainder of our author’s Chondropterygii form
a natural group; the primary divisions of which we

122 CLASSIFICATION OF FISHES.

‘shall now endeavour to make out, and subsequently
demonstrate.

(108.) Of all the cartilaginous fishes yet discovered,
that which seems to make the nearest approach to the
osseous orders, is the Polyodon reticulatus (fig. 7.), a

most extraordinary fish, about a foot long, found in
the Mississippi. It is at once known by the excessive
prolongation of the snout, which is very flat and lan-
ceolate, or broadest in the middle, while its length is
nearly equal to that of the whole body. The skin is
smooth and destitute of scales. The general structure
shows an affinity to the sturgeon, close to which it has
always been arranged; but it differs from that genus in
some important particulars, besides presenting a totally
different form. The maxillary and palatal bones, indeed,
are united; but the pedicle of the mouth has two
articulations. The mouth itself is wide, and is furnished
in the upper jaw with a double, and in the lower with
a single, row of small, but sharp, curved and serrated -
teeth. In all these respects, however, we still have the
general characters of a cartilaginous fish ; but by its
other characters we trace its connection to those whose
bones are osseous. The spiracle, common to the rest
of this order, is so large as to assume the appearance of
the branchial aperture of ordinary fishes; for both
Lacepede and Cuvier affirm that it extends to the
middle of the body. It is covered by a very large,
soft, and pointed operculum, which, on being raised,
exhibits the gills, consisting of five cartilaginous lamina,
with fringed edges, as in the generality of fishes. Like
Acipenser, there is a large swimming bladder: the in-
testine is provided with the spiral valve common to this
order ; but the pancreas, according to Cuvier, exhibits

THE STURGEONS. 123

the commencement of a subdivision into lobes ; in other
words, makes a departure from the cartilaginous structure,
and the nearest approximation yet discovered to the
more complicated form observable in all the osseous
orders.

(109.) The Sturgeons (Srurionipa) form the next
aberrant group, of which, at present, only one genus
is known. All the species are distinguished by being
defended, as it were, by armour, or, at least, having the
body covered by hard bony tubercles. The mouth is
small ; but instead of teeth it is furnished with a horny
prolongation of the jaws, which perform the same office,
and are analogous to what we see in the cheloniform
fishes. The mouth, however, has this peculiarity ,— that,
by its possessing a style with three articulations, it has
the power of being protruded and retracted at pleasure.
The gill-cover is of one oval radiated plate; but the
aperture is comparatively small, and its cover, by being
edged with a membranaceous border, closes the aperture
so accurately as to exclude the air. The food is small
fish and worms.

-(110.) The common sturgeon (A. sturio Linn., fig.8. )

affords that well known delicacy called Caviar, which is,
in fact, the roe of this fish properly prepared and dried.
It is usually inclosed in wax, and in this state is sent
to all parts of Europe. Sturgeons grow to a very
large size, many having been caught that measured
more than twenty feet long. Its form is lengthened
and slender; the snout very long in some species;
and the mouth, as in nearly all the cartilaginous
fishes, placed beneath. Several cirri, or worm-like
appendages, are seated beneath the muzzle, and near
the mouth: this latter consists of a transverse oval

124 CLASSIFICATION OF FISHES.

orifice totally destitute of teeth, but containing a thick
and strong tongue; it is bordered, both above and be.
low, by a strong cartilaginous edge or lip, which has
the power of retracting and closing at pleasure. The
whole body, which is pentagonal, is more or less covered,
according to the species, by strong, large, bony tubercles ;
thus forcibly calling to mind, both in its covering and
the construction of its mouth, the toothless quadrupeds
(Edentata). Sturgeons are natives of the northern
European and American seas; they migrate, during the
early summer months, into the larger rivers and lakes,
and, after depositing their spawn, return again to the
sea. The North American sturgeons may almost be
called freshwater fishes, since they are rarely taken at
any great distance from the shore. In some of the
rivers of Virginia they are so numerous, that Pennant
affirms 600 have been taken in the space of two days,
by merely putting a pole into the water, with a strong
hook at the end, and drawing it up again on perceiving
that it rubbed against a fish. There are regular stur-
geon fisheries, during summer, near Pillau, and in the
river Garonne, on the coast of France. Its flesh is
described as delicious, both as regards delicacy and
firmness. In this country, sturgeons are much more
rarely met with than formerly ; the largest ever taken,
according to Pennant, weighed 460 pounds. . The
fish, when roasted, is said to resemble veal; but that
which we receive from the Baltic and North America is
generally pickled. The sturgeon was a fish in high
repute among the Greeks and Romans: Pliny informs
us it was brought to table with much pomp, and orna-
mented with flowers; the slaves who carried it being
also adorned with garlands, and accompanied by music.
A smaller species, called the sterlet (Acipenser Ruthenus),
found in Russia, is in much higher esteem for the table
than the common species. The soup of this fish formed
one of the favourite luxuries of that gigantic epicure,
prince Potemkin of Russia, who, as Dr. Shaw relates, in
seasons when this fish happened to be unusually dear,

THE CHIMZERIDZ. 125

was contented to purchase it at a price so extravagant,
that a single tureen, forming the mere prelude to his
repast, cost him the sum of 300 rubles*; a sum, we
may add, which, had it been expended in promoting
the happiness of his miserable serfs, might have called
down blessings on the head of this worthless sensualist.

(111.) The third division is represented by the Cut-
MRID#, or sea monsters (fig. 9.), so called from the fan-

tastic shape of their heads, which are ornamented, if this
term may be used, with a singular hoe-shaped appendage
tipt with spines, and analogous to a crest, upon their snout:
in other respects they have the “ closest relation,’ as it
has been well observed, to the sharks ; from which, how-
ever, they essentially differ, in having a still smaller
mouth : the palatine and tympanic bones are merely rudi-
mentary, and suspended to the sides of the muzzle, which
is much advanced, while the upper jaw is represented
only by the vomer. The Chimera borealis (fig. 9. a) is
the chief of three species, remarkable for the singularity
of its appearance, which gives as much the idea of a
reptile as of a fish. It grows to three or four feet long.
The head is very large and obtuse; but the body termi-
nates gradually into a long and slender filament. In

* Gen. Zool. vol. v. p. 377.

126 CLASSIFICATION OF FISHES.

reference to the natural affinities of this extraordinary
fish, the head deserves particular attention: it is very
large, thick, and rises in the shape of a conical py-
ramid: at a little distance from the tip of the snout,
in the male fish only, is a short upright pro-
cess, terminating in a fringe; the whole resembling
a tuft or crest: the mouth is placed rather beneath,
and is small for the size of the fish; it has no true
teeth, but the jaws are furnished with broad bony
lamine ; and these are notched in the margin, so as to
resemble numerous small teeth ; while in front, both
above and below, stand two large, subtriangular, fiattish
cutting teeth. We see, in short, the first indication of
the plectognathiform structure, and of all those other
groups where the teeth are represented by crenated or
serrated bony jaws, analogous to the chelonian reptiles.
The northern Chimera lives in the deep recesses of the
ocean, and is therefore seldom seen to approach the
shores, except during breeding time. It is described as
a nocturnal fish, chiefly searching for its prey at that
season; when it devours the young of the cod, herring,
and other similar tribes. Its flesh is particularly coarse
and uneatable ; but the Norwegians are said to esteem
its eggs, which are mixed up with their pastry. Much
oil is contained in the liver. The C. Australis (fg. 9. b)
inhabits the Southern Ocean. Having now enumerated
the most aberrant forms in this order, we shall proceed
at once to those which are more typical.

(112.) The Seuatipa, or sharks, are the most con- -
spicuous and the most perfectly organised of all the
cartilaginous fishes. Their forms are often gigantic,
and their fierceness and voracity are proverbial: they
are the dread and detestation of mariners; and even
when dead, their aspect is sufficient to excite fear. These
monsters of the deep are nearly all completely carni-
vorous ; and their appetite is so voracious, that they in-
discriminately devour whatever living being comes in
their way It is a well-authenticated fact, that some
of these monsters, at a single bite, have cut a man in

THE SQUALIDE, OR SHARKS. 127

half; and an entire human body is said to have been
found, on one occasion, in the stomach. Fortunately,
however, very few of those found in our temperate
latitudes grow to such a size as to awaken our fears, or
commit injury upon our persons ; but so soon as we
enter the warmer regions, towards the tropics, bathing
in the sea becomes a hazardous, and often a dangerous,
undertaking. The late sir Brook Watson is well-known
to have had his leg amputated by one single bite of a
shark, while bathing in the West Indies: and both
there, and on the opposite coasts of Africa, the ocean
swarms with them. A very few species, however, feed
upon animals that are already dead, and even upon
marine plants. They all swim with great velocity, and
often in vast multitudes, when pursuing shoals of other
fish. Our excellent ichthyologist, Mr. Couch, says he
has heard of about 20,000 of the picked dog-fish
(Spinax acanthias), having been taken in a Cornish
net, called a sein, at one time ; and such is the strength
of instinct, that young ones, not six inches long, are
found, in company with their parents, following shoals
of fish, on which, at that age, they could not prey.*
(113.) The form of all the sharks is lengthened ;
the body and fins being covered with a hard coriaceous
skin, often tuberculated, and sometimes intermixed with
spines or plates; but none have been yet found with
true scales. The substance called shagreen is no other
than the prepared skin of these and other cartilaginous
fish, the different degrees of roughness indicating different
species. The head is always more or less flattened,
generally wider across than the body ; and sometimes,
as in the hammer-headed sharks, enormously dilated.
The snout, more especially, is dilated, and always ad=
vances t considerably beyond the mouth, which is thus
concealed beneath, and can only be seen, or indeed
used, when the fish is turned on one side: this is pre-
_ cisely the case with the rays; and renders it necessary

* Yarrell’s Brit. Fishes, vol. ii. p. 401.
¢ Except in the most aberrant forms.

128 CLASSIFICATION OF FISHES.

that both should turn almost upon their backs, in order
to seize their prey. The teeth of the shark exhibit,
perhaps, the most formidable apparatus for devouring,
of any animal in creation. In some species they are
so numerous, that, upon opening the mouth, the eye
sees nothing but a forest of pointed teeth, any one of
which, if detached, would be sufficient to inflict a most
severe wound: some of these are for the purpose of
seizing, others for tearing; but there are none for grind-
ing, as the food of the shark is always swallowed in an
entire state: the only exceptions to this general rule are
found in those genera (Pristis and Mustelus) which
form the passage to the rays, and where the teeth are
flat, blunt, and tesselated. All the other sharks have
pointed teeth, but differently modified in their form ;
and this diversity, as implying difference of food or
habit, deserves much attention. The gill-covers, as
already observed, do not open as in ordinary fishes: the
branchia, in fact, are completely concealed beneath the
skin ; yet their number may be judged of by certain
oval perforations, placed in a single row on each side,
through which the water is emitted in the act of respir-
ation. Let us now proceed to examine this family in
more detail.

(114.) To professor Rafinesque* belongs the honour
of being the first who ventured to break up the old
Linnean genus Squalus into a number of others ; to all
of which he has attached well-constructed names, and,
in most cases, very satisfactory descriptions. This re-
formation was begun many years before the appearance
of the Régne Animal; but the name and works of Ra-
finesque were then so little known, that M. Cuvier was
ignorant that nearly all his divisions had been anticipated.
As the work wherein these genera were first charac-
terised, is now become scarce, and as Rafinesque’s names
have the undoubted priority of all others, we shall here
lay them before the reader in his own words, more par-
ticularly as he describes two or three which still re-

* Caratteri di Alcune Nuovi Generi, &c. Palermo, 1810.

SQUALID. —RAFINESQUE'S GENERA. 129

main entirely unknown to all naturalists who have
followed him.

1. G. Carcuartias. No spiracles: dorsal fins two ;
anal one: five branchial apertures on each side : tail ob-
lique, unequal.— Obs. This genus is the first among the
Squalini, and contains those species which are the largest
and the most voracious. It is strikingly distinguished
from the genus Galeus, by the absence of spiracles.

2. Dauatias. No spiracles: two dorsal fins, but
no anal: five branchial apertures on each side: tail
unequal, oblique. This genus differs from the last by
wanting the anal fin ; and from that.of Squalus proper,
by the absence of spiracles. Teeth flat, long, acute,
disposed in a single row on the under jaw, and in two
on the upper, where, also, there are others much smaller:
eyes round: the branchial apertures are rather large.
Two species are described, — D. sparophagus and noc-
turnus. The latter has the anterior part of the dorsal
fin spined, and the posterior acuminated ; the head has
numerous pores: habits nocturnal: length seldom above
three feet: the teeth are unequal, acute, disposed in
various ways: dorsal spine united half way to the fin:
branchial apertures narrow. The pores on the head are
very remarkable: they are easily seen, although very
small ; and are round, unequal, and irregularly scattered
on each side of the head, from the tip of the snout to
above the eyes.

3. Trtroras. No spiracles: two dorsal fins, and
one anal: branchial apertures rather large, four on
each side: tail unequal, oblique.

4. Isurus. No spiracles: dorsal fins two, the pcs-
terior adipose; anal fin one, adipose: branchial apertures
five on each side: tail vertical, equally divided, and
lunulate. This genus is remarkably distinguished from
all others in this order, by the form of its tail,—a form
which is not seen in any other, and from which the
name is derived.*

* The only species known to our author is described in the following
words, where he introduces those other characters which belong to the

VOL I. K

aire 4
Pe abe '
| |

Crerictius. No spiracles: two dorsal fins; the
posterior much larger and bilobed; anal one: branchial
apertures five on each side: tail unequal, oblique:
head with two bony appendages in the form of horns. —
Obs. The two appendages, or rather horns, which this
genus bears on the head, give it an aspect of great sin-
gularity, and readily distinguish it from the next genus.*

6. Anopras. “No spiracles : two dorsal fins; the pes-
terior adipose; the ventral is single, and also adi-
pose: branchial openings five on each side: tail as
long as the body, oblique, unequal.” Of this one spe-
cies, A. macrourus, is described, which, our author
remarks, has some affinity to Galeus vulpecula, or
Squalus vulpecula Linn.; but is distinguished by the
absence of spiracles, by its adipose fins, its greater
size (12 or 14 feet), &c. The mouth is small; the
teeth are minute, acute, flat, and disposed in different
ways tf; the eyes are large and much sunk.

7. Heprrancuias. No spiracle: a single dorsal and
anal fin: branchial openings seven on each side: tail
unequal, oblique. Our author does not describe, or ap-
pear to have seen, the only species he thinks belongs to
this genus, which, he says, is the Squalus cinereus of
Lacepede.

8. Gatzuus. Spiracles two: two dorsal fins, and one
anal: branchial apertures five on each side: tail un-
equal, oblique. — Obs. The greater part of the Squaii of
authors are now placed in this genus, which is distin-
guished from that of Squalus (as restricted by our author)
by the presence of an anal fin.

130 CLASSIFICATION OF FISHES.

genus. IsuRUS oxyrynchus.— ‘Grey above, white beneath : snout very
acute: lateral line apparent, and rather curved : base of the tail angulated,
and nearly winged on each side : the branchial apertures are very long and
narrow: each jaw has three rows of teeth near the palate: eyes small and
round: the hinder dorsal opposite the anal. It grows to the length of ten
feet, and is called Pesceetondo.”

* The only species enumerated of this most extraordinary genus (which
seems absolutely unknown to all succeeding writers), is thus described : —

“°C. macrourus. Above bluish black ; beneath white : appendages obtuse,
recurved towards the eyes: tail forming one third the total length. This
rare fish is called by the Sicilians Pesce dzavolo, on account of its horns.
One was caught off Palermo in March, 1806, which measured eight feet,
and weighed six Sicilian cantars.”

t+ ‘* In diversi ordini.”’

SQUALID. — RAFINESQUE S GENERA. 131

9. Hexancnus. Spiracles two: branchial apertures
six on each side: dorsal and anal fin single: tail un-
equal, oblique.—Obs. The lesser number of the branchia,
and the presence of spiracles, distinguishes this genus
from that of Heptranchias, although both have a single
dorsal fin. This is founded on the description of the
Squalus griseus of Lacepede ; which, as it is not a native
of Sicily, nor appears to have been seen by our author, he
does not describe.

10. Ermoprerus. Spiracles two, round: dorsal fins
two, laciniated—the first armed with a spine, the second
nearly opposite to the vent ; no anal fin: branchial aper-
tures three on each side: tail unequal, laciniated, ob-
lique: muzzle obtuse: nostrils with appendages: teeth
small and acute: eyes oval, and deep sunk. This genus,
and the Squalus squatinus Linn., have the least number
of branchial apertures among the whole of the Squaii.

(115.) It is much to be regretted, at the present day
that some of these genera have not been more fully de-
scribed: but it is also true that most of these descrip-
tions are sufficient to identify both the genera and the
species ; and that they are even more precise than those
which were in use twenty-five years ago. It is very
easy to attempt to reconcile some of these genera with
others of their congeners, by attributing imaccuracy to
the author ; and this has been done, in numerous in-
stances, by Cuvier — with what degree of truth will
hereafter appear: but even if we suppose our author
may have overlooked some points, and have been mistaken
in others, there are, nevertheless, some of these genera
whose structure is altogether unique, and too remarkable
to be either confounded or misrepresented ; among these
are Isurus, having an equal lunate tail; Dadlatias,
having spiracles, but no anal fin; Cerictius, possessing
horn-like appendages ; and Etmopterus, with only three
branchial apertures.* Until the existence of such fish

* Upon this alleged fact, M. Cuvier says, ‘‘ Our author is most probably
mistaken, for he describes the Squalus squatina of Linnzus as also having
but three, whereas it has five.”” But before we can make up our minds on
this subject, it will be necessary to show that there is not a species, also,

K 2

132 CLASSIFICATION OF FISHES.

is absolutely disproved, we see no reason for consider-
ing them as purely imaginary. Professor Rafinesque
resided five years in Sicily, and therefore had far better
means of discovering its rarer productions than na-
turalists who have merely staid there for a few months.
Some of these genera, we have no scruple, therefore, of
adopting, while others may be held in abeyance until
they are verified by further observation.

(116.) If we look to the different genera in which
authors have divided this family, with a view to deter-
mine those which are more typical, and such as are
aberrant, we shall have but little hesitation in fixing
upon Pristis and Zygena as forming two of these;
while most authors agree in bringing Squatina also
into the family: this is in accordance, also, with the
views of Cuvier, who has separated the hammer-headed
group from all the other sharks, and placed Pristis and
Squatina in the same rank. There yet remains, how-
ever, the great bulk of the family under his genus
Squalus: these are obviously the most typical sharks,
and, like all such assemblages, contain two distinct
groups or sub-families, which we shall here term the
Squaline and the Centrine; the first being distinguished
by the absence, and the last by the presence, of spi-
racles. These are small temporal orifices, which, when
they exist, are placed immediately behind the eye: their
peculiar use is not clearly known, but they must un-
questionably perform an important office in the economy
of these fishes; because, from their universality in one
of these typical groups, and their absence in the other,
it would seem that nature intended thus to distinguish
them. The two aberrant genera of Pristis and Squa-

in Sicily, with only three apertures, which Rafinesque has supposed to be
the sqguatina of Linnzus, and so described it. Now I think that the ex-
istance of such a species is just as probable, if not more so, as that Rafi-
nesque has overlooked two of the spiracles. I can bear testimony to the
peculiar tact and unwearied zeal of our author, in detecting species closely
allied to each other. I must here again repeat, and the proofs will follow,
that not one half of the Sicilian fishes, described by Rafinesque, were known
to M. Cuvier, who has not only omitted them in his great work, but thrown
discredit on their very existence.

ZYGANINE, OR HAMMER-HEADED SHARKS. 133

tina contain very few species; and they are so much
isolated, when compared with the graduating links of
connection seen among the true sharks, that their pre-
cise situation in the circle is still open to dispute. We
have to regret, also, the same paucity of forms between
these aberrant Squalide and the three aberrant families,
or rather types of families, already noticed; so that,
whether the true sharks (Squalide) are directly con-
nected to Chimera, or to Polyodon, is a question
impossible to be determined at present by simple ana-
lysis. It might, indeed, be thought, on a hasty view of
the subject, that Pristis leads immediately to Polyodon:
but all authors agree, and we think justly, that this sin-
gular fish connects the sharks to the rays; and this will
be apparent when we come to describe it. Squatina,
also, has more the aspect of a ray than of a shark. Zy-
gene, therefore, is that group of the Squalide most
removed from the Raide; and it must, consequently,
stand at the furthest extremity of its own family,—in
other words, at that point which is in the line of pas-
sage to Polyodon. With this group, therefore, we shall
now commence our survey.
(117.) The Zyganine, or
Peel) » hammer-headed sharks (fig.
pre 10. a) present, at the first
glance,amarked and decided
character in the form of the
head, which, as their name ©
implies, may be compared to
a hammer, the body of the
fishrepresenting the handle;
in other words, the head is
flattened, with the sides so
much prolonged that the
eyes, which are at the ex-
tremities, appear placed on
two great peduncles. Cu-
vier remarks that the ani-
mal kingdom presents no
K 3

134 CLASSIFICATION OF FISHES.

other example of ahead so formed ; but this is incor-
rect: the genus Diopsis (b) among insects, gives us a
perfect representation of these hammer-headed sharks:
the resemblance, in short, is so striking, as alternately to
excite our wonder and our risibility at seeing a fly so like
a fish (fig. 10.) In other respects, we find the structure
in general accordance with the rest of the true sharks.
There are no temporal spiracles; but the teeth are
strong and acute, crenated on their edges, and placed in
three rows. The female is oviviparous. The species
are few, and these not well understood. The Z. mai-
Jeus is that which is best known: it inhabits the seas of -
Southern Europe, and grows to twelve feet long. A second
is found in India; and what seems a third is peculiar
to the Australian seas. The most typical species, how-
ever, yet known, has been recently discovered and de-
scribed by Dr. Cantor * as the Z. daticeps (fig.11.). These

are all typical examples; but the aberrant forms, which
have the head more heart-shaped, it will be necessary to
place in a distinct genus.

(118.) The second type of the aberrant sub-families
is the genus Pristis, or saw-fish. This genus has
been placed by all writers between the sharks and
the rays: and with great truth, for it partakes almost
equally of the structure of both— uniting, however, a
peculiarity altogether its own. This consists in the

* An acute and most zealous naturalist, whose materials for elucidating
the fish, serpents, and mollusca of India are particularly valuable; the
drawings and descriptions having been made from the living subjects.

PRISTIS, THE SAW-FISH. 135

enormous prolongation of the snout (fig. 12.), which is
straight, flat, and nearly of equal breadth throughout :

12

— ee = J tf eee = eee !

the tip is obtuse; but the sides are armed with a
single row of strong acute spines, pointing outwards so
as to resemble a very wide-toothed comb.* Cuvier
observes, that the anterior sides are sharp or cutting ;
but this is certainly not the fact in regard to such as
we have examined. With this weapon, as it is said,
the saw-fish attacks its prey, and even encounters the
large Cetacea, or whales. The mouth, placed quite
beneath the snout, is furnished with small rounded
teeth, close together, as in the rays; and, as in that
family, the branchial apertures are placed beneath the
pectoral fins. It possesses, also{, another character of
the rays, in the nasal cartilage already alluded to. On
the other hand, its affinity to the sharks is shown in
the general elongated form of the body; but more
especially by that peculiar character, which distinguishes
the Squalide, of having the pectoral fins totally free
and unattached to the head or snout,—-a formation, how-
ever, which is likewise seen in Squatina. Nevertheless,
the pectoral fins in Pristis are not dilated from the
base, as in Squatina; and the general structure of this
and all the other fins is precisely the same as those of
the true sharks. The temporal orifices are large, and
placed behind the eye; while the teeth, in the gene-
rality of the species, are flat and _ tesselated; the

* A species now before us, from nlieopieal America, has no less than 28?of

these teeth on each side the snout ; it is probably the Pristis pectinatus.
+ Mentioned by Drs. Muller and Henle.

K 4

136 CLASSIFICATION OF FISHES.

mouth and the branchial apertures being placed com-
pletely beneath, — the former under the eyes, the latter
under the pectoral fin,—so that neither of them can be
seen when the fish is laid upon its belly. These fish,
of which there are several species, grow to a large size,
and appear to be pelagic, or rovers of the ocean. Some
inhabit all latitudes, from the coldest to the hottest ; but
no doubt each species has its peculiar geographic range,
although none have yet been found on our own coasts.
The common species is said to attack whales much in
the same manner as the sword-fish ; although it is ob-
vious that the snout, being calculated to cut laterally,
and not to thrust, must be used as an offensive weapon,
in a very different manner: for this reason, we do not
believe the assertion that some writers have made, that
the snout of the saw-fish has been found driven into
the sides of ships like those of the sword-fish ; because
any one who looks to this snout, and observes that the
end of it is quite. blunt, must see such a thing to be
altogether impossible. The species often grow to be-
tween 18 and 20 feet long, and are chiefly distinguished
by the number and form of the tooth-like processes on
each side. Klein first made known the singular fact,
that in the foetal or young saw-fish the snout is folded
back over the head, and the rudiments of the spines are
indicated by tubercles.

(119.) The third of the aberrant forms in the great
family of sharks, is either represented by Squatina, or
by Crossorhinus ; both of which differ from all the
more typical sharks, in having the mouth at the extre-
mity of the muzzle, and not beneath it. That these
two forms, as well as Cestracion, are perfectly analogous,
there can be no question; the only difficulty is that of
determining their natural situation or affinity. We
confess our strong doubts on the propriety of placing
Squatina among the sharks, rather than with the rays,
to which it has certainly, of the two, the greatest
resemblance both in external form and internal structure.
Without, however, entering further into this question

SQUALIN E.——SQUATINA. 137

at present, we may certainly affirm that the general
shape of these animals (of which two species are known)
seems a compound of both the typical forms. The
hinder parts are those of a shark, while the broad de-
pressed head is that of a ray, and it is just of such a
form as we might imagine to intervene between a
Torpedo and a Rhinobates. Our European species,
C. angelorum, is generally called the angel-fish. The
body, but especially the head, is flattened ; and the eyes,
like those of all the rays, are vertical, or placed upon
the crown: behind these are spiracles; while the bran-
chial apertures are not, as in the sharks, on the sides, but
placed beneath: the pectorals are very broad, and the
mouth terminal. Inthe American species (fig. 13.), the

upper jaw has two flattened and somewhat triangular
cirri: the teeth are broad at their base, but slender
and sharp at their points. The Squalus aculeatus of
authors has been also referred to this genus, to which
it is evidently related; but whether by analogy or
affinity appears comewhat questionable. For the pre-
sent, we feel disposed to follow our predecessors in
placing Squatina in this family; where, if it truly
enters, it comes in as the chironectiform type of the ©
whole. circle. Leaving the three aberrant groups, we
shall now proceed to the two which are typical.

(120.) We place the sub-family of Sguatin™ as the
next in order, because it seems connected to the Zyga-
nine by its pointed teeth, and by the want of those
remarkable temporal orifices, or spiracles, which seem

138 CLASSIFICATION OF FISHES.

to mark the primary distinctions of the two typical
groups. This character also happens to be one of the
most obvious; and thus affords the ichthyologist an

easy and, as we believe, a natural guide among the in-

tricacies of the numerous genera that have been formed
out of these fishes. On looking to these, we plainly
perceive that, although they have hitherto all been
termed genera, and therefore placed upon the same rank,
yet that some are much more strongly marked in their
differences than others; so that they form themselves
into little groups, under which two, three, or more, may
be arranged. This it would be very easy to accomplish,
if our object was merely to make an artificial arrange-
ment: but when we attempt to work out a natural
group, such is the state of ichthyological science, that it
almost becomes absolutely necessary to verify what has
been done by our predecessors, by going over the same
ground, and re-examining the major part of these sub-
genera ourselves. In very many instances, however,
this is totally impracticable; and in such cases we have
no other resource left than to take for granted what has
been published, and endeavour to trace the line of affi-
nity by the imperfect materials before us. In the at-
tempt, therefore, which we shall now make to place the
numerous sub-genera of sharks under their genera, pro-
perly so called, the above difficulties must be borne in
mind, and every allowance may fairly be claimed for those
errors which necessarily attend upon a task so peculiarly
perplexing. Enough, however, will come to light in
the sequel, to show that this effort has not altogether
been unsuccessful ; and for the rest, we must leave the
rectification of minor errors of location to time, — to
‘greater knowledge of those forms already known, but
imperfectly described, — and to the discovery of others
which are at present unknown.

(121.) The first genus, if such it be, which we shall
notice, among the Squaline, or sharks having no tempo-
ral orifices, is that of Scoliodon of Muller and Henle,
which seems to bear a nearer affinity than any other to

SQUALIN HE. — SCOLIODON. — SQUALUS. 139

Zygana. We are led to believe that Scoliodon is a
generic, and not a sub-generic, type, from its containing
five species; but as no typical example has been named,
and no notice taken of the form of the head and tail
(characters, in our opinion, of much greater importance
than slight variations in the teeth), our idea of its rank
is entirely conjectural. If Rafinesque’s Tetroras, on
the other hand, has only four spiracles, it may fill the
place here assigned to Scoliodon; which, according to
Muller and Henle*, differs only from the true sharks in
the next genus, by such slight modifications in the teeth,
that, in the absence of further characters, we hardly
venture to incorporate it in our present survey. The
genera Tyiaenodon and Leptocharias, each with only one
example, appear to us — judging from the characters
that have been as yet assigned to them t— no other
than aberrant species ; but this, again, is mere conjec-
ture. It is clear, however, that they all enter into the
present sub-family, as they are destitute of temporal
spiracles.

(122.) The next is the typical genus of the whole
family; and as such we retain to it the original generic
name { of Squalus, in preference to that of Carcharias
proposed for it by Rafinesque, seven years before
M. Cuvier. Here we meet with the most ferocious and
gigantic monsters of the whole family ; among these is
the great white shark, Squalus carcharias, which some-
times grows to the length of twenty-five feet, and which
is a savage and destructive wanderer over the whole
ocean. Its jaws are armed with innumerable cutting
teeth, acutely pointed at their tips, and generally den-
tated on their margins, the base being very wide and

* Mag. of N. Hist. No. xiii. p. 35. — “ Differs only from Carcharias Cuv.
by the teeth being of the same shape in the upper and lower jaw; viz. the
points directed towards the corner of the mouth, with a smooth edge, and
a truncated protuberance, either smooth or indented, on the exterior side
of the base (5 sp.),”’

f b= p36.

t The propriety of retaining the original name of a genus to the typical
group, has been so well advocated by others, that any further observations
of ours would be superfluous.

140 CLASSIFICATION OF FISHES.

compressed. They all have an unequal tail, two dorsal
fins, and one anal. In the pre-eminently typical section,
or sub-genus, the tail is of ordinary length (Squalus
elephas Le Sueur, fig. 14.) ; but in the next sub-genus,

Alopias Raf., to which the Squalus obscurus of Le Sueur
forms a passage, the upper division of the tail is exces-
sively lengthened; a familiar example of which is seen
in the fox-tailed shark of Britain (Alopias vulpes Nob.).*
Independent of this singular development of tail, Alo-
pias is further distinguished by having the snout coni-
cal, not, as in Squalus, broad and depressed ; the teeth
also are less numerous, and are only in two or three
rows. The sub-genus Cericteus of Rafinesque is an
equally distinct, but a much more extraordinary, type,
hitherto found only on the prolific shores of Sicily T: it
is at once distinguished by having two horn-like osseous
appendages on the head, resembling horns; while its
affinity to Alopias is manifested by its oblique, unequal,
but very long tail. All the foregoing types have an
anal fin, and the two dorsal fins are soft; but in Dala-
tias nocturnus of Rafinesque, which he distinctly asserts
has no spiracle, the anal fin is wanting, and the two
dorsal fins are spined. It has been thought by Cuvier,
that the spiracles of this fish have been overlooked, and
that it is, in reality, a species of Spinax ; but we see no
good reason for this belief, and a strong one against

* Figured in Yarrell, vol. ii. p. 379.
+ Cericteus macrourus, Raff. Caratt. p. 12.

SQUALIN Z. — SELACHUS. 141

it. The habits of the fish, which Rafinesque particu-
larly mentions, show that it is a nocturnal feeder ; and
it therefore becomes highly probable that it possesses 2
membrana nictitans, similar to all the groups we have
hitherto noticed ; whereas both Spinaw and Centrina,
according to Muller and Henle, have not that appendage
tothe eye. At all events, we must not believe an author
is invariably wrong, merely because he may have com-
mitted occasional errors; for if we proceed on such a
principle, who is exempt? We shall, therefore, retain
the genus Dalatias, untilit is proved false; restricting it
alone to the D. nocturnus, and viewing it, for the pre-
sent, as the representation of the spiny-finned group of
Centrine, in the family of spiraculated sharks, to which,
in every thing but the absence of spiracles, it seems to
agree.

(123.) The genus Selachus Cuv. is the third of the
Squaline. It has several strongly-marked characters,
and appears altogether a very natural one. Unlike all
the preceding, the teeth of these sharks are conic, sim-
ple, and generally small; that is, not serrated or lobed.
The tail, which in all the preceding genera has been
unequally lobed, now assumes the more regular appear-
ance of ordinary fishes ; its form is lunate, the two lobes
being nearly equal. The third character is to be found
in the extraordinary size of the branchial apertures,
which are so large as nearly to extend half way round
the neck. These characters are developed in the sub-
genera Isurus, Selachus, and Lamna‘, the first of which
appears the true type of the group. We have now
arrived, however, at that extremity of the Squaline

* Oxyrrhina Agass. evidently belongs to this group, so remarkably dis-
tinguished by its teeth ; but I look upon it as not sufficiently distinct from
Lamna to allow of sub-generic separation. Carcharodon, formed on one
species, is unknown to me.

{ It would appear, according to Dr. Smith, that Cuvier has overlooked
the spiracles of his genus Lamna, which Dr. Smith says are present in
that group, although extremely small. There is thus as much uncertainty
regarding one of M. Cuvier’s genera, as in the Dalatias nocturnus of Rafi-
nesque. May not Dr. Smith have mistaken some of the numerous pores,
placed on the head of certain Lamne, for true spiracles? We have no
means, at this moment, of settling this disputed point. _

142 CLASSIFICATION OF FISHES.

which blends into the next sub-family of Centrine, and
Wwe are consequently prepared to expect that the chief
characters of the two groups also blend into each other:
in this expectation we are not disappointed, for we find
that the presence or absence of spiracles now becomes
quite a secondary character. In the true type, which
is probably Jsurus, there are no spiracles; but in Lamna
they first appear (perhaps not in all the species*) to be
minute ; and in Se/achus we still find them very small.
In all these, however, we see the three primary charac-
ters already noticed. Jsurus, however, stands promi-
nently forward as a most remarkable type, having the
snout so lengthened and pointed as to be a representative
of Pristis: in Lamna, the snout, although not lengthened,
is still pointed and conic ; and evenin Selachus, the muz-
zle, according to Cuvier, projects far beyond the mouth.
What other sub-genera enter into this group, we know
not; but it is quite clear that we have now a passage
opened to the spiracled sharks. Before, however, we
quit this division, we may advert to another form,
which seems entitled to be viewed in the light of a
generic type; for, although only one species is yet
known, its form is so remarkable, and so different from
all others, that it must either be placed with the Squatine,
or stand as the most aberrant genus in the present as-
semblage. We allude to Rineodon of Dr. Smith, having
all the characters, as it would appeart, of Selachus, but
with the mouth on the top of the snout. As this struc-
ture is totally at variance with that of the ordinary
sharks, excepting Crossorhinus and Cestracion, we may
fairly conclude, from the location that has been assigned
to it, that it has a relation both to those and to Squatina.
In the Crossorhinus lobatus M. and H.( fig. 15.) or Watts’s
shark, the mouth is also terminal, but the sides are fur-
nished with broad cirri, or lobes. This singular fish

* This supposition is highly probable, and will at once reconcile the oppo-
site statements of Cuvier and Dr. Smith.
+ Mag. of Nat. Hist. No. xiii. p.37. second series.

CENTRINZ.—SCYLLIUM. 143

certainly does not belong to the Squatine, or even te
the same genus, strictly so termed, as Dr. Smith’s
Rineodon ; for the teeth are large, acute, and seem more
to resemble those of our Squalus ; both of the dorsal

fins are placed behind the ventral ; the tail is long; the
caudal fin unequally and irregularly lobed: it only
agrees with Rineodon in its terminal mouth, and the
situation of the branchial openings, which appear very
large, and are all placed before the pectoral fin. Whe-
ther this singular fish naturally intervenes between the
Zyganine and Pristis, or whether it is the most aber-
rant type of the Squaline (in which case it would
represent Squatina and Rineodon), are questions which,
in the present confused state of this family, cannot be
determined.

(124.) We now enter on the sub.family of Crn-
TRINZ, or spiracled sharks, to which we are conducted,
as before observed, by the sub-genus Se/achus, which has
the general structure of Zamna, with the spiracles suf-
ficiently large to become obvious, sithough, when com-
pared to the sharks now before us, they still remain
very small.

(125.) The first genus we shall notice in the line
of affinity is Scyllium (S. canicula, fig. 16.), which,

although agreeing with Lammna in its obtuse and pyra-

144 CLASSIFICATION OF FISHES.

midical snout, and somewhat in its teeth, is at once dis-
tinguished from that sub-genus by its lengthened and
unequally lobed tail, which has the same form as in the
generality of sharks ; like them, also, it has two sof
dorsal, and one anal fin. The long snout seen in Jsu-
rus, appears to be, in some measure, continued in one
of the sub-genera (Pristiurus Bon.) which authors
have placed under Scyllium. These smaller divisions
have been chietly formed upon the different positions of
the dorsal fins, and other minor characters, which, how-
ever important and interesting they may be thought, do
not appear to us, taken by themselves, to lead to any
clear notion on the natural sub-generic types of the
group before us. As we shall notice them in our
synopsis, we need not, in this place, detail their technical
characters.

(126.) The genus Galeus has an obvious and close
affinity to that of Scyllium. They are almost exact pro-
totypes of the true sharks, except in wanting the tem-
poral spiracles: all the teeth are flat and sharp ; but
they vary so much in their minor modifications, that
Muller and Henle have divided this group into four
sub-genera, viz. Galeocerdo and Galeus, where the
teeth are serrated (2 species); Lowvodon, having no
serratures (1 species); and Tvriachis, with the teeth
pointed, as in Scyllium, and without a dimple at the
tail (1 species). The value of these distinctions will
no doubt appear more definite, when the learned and
able ichthyologists who have proposed them, publish
their views more in detail: until then we feel incom-
petent to arrive at any conclusion on the subject.

(127.) The third genus, Centrina, is much more
definite than the two last, and seems to be the most
natural in the present sub-family. It is composed of
all those spiracled sharks which have a spine placed
before each of their dorsal fins, while the anal fin is
entirely wanting: hence it differs from every other
group in this sub-family. But this structure is not

CENTRINZ.—CENTRINA. 145

reached abruptly: there are some which, by having no
spines, evince an affinity to Galeus; while, from being
destitute of an anal fin, they come within the confines
of the present group: these form Cuvier’s sub-genus
Scymnus, which we shall, at least for the present, keep
entire, since the divisions that have been made of it
appear to us* too slightly marked for even sub-generic
separation ; more especially as there are evidently five
divisions, with much more prominent characters, enter-
ing into this genus. Scymnus seems to represent Se-
lachus in some particulars well worth noticing. To
Dr. Scoresby we are indebted for all the knowledge we
possess of the habits of S. borealis, an immense species,
observed by that well-known navigator and philosopher
in the Arctic seas. According to his observations, it
often grows to the length of fourteen feet, and six or
eight feet in circumference. Its chief food is derived
from dead whales and other Cetacea, out of which, at a
single gripe, it scoops masses of blubber as large as a
man’s head: hence it is, that when, on such occasions,
any sailors may be in the water engaged in securing the
whale, this shark is so intent upon claiming his portion,
that he offers no molestation to the fishermen ; indeed,
he is so ravenously fond of blubber, that he has been
known to return to the carcase, even after a long knife
has been run into his body by the seamen engaged in
cutting up the whale. The slight variation in the teeth
of those species which we place in this sub-genus, seems
to mark the transition from the last genus. In Galeus,
the teeth in both jaws are serrated on the external edge,
and inclined outwards ; but in Scymnus, the upper teeth
are straight and narrow, while those in the lower jaw are
crooked, pyramidal, and equilateral: between these,
however, are species having the upper teeth of Scymnus,
and the lower onesof Galeus. -From Scymnus we pass
to one of the typical genera, both of which have the
dorsal fins spined: the first is Centrina, which, as

* Laemargus M. and H., Echinarrhinus Blainv.
VoL. I. L .

146 CLASSIFICATION OF FISHES.

Cuvier observes, has all the characters of Scymnus, but
with the addition of spines. The most common ex-~
ample of this type is the Squalus centrina of Linneus,
—a large, thick-shaped fish, having the spines not placed
in front of the dorsal fins, but partly concealed, in the
skin which covers them: the anterior spine points for-
ward, but the posterior is directed~backward, and the
tips of both are alone naked: the last dorsal fin is
placed over the ventral; and the tail is remarkably
short. We exclude from this sub-genus the Squalus
spinosus and squamosus of authors, the last of which
appears to us the true type of the sub-genus Centropho-
vus of Muller and Henle, the distinctive and most
striking characteristic of which is the body being covered
with hard carinated scales. The sub-genus Somnolentus
of Le Sueur seems to unite this singular form with
Scymnus, of which we consider it only as an aberrant
species. The fourth sub-genus, following Centrina, is
Spinaxw Cuyv.*, where we have again the ordinary form
of the sharks, but with each of the dorsal spines placed
in front of the fins: the snout is rather lengthened, the
tail long and unequal, and the teeth are small and cutting. ©
The most familiar and typical example of this group is
the Spinawx acanthias, or picked dog-fish of our own seas,
of which Mr. Couch has recently given us an interest--
ing account.f It seems to be the most abundant of all
the sharks found on the western coasts, where it is
sometimes seen in incalculable numbers, to the no small
annoyance of the fishermen, whose hooks they cut from
the lines in rapid succession. One of its modes of de-
fence is very singular, and is effected by bending itself
in the form of a bow, for the purpose of wounding
with its spines ; and then, by a sudden motion, it causes
them to spring asunder in opposite directions : so accu-
rately is this effected, that if a finger be placed on its

* It appears to me that the sub-genus Acanth7as Bonap. is the true type
of Spinaz, and that the single one to which Cuvier’s original name is
thus restricted, is but an aberrant species of Spinez.

+ Inserted in Yarrell’s British Fishes, vol. 11. p. 401.

CENTRIN &. — CESTRACION. 147

head, it will strike it, without piercing its own skin.
Its greatest size, however, seldom exceeds two feet. We
cannot subscribe to the supposition of M. Cuvier, that
Etmopterus aculeatus Raf.* is a typical example of
this genus ; for we know that the descriptions of this
author, as before remarked, were never taken from dried
specimens. The fifth of the most prominent divisions
of the spine-finned sharks is the sub-genus Cestracion
Cuy., which we have not yet seen. According to Mul-
ler and Henle, however, it has a prickle before each
dorsal fin; a fact established by the figure given of the
Cest. Phillippii by Lesson (fig. 17.), although not men-

tioned, and perhaps overlooked, by Cuvier ; which is
somewhat singular, as he himself originally defined the
genus.t In addition to this, the teeth are tesselated,
—those on the anterior rows alone, being small and
pointed ; while the mouth, unlike all the other sub-
genera of Centrina, is terminal, or at the extremity of
the pointed muzzle. The Cest. Phillippii is the only
species yet discovered : it is very rare, and inhabits the
coast of Australasia. It is not only analogous to

= “ Ftmopterus aculeatus. All the fins and tail as if laciniated; the
dorsal fins with a detached spine before each; the posterior one almost
opposite the anal. — This is the smallest of all the sharks I have seen in
Sicily, for it scarcely exceeds a foot in length, and is the only one not
eaten. The fishermen distinguish it by the name of Dzavolucchio de mari,
or little sea-devil. The snout is obtuse ; the nostrils are furnished with an
appendage ; the teeth small and acute ; the tail unequal and oblique; andthe
branchial apertures only three.” —Haf. Caratt.p.14. The Squalus uyatus
Raf., as Cuvier observes, is obviously a Spimaz, but seems to me to differ
from our northern Spinaz acanthias. p

+ With such conflicting statements as to simple matters of fact, as those
we have just been obliged to notice, it is almost impossible to determine
the limits of any one natural genus, or even of rigorously determining any
one point in the natura] arrangement of this family.

Lee

148 CLASSIFICATION OF FISHES.

Squatina, Crossorhinus, &c., by the mouth being ter.
minal, and the eyes vertical, but still further resembles
the first, in the lobes of the tail being nearly equal; so
that there can be but little doubt of its being the chi-
ronectiform type of the circle of spine-finned sharks.
The front view of the head, as delineated in M. Les-
son’s Atlas, gives the lateral ridges or elongated lobes an
appearance of horns ; but this merely results from the
peculiar position in which the head is drawn.

(128.) The genus Mustelus is the fourth of this sub-
family; and although, in its general form, it has a close
resemblance to Galeus, it is yet distinguished from all
the other spiraculated sharks by having the teeth fiat and
tessellated, like the rays and the genus Pristis: it is on
his accountt that we consider Mustelus as the represent-
ative of a genus connected, in the most perfect manner,
with that of Centrina, through the medium of Cestracion,,
which unites in itself the chief characters of both,
joined to a peculiarity of its own.

(129.) Regarding the fifth primary group of the
Centrine, much uncertainty prevails, on account of the
conflicting opinions of Rafinesque and Cuvier as to the
question whether Heptranchias has no spiracles, or
whether they really do exist, as asserted by the latter.
it is clear, however, that even if Lacepede, rather than
Rafinesque, is in error on this point, and that Heptran-
chias is but a sub-genus of Hexanchias, the latter name
has the priority over Cuvier’s Votidanus,—having been
published seven years before.* Leaving, therefore, the
presence or absence of spiracles in Heptranchias to be
determined hereafter, we may state that the genus
Hexanchus is distinguished by having no second dorsal
fin, and that it seems to contain two sub-genera: Hew-
anchus proper, having a depressed and rounded muzzle,
and six wide branchial apertures ; and Heptranchias,

x If there is an error in attributing no spiracles to the sub-genus Hep-
tranchtas, that error belongs to Lacepede, and not to Rafinesque, who
founds his generic characters entirely upon Lacepede’s account of his

Squaie perlun (Hist. des Poissons, p. 220.), without having scen the species
himself, which he does not describe.

ANALOGIES OF THE CARTILAGINES. 149

where the muzzle is pointed, as in Zamna, while the ~
branchial apertures, equally large, amount to seven: the
caudal fin, in both, is oblique and unequal.

(130.) We may here close our enumeration of the
most prominent variations in this extensive family ;
and we shall now take a retrospective view of the
whole. It has been our endeavour, with the imper-
fect and often contradictory materials before us, to
trace, in some degree, the real line of continuity, and
the manner in which the different forms blend into
each other. Some of these affinities are much more
obvious than others ; but as even these latter require to
be tested by the theory of analogy, we must now turn to
this sort of relationship as essentially necessary to give
some degree of verisimilitude to our arrangement of the

-Squalide, no less than that of the whole order. We
shall, in the first place, arrange the orders of fishes in one
column, and the families of the Cartilagines in another,
and then see how far the contents of each are analogous
in their most prominent characters.

Peas othe, Analogies. Orders of FIsHEs,
Raide. Back armed with spines. § ACANTHOPTERYGES,
Squalida. Back with soft fins. MALACOPTERYGES,
Polyodonide. Pre-eminently cartilaginous. CARTILAGINES.
Sturionide. Body mailed ; mouth very small. PLECTOGNATHES.
Chimeride. Taii excessiveiy lengthened. APODAL.

(131.) Before the naturalist enters upon the investiga -
tion of these comparisons, we beg to remind him of one
important consideration, that must always be borne in
mind in all investigations of this nature, namely, that we
are to look only to the pre-eminently typical characters
of each group, and not to the exceptions which always,
and inevitably, occur in those which are aberrant. It
is no more meant, for instance, that all the rays are
armed with stings, than that all the Acanthopteryges have
spined dorsals: here the absence of spines is the excep-
tion to the general character, just as their presence is the
exception among the Squalide and the Malacopteryges.

L 3

150 CLASSIFICATION OF FISHES,

We have illustrated this position so fully in the two most
perfect classes of vertebrated animals, and more especially
among birds, that it appears hardly necessary to touch
upon the subject in this place ; but as this volume will,
doubtless. be perused by many ichthyologists who have
not turned their attention to ornithology, it seems ne

cessary to explain to them the leading principle upon
which we universally proceed in analogical comparisons,
and upon which the essence of our theory depends. True
it is, that there are sharks with spined dorsal fins ; and
this fact would appear to invalidate the character we have
given to the family ; but it will be seen that these spined
sharks are not the most typical, and therefore they are
not taken into the account at present: the same may be
said of such as, from being covered with spines, are
analogous to the sturgeons and the cheloniform fishes.
In explanation of all which we may observe, that these
minor variations, belonging only to aberrant forms, are
to be explained by this simple law of nature,—that every
circular group, whether large or small, contains within
itself representations of all other groups; so that if,
among the sharks, there were none with spined rays,
there would be no representation of the order Acanthop-
teryges, and (unless other analogies to that order existed)
the sharks would be an imperfect circle. We hope the
experienced zoologist, to whom all this is well known,
will excuse our again explaining these views to the ge-
neral student, and we shall now proceed to the compari-
son above intimated.

(132.) The most typical forms of the rays, as will
subsequently appear, are those whose backs are pro-
vided with a formidable spine, usually, although im-
properly, denominated a sting. This weapon is placed,
indeed, upon the tail, generally near its base ; but so also
is the first dorsal fin in several types of the sharks; so
that it becomes no more nor less than are presentation
of the first or spinous dorsal fin of acanthopterygious
fishes. It may be here observed, that some of the rays
have two spines, analogous to the two dorsal fins of

ANALOGIES OF THE CARTILAGINES. 151

the Acanthopteryges. That the rays also are the most
typical of the whole order, may be inferred from two
circumstances. Of all the Cartilagines, they have
the broadest snouts, just as the fissirostral or natatorial
types, among birds, have the broadest bills; while the
peculiar form of their body, which may be said to be
surrounded with two immense fins, must give them a
greater celerity of swimming than is enjoyed by all
their congeners. Such is exactly the case with the fissi-
rostral and natatorial birds, of which the swallow, the
goat-sucker, the albatross, and the Tachypetes are familiar
examples — well known to every ornithologist — where,
as in the rays, the organs of flight considerably exceed
the size of the body. There can be no doubt, therefore,
that these analogies are founded in that law of repre-
sentation, which assimilates all these groups to one of
the primary types of the animal creation. If the rays,
therefore, represent the Acanthopteryges, the Squa-
lide, by .which they are immediately followed, must
bear a corresponding relation to the sub-typical order of
fishes ; the chief character of both consisting in their
having the fins soft. The genus Centrinus, indeed, is
furnished with spines: but it is clear, even upon the
bare opinion of Cuvier, that this genus is not typical of
the sharks ; that station being assigned by him to the
Squalus carcharias, and its allies, to which we have re-
tained the original patronymic name of Squalus. The
Squalide, therefore, by following the rays, become the
sub-typical family of the cartilaginous order ; and this
analogy at once explains the relation they bear to the
Ferg among quadrupeds, and the Raptores among
birds. Like these, their representatives, they are pro-
verbially the tigers and panthers of the ocean ; and fre-
quently carry upon them, as it were, the very spots and
markings of those ferocious beasts, as if Nature was de-
termined to make her analogies plain, whether they were
studied or not. ‘These relations of the two chief groups
being thus established, we must be satisfied if those that
are aberrant are less determinate ; because, as the forms
L 4

152 CLASSIFICATION OF FISHES.

contained in the order before us are very few, our ma-
terials for comparison are as 1 to 10 less numerous ;
and yet, upon study and reflection, we shall find that
the same train of analogies can be traced, although,
perhaps, they may appear to some less perfect than in
the instances already explained. But to proceed : —
(133.) It will be seen that the genus Polyodon, which
represents a family, stands opposite to the Chondropte-
ryges, or cartilaginous order. We place this genus close
to the sharks ; M. Cuvier does the same ; and, therefore,
the scruples of those who form their opinions on previous
authority will not be disturbed. But it may be im-
mediately asked, how can Polyodon, which departs in
so many points from the cartilaginous structure, be, at
the same time, a typical representation of that order?
To answer this, we shall cite an accidental remark of
M. Cuvier’s, which, in our opinion, at least, is quite
conclusive. In speaking of the Polyodon, he remarks,
that the spinal column merely consists of one entire
piece, like the lamprey. Now, as one of the greatest
characteristics of the order before us is to have the
spinal column cartilaginous, and less developed than in
any other order, so it results, that the most imperfect
fish, in this respect, among the whole of the known
Chondropteryges, is the Polyodon, which thus represents
them in its own circle. Did this peculiar construction
constitute the only character of the order, then, indeed,
Polyodon would stand at the head, and occupy that sta-
tion we have given to the rays: but this is not the
case, either in nature or in any system. Polyodon
has an enormous gill-cover, with a large branchial aper-
ture, nearly similar to the generality of fishes ; it is,
besides, furnished with an air-bladder ; and thus nearly all
other parts of its structure are directly opposed to the
idea of placing it at the head of the cartilaginous fishes,
merely on the strength of having one of their characters
uncommonly developed: thus, also, we see that every
fact regarding the anatomy of an animal, however bare
and barren it may appear, at first, of ulterior interest,

ANALOGIES OF THE CARTILAGINES. | P53

may yet become of the greatest importance in our en-
deavours to determine the different relations which
subsist between animals, whether by affinity or analogy.

(134.) Our next comparison is between the stur-
geons and the cheloniform fishes, or, in other words, the
Sturionide and the Plectognathes of Cuvier. This
analogy will not detain us; for the very aspect of the
two is quite sufficient to show us we are following in
the right track. The types of both are incased, as it
were, in armour ; the head and body being defended by
large bony plates, which either compactly join at their
sutures, as in Tetraodon, or assume the form and sub-
stance of little targets or shields having a sharp central
spine ; the mouth, in both groups, is very small; and
the absence of true teeth in both is supplied by an acute
elongation of the jaws. This latter character, which is
one of the primary distinctions of the cheloniform fishes,
is found still more developed in the next type of the
cartilaginous order; and this at once brings us to the
only remaining analogy, namely, that between Chimera
and the apodal or anguilliform fishes. Now, it may be
observed, that throughout the whole of the cartilaginous
groups which we have yet noticed, there is not one
which gives us any idea of that slender and attenuated
form which belongs to the eels among fish, and to the
serpents among reptiles; and yet in the Chimera we
actually see a fish having the fore part of a shark, and
the tail, or hinder part, of an eel. Thus does Nature
combine her primary forms : and yet, that analogy should
preserve a due subordination to affinity, the primary ©
characters, as well as the whole aspect, of these singular
shaped fishes, are decidedly those of the true Cartila-
gines, yet so modified as to point out its relations to
other groups. Of all the cartilaginous fishes yet dis-
covered, the Chimere are those only that have the
second dorsal fin very narrow, excessively long, and
gradually tapering to the point of the tail ; being all but
united to the caudal fin. This latter character, as is
well known, pervades the whole of the anguilliform

154 CLASSIFICATION OF FISHES.

fishes ; and both agree in having only one small external
branchial opening. The other peculiarities of Chimera
relate to its three affinities — on one side to Acipenser,
on another to the Raide, and on a third to the Plecto-
gnathes, — all of which will be noticed in their proper
place.

(135.) Such are the analogies, resulting from our
following closely the line of affinity, and upon which
we rest our belief that the five types of the cartilaginous
order represent the five great divisions of the class
Pisces : but on a question of such paramount importance
to the philosophic naturalist, it appears necessary to take
a still wider range ; and, by looking to the whole circle
of vertebrated animals, endeavour to test the correctness
of this series by bringing it into comparison with the
great groups of the Vertebrata. The resemblances, of
course, will be far more remote, because the dissimilar-
ities are immeasurably greater ; but yet, if our arrange-
ment is true to nature, these resemblances, however
faint some may think them, must not only exist in part,
but must follow each other in an harmonious and defi-
nite order. Placing, therefore, the contents of both
groups in separate columns, we shall find some of the
analogies both curious and interesting.

Analogies of the VertEBRATA and the Cartilaginous

Fishes.
Circle of the CHon- - Circle of the
DROPTERYGES. Analogies. VERTEBRATA.
d Partaking most of all tothe struc- ,
eee ture of the Cefacea. Viviparous. } we
G Pectoral fins assuming the form 3
Raide. f of wings. All oviparous. fins.
Posterior part of the body, or the
Chimeride. ) tail, gradually attenuated and +} REPTILES.
pointed.
Sees Most aberrant in their respective
Sturionide. circles. Teeth none. i LS -
P Gill opening very large; gills pecs}
Polyodonide. Fite! z - PIScEs.

(136.) The following points of analogy do not admit
of much illustration, seeing that they are remote ; and

ANALOGIES.— VERTEBRATA TO CARTILAGINES. 155

yet itis most extraordinary to observe the perfect regu-
larity with which they follow each other. Every zoologist
will confess the likeness between the sharks and the por-
poises, even in their external appearance: and while no
fish make such a near approach to quadrupeds as the
sharks, no quadrupeds more resemble true fish than the
Cetacea: this, of itself, is a fact so far beyond dispute,
that we may at once pass on to the next analogy. The
enormous pectoral fins of the rays, and the remarkably
small size of the others, which are nearly obsolete, in-
contestibly prove that in them is concentrated nearly
all the powers of locomotion, and accounts at once for
the excessive rapidity with which they swim: this is
precisely the case with birds ; whose wings correspond
anatomically with the pectoral fins of fishes. The
very appearance of some of the rays shows that nature
intended to make them represent the feathered class ;
and this analogy is so apparent to ordinary observers,
that several have acquired the name of sea eagles, eagle
rays, &c. As the eels obviously represent the serpents,
so do the Chimeride represent the reptiles, the pri-
mary external character of which consists in the tail
being excessively lengthened, and gradually ending in a
point. The Chimeride are the only cartilaginous fishes
yet discovered, that have a tail thus formed ; and they
cannot, therefore, be likened to any of the vertebrated
divisions, excepting the reptiles. The analogy between
the sturgeons and the Amphibia is not only faint, but
even obscure. But this may be easily accounted for
in two ways: first, it is an indisputable fact that the
analogies between two groups of animals thus com-
pared, are almost always weakest between their most
aberrant types ; and secondly, because, when there are so
few species in a group, as in the Sturionide, we have
not the same facilities or materials for determining its
analogies, as when it is more numerous: the points of
comparison, in short, are few; and setting aside the
ignorance under which we may labour, we must, in
all such cases, rest satisfied, if what is really known does

156 CLASSIFICATION OF FISHES,

not militate against our other analogies. So far, how-
ever, is this from being true in the present instance,
that we actually find the Sturionide coming opposite
to the Amphibia, when we bring the circles of the
Vertebrata and the Cartilagines together: there are
even some considerations which strengthen the analogy
thus inferred. The sturgeons, like the Amphibia, have
no true teeth, and they live in two distinct modifications
of the same element,— that is, both in salt and fresh
water. The simple fact, however, of their standing in
the order of affinity (for this is the primary consi-
deration) between the Chimeride and the Polyodonide,
and that the Amphibia hold the same rank between the
reptiles and the fishes, is a sufficient argument that they
represent each other, although we are not prepared to
state the true manner in which this law of nature is
effected. We now come to the Polyodonide and the class
Pisces. Weare to inquire under what view we may
consider. the former as a representation of the latter: it
is not sufficient to say that Polyodon is a fish, because
so are all the Cartilagines. Now, if the question was
asked, What are the most prominent characteristics of
the typical orders, independent of their general form ?
the answer would be, that such fish possessed free
laminated gills, with a large and unconfined branchial
opening. These, then, are the very characteristics of
Polyodon ; and as they are found in no other type of the
cartilaginous circle, it follows that this division, more
than any other we have noticed, gives us the best repre-
sentation of the ordinary and typical structure of the
class Pisces.

(137.) We have had frequent occasion to remark,
while tracing the analogies among quadrupeds and
birds, that, to illustrate all the peculiarities of an animal,
one table of comparisons is not sufficient: many others
would then remain ; and we can only explain these by
instituting other comparisons, and applying further tests
to the accuracy of our theories. Now, the aberrant
groups of the order before us particularly require his,

~

ANALOGIES.-— CARTILAGINES TO BIRDS. 157

especially Polyodon and Chimera, of which nothing
that we have yet said relates to the enormous flattened
snout of the first, or the lobe-shaped crest of the last.
We shall, therefore, now exhibit the analogies of the
cartilaginous types in a new light, by bringing them
into contact with the primary orders of birds.

Familiesofthe 4

GaGannocranvars: Analogies. Orders of Braps.
Squalide. Pre-eminently rapacious. RAPYORES.
faide. Typical of their respective circles. INCESsORES.

. : Males with crests or frontal ap- }
Chimeride. j pendages. RASoREs.
Sturionide. Mouth very small. GRALLATORES.
Polyodonide. Snout or bill excessively broad. NATATOREs,

(138.) The two first set of analogies are so obvious,
that every naturalist will at once perceive them. It
follows, indeed, as a necessary consequence, that if the
sharks represent the beasts of prey, they also represent
the rapacious order of birds ; and that if the Raide are
typical of birds, they must bear the same relation to
that group which is the most perfect among birds.
The rasorial type of form, already so much enlarged
upon in former volumes, is eminently distinguished
from all others by the heads of one or both sexes being
ornamented or defended by unusual appendages, which
among quadrupeds take the shape of horns, and in
birds that of crests. The Chimera borealis exhibits an
appendage perfectly analogous to this, in the singular
fieshy caruncle or lobe which surmounts ‘its snout,
the end of which is beset with numerous short prickles ;
while the tail, as in all rasorial types of the Vertebrata,
is highly and singularly developed. Thus we have,
among fishes, a structure perfectly analogous to the
rasorial order of birds, and to the ruminating order of
quadrupeds ; and as the types of the rasorial birds (the
family of peacocks) are among the most splendid
coloured of the class, so Chimera is the only group
among the cartilaginous fishes whose colours have any
degree of brillianey, The difficulties attending the

158 CLASSIFICATION OF FISHES.

analogies of the Sturionide have been already stated ,.
but we may remark, that the smallness of their mouths
is in complete. accordance with that structure which is
one of the most marked peculiarities of the Grallatores,
or wading birds; while the order Kdentates, among
quadrupeds, — the types of which have their bodies co-
vered with bony scales like the sturgeons, — is an indirect
proof in support of the opinion that all are repre-
sentatives of each other. Lastly, the Polyodonide, and
the natatorial type of birds, are those only which have
the snout or bill excessively broad and uncommonly
flattened. That Polyodon, therefore, is the natatorial
and, consequently, the fissirostral type of the cartilaginous
circle, cannot be doubted, because its snout is much
longer and broader than in any other fish yet disco-
vered ; and we thus get an explanation why, in a group
which is collectively a natatorial type, it should yet
have one of the peculiarities of that type so pre-
eminently conspicuous.

(139.) To trace the analogies of the cartilaginous
families further, might weary the reader, and may be
thought unnecessary by the naturalist; seeing that all
the peculiarities of the two most singular forms in the
sroup, Polyodon and Chimera, turn out to be in per.
fect accordance with those ordinary laws of variation
which nature adheres to in other divisions of the ver-
tebrated animals, and which we hope to trace hereafter
in the annulose circle. There can be no doubt that
innumerable analogies, equally strong, exist between
them and their representatives among the osseous fishes,
which may hereafter add additional force to what has
just been elucidated.

(140.) The analogical relations of the primary divi-
sions of the order being now disposed of, we shall again
revert to the family of Squalide, for the purpose of
ascertaining whether the same system of representation
can be traced in its sub-families. In endeavouring to
determine these latter, it will be remembered that we
have noticed them in the following order: . Zyganine,

ANALOGIES OF THE SQUALID. 159

Squaline, Centrine, Pristis, and Squatina or C'rosso-
rhimus; the first and the two last forming the aberrant
group, while the second and third are considered as the
typical and the sub-typical. Let us place these in one
column, and the primary divisions of the Cartilagines
in another, and then trace their analogies.

Analogies of the Squatip™ to the CARTILAGINES:

Sub-families Families of the

Analogical Characters.

of the SQUALID. CARTILAGINES.
Squaline. Dorsal fins generally without spines. SQUALIDE.
; One acute spine on the first or se-

Centrine. aandinesaitane: tRawa.

.. Snout produced, armed with prickles
Pristine. or spines. 2 : i CHIM#RIDE,
Crossorhine. Mouth furnished with cirri. STURIONIDE.
Zyganine. Head or snout excessively broad. POLYODONIDE.

(141.) There is a somewhat intricate point, which the
last table brings more immediately before us, upon
which we must here say a few words. In the present
infant state of philosophic ichthyology, it is not likely
to claim that attention it will hereafter most assuredly
receive ; but we shall now advert to it, to show it has
not escaped our cbservation. » This point regards the
rank of pre-eminence among the Squalide. It may be
argued, that if the Raide@ are typical of the order
Cartilagines, then it would seem to follow that the
Centrine, which clearly represent them, are also typical
of the Squalide: both are distinguished by their spined
backs, which make them also analogous to the Acan-
thopteryges, the most typical of all the fishes. By re-
garding the Centrine, therefore, in this light, we give to °
all the groups we have just named one and the same
rank ; that is, of being the pre-eminent types of their
own circles : nor does there appear any great objection to
this, if we only look to the groups just noticed. But
how would the case then stand, regarding the analogy
between the rays and birds ? for the latter are most cer-
tainly not the pre-eminent types of the Vertebrata, and
therefore the rank of these two would still remain dif-

160 CLASSIFICATION OF FISHES.

ferent ; that is, the rays would be a typical, and the
birds a sub-typical, group. ‘This latter denomination, as
applied to the class Aves, is so unquestionable, that it
must remain undisturbed. Analogy must always be
made subservient to affinity ; and as this very transport-
ation of the two typical groups has been frequently
observed in ornithology, we must leave it to time, and
a better acquaintance with the theory of variation, to
clear up a question so beset with difficulties.

(142.) The analogies of the two first groups in each
of these columns are, of course, only applicable to the
typical examples of each; while the only exception to
the whole of the Squaline being destitute of spines,
rests on the question whether the Da/atias nocturnus
of Rafinesque has been correctly described as without
spiracles: should this really be an error, then this sup-
posed genus must be abolished, and the above-named
fish will become, as Cuvier conjectures, a species belong-
ing to the Centrine. This question, however, is of no
importance to our present purpose, for we are looking to
large assemblages, not to the peculiarities of the sub-
genera: besides, it is quite clear that, even if some of
the sharks without spiracles have spinous fins, the
greater portion have not; while, as the majority
of those with spiracles also possess spines, this latter
character becomes one of their typical distinctions. In
this manner, the Centrine will, of course, represent the
rays. Now, the nearest approach which is made by the
sharks to the saw-fish, seems to beby the genus Mustelus,
because it has, like Pristis and the rays, tesselated teeth ;
hence we have supposed that they are united by affinity,
although there appears an hiatus between Mustelus and
Pristis, which nothing yet known is caleulated to fill
up. Whether we are correct in this supposition, time
only will show. The analogy of Pristis to the Chimerine
is manifested by the tooth-like processes of their snouts;
those in Pristis assuming the form of teeth, those of
Chimera prickles. It might be thought, indeed, that
Pristis was more analogous to Polyodon, because the

3

ANALOGIES OF THE SQUALID#. 161

only remarkable difference between their snouts consists
in the one having bony or tooth-like processes, while
that of the other is smooth: but this difference is a very
important one, because no fissirostral or aquatic types,
whether among birds or quadrupeds, have the snout
horned.* The spined processes, therefore, of Pristis,
placed on the snout, are completely analogous to the
horns of ruminating quadrupeds, and to the prickles
on the prolonged frontal lobe of Chimera; and both re-
present, however imperfectly or obscurely — for how
could it be otherwise ?—the rasorial birds, and the
ungulated quadrupeds. Between Crossorhinus, Squatina,
and the Sturionide, the analogy is very slight; since
the only resemblance to be traced between them, at
present, is their mutual possession of cirri, or fleshy
barbs, round their mouths. We should almost have he-
sitated — indeed still hesitate in definitely placing
Squatina among the sharks; because it seems to have, in
its general aspect, as already observed, a much nearer
affinity to the rays. Lastly, we come to the resemblance
between the Zyganide and the Polyodonide, which
agree in this one fact,—that both have the broadest heads
or snouts of all the groups we have been comparing.
The forms of the two fishes are certainly dissimilar ;
because the snout of Polyodon is, although very wide,
more remarkable for its length; while that of the
hammer-headed sharks is very short and obtuse, yet
excessively wide. It is clear, however, that, as both are
fissirostral types, they represent each other; although we
by no means feel confident that the precise situation we
have assigned to the Zyganide is the correct one.
(143.) The peculiar difficulties, already adverted to,
in our attempt to arrange the sub-families of Squaline
and Centrine in their natural series, and to designate
their primary divisions or genera, bring with it corre-
sponding difficulty and uncertainty in attempting to trac-

* The Ceratodon (Monodon monoceros), although in the aquatic order
of Mammalia, is but a representation of the rasorial or ruminating type
among the Delphinida, or porpoises,

VOL, I. M

162 CLASSIFICATION OF FISHES.

their internal analogies. Some of the more recently cha-
racterised forms we have not personally examined, and
others are not now before us, so that we had almost de-
termined not to have prosecuted our analogical views
further than to the sub-families: but this might have
given an impression to some few of our naturalists, that
the theory could not be carried further, and that we de-
serted our former declaration, that every group, whether
large or small, if natural, would contain representations
of all others. To show, therefore, that, even in our
present dilemma, there is some ground for this asser-
tion —so fully demonstrated already in the class of
birds — we shall make the attempt. If one or two of
these analogies carry with them an appearance of truth,
our principle, substantially, is gained ; while, for the
rest, if we are in error, these very errors will serve as
land-marks to others, and elicit that additional inform-
ation which is absolutely essential before we can hope to
work out the internal affinities and analogies of the great
number of forms comprised in the sub-families Squaline
and Centrine.

Analogies of the Seuauinaz and the CentTRINZ.

Genera of the Genera of the

Distinctive and analogical

SQUALINE ; 3 pas CENTRINE ;
no spiracles. ELSES: wita spiracles.
Squalus Linn. Typical of their respective groups. Centrina Cuv.
Dalatias Raf. Spines to the dorsal; no ventral fin. Galeus Raf.
Ad Snout or muzzle excessively long, .
Zsurus Rat. } projecting beyond the mouth. sey llzuna Cn.

Mouth at the end of the muzzle
eyes verticai.
(The second dorsal fin opposite the
j} anal: the two last branchial
Scoliodon M. H. i openings placed above the pecto- } Mzstelus Cuv.

Rineodon Smith a Cestracion Cuv-.

ral fin: teeth the same in both
jaws.

(144.) It will tend much to elucidate the above table,
if we first of all briefly recapitulate the reasons that have
induced us to arrange these two series in the order in
which they now stand; so that, before entering on an
explanation of the analogies they bear to each other, we
shall take a hasty glance at the affinities of the genera

ANALOGIES. —SQUALINZA AND CENTRINZA. 163

respectively placed in each column. We must first,
however, remind the naturalist, that the groups in these
columns are what we consider to be genera; and that,
consequently, those numerous sub-genera which have
been proposed by other naturalists, and whose names do
not appear in the above list, are considered by us of
subordinate rank, and form a part of one or other of
these genera. One instance will better explain our
meaning: M: Cuvier’s sub-genus Se/ache is not men-
tioned, because we consider Rafinesque’s Jsurus is the
true typical example of the genus which connects the
Squaline and the Centrine. This union of two circles
is always effected either by the rasorial or the fissirostral
type; and the long snout and forked tail of Jsurus
clearly show that it is of this latter description :
Selache, indeed, is an excellent sub-genus, but it is sub-
ordinate to Zsurus ; and therefore, as every group should
bear the name of its most typical example, we name it,
in the present instance, accordingly. In like manner,
Pristiurus of Bonaparte, from what little has been said
of its form*, appears the true type of Scyllium, because
it is said to have a ‘‘long snout, ’’—the exact character of
Isurus in the opposite circle of the Squaline; so that
we may fairly suppose they are analogous. As our
acquaintance, however, with Pristiurus is so slight, we
shall for the present retain the long-employed name of
Scylliium. But it may fairly be asked, Upon what prin-
ciple do we ground our belief that one group is a genus,
and that another is a sub-genus? and why, in reference
to the above case, has not Selache as great a claim to be
considered the type of a genus as Zsurus? To this
we reply, by stating the especial object of the present
inquiry: our purpose is to show that each of the two
typical sub-families of the sharks—the Squaline and
the Centrine — represent each other in their respective
circles ; and that the subordinate divisions, or genera,

* Muller and Henle, Mag. of Nat. History, 2d series, vol. ii. p. 34.
mM 2

164 ULASSIFICATION OF FISHES.

of one, correspond to those of the other, — but with
this difference, that all of one are furnished with tempo-
ral spiracles, which spiracles are not seen in the other.
Having, therefore, endeavoured, in the first instance, to -
make out the affinities of each among themselves, we then
select, from all the minor groups that have been named,
those which correspond, in some way, to each other:
these we denominate genera, and place all the others as
sub-genera: the distinction, therefore, is neither empy-
rical nor arbitrary ; however we may err in the selection,
the principle upon which that selection has been made
is sound and philosophic. We have little doubt that
nearly all the divisions of Rafinesque, Cuvier, Muller,
Henle, Smith, Le Sueur, &c. will arrange themselves
in the line of affinity, either as genera, sub-genera, or
aberrant species: but we repeat our belief, that our
existing information on this family is not sufficient to
carry us through such an analysis. By far the
greater part of the specimens of sharks, seen in mu-
seums, are miserably preserved, —the natural form
completely destroyed by having the skin either dilated
or contracted ; while the mouth is either closed, so that
the teeth are not seen, or the jaws are taken out, or the
specimens are of young individuals before the teeth are
wel] developed. We look forward, indeed, with much
interest to-the forthcoming publication of MM. Muller
and Henle upon this family, satisfied that in many re-
spects it will add much to our general knowledge of this
group. But we consider the principles of their arrange-
ment, so far as it has been developed, as essentially
artificial, being framed without any regard to the other
groups of ichthyology ; and we look on all systems
founded, as this is, upon the teeth, as liable to much
fallacy, because these organs are well known to vary in
young and adult specimens, as well as in mature old ones,
of species which follow close upon each other in all the
remaining points of their organisation. Our chief desi-
derata, in fact, are accurate drawings, and full descrip-
tions, made from the fresh subject, and from adult

AFFINITIES OF THE SQUALINZA. 165

specimens. This, of course, can only be done by slow
degrees, and by different naturalists in various parts of
the world: but the spirit which is now abroad, particu-
larly among the rising naturalists of our colonies, will
do much to hasten this; and the time will then come,
when the groups of the Squalide will be established on
the best of all foundations, — their analogical resem-
blances. This digression, although long, may not be
thought out of place ; and we shall now proceed, as we
intended, to recapitulate the affinities of these two
groups, before we enter further upon their analogies.
(145.) Let us first take the genera of the Squaline,
or those sharks which have no temporal spiracles.
Commencing with Squalus (improperly named Car-
charias by Cuvier*), we find a numerous assemblage
composed of Alopias Raf., Cericteus Raf., and many
others, imperceptibly leading to Dalatias. Of this
latter, Rafinesque expressly says, that although his
D. nocturnus has spines on the dorsal, yet that it has
no spiracles and no anal fin. It is worthy of remark,
also, that he places this genus close to our Squalus ;
observing, that it differs from that in having no anal
fin, and from Centrina Cuv. (Squalus Raf.), in
having “ no spiracles.” From this we pass to Iswrus,
Raf., distinguished —as are all fissirostral types, both of
birds and beasts—by a very lengthened snout, and a
deeply forked, equal tail. Another character, more ge-
neral in this genus, is the excessive size of the branchial
openings; and both these latter characters are found in
Selache and Lamna. But now, having reached the
passage to the Centrine, we begin to see the incipient
development of the temporal orifices, asserted by Dr.
Smith to exist in Zamna, where Cuvier says they are
not to be found. Next follows Rineodon Sm. which

* We say improperly, because, in this instance, and in numerous others,
M. Cuvier, while he professes to retain the genus Squalus, virtually abo-
lishes it; since he does notipreserve the original name to any one of its
~ divisions,

M 3

166 "CLASSIFICATION OF FISHES. -

Muller and Henle consider is so closely allied to Se-
lache, that they actually place them close together.
Finally, we quote the same authority for placing Sco-
liodon next to our Squalus; for, according to these
eminent naturalists, the one follows the other, and so
nearly coincide in their teeth, that those of Seoliodon
“ differ only in being of the same shape in both jaws;”
— and thus we return to the point from whence we
commenced.

(146.) We now turn to the other column, composed
of the Centrine, having temporal spiracles. At the
head of these stands Cuvier’s genus Centrina; under
which we place, as sub-genera, Spinaw and Seymnus.
This latter, being aberrant, has no spines to the dorsal,
but, in the words of M. Cuvier, it has, in every other
respect, “all the characters of Centrina.” Scymnus
brings us immediately to the genus Galeus (Raf. Cuv.),
under which we may place Wotidanus Cuv., and perhaps
Etmopterus of Rafinesque, as uniting Seymnus, Galeus,
and Centrina. Somniosus Le Sueur, from not having
an anal fin, seems to belong to the same group. And,
indeed, it almost seems that the next genus after Cen-
trina should consist of those sub-genera which have
neither dorsal spines nor anal fins; in which case Galeus
will stand only as a sub-genus connecting Mustellus to
Centrina. We now arrive at that division which leads
to the Squaline ; and we consequently find that some of .
the sub-genera associated with Pristiurus Bon., as Scyl-
lium Cuv., and Chiloscyllium M.and H., begin to have the
temporal orifices very small, so as to blend with Lamna,
and other subordinate forms in the circle we have just
left. In all these, as MM. Muller and Henle have well
observed, the first dorsal fin is never placed before the
abdominal fins. Cestracion, another genus with spined
dorsals, seems to follow the last ; and thus we arrive at
Mustelus, the affinity of which with Centrina is mani-
fested in all but the teeth, which resemble those of the
saw-fish and skates.

ANALOGIES OF THE TWO TYPICAL GRouPS. 167

’ (147.) The result of this disposition of the groups is
seen in the preceding table, which shows the analogies
existing between their component parts. From these it
would appear that each has a division (Dalatias in one,
and Galeus in the other), where the ventral fin is want-
ing, and the dorsal fins are spined. Again, Jsurus and
Pristiurus (which latter we have arranged with Scyl-
lium) are the longest-snouted sharks yet discovered :
while Rineodon and Cestracion represent each other by
the very reverse of this latter character ; for the mouth
of both is deseribed as being at the extremity of the
muzzle; and thus they also represent Squatina and
Crossorhinus. With these striking coincidences before
us, we need feel less regret at not being better informed
on the new genus Scoliodon of MM. Muller and Henle;
but the short characters they have assigned to it sin-
gularly coincide, in ali but the teeth, with those of
Mustelus: and as this latter genus opens a passage to
Pristis, so we may expect that it would possess some
one of its characters; and this expectation is realised by
the structure of the teeth, which are precisely alike ;
Mustelus including the only sharks where these organs
are blunt and tesselated, as in the rays and saw-fish.
(148.) That errors may eventually be discovered in
this imperfect sketch of the natural arrangement of the
sharks, is only what we fully expect; and this, for the
reasons already stated, we should say would be inevi-
table. But whether these errors are few or many,
the main facts which we have sought to establish, of
there being certain types, representing each other, but
without mutual affinity, will remain unshaken ; and
further, that those types correspond to others pervading
every group in ichthyology. We contend not, in this
ease, for details, or for the accuracy of minor com-
binations: all we seek to establish at present, is the
theory of representation ; and for this there seems to be
conclusive evidence. The two typical groups may pos-
sibly possess other characters than the mere absence or
M 4

168 CLASSIFICATION GF FISHES.

presence of spiracles: one may be viviparous, the other
oviparous—(and this deserves much investigation) ; but
still there will be corresponding relations, however these
relations may be exhibited ; and that arrangement
which places them in the clearest light, must always
be that which is nearest to Nature.

(149.) The Rain, or rays, succeed the sharks, to
which, as before remarked, they are closely and inti-
mately united by the saw-fish (Pristis) : they are com-
posed of the rays, properly so called, having the base of
their tail armed with a sting, and of the torpedos, skates,
and thornbacks. The two latter, from being well
known and very abundant in our own seas, will give
the general reader a correct idea of the whole family.
Taken collectively, they may be called the flat fish of
the cartilaginous order, and, in this respect, show a
marked and unquestionable analogy to that family of
osseous fishes. The whole of the species, like the sharks,
are marine ; and several of those found in the warmer
latitudes grow to a very great size. The depression of
their body is fully as great as what we see in the Pleu-
ronectid@, or true flat fish ; but the head and eyes are
symmetrical ; while the pectoral fins are of such vast
magnitude, that they actually extent all round the head
and body, and terminate only at the base of the small
ventral fins, thus giving the body a disk-like form : the
tail is excessively slender ; and the dorsal fins, when
present, are generally remarkably small, and placed upon
their slender tail. In the typical species, the caudal fin
is mostly wanting, as the tail ends in a slender point ;
but in others, as the torpedo skates (Raie) and shark
rays (Rhinobates), there is a small caudal, whose size
seems regulated by the comparative diminution of the
pectorals. The scapule of the pectorals are articulated
with the spinal column, just behind the branchial spi-
racles : the eyes, and the large temporal orifice imme-
diately behind them, are, of course, placed on the upper
surface, at a considerable distance from the snout and

GENERAL CHARACTER OF THE RAYS. 169

the circumference ; but the mouth, nostrils, and branchial
apertures are on the under surface, and are completely
hid when the fish is laid on its belly: the latter organs
are generally five in number, arranged on the sides, and
are of the same form as in the sharks. The rays of the
fins, like the rest of the skeleton, are cartilaginous,
straight, and furnished with numerous swellings or knots.
The mouth is small, and furnished with numerous small
blunt teeth, which are placed in rows, like paving stones
or mosaic, so as to completely cover the lips or edges
of the mouth: the eyes are protected by a nictitating
membrane or skin, which can at pleasure be drawn over
them like an eyelid, —a character which is common to
many of the sharks : at some distance above the eyes are
situated the nostrils, each appearing like a large andsome-
what semilunar opening, edged with a reticulated skin,
and furnished internally with a great many laminated pro-
cesses, divided by a middle partition, and guarded by an
exterior valve: behind the eyes are the temporal orifices
or spiracles, communicating with the mouth and gills ;
these orifices are much larger than those of the sharks,
and often exceed the size of the eye ; and all these parts
taken together occupy a wide extent of surface. The young
are contained in oblong square capsules, of a horny sub-
stance, with a filament, more or less lengthened, at each
of the four corners. It would seem that the female has‘
the faculty of twisting these round the stems of marine
plants or corals, so as to secure the capsule from being
tossed about and drifted by the waves. These cases,
when the young have been exuded, are finally de-
tached, and are often cast upon the shore in considerable
numbers, when they are called sea purses by the common
people.

(150.) Little is known of the natural history of these
singular fishes : inhabiting the depths of the ocean, they
elude the inquisitive eye of man ; and we can only form
a few conjectures by their general structure. We know
that the Pleuronectide, or true flat fish, lie concealed at

170 CLASSIFICATION OF FISHES.

the bottom of the sea, among weeds and mud, and thus
watch for. their prey: hence it may be safely inferred
that habits, somewhat similar, belong to the rays.
The unusual development, however, of their pectoral
fins, places it beyond doubt, that they can pursue their
prey with a swiftness surpassing that of all other fishes ;
an inference which is further strengthened, when -we
remember that these swallow-like fish stand at the head
of the fissirostral type of the class Pisces, corresponding
to the swallows among birds, and the Natantia among
Mammalia. Some of the species grow to an immense
size, —a circumstance that may be accounted for by the
supposition that cartilaginous fish continue to grow as
long as they live. A species of skate, common to the
British seas (Raia batis) is frequently caught of im-
mense dimensions, sometimes weighing two hundred
pounds. But this is nothing to another individual of
this family, which is stated to have been caught in the
West Indian seas, whose length was twenty-five feet,
while its greatest breadth is stated at thirteen ; the tail
alone measuring fifteen feet. The sting rays, of which
this last was probably a species, are, perhaps, the largest
in their dimensions of the whole family. Two spe-
cimens of the Pterocephalus massena Sw. of the Me-
diterranean, were caught near Nice, and seen by Risso,
‘which measured twelve feet long, and twenty-seven in
circumference; the weight of the female was 1250
pounds, but that of the male only 800. The Pteroceph.
Banksianus is a still more gigantic monster; for although
its weight was not ascertained, it is said to have required
no less than seven yoke of oxen to drag it on shore.
There is some evidence, also, that these monsters of the
deep, like the sharks, are destructive to mankind.
Colonel Hamilton Smith relates, that he once witnessed
the destruction of a soldier off Trinidad, by one of these
immense Pterocephali. It would seem that the soldier
wished to desert, and, being a good swimmer, he had
jumped into the sea from the vessel, which then lay at

RAID. — NATURAL ARRANGEMENT. Vis

anchor in the entrance of the Bocco del Toro. The cir-
cumstance occurred soon after daylight, and the man,
being alarmed by the call of a sailor up aloft, endea-
voured to return to the ship ; but the monster threw one
of his fins over him, and he was never seen more.

- (151.) The natural arrangement of the family has
never been yet attempted. MM. Muller and Henle
have just made considerable improvements on the method
and nomenclature of Cuvier, by separating and defining
many of the subordinate types, passed over in the Régne
Animal; but their arrangement has no ulterior object,
and merely aims at characterising the divisions. Having
paid some little attention, therefore, to this family, we
shall endeavour, on the present occasion, to determine
the natural series of its variation, and shall subsequently
show that this is regulated by the same general law that
pervades all the other groups of ichthyolegy. We ar-
range the whole family under the five following divi-
sions: — 1. The Ratna, or true rays; 2. The Myxio-
BATIN&, or eagle rays; 3. The Torrenina, or torpedo
rays; 4. The Seuatina, or shark rays; and, 5. The
RHINOBATINA, or snout rays. The two first of these
are typical, and are distinguished by their very slender
and whip-like tails ; while in the three latter, or aberrant
sub-families (each represented only by a single genus),
the tails are thick, and more or less approach those of
the sharks.

(152.) The barb, sting, or spine, — for it has been
called by all these names, — with which the tail of the
majority of the sting and eagle rays is armed, is a most
formidable weapon, in the shape of a long-headed lance:
it is acutely pointed, and varies in length according
both to the species and the size of the individual: it
is a compressed, hard bone, having the two edges finely
serrated, with the serratures pointing to the head, so as
to tear the flesh upon being drawn out; and it thus in-
flicts a most grievous wound. It is currently understood
by all sea-faring people, that these barbs are poisonous ;

fy CLASSIFICATION OF FISHES.

and so firmly is this believed by fishermen, that upon
catching any of these sting rays, they immediately pro-
ceed to cut off the tail of the fish, or mutilate the spine.
The use of the long naked tail, seen in most of these
fish, is probably to twine round their prey, so as to con-
fine its struggles. Sometimes there are two of these
barbs placed close together ; but in some of the sub-
genera, both are absent.

(153.) The first, or typical, sub-family, containing
the true rays, is eminently distinguished from all the
others, by the pectoral fins being united to the snout
in such a manner as that there is no interval of separa-
tion between them. All our British species, including
the thornbacks, skates, &c., are of this description, and
afford perfect examples of the general form pervading
the whole of this division: in other respects, there are
many variations. The group, indeed, is so numerous in
its contents, that we may even distinguish the genera,
which we shall now enumerate. The trygons, or sting
rays (Trygon Antiq.), divide themselves into three
genera. The first is Trygon, where the breadth of the
body and pectorals is about equal to its length: the tail
is armed with one or two spines, or stings, as they are
called, at the base ; and there is a narrow fin, either
above, or below, or on both sides. Pastinaca Antiq.*
differs from Tygon only in having the tail entirely
naked: the common sting ray of the Mediterranean is
the type of this genus, to which we prefer retaining the
name by which it was known to the ancients. In two
others, described and figured as natives of the Indian
seas by Dr. Russell, the body is somewhat more oval
than in the Mediterranean species, and there are two
spines ; but the number of these do not appear to in-
dicate generic groups; and it not unfrequently happens
that, in such as really possess two spines, one is acci-
dentally broken off. The presence or absence of fins

* Himantura, Muller and Henle.

RAINZ, OR TYPICAL RAYS. V3

upon the tail, therefore, appears to us a more certain
mark for discriminating the two typical genera of T'ry-
gon and Pastinaca, than the number of the stings ; and
this view, we perceive, has been taken by Muller and
Henle. In further proof, also, we may refer to the
two species above mentioned, from India: both are of
the same form, and both have the tail entirely naked ;
yet in one there is but a single spine, while the other
has two. The third genus is that of Pteroplatea,—a
name given by the last mentioned ichthyologists to
certain sting rays, which have the pectorals so very long
as to render the breadth of the fish considerably more
than its length: the tail, like that of Pastinaca, is
always naked, but itis also remarkably short. It is here
that the stings begin to disappear; for although one
species of those which have been described possesses two,
yet in another, from India (Russell, pl. 22.), there is
none whatever. This latter fish, therefore, brings us near
to the genus Raia (R.rubra, fig. 18.), as now restricted
and understood by the moderns. This group, indeed, has
recently been divided into several sub-genera ; but as
the value of these remains to be determined by a _philo-
sophie analysis of the
real types, we do not,
at present, adopt them.
The whole may be
characterised as dia-
mond-shaped fishes,
almost always covered
with prickles or mi-
nute asperities, but
never having the tail
armed with a barbed
spine, as in the three
preceding genera: the
tail, moreover, termi-
nates in a small cau-
dal fin; immediately

abe,
i / j

7
|

|

|

\\
SS

174 CLASSIFICATION OF FISHES.

before which, on a line with the back, are two small
dorsal fins: one of these latter, or the small caudal,
may be expected to disappear in such aberrant species
as approach Pteroplatea on one side, or go off, on
the other, to Anacanthus Ehremb.,—a genus which is
stated to resemble Trygon in every thing but the
possession of a-sting. We are thus enabled to trace
a circular disposition of the whole sub-family ; the
contents of which represent all the primary types of the
cartilaginous order.

(154.) The second sub-family contains the gigantic
Pterocephaline*, or eagle rays. These are the fish
which we have already mentioned as often growing to
such an enormous size, and being as dangerous to man
as the sharks. The form of their body is much like
that of the sting rays, but with this difference,—that the
pectoral fins are not continued so as to encircle the fore
part of the head, which is consequently free; and the
eyes are inserted at the edge of, not within, its circum-
ference. The tail is as slender as in the last group, and
is generally armed with a formidable barb or sting at its
base; in addition to which, there is usually a small trian
gular dorsal fin placed at the base, which is very different
from the long and narrow fin-like membrane seen towards
the end of this part in many of the sting rays. Although
the species are by no means so numerous as in the last,
we yet find five divisions, so well characterised by their
general form, that we shall adopt them on the present
occasion. These gigantic fish are very rarely seen, and
then chiefly in warm latitudes: they seem, indeed, to
be pelagic, for they are seldom taken near any shore.

(155.) We place Myliobates as the first genus, because
it has a closer affinity to the typical rays than any of the
other four ; this is shown by its resembling Péeroplatea
in being much broader than long, owing to the great

* These constitute the genus Cephaloptera of Dumeril ; but as that name,
unluckily, had been previously given by Geoff. Saint- Hilaire to a remarkable
genus of birds, we propose to substitute the present for it.

PTEROCEPHALIN®, OR EAGLE RAYS. 175

development of the pectorals: these fins, however, only
take their commencement immediately behind the eye ;
so that those and the snout are entirely free. The tail
of these fish, of which one species is well figured among
the Indian fishes of Russell, is very slender and exces-
sively long, being near twice the length of the body ;
and it is described as being without any sting or bony

process. Whether this weapon is constantly absent in
certain species, or whether, as on the present occasion,
it had been broken off accidentally or purposely, (for it
is universally considered by fishermen of all countries
to be poisonous, ) is a question which must be undecided ;
certain itis, however, that some of the rays, both in this
and the last sub-families, appear to be totally devoid of
a sting. The next genus is Rhinoptera of Kuhl. In
this the head is equally free; but the snout is so deeply
cleft in front, that in some species it assumes the aspect
of two horn-like protuberances, not in substance, but in
shape. In one species, the R. quadrilobata of Le
Sueur*, here represented from an exquisite plate by that
naturalist (fig. 19.), there are two other processes,
one on each side
the under part of
thesnout (a), which
are obviously the
first development
of those fin-like
paddles seen in
the next genus.
The mouth, both
of this and Mylio-
bates, is placed un-
derneath (6) ; but
the sting does not
appear constant: it
exists in the species

here figured (¢), but

* Amer. Trans. vol, i. pl. 20,

176 CLASSIFICATION OF FISHES.

is absent in that described by Dr. Russell. The two
processes just mentioned conduct us at once to the
genus Ceratoptera M. and H., where these appendages
assume the form and office of lobed fins, as represented
by the same artist (fig.20.), the head is completely

obtuse in front, without any of the lobed appearance
seen in the last genus; while the mouth is at the ter-
mination of the muzzle. On this latter account, these
remarkable fishes have been justly separated from the
true Pterocephali (or the Cephaloptera of Dumeril),
where the mouth is on the under side of the head, as in
all the other genera. We place 4tobates M. and H. as
the last genus, with some hesitation, suspecting that it
really possesses this rank in the present division ; for
it has every one of the characters of Myliobates, super-
added to a caudal sting. But its most remarkable pecu-
liarity is the circumstance of the jaws being dissimilar:
“the lower one,’ as Dr. Russell observes on a species
he has described, “‘ being arched, narrow, and projec-
ing beyond the wider immovable upper jaw: the edges
of both are smooth and without teeth.”* MM. Muller

* Coromandel Fishes, vol.i. p. 5.

RAID. — TORPEDO RAYS. LA

and Henle, however, whose method is chiefly founded
on these organs, gives the generic character to this
group of having one row of teeth in each jaw. We
believe that both these accounts, however apparently
conflicting, may be essentially correct, when applied to
different species; and this only adds another to the
numberless instances that may be cited of the subordi-
nate value which such dental characters possess, when
employed generically.

(156.) The torpedo rays appear to follow next
in the natural series. These fishes have long acquired
celebrity from the powers of electricity they possess ;
while their shape is so remarkably singular, that they
appear mere like gigantic tadpoles than fish: the head
seems of an enormous size, owing to its being completely
surrounded by the pectoral fins ; which latter, from not
being angulated, as in the other rays, make the head
appear, in some species, completely circular: so far,
indeed, the general structure is in much accordance
with the last two sub-families ; but now the tail begins
to assume the usual shape of other fishes; although not
longer than the head and bedy, it is thick and fieshy,
terminated by a distinct caudal fin, and bearing above

it two dorsals: but all these three fins are much smaller

than the ventrals ; these are triangular, and placed on
each side the vent, which is in the middle of the fish.
The situation of the eyes, the mouth, and the branchial
spiracles, is precisely the same as in the thornbacks.

(157.) The torpedos appear to be of many species,
and to inhabit the seas of nearly all temperate and
tropical latitudes. As we cannot well pass over the
extraordinary properties of these fishes, and yet cannot
speak of them from our own observation, the reader will
understand that the following account is abridged from
the best authors who have written upon the subject.
The form of the electric torpedo is much the same as
that of the spotted Indian species ( Sig. 21.): the size, ef
course, varies; its general length is about twe feet ;

VOL. I. N

178
f CLASSIFICATION OF FISHES.

but one mentioned by Pennant
was nearly four, and weighed
fifty-three pounds: the colour
of the upper surface is different
shades of brown, sometimes
marked with obscure ocellate
spots; the under surface ‘is
whitish or fiesh-coloured. Like
others of this family, the tor-
pedo seems to lay in wait for
its prey, partly buried in the
sandy bottom of the sea; and
this is effected by the animal
quickly flapping all its fins, so
as to cast the surrounding
sand partially over its body.
According to Pennant, it preys upon surmullets, plaice,
&c., which have been found in their stomach : in what
manner, however, these swift swimming fish are caught
—whether by a sudden dart of the torpedo from its
ambush, or by exerting its electric faculty — must re-
main undetermined. Before detailing the effects of this
power, we shall give the reader the following abstract of
Dr. Hunter’s description of the organs which produce
them.

(158.) The electric organs constitute a pair of gal-
vanic batteries, disposed in the form of perpendicular
hexagonal cojumns, placed on each side of the head and
gills, from whence they extend to the semicircular carti-
lages of the pectorals; within these limits they occupy
the whole space between the skin of the upper and of
the under surface: they are thickest at the edges, near
the centre of the fish, and become gradually thinner
towards the extremities. Each electric organ, at its inner
longitudinal edge, is a convex elliptic curve; each is
attached to the surrounding parts by a close cellular
membrane, and also by short and strong tendinous fibres,
which pass directly across from its outer edge to the
semicircular cartilages; and they are.covered, above and

THE ELECTRIC TORPEDO. 179

below, by the common skin of the animal, under which
there is a thin fascia spread over the whole organ. This
fascia is composed of fibres, which run longitudinally,
or parallel with the back: these fibres appear to be per-
forated in innumerable places, which gives the fascia an
appearance of being fasciculated: its edges, all round,
are closely connected to the skin, and at last appear to
be lost, or to degenerate into the common cellular mem-
brane of the skin. Immediately under this is another
membrane, exactly of the same kind, the fibres of which,
in some measure, decussate those of the former, and pass
from the middle line of the body outwards and backwards :
the inner edge of this is lost with the first membrane ;
the anterior, outer, and posterior edges are partly attached
to the semicircular cartilages, and partly lost in the
common cellular membrane. This inner fascia is con-
tinued into the electric organ by many processes, and
thereby makes the membranous sides or sheaths of the
columns, which are presently to be described.

(159.) Each organ is about five inches in length, and
at the posterior end three in breadth, though it is but
little more than half as broad at the posterior extremity ;
each consists wholly of perpendicular columns, reaching
from the upper to the under surface of the body, and
varying in their lengths according to the thickness of
the body. The shape of these columns, also, is very
variable ; the greater number are either irregular hex-
agons, or irregular pentagons: their coats are very thin,
and closely connected with each other, having a kind of
loose network of tendinous fibres between the columns,
which they unite more firmly; and this purpose is fur-
ther effected by strong unelastic fibres: the number of
these columns, in different torpedos of moderate size,
appears to be about 470 in each organ, but in a very
large individual they were 1182; they must, therefore,
increase, both in size and number, with the growth of
the animal. Each column is divided by horizontal par-
titions, which appear to contain a fluid: they are not
totally detached from each other, for they sometimes

N 2

180 CLASSIFICATION OF FISHES.

adhere at different places, by blood-vessels passing from
one to another: the number of these partitions in a
column one inch in length, appeared to be 150; this
proportion was so regular in several individuals, that it

eems, as the fish grows, new partitions are added to the
extremity of the column from the fascia: the partitions
are very vascular. The arteries are branches from the
veins of the gills, which convey the blood which has
received the influence of respiration: they pass, along
with the nerves, to the electric organ, and enter with
them ; they then ramify in every direction. The veins
of the electric organ pass out close to the nerves, and
run between the gills to the heart: the nerves inserted
into each electric organ, arise from three very large
trunks placed on the lateral and posterior part of the
brain, and then ramifyin every direction between the
columns.

(160.) The number and magnitude of the nerves,
bestowed unon these organs, in proportion to their size,
must, on reflection appear as extraordinary as the pheno-
mena they afford. Nerves are given to parts either for
sensation or action: if we except the more important
senses of hearing, seeing, tasting, and smelling, which
do not belong to the electric organs, there is no part,
even of the most perfect animals, which, in proportion
to its size, is so liberally supplied with nerves ; nor do
the nerves seem necessary for any sensation which can
be supposed to belong to the electric organs; and with
respect to action, there is no part of any animal, how-
ever strong and constant its natural action may be,
which has so great a proportion of nerves. If it be,
then, probable that those nerves are not necessary for the
purposes of sensation or action, may we not conclude
that they are subservient to the formation, collection, or
mana gement of the electric fluid ? especially as it ap-
pears evident, from Walsh’s experiments, that the will
of the animal does absolutely control the electric powers
of its body, which must depend on the energy of the
nerves.

EFFECTS OF THE ELECTRIC POWER. 181

(161.) The effects produced by this highly siugvlar
organisation were well known to the ancients; but they
— being ignorant of electricity, and prone to invest
every uncommon operation of nature with: an air of
mystery — attributed these shocks to magic, at least,
if we may believe some of their poets; and Pliny,
whose credulity was excessive, affirms that the torpedo,
even when touched with a spear or stick, can benumb the
strongest arm and stop the swiftest foot. The celebrated
Redi, in the 17th century, contributed greatly to dissipate
these exaggerated notions, by elucidating much of the
true history and structure of this wonderful fish ; but
our learned coutryman Walsh, by a series of experi-
ments made before the Royal Society, was the first who
proved that its powers were truly electric. The effects
of the torpedo (he observes). are absolutely electrical, »
forming its circuit through the same conduciors with
electricity, and being intercepted by the same non-
conductors, as glass and sealing-wax. The back and
breast of the animal appear to be in different states of
electricity ; by a knowledge of which circumstance, we
have been able to direct his shocks, though they were
small, through a circuit of four persons, all feeling
them ; and also through a considerable length of wire
held by two insulated persons —one touching the lower
surface of the fish, and the other the upper. When
the wire was exchanged for glass or sealing-wax, no
effect could be obtained ; but as soon as it was resumed,
the two persons became liable to the shock. Number-
less experiments of this sort determined the choice of
the conductors to be precisely the same in the torpedo
as in the Leyden phial ; while the sensation occasioned
by one and the other, to the human frame, are precisely
similar. It is remarkable that the torpedo, when insu-
lated, is able to give us, insulated likewise, torty or
fifty successive shocks from nearly the same part, and
with little or no diminution of force ; and these are so
rapid, that Mr. Walsh says he had taken no less than
fifty in succession. from an insulated torpedo, in the space

N 3

1§2 CLASSIFICATION OF FISHES,

of a minute and a half. All these experiments confirmed
the belief that the electricity of this fish is condensed, in
the instant of its explosion, by a sudden energy of the
animal: the effect appears to arise from a compressed
elastic fluid, restoring itself to its equilibrium in the same
way, and by the same media, as the elastic fluid com-
pressed in charged glass. Notwithstanding the weak
spring of this electricity, Mr. Walsh was able to convey
it through a circuit formed from one surface of the
animal to the other, by two long brass wires and four
persons ; which number, in some of the experiments,
was increased even to eight: every person was made to
communicate with each other, and the two outermost
with the wires, by means of water contained in basins
properly disposed between them for that purpose. It will
be unnecessary to follow Mr. Walsh’s more minute de-
tails of these experiments; and, after all, he observes
that the effects produced on these occasions by the tor-
pedo, resembled, in every respect, a weak electricity.
It was further ascertained that the shocks were much
stronger when the fish was taken out of the water than
when it was emerged in it; or, as our author observes,
“the shocks in water appeared, so far as sensation
could decide, not to have near a fourth of the force of
those that took place at the surface of the water, nor
much more than a fourth of those given when the fish
was entirely in the air, on being raised by the hand.”
Finally, we may observe that this power is possessed,
not only by the young torpedo on its birth, but even
while it is yet a foetus in the body of the parent animal.
This fact was ascertained by Spallanzani, on dissecting
a torpedo in a pregnant state, and which contained in
its ovarium several roundish eggs of different sizes, and
also two perfectly formed foetuses, which, when tried
in the usual manner, communicated a very sensible
electric shock; and this was still more perceptible
when the little animals were insulated by being placed
upon a plate of glass. The electricity of the torpedo
is altogether voluntary ; and sometimes, if the animal

SIZE, ETC. OF THE TORFEDO. 183

is not irritated, it has been affirmed that it may be
touched, or even handled (?), without being provoked
to exert its electric power.

(162.) There can be no doubt that several species
have been confounded under the common name of
Torpedo electrica, which likewise possess the same pro-
perties; but whether all those of the same external
form are likewise electric, is very uncertain. Two
species evidently inhabit the British seas—one of which
is spotted, the other not: both, however, are of rare oc-
currence. A specimen mentioned by colenel Montagu,
taken off the coast of Tenby in Wales, weighed about
100 lbs.* It has been thought that the torpedo is
a slow and inactive fish ; and that, consequently, these
powers have been given it for more readily procuring its!
food, by killing such small fish as pass near it. That this
electric power is so used, as well as for a means of
defence, is highly probable: but it appears to us that
there is nothing in the structure of this fish to render it
slow or inactive; on the contrary, the great develop-
ment of the pectorals, and even of the ventrals,
clearly shows that these fishes must be endowed with
the power of swimming, for a short distance, with great —
rapidity, —fully as quick as any of the true rays or
thornbacks. Nor do we think the following sensible
observations of Mr. Couch militate against what we
here advance :—‘* One well known effect of the electric
shock is to deprive animals killed by it of their organic
irritability, and, consequently, to render them more
readily disposed to pass into a state of decomposition ;
in which condition the digestive powers more speedily
and effectually act upon them. If any creature, more
than others, would seem to require such a preparation
of its food, it is the torpedo, the whole canal of whose
intestine Is not more than one half as long as the
stomach.” + Recently}, this genus has been divided

* Yarrell’s Fishes, vol. ii. p. 411.

¥ Ibid. vol. i. p, 412.
= Muller and Henle’s arrangement. Mag. Nat. Hist. No. xiv. p. 90.

w 4

a eed

154 CLASSIFICATION OF FISHES.

into four sub-genera, the characters of which, however,
do not appear in the paper alluded to.

(163.) We finally have resolved to place the genus
Squatinu between the torpedos and the snout rays (Rhi-
nobates); by which situation it preserves its analogy to
all those sharks which have, like this, the muzzle obtuse,
and the mouth terminal. The annexed representation
of S. angelorum ( fig. 22.), found in our seas, will give

the reader a better idea of this singular fish, than a
laboured description.. Its whole aspect is certainly
more like that of a ray than of a shark. The circum-
stance of the branchial apertures being placed beneath,
joined with the very great development of the pectorals,
and the flattened obtuse head (intermediate between that
of a torpedo and a ray), ail conspire to point out its
natural station to be in the present family. The only
character, in fact, which it seems to possess in common
with the sharks, is that of having the tail fully developed,
and the pectorals detached, in frent, from the head : but
these considerations are not sufficient, in our opinion, to
counterbalance those just stated ; to which may be added,
the depressed form of the whole fish, which shows that its
habits are naturally very much the same as those of all
therays. Cuvier refers the Squalus aculeatus of the Me-
diterranean to this group.—a fish we have not seen; and
Le Sueur has beautifully figured another (8. Dume-
rilli, fig. 23.), which inhabits the coast of America: the
shape, situation, and proportion of the two dorsals and
of the caudal are precisely the same as what we see in

RHINOBATiINZ, OR SNOUT RAYS. 185

many of the torpedos. Of our British species Mr. Yar-
rell says, that it some-
times is caught of the
weight of 100 lbs. ;
that it is very voraci-
ous, and feeds on the
smaller flat fish, which,
like itself, swim close
to the bottom; occa-
sionally, like them,
also, hiding itself on
the loose soft soil that floats over it. We cite this re-
mark, because it is in further confirmation of our belief
that the Pleuronectide represent the Raide, not only in
the disk-like shape and compression of their bodies, but
also in their food and modes of life.
(164.) The third and last aberrant division of the
rays is represented by the genus Rhinobates, of which
the annexed cut of R. Riippellii Sw. (fig. 24.) is a very

=~

=

=
Za

24

ZZ vg
Y,

A ZV
covu.os eo 202 oI

good example of the whole. This group has also been
divided into five sub-genera, of which there is only one
species in each; but the characters appear to us so
slightly defined, that we cannot at present adopt them.
Nevertheless, one of them, named PlatyrhinaM. and H.,
is described as having ‘the body orbicular;’? in which
case it is much more probably a sub-genus of the Tor-
pedine than of the present group; an orbicular body
being one of the primary distinctions of those fishes.
Certain it is, however, that the typical form of the sub-

186 CLASSIFICATION OF FISHES,

family now before us, is to have the snout considerably
elongated, and the body and tail more resembling that
of the true sharks, than any of the other rays; in other
words, it has the head of some of the long-snouted rays
— such, for instance, as the Raia chagrinea * Penn. —
placed on the body of a shark. None of these fishes have
been found in the northern seas, nor are we acquainted
with any from the Mediterranean ; several, however,
occur in the Red Sea, and on the shores of India, of which
representations will be found in Riippell’s, Russell’s,
and Hardwick’s collections of figures. The sub-genus
Rhina probably belongs to the Torpedine, since it has
the muzzle short, large, and round, instead of length-
ened and pointed.

(165.) The circular succession of the Raide is thus
seen to be all but perfect, since the only intervals in the
chain occur between Squatina_ and the two types on each
side of it— Torpedo and Rhinobates. It is clear that
Rhinobates is but the incipient form, as it were, of the
saw-fish ; and as these latter have always been regarded,
and justly, as coming within the confines of the Squalide,
we must place Rhinobates as the last of the Raide.
We have no means of judging, either from specimens,
or a good figure and description, of the true nature
of the Squalus aculeatus, which Cuvier refers to the
Squatine ; — Does it really belong to that genus, or to
a different type among the Squalide? In either case,
its spined back shows its direct relation to the thornbacks
(Raia); while its terminal mouth indicates the same with
regard to Squatina, Cestracion, &c., and the other chi-
ronectiform types. Under the belief, therefore, that the
above series is the natural one, we shall at once proceed
to investigate the analogies resulting from this view of
the Raide, by comparing the divisions with those of a

* This species may be cited as an additional instance of the insufficiency
of arranging the cartilaginous, or, indeed, any other fishes, upon a primary
regard to their teeth : for not only do these organs vary in different species,
and in the very same individual at different ages, but actually in the
sexes: the blunt tesselated teeth of the Raia chagr nea become pointed

in the adult male, while in the female they never alter,— See Yarrell, vol. ii.
p- 416.

ae.

ANALOGIES OF THE RAID. 187

higher denomination, under.which we have placed the
whole of the cartilaginous fishes.

Analogies of the Rays to the CARTILAGINOUS

FAMILI&s.
Sub-families of 3 ; Families of the

Pheinade Analogical Characters. , CARTIEAGINES:

: Head surrounded with the pectoral
Trygonine. Ane. P i RAIDm&.
Pterocephaling. Head distinct from the pectorals. SQUALID.

A ; Fore part of the head Soules: f Pah
Torpeding. DROAGL POLYODONIDE.
Squatine. f ao a protractile, and furnished Pea TURTON
Rhinobatine. Body much lengthened. CHIMERIDE.

We must, in the first place, remind the reader of the
arguments already used in proof that the cartilaginous
order of fishes corresponds to the natatorial order of birds,
where the wings are universally more developed than in
any other type. Now, in proof that this analogy is true,
we see that the pectoral fins, which correspond to the wings
of birds, are more developed among the rays than in the
sharks ; and thus we find not only that the Raide stand
at the head of the cartilaginous order, but that the
Trygonine, from having the pectorals so much developed
as to surround the snout, become pre-eminently typical
— and being so, are the representatives of their whole
family. The eagle rays and the sharks, again, stand
opposite each other ; and we discover an analogical cha-
racter, in the head of both being distinct from the pec-
torals. It is evident, that, in whichever family we place
Squatina, it is analogous to the sturgeons, and to the
eirrated sharks ; for these are the only cartilaginous fish
which have cirri to their jaws; and if Cuvier is correct,
that the Squalus aculeatus is a Squatina, we shall have
another point of strong resemblance to the sturgeons,
which are universally armed with prickles. The broadest
snouts among the rays are seen in the torpedos; and the
broadest, as well as longest, in all the cartilaginous fami-
lies, is in the genus Polyodon; both being the fissirostral

188 CLASSIFICATION OF FISHES.

types of their own proper circles. The snout rays, again,
are the longest, in their bodies, of the Raide—a character,
likewise, which more especially belongs to the northern
Chimera among the sharks; so that the analogies between
both become complete; and the series of affinities, in
which we have placed the groups, is doubly corroborated
by the series of analogies occurring in precisely the
same order.

(166.) The rays, properly so called, is the only one of
all the divisions of the family whose types are suff-
ciently made out to show a circular series ; being the
most typical, it is, as usual, very full of species, and tne
variety of forms is accordingly proportionate. From
ignorance, however, of their manners, and of very many
other points in their structure (which we may hope the
two learned naturalists now engaged on these fishes will
clear up), we cannot trace their analogies, in one instance,
so perfectly as could be wished ; but all the others are
so remarkable strong, that we have no hesitation in
laying the following table before the reader :—

Analogies of the Tryvcontnz and the Ratz.

Genera of the Families of the

nian cane Analogical Characters. Raa
Trygon Antia. Tail with narrow fins. TRYGONINE.
Pastinaca Antig. ‘Tail without any fins. PTEROCEPHALINE.
Pteroplatea M.,H. Muzzle broad and very obtuse. TORPEDINE.
Raia Linn. Back often armed with spines. SQUATINZ.

Snout produced: no spinal pro-

Anacanthus. cess on the tail.

RHINOBATINE.

The division we have formerly made between those
sting rays which have a fin either above or below their
tail, and those in which all vestige of fins disappear,
now turns out to be precisely one of the leading discri-
minations between the two great divisions of the sting
rays, and the eagle rays; all the latter having the ter-
mination of the tail quite naked. The thornbacks, and
the other rays, find their prototypes in Cuvier’s Squatina
aculeatus ; these groups, in fact, being the only ones

THE PLEGTOGNATHES. 189

wherein the back is furnished with spines. The genus
Anacanthus, as its name imports, contains those rays
which have the character of Trygon, but without
their sting: the snout, also,eif we rightly understand
the genus, is produced; so that it becomes at cnce ana-
logous to the Rhinobatine; it is this genus, however,
upon which, not having had the means of examining, a
slight’ doubt may arise; but the connection of the
Trygonine to the Rhinobatine is so unquestionable,
that it matters very little to our present purpose, upon
which link in the chain we fix. for a type, supposing
Anacanthus not to be one. We need not pursue this
subject further, because these analogies carry with
them numerous others, and wil! enable the reader to
pursue the subject through all the chief groups of the
class,

CHAP. VII.

ON THE FLECTOGNATHES, OR CHELONIFORM ORDER.

(167.) Tue order now before us, notwithstanding
the diversity of characters it presents as a whole, may,
nevertheless, be pronounced one of the most natural in
the whole ichthyological circle. Under the name of the
Branchiosteges, it was so considered by Artedi; and
although that great father of our science did not detect
the concealed nature of the operculum, yet his views of
the true extent of the group appear to be more just
and comprehensive than those of the moderns. Our
own opinions, at least, are more in unison with those of
Artedi, who includes in this division the genera Cyclo-
pterus and Lophius. MM. Cuvier, on the contrary, con-
fines it entirely to the Balistide, or cheloniform genera

190 CLASSIFICATION OF FISHES.

(B. ornatissimus, fig. 25.), which constitute his order

Piectognathi: this name, however, we retain, since, by.
the anatomical investigations of this eminent naturalist,
the true nature of the gills were first made known.
The most general characters belonging to this group
will now be noticed in the order of their prevalence.
In the first place, they are universally destitute of
true or imbricate scales: the body is soft and naked,
as in the Chironectide, or frog-fish ; or it is hard and
coriaceous: in the Balistide, or file-fish, the skin is
hard, and scored * into diamond-shaped patterns (fig.
26.) : when magnified (a), the
granulations are distinctly seen,
the interstices being smooth. In
others, the body iseither covered
with spines, or incased with
bony plates, the sutures of which
fit to each other, and do not,
as in ordinary fishes, lay in an
imbricate or tile-like manner upon each other. The
skeleton is neither strictly osseous, nor cartilaginous, but
is a mixture of both structures ; presenting a gradation
from the soft and cartilaginous structure to that which
is hard and truly osseous: the assertion, therefore, that
its “entire general structure is that of ordinary fishes,”
is not borne out by fact ; since M. Cuvier himself
acknowledges, that, in the majority, the bones are semi.

* Scored or reticulated: we use this term to denote the peculiar reticu-
lated markings on the shagreened skin of certain Lalistide, which give
them, at first, the appearance of possessing diamond-shaped scales.

CHARACTERS OF THE PLECTOGNATHES. 191

cartilaginous, or that they take a long time to harden;
and that, in all, ‘‘ very small vestiges of ribs are to be
found.’’ The third is an equally important character:
the operculum and branchia, indeed, exist; but in a very
imperfectly developed state, when we compare them
with those of ordinary fishes ; and they are altogether
concealed by being covered with the thick skin of the
body, which only leaves a small cleft, or spiracle, by
which the water taken into the mouth escapes. Hence
they breathe, like the Cartilagines and the apodal order,
by spiracles. The other characters of the group are
secondary, because they serve more to determine the
family divisions, than to characterise the entire order.
In the Balistide, or tortoise-fishes, the maxillary bone
is soldered to the intermaxillary, which alone forms the
jaws, and to which the palatine arch is united by a
suture with the skull, so that it possesses no power of
motion ; the mouth is thus most imperfect, and is
always very small. In another typical group, the eyes
are also very small, but are placed almost vertically: the
pectoral fins are very large, and often have some re-
semblance to feet, being placed on a sort of peduncle,
which enables these frog-fish to crawl upon the ground:
the mouth opens upwards, and the lower jaw is longest.
In one group only—the most aberrant of all —is there
a variation in the branchia, indicating an affinity to the
class of animals which next succeed, namely, the Am-
puibia.

(168.) Thus characterised, as a whole, the Plecto-
gnathes appear to arrange themselves into the following
natural families :—1.The Balistide, or cheloniform fishes,
having the body oval or round, and almost always
covered with osseous plates or armed with prickles:
2. The Chironectide, or frog-fishes, where the pectoral
and ventral fins, particularly the former, assume the
appearance of feet; the body being thick and smooth:
3. The Lophide, having the head enormously large and
greatly depressed: and, 4. The Sygnathide, of a long
serpent-like shape, covered with hard plates, and the

192 CLASSIFICATION OF FISHES.

muzzle excessively long. The contents of these divisions
are very unequal; and it will be seen that, from the
absence of a fifth type, they do not form a circular
group: but this is a matter of inferior moment;
since it will subsequently appear that the four, above
named, find their representatives in four of all the other
ichthyological circles. It may here be observed, that
very few of this order are found in the European seas,
and that none of them are esteemed as food.

(169.) The family of Balistide, or cheloniform
fishes, is the most interesting, as it is the only one in
which any vivid colouring is found; many of the
species, indeed, are remarkably beautiful: neither is the
form, in general, devoid of symmetry. They are very
numerous in tropical seas, and present many modifica-
tions of form, which have not yet been accurately
defined and arranged; but only one species (the
Capriscus Rondeletii of our celebrated Willughby)
occasionally wanders to our coasts. Having paid much
attention to this interesting family, we are enabled to
determine, as we believe, the five sub-families. The
two first are the Ostracine, or trunk-fish, and the
Balistine, or file-fish : these we regard as typical: they
are distinguished by having the body covered with an-
gulated plates, or hard and reticulated skins ; the mouth
being furnished with real teeth. The three aberrant sub-
families are the T'etradonine, or hare-fish ; the Diodo-
nine, or globe-fish ; and the Cephaline, or sun-fish: the
circular succession of these groups into one is effected
by the Orthagoriscus oblongus*, whose hard skin is
divided into those angular compartments which is the
peculiar characteristic of the Ostracing. We shall now
collect together the few points of general interest that are
at present known respecting these groups; and then
compare them, in their analogical relations, with others.
They have hitherto been much neglected, even in the
latest systems ; and this will account for our not being

* Bl. Sch. pl. 97.

THE BALISTINE, OR FILE-FISH. 193

able, in the synopsis, to determine many of the subor-
dinate forms.

_(170.) The Balistine, or file-fish, are not so
grotesque in ‘their general form as the diodons and
tetraodons, and are much more beautiful fish: the
eolours are generally rich and vivid, and the body is
not armed with spines (B. erythropterus, fig. 27.).

The greatest number are confined to the still waters of
tropical seas, and principally those of India and Ame-
rica. A very singular circumstance connected with this
species has given rise to the name now applied to the
whole group. The typical Balistes have two dorsal fins,
one of which is fronted with a strong bony spine.
Salviani was the first to discover that the bones or rays
of this fin are so contrived as to act in concert, with
considerable force, in suddenly elevating the fin at
the pleasure of the animal: though the foremost or
largest be pressed ever so hard, it will not stir; but if
the last or least ray of all be pressed but very slightly,
the other two immediately fall down with it, as a cross-
bow is let off by pulling down the trigger. For this
reason, the fish is called at Rome Pesce balestra. These ©
fish are provided with true teeth, of which eight are
in each jaw. There are no true ventral fins; but, in
most, the bone of the pelvis is prolonged beyond the
skin, and is even furnished with bony rays connected
by a membrane, so as to constitute, in effect, a true
ventral fin,

(171.) We arrange the Balistine under five prin-

VOU. oO

1G4 CLASSIFICATION OF FISHES.

cipal divisions or gerera; and these include several
distinct modifications of form, which take the rank of
sub-genera. The great number of species, however,
which swarm in the Indian and Pacific Oceans, will con-
siderably augment these minor groups, when their
peculiarities of structure are better understood.* The
two first, or typical genera, are Balistes proper, and
Capriscus, a name employed by Willughby and the
old writers to designate some of these fishes, and which
will be preferable, on that account, to a new one of our
own: both these are distinguished by having the body
covered with large diamond-shaped divisions, scored, re-
sembling network, separated from each other by a suture,
as if the hard skin had been regularly scored: hence
their bodies may be termed mailed and tessellated. In
Balistes, the tail is armed with three or more rows ~of
acute prickles, or lancets, which are entirely wanting in
Capriscus (C. velata, fig. 28.): each of these, again,

contain several sub-genera, readily distinguished in the
different forms observed in the first dorsal and the caudal
fins, and in the structure of the pelvis.f The aberrant
genera, as usual, contain fewer variations of form; all
three, however, are at once separated from the typical
groups by the scale-like reticulations on their body,

* Having long prosecuted, at intervals, a particular analysis of this fa-
mily, with drawings of all the species we can procure, we beg to solicit
from those of our readers who have the means of assisting us, preserved
specimens (either dried or in spirits); and, more especially, the loan of co-
loured sketches or drawings made from the life : we make the same request
in regard to the chetodons, and the silures (Silurid@).

+ The arrangement of M. Cuvier, founded-upon the number of rows of
ancets on the sides of the tail, is obviously artificial, and otherwise ob-
ectionable, particularly as the above-characters are entirely overlooked.

THE OSTRACINE, OR TRUNK-FISHES, 195

above mentioned, being either very minute, or the skin
only granulated. These genera have been already
named by Cuvier, Alutera, Triacanthus, and Monocan-
thus. In this latter genus, some of the forms, as that

of Mon. bifilamentosus Less. (fig. 29.), are singularly
grotesque ;_ but

the sub-genera
have not been
investigated, and
much remains to
be done in deter-
mining their lo-
cation: some will,
doubtless, enter
as aberrant types
in the other ge-
nera; nor is it at all probable that the genus Tria-
canthus should contain only one typical example. If
the ichthyologist wishes to study the relations of all
these new divisions, he will find they follow each other
in the same series as that in which we have noticed the
primary families of the entire order. The analogies,
indeed, of the whole of this family, to that of the Che-
todonide, with which so many writers have incidentally
compared them (one of the best proofs of the analogy
being natural), are most particularly beautiful. But
we have no space for this inviting subject.

(172.) The sub-family Ostracine is composed of the
trunk or tortoise fish (O. argus Riipp., fig. 30.) ; so called

from their bodies being often quadrangular like a trunk or
0 2

196 CLASSIFICATION OF FISHES, —

box, and from the plates by whichit is covered resembling,
in shape, those of the chelonian reptiles, or tortoises.
They are all fish of a small or moderate size, with rather
large eyes ; but very small mouths, armed with a few
conical and real teeth: so different is their internal
structure from that of the osseous fishes, that M. Cuvier
confesses that the greater portion of their vertebre are
cemented together ; and the ribs exist only in a rudi-
mentary state. They have no ventral fins, and they
have but one dorsal. This evident inferiority to the
Balistine at once shows them to be the sub-typical
group. Some of the species are furnished with
horn-like protuberances, giving them a yery grotesque
appearance; and all are so compacily covered with the
impenetrable and immoveable cuirass of the body, that
they have only the power of moving the tail, the fins,
and the thin lips of their little mouth. Most of the
species are found in the Indian seas.

(173.) We now come to the aberrant group, composed
of the Tetraodine, the Diodonine, andthe Cephaline:
these three are distinguished from the former by having
no true teeth, these processes being supplied by certain
lamina of an ivory substance placed inside of the jaws.*
They are also entirely destitute of the squamular plates ;
their body being covered with a simple skin, which is
either rough or beset with spines. In other respects
they have a general resemblance, both in shape and
structure, to the other cheloniform fishes; the dorsal fin,
however, is invariably single. A prejudice against
eating these fish seems to be prevalent in all the coun-
tries where they are found, and also a general belief that
most of the species are poisonous.

(174.) The Tetraodine t, or hare-fish (Tet. diade-

* M. Cuvier remarks, that these lamine of the jaws are essentially true
teeth, united together and succeeding each other as they are successively
worn out by the effect of triturition. If this be true, which there seems no
reason to doubt, it reveals an absolute point of analogy to the gliriform
quadrupeds, where the cutting teeth are renovated nearly in the same
manner.

+ It seems advisable to designate this group as the Tetraodine ; the
Tetraonine being a sub-family of rasorial birds.

THE TETRAODINZ, OR HARE-FISH. 197

matus Riipp., fig. 31.), have acquired this name from
the sharp edges of the jaws being divided in the middle,

SO as to present the appearance of four teeth — two above,
and two below; and this structure also gives them a
remote analogy to the lips of the hare. The tail is
more lengthened than in any other of the cheloniform
fishes, in accordance with what we should expect in the
apodal type: the spines on the body are so short, that
they are mere prickles; while, in some, they merely
assume the appearance of rough asperities on the skin ;
the body is nevertheless, very slimy.* These fishes are
remarkable for having only three gills or branchia on
each side, —a structure which prepares the passage from
the Balistide to the Chironectide, where the number is
precisely the same. ‘The rays of all the fins, except the
caudal, are covered by a thick skin ; and all the species
have small mouths with fleshy lips. The food of all
these small-mouthed fishes appears to be crabs and shell-

* Hamilton’s Gangetic Fishes, p. 5.
0 3

198 CLASSIFICATION OF FISHES.

fish, which they grind with ease by means of their
strong enamel jaws or hard teeth. As a passage from this
to the next sub-family, we arrange the singular shaped
Triodon bursarius Reinw. (fig. 32.), for it partakes of
the general structure both of one group and the other.
(175.) The Diodonine, or globe-fish, have the jaws
not only destitute of any apparent teeth, but are even
without any divisions, so that each remains as one entire
piece ; yet behind their cutting edges, however, is a
slightly rounded protuberance, marked by regular fur-
rows, which supplies the place and office of grinding
teeth: the form of the body is more globose than in any
of this order ; and, when distended, becomes, in some
species, absolutely round: they are all armed with nu-
merous long and acute spines, of which, as M. Cuvier
observes, a horsechestnut is a good miniature resemblance.
These fishes are confined to the tropical and warm
latitudes, chiefly of India ; and none seem to exceed 2
very moderate size. The gills are five in number, and are
very slightly developed. The globe-fish, as well as
the tetraodons, are remarkable for the power of inflating
their body like balloons, to an enormous size, by swal-
lowing the air, and thus filling their stomach. This is
obviously a means of defence against their enemies:
M. Cuvier, indeed, remarks, that, when thus inflated,
they turn topsy-turvy, the stomach being uppermost,
and they float to the surface, without being able to
direct themselves. This, however, seems somewhat im-
probable, and by no means according with what we should
naturally expect from fish so well provided with the
means of actively repelling their enemies. The follow-
ing account of the Diodoa histriv L., by Dr. Hamilton,
seems a much more natural relation of its habits. ‘ This
fish is said to afford an amusing spectacle when taken by
a line and hook properly baited with some small crab or
other crustaceous animal: after having played round the
bait for some time in various directions, it seizes it with
a sudden spring ; but finding itself hooked, it exhibits
every appearance o* the most violent rage, inflating its

THE CEPHALINZ, OR SUN-FISH. 199

body and elevating its spines to the highest possible
degree, as if endeavouring to wound in all directions ;
till, after having tired itself by its vain efforts, it sud-
denly expels the air from its body, and for some time
becomes entirely flaccid. When drawn towards the
shore, however, it redoubles its rage, and again inflates
its body ; in this state it is left on the sand, it being im-
possible to touch it without danger till it is dead.”
(176.) The Cephaline, or sun-fish, are a most sin-
gular group: they grow to an immense size, and look
more like the dissevered head of a fish, than the entire
animal itself. There are but few species yet determined,
only one of which (Cephalus brevis) is found in the
European seas. It grows to a vast size, having been
said to reach the length of eight or even ten feet, and
the weight of 500 lbs.: it is sometimes observed to lie
on its side on the surface of the water ; on which oceca-
sions it may be captured without difficulty. The false
teeth and mouth very much resemble those of the dio..
dons, — theformer being undivided. M. Cuvier affirms
that it is destitute of an air-bladder; in which case it
probably swims at the bottom of the ocean. A new
ean of these fishes, Pedalion gigas Guild* (fig.33.),
of gigantic size,
inhabiting the
_ ~West Indies, will
- = be subsequently
=described. The
¢ sun-fish are sup-
J posed to feed
j principally on
crabs and shell-
fish ; and they
are known to ex-
hibit, during the
night, a high de-

* The late and lamented Lansdowne Guilding, who, to the regret of
science, fell a victim to the ardour of his researches in the island of St.
Vincent,

0 4

200 CLASSIFICATION OF FISHES.

gree of phosphoric splendour. The skin is not armed
either with spines or prickles, but is nevertheless very
rough ; nor can these fish dilate their stomach.

(177.) Before proceeding further, let us see what
results attend this new distribution of the cheloniform
family, by comparing the several divisions, all of which
have been instituted long ago, and adopted by M. Cuvier,
although in a different series to that in which they now
appear. We shall place, therefore, the sub-families in
one column, and the primary orders of fishes in another.

Analogies of the Batistipz, or Cheloniform Fishes,
with the Primary ORDERS.

Sub-families of 2
the BALISTIDz. Analogies. Orders.

4: Balistine. Dorsal fins two, armed with P aca

spines : ANTHOPTERYGES

Dorsal fin generally one; therays
soft.

Caudal, i dorsal, and anal ns? Apopus.

Jaws neither bearing true, nor re-
presenting false teeth.

5. Tetraodine. 2 CARTILAGINES.

2. Ostracine. i MALACOPTERYGES.

2. Cephaline.

4. Diodonine. i PLECTOGNATHES.

(178.) The analogies of the first, second, and third,
are so striking, that nothing need be said by way of
further illustration. In respect to the third, we have
the means of stating, that, as it represents the apodal
order, by the union of its fins, so does it the Syngnathide,
not, indeed, in the form of its body, for nothing can
be more dissimilar, but in a part of its structure to
which we should hardly have looked for a resemblance,
namely, the eye (fig. 33. a): this, we can state, on Guild-
ing’s authority, is of that particular construction seen in
the chameleon reptiles and the syngnathian fishes ; it is
conical and versatile, so that the fish can look afeeeae
Ways at one time,—a most admirable provision for
such animals as are slow and heavy in their motions,
like the sun-fish and chameleons; or, by being fixed,
(like the pipe-fish) require great quickness of sight and

THE CHIRONECTIDZ, OR FROG-FISHES. 201

a wide range of vision. One of the most striking cha-
racters of the Plectognathes is the deficiency of true
scales and teeth; the latter is consequently more appa-
rent in the Dindanide than in any other, eee in
several other respects it is aberrant.

(179.) Again, some very singular coincidences will
arise by comparing this family with the primary groups
of the quadrupeds and birds, but the analogical rela-
tions will, of course, be very remote. The Diodonide,
for instance, are the most aberrant of the whole circle,—
a station likewise held by the porcupines, the hedge-
hogs, and the spined rats, in their own proper circles
among quadrupeds: hence we immediately see one of
the analogical reasons, so to speak, of Diodon having
the longest spines of all the cheloniform divisions ; for,
had it been otherwise, there would, apparently, be want-
ing some one character by which all these diversified
groups would be analogically related. But into these
details it is needless to enter, seeing that, if we have not
erred in the foregoing table, all other analogies, near or
remote, will follow as a matter of course. We leave
the presumed analogy of the Tetraodine and the Carti-
lagines for future determination. When we consider that
little or nothing is known of the manners of these fishes,
and that, if the peculiar form of the eye in the Cepha-
line had not been discovered, nothing would have been
left te show their analogy to the Syngnathide, we need
not be sceptical on this point of our comparison: if
nothing was left to be discovered, our knowledge of
nature would be perfect.

(180.) The second family we consider to be repre-
sented by the CurronEcTID#, or frog-fishes, at present
comprised under one genus ; yet the species and forms
are so few, that its internal relations cannot be made
out. The characters of this group are so different from
the Acanthopteryges, where Cuvier has placed it, that we
cannot discover one solitary character they possess in
common. The skeleton of Chironectes, instead of being
osseous, is, as M. Cuvier admits, semi-cartilaginous: the

202 CLASSIFICATION OF FISHES.

rays of the fins are all soft: the branchial opening, as
in the Balistide, is confined to a small slit or spiracle :
- the operculum is concealed beneath the skin, so that the
branchia themselves are concealed. Not one of these
characters belongs to the typical osseous fishes; while,
on the other hand, every one of them are characteristic
of the order now before us. Like the aberrant Balistide,
the branchial arches are very few—only four in num-
ber; and like them, also, these fishes have the power of
inflating their bodies like a balloon when agitated by
fear or anger. Their remaining characters, however, are
altogether peculiar; and even their very aspect is suffi-
cient to distinguish them (Chir. histrio, fig. 34.) ; they are

the most grotesque—we had almost said the most hide-
ous—of all fishes, and, as their vernacular name of frog-
fish implies, they have nearly as much the appearance of
frogs or toads as of fish; this similarity may be perceived
in the head of Malthe nasuta(fig. 35.). The late Mr. Ben-
nethas very justly insisted on
2. the intimate affinity between
these strange-looking crea-
tures and the file-fish, or Ba-
listide, — an affinity which
» has only been disturbed, as
3 we believe, in the Régne

Animal. The imagination
can scarcely conceive more fanciful forms than such as

LOPHIDA, OR FISHING FROGS. 203

are actually found in this group; and the monstrous
combinations which painters have represented under the
aspect of animals, can scarcely surpass the singularity
of many of these real fish. True it is, that they have
their representatives, like the Balistide, in other groups;
but a slight degree of attention, even to their external
characters, will prevent them from being confounded
with any other; for, as M. Cuvier well observes, inde-
dependently of their semi-cartilaginous skeleton and
naked skin, destitute of any scales, the frog-fish have
each of the pectorals supported by two bones, analogous
to the radius and ulna of the frogs, although, in reality,
they belong to the carpus, and which, in this group, are
longer than in any other. The ventrals, again, are placed
much before the pectorals, and stand, as it were, upon
peduncles ; they are thus enabled to perform the office
of feet. The effect of this singular organisation is, that
these fishes can “ creep almost like small quadrupeds ;
the pectorals, from their position, performing the office of
hind feet ;” and their nature is so truly amphibious, that
they can ‘‘live out of the water for two or three days;”
they are, in fact, so tenacious of life, that they have been
transported alive from the Tropics to Holland, “ where
they sold as high as twelve ducats apiece.’ M. Cuvier,
with his usual anatomic skill, has determined the cha-
racters of many distinct species, confounded under the
specific nameof Histrio,—a name not given, as somehave
supposed, from any fancied activity of these animals —
for they are remarkably heavy and slow — but from the
patched and party-coloured spots with which they are
_ variegated. *

(181.) We separate from the last group the Lophide,
or fishing frogs, which, although few in number, and
evidently connected to the Chironectide, nevertheless
present us with so many peculiar characters, that we
look upon them as representations of a family, rather
than of a genus. These reptile-looking fishes have the
head enormously large, broad, and depressed ; the mouth
very wide, armed with pointed teeth, and furnished

204 CLASSIFICATION OF FISHES.

with cirri: the branchia are only three, and the spiracle |
small; in other respects, their skeleton,—the situation.
of the pectoral and anal fins, the vertical position of the
eyes and of the mouth,— all approximate these hideous
fishes to the Chironectide, with which they were placed
by Linneus in his genus Lophius. The angler, or fish-
ing frog, of Britain, is a good example: it is said to
hide itself among weeds and mud, where, by agitating
the long filiform processes on the head, it attracts the
small fishes upon which it preys. How far the genus
Batrachus of authors, as it now stands, forms a part,
if any, of this group, it is almost impossible to deter-
mine. It seems clear to us, that several distinct types,
naturally belonging to groups in the acanthopterygious
order, have been placed in this; the essential characters
of which are, that the skeleton is semi-cartilaginous, the
skin invariably destitute of true scales, and the pectoral
and ventral fins pedunculated. M. Agassiz has not
failed to remark the affinity of such fish as his Batrachus
punctulatus* to Cottus, Uranoscopus, and Trachinus,
with which we believe they are truly connected ; but so
much obscurity hangs over this group, that we shall
restrict it, for the present, to those which are destitute
ot true scales, however small, and whose ventral fins
are, in some measure, pedunculated.
(182.) We place the Synenatuin# as the last and
most aberrant type of this order, to which it is to be
referred on account of its imperfect skeleton, its spi-
raculated aperture, and the slight development of all
those characters which constitute the typical perfection
of fishes. Place these singular-looking creatures where
we will, they remain, at present, a very detached
group ; while, if the modification of their branchia,
partly composed of little tufts, is of such primary im-
portance, the silurian Heterobranchi of Geoffroy Saint-
Hilaire have an equal claim to be classed as the re-
presentatives of a distinct order. The mode of re-

* Spix and Agass. Brazilian Fishes, pl. 74.

THE SYNGNATHID#, OR PIPE-FISH. 205

spiration, or the form of the branchia, in all animals,
must be variable in that part of the series of beings,
or those links of her chain, where Nature is about to
assume some other form of structure: it consequently
follows that, in such cases, the character which heretofore
she has assumed, sinks into secondary importance, as a
sole instrument for classification, where it is about to
be quitted and exchanged for another; and if we wish to
follow Nature in her own course, we must not merely look
to one of her characters, but to all. The insessorial order
of birds, for instance, are strictly terrestrial, yet there
is one genus among them—the Cincli, or water ouzels
—which are altogether amphibious. If, therefore, we
are to adopt the idea that the Syngnathide, on account
of their branchia, should be considered the represent-
ation of a distinct order, the same rule, by parity of rea-
soning, should be followed in the case of Cinclus: all
the insessorial birds would thus be divided into two
orders, the terrestrial and amphibial ; the Cinclus alone
coming under the latter. Now, as it is by this order
that the class of fishes pass into that of the Amphi-
bia ; so it is not only probable, but almost necessary
to the harmony of the series, that the great difference
in the repiration of these two classes should be marked
by animals presenting a union of both. The branchia
of the Syngnathide@, as we conceive, are precisely of this
description ; and while this one character determines
the situation of these fishes in the natural series, we
must class them in that order, to which, in all other
points, they bear the strongest resemblance of affinity.
Indeed, there is no great innovation in this, because M.
Cuvier himself places the Syngnathide close to the
Balistide. ‘The structure and appearance of this family
are as remarkable as its economy and habits: From
their long slender bodies, they have got the name of
pipe-fish: the snout is excessively lengthened, and is
terminated by a little mouth, opening almost vertically,
and destitute of teeth: the body itself is covered with
a cuirass of bony plates, which renders it angular: the

206 CLASSIFICATION OF FISHES.

operculum is large, but confined ; and the aperture is
merely a spiracle: the branchial rays are much the
same as in ordinary fishes; but the branchia-themselves,
instead of being fringed or pectinated, form a number
of small rounded tufts or fascicles. In what seems the
typical division, a most singular conformation prevails,
discovered so far back as 1785, by Walcott, a learned
naturalist of this country, whose words we shall here
quote. “ The male (of the Syngnathus) differs from the
female, in the belly, from the vent to the tail fin, being
much broader, and in having, for about two thirds of
its length, two soft flaps, which fold together and form
a false belly or pouch. They breed in summer, the
female casting her roe into the false belly of the
male.”* Here the eggs are matured ; and the young,
when ready, escape from the capsules and shift for
themselves. Nevertheless, it appears certain that, as
in the marsupial quadrupeds, the young again resort to
this natural shelter, even after they have quitted it for
the first time. Mr. Yarrell says, “ I have been assured
by fishermen, that if the young were shaken out of the
pouch into the water, they did not swim away, but when -
the parent fish was held in the water in a favourable
position, the young would again enter the pouch.”
Another extraordinary peculiarity of these fish,—at least,
of those of the genus Hippocampus,—is the prehensile
nature of their lengthened and finless tail; they twist
this member round the stems of marine plants, and
in this position dart upon such small insects or other
animals as come within their reach. ‘‘ The eyes move
independently of each other, as in the chameleon : this,
with the brilliant changeable iridescence about the head,
and its blue bands, forcibly remind the observer of that
animal.” This analogy is still more perfect, from the
fact of the chameleon fixing itself, as it were, by the
tail, when looking out for its prey, precisely in the same
manner as the Hippocampus. Analogies so strong be-

* See Yarrell’s British Fishes, vol. ii. p. 328., where the reader will find
many other interesting particulars relative to the group.

THE SYNGNATHID#, OR PIPE-FISH. 207

~ tween fish, reptiles, and quadrupeds, are too remarkable
to be incidental; and it is an extraordinary fact, that, if
we place the circles together, which compose these par-
ticular groups, we find that, at these points, at least, they
stand opposite each other.

Analogies of the SYNGNATHIDZ.

Circle of the Circle of the Circle of the Circle of Class
Plectognathes. Reptiles. Glires. Pisces.
Balistide. Saures. ? Acanthopteryges.
Chironectide. Ophides. ? Malacopteryges.
Lophide., Elanosaures. Gliride. Cartilagines.
? Chelones. Hystricide. Plectognathes.
Syngnathide. Chamelides. Marsupide. Apodes.

Our present business is only with the Syngnathide,
which —if the situation here assigned to them among
the Plectognathes be the true one — is found to repre-
sent the chameleons, by its eyes and its prehensile tail ;
the kangaroos (Marsupide), by its marsupial pouch ;
and the apodal fishes, by its want of ventral fins, its
eel-like form, and its very long tail, often destitute of
a distinct caudal fin. The experienced naturalist will
perceive that the series of these columns are not altered
to suit our present object, but entirely repose on the
analysis of the groups, — these reciprocal analogies hay-
ing arisen, as it were, incidentally ; for they have never,
until now, been in the least suspected.

(183.) It will be seen, that in the first column we
have expressed a doubt as to the family type which
should fill up the hiatus between the Lophide and
the Syngnathide ; and yet there is a most extraordi-
nary fish, of a structure altogether unique, which, if
it belongs not to this class, we know not where to
arrange ; we alludeto the genus Polypterus of Geottroy
Saint-Hilaire, a freshwater fish of the Nile, and which
that distinguished zoologist considers as forming in some
degree a connecting link between the osseous and the
cartilaginous orders: the opinion of such a profound
and experienced observer must always carry with it much
_ greater weight and influence than our own— and the

208 _ CLASSIFICATION OF FISHES.

more so in the present case, because it gives a sanction,
in some sort, to the conclusion we have almost come to,
that this idea of its relations is founded in nature, and
that Polypterus, in reality, may ultimately prove the re-
presentation of that fifth type of the Plectognathes,
which we have long been searching after. Limited as is
our space, the structure of this fish is so remarkable, that
it must not be hastily passed over; the annexed figure of
P. Niloticus ( f9.8 36. ) aided e the following description,

cn LEE

ie

abridged from M. Geoffroy’s own account, are therefore
placed before the reader. The usual length of the Nilotic
Polypterus is about eighteen inches, and its colour is sea
green, paler beneath, and marked with irregular black
spots: it bears some resemblance to certain genera in
the Esox, or pike family, by the shape of its body, &c.;
but these seem to be very distant. The shape is long and
anguilliform—the body being nearly cylindrical: the head
is defended by large bony plates (fig. 37.) ; and the body

covered, or rather mailed, with large strong scales of a.
stony hardness, and so firmly attached to the skin that
it is hardly possible to open the fish with a knife ; so
that the natives only draw off the skin whole, after the
fish has been boiled. The pectoral and ventral fins,
but particularly the former, are attached by a sort of

POLYPTERUS NILOTICUS. 209

strong and scaly base or cubit, allowing the same sort of
motion as in those of the Chironectide. The pectoral.
fins are placed close to the head, and are large, broad at
their base, and much rounded ; but the ventrals are ex-
cessively remote from them, being situated very near the
‘caudal, close to which latter is the anal. The caudal is
rounded, and extends further upon the superior than
upon the inferior part of the tail, where it is met by a
long row of numerous dorsal finlets, which extend along
the whole of the back to within a short space of the
head: the number of these finlets varies from sixteen to
eighteen, each being of an oval shape, and furnished
with a very strong spine at its base or origin, while the
remaining part consists of four or five soft and branched
rays, connected by a membrane: the first, or spiny ray,
at about two thirds of its height, sends off a small as-
cending point or spine. The rays of the caudal fin are
soft and articulated, but so disposed in the membrane
as to allow of little freedom of motion. The eyes are
small and round: the mouth of considerable width; and
the jaws furnished with a row of rather small and sharp
teeth ; while the tip of the upper lip has a pair of small
and short cirri: the vent is close to the caudal fin, and
at the commencement of the anal. The branchial aper-
ture is large ; but in place of a membrane, there is only
a single bony plate, or semicircular arch. ‘The stomach
is long and large; the liver long, and unequally Jobed ;
the swimming bladder double, and loose ; the ovaries
long, and the eggs about the size of millet seeds. This
highly singular fish is very rare, and is called Bichir by
the Egyptians. It is generally supposed to inhabit the
depths of the Nile; usually remaining in the soft mud,
which it is thought to quit only at particular seasons :
its flesh is white and savoury. Not having seen this
type, we can only form some opinion of its relations by
the foregoing description of M. Geoffroy. The only
circumstance which makes us hesitate in placing it defi-
nitely in this order, is the circumstance of the branchial
aperture being large, —a structure altogether unexampled
VOL. I. P

210 €LASSIFICATION OF FISHES.

among all the genera of this order we yet know. The
plates of the body, again, although of such excessive
* hardness, appear, from the figure, to assume the form
and imbricate dispositiom of true scales; and this struc-
ture is again at variance with all the other Plectognathes.
Nevertheless, these deviations in an extreme aberrant
type are not altogether insurmountable, under the cir-
cumstances of its possessing the pedunculated pectoral
and ventral fins of the Chironectide and the Lophide,
joined with their small and almost vertical eyes—a
single ray (if we understand the description) to the
branchia—and the uncommon hardness of its covering.
It has an obvious relationship to the genera Lepisosteus
and Sudis among the Salmonide ; but whether this is of
analogy or affinity we know not. The preponderance
of its characters, looking especially at the imperfect de-
velopment of the branchia, appear to us to be towards
those of the Plectognathes; but this must be consi-
dered and determined by others who have better op-
portunities of investigating the question. Presuming,
however, that Polypterus is actually related by affinity
to the Syngnathide and the Lophide, it will be seen that
certain analogical characters would come out, rather
tending to strengthen this supposition. As we have not,
therefore, tested the order by itself, or by that of the
others in the class, we shall do so in the following table: —

Analogies of the Puectognatues and the Orders of

FIsHEs.

Orders of
FISHES.

Families of
PLECTOGNATHES.
The most perfectly organised: the
types with spinal rays, and with $ ACANTHOPTERYGES.
thin membranes to the fins.

Analogies.

Balistide.

Chironectide. Stata EE SES SDE SE M]MALACOPTERYGES.

Lophide. Body mailed: mouth with cirri. CARTILAGINES.
aay Branchia slightly developed: pec- s

Polypteride(?). toral and anal fins pedunculated. SHELTIE ST ETS

Syngnathide. Tail very long, attenuated. APODES.

It is needless to enter further into the first two of
these analogies, because they must be quite apparent to

SUPPOSED ANALOGIES OF POLYPTERUS. Ona

the reader: respecting the third, he will remember
that the primary character of the Cartilagines is not the
construction of their skeleton, which applies only to
them as fishes, but the great breadth of their head,
which extends to all the classes of the vertebrated circle:
hence, even if the skeleton of the Lophide was not
semi-cartilaginous, yet the excessive size and width of
their muzzle assimilates them at once to the rays and
torpedos, which are the cartilaginous types. The ver-
tical cutlines, in short, of the torpedo and the fishing-
frog, are almost so alike, that, if the details were not
filled up, one could hardly be distinguished from the
other. We have already stated the characters by which
Polypterus may be placed among the cheloniform fishes ;
and the analogy of the Syngnathide to the apodal order
has also been touched upon.

(184.) Our chief attention, however, must be di-
rected to Polypterus ; and, with the hopes of arriving
at more definite ideas regarding its true affinities, we.
shall compare the presumed contents of the order Plec-
tognathes with those of the cartilaginous order, thus : —

Analogies of the PhecroenaTtueEs and the CaRTILAGINES.

Family of . Families of
PLECTOGNATHES. Analogies. CARTILAGINES.
ee The types generally furnished with spined
Balistide. rays on their dorsals. } SQUALIDE.
Chironectide. The fins always fleshy, and with soft rays. Raina.
Lophide. Head or muzzle excessively broad. POLYODONID®.

Banchial aperture very open; body
Polypteride(?). mailed; mouth cirrated; gill mem ¢Srentostp,
brane one-rayed, or none.

Tail long, attenuated ; caudal fin obsolete,

Syngnathida. or none; mouth very small. i CHIM&RIDA,

Passing over the three first, we must confess that
our belief in the alliance of the Polypterus with the -
Plectognathes is considerably strengthened, when we
now bring it into comparison with the sturgeons (Sturi-
onide), — a test we had not applied to it when writing
the former paragraphs, because we wish this arrange-
ment to repose, not upon the analogies, but on the
affinities, of the groups. These two types are the only

p 2

wd Nhe CLASSIFICATION OF FISHES.

ones in their respective circles which have the branchial
aperture unusually large; both have the body much
lengthened ; both have the gill membrane either with-
out any rays, or with merely a single pair; in both,
the caudal fin surrounds the extremity of the spine, and
(what is very remarkable), in both these types, the
upper portion of the caudal is longer than the lower.
Polypterus is not more unlike the other cheloniform
fishes, than Acipenser is unlike the rays and the sharks ;
and yet the latter affinity has been universally acknow-
ledged. Unfortunately, however, we cannot make out
whether M. Geoffroy’s original memoir on Polypterus,
which we have not the means of consu!ting, mentions
any thing of the nature of the skeleton : if this is truly
sub-cartilaginous, the question, we think, would be at
once decided; but if it is osseous, the probability is
diminished of this perplexing type entering into the
present order. There is, indeed, an evident — though,
we think, a distant — resemblance between the form of
Polypterus and some of the Siluride, —a resemblance
which results from the one, apparently, representing the
other, as will be seen in the following comparison : —

Families of the PLECTOGNATHES. Families of the MALACOPTERYGES.
Balistide. Pleuronectide.
Chironectida. Salmonide.

Lophide. Cobitide.
Polypteride@ (?). Silurid@.
Syngnathide. Gadide.

But it is not necessary to prolong this discussion ; and
we shall, therefore, at once proceed to the next order.

CHAP. VIII.

ON THE APODES, OR ANGUILLIFORM FISHES.

(185.) Tue fishes belonging to the Linnean order of
ApopEs,—a name which we shall now revive, —are not

THE APODAL ORDER. 213

more numerous than those of the last, yet they are
equally well marked ; nor can they be blended into any
other of the classes without a disregard to those natural
relations which, as we have already seen, constitute them
a primary division. Their slender, cylindrical, and
serpent-like body, as seen in the eels (Ang. acutiros-
tris Yarr., fig. 38.) generally destitute of scales, and

covered with slime, separates them, at first sight,
from the Plectognathes, or cheloniform fishes ; absence
of ventral fins detaches them equally from the order
Malacopteryges ; while the softness of their fins, the rays
of whichare never spinous, no less separatesthem from the
Acanthopteryges. Finally, from both these latter orders
they are further distinguished by having the operculum
and branchia concealed ; the former being covered with
the common skin of the head, which only leaves a small
slit or spiracle (a), by which they breathe. This latter
is a universal character ; and is the more to be valued,
since, whatever eel-shaped or anguilliform fishes are
found scattered in other types, even though they may
have soft fins, or only the rudiments of ventrals, yet
they never have these two characters united with the
spiraculated aperture.

(186.) The Apoprs, as we have already shown,
occupy that part in the series of fishes which mark
the transition from the cartilaginous to the osseous.
Hence no definite character for them is to be derived
from the nature of their skeleton, except. this, indeed,
—that a gradual progression in its development may be
traced in the different families, from the semi-carti-

1D,

914 CLASSIFICATION OF FISHES.

Jaginous structure of Lophius, to the bony skeleton of
most of the eels ; and, at the same time, a falling off
to the faintest indication of the vertebral column, as in
Myzxene glutinosa ( fig. 39.), which all authors agree in

placing near tothe worms. The order, however, stands
in no need of any additional characters than those we have
pointed out. The common eel may be cited as the most
perfectly typical of the whole; while the eel-shaped
lampreys, as all writers agree, indicate the manner in
which the class of fishes blends into that of the annulose
animals, or insects. It may further be remarked that
this, no less than the other two aberrant orders of fishes,
affords us no example of true deci_uous scales. Never-
theless, as the Apodes blend into an order where this
sort of covering is almost universal, we now begin to
see their incipient commencement. Many of the eels,
for instance, have very small scales, but so covered over
by their thick and fat skin, that they are scarcely visible
when the animal is alive. True scales, on the contrary,
are always placed upon the surface of the skin, and, as
every one knows, may be easily detached, or even rubbed
off. Linneus, indeed. placed all fishes not having
ventral fins within the limits of this order, and hence
rendered it a most artificial assemblage. M. Cuvier,
who still preserves the major part as a distinct and
‘natural family,” has much improved on this arrange-
ment ; although he has still left in the group several
genera, which, to our views, are evidently separated from
their true congeners. Ophidium and Fierasfer, from
having unattached opercula, and wide branchial aper-
tures, we arrange with the Gadide : Ammodytes, in like
manner we refer to the Gymetres, or riband-fish, with

FAMILIES OF THE APODAL ORDER, 915

which, also, we suspect some of Gymnarchi are natu-
rally allied.

(187.) Our general sketch of this order must be very
brief ; for, as we have not space for a lengthened expo-
sition of every family, we shall only dwell upon those
which more especially seem to require elucidation; and
the popular history of the eels, and other well-known
fishes belonging to this group, however interesting,
is not exactly suited to the nature of these volumes.
We shall therefore at once proceed to the natural families
of which the order appears to be composed, referring the
reader to the general synopsis for the characters of the
minor divisions.

(188.) The eels appear to arrange themselves into
two divisions: the one, which we designate the Mu-
renide, having two branchial spiracles in their ordinary
position ; and the Sphagebranchide, or sea eels, where
the branchial spiracles are either close together or united
into one, and in both cases are placed under the throat.
These we denominate the typical groups. The three
aberrant depart more or less from this structure, and
each assumes some of the characters of that particular
order to which it leads: thus, the Gymnarchide have
the head and body compressed, as in ordinary fishes,
and the scales are more developed than in the other di-
visions ; because this group, as it appears, leads to the
acanthopterygious order. The Petromyzonide, or lam-
preys, have the skeleton almost obsolete; the body worm-
shaped, and without scales: some of these lead to the
Vermes, and others to the cartilaginous order. Lastly,
we have the Cyclopteride, or suckers, characterised by
their ventral ring; thus connecting the Gymnarchide to
the Plectognathes, by means of the lump-fish and the
Lophide, or frog-fish. A few general remarks upon
these groups seem to be necessary, because, however
nearly we think they are related, they have never before
been assembled together.

(189.) It will be observed, in our last chapter,
that the Lophide are placed at one extremity of the

‘ig eee

210 CLASSIFICATION OF FISHES.

cheloniform order, and at that particular part which
forms a passage to the Apodes. Now this passage, as
we conceive, is effected by that singular-looking fish—
scarcely less grotesque than the Lophius — familiarly
called the lump (Cyclopterus lumpus*): both these
have their skeleton more or less cartilaginous, particu-
larly the latter, which, by the intervention of Liparis,
becomes intimately connected with Lepidogaster and the
genus Gobiesor. The passage from these to the eel-
shaped genera, is obviously effected by Cuvier’s genus
Alabes: from this we have an uninterrupted series of
forms through the two typical families of the Mure-
nide and the Synbranchide: these latter, again, are
closely connected to the Sternarchide ; and thus we are
led to the last and most imperfect of all fishes, namely,
the Petromyzonide : between some of these latter and
Liparis, as well as with our new genus Rupisuga, there
is an obvious affinity; and thus, having returned once
more to the Cyclopteride, the outlines of the circle be-
come sufficiently marked to make us believe that the
whole form a natural group.

(190.) Nevertheless, from not having finished, at
present, the analysis of this order in its more minute de-
tails, we feel some lurking doubts as. to the precise situ-
ation of the family Petromyzonide, or, rather, of some of
its genera. The analogies, however, of the other groups
to those of the orders, appear sufficiently strong to lay
before the reader: they may be stated as follows :—

Analogies of the Apopat Order.

Families of , Sepee Orders of
the APODEs. Analogies. FISHES.
PEt Skeleton osseous: the most ? eae
1. Murenidz. highly orsaniseds 5 ACANTHOPTERYGES.

2. Gymnarchide. Sub-typical. MALACOPTERYGES.

Head broad; the snout de-
pressed and obtuse.

¢ > . s a.
Eyes minute; mouth aoe Prebeacuniae
circular.

5. Sternarchide. Tail excessively long. APODES.

3. Cyclopteride. i CARTILAGINES.

4. Petromyxonide.

* See the figure, and an interesting description of its habits, in Mr.
VYarrell’s Brit. Fishes.

ANALOGIES OF THE APODAL ORDER. ON

It is somewhat singular that the first, third, and fifth
of these analogies are more clear than the intervening
ones: and yet this very circumstance, instead of weak-
ening, rather tends to strengthen, the probability of the
whole being correct; for it cannot be doubted that M.
Cuvier is perfectly right in placing Gymnarchus imme-
diately after Murena: and as all writers, ancient and
modern, agree in the opinion that the lampreys and the
Myzxene, of all fishes, are those most allied to worms,
so they become the most aberrant of the order. Again,
the skeleton of Myzene is so slight, that it is not even
cartilaginous ; while that of Liparis, among the Cyclo-
pteride, is equally imperfect, being almost gelatinous. It
may be further remembered, that very small eyes is one
of the characteristics of the cheloniform type ; witness
the whole of the Balistide and the Chironectide, nu-
merous genera of the Siluride, &c.: now this limited
vision is carried to its highest imperfection in the lam-
preys; for some of the genera are actually blind, having
no eyes whatever. Finally, of all the apodal order,
length of tail, which is one of its primary character-
. istics, is especially developed in Sternarchus, so named,
because, although it is a long fish, the anus is close to
the sternum. Nevertheless, we wish it to be remem-
bered on this, and on all other occasions, that analogies
(however necessary to the confirmation of an arrange-
ment supposed to be natural) are of inferior importance
to affinities.

(191.) In regard to the situation of Petromyzon,
which we have removed from the Cartilagines, it is quite
clear that no one would have arranged it in the same
order with sharks and rays, but for the similarity of its
branchia. Now, if this part of its structure is really of
such importance as to decide its place in nature, it
may be fairly asked, Why is not this absolute rule acted
up to in the case of Mywxene, whose branchial apertures
are like those of the eels? These two genera cannot be
separated: and is not Mywene infinitely more allied, in
all parts of its structure, to Gymnarchus, than Petromy-

218 CLASSIFICATION OF FISHES.

zon is to the sharks? No one will dispute this. What,
therefore, is the logical deduction, but that the Petro-
myzonide should be naturally arranged with those fishes
with which, in their general organisation, they have the
nearest resemblance? The structure of the branchial
apertures, indeed, in this group, are so variable, that
they absolutely become hardly sufficient to characterise
a genus ; much less to determine an order. Among the
Gymnarchide, or sea eels, for instance, they are close to-
gether and under the throat in Sphagebranchus; united
almost into one in Monopterus ; single, and round, in
Synbranchus ; before the pectorals in Gymmnotus, and be-
hind them in Murena: in short, they almost assume
every possible form and situation within the limits of a
few genera,—a clear proof how completely secondary these
characters become in the present group. It would seem,
indeed, that Nature, upon leaving the annulose circle,
and entering that of the fish, intended to show us all the
forms of variation in the first group, which she after-
wards employs to characterise higher divisions: this
she has done in the class Acrita, as Mr. MacLeay has so
beautifully illustrated *; and the same remarks may be
made applicable to the group before us: the least or-
ganised of all the fishes, as the Aecrita are the least
among animals.

(192.) But there is another inference to be drawn
from the peculiar structure of the lamprey, sufficiently
important to merit a separate notice. In a former vo-
lume, we have stated the proposition that one of the
primary laws of the circular succession of all groups is,
that the three aberrant divisions constitute a circle of
their own, independent of their connection to the other
two; from which it follows, that the primary circles in
every group (when that group is perfect in all its parts)
are three, although they appear to be jive. We have
shown that this is more than probable in the union of
the monocardian animals ; and even Cuvier confesses
the affinity of certain of the cartilaginous order to rep-

* Hor. Entom. See also Classif. of Animals.

TYPICAL FAMILIES OF THE APODES, 219

tiles: and when we consider that some of the foreign
eels have actually no fins whatever, the enly essential
difference that remains between them and serpents is
their diverse modes of respiration. This theory, again,
receives strong support, if not direct confirmation, by
the structure of the branchia in Petromyzon, which is
precisely that of the cartilaginous fishes; it thus effects
the union of that order with the Apodes, so that the
three aberrant divisions of the whole class become united
into one circle: there is, in short, no other mode of ac-
counting for this singular departure of Petromyzon from
Myxene and Gastrobranchus,—two genera with which, in
all other respects, it is so naturally and confessedly allied.
Viewed in this light, the apparent anomaly becomes abso-
lute harmony ; since, were the branchia of the lampreys
like those of the Mywene, no passage whatever could be
traced between the Apodes and the Cartilagines.

(193.) Without entering, in this place, into sci-
entific details, or lengthened popular descriptions, we
may yet make a few general observations on the dif-
erent families under which, for the first time, we have
distributed the genera. The typical families, Mure-
nide and Synbranchide, comprise all the true eel-
shaped fishes, having serpent-like bodies, long and
cylindrical: they are either naked, or with scales so
minute as to be barely perceptible. In the first,
the branchial spiracle, or opening, is situated as in the
generality of fishes, that is, on the sides of the neck,
close to the pectoral fin (fig. 38. a); but, in the latter
family, they are always placed on the under part of
the throat, and thus are close together. Their general
aspect is so like that of reptiles, that they may be termed
serpent-eels, in contradistinction to the former, or true
eels. The habits of the Murenide, in general, are pretty
much the same as those of the common eel and the
conger. Some few are confined to fresh waters, but the
majority live near the mouths of rivers. The eels, pro-
perly so called, have pectoral fins ; but in the Murene,
or sea eels, no pectorals exist, as in Gymnothorax Zebra

220 CLASSIFICATION OF FISHES.

(fig. 40.). From the peculiar structure of their gills,

=, 9S

———__

<a

Seg

the eel is capable of living out of the water a consider- ©

able time—some say many days; and when the waters of
one of their haunts are dried up, they are well known to
have the power of making their way by land, during the
night, to other localities where water exists. They feed on
almost any thing— subsisting both upon living and dead
animals, and also on aquatic plants. The genus Alabes
is particularly remarkable from having a small concave
disk between their pectoral fins ; so that the connection
between the eels and the suckers (Cyclopteride) is ren-
dered complete.

(194.) The circle of the Murenide contains by far
the greater number of the apodal fishes ; and presents
us, under the general form of the eel, with a great di-
versity of modifications. In the most perfect or typical
division, the sub-family Anguilline, or true eels, the pec-
toral fins are always present, as in the conger (fig.41. ):

the head is depressed, the muzzle rather obtuse, and the
teeth small: the branchial spiracle is in the form of a
slit(a), placed just before the base of the pectoral fin,
but rather below it ; and the nostrils are always tubular ;
the dorsal, caudal, and anal fins are united into one:
but in the new genus Aviosoma Sw. the nostrils are
not tubular, and the branchial aperture is in front of the

THE MURZNIDA, OR EELS. Page|

pectoral. Several of these fishes inhabit the Sicilian
shores ; and they are richly coloured with silver reflec-
tions, very different from the lurid hues of the true eels.
Another extraordinary form is seen in our Leptognathus
oxyrhynchus (fig. 42.), which immediately reminds us

42 SS WERE WHER.

a GEN \ ‘i \ EN Yn \ \
ese) ll NS vt

ay : ee \
et hyp yyy | js
Z Wy y
Li
patel ss shslss bee at™“~—ywyy,
a

of Laurida, Sphyrena, and all such analogous types ; it
has the pectoral fins of Anguil/a, but with the naked tail
of Ophisurus Lacep.: this, also, seems one of the rare Si-
cilian fishes unknown to modern writers. Several others
will be found designated by professor Rafinesque. The
next sub-family of Murenine contains those eels which
are altogether destitute of a pectoral fin; but the two
branchial spiracles are placed, one on each side, in the
same situation as the last. The name of this division is
taken from those species which were so highly esteemed
by the Romans; and therefore M. Cuvier has very pro-
perly rejected those of more recent nomenclators. *
Among these there are even more variations than in the
Anguilline, not merely in the structure of the teeth, but
in the form and relative proportion of their fins. The
most typical have the
dorsal and analunited:
but in the singular
genus Dalophis Raf.
(fig. 43.), with which
M. Cuvier says he is
not acquainted, the
end of the tail is com-
pletely naked (a), as
in our Leptognathus ;
while in WNettastoma Raf., which was likewise un-

* Gymnothoraxz Bloch; Murenopsis Lac.; Anguilla Raf.

Le ocd CLASSIFICATION OF FISHES.

known to him, the branchial apertures are even more
beneath the neck than in Dalophis, although the fins are
like those of Murena. These instances, taken almost
at random, will be sufficient to show how very little is
yet known of the European genera and sub-genera or
this order, and may stimulate foreign ichthyologists to
explore more distant seas, where numerous others will,
no doubt, be discovered.

(195.) The Synbranchide, or serpent eels, are all
marine. and numerous species are scattered in all the
temperate and tropical latitudes: excepting the Petro-
myzonid@: they are the least organised, as fish, of any
in the order, for some of the fins, exclusive of the ventral,
are often wanting ; and in Cecilia Lac. all these organs
of motion totally disappear. The species contained in
this division, hitherto determined, are few ; but we be-
lieve very many have been overlooked, more especially
in the Régne Animal, where we find no notice taken of
these discovered on the Sicilian coast by professor Ra-
finesque, who has characterised several excellent sub-
genera, to be found in our synopsis.

(196.) The Sternarchide are so named from the
body being so excessively short that the vent is close to
the sternum. Nevertheless they are very long eel-shaped
fishes, although the body is more or less compressed ;
sometimes (as in Carapus) covered with visible scales,
and having altogecher something of the appearance of
ordinary fishes. In all these the spiracles are lateral, and
they all exhibit a tendency to blend into the acanthopte-
rygious, or spine-rayed order. The famous Gymnotus
electricus, or electric eel cf South America, seems to
belong to this division, which (if its analogies are ex-
amined) will be found to represent the torpedos in the
circle of the Raide, and the electric silures in that of
the Siluride. All the species hitherto known—and they
are but very few— occur out of the European range, ex-
cept, indeed, that curicus genus Leptocephalus, which
clearly represents, in this family, that of Amphiowus in
the next.

THE LAMPREYS. 293

(197.) The Petromyzonide, or lampreys, for the
reasons already assigned, we have placed as the most
aberrant family in the order. Their very low state of
organisation renders them the most imperfect of all
vertebrated animals, or, at least, of the whole class of
fishes. Their skeleton is so soft as not even to be car-
tilaginous: the vertebre are indistinct, and are per-
forated by a central tendinous cord, filled with a muci-
laginous substance ; the vertebral column thus becomes
a series of rings, and is hardly more solid than the mu-
cilage within. The gills, instead of being pectinated,
as Cuvier remarks, more resemble pouches, resuiting
from the union of one of the faces of one gill with the
opposite face of the neighbouring gill. In some there
are seven branchial spiracles on each side, but in others
only one. Their external form, however, is quite suf-
ficient to make them immediately known. The body
is eel-shaped, devoid of ventral and pectoral fins, er, in-
deed, of any true fin; for that elongation of the skin
which forms the dorsal and unites to the ventral is devoid
of any rays: the mouth is circular, placed on the lower
part of the head, and forms a maxillary ring. Such
as have the mouth armed with rows of strong teeth and
tubercles, like the true lampreys, are able, by this ap-
paratus, to adhere to stones and other substances with
astonishing tenacity ; by the same means they are said
to attack the largest fishes, which they pierce and devour
by their rasp-like teeth. Of this very remarkable family,
three principal types, or genera, are only known: the
first is Petromyzon, or the true lampreys, having several
lateral spiracles; the second is Myxene *, where the
spiracles are only two; the third is represented by the
Amphioxus of Mr. Yarrell: the whole may be charac-
terised by having the skeleton almost mucilaginous ; the

* Subsequently named Gastrobranchus by Bloch, who has admirably
illustrated its internal structure. This group, as seen in the Régne Animai,
is another instance of the Linnean genus Myzene being nominally retained,
but yirtualiv abolished: Cuvier divides Myzene into three genera, but does
not retain the original name to any one. The nomenclature and the
arrangement appear to us equally objectionable.

994 CLASSIFICATION OF FISHES.

mouth placed beneath, presenting, when closed, the ap-
pearance of a fissure placed longitudinally, and not, as
in all other fishes, transversely with the body: the eyes
are very small: the rays of the fins are obsolete, or
nearly wanting ; and there are no ventrals or pectorals.
~ Between Petromyzon on one side of Amphioxus, and
Myzxene on the other, there are still wanting two types
to complete the circle of this family : one of these types
would also fill up the only link wanting between Amphi-
oxus and Leptocephalus; while the other would connect
Amphioxus with Liparis and the lump-suckers.

(198.) The Cyclopteride, or lump-suckers, form the
concluding family of this order. Like the last, their
skeleton is so soft, that some of these fishes are said to
dissolve after death into nothing but jelly, or mucilage:
like the lampreys, also, they are adherent, or suckers ;
but this faculty, instead of lying in their mouth, is
transferred to the pectoral and ventral fins, both of
which, in being united into a circular disk ( fig. 44.),

: form two power-
ful suckers, by
which these ani-
mals adhere to
rocks, stones, or
other substances, and even to the hand of those who
capture them. They are smooth, destitute of scales,
and of an ugly appearance. Sometimes the disk, as
in the genera Liparis and Cyclopterus, is only single ;
but in Lepadogaster* and Rupisuga, it is double.
Like all the fissirostral types, or their represent.
atives, the head of these fishes is uncommonly large
and greatly depressed, although the body is compressed :
the snout is rather lengthened and obtuse; so that, in
short, we are presented with such a miniature resem-

* The genus Piecephalus of Rafinesque appears to differ from this, in
having the ventral or abdominal fins forming a semicircular plate, whose
concavity is turned towards the head, and furnished with scattered. cup-
shaped suckers (sparse di cupule succhéianti) ; ; there is no operculum, but
a three-rayed membrane; and the tail is heart-shaped, or emarginate. —
Raf.Carati. p. 69.

ON THE ORDER MALACOPTERYGES. 995

plance to that of the generality of sharks, that the out-
line of the head of one would almost serve for that of

the other. A new species from Sicily, — the specimen
of which, sent to the British Museum, but now lost, —
is here figured from our original drawing (fig. 45.).

(199.) The union of the sucking fish with the eels,
as before observed, is effected by Cuvier’s genus Alabis,
and Rafinesque’s Piescephalus, This latter is placed by
Rafinesque with the eels, with the observation, that it
has several characters of conformation with Lepadogas-~
ter; and there is good reason for the remark, for both
have the power of adhering, by means of concave disks
on the throat. Although we have never seen this ex-
traordinary fish, we feel perfectly sure that future inves-
tigations in the Mediterranean will bring it again to
light. But whether we take this for the connecting link,
or Cuyier’s genus Alabis, we find the suckers brought
into immediate union with the eels, —thus uniting all
the apodal families into one complete circle; and so
perfectly is this effected, that we may at once dismiss
the subject, and pass onward to the next order.

CHAP. IX.

GENERAL ACCOUNT OF THE MALACOPTERYGES, OR SOFT-FINNED
ORDER 5 AND OF THE ANALOGIES OF THE FAMILIES.

(200.) Tue great order of fishes characterised by

Artedi as the Malacopteryges, is composed, as already

observed, of those whose fins are supported by soft or
VOL. I. i)

226 CLASSIFICATION OF FISHES.

articulated rays. The branchial opening is always wide*,
with the gills perfect ; and although in some there are
no teeth, the jaws are never prolonged into sharp plates, —
as in the chelonian fishes. This is the chief typical cha-
racter, and the exceptions are very few. In some, as in
the Siluride, the first rays of the dorsal and pectoral fins
are represented by bony spines, the sides of which are
crenated, or toothed, like a saw. In the flat fish (Pleu-
ronectide) the rays are semi-spinous ; and even among
the most typical families, the first two or three dorsal
rays are rigid: yet all these deviations take not from
the fact, that the whole of these fishes are known by
the absence of spiny rays, placed after the first or second
in any of their fins. Thus characterised, we may at
once take a general view of the primary divisions under
which we shall now arrange them.

(201.) The soft-rayed fishes, although composing a
circle of equal rank to that of the spinous rayed, are
yet so inferior in point of extent, that they do not, in
all probability, amount to more than one fourth of the
number comprised in the great typical circle of the
Acanthopteryges, or spine-rayed fishes: they are inferior
to them, also, in the elegance of their shapes and
colours; but, on the whole, are superior in point of
utility to man, since they comprise by far the largest
proportion of such as furnish him with food. When we
enumerate the salmon, cod, turbot, herring, and carp
tribes, as belonging to this order, we absolutely name
nearly all those which not only supply food to the great
bulk of mankind, but in whose capture thousands of
men and fleets of vessels are exclusively engaged: the
greatest part of these are, of course, marine; but it is
also a natural character of this order, that it likewise
contains nearly the whole of those families which live
exclusively in fresh water.

(202.) The primary divisions appear to be as fol-
lows: the first, and most typical, are the Salmonide,

* Except in that group which leads to the cartilaginous order.

PRIMARY DIVISIONS OF THE MALACOPTERYGES. 227

or salmons; while the second, or sub-typical, seem to
be the Pleuronectide, or fiat fish ; then follow the three
aberrant groups, represented by the Gadide, or cods, the
Siluride, or cat- fish, and the Cobitide, or loaches. That all
these groups are united into one great circle, is evident :
for although, in tracing the series, we shall find an hiatus,
which nothing yet known can fill up, there is yet such
circumstantial evidence proving the series of those forms
which we already know to be natural, that we can only
look upon the inequality of the links as arising from
one or other of those causes elsewhere explained. We
here allude to the interval between the Salmonide and
the Pleuronectide, or flat fish. All ichthyologists agree
in considering these latter to be the most isolated group
among fishes, just as the Psittacide, or parrots, are
among birds; and for the same cause, namely, that there
are no forms among them so aberrant as to mark beyond
doubt the character of the group by which they are pre-
ceded, and that, again, by which they are followed. In
deciding, therefore, the probable station which such an
apparently isolated group would hold, we must have re-
course to inductive reasoning. First, then, there can be no
doubt on the acknowledged fact, that the Pleuronectide
belong to this order, not merely because all writers have
so placed them, but because they would interrupt the se-
ries of the other circles; and, further, because they have
some affinity to the Gadid@, near to which M. Cuvier,
following all his predecessors, has placed them; both
having the anatomical character of the ventral fins being
attached to the pectorals, and the pelvis immediately
suspended to the bones of the shoulder. This affinity,
therefore, being established, we have only to follow the
thread of progression from Gadus to the next and to
the next family, until, having gone as far as we can,
and successively established our groups as we proceeded,
there is no other conclusion to be made than this, — that
where the line of affinity becomes lost, is precisely where
those forms which should lead us back again from our
starting post is wanting. Now, this is a precisely ana-
Q 2

298 CLASSIFICATION OF FISHES.

logous case to that of the Psittacide among the scan-
sorial birds, where we have the series as much interrupted
on one side, but not quite so imperfect on the other ;

-and yet no ornithologist would think of placing the ©

parrots in any other situation than between the wood-
peckers and the toucans. But let us look to this ques-
tion in another point of view. Let us suppose that-al}
the aberrant types* of the rays (Raid@) were unknown
or destroyed, and that the only representations of them
now in existence were the skates or rays: looking,

then, to these only, and to the sharks, how slight—how _

questionable — would be their affinity! One would ima-
gine that, if they were really related, whole families
of intermediate forms would be necessary to connect
them ; and yet how completely has nature effected this
by such forms as Rhinobates and Pristis,—two little
groups which blend the form of the rays and the sharks
so completely, that ichthyologists are even undetermined
where one ends and the other begins. And so, may we
fairly presume, is the case with the Pleuronectide and
the Salmonide. It will be subsequently explained on
what grounds we believe these two are the typical groups
of the present order ; and two or three genera would be
quite sufficient to unite them as perfectly as are the
Raide and the Squalide.

(203.) With the above exception, therefore, we shall
find the circle of the malacopterygious fishes sufficiently
perfect. From the Salmonide and the Pleuronectide we
pass on to the Gadide. This latter affinity has long been
admitted ; and the connection, not at all remote, is
further established by the holibut,—a well known and
gigantic flat fish, beginning to assume the thick and
lengthened body of the cod and haddock. We quit the
Gadide by means of Brotula and Oligopus, — genera
which blend their own group in the most perfect manner
with that of the Siluride. It is here that we have the
most aberrant forms of the order: the first rays of the

* Torpedo, Squatina,and Rhinobates.

SALMONIDZ, OR SALMONS. 229

dorsals and pectorals are not only spines, but become
bony, lance-headed plates; and the other rays are gene-
rally so thick and strong as to possess little flexibility.
From these the passage to the loaches is remarkably
gradual ; and as we now arrive at that part of the circle
which touches the cartilaginous order, we accordingly
find that these fishes show a decided relation to that
group by their viviparous nature. All writers place the
loaches in immediate conjunction with the Cyprine, or
carps, which belong to the Salmonide: and as these
latter form a circular group of themselves, we return
from whence we began; and thus we find all the divi-
sions, on one side of the Pleuronectide, sufficiently well
united, although capable of containing othef and more
intimate links of connection. Before proceeding further,
we shall briefly notice the distinguishing characters of
the families whose affinities we have just endeavoured to
trace, and then see how far their analogies are con-
formable to the theory of representation.

(204.) The Salmonide not only include the salmons,
but nearly all the freshwater fish of Europe, and a
great proportion of those peculiar to the lakes and
rivers of other countries. They are distinguished, as a
whole, by having all the rays of their fins soft ;
or, if any exceptions exist to this character, it is
merely found in some of the perches, where the first
and second dorsal ray is simple and rigid.. They have
the body protected by large and well-formed scales, by
which they are distinguished from the cat-fish (Silu-
ride), and have not that fieshy covering to the fins so
general among the Gadide: they differ from the
loaches in their being oviparous, and in certain other
anatomical characters hereafter to be noticed. This
we look upon as the most typical division of the whole
order; and it preserves this character in being most
numerous in species, and most diversified in its forms.
it contains the different groups known under the fami-
liar names of carps, trouts, salmons, pikes, and her-
rings.

Qa 3

230 CLASSIFICATION OF FISHES.

(205.) The second, or sub-typical group, are the
Pleuronectide, or flat fish. These are at once recog-
nised by their remarkably flattened bodies, of an oblong —
or rhomboid shape: the circumference is almost com-
pletely margined by the dorsal and anal fins, the rays of
which are, for the most part, spinous: the eyes are
placed on the same side of the body; and this side,
which is the upper surface, is coloured so as to resemble
the ground upon which these fishes lie in wait, among
mud, sand, and weeds, to seize ther prey: the under
surface, from never being exposed to the action of the
light, is always colourless, and generally of a fleshy
white: the scales are small and well-defined: and the
whole of the species are marine. They are, however,
very few in number when compared with the Sal-
monide, and present but few variations in their general
structure.

(206.) The Gadide, or cod-fish, also form but a
small family, although with several well-defined vari-
ations of structure. They have the usual shape of
ordinary fishes, and are entirely marine. The small
scales of the body are covered, and often nearly
- concealed, by a mucous skin, which also extends over
the fins, and gives them a thickened and fleshy cha-
racter, not found in the Salmonide. The head and
body is generally but slightly compressed, the eyes
large, and the mouth very wide. The ventral fin in
this group, is very small, and generally terminates in a
pointed fleshy filament, more or less lengthened; and
three out of the five rays usually found in this fin are
sometimes wanting : this character, hitherto overlooked,
we shall subsequently find to be of much importance. The
Gadide present considerable variation in the apparent
number of their dorsal fins ; in some there are three, in
others two, and in a few only one; but the interval
of their divisions are so small, that, perhaps, it would
be more correct to consider them as having one long
dorsal, variously cleft: the snout is often furnished
with small cirri, or round worm-shaped filaments.

PRIMARY DIVISIONS OF THE MALACOPTERYGES. 23]

The cod, haddock, and whiting, are familiar examples ;
and nearly all are confined to the seas of cold or tem-
perate latitudes.

(207.) The Siluride, or cat-fish, in many respects,
have a strong resemblance to some of the last: like
them, they are furnished with cirri on the head, but
often prolonged to an enormous length: the body is
generally soft and mucous, but the head is hard and
bony ; and although there are no true scales, the head
of very many, and the whole body of the typical species,
are covered with hard bony plates, which either serve the
office of a helmet or a complete coat of mail. The
species are very numerous in the great rivers of hot
climates, more especially in those of India ; and they
swarm in the Ganges: one only has been found in the
European range ; so that we may look on it as a tropical
family. The head is greatly depressed, so as to ex-
hibit, when viewed in front, some slight resemblance to
that of a cat, from whence the vernacular name of cat-
fish.

(208.) The Cobitide, or loaches, form a small family
of freshwater fish, well distinguished from the Siluride
by their elongated and somewhat rounded body, the
compression of the head, and the possession, in general,
of true scales: they differ from all the other soft-finned
fishes, by being viviparous. The primary type, how-
ever, of this family, appears to be Anableps: their
mouth is small, and furnished with cirri; and the
aperture of their gills, like that of the eels, is merely a
lateral slit behind the pectoral fin, confined by a skin at
both extremities: the generative organs of the male
have a close analogy to certain of the cartilaginous fishes.
The passage from this family to that with which we
began, is effected by certain genera, as Pecilia, Lebias,
&c., which have the above characters united with
the oval body of the carp (Cypring); and, indeed,
these two families are so connected by their aberrant
types, that all writers have arranged them close to-
gether.

2 4

232 * CLASSIFICATION GF FISHES.

(209.) Let us now arrange this series, in the first
place, in juxtaposition with that of the orders of the
whole class, and we shall then find that all the parti-
culars we have stated have a double reference, —one to
their actual characters, and another to their analogical
relations.

Analogies of the Mauacopteryces and the Orders of

FisHes.

Families of the pape Orders of
MALACOPTERYGES. Analogies. FISHES.
Salmonide. Rays of the fins soft. MALACOPTERYGES.
Pleuroneciide. Rays of the fins rigid. ACANTHOPTERYGES.

: Ventral fins small or none;
Gadide. scales imbedded in the skin. +} APODES.
Siluride. Body mailed. PLECTOGNATHES.
Viviparous ; mouth beneath the
ae muzzle, which is broad and re,
Cobitide. depressed ; pectoral fins very CaRTILAGINES.. |
large.

We have already shown that the Salmonide are those
fish which have the fin-rays soft ; and as they are the
most highly organised of the Malacopteryges, they con-
sequently represent the perfection of their own order.
The Pleuronectide, in a manner no less singular than
beautiful, thus turn out to be prototypes of the Acanthop-
teryges ; for Cuvier follows his predecessors in placing
them in this order, to which, notwithstanding their
spined rays, they undoubtedly belong. The relation of
the Gadide to the apodal or anguilliform order, when
attentively considered, will be found no less undeniable.
The Apodes, besides their eel or serpent formed body,
are mainly distinguished by the total absence of the ven-
tral fins: their body is slimy: the scales, which are very
small, appear to be imbedded in the flesh, or covered
by a fat skin, which extends also over the fins. Now,
the Gadide have as many of these characters as it is pos-
sible for fishes to have, whose situation is in this order:
of all the Malacopteryges, they have the most imperfect
anal fins. In the forked hakes, or Phycis, it is reduced to

ANALOGIES OF THE MALACOPTERYGES. 233

a single ray; and even in the more typical forms, as Ga-
dus and Mustelus, the three hinder rays are often so short
as to appear obsolete: their slimy body, fleshy fins, and
minute scales, are all so many characters possessed by the
eels; while this resemblance is carried so far in the rock-
lings (Motella), that the forms of both are nearly alike ;
both having the body very long, and the anal, dorsal, and
caudal fins nearly, if not quite, united. The Siluride
are no less strikingly analogous to the Plectognathes, or
cheloniform fishes. In both there are no true scales, but
in both are the typical groups incased in a coat of mailed
plates; so that Loricaria isas perfect a prototype of
Ostracion, as the half-mailed Pimelodes are of Balistes.
Finally, we come to those soft-rayed fish, whose mode
of generation separates them from all the others of their
own order, and likens them to the cartilaginous or chon-
dropterygious fishes: these are the Cobitide, or loaches —
one of the most remarkable groups of fish in the whole
order. Whether we consider the peculiarity of their
external or internal anatomy, we can only feel astonish-
ment that neither one nor the other should have given
them a more prominent station in our modern systems
than they have hitherto held. To place viviparous and
Oviparous fish merely as genera following each other,
appears just as natural and consistent as if we arranged
the flat fish and the skates as cognate families, merely
because both are flat, and have the fins surrounding their
body.

(210.) If the validity of the foregoing comparisons
are admitted,—and they appear to us as true to nature as
any of those already brought forward among the more
perfect vertebrated classes, —it follows, as a necessary
consequence, that the families of the malacopterygious
fishes follow each other in the same order, also, as do
those of the Cartilagines. We have just glanced inci-
dentally at the similitude between the flat fish and the
rays ; let us see, therefore, if this is merely fanciful, or
founded in nature: a comparison of the two groups
will determine this question.

1
| 4
ae

¥

934 CLASSIFICATION OF FISHES.
Orders of Families of = Families of the
FIsHEs. MALOCOPTERYGES. Analogies. CARTILAGINES.

Body lengthened ;
> < : dorsal fins fully€ 5
MALACOPTERYGES. Salmonide. developed,one or Squalide.
two.
ACANTHOPTERYGES. Pleuronectide. ay ele Raide.
; flat, rhomboid. wanes
Tail long, neariy :
5 surrounded bye Chimeride.
the ventral fin.
Body mailed, teeth
small ; mouth ¢ Sturtonide.
with cirri.

AFODES. Gadide.

PLEGTOGNATHES. Sduride.

{ Mouth beneath the
a) | muzzle ; pata |
CARTILAGINES. Cobitide. ~ ing of the carti- $} Prionide.
| laginous struc- |
ture.

We have already compared the primary orders of
fishes with those of the great classes of the animal
kingdom, and also with the families of the Cartilagines ;
nothing further, therefore, need be said on this subject,
than to remind the reader, that these analogies give
him a clue by which he can trace the most remote rami-
fications of these relations in the classes of birds and
quadrupeds. In the present instance, nevertheless, we
have introduced the orders of fishes in a separate column,
to show more perfectly the wonderful harmony and wnity
of design—far greater than the wit of man could de-
vise — which pervades these otherwise singularly varied
groups. Here, in fact, we see that, by simply placing
the soft-finned fishes in juxtaposition to the cartila-
ginous order, we have the Pleuronectide standing op-
posite to the Raide, as their bond fide representatives.
The Gadide represent the slender-tailed Chimeride,—
the latter being the on/y cartilaginous fish, yet discovered,
having an eel-shaped tail— that is, gradually attenuated
from the belly to a point, and bordered beneath by a
long ventral. The mailed Siluride so completely repre-
sent the mailed sturgeons, that the Loricaria rostrata
of Spix might easily be taken, at the first glance, for an
Acipenser ; and what is still more singular, the sturgeons,
and the genera which represent them*, are the only

* Sguatina, Crossarchus, &C.

ANALOGIES OF THE MALACOPTERYGES. 935

family of the Cartilagines that have cirri or barbels to
their mouth; these appendages being more developed
among the Si/uride@ than in any other fish in existence.
The analogy of the Cobitide to the cartilaginous fishes
has already been stated ; and although their direct rela-
tion to Prionodon is far less apparent than any of those
just noticed, still, if these are correct, it follows that
either this or some other extraordinary type will effect
the union. The Prionide, in fact, at present contain
only two fishes ; and it is yet undetermined whether
these, like all the other Cartilagines, are viviparous: the
probability is, that they really are so; and therefore even
one such character is quite sufficient to show some re-
lation to the Cobitide. This point, however, must yet
be considered as debateable, until the real station of
Polypterus is more clearly determined than it now is.
(211.) Lastly, it seems further desirable to test all
these relations, by tracing the analogies of the Mala-
copteryges with the Plectognathes, or cheloniform fishes :
the two circles, if placed in a linear series, will stand

thus :—

Analogies of the Sort-FINNED,and CHELONIFORM FiIsuHEs.

Order of Analogies. Order of the

Malacopteryges. Plectognathes.
SALMONIDZ. The fins with soft rays. BALIsTIDE
PLEURONECTIDE. Dorsal fins with spines. CHIRONECTIDE.
GADIDZ. Head broad; mouth very wide. lLopuip#.
SILURIDE. Body covered with mailed scales. SYNGNATHIDZ.(?).
CoBITIDz. ——-—— Unknown. ©

We have already adverted to the little diversity of
form — in comparison to what we see in the two typical
orders —that exists among the tortoise-formed group ;
and this fact at once accounts for their analogies being
less obvious than many others. Certain, however, it
is, that the Salmonide@ are as typical of one order as
the Balistide are of the other, or the Plectognathes :
the ugly aspect and unsymmetrical head of the Pleu-
ronectide, again, find their representatives in the still
more hideous Chironectid@; and both have fins with

236 CLASSIFICATION OF FISHES.

simple rays, and of little flexibility. Raniceps, among
the Gadide, is a stilljmore striking prototype of the
common Lophius piscatorius, or fishing frog; an analogy
so obvious as to be conveyed by their respective names.
The mailed Stluride, forming the typical Loricarine,
find their representatives among the cheloniform fishes,
in the singular families of Syngnathus and Pegassus.
Nor does their analogy rest alone on their external ana-
tomy. Cuvier, as we have already shown, has separated
the Sygnathide from all other fish, on account of their
branchia assuming the form of tufts ; and yet, although
he is perfectly aware that the very same deviation from
the ordinary branchia of fishes is found in another
group, he merely considers the latter as only deserving
of a simple generic distinction. The genus we allude
to is that of Heterobranchus of Geoffroy Saint- Hilaire,
which, in addition to the ordinary branchia, have
others attached to them, resembling tufts, considerably
ramified ; so that these fishes combine the branchia of
two distinct classes, namely, that of Pisces and of Am-
phibia. Among fish, their only prototypes, in this
respect, are the Syngnathide, which, as we have already
shown, are also the representatives of the amphibious
reptiles. If primary divisions are to be made on such
apparently anomalous characters, Heterobranchus has
precisely the same claim for such a distinction as Syng-
nathus : but the fact seems to be, that this structure, far
from being anomalous, is what we should expect in groups
that stand at the greatest distance from their respective
types; and from its thus occurring both in the Syngna-
thide and the Siluride, we discover that it is in perfect
harmony with the usual course of natural variation, — a
variation which frequently makes one group represent
another in the most unexpected and singular manner.
The cartilaginous type of the cheloniform fishes, if not
Polypterus, appears to be undiscovered ; and this may
account for there being no group in that order which
represents the Cobitide, which form the cartilaginous
type in the circle of the Malacopteryges.

THE SALMONIDZ. — CHIEF DIVISIONS. oy

(212.) We shall now take a more particular view of
the several families composing this order ; always noticing
the most remarkable or prominent forms in each, and
occasionally pausing to trace the analogies of such as
appear more particularly interesting, or as necessary to
show the reasons of our arrangement being so different
from those of our predecessors. The families will be
reviewed in the same order as we have already noticed
them, viz. — 1. the Salmonide ; 2. the Pleuronectide ;
3. the Gadide; 4. the Siluride; and, 5. the Cobitide.

(213.) The Satmonip&, or salmons, appear to resolve
themselves into five principal groups or sub-families,
all of which are represented by the Linnean genera
Cyprinus, Salmo, Clupea, Esox, and Mormyrus. The
few characters common to them all have been already
intimated : where so much diversity of structure exists,
a corresponding difference of habits will be found ; and
these had better be noticed under the separate divisions
of the family.

(214.) The Cyprine, or carps, form a most exten-
sive division of fish, entirely confined to fresh waters.
Their numbers are much more abundant in the old
world than in the new, and many species inhabit the
rivers and lakes of temperate Europe. The carp (Cy-
prinus Carpio Linn., fig. 46.), perch, roach, and several

other native fishes, are familiar examples of the genera
construction of the whole. They are the most herbi-
vorous of all fish—feeding chiefly upon aquatic vege-
tables, like their prototypes the eels ; to which, although

238 CLASSIFICATION OF FISHES,

they have not the same shape, they appear really
analogous: like them, also, they have thick fleshy fins,
and a slimy mucous substance spread over their body :
they further resemble the eels in having few or no real
teeth, and both feed on the same substances. The
mouth of the Cyprine is always very small, and the
jaws destitute of teeth* ; but they have strong powers
of mastication, from the inferior pharyngeal bones being
provided with a few large teeth, adapted for pressing
their vegetable food: the stomach is simple, and with-
out ceca. In external characters, they differ from the
salmons, by having a single dorsal fin ; the majority,
also, have very thick fleshy lips, sometimes furnished with
barbelst: the scales are generally large, the body ovate,
the head thick and obtuse, and the ventral fin consider-
ably behind the pectoral ; it is generally, indeed, placed
intermediate between the pectoral and anal. The charac-
ters of the two typical genera we have not yet clearly
determined; but we suspect that the true Cyprine are
almost peculiar to the Old World, and that Catastomus,
‘ with. its sub-genera, represent them in America. Cy-
prinus, even as thus restricted, constitutes a very large
group, which, notwithstanding the minute divisions that
have been made among the European species, requires a
complete revision. If the eighty-three species, which
Dr. Hamilton alone has discovered in India, were to be
divided on the same plan as has been done with those
of Europe, those alone would amount to twenty or
thirty sub-genera. Some of those proposed by Cuvier
may be adopted, at least for the present; but we must
confess our belief that his arrangement of this family
is any thing but natural. The genus Cyprinus is dis-
tinguished from that of Catastomus, by not having the
lips nearly so thick, or the under one hanging down and
wrinkled in numerous folds ; Catastomus, also, is entirely

* Except, of course, in the fissirostral type, or Erythrinus of Gronovius.
These, in numerous instances, are so small as to escape detection ; and
from their being present or absent in species which have the greatest afii-
nity to each other, cannot be made use of as exclusive characters, even for ,
sub-genera. This is aiso the opinion of Dr. Hamilton (Gang. Fishes, p. 257.).

CYPRINZE, OR CARPS. 239

destitute of barbels, and the snout always projects beyond
the mouth. Among the sub-genera of Cyprinus, which
may perhaps be retained, is Barbus, represented by the
common barbel of our own rivers ; these processes, or
cylindrical cirri, being placed on the sides of the mouth :
with these we associate the gudgeons and the tench, as
aberrant forms ; but nothing definite can be determined
on this point in the present state of their arrangement.
The beautiful and masterly figures of Le Sueur of ten
species of Catastomus may be consulted with advantage,
both by ichthyologists and artists. The genus Leuciscus
of Klein seems also to be one of the primary divisions of
the carps: it is very numerous; but only one of its
sub-genera has, as yet, been incorporated in our sys-
tems. Dr. Hamilton’s genus Chela belongs to it, and
others will be found in our Synopsis. The bleak, roach,
and smelt, are all native examples; and they are at once
distinguished from the two preceding genera, by their
lips being destitute of barbels; and their thin fins,
which are without the anterior spinal ray. Those that
are found in India have such a strong resemblance to
the herrings, as observed by Dr. Hamilton, that they
are either related to them by affinity or by strong
analogy. We believe, however, that the relation is
analogical. As we proceed to the more aberrant carps,
we find the genus Erythrinus of Gronovius (E. teniatus
Spix, fig. 47.), uniting the Cyprine with the Salmoni-

WYSE
y Y

de, by their strong teeth and large mouth; while Gono-
rynchus of the same author somewhat resembles Ery-
thrinus, with the small toothless mouth of a carp. Lastly,
the most aberrant type may possibly be that of Sudis,
a small genus of freshwater fishes found in the rivers of

240 CLASSIFICATION OF FISHES.

a

tropical latitudes. They have, says Cuvier, “all the
characters of Erythrinus,” except that their dorsal and
anal fins, both long and narrow, are placed close to the
tail: he arranges them with the Clupeine ; but they
seem to us more naturally related to Cyprinus and
Salmo. The last genus, Sudis, is remarkable for its
lengthened cylindrical body, covered with large osseous
scales, indicating its analogy to the cheloniform fishes,
and to the flat-headed Siluride : its mouth opens some-
what vertically, as in Chironectes and Uranoscopus ; and
on these grounds we suspect it is the type of all these
fishes in the present sub-family: only three species have
yet been found.

(215.) The Salmonine, or salmons, have many close
relations to the carps; nevertheless their larger mouth,
their thin lips, and well-defined teeth, show that their
food and economy are very different. In this division are
arranged all such fishes of this family as have a small
adipose dorsal fin, placed half-way between the first
dorsal and the caudal. We do not feel confident, how-
ever, for reasons hereafter stated, that this is a truly
natural arrangement; although, in the present state of
ichthyology, it is a.very convenient one for discrimi-
nating the genera. The different groups of salmon are
distributed more abundantly in the Old than in the New
World: the history of those that are common to our
own country has been repeated so often as not to require
notice in this place. The salmon, salmon trout, and
chars are well known for their rich and delicate flavour:
the first, more especially, produces a source of much
wealth to those concerned in its fishery.

(216.) It is evident, that if Erythrinus is to be
placed among the carps, on account of its single dorsal
fin, it must be an aberrant type; and whether we ulti-
mately assign it to that circle, or to the one now before
us, it forms a link of connection between the two. There
are two or three very remarkable sub-genera of fiuvia-
tile fish in the rivers of Tropical America, which seem to
have as much of the aspect of perch as of salmon: they

CHIEF FORMS AMONG THE SALMON. Q41

have a very small mouth, and thick lips; but some have
strong teeth ; and all have a small adipose fin. These
appear to form one of the primary divisions of the Sa/-
monide, and to be further distinguished by their geogra-
phic distribution, which is confined to the New World;
while the genuine salmon, and its sub-genera, are more
particularly appropriated to Europe, and the more tem-
perate latitudes of Asia and America. Among these
first we may glance at Serrasalmo, remarkable for a
thick, oval, and high body, and very large triangular cut-
ting teeth ; the belly is compressed, and serrated like the
edge of a saw, from whence their name. These fish,
like Hrythrinus, may be called the sharks of their own
family ; and they are stated to be so fierce, that they
will destroy water-fowl, and even attack the Indians
when bathing in the rivers, although, from their com-
paratively small size, they can only inflict wounds in
the flesh. The other sub-genera belonging to this group
are Myletes (Cuv)., Tetragonopterus (Artedi), which
have the belly serrated ; and Characinus (Artedi*), Le-
porinus, Anodus, and some other aberrant forms or
species, which have small mouths and more delicate
teeth: one or two, indeed, have no teeth whatever in
their jaws ; while, in the very small size of the mouth,
and the thick lips, they have all the aspect of carps, but
with a small adipose fin. In the European division,
we have the common salmon, the trout, and the char:
in these, the body is more Jengthened, the mouth mo-—
derately wide, the lips thin and bony, and the teeth
small, edging the margin of the jaws: many sub-genera
have been proposed, among which is Osmerus of Artedi,
which includes the small salmons called smelts. Core-
gonus, of the same excellent ichthyologist, seems to
represent, in Kurope, the salmons of Tropical America;
for their mouth is very small, and the teeth are fre-
quently wanting: they are chiefly found in the conti-
nental lakes. The Argentina of Linneus is a small

* Subsequently called Curimata by Cuvier.
VOL. I. R

ZAQ CLASSIFICATION OF FISHES.

semi-opaque fish, long celebrated for its richly silvered
swimming bladder; it is so brilliant as to be seen
through the body, and has long been used in the manu-
factory of false pearls. Hitherto we have noticed only |
such salmon as have a small, or, at least, only a mode-
rate sized mouth: but there is another race, whose food
must be entirely animal ; the mouth is excessively large,
the gape opening far beyond the eye; and the sides of
the jaws, as well as the inside, are armed with long
slender teeth, of different sizes, and moving backwards
at their roots: this gives them an appearance of being
flexible, but they are not so; for if an attempt is made
to bend them forward, they become as firm and hard as
if they were rooted in the jaw: such is the nature of
the teeth in the genus Laurida of Aristotle, which was
the Salmo Saurus of Linneus.* The use of this struc-
ture seems to be, that the fish may swallow its prey at
once, and that it may glide down the throat without
being impeded by the numerous teeth it meets with in
the passage. The enormous gape, indeed, of these
fishes, shows that they swallow others of'a dispropor-
tionately large size; and the teeth, being so very slender
and acute, are only used for the purpose of capture.
One species, the Salmo Saurus, of the old writers
(Laurida Mediterranea Sw.), is found in the Mediter-
ranean, and presents no very marked difference in struc+
ture from others found by Spix in the Brazilian seas:
the head is covered with strong bony plates, or, rather, it
appears to be naked; and the scales are firm and hard.
The Lauride are also generally remarkable for the
smallness of the pectoral, and the great size of the
ventral fins. Although marine fishes, they have an
evident relation to Erythrinus, in their round and cy-
lindrical body, large mouth, and bony head; as well as
to the true salmons of Europe; the former being a

* M. Cuvier has not only rejected the elassic name given by no less a
maturalist than Aristotle to this group, but uses one (Saurus, a lizard),
which, under any circumstances, is totally inadmissible’: we may just as well
employ Avis, Amphibia, Reptilia, &c. to designate ichthyological genera.

‘ON THE TEETH OF THE SALMON. 243

relation of analogy, while the latter is obviously one of
affinity.

(217.) The foregoing observations will give the
general reader some idea of the most remarkable fish
included in the two typical divisions of the Salmonide:
but this manner of treating the subject is so dry, and
so little calculated to excite the attention of the natural-
ist, that we shall now attempt to investigate the whole,
with reference to the natural series, and to the different
relations of the minor groups. True it is, that our ana-
lysis has not been carried so far into this family as into
many others ; and therefore we must have recourse to
that mode of investigation already explained.* This
has partly arisen from the insufficiency of those charac.
ters which have been given to the numerous sub-genera
recently proposed, the majority of which are made to
rest entirely on the shape of the teeth: thus it is that
natural groups have been broken up into smaller ones ;
and these being all considered of the same rank, subor-
dination in their value has been lost sight of. If the
teeth are really of such primary importance in classifi-
cation, why are they viewed so in one instance, and not
in another—not in different orders and families, but in
the same genus? Why, for instance, is Myletes to be
separated from Serrasalmo solely “on account of their
teeth,’ when, in the very next sub-genus, Hydrocyon,
M. Cuvier states that ‘“‘ some have a crowded range of
small teeth on the maxillaries and palatines; others, a
double range on the intermaxillaries and lower jaw, and
none on the palatines; others, a single range on the
maxillaries and lower jaw ;” and, finally, “‘ others have
absolutely no teeth whatever, except on the intermaxil-
laries and lower jaw.” Here, then, is a group confess-
edly varying in the teeth of almost every species, yet to
which no definite characters, taken from the shape or
fins, are given, by which the naturalist can possibly com-
prehend the extent of the author’s meaning. For our
Own parts, we confess our perfect inability to compre-

* Page 3. (2.)
R2

DAA . CLASSIFICATION OF FISHES,

hend, from the definitions in the Régne Animal, the
greater part of the sub-divisions of the old genus Cha-
racinus, proposed by its learned author, — a genus which
is really abolished, although nominally retained. We
are obliged to make this exposition of the utter insuffi-
ciency of such characters, in justification of our opinion,
so often expressed and acted upon, on the secondary va-
lue of M. Cuvier’s dental system, and of the inconsis-
tency, seen in the examples just quoted, of employing and
rejecting such characters in two genera absolutely placed
close to each other. To act in opposition to high sei-
entific authority, without assigning reasons, implies not
only great presumption in the dissentient, but contempt
towards his predecessors.

(218.) On taking a comprehensive view of all those
genera which accord, more or less, with the characters
assigned to this numerous division, and placed in it by
M. Cuvier, we shall have no difficulty in determining
that Salmo, Characinus ( Artedi), Xiphostoma, and Ster-
noptyx are the most prominent or dissimilar types of
form among all those which have an adipose dorsal: to
these, for reasons which will afterwards appear, we shall
add Sudis, placed by M. Cuvier between Amia and Os-
teoglossum (Vandel.). We shali now endeavour to assign
to each of these their determinate characters, and to
assemble the minor groups under those to which they
appear most allied. The two typical divisions appear
to be Salmo and Characinus; while the three aberrant
are here considered as Xiphostoma, Sternoptyx, and
Sudis. |

(219.} The genus Salmo, with its subordinate types
or sub-genera, first claim our attention. In this group
we include all those divisions of Cuvier, whether they
are called genera or sub-genera, which more or less agree
with the general structure of the salmons, chars, and
trouts of Europe. On generalising the characters of
these fishes, it will be observed that the body is of a
lengthened oblong shape; the head small, and the
muzzle narrow and rather pointed, without being length-

THE SUB-GENERA .OF SALMO. BA5

ened: the two jaws are of nearly equal thickness, or,
rather, the under one is narrower and weaker than the
other: the mouth is wide, and cleft in nearly a hori-
zontal direction: the first dorsal fin is in the middle of
the back; while the anal fin is so short, as rarely to be
Jonger than the breadth of its anterior. These charac-
ters may be seen in full perfection in the salmon (Salmo),
with which we include the smelts (Osmerus) and Co-
rygonus: Mallotus has them only in part; the chief devi-
ation being in the anal fin, which is more lengthened.
‘The true genus Lawrida of Aristotle, as well as Ana-
stomus of M. Cuvier, complete the divisions of this genus.
We consequently do not adopt Thymaiius, or admit that
the smelt is any thing more than an aberrant species of
Salmo, just as Aulopus is of Laurida. Thus circum-
scribed, the European salmons with short anal fins are
all referrible to the following sub-genera, viz. Salmo,
Laurida, Anastomus, Mallotus, and Corygonus; the
affinities and rejations of which appear to be as follows.

(220.) The European salmons, forming the genus
Salmo, have a small sharp head, with a mouth so
large as to have the gape often extending beyond
the line of the eye. The first dorsal is nearly central,
and the ventral immediately under it; the anal fin,
in like manner, corresponds in its situation to the
second or adipose dorsal; while the pectoral is small,
pointed, and placed very low down towards the throat.
The teeth are very numerous in all; being placed,
generally, on ali the maxillary bones, the palatines,
and the vomer: in some species, like the smelt, there
are very few teeth in the latter part ; and the lower
jaw, in both sexes, is longer than the upper; but
as this is a typical group, such slight variations are
always to be expected. In the Saimo tumbdil (Bloch,
436.), we have the first material deviation from the
typical characters. The pectoral ‘is placed higher up
towards the back; the ventral fin is before the first
dorsal; and although not greatly developed, is yet as
Jarge as the pectoral: the lower jaw, as in the smelt

R 3

246 CLASSIFICATION OF FISHES.

(Salmo Eperlanus), is longest; but, in all other respects,
the head of this fish is a compound of Salmo and
Laurida,— two genera widely separated by Cuvier, but:
which actually pass into each other in the most gradual
manner: the Salmo fetans (Bloch, pl.384. f.2.) brings
us, in fact, at once into the genus Laurida (Laurida Me-
diterranea Nob., fig. 48.), the peculiar and discriminat-

ing character of which is, that the ventral fin is so much
developed as to be considerably larger than the pec-
toral, near to which it is placed, —and not, as in Salmo,
immediately under the first dorsal. Some of the Lau-
ride of America (L. microps, fig. 49.) have the eyes very

small, and the teeth (a) crooked and hooked, or, rather,
half barbed ; the mouth being excessively wide: while the
jaws in others are unequal. In all, however, the sides
of the head are covered with scales, and the ventral fin
is very large. For convenience we may retain the sub-
genus Aulopus, although it consists of but a single
species, and has such a close resemblance in most parts
of its structure to Laurida, that it seems to us to
belong to the same generic group. M. Cuvier supposes
this fish to connect the salmons with the cods ; but we

SALMONIN&.—LAURIDA, QA7

do not see in what manner it bears any relation to the
Gadide, in any other respect than. its small teeth. It
has the large and firm scales, the ample ventrals, the
small pectorals, and the adipose dorsal of Laurida ; from
which it chiefly differs in the smallness of the teeth,
and in having (like Laurida microps Sw.) the ventral
placed immediately under the first dorsal; two charac-
ters which bring it closer to Salmo than to any other
group. The singular protrusion of the point of the
chin, in the lower jaw, establishes a remote relation of
analogy between Aulopus and Sphyrena ; and all these
circumstances tend to confirm us in the views we have
taken on its true affinities. It should here be observed,
that as Laurida stands as an aberrant group, so we find
it loses one of the typical characters of the European
salmons, and begins to assume another of the American
ones: the anal fin, instead of being short, is always
longer than in Salmo; and the teeth differ from all
others of the group, by being excessively crowded, of
unequal lengths, and pointed or moveable at their
roots. The two groups appear connected by such fish
as Laurida minuta (fig. 50.), where the anal and ven-

—

SS vanes = SS —
Bort OAK BN CA
= Z NC ATALT RO Ue

tral fins more resemble those of the true salmons. The
species of Laurida are not numerous, and appear to
be altogether marine, without growing to more than two
ot three feet in length: they do not ascend or enter
freshwater rivers, like the salmons ; but, being oceanic,
are found both in the Mediterranean and Atlantic
Oceans. Following Laurida we place the genus Anasto-
mus Cuv.; not because there are any species of one or
the other which evince any affinity to each other, but
because this is the only genus which, possessing the
R 4

948 CLASSIFICATION OF FISHES.

short anal, and some other characters of the European
salmons, has the mouth completely vertical. It will be
seen that, throughout all the natural groups of ichthy-
ology, one type with the mouth opening vertically is
always to be found ; and this variation is so prevalent,
that whenever it is met with, we may be almost sure
it is the chironectiform type of that circle of which, in
all other respects, it possesses the leading characters.
M. Cuvier has, therefore, very judiciously made the S.
anastomus of Linn. (fig. 51.) the type of a genus distinct

from that of Gasteropelicus, although he was not aware
of the reasons which induce us now to adoptit. In
all other respects but its narrowed head, and very small
vertical mouth, Anastomus has the general shape of
the true salmons ; the size, form, and relative propor-
tion of the fins being the same: it cannot be said, how-
ever to have the “form of the 8. Thymadlus,’ because
the first dorsal fin is not at all lengthened ; nor is it so
proportionably high as in that fish. We next come
to the sub-genus Mallotus, which has also the general
structure of Salmo, but with the remarkable character
(unique in this group) of rounded pectorals ; all the
other salmons having these fins pointed. Only one species
of this singular type is known, — the Mallotus Gren-
landicus, or Greenland salmon, — which Bloch (pl. 381.)
describes as a small fish, mostly used as a bait for cods.
This type has the further peculiarity of the tail or caudal
fin not being symmetrical with the pectoral; the latter
being rounded, whereas the former is forked; while the
anal is more lengthened than in any of its congeners:

SALMO.—ANALOGIES OF THE SUB-GENERA. 249

the teeth, as in Anastomus, are small and crowded ; and
the lower jaw is even longer than the upper one, although
the form of the head is like that of the ordinary salmon.
The transition from this type to Coregonus * (Artedi) is
beyond dispute. Cuvier, in fact, makes them to follow
each other, with the intervention, indeed, of the Corego-
nus Thymallus, or the grayling, which he places as a
distinct sub-genus; but which we consider the most
typical example of the genus from which he has separated
it. In this view we only adopt the opinion of Artedi,
and, more recently, of Dr. Richardson. Coregonus, then,
is composed of the graylings, well distinguished from
the true salmon by having a very small mouth, furnished
with a few slender teeth, which are altogether wanting
in some of the species: the under jaw is always the
shortest ; and the gape of the mouth is before, instead of
behind. the line of the eye: in all other respects the ge-
neral structure is completely similar to that of the large-
mouthed salmons; so that, having returned to the type
we originally commenced with, we may place the whole
in a column, and trace the analogical relations of the
series to the primary orders.

eae Suan Analogies. Orders of FisHEs.
Sana: ee ee the most highly ee NGANTHOp rE ORS
Coregonus. Sub-typical. MALACOPTERYGES.

/ Body lengthened; pectoral fins
Maliotus. annded: d i APODES.

Anastomus. Mouth very small, opening vertically. PLECTOGNATHES.
Voracious; mouth large; head de-

Laurida. pressed.

i CARTILAGINES.

cm

If the question was asked, whether Salmo or Coregonus
was the most highly organised group, no difference of
opinion could possibly arise; because a fish that has all
the parts of its mouth well armed with teeth is unques-
tionably more perfect, that is, of a superior organisation
to one that has few or none of these organs: now, this
is Just a parallel case with the Acanthopteryges and the

* This genus was founded by Artedi,—not, as is supposed, by Cuvier.

1

250 CLASSIFICATION OF FISHES.

Malacopteryges, or the spiny and the soft-rayed orders.
Other analogies may possibly exist, but these are quite
sufficient for our present purpose. Mallotus, it will be
remembered, is the only division of these salmon which
possesses very ample rounded pectoral fins; and the
apodal is the only order to which, from this structure
being absolutely universal, this type can be compared.
It is further remarkable, that M/allotus has the body very
much lengthened,—another point in which it shows an
analogy to the eel-shaped fishes of the apodal order:
the length of the anal fin (always more developed in
this than in any of the primary types of fishes) likewise
favours the supposition that Madlotus is the anguilli-
form type of the salmon. The analogy of Anastomus
to the Plectognathes is so evident, that no additional il-
lustration is necessary. There now remains only Lau-
rida, which stands opposite to the cartilaginous order.
If any of our readers wishes to be convinced that these
are the sharks of the salmon race, let him look at the
head of one of these species (fig. 48.), and, without being
exactly able to explain in what this analogy consists, he
will be convinced that it is founded in nature. One
important character of these salmon, not mentioned by
our predecessors, is the great flatness of the head, and
the almost vertical position of the eyes: the formidable
nature of the teeth, which, in comparison to the size
of the fish, are excessively large, is another of the
many points of resemblance between Laurida and Squa-
lus, and this is accompanied by that destructive warfare
which each, in its way, carries on among other fishes.
Thus we see that the theory of analogy confirms our
disposition of these groups, and sanctions us in rejecting
several of the sub-genera of the Régne Animal.

(221.) To give additional support to our present
arrangement of the foregoing fishes, we shall now com-
pare them with the primary divisions of the entire family
of Salmonide, in which, it will be remembered, we also
bring in the herrings (Clupeine), the pikes (Esocine),
the snout-fish (Mormyrine), and the carps (Cyprine).

-

SALMO.—ANALOGIES OF THE SUB-GENERA. 251

On placing these two groups together, we shall get the
following mutual resemblances : —

Py an : Analogical Characters. See tie
Salmo. The pre-eminent type of the SALMONINE.
Coregonus. ‘Teeth verysmallornone; mouthsmall. CLUPEINa.
Mallotus. Pectoral fins rounded. CYPRINE. ~

uzzle produced; mouth terminal

’
very small. F MorMyRINz&.

Anastomus. f -
Head flattened; ‘mouth very large ;
Laurida. ; teeth long, sharp, slender, exceed: ( Esocrss,
ingly numerous.

Our restricting the typical salmon to those with large
mouths and numerous teeth, makes them representatives
of the whole family, so that nothing further need be
said in confirmation of the first analogy. The small
mouth, never cleft beyond the eyes, with the scales
much larger than in the true salmon—setting aside the
general appearance of the fish — renders Coregonus a per-
fect representation of the Clupeine, or herrings: hence
the specific name of Clupeoides, given by Pallas to one
of the species. Again, Mr. Yarrell observes, the ap-
pearance of the gwyniad (Coregonus fera) “is not unlike
a herring. Indeed, the common people of the lakes
where this fish is found, call them the freshwater her-
ring, and preserve them in the same manner, with salt.”
Facts like these are always valuable; for there must
always be something real in resemblances that have ac-
quired popular credence. . A rounded pectoral fin is as
prevalent among the carps (Cyprine) as a pointed one
is characteristic of the salmon: we may therefore com-

_ pare Mallotus with the carps, for it is nearly the only

salmon * possessing this analogical character. There
may, indeed, be some doubts on this analogy; but there
can be none with regard to the next, or that between
Anastomus and Mormyrus; for the former has quite the

incipient trumpet-shaped snout of the latter, while in

* The parr or samlet (Salmo Salmulus), so ably determined as a distinct
species by Mr. Yarrell, is represented, indeed, with rounded pectorals; but
as no mention is made of this unusual deviation in the description, some
uncertainty hangs on the subject. I have never seen this fish. See Yarr,
Brit. Fishes, vol. ii. p. 42. 3

woe : CLASSIFICATION OF FISHES.

every other character it is a true salmon: this analogy,
in fact, is particularly strong, yet not more so than that
between Laurida and the Esocine, or pikes. Both
these have enormous mouths, presenting, when opened,
a forest of teeth of all sizes, and in all the internal
parts ; so that the Lauride are the pikes of the sea, as
much as Esox is of the fresh waters. This result,
which is the consequence of restoring Laurida to its
natural situation, close to the typical salmon, is exceed-
ingly interesting, because it establishes the two most
ebvious analogies: one being remote, as to the sharks; the
other approximate, being to the pikes: We look on this
part of our arrangement as the most certain affinity in
the whole series: and we shall now proceed to the next
group.

(222.) To the second great genus of the salmons
we retain the original name of Characinus, given to the
majority by Artedi. The number and variety of these
are even greater than those of the last genus; but they
are all natives of Tropical America. Unlike the Eu-
ropean salmon, the majority of these fish are short and
deep in the body, so that some assume the discoid form
of the flat fish : the mouth is short and very blunt ; and
although not wide, is moderately large: the commis-
sure, instead of being straight and nearly horizontal, is
considerably angulated, and obliquely vertical. They are
further distinguished from the European types by the
superior length of the anal fin, which is generally four or
five times as long as it is broad ; whereas, in the genus
Salmo, this fin is always short, at least in the typical
examples. The only exception to these characters
occur in certain Brazilian river fish, placed by Spix
and Agassiz in the sub-genera Curimata, Leporinus,
Anodus, and Prochilodus: these are all of an oblong
shape, much like that of Coregonus, which they further
resemble in the shortness and breadth of their anal fins,
and the smallness of their mouth. The natural station
of these sub-genera, and their relative rank, is too diffi-
cult to allow of our arriving at any certain conclusion :

5

CHARACINUS.— THE SUB-GENERA. 253

-M. Cuvier places them all in his sub-genus Curi-
mata; but although some, as C. fasciatus (Spix,.
pl. 36.*), evince a resemblance to Anastomus, by their
slender snout, and excessively small mouth, almost ver-
tically cleft, it is very questionable whether Prochilodus
is also allied to the others: two species are figured
by Spix, which, in their small mouth, thick fleshy lips,
and absence of teeth, no less than in their body and
fins, perfectly agree with the carps; the only difference
being a very small adipose dorsal, placed immediately
above the short anal. We see no possible affinity be-
tween these singular fishes and the Salmo Thymallus,
with which M. Cuvier has compared them: they have
not, as in that, the first dorsal fin high, long, and unu-
sually developed ; it is, on the contrary, of the same
ize as the ordinary species of Coregonus ; and we feel
by no means sure, whether the majority of the above-
named salmon, although natives of South America,
should not be placed with the European group. On
the other hand, it must still be remembered, that all
these tropical salmon differ from ours, in having no
teeth on the tongue; and that the’number of rays in
their gill membrane is rarely more than four or five :
the wide separation, also, of the latitudes they respect-
ively inhabit, must not be overlooked ; so that, upon
the whole, we may safely conclude them to be the Ame-
rican representives of Coregonus, just as Catastomus is, in
the New World, of the European and Asiatic Cyprine.
Now, with the exception of these fishes, the whole of
those now under consideration agree in having the
characters already assigned: that is to say, the snout is
short, thick, and obtuse ; the mouth angulated ; and the
anal fin more or less lengthened. From all the divi-
sions made by Cuvier of Artedi’s genus Characinus, we
select the following as the most dissimilar to each other,
and these we consider as types of form, viz. Characinus
Artedi, Servasalmo Lac., Chalceus Cuv., Gasteropelicus
Bloch, and Cynodon Spix and Agassiz. In this selee-
* Also Anodus latior, pl. 41., and A. elongatus, p\. 40., of the same author.

254 “CLASSIFICATION OF FISHES.

tion we have been guided entirely by outward structure ;
since the different modifications of the teeth, as M. Cuvier
himself confesses, “‘ are varied in the most surprising ©
manner ;”’ so that almost every species differs from its
congener, and proves the utter futility of attempting
to make them the bases of generic characters. Now,
as the form of the body and fins partakes also, in
some degree, of this unusual variation, it may be
better to view this division as assuming the rank of a
sub-family, and to consider the above-mentioned groups
as genera, rather than as sub-genera: this will enable
us to specify the minor types, and to designate them, we
hope, in a more comprehensible manner than has
hitherto been done.

(223.) The Characine, or tropical salmon, appear
typically represented by the genus Serrasalmo of Lace-
pede. These are a group of large, stout fish, whose depth
is more than half as much as their length: the snout
is blunt; the head small ; the gills very large, bony, and
naked: the mouth opens obliquely downward : the upper
jaw is small ; but the lower one is excessively thick and
strong, being armed with sharp triangular cutting teeth
double the size of those in the upper jaw: the pectorals
are placed very low, close to the belly ; and the ventrals,
which are only half as large, are immediately under
the first dorsal fin; and both these latter are pointed :
the anal is long, broad in front, and gradually nar-
rowed behind: the scales are small; and the belly is
sharp and dentated like the teeth of a saw: in some,
as in S. nigricans (Spix, pl. 30.), there is a very short
procumbent spine, pointing forwards, at the base of
the first dorsal; while, in others, as in S. ferox Nob.
(Spix, pl. 28.), these anterior spines point backwards,
and three or four of the first dorsal rays are spinous.
Those fishes which form the sub-genus Myletes Cuv.,
have all the above characters, except that the procum-
bent spines are wanting, the teeth are blunt, and
the upper jaw is strongly angulated. The sub-genus
Tetragonopterus of Artedi differs only from the last in

SUB-GENERA OF CHARACINUS. 255

its mouth, which is smaller ; in its teeth ; and in not
having the belly carinated. These are followed by
Chalceus, which differs materially from all the fore-
going, by having the hinder part of the body lengthened,
and the pectoral fin so long as to reach to the vent:
the unusual development of this fin, as seen in C. an-
- gulatus (Spix, pl. 34.), is very remarkable, and de-
serves much attention, because it will materially help to
- determine the station of this type in a natural arrange.
ment. We retain the original generic name of Chara-
cinus to those which have been subsequently called
Curimata by Cuvier. These fishes, whose teeth vary
in almost every species, are nevertheless readily dis-
tinguished from all the other salmons, by their elon-
gated form, by the smallness of the mouth, and,
more especially, by having the anal fin as short as any
of the European salmons: their external similarity, in
fact, to Coregonus, indicates a clear relation of analogy,
if not of affinity. Following these we place Piabu-
cus Cuv., which has the general shape, mouth, &c. of
Characinus, but differing in the anal fin being very
long. Allied to all these small groups, is that of Gas-
teropelicus, —a singular little fish, having the shape of
Serrasalmo, but with the belly remarkably protruding ;
while the head is so raised upward, that the mouth
becomes completely vertical: the anal fin, as in the major-
ity of these groups, is very long and narrow; and the
belly sharp, without being serrated. Finally, we come
to the genus Cynodon of Spix, whose aspect is altogether
different from any of the preceding fishes: the head
is large ; and the wide mouth, which opens downward,
is armed, as in Laurida, with long slender teeth of
different sizes: the body is lengthened, and the anal
fin is uncommonly long. The two very remarkable
species figured by Spix*, are included by Cuvier in
‘his genus Hydrocyon ; but this group contains such a
heterogeneous assemblage of fishes, with little or no
affinity to each other, that it would almost seem to have
* C. vulpinus, Spix, pl. 26.; gibbus, pl. 27.

256 CLASSIFICATION OF FISHES.

been made a receptacle for all such salmon as could not
be arranged under the other divisions: for this reason we
cannot possibly adopt it. The genus Citharinus of the
same author is so slightly defined, that we can form no
idea of its true affinities; it seems to have both the adi-
pose fin, and the greater portion of the caudal, covered
with scales : as this latter character belongs to some spe-
cies of Cynodon, we may presume, not having seen an
example, that they are in some degree related.

(224.) We have now enumerated the greater part of
those American salmon, which, from having the anal fin
lengthened, will enter into our definition of the CHara-
civ, Without venturing to determine the value of
this group collectively, or even to separate the sub-
genera from the genera, we are yet led to believe that
this is by no means an artificial assemblage: the
natural succession of the types may possibly prove dif-
ferent from that series in which we have arranged
them ; and even some, as already hinted, may be found
eventually to belong to other divisions : and yet, with all
these difficulties and uncertainties in our way, there
is some reason to believe that the Characine really
contain representations of the Sa/monine: and that a
little attention to the peculiarities of the forms in each
will materially confirm this idea, will be apparent from
the following considerations.

(225.) In the first place, it will be observed, that the
most typical salmon of Europe are those which, like the
common species, have a widely cleft mouth ( fig. 52. 5),
with the jaws, and all the bones of the palate and throat,
covered with teeth; so much so, indeed, that Cuvier
has well observed, ‘‘ they are the most completely
dentated of all fishes.” Now, if we look to the Ameri-
can salmon, we shall find that the extraordinary deve-
lopment of these organs takes place likewise in the
Serrasalmi: the teeth of these ravenous fishes are as
formidable, in size and structure, as those of the
sharks ; taking into account the relative size of the two
races: the tongue, indeed, in those of the American rivers,

ANALOGIES OF THE SALMON. Qh

is smooth ; but then the
teeth of the jaws are large,
triangular, cutting, and
dentated: in some, in-
deed, as in Myletes Cuv..,
the teeth are so strong as
to be employed in mas-
tication. The form, num-
ber, and situation of these
teeth vary in a remarkable
manner ; but, from being
present in all the species
and sub-genera, we may
at once decide that the
Serrasalmi are the most perfectly dentated of the
American division, just as the true salmons are of
such as occur in Europe. The affinity between Ser-
rasalmo and Characinus (fig. 52. a) is as intimate, ac-
cording to the way in which Cuvier has placed them in
his system, as that which unites Salmo with Coregonus:
now, the teeth in both these groups are always smaller
and fewer than in the two preceding ; they have both
a small mouth ; the scales of both are larger; and in
both, the teeth are often altogether wanting: this re-
semblance, in fact, between the two is so strong, that
we might almost think they were united by affinity,
because it will be remembered that the different types
ot Characinus are the only fishes among the American
salmon which have the anal fin short. There is, how-
ever, a peculiar aspect about these latter, by which the
practical ichthyologist will readily detect them, inde-
peprdent of their geographic range, and the remarkable
difference in the number of the rays to the gills. In
Coregonus, these rays amount to seven or eight ; but in
Characinus, and all the American salmon, they do not
exceed, according to the best authorities*, more than
four or five; nor have the American fishes any teeth

* Artedi, Cuvier, Bloch: the former chiefly founded the genus on
this very peculiarity.
VOL. I. Ss

9258 CLASSIFICATION OF FISHES.

upon the tongue. Let us next compare Piabucus and
Mallotus. Of all the types of the Characine, Piabucus

is that which has the longest anal fin, which, in two out —

of the three species figured by Bloch*, is equal te one
half the length of the body. Now it will be remembered
that Mallotus is the only type of the European salmon hay-
ing this fin very conspicuously lengthened. The pectoral
fins of both are more than usually developed ; but in one
they are pointed, and in the other rounded: while the two
types, in other respects, are sufficiently distinct to check
any suspicion of their being any real affinity between
them. The slight “ace Peenn es of the belly in Piabucus
argentinus appears an indication of its being followed
by Gasteropelicus, which, in the vertical position of its
little mouth, is such a complete representation of Ana-
stomus, among the true salmon, that whatever may be
the situation of the two, they must always stand as re-
presentatives, not merely of each other, but of all the
chironectiform types in ichthyology. There yet remains
the singular genus Cynodont of Spix among the Ame-
rican salmon, which differs from all others by the ex-
cessive wideness of its mouth, armed with long, slender,
pointed teeth of various sizes. Its long anal fin shows
its relation to Characinus, and the slight protuberance of
its belly to Gasteropelicus ; but if these two peculiarities
are set aside, the reader will immediately recognise in
the foregoing description a counterpart of Lauwrida,
whose wide mouth and long slender teeth are altogether
unique among the salmons of Europe.

(226.) That the above analogies may be brought to-
gether at one view, we now subjoin the annexed table,
leaving it for future ichthyologists to determine the rank
of the different groups it contains.

* P. argentinus, Bl. 382., fig. 1.; bimaculatus, ib. fig. 2.

+ The genus Synodus of Gronovius is supposed by Cuvier to rest upon a
fish which he thinks is a Laurida, whose small adipose {dorsal had ac-
cidentaily been omitted by:the artist, or broken in the specimen. This may
possibly be the case; and yet the remarkable correctness of all the figures
» of Gronovius, and his great accuracy as an ichthyologist, renders it equally
probable that this figure represents a fish unknown to modern ichthyolo-
gists,

XIPHOSTOMAX. , 259

Analogies of the EuRopEAN and AMERICAN SALMON.

American Salmon. Resemblances. European Salmon.

Mouth small; teeth minute. s
often wanting. >} CorEconus Art.

Mouth larger ; teeth strong :
numerous. ” Samo Linn.

CHARACINUS Artedi.

SERRASALMO Lac.

a A s Mouth excessively large ; P
= ES ODUS Sid i teeth long, eciee ncrnae | Est Li
GasTEROPELtcus Bi. Mouth small, vertical. ANASTOMUS Cu.

Prasucus Cuv. Anal fin very long. Mautotus Cuv.

(227.) We now come to the three aberrant types of
this sub-family, represented, as we conceive, by the
genera Xiphostoma Spix, Sternoptyx Herm., and Sudis
Cuv. The first of these is represented by one of the
most singular forms among the salmons; and so strongly
does it recall to cur mind the form of the Xiphias, that
it may be termed the sword-fish of the salmons. It
will be seen, from the annexed.cut (fig. 53.) that the

body is elongated; and although both jaws are con-
siderably lengthened, the upper one slightly exceeds the
other, and terminates in a smal] but very distinct point:
the angulated upper jaw, so conspicuous in the last
group, is also continued to this, and there is a small
adipose fin ; here, however, all similarity between them
ceases. The first dorsal fin is placed much nearer the
caudal, and both that and the adipose fin are situated
rather behind the ventral and anal: the teeth are very
remarkable: being all equal, with their points directed
backwards: the ventral and anal fins are triangular, and
of equal size; while the hard compact scales, marked
with longitudinal elevations, remind us immediately of
Laurida. In the sub-genus Hepsetus Sw., which in-
cludes the African or Old World representatives of the
American Xiphostomi, the jaws, although narrow and
s 2

260 ‘CLASSIFICATION OF FISHES.

‘rather attenuated at their tips, when viewed laterally,
are nevertheless considerably wider above ; but they are

not lengthened: both are equal, and armed with sharp |

slender teeth of unequal sizes. In these fishes, the first

dorsal is still further removed from the head than in

Xiphostoma, and the upper jaw is more angulated: ‘the
appearance of all these fisn reminds us so much of pikes,
that they may be termed pike-salmon: but few species
are known, and they all appear to inhabit the tropics of
the Old World.

(228.) The genus Sternoptyx is one of the most sin-
gular forms in this or any other order, and yet it is by
no means so anomalous as its first aspect might tempt
us to believe. Let the reader only imagine a highly
exaggerated figure of Gasteropelicus, and he will have
a very good notion of the general shape of these sin-
gular fishes. We are not aware of any other figures of
the two species already described, than those which are
to be found in most works, copied from Hermann ; and
on this account we regret the more that several speci-
mens of two other new species we discovered in the
Mediterranean, and deposited in the British Museum,
are now no longer in existence: as we depended upon
these for subsequent descriptions and drawings, we are
compelled partly to transcribe what Cuvier has said of
the species known to inhabit the warm parts of the
American coasts, for hitherto no one appears to have
detected them in the Mediterranean. The annexed cut
( fig.54.),copied from
the rude figure of
Hermann, will tend
to elucidate the fol-
lowing anatomical
description given by
Cuvier. These fishes,
he observes, have a
very deep and considerably compressed body, with the
mouth directed upwards: the humeral form a sharp
ridge in front, terminated below by a small spine; the

STERNOPTYX. 961

bones of the pelvis form another, also terminated below
by another small spine, in front of the ventrals, which
are so small as to have been formerly overlooked: along
_ the ridge of the pelvis, on each side, is a series of small
fossettes, which have been regarded as a festooned fold
of the sternum, and has given rise to the generic name.
In front of the first dorsal fin is an osseous or mem-
branaceous ridge or crest, which appertains to the anterior
interspinals; and behind this fin is a small membraneous
projection, which represents the adipose fin of the
salmons: the sides of the mouth are formed by the
maxillaries. Of the two described species, S. diaphana
has small, even, and crowded teeth, and five rays to the
gills: its form is singularly oblique, for the mouth pro-
jects beyond the ventral line. The other, S. Olfersii,
has hooked teeth, and nine rays to the gills. The
Sicilian species, before alluded to, is so exceedingly rare,
that we only met with them twice during six years,
and both times in the same situation, cast up on the
shore cpposite Reggio, in the Straits of Messina, after
violent storms: on one of these occasions, near a dozen
specimens were found, most of them much broken by
the action of the surf: excepting for their bones, they
would have been little thicker than a wafer, and their
colour was of the most brilliant silver. In all proba~
bility they live in very deep water, for their whole
structure is analogous to that of some of the riband-fish,
the eyes being excessively large, and the fins very brit-
tle. We remember comparing them with the descrip-
tions in Shaw*, and observing some differences ; but
depending on the power of describing them hereafter,
we neglected taking any notes; and the specimens being
now lost, as already stated, further information upon
them, from us, at least, becomes impossible.
(229.) The singular genus Swdis is placed by Cuvier
mm the herring family (Clupeine), close to Erythrinus
and Amia; it has an evident affinity with the two last.

_* Shaw, probably on the authority of Hermann, states that there is no
gill membrane. — Gen. Zool. vol. iv. p. 112.

s 3

262 CLASSIFICATION OF FISHES.

but we cannot discover any relation it bears to the
herrings,—even in a solitary character: its depressed

head, large mouth, and strong teeth, and even some-

thing in the position of its fins, would lead us, in the
first instance, to arrange it among the pikes (Esocine) ;
its relation, however, to Evythrinus appears, upon the
whole, more close; and as we have placed this latter
‘genus as an aberrant form in the circle of the Cyprine,
so do we arrange Sudis as the connecting link between
the salmons and the carps. Whether this is its true si-
tuation in nature, it is impossible, in the present state
of things, to determine ; but it appears much more na-
tural (when we consider its resemblance to Erythrinus,

and of this latter to Gonorynchus) than to associate it with .

the herrings. The Sudis gigas (fig. 55.) is the largest

55

of four or five species which seem to be distributed in
the fresh waters of America and of Africa. The typical
form to which it shows the nearest approximation is
clearly the anguilliform ; and as we have no genus in
the primary divisions of the Salmonide which represents
those fishes, we confess that this consideration has ma-
terially influenced us in giving this station to Sudis.
The scales are very large, strong, thick, and bony: the
bones of the head are hard, naked, and rough : in some
the muzzle is oblong, and in others shorter; while that
of S. gigas is evidently depressed. In S. Niloticus, ac-
cording to Ehrenberg, there is “a singular funnel spi-
rally convoluted, which adheres to the third gill,” which
Cuvier, with much probability, conjectures is analogous
to those which he has so ably and beautifully investigated
in the genera Anabas, Ophiocephalus, &c. We have
not yet come to our exposition of the spine-rayed order
(Acanthopteryges), and therefore any partial exposition
of its analogical characters would be premature; but if

=

ANALOGIES OF THE SALMONIDA. 263

the reader subsequently compares the circle of the Sal-
monide with that of the tribe Macroleptes, he will find
that Sudis stands opposite to Anabas and the other laby-
rinthiform fishes, as their representative among the soft-
rayed families, or the Malacopteryges. This analogy
may, nevertheless, truly exist, and yet the precise situation
of Sudis may be in some other circle; although, for the
present, we believe it is really where we have placed it.

(230.) Such are what we consider as the three great
aberrant divisions of the whole sub-family of the Sa/mo-
nine; and we are now to seeif they tally with those two
whose types we have more particularly investigated,

namely, the Salmonine and the Characine.

Primary Divi-

sions of the Sat- Analogical Characters. CINE On Sener

ae SALMO. CHARACINUS.
SALMONINE. Nat a teeth vacate Salmo. Serrasalmo.
Mouth small; teeth often ‘ Be
CHARACINE. wanting. { Coregonus. Characinus.

Mouth very large; teeth
XIPHOSTOMA, 5 many, long, slender, un- Osmerns Cynodus.
equal.

STERNOPTYX. Mouth small, vertical. Anastomus. Gasteropelicus .

Body lengthened eel. Mh
SupIs. shaped; ventral long. } Mattotus. Piabucus.

(231.) Lastly, as it will tend much to strengthen our
disposition of the Salmonid@, we shall compare the five
groups in which we have arranged them with the five
principal divisions of the whole family ; because, if the
series in these are natural, they must possess some points
of mutual resemblance.

Genera of the Sub-families of the

Analogies.

SALMON. SALMONIDE.
Salmo. j Typical of the Salmonine.
Body and belly much com-
Characinus. pressed, the latter often ¢ Clupeine.
serrated.
Muzzle depressed above ;
Xiphostoma. teeth numerous; dorsal Bn francine
placed nearest the tail.
Sternoptyz. Mouth vertical. Mormyrine.
Sudis. 2 Cyprine.

The circumstance of the European salmons being
placed by all writers at the head of this family, as re-
s 4

x

264 _ CLASSIFICATION OF FISHES.

presentatives of the whole, does away with the necessity .
of any further elucidation on this point. This view of
the subject, however, is fully confirmed by the beautiful
manner in which the salmon of America (Characinus)
represent the herrings (Clupeine): both these groups
have the body much compressed, more especially on the
belly ; and numerous examples in both occur of this part
being not only sharp and carinated, but serrated like a
saw: all the serrasalmos in one, and the saw-herrings
in the other, answer to this description. We before re-
marked the similarity of Xiphostoma to the pike, not
merely in its formidable teeth, but its depressed upper
snout, and the backward position of its first dorsal ; so
that here, again, the analogy turns out to be strictly -
conformable to the natural series. We know so little
of Mormyrus, that we are at a loss to discover in what
respect it more immediately resembles Sternoptya ; cer-
tain, however, it is, that if the existing descriptions are
correct, Mormyrus has the smallest mouth of any
genus in the salmon group; and if we seek for this
character in its highest state of development, we find it
only in the chironectiform types, of which Fistularia,
Centriscus, Gomphosis, &c. are notable examples: these
genera, as will be seen hereafter, come in as the chironecti-
form types of their own groups, representing Balistes, &e.
by their very small mouth, and Chironectes by its vertical
direction. Swdis, in like manner, can only be assimilated
to the carps through other groups. Having already en-
deavoured to prove that the Cyprine are the represent-
atives of the eels among the salmons, we come also to
the same conclusion respecting Sudis. The analogy of
the first is shown by its food and habits, while that of
the last is manifested by the lengthened form and general
aspect. Fond as we are of tracing analogies, we cannot
let them interfere with what appears to us natural af-
finities ; and as we join with Cuvier in believing that
Sudis is closely allied to Erythrinus, and that these, by
means of Gonorynchus Gronov., pass into the Cyprine,
—we cannot substitute any other type in the place of

THE CLUPEINZ, OR HERRINGS. 205

Sudis, merely to give an additional air of perfection to
the foregoing table. We now close our survey of the
salmons, and proceed to the next family.

(232. > The Crurrin@, or herrings, appear to hold a
natural station between the salmons (Sa/monine) and the
pikes (Esocine). This situation has been assigned to
them by the best ichthyologists, and thus we have the
analogy strictly preserved in the circle of this class ; for,
upon comparing it with the spine-rayed fishes, we find
the mackarel (Scomberide) come exactly parallel to the
Clupeine, or herrings. In this very natural and strongly
marked group, the adipose fin is entirely wanting, and
the single dorsal is placed in the centre of the back, as in
our new species, Clupea aurovittata( fig.56.), which is as

nto

SNA
KOO ON,

NO) Nis y)
iy
vy) yyy)

> Boy Aan a es

typical as the common herring. All the species are ma-
rine, and very seldom ascend rivers beyond the influence
of the sea. The body is oblong oval, covered with large
deciduous scales: the belly sharp, and generally serrated :
the mouth, in its position, is mostly oblique, or inclining
more to the vertical direction than to the horizontal;
but it varies considerably in size, and in the armature of
the jaws : the teeth, however, when they exist, are always
small and slender. Cuvier remarks that the upper jaw
is formed like that of the salmons,—in the middle by in-
termaxillaries without peduncles, and on the sides by the
maxillaries. The opening of the gills is remarkably wide ;
and thus, asin all fishes so constructed, the herrings are
known to die a few minutes after they have been taken
out of the water. The bones of these fishes are more
numerous and slender than of all others.

(233.) The natural history of the whole family, as
far as known, bears a general resemblance to that of the

266 CLASSIFICATION OF FISHES.

common species, among which the herring and the
pilchard are the best known; but before we proceed to
this subject, we shall first enumerate the chief divisions
of this sub-family— numerous in species, but far less
diversified in outward appearance than the last. We
arrange the whole under the following genera, some of
which contain minor divisions, or sub-genera :—1.
Clupea, or the true herrings, having the body compressed,
the belly sharp and carinated, and often serrated ; the
dorsal fin in the middle of the back; and the mioate
opening in an oblique direction : the caudal fin is always
distinct from the anal, which, as in Characinus, is gene-
rally long.—2. Elops, or salmon herrings, possessing the
general form of the last; but the body is not compressed,
nor is the belly either sharp or serrated: the mouth is
moderate in size, furnished with teeth; and the anal fin
shorter. —3. The genus Chirocentrus differs materially
from the two former, in the very backward position of
the dorsal fin, which is almost as near the caudal as that
of the true pikes. —4. Pr istogaster (fig.57.), where the

small mouth is com-
pletely vertical, and the
belly curved outwards,
is serrated as strongly
as in any fish yet dis-
covered: that this ge-
nus represents, in the
most striking manner,
Ster noptyx, Gasteropelicus, and Anastomus in the last
family, is abundantly evident ; and yet in many re-
spects it seems so closely allied to some other kindred
forms among the aberrant sub-genera of Clupea, that
we have strong doubts whether it forms one of the
primary types of this family; no other, however, that
we know of, is a more perfect representation of the
chironectiform type.—5. Last of all, we place Osteoglos-
sum of Vandell (Ichnosoma Spix) as the genus more
immediately connecting this family with the last: it
differs from all others by having the ventral, and some-

GENERAL ACCOUNT OF THE HERRING FAMILY. 267

times almost the dorsal, united to the caudal fin : like all
such fishes, the body is more or less lengthened and eel-
shaped ; but its great compression, its large scales, and
the general aspect, clearly show an affinity to the herrings ;
while it also reminds us both of Laurida and Esowx, by
its numerous, slender, sharp, and unequal teeth.
(234.) The herrings, taken collectively, are moderate
sized fishes, the greater number not exceeding in size
that which is so well known on our own coasts ; yet a
few others, found in tropical seas, attain to the gigantic
length of from ten to twelve feet: many of them, how-
ever, are small; and of these the sprat and white bait
(Clupea alba Yarr.) are native examples. It appears
that the whole of this family, so far as known, are car-
nivorous; yet the animals upon which the greater part of
them feed, are very small: we argue this from the
excessive minuteness of the teeth, and even. in the total
absence of them in very many of the herrings; and
this idea is confirmed by the fact of immense quantities
of minute shrimps, resembling our sand fleas, having
been found in their stomachs: it seems, also, that they
greedily devour the roe or spawn of other fishes; since
large quantities are said to be imported into France from
the north of Europe, for the purpose of attracting
pilchards to the nets.* There are several genera, how-
ever, whose teeth are much more developed, which, with
their very wide mouths,'leads to the conclusion that they
feed upon other fishes and larger prey : among these are
the anchovies (Engraulis), Butirinus, Foca, and par-
ticularly Chirocentrus, whose teeth, in fact, much more
resemble those of the pike: the habits and economy,
however, of allthese are quite unknown. There is not
much diversity among the herrings in the form and
disposition of their fins: the dorsal is always single, at
least in the more typical genera, and most generally
placed towards the middle of the back ; this we make
the strongest mark of discrimination between them and
* For this, and several other facts connected with the herrings, see

Yarrell’s British Fishes, —a work which abounds with similar interesting
anecotes.

268 CLASSIFICATION OF FISHES.

the pikes, where this fin is invariably inserted very near
to the caudal: the pectoral is always pointed, and that
of the tail (except in Osteoglossum) is uniformly forked:
the ventrals are very small, sometimes obsolete, and
rarely wanting; while the lateral line is either very close
to the belly, or is not discernible. The unusual sharp-
ness and rigidity of the belly in these fish, if the
accounts of the mode in which they deposit their spawn
can be relied upon, is easily explained: when the season
for this operation commences, the herrings, abandoning
their winter quarters, proceed in large troops to the
breeding grounds ; there they commence rubbing their
belly against the ground, and, as if under great excite-
ment, they rapidly vibrate their fins, agitate their bodies,
and imbibe and reject the water through their gills with
unusual vivacity. The food of the British herrings has
been already noticed: those of Norway feed upon an-

other species of minute crab, named by Otho Fabricius, ©

from this circumstance, Astacus harengum. The num-
ber of these little creatures, during summer, is so near
infinity, that in taking up a jug of sea water it will be
often found to contain thousands. So partial are the
herrings to these insects, that they follow them where-
ever they are driven by the currents or tides; and by
feeding upon them continually, the belly of the fish
acquires a tinge of red, occasioned, according to Stroem,
by a reddish humour contained in these little creatures :
that putrefaction proceeds more rapidly in such herrings
as have been caught with their bodies thus filled, may
be readily supposed, for the same takes place in all other
animals ; and it is well known to cooks, that the best
method of keeping any animal for a long time fit for
eating, is to clear out the contents of the stomach.
(235.) Recent investigations have gone far to prove
the inaccuracy of those wonderful accounts, given by
Pennant and others, on the migrations which the her-
rings and pilchards were supposed to make every year,
from our own shores to the Arctic regions. The facts
upon which this was founded seemed to favour the

NATURAL HISTGRY OF THE HERRING. 269

supposition ; but in this, as in many other cases, these
very facts remain unimpeached, while they furnish
directly opposite inferences. They do, indeed, only
appear in immense shoals on our coasts at certain sea-
sons, and, to all appearance, disappear as rapidly as
they came; and yet it now appears that this migration
is only from deep to shallow water, and that the herring,
comparatively, is a domestic resident in our own seas.
Numerous observations, too long to be here repeated,
establish the fact, that the herrings inhabit the deep
parts of all our coasts throughout the year; since indi-
viduals have been caught in every month. The great
armies, however, of these fish, only come near the coast
in summer, for the purpose of depositing their spawn ;
and here, as Mr. Couch so justly observes, we cannot
but admire and praise the goodness of Divine Provi-
dence, by which these and many other fishes are brought
to the shores, within reach of man, at that particular
time when they are in their highest perfection, and
therefore best fitted to be his food. On these occasions,
the shoals may be compared to vast armies, led on by
the largest and most vigorous, and followed by the rest,
which are sometimes so numerous as to cover the sea
for miles; so that, on entering confined bays of the shore,
immense quantities have been stranded and crushed:
these are followed and assailed on all sides by birds and
hosts of ravenous fishes, such as the different species of
sharks, porpoises, &c., who gorge upon their feeble un-
resisting prey ; yet the numbers are so much beyond all
calculation, that their ranks are never thinned. Large
quantities are captured cn our own coasts ; but these are
far exceeded by the fisheries of Sweden and Norway,
where it is said that near 400,000,000 have been taken
in one year, and 20,000,000 in a single fishery. Go-
thenburg, in Sweden, is celebrated for the great abun-
dance of its herrings, of which there has been taken, in
one year, the almost incredible number of 700,000,000.
It is supposed that those taken to the northward of our
own coasts are finer than those of the south; and hence

270 CLASSIFICATION OF FISHES.

it is said that the fishermen of Scotland go out to meet
the shoals as far as the Crkney and Shetland islands,—
a useless labour, one would imagine, seeing that these
very fish, in a few days, would reach the southern parts
of Scotland.

(236.) The spawning is over, by those few, compara-
tively, who escape their numerous enemies, of which
man is the chief, by the end of autumn ; and this being
accomplished, they again return to the depths of the
ocean—or, at least, are no more seen, until the following
year. Mr. Yarrell observes, however, that the young
abound in shallow water, all round our shores in the
summer months, and that theyremain inthe mouth of the
Thames during their first autumn and winter. Perhaps
the most conclusive evidence against the migratory
habits of this fish, at least from the north, is furnished
by the fact, that they visit the west coast of Cork in
August, which is earlier than the arrival of those which
come down the Irish Channel, and long before they ap-
pear at other places much further north. Mr. Wilson
observes, that the herrings caught upon the east coast
of Scotland are much inferior to those taken on the west
coast, and more particularly to those of Loch Fine, and
other lakes of Argyleshire. We were assured of this,
also, by our friend, Robert Findlay, esq., of Glasgow,,
who further stated that they were caught at different
periods of the year. A question naturally arises out of
these facts, — May theynot be of different species? ‘* Dr.
Knox states,” continues Mr. Wilson, “ that the herrings
taken near the Firth of Forth are foul, or are engaged
in spawning ; while those of the west coast, in the same
season, have the organs of reproduction very slightly
developed *:” and he conjectures that the particular
crustaceous animal which forms their favourite food, may
exist abundantly in the bays of Western Scotland, but -
either not at all, or not in sufficient quantities, along the
eastern coast. The time of spawning, according to Wil-
son (who is probably speaking of the Scotch herrings),

* Encyc. Brit. art. Ichthyology, p. 214.

HERRING. — PILCHARD. Pgh |

“seems to vary considerably, both in the same and in
different districts ; so that we may have spring, summer,
and autumn herrings, as we know they have in some
parts of the Baltic.”* We cannot believe, until the sub-
ject is completely investigated, that all these variations
are met with in one species (Clupea harengus, fig. 58.),
which is the same
as the Cornish her-
ring, whose time of
spawning, and con-
sequent appearance
on the west coast of England, is always the same, how-
ever much they vary in the locality they choose, or in
the comparative numbers in which they appear ; in both
these latter circumstances, but not in the former, they are
proverbially capricious. The ancients do not appear to
have known either the herring or the pilchard ; although
there is a species sometimes met with on the Sicilian
coast so exactly like the latter, that even a professed
ichthyologist may take it for the same (fig. 57.) : as we
only met with it on two or three occasions, and that in
no abundance, in the fish-market of Palermo, we con-
clude it is not only rare, but does not live in shoals.
(237.) The pilchard is another fish of this family,
and a much more important one to a large part of the
population on our western coasts. According to Mr.
Couch, whose valuable and mest interesting history +
furnishes us with much of the following account, the
pilchard fishery, in the year 1827, employed, upon a
fair average, no less than 10,521 persons; while the
total amount of capital invested was calculated at
441.2157. Few persons, we imagine, would have any
idea of such enormous amounts, seeing that this fishery
is carried on in open boats, on a far less extensive scale
than those for cod on the banks of Newfoundland, or
for whales in the Arctic seas. Fishing, like all other
things, upon which the results do not depend upon

* True; but there are, according? to Mr. Yarrell, three species of her-
rings in the Baltic, and not one, as our author supposes.
f Insertedzin Yarrell’s Fishes, vol. ii. p. 96.

272 CLASSIFICATION OF FISHES,

fixed laws, is hazardous and uncertain ; and Mr. Couch,
who resides upon the spot, appears to think, that, one
year with the other, the greater part of the Cornish
proprietors of the pilchard nets (or seins) obtain no more
than their expenses; but when there is a profit, it is
usually considerable ; and thus every one hopes for a
prize. The pilchard was supposed to migrate like the
herring, but its range is now ascertained to be even
more limited ; it may be truly called a British fish, for
it seldom wanders through the Straits of Dover in a
northward direction, and its most southern range extends
only to the coasts of France and Spain, where they are
never found in any considerable numbers. It would
seem to be found in Scotland, as Mr. Yarrell considers
it the gipsy herring of that country ; but he makes no
mention of the localities in that kingdom where it has
been captured. On the coast of Cornwall it seems to
be found at all seasons: they sometimes congregate in
immense numbers in March, and, in some years, thou-
sands of hogsheads have been taken at that season; but
it is only in July that they regularly unite permanent
in society for the rest of the year: it is then that the
regular fishery commences, and it is continued until the
equinoctial gales of autumn render its further prosecu-
tion impracticable. When thus united near the coast,
observes Mr. Couch, the pilchards assume the arrange-
ment of a mighty army, with its wings stretching
parallel to the land; the whole being composed of num-
berless smaller bodies, which are alternately joining the
main body, shifting their position, and again separating.
There are three chief stations to which they resort, and
which have a separate influence on the success of the
fishery: one is to the eastward of the Lizard Point ;
the second is between this and the Land’s End ; and
the third is on the north coast of Cornwall, towards St.
Ives. It is no uncommon circumstance for one of these
districts to be full of fish, while no shoal is to be met
with in the others. It was formerly the custom to sta-
tion men on such elevated situations near the sea as

PILCHARD FISHERY. 273

would enable them to see the course of the pilchards,
and direct the fishermen, by concerted signals, how best
to surround them ; but this custom now only exists, as
Mr. Couch affirms, at St. Ives.

(238.) The fishery is carried on in common fishing
boats, with four men and a boy, who commence their
operations a little before sunset ; and the nets are drawn
in about two hours, to be again shot, or laid, as morn-
ing approaches. The number of fish thus taken in a
night’s fishing, by these drift boats, is, of course, uncer-
tain; from 5000 to 10,000 is considered moderate, but
double this number are often captured; while 150,000
fish for one boat during the season is reckoned favour-
able. There is another method employed, called sein-
fishing, to prosecute which three boats are provided :
two of these are each provided with a sein, or net; while
the other is merely used for the purposes of enabling the
head fisherman (or master seiner) to be rowed about,
and make observations. These three boats proceed in
the afternoon to some sandy bay, where they cast anchor,
and watch for the fish. The presence of the pilchards
is discovered either by the peculiar rippling of the water,
the colour it assumes, or by the leaping of the fish
themselves a little above the surface. So soon as they
are discovered, the head fisherman proceeds to ascertain
the size of the shoal, and the direction it is taking:
this done, the greatest activity is immediately used to
throw the nets in a line across the course of the fish,—an
operation which, notwithstanding the size of these nets,
is generally performed, from long practice, in less than
five minutes: the crew of the two larger boats ‘are then
employed in warping the ends together, while those in
the third boat, which is behind, by dashing and beating
the water, frighten the fish away from that part only
where they could turn round and escape. Whether the
shoal be large or small, the trouble is thus the same. The
net is then closed, and the ends laced together: if the
fish are numerous, and the sea or tide strong, the whole
is secured by grapnels; and when the evening has closed

VOL. I. T

274 CLASSIFICATION OF FISHES,

in, and the tide is low, they proceed to empty the nets.
The fish soon become so exceedingly agitated, and so
great is the accumulated force of their numbers, that,-if
it has been a full haul, the utmost caution is necessary
lest the nets should burst or sink. When brought to the
surface, the voices of the men are lost in the noise made ~
by the fish in struggling to escape, and in dashing the
water. They are taken out in buckets (or flaskets), and
only in sufficient quantities to load the boats: the rest
of the fish are left in the net until the first are salted ;
another cargo is then taken out: and thus a week may
possibly elapse before the whole of the capture is se-
cured; part being taken out every night. Nearly the
whole of the pilchards thus procured and salted are ex-
ported to catholic countries, but chiefly to those of the
Mediterranean, where we have often eaten them. The
numerous fasts of the Romish church cause an immense
consumption of fish, both fresh and salted, ail the year
round. This appears still more. striking, when it is re-
membered that, besides the extensive fisheries of the
Sardi (a small species of Clupea) and of the tunny,
which the Italians themselves carry on, and pickle, they
require immense quantities of stock-fish or cod from
Newfoundland ; and all these independently of the pii-
chards of Cornwall, the greater part of which are sent
to them. The quantities of these are so immense, that,
to prevent our being suspected of exaggeration, we shall
cite Mr. Couch’s own words. ‘“ The quantity of pil-
chards taken is sometimes incredibly large. A fisher-
man, now alive, was once present at the taking of 2200
hogsheads of pilchards in one sein; but the greatest
number heard of, as taken at one time, is stated by Bor--
jace at 3000 hogsheads. Each was formerly calculated
as containing 3500; it was then changed to 3000, and
is now 2500 fine fish; but it is scarcely necessary to
say they are not counted.* An instance has been known,

* “*Tn reference to this anecdote, Pennant has made an astounding error,
in reckoning, by mistake, 55,C00 fish to a hogshead, instead of 3500.. The
hogsheads, probably, are made of such a uniform size, as to contain this
exact number; — few more or less.” — Yarreli.

CLUPEIN&.— THE GENERA AND SUB-GENERA. 275

when 10,000 hogsheads have been taken on shore in
one port in a single day; thus providing the enormous
multitude of 25,000,000 of living creatures drawn .at
once from the ocean for human subsistence.”

(239.) We shall now enter into the detail of this
interesting group,— interesting from the importance
it possesses in an economic point of view, and doubly
so to the ichthyologist, because it is one of the most
perfect circular groups which we shall have to lay
before our readers. On this account, and from the
analysis we have been enabled to make of the whole,
we shall not merely enumerate the sub-genera, but
trace, in the principal or typical group, the series of
those links by which these latter types are united.
_ The primary divisions already enumerated, we consider
as genera; the lesser ones, consequently, we view, with
Cuvier, as sub-genera. We commence with Clupea,
under which we place all those herrings that have the
teeth either minute or altogether wanting ; the body is
also much compressed, the belly sharp or serrated,
and the dorsal fin placed in the middle of the back.
By these characters we distinguish the true herrings from
the salmon-herrings,—a name we apply to those whose
teeth are very conspicuous ; for although the aberrant
forms of Megalops have the vody or belly serrated, as
in all of the sub-genera of Clupea, yet the former have
well-defined teeth, which are not perceptible in the latter,
except, indeéd, in Thryssa, or that sub-genus which con-
nects the two groups. ‘The position, also, of the dorsal
and anal fins, hitherto so little regarded that no notice
whatsoever has been taken of them, separates Clupea
from Chirocentrus, &c.; so that the group becomes
very definite.

(240.) Commencing with the herrings of Britain,
as the true type of the genus, we observe the dorsal
fin lunated, and placed nearly in the middle of the
back ; while the ventral fin (little inferior to the pec-
toral) is directly under it: the anal, in comparison to

Tee

276 CLASSIFICATION OF FISHES.

the other types, is rather short, so as not to exceed the
length of the dorsal. All these characters are pos-
sessed by the shads, which have merely a slight emar-.
gination on the upper jaw,—a variation so slight, when
compared to numerous others in this group, that we
cannot possibly adopt the sub-genus Alosa*, or rather
Clupanodon ; for the latter name has the priority of the
former by many years. From these to Chatoessus the
transition is very gradual: the change consists in the
prolongation of the terminal ray of the dorsal, which
exceeds the others so much as to become twice or thrice
their length. These fish are generally much broader than
the herrings, as in C. thryssat: the relative size and
position of the dorsal and ventral fins are the same as in
Clupea ; but the anal fin is considerably longer, and the
mouth usually much smaller. They are all natives of
warm seas, chiefly of India. In two or three, a remark-
able deviation from the usual equality of the jaws is
observable: the snout or point of the upper jaw projects
beyond the under, — a deviation of structure which can
only be explained when we look to the singular herrings
forming the sub-genus Thryssa; this group is a very
remarkable one on many accounts. In the first place it
has an obvious alliance to the anchovies, upon which
account Cuvier has placed them following each other,—
a station which we shall subsequently show is truly na-
tural: this affinity is indicated by an excessively wide
mouth ; the edges of the jaws, particularly the upper,
or maxillaries, being sometimes armed with a single
row of short, isolated, acute teeth : these, however, must
not be considered as a primary character, because they
are sometimes wanting ; while in other species, as
Thryssa Hamiltonii Gray}, they appear distinctly de-
fined; the tip of the snout, also, is sometimes, as in the
last species, rather advanced over that of the lower jaw,
similar to what we see in the anchovies, while in others

* If sucn deviations of structure are sufficient for sub-genera, that of
Thryssa alone might furnish five or six of equal value.
+ Bloch, pl. 404. t Ind. Zool. vol. ii. pL 92. fig. 3.

CLUPEA: — THE SUB-GENERA. dal 7

the mouth is nearly as vertical as in Pristogaster. Now
the question is, How can these deviations of structure be
explained ? and what affinities do they indicate? The
advanced snout of Chatoessus nasus and latus, seems to
prepare us for the same structure in some of the Thryssa,
while the more vertical mouth of the typical species of
Thryssa plainly intimates, we conceive, that the next type
which follows in the series is Pristoguster. The cir-
cumstance, again, of some of the Thryss@ having teeth,
and others none, is at once accounted for, when we
recollect that it is through this group we pass from
the toothless herrings, or Clupea, to the salmon-herrings,
or Elops. This variation, therefore, must consequently
happen, because it is precisely in this part of the series
that the teeth began to appear. Cuvier has simply ob-
served of Thryssa, that it only differs from the anchovies
with serrated bellies by the greater prolongation of their
maxillaries. This is such an indefinite character, that
we have placed all the serrated anchovies in Thryssa,
leaving only such as have the belly smooth (like the
European) in Engraulis. Thryssa is further dis-
tinguished from Clupea and Chatoessus by the position of
the ventral fins: these, instead of being placed under
the dorsal, are situated before it, and are so small as only
to be half the size of the pectorals; thus, again, we -
are prepared for Pristogaster, where these fins totally
disappear: the mouth is still more vertical, or, rather, it
is completely so; while the dorsal fin, which is very
short in Thryssa, is now so much reduced, as, in one
instance, to be totally wanting.* The Indian species of
Pristogaster are oblong or lengthened fish, while those
of the Atlantic are short and broad: these latter, we
suspect, are the true types ; since the deep, prominent,
and arched belly, which is the chief character of these
fishes, is more developed in the American than in the
Indian species: the mouth, as just observed, is com-
pletely vertical ; and this, with its small size, and its

* Apterygia, Gray, Ind. Zool, iF
tT 3

278. CLASSIFICATION OF FISHES.

toothless jaws, separates Pristogaster both from Thryssa
and Engraulis. The absence of teeth, again, indicates
a return to the typical form of Clupea. But there is yet
another modification of form which is necessary to effect
this union. This is seen in certain herrings which have
all the characters of Pristogaster, except that they have
a remarkably small ventral fin — so small, indeed, that
it may be considered rudimentary : to these we give the
name of Platygaster. In some, as in P. verticalis*, the
mouth is completely vertical; in others it is less so
(P. Africana) ; while in some, which approach the true
herrings, the mouth (asin them) is only oblique. As the
progress of this transition from Pristogaster to Clupea,
through Platygaster, is not only particularly beautiful, but
highly important to our present purpose, we shall call
the attention of ichthyologists to the steps by which it is
effected ; for it fortunately happens that this can be done
by the a of the figures of such species as have already
been published. In the first place, let the reader turn
to the 192d plate of Russell’s Indian Fishes, where he
will find, at fig.2., our Pristogaster elongata ( Tardoore
Russ.), a fish which Cuvier himself cites as one of the
types. Now the figure immediately above this ( Platygaster
verticalis Sw.) is absolutely a Pristogaster in its form,
its vertical mouth, and its long anal fin; while it isa
Platygaster in its very small ventral fin, placed a little
before the dorsal: the position, also, of this latter fin is
more towards the middle of the back than in Pristogaster
elongata; and thus it agrees with Clupea. The very
same structure is seen in Platygaster macropthalma
Sw. i We then, in P. affinis§, get the typical structure ;
the minute ventral fin being nearer to the pectoral
than to the dorsal, and the mouth less vertical than in
the two former species. Platygaster Africana|| leads
us a step nearer to Clupea, the ventrals being rather
nearer to the line of the dorsal than to the pectoral. This

* Russell, pl. 1¢2. t Clupea Africana, Bloch, 407.
t langar loo, Russell, pl. 191. § Clupea affinis, Gray, Ind. Zool. =
II Clupea Africana, Bl. pl. 407.

CLUPEINZ. — THEIR ANALOGIES. 279

fish brings us at once to Clupea, by that singular species,
the Clupea Sinensis of Bloch, which is a true herring,
having, however, the same broad form as P. Africana:
the ventral, although placed beneath the dorsal, is yet
very small, being only half the size of the pectoral ;
while the anal, although not so long as in that fish, is
yet longer than in any other example of the typical
Clupee. Here, then, all further distinction ceases, for
we actually return to the first type we began with:
the circle of the series is closed ; and we find Clupea,
Chatoessus, Thryssus, Pristogaster, and Platygaster so
closely and intimately blending into each other, that we
scarcely can say where one begins, or where the other
ends.

(241.) Affinity, more especially when so strongly
manifested as in these instances, must always take place
of analogy. We have, therefore, laid before the naturalist
these details, before we premised any thing of the results
or inferences that may be drawn from them ; let him go
through the series himself, and then, if we are not greatly
deceived, he will be fully sensible of its representative
nature, and will hardly need the following confirmation
of it, drawn from the analogies which this group presents
to all the others we have given of this family: those, how-
ever, of the principal divisions of the Clupeine are so con-
clusive, that we cannot refrain from drawing them up.

Analogies of the Sup-cenera to the Genera of C1iv-

PEIN ZA,
hae Analogs ae
ee f LEC eons type of sole Cinceis
Chatoessus. D ore linprolonged into a nea ELops.
Thryssus. Se , opine, ee eee: OsTEOGLOssUM.
Pristogaster. aie pea a ae re yer i ODONTOGNATHUS.

Platygaster. f Ventral ape Si gerae, pul f CHIROCENTRUS.

We know not which of these two expositions are most
T 4

280 CLASSIFICATION OF FISHES.

in accordance with all we have said on the nature and
properties of natural groups, — the close affinities between
each of the types in the first column, or the beautiful
analogies which they find in the primary divisions of
the Clupeine. We here deal, in the first instance, with
facts, and facts only;— the inferences are an after
consideration ; but they are just as strong, to any mind
which understands the nature of inductive reasoning, as
the facts themselves ; and to those who do not compre-
hend the Baconian philosophy, all argument is useless.
Addressing ourselves, therefore, to the former, how
singularly does Fhryssus and Osteoglossum represent
each other by their enormously wide mouth — cleft in
an oblique direction, and armed with single detached
teeth: this structure, with the slight projection of the
muzzle, immediately reminds us of the sharks and the
pikes, which these fishes, as being representatives of the
cartilaginous type, so singularly represent, as well as
Xiphostoma, and numerous others. The completely
vertical mouth of Pristogaster, with the entire absence
of the ventrals, reminds us immediately of Chiro-
nectes by the first character, and of Balistes by the
second ; while, if we look for a repetition of these forms
under a different modification among the salmons, we
are at once presented with Sternopty#, more especially
resembling the American Pristogaster Martii of Spix.
The disappearance of the ventrals, and the great deve-
lopment of the anal fins, no less than the superior length
of the tail (or, what is the same thing, the proximity
of the vent to the pectorals), is a very general character
among the apodal fishes ; and one or both of these cha-
racters are also seen in Platygaster and Chirocentrus ;
although, as the latter, by the backward position of its
dorsal, passes into the pikes, it assimilates even more to
that group than to Platygaster. We had almost forgotten
to notice the wonderful resemblance between Chatoessus
and Megalops,—a resemblance so strong to a superficial
eye, that none but an ichthyologist would detect their
absolute difference : both have the elongated form of the

ELOPS. — THE SUB-GENERAe 981

herring, joined to the prolongation of the last ray of the
dorsal into a filament ; yet in Megalops the mouth is
strongly armed with teeth, and the belly is neither com-
pressed nor carinated ; while in Clupea the jaws are all
but toothless, and the ridge of the belly is sharp and
serrated. Clupea, in its most typical examples, repre-
sents, of course, the whole group; so that, whatever
minor divisions may, and possibly do, enter into this
circle, the prominent variations intimately correspond
(so far as the nature of the group will admit) with all
that has been said of the primary types of ichthyology,
or, rather, when these views are extended, of those of
the whole vertebrated circle.

(242.) Leaving the toothless herrings, we come now
to those which have well-defined and often numerous
teeth; the majority of which, also, are without the
sharp serrated belly which pervades the whole of the
last division. We look upon the Linnean genus Elops
as exhibiting the most typical structure of this group,
associating with it Butirinus, asa subordinate form: fol-
lowing these we place Megalops, Notopterus, Trichosoma,
and Engraulis, all of which are at once known by possess-
ing determinate or well-defined teeth.* There is not
sufficient information:on these fishes to allow of our
tracing the series so effectually as in the last; but M.
Cuvier places them close together t, and we shall now
proceed to show how intimately they are all allied. The
reader will remember that the genus Thryssa was the
only one of the last group which had a wide mouth and
distinct teeth ; and that it consequently opened a pas-
sage from Chatoessus to the true anchovies. Engraulis,
therefore, will be the first type of our present division
after leaving Clupea : we restrict this sub-genus to those
anchovies of which the common Mediterranean species

* This must remain questionable in regard to our new genus Trichosoma,
the Engraulis Hamiltonii of Gray, Ind. Zool., because, as no description of
this fish has been published, we can judge only from the figure; but as
Mr. Gray associates it with the anchovies, we may presume that its teeth
are the same.

+ Cuvier’s series is as follows : — Notopterus, Engraulis, Thryssa, Mega- .
lops, Elops, and Butirinus.

‘

282 CLASSIFICATION OF FISHES.

is the type, and thus confine it to such as have the belly
not serrated: possessing much of the general form of
Thryssa, these small fishes are nevertheless much more
slender ; the cleft or commissure of the mouth is more
horizontal ; the top of the muzzle more projecting over
the mouth ; and the jaws or maxillaries, as Cuvier well
remarks, less prolonged: the anal fin, which is long in
Thryssa, is shorter in Engraulis. M. Cuvier alludes to
an anchovy found in America, his EZ. edentatus, which
is without any teeth: we have not seen this; but as it
would appear to have a serrated belly, we should place
it as an aberrant Thryssa, forming another link in the
chain of those species which unite the toothed Thrysse
with the toothless Chatoessi. Leaving this sub-genus,
we pass on to. Elops and Butirinus, because M. Cuvier
observes of the latter, that ‘“‘the muzzle is prominent
like the anchovies :’’ both, however, differ from those
fish, in the mouth being smaller. On looking to the
figure of Elops saurus (Bloch, 393. f.1.), the ichthy-
ologist will be fully persuaded that its affinity is with
Engraulis, and that its analogy is with Clupea: its
rather wide mouth, opening horizontally ; the great pro-
longation of the maxillaries, which reach far beyond the
eye ; and the distinct teeth, establish the first of these
relations: while the size and position of the ventral fin,
as well as the shortness of the anal, render this fish so like
a true herring, that if the snout was concealed, and the
belly serrated, it would pass for such. Of Commerson’s
genus Butirinus, which we have not seen, Cuvier inti-
mates that it has all the characters of Elops, except in
having “ the muzzle prominent, the mouth but slightly
cleft, and the tongue, vomer, and palatines paved with
rounded teeth set close together.’’ Now, as there must
be aberrant species between the types of Engraulis and
Elops, we look upon these fishes, at present, as being
such ; but it will still remain a question which of these
is the true type, or, rather, which is subordinate to the
other: the form of the jaws in Butirinus would cer-
tainly lead us to place it next to the anchovies, while

ELOPS. — THE SUB-GENERA. 283

the elongated lower jaw of Hlops sawrus opens a pas-
sage to the sub-genus Megalops: in this we still have
the rounded belly of the last fishes; but from them, and
ul others of the toothed herrings, they may at once be
listinguished, on account of their possessing a dorsal fin
shaped exactly like that of Chatoessus—that is, with the
ast ray prolonged into a filament: all the species have
sonspicuous teeth, close and even upon their jaws ; and
some also have them on the palatines: the gill mem-
oranes, as in H/ops, has a great number of rays, but these
vary among the species; while in some the last ray of
the anal corresponds to that of the dorsal, in being also
orolonged into a filament: this deviation is particularly
seen in an immense species found in America, which
sometimes grows to the length of twelve feet. Other
species occur in the Atlantic; and Dr. Russell has
figured some from India. The next type in our series
is that of Notopterus Lac., founded upon a rare In-
dian freshwater fish, considered by Pallas as a Gym-
notus*, but removed by Cuvier to this family. Never
having seen a specimen, we must take for granted that
the great ichthyologist of France is correct in this view
of its affinities ; and, judging from the figures that have
been published, we join in the same opinion. The
general aspect of this fish is that of a herring; but it
differs from the three preceding types of this division
oy having the carinated edge of the belly serrated, the
ventrals remarkably small, and the anal long, narrow,
ind united to the caudal fin: this latter character we
1ave not hitherto noticed in the whole of this family,
ind for subsequent reasons it deserves great attention:
10t only the jaws, but the palatines and the tongue, are
ill armed with teeth ; those on the two former are fine,
yut those upon the tongue are strong and hooked. In
nost of the toothed herrings, the number of the branchial
ays are remarkably numerous; but in this, according to
yuvier, there is but one, which is strong and osseous.
We have now enumerated four of the types, and we
* Pallas, Spec. Zool. vol. vi. pl. vi. fig. 2. Bontius, Ind. p.78.

984 CLASSIFICATION OF FISHES.

want a fifth to enable us to return again to the ancho-
vies, with which we began our survey. We should
suppose, from theory, that such a type should present
us with some of the characters of Notopterus, joined to
others which would be more characteristic of Engraulis,
and these so combined as to present a structure altogether
peculiar: this, we repeat, would be our theoretical notion
of such atype; but is it only theoretic, or is there, in
reality, such a fish ? This question we can now positively

answer in the affirmative: the Engr. Hamiltonii of Mr.

Gray (fig. 59.) has precisely that union of characters

which we have described; and although there isnothing as
yet to guide us but the figure*, we cannot but be struck
with the belief that it represents a form intermediate be-
tween Engraulis and Notopterus: ithas the head of the
former, the snout being considerably advanced beyond
the lower jaw; while it has the long anal fin of the
latter, and this fin, moreover, is so united to the caudal,
that, like otopterus, it might have very well been
arranged, by the old authors, among the Gymnoti. Its
other characters are still more remarkable, and renders
it the most extraordinary fish in the entire family of
Clupeide. We have had, in fact, some hesitation in
placing it here, under a suspicion that, instead of being
a secondary form, as we now arrange it, among the
toothed herrings (Elops), it was, in reality, one of the
primary types of the whole sub-family. Nevertheless, it
has been our rule, in all such cases, to be guided in the
first instance by what appears the greatest affinity; and
under the impression that Trichosoma is more connected
to Engraulis than to any other of the herrings, we
* Gray, Ind. Zool, vol. i. pl. 85. fig. 5., here reduced. _

ELOPS. — THE SUB-GENERA. 285

follow Mr. Gray in approximating it to them. The
preceding cut ( fig. 59.), which is a reduction of the figure
alluded to, will show this affinity, and will supersede a
more particular description of its form, &c. The remark-
able fascicle of long, slender, and detached filaments, in-
serted close to the pectoral fin, is altogether unexampled
in this family, or, indeed, in the entire order of Ma/lacop-
teryges, and yet we have precisely the same structure in
Polynemus : in both the genera these filaments may,
perhaps, be analogous to the digitated processes in the
gurnards (Tviglide): the smallness of the pectoral fin
is another anomalous character, of which we are un-
acquainted with any other example; while the length
and attenuated form of the tail would almost lead us
to consider this fish as an anguilliform type: still the
similarity it bears to Engraulis and to Notopterus, as
we at first mentioned, cannot be got over; and, in-
fluenced by these relations, we decide upon placing
Trichosoma as the representation of Polynemus among
the herrings. This adberence to what we deem an
affinity, will not altogether destroy the analogy of the
sub-genera of Clupea to those of Elops, as the following
table will show :—

CLUPEA.

: : ELops.
Head or Analogies. Teeth conspicuous.
Dorsal fin with the upper margin

1. Clupea. f simply lunate. E Elops.

Dorsal fin with the last ray pro-
2. Chatoessus. j longed into a filament. } Megatops.
Teeth variable; mouth very :
large ; snout slightly produced. i Engraults.
Belly strongly serrated ; anal fin
4. Pristogaster. } very long, almost or quite united ¢ Trichosoma.
to the caudal.

5. Platygaster. Ventral fins almost imperceptible. Notopterus.

3. Thryssa.

We have already said sufficient on the analogies of
Clupea to Elops, Chatoessus to Megalops, and Thryssa
to Engraulis, to render any further remarks unnecessary.
These analogies are indeed so obvious, that it would be
a waste of words to bring forward additional evidence.

286 CLASSIFICATION OF FISHES.”

The same, however, cannot be said of the resemblances
between Pristogaster and Trichosoma: the horizontal
mouth of the latter is directly opposed to the vertical
one of the former; while the ventral fin, which is
altogether wanting in Pristogaster, is actually of a
larger proportionate size in T’richosoma than in any other
type of the whole family with which we are acquainted.
The analogy, therefore, if such it be, between these two
types, must, at the best, be looked upon as remote — we
may almost add questionable. We could get over this
difficulty, it is true, by substituting Odontognathus for
Trichosoma, because that type is an unquestionable re-
presentative of Pristogaster ; but this would, as we
conceive, be sacrificing affinity to analogy ; or, in other
words, would be separating Jvichosoma from those
fishes to which it has every appearance of being truly
allied, merely for the purpose of perfecting our analogi-
cal table. Another consideration has much influence on
our mind in this decision, which, as it tends to illustrate
a very important character in T7ichosoma, we shall now
lay before our readers. The character to which we
allude, lies in the long filaments near the pectoral, which
we cannot but suspect are really analogous to those pro-
cesses among the T'riglide, or gurnards. Now, it will
be subsequently shown that the whole of these genera
compose the most aberrant type of the acanthoptery-
gious, or spine-rayed, order of fishes— analogous, in fact,
to the situation we have here assigned to Tyichosoma :
to render this more apparent to the reader, we shall
here place the two groups in juxtaposition, for the sole
purpose of showing that, when so placed, these two
points turn cut to be parallel with each other.

Macroleptes. Elops.
Microleptes. Megalops.
Gymnetes. Notopterus.
Canthileptes. (Triglide, &c.) Trichosoma,
Blennides. Engraulis.

As any attempt to explain the whole of these pre-
sumed analogies would lead us from our more immediat<

CLUPEID &. — OSTEOGLOSSUM. 287

purpose, and would, in fact, be anticipating our future
exposition of the order Acanthopteryges, we shall pass
them over, and at once proceed to the other groups of
the Clupeine. vat

(243.) We have now disposed of the two most typical
groups of the family before us, and shall proceed to the
other three, which we consider are aberrant. We com-
menece with Osteoglossum, because it follows that of the
toothed herrings, represented by Elops and Megalops.
The annexed cut (fig.60.) of this singular type (O. bicir-

rhosum Sw.*) will give au accurate idea of its genera
structure. In the continuity of the dorsal, caudal, and ana:
fins, as well as its lengthened tail, it reminds us of the
eels ; while its large oblique mouth, and very short muzzle,
present some resemblance both to Laurida and Thryssa:
the teeth, however, although numerous and sharp, are
not very unequal in their length. Cuvier remarks o!
this type, that it has many relations with Sudis; but he
has not stated what these relations are, and to us they
appear few and very remote;—the one is nearly cylin-
drical, the other much compressed ; and their general
physiognomy is very different. There is no fish, yet
discovered, where the scales, in proportion to the size of
the body, are so enormously large asin this type; and
Spix mentions that they are very hard. Cuvier remarks
that the tongue is osseous, and singularly rough, from a
multitude of short, straight, and truncated teeth, with
which it is covered ; so that, as he conjectures, it serves
like a rasp to reduce fruits to a pulp, or to express their

* Ichnosoma bicirrhosum of Spix, pl. 25.

288 CLASSIFICATION OF FISHES.

juices: upon what authority, however, this is made a
frugivorous fish does not appear, and the supposition
seems to us highly improbable. The great size of the
pectorals, when compared with the ventrals, is worthy
observation ; as also the interruption (if it really is so
in nature) between the dorsal and the caudal: this we
have preserved in the figure, which is reduced from that
given by Spix; but we suspect that in a perfect state this
interval is filled up so as to render all the three fins
continuous.

(244.) The next genus which we place as aberrant,
is Odontognathus, arranged by Cuvier close to Pristo-
gaster: that there is some sort of relation between the
two is very obvious; but whether this is of strong ana-
logy or absolute affinity we do not know. The only
figure existing of this strange-looking fish is that of
Lacepede’s, which Cuvier says has been taken from a
badly preserved specimen: this we can readily believe,
since the maxillaries are represented as assuming the
appearance of two porrect horns in front of the mouth,
—a structure which no fish can possibly have; and yet
the size and form of these maxillaries appear to be so
different from those of Pristogaster, that we are induced
to think the resemblance is only analogical.— the more
so, as these maxillaries are stated to be “‘ armed with
small teeth directed forward,’ of which no instance is
known to exist in Pristogaster, where the mouth is very
small, and the teeth altogether wanting. The mouth
must be completely vertical; the anal fin is almost
united to the caudal ; and the dorsal is so small and
brittle as to be ‘‘ almost always obliterated.”’ Only one
species is known, which is rather small, much compressed,
and comes from Cayenne. In the third and last aber-
rant division we place two sub-genera, Chirocentrus Cuv.
and Hyodon: these, although somewhat different from
each other, may be immediately distinguished from all
the other types of this family by their numerous teeth,
but more particularly by the very backward position of
the dorsal fin, which is almost as near the caudal as in

CLUPEINE. — CHIROCENTRUS. — HYODON. 289

the pikes: it is, in short, by these genera that we con-
ceive the sub-families of the Clupeine and the Esocine
are united ; they are herrings in their shape and general
aspect, but pike in the position of their dorsals, and
almost in the armature of their mouth. Chirocentrus
is represented by some few fishes found in the East
Indies: one of these is particularly described and figured
by Russeli, under the name of Wallah (Chirocentrus
Russellii Sw.). According to this author, it has a long
knife-like body, which is entirely destitute of scales ;
the mouth is large, very oblique, and with the lower
jaw longest; in the upper jaw are four long, projecting,
lanceolate teeth, with many small marginal ones behind ;
in the under jaw they are long, distant, and reflected ; the
tongue is small, ovate, and smooth*; the palate is also
smooth: in others the tongue, according to Cuvier, is
“bristled with pectiniform teeth.’’ The body of both
these is much lengthened ; the belly sharp, but not ser-.
rated ; the dorsal close to the caudal; and the ventrals
extremely small.
(245.) The ae ercikeat clodalis Le Sueur, fig.61.)

He
cin Hg ‘i ay
a ii (K Ha ae
( i (
Ae
rite aany Cite

approximates to the last by having the dorsal situated
towards the caudal ; but the form is more like that of the

* Cuvier cites this very description of Russell’s Wadlah for the “single
species that is known” of this type; the teeth of which he describes as
follows: ** The intermaxillaries and maxillaries are both furnished, as well
as the lower jaw, with a range of strong conical teeth; two of which, in
the middle of the upper range, and all below, are of extraordinary length :
the tongue and branchial arches are bristled with pectiniform teeth.”? It
is clearly impossible that this can be the species described by Russell, which
has the “ tongue smooth.” It is more than probable, also, that the other
synonyms of Cuvier refer to different species, and that several, in short,
exist in the Jndian seas. It is such variations as these, in the teeth of fishes,
which show how often they are merely specific —not generic— characters.

VOL, I. U

290 CLASSIFICATION OF FISHES.

ordinary herrings; and Cuvier, while he places it next to
Chirocentrus, assimilates it to the salmon-trout, by the
“ hooked teeth on the jaws, the vomer, the palatines,
and the tongue.” The best account of these curious
fishes. however, will be found in Le Sueur’s own words.
The hyodons, as he observes, “ inhabit the river Ohio
and the lake Erie, where they go under the popular
name of herrings. They have, in effect,’’? continues our
author, ‘‘ much resemblance to Clupee in their co-_
lour, their large eyes and scales, and the compressed
form of their body, —and with which genus they may be
confounded on a superficial view: but it is easy to distin-
guish them by the absence of the carinated abdomen ; by
their extremely short intermaxillaries-and maxillaries,
which are articulated together; and by every part of the
mouth being strongly toothed, as in the salmon family of
Cuvier. They have in a great measure the habits of
these last; as, like them, they appear to prey upon living
animals, particularly insects, which they take on the
surface of the water. The stomachs of several of these
fishes, which were examined, were filled in the spring
with Scarabei and the larva of Ephemera; the perfect
insects of the latter, at that period, being observed, in
immense multitudes, swarming over the surface of the
Ohio. The want of an adipose fin in our fishes,” con-
cludes this excellent zoologist, ‘‘ excludes them from the
genus Salmo: by their teeth they seem to approximate
to the genera Chirocentrus and Erythrinus of Cuvier
and Gronovius, and the Amia of Lacepede: but they
differ from the first by the vomer being furnished with
teeth ; from the second, alsc, by the teeth; and from
the third by the pectorals, the dorsal, the teeth, the gill-
covers, &c.” He then describes two species with great
accuracy, H. tergisus, and clodalis (fig. 61.). For
the present we follow Le Sueur and Cuvier in associ-
ating this genus with Chirocentrus, not from any convic-
tion that their resemblance is one of affinity, but, until
the Salmones are better understood, and the sub-genera
naturally arranged, we think it preferable not to make any

CLUPEINE. — HYODON. BG

further innovations on the existing dispositions of these
groups. We have a strong conviction, however, that
Hyodon will hereafter be brought in among the aberrant
types of the Salmonine, with which, in every thing but
its single dorsal fin, it bears, both externally and inter-
nally, the strongest resemblance, —a resemblance which
extends to its habits, food, and the fresh waters to which
both are more especially appropriated. If we ventured
a conjecture upon its true station, we should place it
between Osteoglossum and Xiphostoma. 'The absence
©r presence of an adipose fin is considered the sole dis-
tinction between the Cyprine and the Salmonine, and
between the latter and the Clupeine—and it is, doubt-
less, one of the most typical characters that we yet know
of ; but at the confines of each of these groups we must
lock for considerable variation in this respect, because
every naturalist is fully aware that when nature is about
to quit one type of form, she modifies her structures in
such a way that many of the strongest characters she
had been employing are lost, and are exchanged for
others which only exist in their full perfection in the
next’ group to which she is advancing. Besides this,
we have several instances of two closely allied genera, in
which one has two dorsal fins, and the other only one.
The most striking of these that at present occurs to our
mind is in the case of Loricaria and Hypostoma among
the Siluride,—two types which Cuvier places only as
sub-genera, although the latter has an adipose fin,
while the former has none. We have before expressed
our suspicion that some of the sub-genera of Brazilian
salmon, as Prochilodus Agass., and even Anodus, may
eventually be found to enter among the carps; and for
the above reasons Hyodon would not be absolutely ex-
cluded from forming an aberrant group among the sal.
mons, merely because it has not an adipose fin. How-
ever this may be, we feel perfectly satisfied on the
situation we assign to Chirocentrus, whose whole struc-
ture is intermediate between the herrings and the pikes.

(246.) Having now closed our survey cf the entire

u 2

292 CLASSIFICATION OF FISHES.

family or sub-family of Clupeine, we shall just advert
to the analogies which seem to result from the disposition
we have made of the various groups, by instituting a_
comparison between the

Analogies of the CLUPEINZ and the SALMONIDS.-

Primary Types of the ac Primary Types of the
HERRINGS. LEIS Ee SALMONS.
Body much compressed ; mouth :
Clupea. ) small, obliquely vertical 2 boa ¢ Pee or Ser-
sharp and serrated. pilot eG
Mouth larger, more horizontal ;
Elops. ; belly (typically) smooth ; body ¢ Sedma
not much compressed.
. Mouth excessively large; teeth
Osteoglossum. strong, numerous; head com | Xiphostoma =
pressed.
ay 1 ical ;
Odontognathus (?). f eu pede BEBE) EI ES Be Sternoptyz.
Chirocentrus. Dorsal fin close to the caudal. Sudis (?).

The reader will remember that we have considered
Lacepede’s genus Serrasalmo as one of the typical re-
presentatives of the American salmons ; and it is interest-
ing to see how much they accord with that of Clupea:
both have the body greatly compressed, and much
broader than any of their congeners ; both have the mouth
small and obliquely vertical—at least such is the direction
of the lower portion of the jaws in Serrasalmo; and
both have the very unusual character of the belly being
sharp and serrated. In £lops and Salmo,-on the other
hand, the form of the body is more lengthened ; the
belly is neither sharp nor serrated ; the mouth is cleft
almost horizontally ; and the very aspect of the two
groups show an intimate resemblance. This, indeed,
cannot be said of the external form of Osteoglossum and
Xiphostoma; and yet there are points which intimate a
similarity of structure, and, consequently, of habits: each
has the largest mouth in its own circle ; the teeth in both
are slender, unequal, and numerous: but the truth is, that
each contains very few species; so that, as there are no
very aberrant examples, there are no intermediate links of
connection. We before remarked, that Cuvier intimates
a resemblance between Osteoglossum and Sudis ; and we

ESOCINE, OR PIKES. 293

at first imagined this was analogical, because both have
the dorsal and anal fins very close to the caudal, and some
of these fins, in Osteoglossum, are actually united ; but
then the great breadth and depression of the head in
Sudis is so directly opposed to the narrowness and com-
pression of this part in Osteoglossum, that we incline
‘more at present to assimilate the latter to Xiphostoma,
and the former to Chirocentrus. The question, how-
ever, may be left open to discussion, because, at present,
Sudis is such an isolated form, notwithstanding its ge-
neral similarity to Erythrinus, that its other affinities re-
main uncertain. The analogy of Odontognathus to Ster-
noptyx is more satisfactory ; and will equally hold good,
whether it be ultimately retained as a primary type of the
Clupeine, or as one subordinate to Pristogaster. These
three last analogies, it must be remembered, regard
aberrant types ; and such types are always more varied —
more disconnected, as it were, among themselves — and
therefore more difficult to determine by a graduating
series of connecting species, than either typical or sub-
typical groups, wherein the intermediate modifications of
form are always more numerous.

(247.) The Esocin a, or pikes, succeed the herrings :
they constitute, in our present arrangement, a sub-
family ; and although, in point of numbers, they appear
much more restricted than either of the three families we
have already disposed of, the variations in their structure
are so remarkable, and the gradations between them so
few, that their natural arrangement is proportionably as
difficult, and an artificial one is easy. Where the line
of continuity is lost or not discovered, the different forms
will appear isolated, and will then furnish the most po-
sitive characters ; but when these forms are modified in
a variety of ways, so as to present ramifications of dif-
ferent relations, we have some clue to the natural series.
The most prevalent character of all the fishes that have
been classed among the pikes, is that of the dorsal fin being
placed very far backward, so as to be close to the caudal ;
while the anal fin is immediately below it. The unusual

u 3

204 CLASSIFICATION OF FISHES.

disposition of these fins influences that of the others:
the ventrals are thus placed in the middle of the body,
or about half way between the pectoral and the caudal :
the latter is always forked ; and the former, in general,
pointed. The whole of these fishes are excessively vo-
racious and destructive to others: hence Lacepede has
justly said of the common pike, that it is the shark of
our ponds and rivers. We accordingly find the mouth to
be particularly large, the snout often greatly lengthened,
and the teeth, in nearly all instances, numerous and sharp. -
There is no instance of a second dorsal or adipose fin,
as in the salmons ; or of the belly being sharp or serrated,
as in the herrings. As to other parts of their structure, it
may be mentioned that the margin or edge of the upper
jaw is formed by the intermaxillary bones ; or, at least,
when this is not the case, the maxillaries are without
teeth, and partially concealed. Excepting the common
pike, and a few others nearly allied to it, the whole of
the remainder are marine fishes.

(248.) We have already said that the connecting
links between the different genera placed by authors in
this group, are very few: nevertheless, by the help of
these, and of the mode of variation more clearly to be
traced in the other divisions, we may arrive at more
definite notions as to the probable cause of the natural
series than would at first be expected. We have already
seen in Chirocentrus, the last genus among the herrings,
that the fins are placed almost precisely in the same
situations as those of the pikes, although the sharp belly
and other characters assimilate those fishes to the Clu-
peine. Now this link in the chain is most important ;
since it not only determines the connection between
these two sub-families, but also guides us, in some
measure, to look for that group among the pikes which
shows the nearest affinity to the Clupeine. Cuvier ap-
pears to have had no hesitation in arranging the Exoceti,
or flying fish, with the Zsocine, close to those long-
snouted genera which comprehend the gar-fish. There
can be no doubt, however, that the true types of the whole

ESOCINA, OR PIKES. , 295

of this sub-family are the fishes represented by the Lsow
belone of Linneeus, —a rank which they derive from their
exact analogy to the Xiphiane, or sword-fish, in the cor-
responding circle of the acanthopterygious order; and
therefore, correctly speaking, the family name of Esox
should have been retained to this most typical group.

t is one of the beauties of the system of representation,
that the typical forms of an extensive circle, in cases of
this sort, may always be determined by the simple and sure
method which nature has herself taken of pointing out
her own analogies. The name of Hsox, however, being,
by long usage, so universally affixed to the freshwater
pikes, we shall so retain it, distinguishing the gar-fish
by that of Ramphistoma, long ago given them by Rafi-
nesque. Following these three genera we shall place
those of Stomia and Chauliodes, whose obtuse mouth we
have been in some measure prepared for by Esov. In
regard to the fifth or last type, much uncertainty pre-
vails: the genus Diplopterus of Mr. Gray may possibly
be the true one; and yet the great elongation of the jaws
in Lepisosteus, and its depressed muzzle, seems to bring
it much nearer to the gar-fish and the pikes than to any
others of this order. Cuvier, indeed, places it at the
end of the soft-rayed families, intermediate between
Osteoglossum and Polypterus, but without venturing to
intimate any supposed affinity with either ; and it is plain
that he places these three together, not as having: any
real connection, but as being in some measure related
to the groups that precede them. If such an accom-
plished ichthyologist, with all the materials of the
French Museum at his command, could not determine
the natural station of this singular genus, we may well be
pardoned for being equally unsuccessful.

(249.) These extraordinary creatures, the fiying fish,
forming the genus Hxocetus Linn., will first be noticed,
both as to their habits and their classification. By
Linneus they were placed much nearer to the herrings
than they have been by Cuvier, who arranges them in a

u 4

206 - CLASSIFICATION OF FISHES.

different family, close to the gar-fish (Ramphistoma*
Raf.). That they have an intimate relation to both of
these groups, is very obvious ; but it is not so easy to de-
termine to which they naturally belong. If we regarded
their sharp carinated body, their small obliquely cleft
mouth, the peculiar form of the maxillaries, the small-
ness or total absence of the teeth, and even their large
deciduous scales, we should at once place them among
the herrings ; while, if we attach more importance to
the backward position of the dorsal fin, and the singular
carinated lateral line placed close to and on each side
of the belly, we should adopt Cuvier’s idea of arranging
them close to Hemiramphus,—more especially as the pre-
ponderance of characters are certainly in favour of these
fishes belonging to Esocing, but at that extreme point
where they pass into the herrings. Some very import-
ant analogies, also, will result from this arrangement,
which, as it was made by Cuvier, becomes totally unin-
fluenced by such considerations. As Chirocentrus was
the last type among the herrings, so does Exocetus be-
come the first among the pikes: from the absence of
intermediate or graduating forms, the connection is not
very obvious ; and yet, when we look to the profile of
their heads, the depression of the crown, the sub-vertical
direction of the mouth, the sharpness of their belly,
and the position of their dorsal, anal, and ventral fins,
we see a manifest relation between them ; although in
one the teeth are highly developed, while in the other
they are almost or altogether wanting.

(250.) The Exoceti, or flying fish (Exc, evolans Linn.,
fig.62.), however, are chiefly remarkable for the enormous
development of their pectoral fins, by the aid of which they
are sustained in the air during a short time—when they
have more the appearance of birds than of fish ; so that if

* Esox belone Linn. The impropriety of calling this group by such a
name as delone, need not be pointed out. M. Cuvier’s names are in
general so well chosen, and so classically constructed, that we always feel
repugnance in proposing to substitute others for the very few which are
faulty. In the present case, however, as in that of Laurida, M. Cuvier’s

names have not even the claim of priority, for Ramphistoma was proposed
seven years before that of Belone.

THE FLYING FISH. 207

we have birds which swim, these are fishes which fly. A
great deal more has been said and written on this interest-

—————=
—SS>>

VX)
fy) iy oD Wee
a) Yay

sal we
PAY WAR ALIAS

——
A ii BOD vA a

JY)

ing subject than it would be necessary to repeat, but for the
purpose of rectifying error ; and having had numerous
opportunities of witnessing these fishes in their native
seas, what we shall now state will be the result of per-
sonal observation. It has been said, indeed, that the
name of flying fish, given to this group, is an error, be-
cause they only leap into the air, where they have not
the power of sustaining themselves at will; but this is a
mere quibble. An animal which can make its way in
the air, in the general acceptation of the word, certainly
flies, although that flight is not sustained by the motion
of the members, and ceases when there is no longer
power for continuing the exertion. The idea that this
power ceases when the membrane which connects the
pectoral rays is dry, is very possible, but we do not
think it can be determined as a fact. Thirty seconds is
the longest time absolutely mentioned for these fishes to
be out of the water*; and although we never precisely
ascertained this, we do not think that the time is under-
stated. Now, as the atmosphere under the equinoctial
line, where these fishes most abound, is almost always
excessively damp and moist, it is highly improbable that
the membrane of their fins would dry so very rapidly
as in half a minute after they have been saturated, as it
were, with sea water, immediately on the rising of the
fish. It is said, also, that the fins are merely used a

* Bennett’s Wanderings, vol. ii. p. 30.

208 CLASSIFICATION OF FISHES.

parachutes, and do not, as in birds, propel the fish for-
ward by repeated motion: this, again, admits of doubt:
the flight of these fishes, although short, is very rapid, —
almost as much so as that of a swallow; and every one
knows that these birds will go over a good extent of
ground with little or no beating motion of the wings.
In crossing the line, in the year 1816, we were very
anxious to ascertain this point in the economy of the
flying fish ; but although we had them before our eyes
almost every quarter of an hour for a week, their flight —
was so rapid, that, at the nearest distance they ever were
to the ship, we found it as utterly impossible for the
eye to determine this question, as it is to see the vibra-
tion of the wings of a fly. Our impression is, that this
act of flying is effected in two ways: first, there is a
spring or leap, by which the fish is raised out of the
water; and then, that the pectoral fins are spread, and
are employed to propel the fish in a forward direction,
either by a few flappings, or by that motion which is
analogous to the skimming of swallows. That this could
not be continued when the moisture of the fins began
to be absorbed, is quite obvious ; but we think that it is
only discontinued until the fish suppose themselves to
be out of danger. It is quite true that they have not
the power of elevating themselves in the air in an un-
dulating direction, as we see in swallows, — the course of
their flight being always that of a very slight arch, the
height of which, we believe, varies with the species. We
have frequently seen great variation in the height which
flying fishes ascend: those towards the equinox of Ame-
rica have a low flight; so that, although innumerable
flocks rose round our vessel in all directions, not one
ascended sufficiently high to fall into it. On the other
hand, it has been mentioned, that they have fallen into
ships which were from fourteen to twenty feet above the
water: it may be as well to observe, that this could
only have originated in the flying fish having been im-
peded in their course over the vessel by the ropes or
other tackle ; because their return to the water is always

THE FLYING FISH. 299

very gradual, so that the last ten or fifteen yards of
their course is almost parallel with the waves. It is
seldom that more than 200 or 300 rise up at once, so
that it is an exaggeration to magnify this number to
thousands. Neither do they fly in all possible direc-
tions ; for their flight, with very little variation, is always
straight forward, and they only diverge a little to the
right and left when they are about to re-enter their more
natural element. Sometimes, indeed, they fly off in an
obliquely angular direction from that which they at first
took. This is an important fact, because it proves that
this flying is not merely effected by a leap, but by the
action of the fins and tail, just as these members are
used to influence the course of birds: the forked struc-
ture of the caudal fin has an obvious connection with
this power, but in what manner the pectorals are used
we are altogether ignorant. We have no doubt that
more than double the number of species of Ewocetus
really exist above those that have been described; and
we have to lament the loss of three, at least, that
formed part of our Brazilian collection : those of India,
the Mediterranean, and the Pacific, are probably all
different. Some of these, forming our genus Cypsilurus,
are singularly characterised by the possession of simple
or lobed cirri or barbels proceeding from the lower jaw.
We here insert the cut of one of these (C. appendicu-
latus, fig. 03. : described by Wood* as inhabiting the
American seas.

rei
pee

* Journal of the Acad. of Nat. Sciences of Philadelphia, which for
brevity we cite as Amer. Trans. vol. iv. p. 283. _
\

300 CLASSIFICATION OF FISHES.

(251.) The genus Ramphistoma, or gar-fish, have but
a very slight connection, so far as their external form is
concerned, with the flying fish ; and yet there are some
points of anatomical coincidence, which, in the absence
of intermediate forms, bring them together. They are
long, narrow, and compressed fishes, at once known by
the excessive prolongation of their jaws, which in this re-
spect are precisely analogous to those of the sword-fish
(Xiphine), which represent them, in fact, in the opposite
circle of the acanthopterygious or spine-rayed fishes.
Although not a numerous group, the gar-fish appear to
contain three or four sub-genera, —two of which, Ram-
phistoma, properly so called, and Scomberesox, occur on
the British coast. The first of these is the Esox belone
of Linneus,—a fish very often seen in the London mar-
kets in the spring, and remarkable for the beautiful
green colour of its bones: it seldom exceeds two feet
long. No ichthyologist seems yet to have been suc-
cessful in ascertaining the precise nature of its food,
although there can be no doubt, from the structure of
the teeth, that it devours small fish. Mr. Couch,.who
has observed its habits, informs us*, that ‘it swims
near the surface at all distances from land, and is not
unfrequently seen to spring out of its element; its
vivacity being such that it will for a long time play about
a floating straw, and leap over it many times in succes-
sion. When it has taken the hook, it mounts to the
surface, often before the fisherman has felt the bite;
and there, with its slender body half out of water, it
struggles, with the most violent contortions, to escape :
when newly taken it emits a strong smell.”” The gar-fish
seem widely dispersed, for they occur in the Atlantic
and in Tropical India; but we found no species in the
Mediterranean. Of the sub-genus Scomberesox, also,
only one species is British: it chiefly differs from the
last in having the hinder portion of the dorsal and anal
fins divided into those finlets which are so conspicuous

* Yarrell’s Fishes, vol. i. p. 392.

ESOCINE. — HEMIRAMPHUS. 301

among the mackerel. The sub-genus Hemiramphus Cuv.
is another subordinate type of this genus ; and its sin-
gular mouth is perfectly unique among fishes: the lower
jaw is of the same proportionate length as in the gar-
fish, but the upper is so short that it appears as if
broken off almost at its base. It would be highly inter-
esting to know those particular habits which require a
structure of mouth so different from all other fishes ;
but this remains unknown. Various species are found in
the tropics of both hemispheres: one of these, the Hem.
Brasiliensis? (fig. 04.), we have examined in a fresh

ATT TOME

state, and thus been able to detect a singular peculiarity,
not yet noticed. On both sides of the lower jaw (which,
in its depressed shape, resembles the upper one of a
saw-fish deprived of its spines) is a thin membranaceous
fringe or skin, very delicate, and which is half the
breadth of the jaw itself: it is quite clear that this jaw
is not used either to secure the food upon which this
fish may feed, for the point of it is quite obtuse ; neither
can it be employed to thrust into the sand or other
substances, for then this membrane would be destroyed
immediately: besides, the circumstance of the other
gar-fish swimming close to the surface of the water,
shows that their food is not found at the bottom ; and
this also must be the case with Hemiramphus, whose
shape and general structure, in every thing but its mouth,
is perfectly the same as Ramphistoma. Now itis a re-
markable circumstance, that we have a genus of birds,

302 CLASSIFICATION OF FISHES.

equally unique in its own class, where the mouth is
similarly constructed: in Rhyncops, or the skimming —
terns, the upper jaw, in fact, is considerably shorter than
the lower ; and these birds skim along the surface of
the sea to feed upon those minute animals, which are
only to be found there: this well-authenticated fact,
which we have elsewhere enlarged upon, throws consi-
derable light upon the probable habits of these fishes,
which every induction of reason leads us to believe
habitually feeds much in the same manner, and on the
same description of animals — that is, on such as float
upon, or swim very near to, the surface.

(252.) Quitting the gar-fish for the present, we pass
to the fluviatile pikes, to which belongs the well-known
fish of that name (Esow lucius Linn.), so common in our
lakes and ponds. The boldness and voracity of this
fish is proverbial ; it not only gorges itself with all others
that it can swallow, but attacks other animals, as if from
mere savageness, or as if, accustomed to overcome and
devour every cther fish that it was in the habit of
encountering, it had lost the instinct of discrimination.
Mr. Yarrell has collected several remarkable instances
in proof of this, to which we must refer the reader.
It has been known not only to seize ducks, water hens
(Fulica), and other aquatic birds, but even to make
unprovoked attacks upon man, and retain its hold with
all that pertinacity which would seem as if it could
conguer and devour a being ten or twenty times larger
than itself. There is a story, often repeated, told by
Gesner, that a pike was once caught at Heilbrun, in
Germany, which had a brass ring attached to it, inti-
mating that it was put into the lake in the year 1230;
so that, being captured in 1497, it must have been 207
years old. One would have been incredulous on this
subject, but Gesner further asserts that the skeleton,
nineteen feet in length, was long preserved at Manheim
as a great curiosity. It would be well worth the trouble

- of inquiry, of any fragments of this gigantic monster are

ESOCINZ. —- LEPTODES. 303

yet in existence, or if any records regarding it exist at
that place : for ourselves, we confess our entire disbelief
that such a pike, and of such an age, ever existed. The
largest, we believe, that has ever been captured in this
country, was the famous one caught by colonel Thornton
in one of the Scotch lakes, which measured exactly four
feet four inches from eye to fork: the colonel says, that
on opening his jaws, ‘‘ so dreadful a forest of teeth, or
tusks, 1 think I never beheld.” The shape of the pike
is much more like that of an ordinary fish than of the
Ramphistome: the snout and jaws, indeed, are rather
lengthened, but they are blunt, depressed, and large ;
upon opening the mouth, which is very wide, it appears
to be absolutely lined with teeth of all sizes, covering
the jaws, palate, and vomer, or throat: the size and de- '
pression of the head must be particularly noticed, since
it far exceeds that of any other fish in this order, and
will be subsequently adverted to. It is fortunate for
other fishes, that there are very few species of pikes ; and
the wonder really is, how any others can live in the
same waters with such a depopulating monster. It has
been ascertained that eight pikes, of about five pounds’
weight each, consumed near 800 gudgeons in three
weeks.

(253.) The genus Leptodes* is the next form among
the pikes to which we assign a primary rank. Very little
has been published of this extraordinary fish, and it
would appear that even Cuvier himself had never seen it ;
while the only figure existing is the rude and ill-drawn
one of Catesby, which has been copied repeatedly into
other works. Among the numerous fishes we collected in
the Mediterranean was one of these; but it has shared the
fate of nearly all the others, and seems to be no longer
in existence. Fortunately, however, a coloured drawing
was made from the fresh specimen, which is reduced

+ The name of Chauliodus, given to this type by Schneider, is particularly
expressive, but unfortunately it had long been used to designate an equally
remarkable genus of neuropterous insects by Latreille ; so that we have no
other alternative than to propose another — Leptodes

304 CLASSIFICATION OF FISHES.

in the annexed cut (fig. 65.) ; and the head is also repre-

sented, but of the natural size (fig. 66.). These figures will
convey to the reader
a much better idea
of this extraordinary
creature than any
detailed description.
The head, but more
Bs especially the mouth

S35 and teeth, are enor-
ZT mously dispropor-
tionate to the size of
body ; and, indeed, the latter members are larger than
in any other fish yet discovered. There seems also suf-
ficient difference between the species known to Catesby,
and the Mediterranean one now figured, to justify us in
believing they are different species: the absence of a
terminal spoon-shaped membrane at the tip of the first
dorsal ray in the former may be accidental; but this
cannot be said of the difference in the size and proportion
of their scales, or the shape of the caudal fin, which is
lunate in L. Sloanii, but cleft to the centre of the base in
L. Siculus ( fig.65.): in the latter, also, the dorsal fin is
inserted much nearer to the pectoral than it is to the ven-
tral; but in Sloanii it is just Intermediate between the
two. These fishes are very rare, and are only seen, like -
Sternoptyx,Gymnetrus, and other strange-looking genera,
after violent storms, which have agitated the bottom of
the sea, and cast these delicate fishes upon the beach. In
the course of five years we never met with more than

ESOCINA. — LEPISOSTEUS. ——ITS ANALOGIES. 305

two individuals lying dead upon the Isthmus of Messina.
The sub-genus Stomias is evidently of this type, but dif-
fers from it in having the dorsal fin situated as in all the
other pikes. Risso describes two species inhabiting the
Mediterranean, neither of which we had the good fortune
to meet with.

(254.) The last genus which we bring within the con-
fines of this family is Lepisosteus* Lac. (fig. 67.) There

ean hardly be any doubt that this remarkable fish belongs
to the pikes, although Cuvier, with singular infelicity of
arrangement, places it immediately after Osteoglossum.
The only question seems to be as to its more immediate
allies, and the rank we should assign to it. In its form,
and in the disposition of its fins, it immediately reminds
us of the gar-fish ; but then the body, which is nearly cy-
lindrical, is entirely covered with diamond-shaped scales
as hard as stone: the edges, or outer rays, of all the fins
are defended with spine-like scales, quite analogous to the
spined fins of the Siluride, while the muzzle, although
long, is broad and depressed: both jaws are internally
covered with numerous rasp-like teeth, with a row of
larger ones intermixed, and placed at their edges. There
is no gar-fish yet discovered having any thing like this
structure, and we therefore view Lepisosteus as a pri-
mary rather than as a secondary type among the Hsocine.
Rafinesque shortly describes several species as inhabit-

* The figure given by Lacepede, and copied in the Ency. Brit. pl. 305.
fig. 4. is entirely erroneous ; for it is, by some strange mistake, turned upside
down, so that the pectoral and ventral fins seem to be a second and third
dorsal fins: that of Bloch, pl. 390., is a tolerably good representation.

VOL. I. x :
@

306 CLASSIFICATION OF FISHES.

ing the fresh waters of America * ; but as no figures of
them have been published, or no specimens have reached
England, we know, as yet, very little about them.

(255.) The other genera or sub-genera placed by
authors in this family, will be noticed in the systematic
part; and now, having selected what appear to be the
most prominent types of form, we will place these toge-
ther, and ascertain what analogous resemblances to them
can be found in other groups. In the first instance, we
shall compare the whole of the tribe or family of Sal-
monide with the circle (subsequently to be made out) of
the Scomberide, in order to show the relation of the pikes
to the sword-fish.

Sub-families of the : Types of the
Salmonide. Analogies. SCOMBERIDZ.
2 ade Mouth large; teeth strong; belly :
SALMONINE. f mah compressed. > f Tiymnus.
Saree ee cee aie small; belly f Scniher
Mouth large; jaws excessively
Esocinz. developed, one or both being ¢ Xiph
very long and pointed.
= Mouth very small ; muzzle greatly ? ,. .
MorMYRINZ. lengthened. = Fistutaria.
CYPRINZ. ? Lepidosaurus (?).

At present we must confine ourselves to one of these
analogies, or that between the pikes and the sword-fish,
than which nothing can be stronger. The analogy of
Mormyrus to Fistularia, both with their long snouts,
yet little mouths, is also sufficiently evident; and there
is a remote resemblance between the two first groups :
the relation, however, to the two latter we cannot make
out, but this point will be returned to hereafter.

(256.) We shall next compare the foregoing divisions
of the Esocine with the malacopterygious families ;
chiefly with a view of showing that Lepisosteus is the
representation of the Siluride in its own family.

* Ichthyologia Ohiensis, or Nat. Hist. of the Fishes inhabiting the River
Ohio and iis Tributary Streams. Lexington, 1820.

ANALOGIES OF THE PIKES. 307

Circle of the Types of the

Malacopteryges. Analogies. Esocinz,
SALMONIDE. Typical. Ramphistoma.
PLEURONECTIDE. Sub-typical. Esoz.

Dorsal fin placed close to ae
Gapipz. head ; mouth very large. Leptodes.
Snout depressed; body covered
SILURIDZ. with osseous plates or sates; { epsosten (2)5
fins with spines on the first ray.
CoBITIDz. Mouth very small. Exocetus.

We thus find Lepisosteus to be among the pikes, what
Loricaria is among the silures; because, as the latter
genus is pre-eminently typical, it of course stands as the
representative of its own family. We must remember,
also, that Lepisosteus is the only pike, or, indeed, the
only genus yet discovered in the whole of the Salmonide,
that has the first ray of the fins spinous; and this pe-
culiarity of structure makes them representations also of
the cheloniform fishes, or the Plectognathes —the types
of which, as in the Balistide, have the first dorsal ED)
almost always armed with prickles.

(257.) Lastly, we may compare the divisions of the
pikes with those of the herrings; for although the ana-
logies, in one respect, cannot be made out, the others are
very observable.

Genera of the Genera of the -

Pikes, Analogies, HEERINGS.
f Body considerably compressed st
TOMEI, 1 teeth minute or none. * 5 Clupea.
Esoz. Body rounded ; teeth strong. Elops.
Leptodes. MGR ee erees and f Osteoglossum.
Lepisosteus. ? Pristogaster.
Exocetus. Belly Hee ssed ; dorsal fin n ee Chirocentrus.

It is the total absence of all resemblance between Le-
pisosteus and Pristogaster which makes us believe that
the former enters as a sub-genus among the gar-fish, or
the Ramphistome, and that the real type between Hzo-
eetus and Leptodes has either not been discovered, or is
unknown tous. This, however, has not the least effect

ee

308 CLASSIFICATION OF FISHES.

in destroying the claim of Lepisosteus to be a tenuiros~
tral type, or that which, in ichthyology, is to represent
the primary order of Plectognathes ; for whether it be
placed with the gar-fish, or stand by itself, as we have
for the present arranged it, it becomes the most aberrant
in either group, and thus possesses all the analogies we
have more especially pointed out. There is a little fish,
indeed, published by Mr. Gray under the name of Diplo-
pterus pulcher (fig. 68.), which, from the position of its

dorsal and anal fins, seems to enter among the Esocine ;
but no description having been given of it, we are fearful
of hazarding any conjecture on its affinities, seeing that
the facts upon which that could be done are insufficient.
We deem it best, however, to mention it in this place,
in the hopes of directing the attention of ichthyologists
to the subject.

(258.) On the last division of this extensive family,
which we have named the Mormyrine (Mor. elongatus
Riipp., fig.69.), we can say but little. The fishes we in-

clude under this denomination form the Linnean genus
Mormyrus, which has been also preserved entire by all

MORMYRINZ. 309

writers. That this group is related to the Esocine, and
yet has a claim to be considered perfectly distinct, is ad-
mitted by Cuvier, who places them atthe end of the pikes ;
observing that “they will very probably give occasion
to form a particular family.” That they are also related
to the Cyprine, or carps, may be inferred from Linneus
having named one of the species M. Cyprinoides. Both
these opinions are reconciled by the situation we now
assign them, as an intermediate or connecting group be-
tween these two sub-families. The whole of the species
are confined to the fresh waters of Tropical Africa,
chiefly those of the Nile and the Senegal; so that, never
having seen them in a fresh state, and but in a cursory
way in museums, we shall give M. Cuvier’s description
of their structure nearly in his-own words. These,
he observes, are fishes with a compressed, oblong, scaly
body: the tail is slender at the base, and enlarged to-
wards the caudal fin: the head is covered with a naked
and thick skin, which envelopes the opercula and the
rays of the gills ; so that the aperture is merely a vertical
cleft, the branchial rays being five or six: the aperture
of their mouth is very small, almost like that of the ant-
eaters; and the angles are formed by the maxillary
bones: some slender teeth, with their tips emarginate,
are on the intermaxillaries and the lower jaw; while
there is a long band of other teeth, small, and crowded,
upon the tongue and under the vomer: the stomach is
like a rounded sack, followed by two ceca ; and there is
a long and slender intestine almost always enveloped in
much fat. Among these fishes, thus generally charac-
terised, there appear several remarkable variations.
The most singular are those which have the muzzle long
and cylindrical ( Scrophicephalus longipinnis Sw., fig.'70.) ;
but among these some have the dorsal fin short, while
in others it is lengthened. Another sub-division has
the muzzle short and rounded; while in a fourth there
is a gibbous projection on the upper part of the muz-
zle, which extends beyond the mouth. The resem-
blance which the long-snouted or typical species bear
x 3

310 CLASSIFICATION OF FISHES.

to the genus Centriscus is so great, that Schneider has
actually called one of them Centriscus Niloticus ; so that

this resemblance being admitted, and the situation of the
whole group intimated by the authorities we have cited,
we may feel some degree of confidence in the station now
assigned to these fishes. The resemblance of the Mor-
myrine to Centriscus, Fistularia, Syngnathus,Gomphosis,
and all other long-snouted fishes, is too obvious to be ex-
patiated upon ; while no question can remain that this
resemblance is one of analogy, and not of affinity, since
Cuvier and most other ichthyologists agree in placing
Mormyrus close to the pikes and carps. Whether the
singular genus Diplopterus Gray enters among these,
or forms a part of the Fsocine, is an interesting but an
inferior question: we are now dealing with large assem-
blages ; and if these are once determined, the more mi-
nute details and relations can be subsequently worked
out.

CHAP. X.

ON THE PLEURONECTID#, OR FLAT FISH, AND THE GADIDZ,
OR CODS.

(259.) Tue PLEvRoNecTIDs, or flat fish, succeed the

salmon family. Although the greater part are savoury

and even delicate eating, it must be confessed that they

are most unsightly fish, totally devoid of that graceful

THE PLEURONECTIDZ, OR FLAT FISH. 311

form and symmetrical colour which belong, more or less,
to nearly all others of this class of animals. The form
of the flat fish is so well known to every one, from the
-frequency of their being served up at our tables, that
there is no occasion to enter on a tedious description.
The genera, comparatively, are very few ; and when we
have seen a plaice and a sole, we have seen the two most
dissimilar forms yet discovered in the whole family.
On a cursory view, it appears as if. the body of these
fish was flattened in the same way as that of the skates,
but this is only an analogical resemblance ; that of the
Pleuronectide is compressed vertically, while that of
the Raide, or skates, is depressed horizontally : this dis-
tinction should always be kept in mind when we are
comparing the repesentations of these fishes in other
families ; for many of the Chetodonide (which are the
flat fish of the spine-rayed order of Acanthopteryges) are
just as much compressed, and some (as Psettus Cuv.)
even more than the Pleuronectide ; but then the position
of the eyes, both being placed on one side of the head,
renders their distinctions very obvious. It is impossible,
in fact, to conceive a more beautiful union of analogical
characters than are to be found in this singular-shaped
group. Placed as one of the types of the order Mala-
copteryges, the Pleuronectide should bear a relation
both to the Chetodonide, or chetodons, on one side, and
to the Raid, or rays, on the other: this they accord-
ingly do, by having the body excessively high and com-
pressed, the fins partially covered with scales, and the
rays semi-spinous: on the other hand, they show a
marked resemblance to the Raide in having both the
eyes placed on one and the same surface of their bodies,
and in one of these surfaces being of a different colour
from the other. These two characters, throughout the
whole class, are solely possessed by the Raide and the
Pleuronectide ; and these analogies not only corroborate
the correctness of the situation we have assigned to each
of these groups, but are borne out by the similarity of
the habits of the Raide and the Pleuronectide, in laying
x 4

312 CLASSIFICATION OF FISHES.

fiat upon the bottom of the sea, wallets in ambush for
their prey.

(260.) The flat fish, besides the pecuiariis in the
situation of their eyes, have some other characters of
a unique nature. The two sides of the mouth are not
symmetrical ; and sometimes one of the pectorals is only
half the size of the other; and this inequality extends to
the bones of the cranium, but which, according to Cu-
vier, are the same, in other respects, as in ordinary fishes.
The general sameness in the external structure of these
fishes renders it very difficult, if not impossible, to
make out the principal types; and therefore, as the
technical definitions of the genera will be given in the
synopsis, we shall here confine ourselves to a few general
remarks on the whole family.

(261.) The geographic distribution of these fishes
is in almost all temperate and tropical seas; but they
seem to diminish towards high northern and southern
latitudes. They are abundant with us, and very common
in the Mediterranean. The largest species, we believe,
that has yet been discovered, is the holibut (Hippoglossus
vulgaris Cuv.), which is often seen suspended in the fish-
mongers’ shopsin London. On the coast of Norway this
fish often attains the weight of 5001b., and Mr. Yarrell
mentions one that was taken near the Isle of Man and
sent to Edinburgh in 1828, which measured 7 ft. 6in.
in length, and weighed 3201b. We know too little of
the foreign flat fish to say whether other species attain
to this size, but those on the Brazilian coast are of the
ordinary dimensions ; and the holibut is, we believe,
unknown in the Mediterranean. The turbot, as an
edible fish, is the most celebrated of all the numerous
species of the Atlantic. We cannot, however, coincide
in the belief that this delicious fish was known to the
Athenians ; for although the Pleuronectide, as a whole,
are common in the Grecian and Sicilian seas, we never
saw the turbot there, nor ever heard of its being captured.
A great deal of interesting information on the fishery of
this article of luxury has been given in Mr. Yarrell’s

PLEURONECTIDA IN GENERAL. 313

volumes, which the reader would do well to consult.
The manners of all these fish, as well as of the soles,
appear to be much the same. They keep close to the
bottoms, generally choosing such as are sandy ; here,
partially covered by the surrounding sand or soil, which,
by the peculiar shape and construction of their fins, they
are enabled to throw about them, these fishes lie in
ambush, watching for any prey which may come within
range of a sudden dart: at other times, however, these
sedentary habits are laid aside, and they shift their
hunting grounds, often in large shoals, when one sort of
food is exhausted, and another is sought for. Mr. Yarrell
observes, no doubt from good authority, if not his own,
that, when near the ground, the plaice (Platessa vulgaris
_Cuv.) swim slowly, maintaining their horizontal position ;
but, when suddenly disturbed, they sometimes make a
rapid shoot —changing their position from horizontal to
vertical: if the observer happens to be opposite the
white side, they may be seen to pass with the rapidity
and flash of a meteor; but they soon sink down, resum--
ing their previous motionless horizontal position, and are
then not distinguished any more than the rest of the family,
owing to their great similarity in colour to the surface
on which they rest. The food of the flat fish is exclu-
sively of an animal nature; but as their mouth is of
moderate size, and their teeth small, the things they feed
upon are of proportionate dimensions: young fish and
crabs, small and soft molluscous and radiated animals,
are what are generally found in their stomachs.

(262.) The resemblance between the colours of the
flat fish, in general, to those of the ground they repose
upon, is so admirably ordered, as to claim both atten-
tion and admiration. The upper surface, or that which
is exposed to view and to the action of the light, is inva-
riably of some shade of earthen brown, or of greyish
sand colour ; this is broken by dots and blotches, either
light or dark, blackish or reddish, but always so disposed
as perfectly to resemble those under-shades, as they may
be called, which are caused by the inequalities of the

314 CLASSIFICATION OF FISHES.

ground and the presence of particles of different tints —
that may be upon it. Thus, whether we contemplate the
God of Nature in his most sublime productions, or in
those provisions which He makes for the well-being of
his most irrational creatures, the same principle of de-
sign — the same absolute perfection in execution —is
equally conspicuous. This exquisite finish is bestowed
upon millions of creatures which the eye of man ‘ hath
not seen ;” “‘ nor hath it entered into his heart to con-
ceive ’’ the faculties and the instincts they possess, still
less to form ideas on all the reasons of their creation.
Such knowledge, indeed, we cannot attain to in this
stage of our existence ; but the good shall most assuredly
enjoy it in their next.

(263.) The Gapipa, or cod-fish, form our third
division of the malacopterygious or soft-rayed order of
fishes. The general construction of these has already
been intimated (p.230.). They form the first of the aber-
rant division of the order, and are placed by Cuvier next
to the Pleuronectide : that eminent anatomist, indeed, |
has included them both in one order, distinct from the
Salmonide, because the ventral fins in both are placed
under the pectorals ; nevertheless, as we find that by this
character the lump-fish, suckers, &c., forming the Cy-
clopterid@, are made to follow the cod-fish, it becomes’
plain that such a principle of arrangement unites dis-
cordant groups, and cannot be followed in a natural
system. We shall first make a few observations on the
general peculiarities of this family, and then explain its
component parts.

(264.) The cod-fish are chiefly found in the cold
and temperate seas of the northern hemisphere, for
none have as yet been described as inhabiting India or
the great Pacific Ocean. The common cod, whiting,
haddock, hake, and several other well-known species,
although of a small size, belong to this family, and are
well known for the abundant supply their prolific num-
bers furnish to man. The fishery for cod on the banks
of Newfoundland is much more important than that of

THE COD FISHERIES. $15

the herrings on our own coasts, since it gives employ-
ment to more men and a greater amount of capital. It
has been estimated that 20,000 sailors are annually
employed in this fishery, which is carried on, not in
decked boats, but square-sailed vessels. An official
report of the French. minister stated that, in 1792,
no less than 210 vessels, amounting to 191,153 tons
burthen, sailed from the ports of France with the sole
purpose of prosecuting the cod fishery ; and it has been
supposed that more than 6000 vessels of all nations are
so employed, partly on the coasts of Norway and Sweden,
-but chiefly on those of Newfoundland and the adjacent
parts. Thirty-six millions of fish are supposed to be
thus captured, salted, and dried, which are carried to all
regions of the world. We have eaten them, under the
name of stock-fish, in all parts of the Mediterranean,
brought by our English vessels ; and they are to be had
in all parts of the Brazilian empire—being carried on the
backs of mules from the sea coast into those provinces of
the interior where fresh fish cannot easily be procured.
The annual destruction of such innumerable hosts of cod
might be supposed, by some, to threaten the total ex-
termination of the species, but a bountiful Providence
— which has signally appointed this to be one of the
most useful fishes to man —has given to it the most
extraordinary powers of reproduction. We have stated
that the annual captures may amount to 36,000,000:
how, it is on record that 9,000,000 of eggs have been
found in the roe of one female; so that, if only one half
of these were hatched and grew to maturity, nine female
cods would supply the destruction occasioned by the
captures of all the fisheries in one year. So far, there-
fore, from their extermination being probable, the only
~ wonder is that they have not so increased, in a series of
ages, as to fill the ocean, like the sand upon the shore.
We can only suppose that they are kept within due
limits, not by man, but by the millions on millions of
fry which are fed upon by innumerable other fishes and
other marine animals: sea birds prey upon them when

316 CLASSIFICATION OF FISHES.

older ; and sharks and other fishes attack and devour
the full grown ones. To support such countless swarms,
there must also be a corresponding fecundity in those
animals upon which they feed : these appear to consist of

crabs, worms, shell-fish, &c., which frequent the bottom,

near to which the cod is almost always found. Great
numbers are caught all round our own coasts, particularly
on the north and west of Scotland, where, as Mr. Yarrell
says, most extensive fisheries are carried on; so that, even
in the United Kingdom — which can only be compared
to one of the suburbs of the great metropolis of the
cod, which is Newfoundland — the catching, curing,
and sale of this fish employ thousands of individuals.
The cod is caught invariably by hook and line; and
they are so voracious, that they bite at almost, any bait.
On the banks of Newfoundland, one man will some-
times catch from 400 to 550 fish in ten or eleven hours ;
and the master of some fishing vessels told Mr. Yarreli,
that “ eight men, fishing under his orders, off the
Dogger-bank, in twenty-five fathom water, have taken
eighty score of cod in one day.” “ The largest cod-fish,”
obseryes the same author*, “ I have a record of, weighed
60 lb.: it was caught in the Bristol Channel, and
produced five shillings ; it was considered cheap there
at one penny the pound.” In Pennant’s time, how-
ever, the price was even less; for he mentions one caught
at Scarborough, which weighed 78 lb., that was sold
for one shilling. How satisfied would be the inha-
bitants of many of our remote inland towns to pay
five times these prices for slices of this. most delicious
fish ! .

(265.) The fins of the Gadide, unlike all others of
this order, excepting the flat fish, are thick and fleshy,
being covered by the common skin of the body. Hence
the rays of many, being slender and close together, can-
not always be counted—at least with any degree of cer-
tainty: the mouth is always large; and the jaws, with
' the fore part of the vomer, are furnished with several
* Brit. Fishes, vol. ii. p. 147.

_

THE GADIN#/, OR TYPICAL CODS. oly

rows of pointed, irregular, rasp-like teeth of different
‘sizes, but none of them very long: the aperture of the
gills is always large. It is in this group only, of all
the soft-rayed families, that we find three dorsals; but
their ventrals are almost always imperfect, — that is to
say, three or four of the rays are either excessively
small or totally wanting. This circumstance, with
their fleshy fins, their depressed head,- and their length
of tail, are all so many proofs of their analogy to the
apodal or anguilliform order.

_ (266.) The whole of the genera defined by modern
ichthyologists, together with two others now intimated
for the first time, we shall arrange under the five fol-
lowing divisions, which may be considered as sub-fami-
lies: —1. The Gadine, or typical cods, having always
two anal fins, and generally three dorsal ones. —2. The
Merlucine, or the hakes and rocklings, where the dor-
sal fins are only two: these, being the two typical forms,
possess the additional character of having five distinct
rays to their ventral fins, although the posterior ones are
usually very small, while the first or second is much
lengthened and pointed.—3. The Phycine, or forked -
hakes, so called from the ventral fins being each com-
posed, apparently, of a single ray forked towards its
middle. —4. The Brosmine, where there is only one
dorsal ; but the ventral is with five rays.—And, lastly,
the Brotuline, or eel-shaped cods, having the dorsal,
anal, and caudal fins united.

(267.) The genus Gadus of Linneus, as represented
by the common cod (G. morrhua), stands at the head of
the entire family, as well as being the type of the Gadine:
with this well-known fish we associate the dorse, haddock,
pout, poor, and speckled cod of Britain, together with
certain other species found in the Mediterranean, two
of which do not appear to be described. All these have
six rays to their ventral fins; but the two first are only
well developed ; and these being lengthened by a fleshy
filament, give the fin a very pointed shape: the caudal
fin is always more or less lunated ; but in one species

318 CLASSIFICATION OF FISHES.

(Gadus furcatus Sw., fig. 71.) we discovered in Sicily,
the tail is forked: they have, moreover, a short cirrus,
or barbel, at the point of the lower jaw; but this 1s

wanting in the next genus, Merlangus Raf., although,
in every other respect, the structure is the same. The
next type we have named Ti/esia, after its first de-
scriber, professor Tilesius, well known as one of the
most eminent naturalists that Russia has produced.
There is something so peculiar in the elongated form of
the Gadus gracilis of this author *, joined to the
truncated form of the caudal fin, that we venture for
the present to keep it distinct from Gadus, with which,
however, it agrees in having a barbel and three dorsal
fins ; but these latter are represented as all of the same
size, — a proportion not observable in any other of
the cods, where the first dorsal is always higher and
shorter than either the second or third: the trunc-
ated or slightly rounded tail of this fish prepares us for
Lepidion Sw., represented by a most singular species of
cod, described as very rare in the Mediterranean by
Risso, who has likewise given a rude figure of it: in
this the dorsal fins are only two; and the two anal fins
are so much united, that they appear almost as one
that is deeply cleft. Risso describes it, however, as a_
Gadus, which, in all other respects, it resembles. It is
a remarkable circumstance, that, of the two species he

* Icones et Descrip. Piscium et Vermium Zoop. Camtschaticorum Pe-
tropoli, 1810. This rare work is in the library of the Linnzan Society.

GADIDA.— MERLUCIUS. 319

mentions, one has the tail truncated or slightly rounded,
while in the other it is, as he says, bifid: we reconcile
this difference by supposing that his Gadus Lepidion (our
Lepidion Rissoii) connects with Tilesia ; and that his
Gadus Moro (our Lepidion Moro), which has the *‘ cauda
bifida,’ prepares the way to the last type, or Cephus.
The only species yet discovered of this extraordinary
type is the Gadus macrocephalus of Tilesius (fig.'72.) :

the head is so enormously large, that it is nearly half
the length of the whole fish, and is much thicker than
any part of the body ; the crown also is depressed ; and
the whole fish immediately gives the idea of a gigantic
Raniceps ; but having the three dorsals, and all the
other points of structure of the true cods, except that
the tail is truncate, and the gills covered with scales.
(268.) The Merlucine are less numerous in species,
and in the variation of their forms ; and thus we know
of only three genera. To the first of these, named by
Rafinesque Merlucius*, after the Gadus Merlucius
of Linneus, belongs the common hake, peculiar to
northern seas, with which the Mediterranean hake (M.
_ sinuatus Sw., fig. 73.), now for the first time described,

* Carattari, &c. Palermo, 1810.

320 CLASSIFICATION OF FISHES.

has hitherto been confounded by all writers: we pre-
sume this is the species, which, under the belief that it
was the common one, Cuvier says is abundant in the
Mediterranean. A third species, under the name of M.
Smiridus *, is mentioned by Rafinesque, as being rarely
found on the Sicilian coast ; but his description, unfor-
tunately, is so short, that the only definite character
we can discover is that the dorsal fins are nearly
equal: this we have never seen, as it was caught on a
part of the coast we did not explore. The next genus is
Lota: it is composed of more elongated fishes than any
of the preceding: their body, as well as the form of the
head, has more analogy to the eels, and the caudal fin is
rounded; the under jaw, as in Gadus, is furnished with a
cirrus. The ling seems the largest, and certainly the best
known, of this genus. We remember catching many of
this fish off the southern coast of Ireland, by a common
hook and line thrown out from the vessel: it is a par-
ticularly voracious fish, and is generally from two to four
feet long; but Pennant mentions one that measured seven
feet. The other British species, Lota vulgaris, or the
burbot, is the only species among the British Gadide
that is fluviatile: we shall quote Mr. Yarrell’s remark on
this species, in further confirmation of our theory that
this family represents the apodal order among the soft-
rayed tribes. “‘The burbot,” observes this excellent ich-

thyologist, “is not unlike the eel in some of its habits, —
concealing itself under stones, waiting and watching for
its prey ; it feeds, also, principally during the night,
and, like the eel, is most frequently caught by trimmers
and night lines.” The third and last genus which enters
into this division is that of Motella, or the rocklings:
these are much smaller and even more eel-like fishes than
the foregoing: they are peculiarly distinguished by
having cirri at the tip of both jaws, and by the singular
structure of the first dorsal fin, which is altogether

* MM. Smiridus. ‘Capo quasi troncato diagonalmente, ale dorsale quasi
uguale. "—Carait. p. 25.

ee

GADIDA. — MOTELLA. — PHYCINZ. 321

unique among fishes: the rays are excessively slender,
and are composed of fine fleshy filaments, without any
internal bony support ; so that they can scarcely be dis-
cerned when the fish is out of the water: the first ray
is always the longest; and it exactly resembles in shape,
appearance, and substance, the cirri or beards on the
snout. Several species occur on our own coasts, and
others inhabit the Mediterranean: one of these (Motella
fusca Sw., fig. 74.) is here represented, as a perfect type

QE SSS

~ SXQQY SST“
RS QWS WSs

of the whole group; its description will be found in
the Appendix.

(269.) The Phycine, or forked hakes, form a natural
succession to the rocklings: they are not so much
distinguished by their greatly depressed head and their
two dorsal fins, as by the peculiar construction of the
ventrals, which consist but of one long and cylindrical
ray, divided about half way into two unequal parts.
Between these and the last we have the intermediate
genus Raniceps of Cuvier, whose first dorsal is like that
of Moteita, but whose depressed head may perhaps place it
within the confines, as an aberrant genus, of the present
group. Its ventrals also partake of this intermediate
character : there are, indeed, six rays, as authors assert,
in each fin; but three of these are so minute as to be
nearly obsolete, while the two outer are long and de-
tached, so as to resemble the forked single ray of the next
genus, Raniceps thus becomes the link of connection
between Motella and Phycis ; and the possession of a
single cirrus on the lower jaw makes the passage more
gradual from the bearded rocklings to the forked hakes,
which have none of these appendages. There seems to
be much confusion among the species of Phycis, ori-

VOL. I. Y

$22 CLASSIFICATION OF FISHES.

zinating in a great measure from the prevalent custom
of assimilating the species peculiar to the Mediterranean
with those found in our northern seas. Cuvier, indeed,
being aware that there is one species found in the
Mediterranean different from ours (his Phycis furcatus),

justly enough supposes that it was the true Blenziius
Phycis of Linneus ; but we have reason te believe that
no less than three inhabit the coasts of the south of
Europe. Two of these we discovered in Sicily, and
shall subsequently describe in detail. One of them, our
Phycis longipinnis ( fig. 75.), will serve as a typical re-

presentation of the whole ; and the other, Physis Siculus
Sw., will be found in the Appendix. We believe that
others from the same seas will be hereafter detected, so
soon as the Mediterranean fishes undergo a more rigid
comparison with those of the German Ocean, than has
ied been deemed necessary.

270.) The Brosmine, at present, are represented only
bya bangle genus, of which the Gadus Brosma of Linneus
forms the type. We have placed this as a represent-
ation of one of the principal divisions of the family,
because it seems to hold an intermediate station between
Phycis and the next group. The first dorsal or anterior
fin here disappears, leaving only one, which begins im-
mediately above the pectoral, and only terminates at
the very commencement of the caudal: the ventrals are
like those of Lota and Motella, but as fleshy as in
Phycis; and there is a single cirrus on the lower jaw.
The only species known, or, at least, that has been de-

GADID =. — BROTULINEA. 323

termined, is a northern fish, seldom reaching to the
length of three feet and a half.

(271.) The Brotuline, or eel cods, will terminate
the series. In Brosma we have been prepared, by the
close approximation of the dorsal and anal to the caudal
fin, for that union of all three which takes place in the
group before us. This is precisely the structure com-
mon to the true eels; and thus we have a. perfect re-
presentation of the apodal order in the circle of the
Gadide. No British example of this type exists. The
only two genera that can certainly be placed in this group
are Brotula Cuv. and Pteridium Scopoli. Of the first,
only one species is obscurely known ; it has six barbels,
and is found on the coast of Cuba.. Pteridium Scop., on
the other hand, is a Mediterranean genus ; and although
so long established, both by Scopoli, and subsequently
(under the name Oligopus) by Lacepede, it is altogether
omitted in the Régne Animal. Risso describes and figures
one species (P. niger), which perfectly resembles, in its
general aspect, the genus Ophidium, except in having no
cirri under the chin, and in possessing small, slender
ventral fins, composed of a single ray.* It will subse-
quently be shown, when we come to treat of the genus .
Ophidium, that this is the precise point of unicn be-
tween the two great orders of osseous fishes, or that
which forms the passage from the Malacopteryges to
the Acanthopteryges. The silvery Ophidium of the Me-

iterranean, in fact, has so strong a resemblance to the
present family, that it may almost be termed a Gadus
without ventrals: it grows to as large a size as some of
the Phyces, and has just the same formed head, mouth,
and teeth ; the same silvery body, minute scales, and
fleshy fins ; they live in the same situations; and they
are so alike in taste, — the flesh of both being the same
as that of the haddock, —that they cannot be distin-
guished when cooked. Cuvier brings into this family
the remarkable genus Macrourus of Bloch, which
agrees with Pteridium so far as to have the caudal fin

* Ichthyologie de Nice, p. 142. pl. 11. fig. 41.
wee

324 - CLASSIFICATION OF FISHES,

united to the second dorsal and to the anal, both of
which are very long. In all other respects, however,
the species of this genus differ so very much from the
general structure of the Gadide, that we cannot think
they are naturally located with them. They may,
indeed, constitute the most aberrant type, or that which
we have assigned to Brosmius ; but our impression is,
that Macrourus represents the anguilliform or apodal
division of the Tvriglide, in which group we have’ there-
fore placed it.

(272.) Without going into an extended exposition of
the analogies of this family, it will be sufficient to call
the naturalist’s attention to those general points of re-
semblance to the higher types or divisions, which are
presented by what appear to be the leading genera of the
Gadide. Thus we have the prevalent character of
the cartilaginous order indicated in the very broad and
depressed muzzle of Phycis and Raniceps; while that of
the apodal is clearly symbolised in the eel-like form of
Brotula and Pteridium, where the caudal fin is con-
founded, as it were, with the dorsal and the anal. Fur-
ther, if we confine our attention to the five sub-genera,
here indicated, of the genus Gadus, it is not difficult to
trace a resemblance to the higher divisions of the whole.
This will be best seen in the following table, with which
we may conclude our sketch of this family.

Sag Analogical Characters. sited pepe
Typical: ventral fins two; mouth with 2
Gadus. - > i GADIN=.
barbels.
Merlangus. No barbels. MERLUCINE.
Tilesia. Body elongated. PHYSINZE.
Lepidien. wae ae dissimilar from their respective } Brosainx.

Cephus. Ventral fins much pointed. BROTULIN2.

325

CHAP. XI.

ON THE FAMILIES OF SILURID4 — THE CAT-FISH, OR SILURES;
AND OF THE COBITIDZ, OR LOACHES. f

(273.) Tue Sinuripa, or cat-fish, are far more nu-
merous than the Gadide@, nor do they yield to the
Salmonide in the number of their species or the diver-
sity of their forms. They are entirely fluviatile, or, at
least, have never yet been found beyond the estuaries of
the great rivers. Of all fish yet discovered, they have
the longest cirri or barbels, — appendages which are
doubtless used to allure the prey upon which they subsist,
and which, mistaking these slender filaments for worms,
bring themselves, unconsciously, within the grasp of their
hidden foe. The Siluride, from all we know of their
manners, lie concealed in the mud ; and hence they are
most numerous in such of the tropical rivers as flow
over soft ground, and whose course is not rapid. Some
of the Asiatic species are more especially found in ponds,
tanks, and even ditches : here they lie concealed in holes
along the bank ; or are half hid beneath the mud and
weeds at the bottom. Only one species, the Silurus
Glanis Linn. ( fig.76.) has yet been found in Europe: it
is the largest of all the freshwater fish yet discovered: and
seems more especially appropriated to the great rivers of
Austria, where individuals have sometimes been captured
of an enormous size; in Pomerania they have been
taken from twelve to fifteen feet long, with a mouth
sufficiently capacious, as it is said, to gorge a child of
six years old; another, captured at Writzen on the
Oder, is stated to have weighed 400lb. The flesh is
white, and of an agreeable taste. An attempt was made to
¥ 3

326 CLASSIFICATION OF FISHES.
naturalise this spe-

for which purpose
several young ones
were brought from
Hungary, and turned
into the river: this
plan has beenin some
degree successful ;
for although the
fish have not mul-
tiplied, in their new
abode, so rapidly as
was anticipated, yet fine individuals are occasionally
caught there, and even transmitted to the Parisian
market. In India, most of the larger species are eaten
by the natives, and many by the Europeans, notwith-
standing the prejudice arising from their lurid colour
and repulsive shape. In Britain, we have no proof
that this species has ever existed ; for although Mr.
Yarrell has introduced it into his valuable enumeration
of our native fish, he very justly questions the fact of
its having been known to Sibbald, who has probably
mistaken she burbot for the “ Silurus, sive Glanis,” of
the ancients.

(274.) The form of the majority of these fishes is
altogether peculiar, or, at least, we only find partial re-
presentations of them in other families. The mouth is
small, furnished with fascicles of minute teeth, often so
imperceptible, that they have justly been compared to
the pile of velvet: these teeth are variously shaped
and disposed, but without any of that uniformity which
induces us to look to them as organs deserving a
primary consideration. In species bearing the closest
approximation in all other respects, one will possess
teeth, while the other has none*; even when present,

z
they are so very minute as not to be clearly defined,

_* Bagris, Pimelodus, &c. of M. Cuvier are striking instances of the impos.
sibility of classing these fishes by their teeth alone.

cies at Strasburg ;

THE SILURIDH IN GENERAL. 327

except under a high magnifier: hence we may safely
infer, that the major part of the Silwride@ swallow their
food entire ; and that, as the mouth is moderate, the
fishes they capture must be of a small size. It is more
than probable, also, that several of them feed as much
upon vegetable as upon animal substances ; something
in the manner of carp and eels. The worm-like and
flexible cirri, or barbels, of the Silurid@, as before ob-
served, is another of their most remarkable distinctions:
those on the upper lip are always the most developed,
and among the Pimelodine are often as long as the
whole body: these are supported by the intermaxillary
bones; and, from their great length, command a much
wider range than those on the under lip, which are
usually shorter. The head is exceedingly depressed,
being often almost flat; so that its height, when viewed
in profile, is not more than one fourth, and even less,
of the vertical breadth: the mouth is small or moderate,
and the eyes by no means equal to those of the generality
of fishes. The body is always destitute of true scales,
even of those obsolete ones seen in the neighbouring
group of the Gadide, or cods; yet, if naked, it is very
slimy ; and in the typical group, or the Loricarine, the
whole head, and the greater part of the body, is mailed
by hard bony plates, which makes them appear like the
mailed ant-eaters of India, and apt representatives of ©
the chelonian reptiles and the loricated fishes. The
Siluride, as we have already shown, are the most
aberrant of all the soft-finned order: this at once ac-
counts for the fact, that the great majority of the species
have the first ray, both of their dorsal and their pec-
toral fins, not only spined, but usually very thick ; and
these rays are rendered sometimes more formidable, as
weapons of defence, by having one or both edges finely
toothed, with the points directed inward. Dr. Bu-
channan Hamilton, in reference to this peculiarity of the
cat-fish, conjectures that, ‘in general, every time that
they are employed, the animal must suffer considerably,
as, in most of the species, these prickles terminate in a
y 4

328 CLASSIFICATION OF FISHES.

flexible substance, somewhat resembling whalebone ; and
before the prickle can penetrate an enemy, this flexible
point, which supports part of the membrane, must be
broken. Although, therefore, it is probable, that in case
of such accidents, the flexible part may be soon reunited
to the prickle, yet it is not likely that the animal should
have recourse to the use of such a weapon, except in
emergency, and perhaps never as a weapon of attack.’’*
To us, however, this subject appears in a different light.
From observations made upon the American Siluride, it
appears, that although these rigid spines, in one sense,
are terminated by a flexible process, yet that this pro-
cess is more an additional appendage to the spine, than
an integral part thereof; it is, in fact, so articulated,
that it can be bent sufficiently back, and that without
any injury, to admit the spine being used as a pow-
erful weapon of offence, — the soft appendage by
which it is surmounted, returning again to its usual
position so soon as the spine is disengaged from any
substance it has penetrated. This soft part of the ray,
in fact, might, with more propriety, be termed as much
articulated to the spined or bony part, as if it moved upon
a spring: it may be easily pressed backward, but not
forward ; and in every position it leaves the attenuated
point of the spine itself completely free. To illustrate
this very singular peculiarity, which does not hitherto
appear to have been noticed, we annex the accompanying
sketches of the dorsal (a) and pectoral (c) fins of our new °
genus Breviceps; that at a (fig. 77.) showing the
spine in a state of repose ; while 6 represents it with
its soft and articulated termination bent backwards,
leaving the point of the spine entirely naked. We
cannot say how far this structure is prevalent among
those Siluride we have not personally examined ; but
there is every probability that it may be general. Be-
sides, if we take a more general view of the question,
there is nothing in nature to make us believe that the

* Gangetic Fishes, p. 139.

SPINES OF THE SILURES. 329

means provided by Almighty Wisdom for the defence,

and consequent protection, of his creatures, should be
accompanied by injury, pain, or distress in its use.
(275.) The classification of this family, as it at present
stands, is involved in greater confusion than it is to be
found, perhaps, in any one group of ichthyology. Several
of the subordinate divisions were pointed out by the
illustrious Artedi, the father of scientific ichthyology,
and were adopted by Linneus in the earlier editions of
the Systema Nature ; but, from a mistaken idea of
simplification, which the celebrated Swede appears sub-
sequently to have conceived, he incorporated all these
genera under the common one of Silurus. In this,
however, he was not followed by either Gronovius or
Bloch, each of whom characterised several others.
Lacepede next separated the Pimelodes from the Siluri ;
but, by mistaking the true distinction of these two groups
and founding his primary character upon one of
secondary import, his arrangement, as will subsequently
appear, became artificial. All these, with two or three
additional divisions, were incorporated in the first edition
of the Régne Animal. It is somewhat singular, notwith-
standing the important additions since made to this fa-
mily—not merely in species, but in the discovery of new
and extraordinary types by Humboldt, Spix, Agassiz, and
Buchanan, as well as intimations of others in the French
Museum — that Cuvier should have merely noticed a
few of these discoveries in the way of incidental notes ;

330 CLASSIFICATION OF FISHES.

so that his arrangement of the Siluride is virtually the
same in the last edition of his Réegne Animal as it was
in the first. Great alterations, no doubt, would have
been made in his general work on ichthyology ; for it
must have been evident to himself, that, as the family
now stands, it is much in the same state as those of
Falco, Sylvia, and Muscicapa, among birds, once were,
when Linnean authority was considered paramount to
that of nature. With a hope, therefore, of laying
some foundation for a natural arrangement of this
group, to which for many years we have felt much
attached, we shall now submit to the ichthyologist our
latest views on its internal and external relations. Al-
though this will be the result of much personal inves-
tigation, it would be injustice not to add that this
could never have been accomplished, but ‘for the inva-
luable labours of Spix, Agassiz, and Hamilton, whose
admirable descriptions and characteristic figures have
made us almost as well acquainted with the Siluride of
Tropical India and America, as if we had personally
examined all the subjects themselves.

(276.) The natural station of the Siturmp#, in the
circle of the present order, has already been investigated.
By occupying an intermediate station between the Gadi-
de, or the cod-fish, aud the Cobitide, or loaches, it be-
comes the most aberrant family of the circle; and thus
corresponds to the order of cheloniform fishes, or the
Plectognathes, and to the class of Amphibia. These
analogies, which are particularly beautiful, at once ac-
count for the fact of our finding among the Siluride
a far greater proportion of mailed fishes than exist in
any other group, either of the Malacopteryges or of the
Acanthopteryges: it is this peculiar character, in short,
which marks the typical perfection of the whole. Our
first division, therefore, or sub-family, is composed of
the Loricarine, or mailed cat-fish: they are distin-
cuished, like the Lor. plecostomus Linn. ( fig. 78.), by
the head and body being covered, more or less, by hard
osseous plates, forming large rude scales ; and especially

\
SILURID®,—DIVISIONS OF THE FAMILY. 331

by the situation of the mouth, which is always placed
beneath. The sub-family of Pimeloding next succeeds :

this is a much more numerous division than the last, from
which, although the body is sometimes partially mailed,
it may be further distinguished by the mouth being ter-
minal, or but seldom placed beneath the snout. It like-
wise comprises all those silures whose belly, as in the
generality of fishes, is of equal length with the tail,
so that the vent is nearly central between the pectoral
and the caudal fins: this structure carries with it,
almost universally, a peculiarity in the proportionate
length of the anal fin, which is never very long, as in the
succeeding division ; nor are the number of dorsal fins
ever less than two, the hinder of which is always adipose.
The third sub-family includes the true Silurine, or
anguilliform cat-fish, well represented by the genus
Plotosus Bl. (fig. 79.). They are immediately known

from the Pimelodi, with which they have hitherto been
confounded, by the excessive length of their tail, which is
always more than two, and often four, times the length
of the abdomen: this structure produces a correspond-
ing modification of the ventral fin, which is always
exceedingly long and low, so as in nearly all cases to ex-
ceed one half the length of the entire fish. Like the
Loricarine, we find that the two typical genera of the
Silurine are distinguished by the presence or absence of
a small adipose fin ; although, in all other respects, their

332 CLASSIFICATION OF FISHES.

organisation is essentially the same. The great deve-
lopement of the tail, however, from being the primary
characteristic of the whole group, may be traced in all
the aberrant types, notwithstanding the numerous vari-
ations they present in other parts of their structure.
Quitting these, however, by means of the eel-like genus
Plotosus, we enter at once into the fourth sub-family, re-
presented by the genera Aspredo of Artedi (Asp. Gronovii
Sw., fig. 80.) Platystacus (in part) of Bloch, and two

Sa /
or three others. The first of these is probably the type
of the whole: they are distinguished by the excessive
smallness of their eyes, which are vertical, or placed
close together on the top of their head; and the aper-
ture of the gills is merely a simple cleft of the skin, as
in the Plectognathes, or cheloniform genera. Here, also,
we arrange the remarkable genus Astroblepas of Hum-
boldt*, one of the most extraordinary fish in the whole
family, since its connection to certain Mysti of Grono-
vius is quite evident. In the Heterobranchus 5-tenta-
culatus of Spix, the head is protected or mailed hy bony
plates, as in the Heterobranchi of Egypt; while its length-
ened muzzle shows such an affinity to the Sorubine, that
we consider its real affinities to be with these latter fishes.
The genus Sorubium of Spix we place after the Aspre-
dine: in these the dorsal fins are two—the hinder one
being either adipose, or with imperfectly developed rays:
the head, however, has now become uncommonly long
and large, the muzzle obtuse and dilated, and the upper
jaw considerably longer than the under. We cannot fail

* This is not alluded to by M. Cuvier, even in his notes.

ANALOGIES OF THE SILURIDZ. 833

to recognise in these characters the genuine indication of
the broad-headed or cartilaginous type of animals; for
they are not only the primary external marks by which
that type is distinguished, but are almost peculiar to this
division of the Siluride. The same form, indeed, but
differently modified, is seen in the Sturisoma rostrata*;
so that we once more enter among the Loricarine, with
which we began our survey, and the circular succession
of all the sub-families becomes sufficiently evident.
(277.) Let us now compare the preceding divisions
with the primary orders of fish, placing each in the
natural series of their succession. ‘This comparison,
also, will elicit several other characteristics of the Si-
lurian groups which have not been yet touched upon.

Sub-families of 5 Orders of
Sane. Analogical Characters. Bane

Loricarine. §The two most typical groups in ? MALACOPTERYGES.
Pimeloding. ¢ their respective circles. ACANTHOPTERYGES.

Eat ae Tail excessively long; the caudal
Siurine. anal, and dorsal fins often united. { ADDS:
Aspredine. eves small, wetticall;/eils seen PLECTOGNATHES.

: Muzzle long, broad, protruded

Sorubine. beyond the ara f CARTILAGINES.

We need not again advert to the difficulty that arises,
in some cases, of determining the more particular ana-
logies of the two typical groups of one circle, with those
of another circle ; which, as we have had frequent oc-
casion to observe, appears to be sometimes reversed.
In such cases we must fall back upon the strength of
our affinities, and leave the analogies to be explained
in a more advanced state of the science. No one, how-
ever, can deny that the Loricarine and Pimelodine follow
each other, and this in as unquestionable a manner as
the soft-finned fishes are succeeded by the spined order.
These points, therefore, being determined by absolute
affinity,— an affinity acknowledged by all writers,— it
follows, from the collateral evidence derived from the
other groups, that these four are analogous to each

* Loricaria rostrata of Spix.

834 CLASSIFICATION OF FISHES.

other. As the Siluride are the most mailed of all the
soft-finned fishes, they thus become analogous to the
tortoises, or chelonian reptiles, and to the cheloniform
fishes ; hence it is natural to suppose that the Lovi-
caring, from being the most mailed, are those which
should stand at the head of the family. On the other
hand, the Pimelodine, in point of general structure,
appear to be the most perfectly formed, or, rather, the
most highly organised, of the two: their body, indeed,
is naked, but they have all two distinct dorsal fins ; and
those cirri, which are few and short, or altogether want-
ing, in the Loricarine, are here developed to an enor-
mous length, more especially in those of the New World.
Leaving these, we may proceed to the other comparisons,
where the analogies are so strong that they must strike
every one who has paid any attention to this most fas-
cinating department of zoology. The Silurine, indeed,
are not, like the eels, destitute of the ventral fins, for
then they would actually become apodal fishes ; but
the great development of their tail, their rounded
caudal fin, their dark lurid hues, and their natural
habits, leave nothing more to be desired on this head.
It is proper, also, to observe in this place, that what has
previously been said on the habits of the European
species (Si/urus glanis), must be looked on as more par-
ticularly applicable to this division, rather than to the
other sub-family of the Siluride ; about which, in fact,
we really know little or nothing. The analogy between
the Aspredine and the Plectognathes is rendered per- —
fectly conclusive by the structure of the gill-covers,
which in these two groups are immoveable. Again, it
may be observed as a general rule, that, on comparing
any natural group of fish with the primary orders of the
class, we shall almost invariably find that those whose
eyes are small, and situated at the top of the head, turn
out to be the most aberrant in their own circle: now,
this is explained by looking to the order Plectognathes,
where we find both the Balistide and the Chironectide
possessing this structure in a remarkable degree. The

SILURIDH.— THE LORICARIN. 335

same peculiarity is apparent in the most aberrant of all
the apodal fishes, where the eyes, as in Mywine, are
totally wanting. All these facts are in full accordance
with our theory (or, rather, we should say they have
been the foundations of the theory), that the most
aberrant group of a circle is always that which is least
organised: the class Acrita in the animal kingdom, the
Amphibia among the Vertebrata, the Vermes among the
Annulosa, and innumerable other instances, are too
well known to require enumeration; and we now have
a further instance in the circle of the Siluride. Fi-
nally, we come to the relations of the Sorubine with
the cartilaginous order of fishes, and the fissirostral
type of birds: the discovery of this group in the rivers
of Tropical America, by the lamented Spix, and the
masterly description of its peculiar structure by Agassiz,
has thrown a light upon this family, we have been
waiting for several years ; while the no less important
discovery of Sturisoma, by the same traveller, establishes
two of the most important facts in the natural arrange-
ment of the family ;— one showing us a group with
the flat protruded muzzle of the sharks; the other,
the manner in which the whole of the groups we have
now reviewed, close into a circle. Did our space permit
us to extend these analogies to the T’riglide, the order
Plectognathes, and some other groups, many new and
interesting illustrations would result; but there is so
much to be said on the internal relations of the Siluride
among themselves, that we must proceed at once to a
more detailed account of the variations of structure ob-
servable in each of the sub-families we have now de-
signated.

(279.) At the head of the Loricarine, or mailed cat-
fish, stands the genus Loricaria, distinguished by having
the whole of the body (excepting the belly), together
with the head, covered by large osseous plates, re-
sembling rude scales; the back being furnished with a
single dorsal fin. There are not many species, and there-
fore it might seem premature to divide the contents of

336 CLASSIFICATION OF FISHES.

so small a group into sub-genera; yet as this has already
been done, in part, by MM. Spix and Agassiz, we shall
adopt their divisions. Nearly all the types or sub-genera —
of Loricaria appear to be discovered; and the addi-
tional interest that attaches to them from this cir-
cumstance, renders it advisable to designate them - by
sub-generic names. One of the most remarkable of these
is Acanthicus Sp., where the whole head and body are
covered with short acute prickles, placed on the surface
of the osseous plates, and even on the first ray of the
fins: the caudal fin is very large, deeply lunated, and
has the outermost rays greatly lengthened: the cirri are
only two, and very short ; but even these are slightly
barbed, on one side, with setaceous hairs. This extraordi-
nary fish was discovered in the great river of Amazons.
Rhinelepis of the same author is entirely without these
prickles ; but the edges of the plates are crenated ; the
two cirri very short and fleshy ; and the tail moderate
and only slightly lunate. The Plecostomus of Gronovius*
is another, and a most interesting form, at once dis-
tinguished by the great length of its tail; while the
caudal fin has one of its external rays prolonged into
a filament, as in Acanthicus: this singular fish is
probably an aberrant type for it has no cirri; and it
thus opens an immediate passage’to the next genus, Hy-
postoma, where it is met by the Hy. etentaculata of Spix,
equally characterised by the absence of cirri, yet having
an adipose dorsal fin, which is the peculiar distinction
of this genus. There are only two species of these
double-finned Loricarie yet known from South America.
Following this, we now first characterise the genus Hop-
lisoma (H. punctata Sw., fig. 81.), the type of which is
the Cataphractus punctatus of Bloch. Unlike all the
mailed silures, the mouth is terminal, as in the generality
of fishes: but, it differs from all the others, in having the
the body compressed: the cirri are well developed; and
although the anal fin is short, the tail is long, so that
the vent is very close to the ventral fin: this is, in short,
* Zooph, pl. 2. fig. 1, 2.

SILURIDZA. — THE LORICARINA. SOF

a very distinct type, whether we regard the compres-
sion of its head and body, or the singular scales with

which it is covered. It is followed by another, to
which a variety of names has been applied, with little
regard either to priority, or propriety, of nomenclature.
To give, for instance, the name of Callichthys, implying
a beautiful fish, to one of the most ugly in the whole
family, is manifestly absurd: we shall, therefore, fol-
low Bloch and Lacepede, and retain the old name of
Cataphractus to that strange-looking fish, figured by
Bloch (pl. 377. f. 1.) with the specific name of depres-
sus. Its eyes are very small and nearly vertical ; the
head large and greatly depressed ; and it possesses,
altogether, the strongest possible resemblance to the
Aspredine: the mouth, indeed, can scarcely be said to
be beneath; but the dentated plates which cover the
whole body, in two series on each side, are precisely of
the same pattern and structure as those of Hoplisoma,
and plainly distinguish it from all the types of the Aspre-
dine. Last of all, we have the new genus Sturisoma
Sw., represented by the Loricaria rostrata of Spix, dis-
tinguished at once from all the foregoing, by its long
obtuse snout, which projects so considerably beyond the
mouth, that it has the greatest possible analogy to the
sharks. On looking to the slender form of this type,
and of the Loricaria maculata of Bloch, we see at once |
that the series of this sub-family forms a most perfect
circle: the last-named fish, in fact, would be a Stu7i-
soma, had it not the short muzzle of Loricaria.
VOL. «I. Z

338 CLASSIFICATION OF FISHES.

(279.) We are thus brought back again to the point
from whence we started ; and we find that the genera of
the Loricari@ describe a circle, whose analogies may be
thus expressed : —

Genera of the Sub-families of

LoRICARINE. Analogies. the SrLurIpE.
ip ee f Hod; seamed mailed; mouth piesa Ponckue
Hypostoma. f aoe always two; the a Practice

; Tail very long; vent close to the pec-
Hoplisoma. trata, & { siuunisz.
Cataphractus. Eyes very small, nearly vertical. ~ASPREDINE.
Siurisoma. Snout lengthened, depressed. SoRUBINE.

The representation which these genera of the Lo-
ricarine gives us of the primary divisions of the
whole family, developes their analogy, likewise, with
the primary orders of fishes. It is, indeed, a most in-
teresting circumstance to find that in such a group as
this we should have a representation of the Gymmetes, or
riband-fish, in the compressed shape of Hoplisoma, so
totally at variance with all the other genera; while the
small vertical eyes, and the general aspect of Cata-
phractus, remind us immediately of Aspredo.

(280.) The sub-family of the Pimenopinz is dis-
tinguished by positive and negative characters, or, rather,
of two which are universal, and of two which admit of
some partial exceptions. The mouth, in the first place,
is always terminal: the body, likewise, is naked ;
for although, in consequence of their close affinity to
the Loricarine, some few of these fishes have a
single row of small plates upon their sides, and even
on their head, yet these plates never extend beyond
the nape. They have all two dorsal fins; the hinder
being completely adipose, or without any vestige
of rays. Their more striking peculiarity, as already
remarked, is in the length of the belly and tail being
equal, so that the anal fin is never of that exceeding
length which it is in all the Silurine. Some idea of
the numerous modifications of form that enter into this
group, may be gathered by looking to the contents

SILURIDZ. — THE PIMELODINZ. 339

of M. Cuvier’s divisions (or sub-genera, as they are
termed) Pimelodus and Bagrus, — two artificial groups,
which we have found it impossible to understand: to
us they appear te be common receptacles for all Siluri
having two dorsal fins. Commencing, then, our survey
with ach Pimelodi- as seem to have the greatest
affinity with the last’ sub-family, we may notice the
genus Synodontes, where the head is mailed as far as the
first dorsal fin; but the body is entirely naked; the
‘ muzzle is naked and unusually narrow ; while the
lower jaw is armed with numerous hooked flexible
teeth, of a structure altogether unique, as M. Cuvier
observes, among fishes: it would appear, however, that
this is not strictly the case, for the teeth of Hypostoma
etentaculum of Spix are similarly formed ; while they
are, no doubt, articulated or flexible at their roots, in
the same way as in Synodontes ; otherwise they would
be useless, as the hooks at their end could not act.
Close to this type, well represented by Syn. Ruppelli
( fig. 82.), Cuvier places Lacepede’s genus Agenieosus* ,

founded on the military or horned silure of Bloch
(pl. 262.), and with which he associates the Pimelodus Si-
londia of Hamilton (pl. 9. f. 50.) ; but as both these
fishes have the belly much shorter than the tail, ie, do
* Ageniosus of our Synopsis.
Viewed

340 CLASSIFICATION OF FISHES.

not enter among these types. The two primary divisions
of this sub-family may be thus characterised : — In the
first, —to which we retain the name of Mystus, as
originally proposed by Artedi and all the old ichthy-.
ologists, —we see the general form of Pimelodus, united
with some of the characters of Loricaria: the parts
before the dorsal fin, for instance, are protected by bony
plates ; and on each side of the body is a series of
smaller ones, placed in a single row, so as to form a
lateral line, armed with prickles, as in our M. costatus -
(BI. 376.). This structure, in fact, is precisely analogous
to the plates of the genus Tachinus among the Scom-
beride, and many other instances of spiny or raised
lateral lines to be found in analogous groups. The simple
plates upon the head are not, however, confined to this
genus ; nor can they be looked upon as a primary generic
character; since. they occur in most of the Indian
Pimelodine, and even in some of the Silurine. In the
next or most typical genus, Pimelodus, the sides of the
body are always naked: the species of this group are
exceedingly numerous in the rivers of India, and several
are found in those of Tropical America. We have
failed, however, in discovering such characters as would
separate these species geographically: for although the
majority, if not all, of those with the head entirely
naked seem restricted to America, yet those in which
this part and the nape are covered with a buckler, are
found, with little or no essential variation of structure,
in both hemispheres. Among the Indian species, minor
differences will be observed, sufficient to constitute sub-
genera, easily recognised, and therefore much better
- understood than if we attempted to arrange them by their
teeth, —an attempt which has so signally failed in the two
““sub-genera” just alluded to, viz. Pimelodus and Bagrus.
In some of those Indian species which we suspect are
typical of the genus Pimelodus (as we now propose to re-
strict it), we observe that the adipose dorsal fin is of such
considerable length as almost to fill up the interval be-

GENERAL REMARKS ON THE PIMELODINA. 341

tween the first dorsal and the caudal, while it is very low
at both its extremities.* Another small group has the
adipose fin much shorter, and triangular: the cirri are
only four ; and the anterior rays of the first dorsal, pec-
toral, and caudal fins greatly elongated.t A third as-
semblage have the adipose fin very small, oval, and
pedunculated ; and here the number of cirri varies from
eight to four.{ It deserves some attention, that, in the
descriptions of the above, notice is generally taken of a
strong sharp spine, attached to the gill covers,—a charac-
ter which does not appear to be found in those Pime-
lodine of Tropical America which have been described
and figured by Spix and Agassiz. We feel convinced
that there must be some such difference between the
Pimelodine of these two hemispheres ; but this, as well
as many other points of structure, have been hitherto
passed over by our predecessors, so that we can come to no
determination on this point, at least in respect to such
species as we have not personally examined. As for
attempting to determine the natural sub-genera by the
serrature or smoothness of the dorsal and pectoral spines,
or even by the presence or absence of naked plates upon
the head, we have altogether failed in the attempt. The
admirable descriptions, indeed, of Dr. Hamilton, who
enumerates no less than pie Pimelodi found in
the Ganges alone, is quite sufficient to convince the ich-
thyologist that the above characters are merely specific,
and therefore quite unfit to be employed for the con-
struction of sub-genera. One species, in fact, will some-
times differ only from another in having the pectoral
spine either smooth or serrated; a third, in the obvious
line of affinity, will have it barbed on one side only, or
on both ; and sometimes the side where the dentations
are situated, is different in one species to that of the next.
The same remark is applicable to the teeth, even on M.
Cuvier’s confession ; and the descriptions of Dr. Ha-

* Gangetic Fishes, pl. 11. fig. 67. + Ib. pl. 7. fig. 62.
t Ib. pl. 31. fig. 57., pl. 23. fig. 60. 66. Russell, No. 169.

Zo

342 CLASSIFICATION OF FISHES.

milton are to the same effect. The teeth, in fact,
throughout the whole of the Siluride, are so minute,
that their structure can seldom be determined, except
under the power of a high magnifier. The value of a
character can only be determined by its extent; and,
decording to this rule, we find that it either serves to
designate a family, or a species. The fins of fishes, as
every one knows, are analogous to the wings of birds;
and we were the first, in conjunction with professor
Temminck, who’employed the modifications of the pri-
mary quills as characters for the discrimination of the
natural groups among birds. There is consequently
good reason to suppose that the same importance should
be attached to the structure of these organs among fishes;
and the proofs offered in this volume, for such an hypo-
thesis, are not few. The number of the cirri among the
*Siluride seem more constant than the characteristics
of which we have just spoken: but they do not appear
to be sufficient for sub-generic distinctions, when not
supported by other and more constant characters.
(281.) Leaving now the two typical genera of the
Pimelodine, we may particularly mention three others,
each of which presents some very prominent and pecu-
liar characters ; these we shall look on in the light of
generic types—not so much from our belief that the two
first are really so, but that they may be kept for the
present distinct from the two presumed typical groups.
The first of these is Siser*, which has the head broad,
depressed, covered with bony plates marked with tu-
bercles and ridges, and terminating behind in three nar-
row processes. The cirri are more numerous than in
any other Silurvi yet discovered — amounting to no less
than fourteen, all of which, however, are shorter than
the head ; the lips fleshy; and there are no teeth: the
eyes are small, and very high. Between the head and
first dorsal fin is a bony plate; and near the latter are
five small plates, nearly united, and disposed in a double

* Sisor rabdophorus, Hamilt., p. 208.

SILURIDZ. — THE PIMELODINZ. 343

row: the back, between the first dorsal and the caudal,
has a tuberculated bony ridge: the vent is before the
middle. The dorsal fins are two: the anterior has the
first dorsal ray hard at the base, but soft at its point,
the hinder part being indented ; the second dorsal has
only one short prickle, with a short membrane behind :
the pectoral spine is strong and doubly serrated ; ventral
fins small; anal fin of six rays only. The caudal fin is
very remarkable: it is lunate, and of ten rays; the
uppermost one of which is so much prolonged, like a
whip or rod, as to exceed the whole length of the head
and body. This great developement of the caudal or tail
fin, and the partially mailed plates upon the nape, lead
to the belief that this extraordinary fish forms the pas-
sage, in conjunction with Synodontes, between the Lori-
caring and the Pimelodine. We have not, however,
yet seen a specimen ; and unfortunately Dr. Hamilton’s
description, which we have here abridged, is unaccom-
panied by any figure. There is some distant resem-
blance between Sisor and the Silurus Bagre of Bloch:
this latter, however, we shall for the present keep dis-
tinct, under the name of Breviceps. In this the cau-
dal fin is also lunate; but the points are not extended,
like those of the first dorsal and pectoral rays, which
reach to a very considerable length: the head is par-
ticularly short and obtuse, with the eyes very remote;
and there are only two pairs of cirri, one of which is
very short. The upper jaw is slightly longer than the
under: it is very large, and armed with numerous sharp
teeth, very small, placed in two broad rows on the upper
jaw, and in one on the under: finally, the tongue is
very large, thick,.and rounded. We mention these par-
ticulars in this place, from a belief that this type shows
us the fissirostral genus of the Pimelodine, or that by
which the sharks are represented in this sub-family.
The third genus is that by which we distinguish the
Pimelodus cyclopium of Humboldt (Cyclopium Hum-
boldtii Sw., fig. 83.), which, because it has an adipose fin,
has been suffered until now to remain unnoticed, even as
Zz A

344 CLASSIFICATION OF FISHES.

a sectional division of “ the great genus Pimelodus.’ A
more beautiful representation of the Aspredine, and

especially of Astroblepus, cannot possibly be imagined :
the cirri are only two; and, but for its adipose dorsal,
this singular fish so much resembles Astroblepus, that it
might be placed in the same genus.

(282.) Imperfect as is this sketch of the Pimelo-
dine, we feel persuaded, that, when all the fishes that
have been referred to this group are well understood, it
will turn out to be one of the most perfect of all the
circles in ichthyology. The forms slightly intimated
by Cuvier under the heads of Pimelodus and Bagrus,
are sO numerous, so varied, so singular, and their cha-
racters are so differently combined, that every shade of
variation promises to exist in these two magazines of
forms. With the hope of finally working out the whole
of these, we have long been collecting drawings and spe-
cimens of all the species; and we take this opportunity
of soliciting the assistance of all those wellwishers to
Science, who, by residing near the great rivers of India
or America, have it in their power to assist us with
correct drawings made from the fresh subjects. Nu-
merous as are the Siluride yet discovered, we believe
that as many more remain unknown; and that, when all
the subordinate types are discovered or ascertained, each
genus will possess its five sub-genera, representing those
of the sub-family we shall now enter upon.

(283.) The true Srnurin# form that division of the
family which represents the eels or apodal fishes: this
relation is at once manifested by the excessive length
of the tail, in comparison to that of the abdomen (as

THE SILURINZ, OK EEL SILURES. . 345

seen in Silurus laticeps, fig. 84.): it consequently re-

sults that the vent is much nearer to the pectoral! fin
than to the caudal, and that the anal fin is unusually
long: this is almost the universal character of every
fish within the circle; although in some of the aberrant
genera it is, of course, less conspicuous than in the
more typical. The first type that meets us, after quit-
ting the Pimelodine, is Ageniosus, at present composed
of only three known species, which belong to two sub-
genera. In Ageniosus proper, we place the Silurus mi-
fitaris of LLinn.*, remarkable for having the intermax-
illaries developed in the form of two suberect bony
and serrated spines, resembling horns, which are placed
before the eyes: the head, as in Breviceps, is broad and
depressed ; the mouth very wide; and the first ray of
the dorsal greatly elongated. The second sub-genus we
have named Siloniat (S. lurida, fig. 85.), as apparently

the true type of the genus. Although in this type there are
two cirri, yet they are so small as to be nearly obsolete :
it thus seems to connect Ageniosus to the genus Pachyp-

* Bloch, pl. 362. + Pimelodes silondia, Hamilt. pl. 7. fig. 50.

346 CLASSIFICATION OF FISHES.

terus, where the cirrci are fully developed. In other re-
spects, the general form of Ageniosus and Silonia are very
dissimilar. In the latter the head is short and thick, but
very slightly widened or flattened: the whole fish, in-
deed, is much compressed, and the lower edge, as a
Hamilton says, is ‘‘ sharp, somewhat like a knife,”’-o0

in other words, carinated: the mouth is moderate ; a
eyes are large; and the whole appearance of the other
parts is more in accordance with ordinary fishes than
with any of the present family. One species is dia-
phanous *, the viscera being inclosed in a bright silver-
coloured membrane, Sees to that of Argentina:
the two cirri are even more minute than in the last,
being scarcely visible without the aid of a magnifier ;
and in both these speces the gill membrane has nume-
rous rays. As the other subordinate types of this genus
remain to be determined, we shall hazard no conjectures
upon them, but at once proceed to the next, which we
propose to designate Pachypterus. In this there is stilla
second adipose dorsal fin, but the cirriare very conspicuous ;

and the tail is considerably more lengthened, so as to
be often near three times the length of the abdomen: it
is bordered beneath, for nearly its whole extent, by the |
anal fin. We cannot, at present, determine more than
two of the sub-genera belonging to this group ;—one from
India, which is the typical, includes many species ; the
other, whichis the 7 ENTE ITS of Spix (fig. 86.), andof

which two are known, seems peculiar to the rivers of Ame-

rica. The second dorsal fin is reduced to so small a size,

thatit becomesalmost obsolete. Wenextpass tothe genus
* Chandramara, Ib. p. 162.

THE SILURINZ, OR EEL SILURES. SAT

Silurus, the most. typical group in this sub-family, all
of which are known by a single and very short dorsal
fin, a very long anal fin, and a distinct caudal, either
reunded or lobed. We may enter the genus Silurus by
the Silurus garua of Dr. Hamilton, which forms the
type of our sub-genus Clupisoma: this singular fish
is almost a counterpart of Silonia, except that it has but
a single dorsal fin, and the muzzle is provided with
eight moderately long cirri. Had we not analysed this
group more than the others, we should certainly have
imagined these two singular types passed into each other ;
and, indeed, so long as the circularity of the genera
Pachypterus and Ageniosus remains undetermined, there
is still a probability of such being the case: yet this,
although it might somewhat affect the contents of those
two genera, would not alter this; for in its fins and cirri
it is so truly a Si/urus, that even Dr. Hamilton refers it
to that genus. Following this, we place the sub-genus
Callichrus Ham. , distinguished by all the species having
a forked tail. T hese fish, as the doctor observes, are
rather handsome, and have little or nothing of that lurid
appearance by which the neighbouring species are dis-
tinguished: they are all very rich, fine-flavoured food ;
and grow to from nine to twelve inches in length: they
are chiefly found in the ponds and ditches of Bengal;
and, no doubt, many more than the five species already
known remain undiscovered. The caudal fin is always
lobed ; the sides of the body are sub-diaphanous ; the
head is only slightly flattened (thus presenting an affi-
nity, in these two last characters, to Clupisoma), the
mouth large, and the upper jaw much longer than the
lower: the eyes are small, and are on the sides, noe
towards the top, of the head; while the gill membran-
has nine rays. Quitting these bright-coloured fishes,
we enter among those to which we retain the sub-ge-
neric name of Silurus. They are at once distinguished
from the last type by having the caudal fin rounded : the
body is much compressed, while the head is remarkably.
depressed: the mouth is moderately large, and opens

348 CLASSIFICATION OF FISHES.

horizontally: the first ray of the dorsal and pectoral fins
varies in almost every species; sometimes it is smooth,
and sometimes serrated on one or on both sides. The
head does not turn upward; and the jaws are about
equal. The Silurus glanis, already mentioned, together
with S. fossilis (Bloch, pl. 370. f. 2.) and S. laticeps,
fig. 84.), are typical examples of this structure. Follow-
ing this, we place the sub-genus Malapturus of Lacepede,
represented by the electric silure of the Nile,— a fish
which is said to possess all the properties of the torpedo
and of Gymnotus electricus, although in an inferior de-
gree. This sub-genus is distinguished from all the rest
of the Silurine, by having the small dorsal fin entirely
adipose, and placed close to the end of the tail, adjoin-
ing the caudal fin. The head is turned upwards, but
the upper jaw is manifestly longer than the under.
Two very interesting species* have recently been figured
from general Hardwick’s collection of Indian drawings:
in one of these there appears a series of small spines
placed all along the back, in front of the dorsal; but as
this is the only apparent departure from the typical
structure, we retain it in the present division. The
spines in M. Cuvierii are certainly analogous to those on
the back of the Acanthonotus of Bloch, but we cannot
for a moment entertain the supposition that these two
fishes belong even to the same order.

(284.) The very small size of the ventrals in the
sub-genus Malapturus deserves notice, for they are
almost obsolete, being nearly one fourth less than the
pectoral fins. The last type of form which we imagine
to enter within this group is Pusichthyst ; it has, indeed,
the body and fins of Callichrus ; but the snout is turned
upwards; and although the head is unusually broad, the
mouth is excessively small, and opens vertically, just as

* Malapterurus (Ailia) Bengalensis Gray, Silurus (Acanthonotus) Cuvi-
ertiaGray, Ind. Zool., vol. i. pl. 85. fig. 1,2. The probability is, that Mr.
Gray did not, at the moment, recollect, that the name of Acanthonotus
had been already used.

t+ The Arabians call these fishes Schilbe, but Cuvier, by some oversight,
has omitted to give them a classic name: their humped back has sug-
gested that of Pusichthys.

THE SILURINZ, OR EEL SILURES. 349

we see in the genera Mormyrus, Chironectes, Uranosco-
pus, Trachinus, and numerous others. So completely,
indeed, do these Silurine remind the ichthyologist of
the chironectiform type, that Dr. Riippell has given the
specific name of Uranoscopus to a species he has recently
discovered in the Nile. This testimony of its analo-
gical relation to the order Plectognathes is highly satis-
factory, since it is given, as it were, incidentally, or as
an insulated fact, without any idea of the inference we
have drawn from it.

(285.) We now pass on to the fourth division or
genus of the Silurine, to which we retain the name of
Plotosus Bloch. With but one exception * (Heterobran-
chus bidorsalis Geoft., fig. 87.), all tne fishes it contains

have but a single dorsal ; but then this fin, instead or
being short and high, is very long and low,—so long,
indeed, that it extends to nearly the whole length of the
back ; the anal fin is nearly of the same length; and
both either terminate just before they reach the caudal,

SL
S

as in the sub-genus Clurias (fig. 88.), or are actually
united to that fin, as in Plotosus (fig.’79.), the head is still

* M. Cuvier has placed under Heterobranchus, certain fishes belonging to
the genus Clarius of Gronovius, which was subsequently called Macropte-
ronotus by Lacepede. As,I consider these latter to belong to a distinct
type, they are so designated in the Synopsis, under the original name of
Clarias, imposed on them by Gronovius, which I see no occasion for alter-
ing. If these fishes have the same ramified branchia as Geoffroy’s Hefero-
branchus bidorsalis, an additional sanction is given for placing the latter
fish in the genus Plofosus, notwithstanding its possessing two dorsal fins.
I must confess, however, that I am not quite satisfied on the true relations
of this singular type.

350 CLASSIFICATION OF FISHES.

greatly flattened, and the eyes small; but the lower jaw
is shortest, and the cirri are usually eight.* Here, then,
we have the true anguilliform or eel-shaped silures, dis-
‘tinguished in the most remarkable manner from all
those we have yet noticed ; and yet so closely allied are
they to the last genus, in every thing but their long
dorsal fin, that their affinity becomes apparent to every
one: the subordinate types or sub-genera, however, are
as yet very few, and therefore we cannot state how far
the contents of the circle represent that of the last, nor
can we trace the connection to the fifth genus,—a genus,
however, too remarkable to be confounded with any
other; we allude to Cetopsis of Agassiz (fig. 89.). These

silures present us, in their long anal fins, with the primary
character of the Silurine ; but in all other respects show
a marked resemblance to Cyclopium in the circle of the
Pimelodine, and to Astroblepus among the Aspredine :
like these, and all other of their representatives, the eyes
are remarkably small—indeed,so minute, that they appear
as mere specks in the skin, — and their situation is nearly
on the top of the crown: the dorsal fin is single, trian-
gular, and placed near the crown, which is thick and
obtuse: the mouth is large, and the gape obliquely ver-
tical: there are no dorsal or pectoral spines: the tail is
hardly longer than the body ; and the four cirri are so
very short, as not to be so long as the head. Now, it
is to these two last characters that we must call the na-
turalist’s attention. In commencing our survey of this
sub-family, it will be remembered that the first genus,

* Batrachus, Bloch, pl. 370. fig. 1.; M. Haméltonz, Hamilt. pl. 26. fig. 45.

ANALOGIES OF THE SILURINA. 351

Ageniosus, is remarkable for two things ;—one, the slight
excess of the tail, in regard to its length, over that of
the belly ; and the other, the almost total disappearance,
in Clupisoma, of the cirri: these two characters are pos-
sessed by Cetopsis ; and by these points of affinity we
consider they pass into each other, and close the series
of the Silurine, which thus becomes, in regard to its
genera, a circular group.

(286.) While the affinities we have been tracing,
and the forms of the genera and sub-genera, of the sub-
family Silurine,.are still before the reader, we will
finish our exposition of the whole by locking to their
analogies. Commencing with the genera of the Silurine,
we find they succeed each other in the following order ;
and by placing the sub-families of the whole group also
in their own series of affinity, we shall get the follow-
ing parallel analogical characters :—

Genera of the Sub-families of

Analogical Characters.

SILURINE. the SILURIDE.
Pachypterus. Dorsal fins universally two. PIMELODINE.
Silurus. : Only one dorsal fin in the typical forms. LorIcARINZ.

* Rostrum round, obtuse, projecting*; ae ees
Plotosus. mouth placed eneuth 22 2a SORUBINE.
Cetopsts. yes small or minute, placed towards a ASenronniad
Ageniosus. Body unusually compressed. SILURID.

Some interesting facts, connected with the general
structure and relations of these groups, will result from
a more particular exposition of these analogies; resem-
blances which, we confess, did not occur to us until after
we had deemed it necessary to submit our arrangement of
the Siluride to this test. In comparing Pachypterus
with the Pimelodine, we see that, throughout both
groups, the second adipose dorsal fin is invariably pre-
sent ; and that there is the strongest resemblance be-
tween the two is sufficiently established, by the fact of
all authors having hitherto actually referred them to the
same genus,—an error that has solely arisen from over-
looking the great difference in the developement of the

* Russell’s description of his Platystachus anguillaris, vol. ii, p. 51.

352 CLASSIFICATION OF FISHES.

tail, existing between the Pimelodine and all the true
Silurine. In the typical examples of the two next
groups which stand opposite each other, namely, the
genus Silurus and the sub-family Loricarine, the
adipose dorsal does not exist; and it is further to be
noted, that these two genera—Silurus and Loricaria
—are the pre-eminent types of their respective circles.
Plotosus and Sorubium represent each other most com-
pletely by their ‘‘very large, long, and depressed
head * ;” and by the snout projecting over the lower
jaw, so as to render the mouth inferior, as in the Squa-
lide and the other typical cartilaginous fishes. Equally
beautiful is the analogy between Cetopsis and the minute-
eyed genera Astroblepus and Aspredo ; and we find a
perfect representation of all three in the genera Cyclo-
pium, as well as in the types of the cheloniform fishes.
We next find Ageniosus standing, in our table, opposite
to the Siluride ; and this analogy at first sight appears
forced, because, as the silures represent the anguilliform
fishes, it would seem to follow that Plotosus, from
having all the hinder fins united, should be compared to
them, rather than to the Sorubine: but further con-
sideration on the nature of the anguilliform type will
show this reasoning to be fallacious. Great length of
tail is,indeed, one of the chief indications of the anguilli-
form structure ; but itis not the only one, for, through-
out the greater part of the representations of this form
among the acanthopterygious order, the body is exces-
sively thin; and analysis proves that the highest de-
velopement of this particular character is, that the body
becomes semi-transparent or diaphanous. Now, by far
the greater part of the Silurine are remarkably thin
fishes, — that is, the body is very much compressed,
although the head is remarkably depressed. But it is
only in our genus Silonia (one of the types of Agenio-
sus), that we find both head and body compressed, and
this to a much greater degree than in any other forms
of the sub-family we are now investigating. Hence it

.* Russell thus describes the head of Plotosus.

SILURUS.—THE SUB-GENERA. 353

follows that, if Silonia had not been discovered, we
should have had no precise representation of Cepola and
"all the other riband-like fishes ; so that, under this view
of the subject, what appeared to be an anomalous form
among the Si/urine, turns out to be one almost neces-
sary to preserve that system of representation, and that
uniform variation of structure, we trace in all the other
groups.

(287.) As we have already shown that the primary
divisions of the whole family represent the primary
order of fishes, we need not submit the genera of the
Silurine to the same test; because, if they represent, as
we think we have now proved they do, the sub-families
of their own circle, their analogy to the orders becomes
established. We shall, therefore, now call the reader's
attention to the different sub-genera of the restricted
genus Silurus, in order to inquire whether these sub-
genera do not, in the same manner, represent the genera
of the sub-family ; the affinities of these sub-genera have
already been explained. We shall now, therefore, glance
at their analogies.

The sub-genera of Siluxwshavebeen The genera ofthe Siluring as already

. seen to follow each other thus— _ stated, stand in the following order—
Callichrus. PACHYPTERUS.
Silurus. SILURUS.
Malapturus. PLOTOSUS.
Pusichthys. CETOPSIS.
Clupisoma. AGENIOSUS.

Our proposition is, that each of these divisions are
mutual representations of each other, independent of all
those affinities by which the contents of each separate
series are united ; and we shall now endeavour to sub-
stantiate its truth. In the first place, we see Callichrus
standing opposite to Pachypterus; and we find that both
have the body brightly coloured, and the caudal fin
lobed, in contradistinction to Siluwrus, where the colours
are dark and lurid, and the caudal fin rounded. That
there is the strongest resemblance between Malapturus
and Plotosus, may be established by the authority, if

authority is required, of Bloch, who places them close
VOL. I. AA

354 CLASSIFICATION OF FISHES.

together ; in which respect, notwithstanding the essen-
tial differences of their structure, he is closely followed
by Cuvier. We now come to Pusichthys: when we look
to the small and nearly vertical mouth, we at once see an
exemplification of those peculiarities by which Urano-
scopus, and all chironectiform types are so singularly
distinguished ; this, in fact, is the only sub-genus of the
above group that has the mouth opening vertically, while,
by the turning up of the snout, it preserves a resemblance
to Malapturus. The analogy of Clupisoma to Agenio-
sus (or, rather, to its type, the sub-genus Silonia) is very
striking ; the exceeding long cirri of Clupisoma and its
single dorsal fin, at once show its affinity to Silurus ;
and yet, setting these peculiarities aside, Clupisoma and
Silonia are so like each other, and yet so different from
all their congeners, that they appear almost related by
affinity. Cilupisoma has the same compressed head and
body, the same fulness of the eyes, the same sharpness
on the edge of the belly, the same semi-transparent sides,
&c., as we see in Silonia; yet it has but a single dorsal
fin: we are at no Joss to recognise, in both, a type of the
Gymnetres, or riband- fish. Further comment upon these
analogies would be superfluous. We now return to
the affinities of the remaining sub-families.

(288.) The fourth principal division of the Siluride
is that of the Aspredine; the typical distinctions of
which are as follows: not only the head, but the ante-
rior part of the body, is considerably flattened, while the
tail preserves some resemblance to that of the last sub-
family, the Si/urine,in being generally slender ; the eyes
are remarkably small, and placed on the crown of the
head, so as to become vertical; the branchial aperture
is spiraculated, as in the cheloniform fishes, and consists
of merely a slit of the skin; while the operculum itself,
“* unlike,” as Cuvier observes, “all other osseous fishes,
is immovable.” These are the positive characters of
these singular fishes; but in other respects they vary
considerably. In our new genus Cotylephorus (C. Blochii,
fig. 90.), which connects these to the Silurine, the anal

SILURIDH.—THE ASPREDINZA. 355

°

fin is very long; but in the more typical forms, such as in

ee

2°
Boe OSPR BOLO OS

Eee an ee L
iN) ft Se @& ig

a ms Zan ey

SSE S

SS

the true genus Aspredo Art. (fig. 80.), this fin is short.
There is but one dorsal fin, which is small, and placed
nearest the head; the first dorsal, and the first of the pec-
toral rays are generally very strong, and more lengthened
than usual, and the barbels or cirri always present, although
variable in their number ; the muzzle projects slightly
beyond the under jaw, which is transverse. Cuvier
observes that the immobility of the operculum is occa-
sioned by the pieces which compose it being cemented,
as it were, to the tympanic bone and to the pre-oper-
culum. These characters apply, more or less, to the fol-
lowing genera :— Platystachus Bloch, Astroblepus Humb.
( fig. 91.), and Aspredo of Artedi and Gronovius ; several

ES

7

i> OE
Uf

yy

Hf
'
Y
{tly Ly
(
|

others will doubtless be added, when the subordinate
forms in this family are better understood. Now, it
will be remembered that we have already designated
several genera, equally remarkable with these for the
smallness of their eyes, and their depressed head ; such,
for instance, as Cataphractus, Cyclopium, and Cetopsis ;
but 3 in none of these do we find the depressed body,
AA 2

356 CLASSIFICATION OF FISHES,

the fixed gills, or the spiraculated aperture of the
Aspredine: and all their other points of structure agree
with the sub-families in which we have arranged them.
Thus, Cataphractus has the perfectly mailed body and
the two dorsals of the Loricarine; Cyclopium has the
naked compressed body, and the adipose fin of the Pi-
melodine ; while in Cetopsis there are all the general cha-
racters of the anguilliform Silurine, except that it has
the minute vertical eyes of Aspredo: hence it follows,
that all these become representations only of the present
group in their own circles ; they have, in short, a strong
resemblance of analogy to Aspredo, but without any ab-
solute affinity, seeing that the preponderance of their
general characters, in all other respects, places them else-
where. The passage from the Aspredine to our next
division, appears to be made by two singular-looking
fish, which Gronovius erroneously places in the genus
Mystus: one of these has the minute vertical eyes of
Astroblepus and Aspredo, with the body and fins of
Pimelodus, or Sorubium,; while the other, in all exter-
nal points, evidently belongs to the same type as Spix’s
Heterobranchus ; for both these latter have the elongated
head, and the unequally lobed caudal, rounded at the
end of its divisions, which is such a general character
among the Sorubine, although no fishes among the
Pimelodine, that we know of, have the caudal fin so
consiructed ; neither is the first dorsal ray, as in these,
slender and unarmed. We may thus trace the passage
from the Aspredine, and proceed to the next group.
(289.) The Sorvsin= constitute the last division of
this very extensive and intricate family. Although
possessing something of the general appearance of the
Pimelodine, they are readily distinguished from them,
and indeed from all the other silures, by an unusually
long and large head, having the muzzle very broad,
flattened, and prolonged cver the under jaw; so that
the mouth, by this formation, is situated beneath (fig.
92.). For the comparatively recent discovery of this
highly interesting group, we are indebted to the re-

,

SILURIDA. — THE ASPREDINZ. 357

searches of Spix, and the masterly descriptions of Agassiz.*
Considering the immense number of forms, crowded,

as this also is, in the old genus Pimelodus, it would be
quite premature to judge of the contents of this sub-
family merely from the few species that have as yet
been placed in it. It is sufficient, however, for our
present purpose, that these are too remarkable to be
arranged under any other division of the family, more
especially as they are the only silures having that large,
long, flat muzzle, which characterises the cartilaginous
order of fish, the fissirostral type of birds, the aquatic
order of quadrupeds, and the saurian reptiles. It is
by this form, also, that the great circle of the Siluride
is closed; for, in the genus Sturisoma, we have the
muzzle and elongated form of Sorubium, joined to the
mailed body of the Loricanine.

(290.) Before concluding our survey of the Siluride,
we shall here introduce a few remarks upon two singu-
lar types, which will enter among these fishes; but of
whose precise situation we feel doubtful. These are
Eremophilus Humb., and Heterobranchus Spix: both
these are described in two separate works, valuable,
indeed, as specimens of typography tT, but of such enor-

* We find an incidental notice of this group of Spix’s in the Régne Anz-
mal, under the wrong name of Sorubim, no doubt a misprint for Sorubiwm..
+ There is not a more serious drawback to the studies of the great ma-
jority of naturalists, than the publication of those national works, as they
are called, which are intended to commemorate the scientific expeditions
fitted out at the expense of the European governments. From the idea (we
think a mistaken one) that no expense should be spared to render these
publications as magnificent as possible ; they are published at such a price
as absolutely to debar all but princely naturalists from deriving any real
benefit from them. The grand ouvrage sur l’ Egypte, and the folios of
Humboldt, are well-known instances of this typographical luxury ; not to
mention others of our own country, published by subscription, at the cost
_of between two and three hundred pounds. These publications, indeed,
are professedly intended to promote science, but we have long thought

AAS

358 CLASSIFICATION OF FISHES.

mous expense, as to be beyond the reach of private pur-
chase ; nor are specimens of either of these fishes to be
seen in the London museums. The annexed cut of Eve-
mophilus ( fig. 93.) will show how much more it resembles

the Siluride, in general structure, than any other family ;
while the smallness of its eyes, and its single dorsal
fin, gives it every appearance of belonging to the sub-
family of Aspredine : that it will form a distinct type,
cannot be questioned, although it has not even been
mentioned by M. Cuvier. If the figure is correct, it
would seem that the pectoral fins are placed on a pe-
duncle, so as to form a joint at the base, in precisely
the same manner as those of the frog-fish, or Chironec-
tide ; and hence we are led to infer that it is the re-
presentation of those fishes in the circle of the Aspre-
dine, in which case it would stand as the most aberrant
type, intermediate, perhaps, between Cotylephorus and
Mystus of Gronovius. Of the genus Heterobranchus
we have already spoken. Of the JH. bidorsalis, our
figure (87.), reduced from that in the great work upon
Egypt, will give the reader a correct idea of its general
form: but not having the means of consulting the

that they produce a contrary effect. Five or six hundred pounds will but
barely supply the working naturalist with such books as he must absolutely
possess, if his researches are extensive, and he writes for the public: the
additional purchase of such works of luxury as we here allude to, is, there-
fore, quite out of the question ; and thus, they do not advance, but injure
and impede science, by being so high priced as to debar nearly all natur-
alists from their possession. Fortunately, however, this extravagant taste
is giving way to a more rational and useful mode of publication. The
enterprising proprietors of THE NatTuraList’s Liprary have set: an
example which we trust to see followed: they have, perhaps, erred in
the other extreme; but this is, considering the pernicious effects of the
folio system, a minor evil. All we now want is a series of ov7ginal volumes
of the same sort, to place the study of zoology, in all its branches, within
teach of every one who desires to cultivate it.

SILURIDZ.—THE SORUBINA. 359

original definition of this group by Geoffroy Saint- Hilaire,
we shall here repeat what has been said on it by Cuvier.
He observes: “ It has the head furnished with a rough
flat buckler, broader than in any other Silurus, because
the frontals and parietals give out lateral plates, which
cover the orbit and the temple: the operculum is still
smaller than in the preceding genera (as Doras, Age-
niosus, &c.); and what distinguishes these from all
other fishes, is the peculiarity observed by M. Geoffroy,
that, independently of the ordinary branchie, they have
(an) apparatus ramified like trees, adhering to the
superior branch of the third and fourth branchial arch,
and which appear to be a sort of supernumerary gills:
for the rest, their viscera resemble those of the other
Siluri. Their branchial membrane has from eight or
nine to thirteen or fourteen rays: their pectoral spine
is strong and dentated; but nothing of this is to be
found in the dorsal and anal: their body is naked and
elongated, as are also the dorsal and anal fins; there
is no spine to the dorsal: the caudal is distinct. Those
which are known, have eight barbels: they come from
the Nile, from Senegal, and from some rivers of Asia.
Some, (forming the genus?) Macropteromus Lac.
(or?) Clarias Gronov., have but a single dorsal, alto-
gether radiated: others have a radiated and an adipose
dorsal.” The Heterobranchus 5-tentaculatus of Spix
(fig. 94.) appears to belong to a different division of

the family ; and the annexed cut will show it has a

much greater resemblance to the Sorubine than to the

Silurine. In respect to the singular structure of the
AA 4

360 CLASSIFICATION OF FISHES.

ramified branchia, it is even a more extraordinary de-
viation from the usual form than that of Syngnathus ;
so that, if the latter genus deserves to be placed as a
distinct division of the osseous fishes, according to
Cuvier’s system, Heterobranchus should form a third.
We believe, however, that this variation in the bran-
chia is not simply confined to one type only of the
Siluride, but to several; at least, the fishes that are
said by Cuvier, Geoffroy, and Spix to possess these —
ramified branchia, are widely different in nearly all
other parts of their structure. It is, in short, by this
means, that nature indicates the analogy which the
aberrant Siluride, as a whole, bear to the amphibian
sirens, which they represent; just in the same manner as
the genus Cinclus among perching birds, by its aquatic
or amphibious habits, represents the grallatorial order.
To make this circumstance a reason for arranging
Cinclus as a primary division of the perching birds,
would appear to us as great a violation of nature as
that of making Syngnathus, and all these Heterobranchia,
types of distinct divisions in the osseous fishes.

(291.) The family of the Copirmp2, or loaches,
have the greatest affinity to the last, near to which al!
ichthyologists have placed them. Like the silures,
they are fluviatile fish, generally lurking close to the
ground ; and they probably feed in the same way, for
the mouth is generally furnished with barbels, and the
body is slimy. In all other respects, however, they
materially differ, both from the Siluride and the
Cyprine, or carps, in their structure, but more especially
in their mode of propagation ; for they are all vivi-
parous, like the cartilaginous fishes, and thus stand alone
among the malacopterygious families, as presenting
almost a solitary exception * to the oviparous nature of
all the others. -The analysis we have given of the
Siluride demonstrates that this remarkable group does

* Some of the Silurid@ are stated to be also viviparous ; but we know too
little of these fishes to sanction the belief that this mode of propagation is
general among the great majority of the genera.

THE COBITIDZ, OR LOACHES. 361

not belong to them; and although we are not pre-
pared, at present, to adduce equally conclusive evidence
that the loaches are excluded from the Salmonida, our
investigations, as we have shown, have been carried
sufficiently far to render such a supposition highly im-
probable. The viviparous nature of the Cobitide ne-
cessarily carries with it certain peculiarities in the
generative organs, not necessary, indeed, to be here
detailed ; yet of great importance in determining the
rank of this family in the circle of malacopterygious
fishes.

(292.) The loaches, in comparison to the silures,
are a very small group, whether we regard the number
of species and forms already known, or the size of the
fish themselves. They appear to be more numerous in
the Indian rivers than in those of Europe, while those
genera, which are found in the New World, are altogether
peculiar. The common loach of Europe (Cobites bar-
batula Linn., fig.95.), not unfrequent in England, is a

perfect example of the typical structure, and Dr. Hamil-
ton has given us a few particulars of other genera,.
which are almost exclusively found in India. The com-
mon loach is generally met with in small running brooks,
where, lurking under stones, it searches for worms and
insects ; and it swims rapidly when disturbed. It spawns
early in the spring, and is very prolific: although it
seldom exceeds four inches in length, its flesh is es-
teemed a great delicacy ; so much so, indeed, as to have
induced many wealthy persons on the Continent to trans-
port them to their own waters. It is said that they owe
their introduction into Sweden to Frederick I., who, at

362 CLASSIFICATION OF FISHES.

much trouble and expense, had them brought from
Germany for this very purpose. The pond loach, Cob.
fossilis, is another European species, but has not yet
been found in Britain ; it is larger than the last, and
sometimes grows to a foot in length. When the little
pools it inhabits are frozen, or even dried up, it hides
itself in the mud, where its tenacity of life enables it to
live a long time: during stormy weather, like several
other ground-feeding fish, it comes to the surface and
agitates the water. In reference to this fact, Mr. Yarrell
has well observed that “ such fish as habitually reside
near the bottom of waters, have a low standard of re-
- spiration, and a high degree of muscular irritability.* In
such animals there is reason to believe there also exists
great susceptibility of any change in the electrical rela-
tions of the medium in which they reside.” This, indeed
is proved by the restless movements of eels, loaches,
and other ground fish, during storms of thunder, &c.
which, as they effect a change in the electrical state of
the atmosphere, extend, in all probability, the same in-
fluence, at least in a considerable degree, to the water.
The pond loach, in fact, verifies this latter supposition,
for, according to Ehrman, it is constantly swallowing
air, which it discharges by the anus, after it has been
changed, by passing through the body, into carbonic
acid. Besides the peculiarity in the mode of propaga-
tion, as already remarked, which the loaches possess,
there are other points in their anatomy which equally
forbid us to class them in any other family. Mr. George
Daniell has communicated to Mr. Yarrell the following
peculiarities in the osteology of the common loach, which
we shall here insert. <“‘ Attached to each outer side of
the first and second vertebre is a hollow sphere of bone
of equal size, between which, on the upper surface, the
vertebre are distinctly seen ; although the union of the
two spheres underneath hides the vertebre when looked
towards from below. These circular bones, which are
hollow, and the smooth insides of which can be seen

* British Fishes, vol i. p. 22. 377.

THE COBITIDZ, OR LOACHES. 363

through a horizontally elongated aperture that exists on
the outer side of each,—these bones are analogous to
the scapule: to their outer surfaces the bones of the
proximal extremity of the pectoral fins are articulated ;
and the fin, being moved by powerful muscles, produces
that rapidity of motion observable in this little fish.”
Another peculiarity, existing in the upper surface of the
head, is the want of union in the two parietal bones at
the top,—a deficiency which has been noticed by the late
celebrated naturalist, Guilding, to occur in the Iguana
tubereulata, or common guana of the West Indies.
“This peculiarity in the loach,” observes Mr. Yarrell,
“‘ is another instance of a relation in structure between
the fishes and reptiles.”* How far these peculiarities of
the true loaches extend to the genus Anableps, and the
other groups, is entirely unknown.

(293.) The natural arrangement of this family, from
the paucity of forms yet known to belong to it, can-
not, as yet, be attempted. We must therefore rest sa-
tisfied with taking a hasty glance at those few genera
which, from their affinity to the typical Cobitine, or by
being known to be viviparous, appear to form broken
links in that chain of affinity which unites the loaches
to the carps.

(294.) The extraordinary genus Anableps, or the
double-eyed loach, is probably one of the typical ex-
amples. It was separated from the genus Cobites by Ar-
tedi T, and Gronovius has given three admirable figures of
it in his valuable work.{ Its body is cylindrical, covered
by hard scales, and having the mouth furnished with
numerous small teeth, while others, small and globular,
are placed on the pharyngeal bones. It is chiefly re-
markable, however, for its eyes, which are large, very
prominent, and placed close to the snout; the frontal

* Brit. Fishes, vol. i. p. 380. We may also add, that this is likewise
another evidence of the original union of the three aberrant circles of the
Vertebrata into one great circle, as conjectured by us some years ago ; so
that the classes Pisces, Reptilia, and Amphibia would form a circle, were
there not so many forms extinct among the saurian reptiles.

+ Not, as‘Cuvier intimates, by Bloch.

t See Gronov. Zoophy. pl. 1. fig. 1—3.

364 CLASSIFICATION OF FISHES.

bone forms a projecting ledge over them: the cornea
and the iris are divided into two portions by transverse
bands ; so that, by having two pupils, the eyes appear to ~
be double. Nevertheless, as Lacepede truly observes,
there is but one chrystalline, one vitrea, and one retina—
a structure, however, of which there is no other ex-
ample among vertebrated animals. The organs of ge-
neration and the bladder of the male, as M. Cuvier
remarks, “have their excretory canal in the anterior edge
of the anal fin, which is thick, long, and clothed with
scales; its extremity is pierced, and doubtless serves
the purposes of coition.” The female is not only vi-
viparous, but produces the young in rather an “ ad-
vanced stage of organisation.”’ Only one species of this
singular genus is yet known.

(295.) We have already given a general account of
the loaches, all of which are left under one genus by
Cuvier. We have, however, in consequence of the
recent investigation of some Indian species, detected two
principal groups and several minor variations, which
constitute sub-generic distinctions, and have arranged the
whole under a sub-family. The genus Cobiies is dis~
tinguished by having no armature on the head; those
with a broad, rounded tail-fin seem peculiar to Europe,
while such as have this fin lobed or forked have hitherto
been found only in India. The genus Canthophrys, as
its name implies, has a hard spiny process, with one
or more points, placed immediately beneath the eye,
where, in a state of inaction, it reposes in a groove. Of
these, which comprise three sub-genera, chiefly Indian,
many are elegantly striped. Dr. Hamilton observes that
they are more beautiful fishes than the ordinary loaches,
which they but slightly resemble in their appearance,
and differ from them still more in their habits, “ espe-
cially in swimming higher, and in not remaining so
stationary at the bottom.” The fishes of this sub-genus
are much compressed, and at first sight have a strong
resemblance to some of the blennies ; these two groups,
in short, appear mutually to represent each other.

THE LOACH FAMILY.— MOLINISEA. 365

(296.) The sub-family Peciline contains a few
freshwater fishes (P. multilineata Le Sueur, fig. 96.),
of very small
size, peculiar to
America. In
outward appear-
ance they bear
very little re-
semblance to the

ordinary loaches,
except, indeed, in the breadth and thickness of their
tail. M. Cuvier, however, attests that they are also
viviparous ; and, as he places them close to the Cobites,
we have been induced to do the same. These genera,
obviously allied to each other more than to the typical
loaches, have some few of the characters belonging to
this family: their mouth is very small; the lips are
fleshy ; and the eyes close to the snout, which is small
and horizontally flattened: on the other hand, they have
no cirri—a circumstance which shows they are not
ground-feeders. Their body is broad, oval, and often
very high in the middle; and the dorsal fin, in the
typical genus Molinisea (M. latipinna Le Sueur, fig.
97.), is so remarkably developed, that we consider this

genus as 4 representation of the riband-fish, more
especially as their eyes and gill-covers are very large.’

366 CLASSIFICATION OF FISHES.

With these two genera, we follow Cuvier in placing an-
other, which he has named Lebia (L. elipsoides, fig-98.),
which differs in nothing from Peeilia but in having

the teeth (a) serrated —a very questionable ground
for generic difference when unsupported by any other
character. We esteem ourselves fortunate in being abie
to present our readers with correct figures of these rare
and interesting fishes, all represented of their natural
size, from the masterly delineations of Le Sueur.
(297.) Hitherto, however, notwithstanding the ana-
logy of these viviparous genera (and more especially of
Anableps) to the cartilaginous order, we have men-
tioned no fish which externally possessed any resem-
blance to the latter, or gave us any idea that the
Cobitide really exhibited any such modification of
form as a fiat head, large pectorals, and a transverse
mouth, placed beneath a projecting snout, — characters,
in fact, which every one knows are prevalent through
all the cartilaginous types, and, therefore, peculiarly cha-
racteristic of them. There are, however, among the
Indian drawings published by Hardwick and Gray,
the figures of two species of their genus Balitora,
which completely realises these particulars. No de-
scriptions to these plates having yet been published, and
not being aware that this genus has ever been defined
by Mr. Gray, we can only be guided by the figures ; but
these leave us in little doubt as to the natural station of
these singular fishes among the malacopterygious fami-
‘lies. A glance at the figures here given of Balitora Bricei

THE LOACHES. — BALITORA. 367

99

al ii Pui
—

6222

( figs. 99, 100.) will be sufficient to show that no fish:
yet discovered of
the _ soft-rayed

“order makes so
near anapproach
to the external
shape of a carti-
laginous fish as
this; while the
circumstance of
its living in
mountain streams *, joined to its single dorsal fin, small
scales, and general habit, sanctions the idea that it enters

* within the confines of the family of Cobitide, of which
it forms the platyrostral or cartilaginous type. Its ana-
logy to Calyonymus and Liparis is likewise too obvious
to require explanation ; but, from both these, we con-
sider it is far removed, were it only from the single
circumstance of possessing small, but well-defined,
scales. That there may be other links, yet discovered,
between the malacopterygious and cartilaginous orders,
seems to us highly probable. Our simple proposition
in short is this, that, of all fishes yet known belonging
to the former of these orders, those of the genus Balitora
make the nearest approach, in their external structure,
to the latter order: we therefore place it as the last of the
Cobitide : and, as all authors agree in the affinity of
Polyodon to the sharks, so do we arrange that genus as
the first on the list of the aberrant cartilaginous genera,
after quitting the Malacopteryges or soft-rayed order.

(298.) We shall now terminate our survey of the

* T imply this from the expression “‘ mountain streams” upon the plate.

®

368 CLASSIFICATION OF FISHES.

soft-rayed order, and our first volume, by a few general
remarks on the Siluride. For many years previous to
that analysis of the Malacopteryges, which we have
laid before tae ichthyologist in the foregoing pages,
we had entertained a strong idea that the Siluride
connected this great division of osseous fishes with the
cartilaginous order, on account of the general depression
and breadth of their head, the inferior situation of the
mouth in several genera, and the viviparous nature of |
others. But, upon analysing this family, in conjunction
with that of the Gadide, we were induced to relinquish -
this idea. Affinity of general structure must always be
the primary object of the naturalist’s researches: if this
is studiously followed up, and successfully traced, the
analogies he is desirous of possessing, to strengthen his
views of affinity, are sure to follow; because they exist
throughout all nature, material or immaterial. The
close connection of the Gadide with the Gymmetres, or
riband-fish, by means of Cepola in one, and Brotula in
the other, proves at once the situation of these two fa-
milies to be annectant between the two great divisions
of osseous fishes. Now, it is clear that the Siluride
show an affinity to the Gadide sufficiently strong to
authorise our placing them in succession; and it is
equally clear that the Cobitide intervene between these
and the salmon family. On these considerations, there-
fore, we arranged our groups, and on these affinities do
we take our stand; not, indeed, presumptuously, but
with the most ardent desire of acquiring further know-
ledge, and a vivid impression of the great things which
will hereafter be done in this little-known department
of zoology, when we shall have long passed away to
purer and brighter regions, removed alike from the cen-
sure of opponents or the applause of friends.

END OF THE FIRST VOLUME.

Lonpon: 7

Printed by A. SPOTTISWOODE, & ,
New-Street-Square. pS —

ee ten an a

i Vie?
ua

ae” We

VACe as
oh

wv
Be,

4

y

te
\
4 7
a a

r Awa:

+ \
i
/ oy '

YY ise Rey hts ¢

a tS aA? f Ne.

> Orns oP Ih ieee Sit ae