The sharks have been referred to as possessing heterocercal tails,
but, though this is true of the majority, within the limits of the
group all three types of tail-fin occur, from the protocercal tail of
the fossil Pleuracanthids and the living _Chlamydoselachus_ to the
highly developed, practically homocercal tail of the ancient
_Cladoselache_(fig. 2).
The praecaudal portion of the fin-fold on the dorsal side of the body
becomes broken into numerous finlets in living Crossopterygians, while
in other fishes it disappears throughout part of its length, leaving
only one, two or three enlarged portions--the _dorsal_ fins (fig. 4,
d.f.). Similarly the praecaudal part of the fin-fold ventrally becomes
reduced to a single _anal_ fin (a.f.), occasionally continued backwards
by a series of finlets (_Scombridae_). In the sucker-fishes (_Remora_,
_Eckeneis_) the anterior dorsal fin is metamorphosed into a sucker by
which the creature attaches itself to larger fishes, turtles, &c.
[Illustration: From _Cambridge Natural History_, vol. vii., "Fishes,
&c.," by permission of Messrs. Macmillan & Co., Ltd.
FIG. 4.--_Tilapia dolloi_, a teleostean fish, to illustrate external
features. (After Boulenger.)
A, Side view. g.r, Gill rakers.
B, First branchial arch. l.l, Lateral line organs.
a.f, Anal fin. n, Nasal opening.
c.f, Caudal fin. p.f, Pelvic fin.
d.f, Dorsal fin. p.op, Preoperculum.
g.f, Gill lamellae. pt.f, Pectoral fin.]
The paired fins--though more recent developments than the median--are
yet of very great morphological interest, as in them we are compelled to
recognize the homologues of the paired limbs of the higher vertebrates.
We accordingly distinguish the two pairs of fins as pectoral or anterior
and pelvic ( = "ventral") or posterior. There are two main types of
paired fin--the _archipterygial_ type, a paddle-like structure supported
by a jointed axis which bears lateral rays and exists in an unmodified
form in _Neoceratodus_ alone amongst living fishes, and the
_actinopterygial_ type, supported by fine raylike structures as seen in
the fins of any ordinary fish. The relatively less efficiency of the
archipterygium and its predominance amongst the more ancient forms of
fishes point to its being the more archaic of these two types.
In the less highly specialized groups of fishes the pectoral fins are
close behind the head, the pelvic fins in the region of the cloacal
opening. In the more specialized forms the pelvic fins frequently show a
more or less extensive shifting towards the head, so that their position
is described as thoracic (fig. 4) or jugular (_Gadus_--cod, haddock,
&c., fig. 5).
[Illustration: FIG. 5.--Burbot (_Lota vulgaris_), with jugular ventral
fins.]
The median fin, especially in its caudal section, is the main
propelling organ: the paired fins in the majority of fishes serve for
balancing. In the Dipneusti the paired fins are used for clambering
about amidst vegetation, much in the same fashion as the limbs of
Urodeles. In _Ceratodus_ they also function as paddles. In various
Teleosts the pectoral fins have acquired secondarily a leg-like
function, being used for creeping or skipping over the mud
(_Periophthalmus_; cf. also Trigloids, Scorpaenids and Pediculati). In
the "flying" fishes the pectoral fins are greatly enlarged and are
used as aeroplanes, their quivering movements frequently giving a
(probably erroneous) impression of voluntary flapping movements. In
the gobies and lumpsuckers (_Cyclopteridae_) the pelvic fins are fused
to form an adhesive sucker; in the _Gobiesocidae_ they take part in
the formation of a somewhat similar sucker.
The evolutionary history of the paired limbs forms a fascinating
chapter in vertebrate morphology. As regards their origin two
hypotheses have attracted special attention: (1) that enunciated by
Gegenbaur, according to which the limb is a modified gill septum, and
(2) that supported by James K. Thacher, F. M. Balfour, St George
Mivart and others, that the paired fins are persisting and modified
portions of a once continuous fin-fold on each side of the body. The
majority of morphologists are now inclined to accept the second of
these views. Each has been supported by plausible arguments, for which
reference must be made to the literature of the subject.[2] Both views
rest upon the assumed occurrence of stages for the existence of which
there is no direct evidence, viz. in the case of (1) transitional
stages between gill septum and limb, and in the case of (2) a
continuous lateral fin-fold. (There is no evidence that the lateral
row of spines in the acanthodian _Climatius_ has any other than a
defensive significance.) In the opinion of the writer of this article,
such assumptions are without justification, now that our knowledge of
Dipnoan and Crossopterygian and Urodele embryology points towards the
former possession by the primitive vertebrate of a series of
projecting, voluntarily movable, and hence potentially motor structure
on each side of the body. It must be emphasized that these--the true
external gills--are the _only_ organs known actually to exist in
vertebrates which might readily be transformed into limbs. When
insuperable objections are adduced to this having actually taken place
in the course of evolution, it will be time enough to fall back upon
purely hypothetical ancestral structures on which to base the
evolutionary history of the limbs.
The ectoderm covering the general surface is highly glandular. In the
case of the Dipneusti, flask-shaped multicellular glands like those of
Amphibians occur in addition to the scattered gland cells.
A characteristic feature of glandular activity is the production of a
slight electrical disturbance. In the case of _Malopterurus_ this
elsewhere subsidiary function of the skin has become so exaggerated as
to lead to the conversion of the skin of each side of the body into a
powerful electrical organ.[3] Each of these consists of some two
million small chambers, each containing an electric disk and all
deriving their nerve supply from the branches of a single enormous
axis cylinder. This takes its origin from a gigantic ganglion cell
situated latero-dorsally in the spinal cord between the levels of the
first and second spinal nerves.
_Cement Organs._--The larvae of certain Teleostomes and Dipnoans possess
special glandular organs in the head region for the secretion of a
sticky cement by which the young fish is able to attach itself to
water-plants or other objects. As a rule these are ectodermal in origin;
e.g. in _Lepidosiren_ and _Protopterus_[4] the crescentic cement organ
lying ventrally behind the mouth consists of a glandular thickening of
the deep layer of the ectoderm. In young ganoid fishes preoral cement
organs occur. In Crossopterygians there is one cup-shaped structure on
each side immediately in front of the mouth. Here the glandular
epithelium is endodermal, developed[5] as an outgrowth from the wall of
the alimentary canal, closely resembling a gill pouch. In _Amia_[6] the
same appears to be the case. In a few Teleosts similar organs occur,
e.g. _Sarcodaces_, _Hyperopisus_,[7] where so far as is known they are
ectodermal.
_Photogenic Organs._--The slimy secretion produced by the epidermal
glands of fishes contains in some cases substances which apparently
readily undergo a slow process of oxidation, giving out light of low
wave-length in the process and so giving rise to a phosphorescent
appearance. In many deep-sea fishes this property of producing
light-emitting secretion has undergone great development, leading to the
existence of definite photogenic organs. These vary much in character,
and much remains to be done in working out their minute structure. Good
examples are seen in the Teleostean family _Scopelidae_, where they form
brightly shining eye-like spots scattered about the surface of the body,
especially towards the ventral side.
[Illustration: From _Trans. Zool. Soc. of London_.