;;->■■■ ^ f < I VV.V;,-M;!' ■ _ •. _r-

Marine Biological Laboratory

Accession No. .^.kSDR

Given By The Blal-clston Ho.
Place Philad filphiri, Pa .

■ ni

ni

_ D

^lakiston boolis are hound in high
grade materials which are sturdy, verc'

min-proof and water resisting

they can be cleaned with soap and
water.

BRAIN AND BODY OF FISH

^qlBO Ho yboa bns nisiB.

1 iN AMI'

HOAOJT

Nl PATTFT?v

gjIAO

jaa^Aa

S ^ ^ H

2 2 3 <

"-J 33 s CO

^ H S fc

o

Brain and Body of Carps

ROACH

CARP

BARBEL

§ > y H

H M 3J f

W G H h

ffi C/3 Cf!

w £5

V /S

BRAIN AND BODY
OF FISH

A STUDY OF BRAIN PATTERN
IN RELATION TO HUNTING AND FEEDING

IN FISH

By
H. MUIR EVANS, M.D. (Lond.), F.R.C.S. (Eng.)

Fellow University College, London

THE BLAKISTON COMPANY

Philadelphia
1940

MADE IN GREAT BRITAIN

PRINTED BY THE LONDON AND NORWICH PRESS, LIMITED
ST. GILES WORKS, NORWICH

FOREWORD

In this short introduction to the comparative study of the brain
of fish we have confined oiu* observations to bony fish. The reader
will find no reference to the brain of the cartilaginous (elasmobranch)'
fish. Our old fi'iend the dog-fish will not appear. This unpleasant
animal may be the cause of the backward state of our interest in
the neurology of fish. Fifty years ago Ray Lankester was making:
pioneer observations on the brain of the dog-fish, but there seem to
have been few scientists who have followed liis trail. It is probably
true that if investigations had been made on the goldfish or th&
roach the interest in these attractive fish would have led to a wider
knowledge of the brain structure of fish. This book is a study
of the brain of our familiar fresh- water fish, and the common food
fishes of the British Isles. Salmon and trout, having a publication
of their owti, do not appear. All the figures I reproduce are from
my own specimens and drawings with the exception of two text-
figures of the brains of the cod and mormyrus : for permission to
redraw these I have to thank the Curator of the Royal College
of Surgeons of England. In the discussion on the Silence of the
Sea, I have referred to passages in Sir WiUiam Bragg's Lectures
on " The World of Sound " and to Dr. Beatty's " Hearing in Man
and Animals." I have also consulted Cunningham on this subject
in the work " Reptiles, Amphibians and Fishes " edited by him.
References to other authors are mentioned in the text. A large
proportion of the observations described in this work have already
appeared in the Proceedings of the Royal Society of London, but
the text has been entirely re-wTitten with a considerable amount
of new material.

H. MUIR EVANS.

INTRODUCTION

In the course of a series of investigations into the physiology of the
swim-bladder in the Carps or Cyprinoids, I was led to study afresh
the f mictions of the Weberian ossicles, a series of small bones which
connect the anterior sac of the swim-bladder in these fish with the
internal ear. Tliis latter study involved the dissection of the brain,
and I thus became aware of the very diverse forms of the hind-brain
in this family. In looking up the literature of the subject I came
upon a reference to a paper by a Frenchman, P. Savoure, communi-
cated to an obscure provincial society, the Bull. Soc. Sci. Quest,
Rennes, in 1912, which fortunately was in- the library of the Royal
Society of London. This paper described the brain of a number of
members of the Carp family, and showed the diversity of form in the
liind-brain, but the illustrations were on a small scale and semi-
diagrammatic. On comparing these drawings with my own of
similar species caught in English rivers, I found that my observa-
tions did not agree with those of the French observer in several
instances, and further, that his descriptions were based entirely on
superficial examinations and had not been supplemented by micro-
scopic investigations of serial sections. This being the case it
appeared to me that there was a fertile field for research into the
neurology of the Cyprinoid brain, and for an attempt to associate
the various forms or patterns of the brain with the habits and diet
of fish, which presented so many opportunities for examination. I
was thus led to undertake a series of neurological studies on fishes
by means of comparative anatomy. This was the method employed
by Sir Charles Bell, the first neurologist, as he has been called. When
a child I was given by my mother " Bell on the Hand." When I
became a medical student, Alexander Shaw, Bell's assistant and
collaborator, gave me Bell's " Anatomy of ExjDression," with annota-
tions in his own hand. Some years ago I purchased at the sale of
Lord Lister's Library a copy of " Bell on the Nervous System,"
which had been presented to James Syme by the author, and bears
his hand-WTiting : this I treasure, not only for its own sake, but for
the triple association of three illustrious men. There are probably
few medical students at the present day who reahse that Bell's

7

8 INTRODUCTION

paralysis is named after the man who first described the true function
of the facial nerve, and that the same man showed by dissection the
nerves of muscle sense and described the fundamental difference
between the anterior and posterior roots of the spinal nerves. Bell
wished to be known to posterity as an anatomist, not as an experi-
mentalist. " Anatomy is already looked upon with prejudice by
the thoughtless and ignorant. Let not its professors unnecessarily
incur the censures of the humane." " In a foreign review of my
former papers the results have been considered as a further proof in
favour of experiments. They, are on the contrary, deductions from
anatomy ; and I have had recourse to experiments, not to confirm
my own opinions, but to impress them on others. It must be my
apology that my utmost efforts of persuasion were lost, while I
urged my statements on the ground of anatomy alone. I have
made few experiments ; they have been simple and easily performed,
and I hope are decisive." In another place he says, in favour of
anatomy, that " it is better adapted for discovery than experiment,"
and illustrates his contention by comparing astronomy and chemistry
and considers that ' ' anatomy is more allied to the former inasmuch
as things are obvious."

Another paragraph extols the work of Monro and Hunter, and
adds, " Let us continue to build on that structure which has been
commenced by their labours, and which the undeserved popularity
of the continental system has interrupted." " The whole history
of medical literature proves that no solid or permanent advantage
is to be gained either to medical or general science by physiological
experiments unconnected with anatomy." We will quote another
passage in full, " He who discovers a new nerve, or furnishes a
more accurate description of the distribution of those already known,
affords us information in those points which are most likely to lead
to an accurate knowledge of the nervous system. For if we consider
how various are the origins of the nerves, although all arising from
the brain, and how different the circumstances attending them, we
must suppose a variety of uses to arise out of this peculiar structure.
In this manner is the nervous system to be studied. For there is an
internal change in accordance with outward organisation, whilst
the system or great plan does not alter. An animal, or class of
animals, may have a particular organ developed, and with the
external apparatus there is a corresponding or an adjusted condition
of the appropriated nerve. Another class may be deficient in the
external organisation, when we shall look in vain for the accompany-
ing nerve ; it is contracted or hardly visible ; but with all this the
system is unchanged." These profound truths were written over

INTRODUCTION 9

a hundred years ago. The experimental method is now predominant,
and the niorphologist is abnost a museum specimen, yet there are
still some minds who see in comparative anatomy a valuable aid
in the investigation of function. The late Prof. J. H. Haldane said
that comparative anatomy may be made as valuable as experimental
physiology. " Future anatomy both normal and morbid will
certainly set itself to investigate the physiological relationships
which are inseparable from structural manifestations, and anatomy
will then be as much an experimental science as is ijhysiology."
Experimental biology can only be pursued in a laboratory where
the investigator can follow his pursuit within the law. But there
is a vast field for a worker in comparative anatomy in its relation-
ship to physiology and the investigation of function. My work
embodies these suggestions, and has been carried out in the scant
leisure allowed by a busy professional life. 1 have been encouraged
by the interest and support of Sir Henry Dale, the late Prof. Boycott,
Sir Leonard Hill and Dr. Tate Regan. I have also received much
help from members of the Staff of the Ministry of Agriculture and
Fisheries at Lowestoft, particularly I would mention Miss Thursby
Pelham and INIr. Michael Graham, who kindly read the manuscript
and made many valuable suggestions, and finally Mr. Clarke, the
chief laboratory assistant, without whose technical assistance
my work would not have been possible. My interest in the subject
of hearing in fish dates from my young days when I studied Otology
in Berlin, and worked on the literature of the cochlea at the British
]\Iuseum, where I first became acquainted with the work of Bell on
Hearing. It is often forgotten that the resonance theory of hearing
first propounded by Cotugno, in 1761, was reaffirmed by Bell in 1826 ;
but it was not until Helmoltz, in 1863, brought his physical insight
t3 bear on these views that this theory was accepted, which has now
become exclusively associated with his name.

;^CAi

CONTENTS

Foreword
Introduction

PAGE

5

I. Intelligence and Brain Pattern of Bony Fish
II. Brain Pattern — Continued .

III. The Carps ....

IV. The Carps — Continued .
V. Hearing in Fish ...

VI. Accessory Organs of Hearing

VII. The Central Acoustic Lobe

VIII. The Silence of the Sea and the Voice of
Fishes ....

IX. Flat-Fishes

X. Flat-Fishes — Continued

XL The Cod Family

XII. The Cod Family — Continued

XIII. The Hake, the Scabbard Fish, Prometheus

AND NeSIARCHUS

XIV. The Eel. Anguilla Vulgaris
XV. The Anatomy of Gustation

XVI. Hearing, Equilibrium and the Cerebellar

Functions
XVII. The Valvula Cerebelli
XVIII. The Pituitary Body .
XIX. The Problem op Pain in Fishes
XX. Retrospect and Conclusions

13
21
28
35
42
51
58

69

75

84

94

102

113
121
126

132
138

142
149
152

11

OOJ393

BRAIN AND BODY OF FISH

CHAPTER I
INTELLIGENCE AND BRAIN PATTERN OF BONY FISH

The subject of behaviour in animals has in recent years attracted
the attention of biologists in this country and also in America, and
a large literature devoted to its discussion has arisen. Recent
studies on fish have been made at the Marine Laboratory at Ply-
mouth, in which experiments have been made to test the ability
of fish to master the intricacies of a maze.

Another type of experiment has shown that if a predatory fish
and its prey are put in the same tank, but separated from each other
by a glass partition wliich prevents the hunter from reaching its
would-be victim, after a short time the fish will keep to its own side
of the partition, even after this obstacle has been removed. These
observations seem to prove that a fish has a capacity to learn, to
associate facts and direct its movements up to a certain point,
according to experience. Nevertheless, one hears not infrequently
the remark " I didn't know fishes had brains " ; and to illustrate the
brainlessness of fish, or perhaps their want of feeling or perception
of pain, the tale is told, how a fish hooked with a certain bait, may,
when returned to the river, be caught again with the same bait after
a short interval. This conclusion does not bear criticism, because
M'hen a fish is hungry or greedy, or, as a fisherman would say, is
biting freely, there is no doubt it appears stupid, but that same fish
when not hungry, will be most cautious in examining the bait and
subject it to the closest scrutiny, so that the finest gut and cleanest
bait are required to lure it to its capture. When hunger is in com-
mand, discretion is forgotten. As bearing on the intelhgence of
fish some remarkable studies have been recently made on the homing
instinct of salmon. The most convincing experiments have been
made on the cliinook or king salmon in Canada, where in the
Columbia River system we find the Lower Columbia River, ninety
miles long, receiving the water of two systems. The salmon fre-
quenting one system are " spring run," and those of the other are
*' autumn run " fish. If eggs were transplanted from one system

13

14 BRAIN AND BODY OF FISH

and hatched in the other, the fish from the transplanted eggs, after
tlieir journey to the sea, returned to the river in which they were
brought up and not to the river in which they were spawned.
But they retained their hereditary habit of migrating in the spring
instead of in the autumn as do the native fish (E. S. Russell),

Young salmon in fresh water get to know their home waters and
recognise them on their return. They remember the way back to
their home river after their extensive journeyings in the sea.

We see, therefore, that the behaviour of fish is not simply deter-
mined by seasons, currents, temperature, and their respiratory
needs, as some biologists maintain. Their migrations are as difficult
to explain as those of birds. Moreover, like birds, there are many
fish which build most efficient nests. The sea stickleback not only
builds a nest but binds it together with a thread spun from its own
body. The male Lumpfish or Cockpaddle fans the eggs and drives-
away all intruders. In the Nile one of the Mormyridse known as
Gymnarchus, makes a floating nest of grasses, and the whole nest is
somewhat like the ark of bullrushes which served as a cradle for
the infant Moses (Cunningham).

So it appears that nest building, parental care, and the homing
instinct are common to fish and birds. A further attribute common
to both is song ; it is true that our knowledge of sounds produced
by fish is slight, but the hearing capacity of some fish has been
stucUed by modem methods and reveals a remarkable range of
hearing. The Squeatague produces a deep drumming sound, but
only the male, so it is probably used in courting. The Maigre, a
Mediterranean fish, is well known for its powers of " song," which
it is believed to have given rise to the legend of the Sirens which
beguiled Ulysses. Sounds are also produced by the Sea-robin
Prionotus.

I have chosen the title of this book, " The Brain and Body of
Fish," because this condenses in a phrase the remarkable fact that
the external conformation of the brain of a bony fish indicates its
habits and mode of feeding. In other words, the study of the external
condition of the brain in bony fish is an index of the development
of those organs belonging to the various sensory faculties. The life
of a fish consists of two main activities which may be conveniently
described under the headings feeding and breechng. The first is
necessary for the life of the individual, and the second for the future
of its race. In other words nutrition and reproduction make up
the life history of a fish. We shall find that not only is hunting
a highly specialised function in fisli, but that the various methods of
hunting are associated with a varying pattern of brain. For instance.

INTELLIGENCE AND BRAIN PATTERN 15

if we come to study two familiar fomilics of flat-fish, the sole and the
plaice, it will be found that their hunting methods differ widely
and that this difference is reflected in their brain structure. The
sole is nocturnal in its habits, and it hunts by tapping the sand
in its search for Morms, just as the thrush does on the lawn in front
of your window, and it has an acute sense of smell, which is evidenced
by its elaborate nasal organ. Compared with the plaice its eyes are
very small. The conclusion is reached, therefore, that the sole
hunts by smell and touch.

This is not the whole story, but it is sufficient for our present
purpose. The plaice on the other hand has large, prominent, and
movable eyes, and, as we shall find later, a marked sense of taste,
due to the presence of taste-buds, so we conclude that the plaice
feeds by sight and taste. Corresponding to the increased import-
ance of these different senses to these fish, the central areas in the
brain connected with the sense organs are enlarged, and thus a
definite type of brain pattern results. Other interesting correla-
tions between feeding habits and form of brain are to be observed
in the Cod family, in which there is a gradual change in the pattern
of the brain as we pass from the shell-fish eating members, through
the mixed feeders, to those purely predacious. The principle of the
enlargement of special sense areas in the brain in accordance with
an increase of function is also of service in elucidating problems like
the question of hearing in fish. Certain fish have elaborate mechan-
isms which are supposed to be accessory to the organs of hearing ;
if it is found that in these fish there is a special area more developed
than in fish devoid of these mechanisms, we may assume that this
area has some acoustic significance. Examples of this condition
are to be seen in the Carps and the Herring tribe among fishes
of the British Isles and seas, and in the interesting family of fish
kno^^"n as the Morm^Tidse of northern Africa. In fact, the com-
parative study of the brains of bony fishes is full of unexpected
revelations. So much may be said as an introduction to the subject
of brain form, in relation to hunting and feeding. Is there any
area in the brain that can be identified as associated with the function
of reproduction ? Modern research into the function of the pituitary
gland directs the attention to the condition of this gland in fish and
evidence will be brought forward to show that this gland undergoes
certain changes in those fish that are about to migrate for the 23ur-
poses of reproduction.

The relation of form to function in determining the pattern of the
brain of a fish will be described in the following pages. In order to
help the reader to follow the argument it will be necessary to give

16 BRAIN AND BODY OF FISH

an outline of the structure of the brain of a typical bony fish, and
we shall endeavour to do tliis without introducing any but the most
necessary technical terms.

The spinal cord of a fish may be likened to the long stem of a
clay pipe, this, instead of being hard like baked clay, is of the con-
sistence of a cream cheese and of the same colour. Instead of a
bowl at the anterior end there are a series of thickenings of the
walls, which for the most part appear on the upper or dorsal aspect.
These swellings or lobes may be single, but more frequently there
are two symmetrical protuberances facing each other, which is
described as bi-lobed. These swellings are not necessarily associated
with any increase in size of the central canal, although at two points
of the brain, as the anterior part is called, the neural canal widens
out into spaces called ventricles.

In the liinder part of the brain, which is known as the medulla
oblongata, another condition is noticed. In the mid-line there is a
longitudinal slit in the dorsal wall of the hinder ventricle, and the
thickenings of the walls thus seem to take place in the free margin
of the split tube ; there is thus left an open space or hiatus, which,
however, is covered in by a vascular membrane. This opening
out of the fourth ventricle, as it is called, is known as the rhomboid
fossa.

The various lobes will now be described, and starting from the
simple spinal tube and passing forwards we first notice two lateral
protuberances (Plate 1, Fig. i), the vagal lobes. This view is drawn
from a specimen from which the roof of the bony cranium has been
removed, and the brain is looked down upon from above. These
lobes in the carp are pear-shaped in outline, and their anterior ends
are seen to be separated by a central globular lobe kno%^'n as the
facial lobe. In the catfish (Plate 1, Fig. ii) the vagals are globular
and there are two facial lobes lying in front of them. It is found
that in certain of the carp family that the facial lobe has a partial
median division, as in the gudgeon, so that the single median facial
of the carp and most Cyj^rinoids is due to the fusion of two lateral
segments. The condition in the cod (Plate 1, Fig. iii) is somewhat
different, as the vagal lobes do not form so prominent a swelling
dorsally, as we shall have occasion to note when the cod brain is
discussed.

The vagal lobes receive nerves of sensation, afferent nerves
and also efferent nerves wliich pass outwards and are nerves of
motion ; so that we say that the vagal lobes have both sensory and
motor nerve roots.

The facial lobe in fish is peculiar in that it receives all the sensory

IXI'ELLIGENCE AND BRAIN PATTERN

17

fibres from certain peripheral sense organs kno^\•n as taste-buds
Avhioli are situated on the anterior part of the mouth, the lips,
and barbels (when present) and even the skin. Its function is
gustatory, i.e. the centre of taste perception, altliough the facial
nerve still sends a few motor fibres to certain neck muscles.

This puzzles the human anatomist, who looks upon the facial
nerve as predominantly the motor nerve of the face muscles ;

PLATE 1.

OB

PEB

Cod. Catfish. Carp.

P. — Pallium. PEB. — Primitive end-brain. OL. — Optic lobe. C. — Cerebellum.
AT. — Acoustic tubercle. S.S.L. — Somatic sensory. FL — Facial lobe. VL —
Vagal lobe. OB — Olfactory bulb. CLM — Cerebellum (removed in Fig. iii,
turned forwards in Fig. ii).

however, even in man the facial nerve still receives gustatory im-
pulses from the anterior portion of the tongue by means of a recurrent
branch, known as the chorda tympani.

It is quite obvious how this apparent contradiction in function
has arisen. In fish there are no facial muscles but an elaborate
gustatory system, on account of the nature of the medium in which
it lives ; whereas in man the facial, which by Sir Charles Bell
was called the respiratory nerve of the face, not only supplies
the styloid and hyoid muscles of the neck, but also has taken on the
supply of all the facial muscles as so fully described by Bell in his
" Anatomy of Expression."

The course of the facial nerves within the brain of fish is so
characteristic that they are a useful guide to the recognition of the
facial lobe or lobes and their relation to other structures. In a

18 BRAIN AND BODY OF FISH

position anterior to the facial lobes there is a bilateral area receiving
sensory fibres from the skin, this can be seen in the drawing of
the cod's brain, and it is known as the somatic-sensory or fifth lobe ;
it is not apparent in the carp or catfish when seen from above, but
it is clearly evident in sagittal sections.

The next prominent lobar swelling is a large central protuberance
smaller in the carp than the catfish, and forming a tongue-like body
overlapping the medulla oblongata in the cod ; this is known as
the cerebellum.

From either side of the cerebellum leading to the sides of the
somatic-sensory or fifth lobes are marked rounded prominences,
varying much in size in different species, and known as the acoustic
tubercles. Anterior to the cerebellum are two large globular swell-
ings known as the optic lobes, which receive the large optic nerves
on their ventral aspect, and at the anterior end of each oj^tic lobe
are smaller lobes which form the primitive end-brain into which
enter the olfactory nerves, leading from the olfactory bulbs, or are
in continuity with the olfactory lobes.

As this end-brain puzzles even a trained biologist, we will
further explain that the end-brain in which the olfactory centres
lie consists generally of two sections, the anterior being the olfactory
bulbs which are the terminations of the olfactory nerves, and the
posterior the larger, the jjrimitive end-brain, in wliich the walls
of the neural tube are thickened. The olfactory bulbs usually lie
close to the end-brain as is seen in the drawing of the brain of a
plaice, but in some bony fish the nasal sacs are far removed, as is
seen in Plate 1, Fig. I, the drawing of the brain of a carp, where the
bulbs appear separated by the pallium, and the olfactory stalks
from the primitive end-brain.

If we now look at Plate 2, which is a drawing of the brains of
a carp and a cod viewed sideways when a sagittal section is made
through one half of the brain we shall be able to visualise the above
described organs in another dimension.

In man the sensory columns of the spinal cord lie posteriorly,
but these columns become dorsal in fish and so we find the sensory
centres of the brain appearing on the dorsal aspect in fish.

If we examine the diagrams of Plate 2 it will be seen that in
both cod and carp the tecta optica which form the prominences
of the optic lobes are similar, but the forward projection of the base
of the cerebellum, known as the valvula cerebelli, is bigger in the
carp than the cod and further in the carp tends to separate the
tecta of either side.

When the medulla oblongata of the cod is examined, the sldn

INTELLIGENCE AND BRAIN PATTERN

19

area or somatic-sensory lobes are prominent, while the facial and
vagal lobes make no j)rominent elevation ; in the carp the skin
area is small, but both the facial lobe and vagal lobes produce well-
marked prominences. In the diagram it is clearly seen that all
the sensory centres, namely the olfactory lobes, optic lobes, acoustic

TO/>i

PLATE 2.
V4LC. CLM

ON

M'V. Pir sv

Brain of Cod in sagittal section.

T0/>|- V/\L.C

L.IIVF

o:n' ••■ «v

Brain of Carp in sagittal section tlirough the middle of a vagal lobe.

OLF — Olfactory tract. P — Pallium. CS. — Corpus striatum. T.Opt. — Tectumi
opticum. Val.C. — Valviila cerebelli. CLM — Cerebellum. SS — Somatic-sen-
sory lobe. LF — Lobus facialis. LV — Lobus vagalis. ON — Optic nerve.
PEB — Primitive end-brain. Pit. — Pituitary gland. SV — Saccus vasculosus.
L.IXF. — Lobus inferior. III.V. — Third ventricle. IV. Vent. — Fourth ventricle,

tubercles, the somatic-sensory lobes and the facial and vagal lobes
are situated dorsaUy. Situated ventrally, are several im2)ortant
bodies ; posteriorly the first enlargements are too well marked
lobes, the lobi inferiores, in front of which lies the saccus vasculosus.
Then comes the pituitary body or gland wliich in recent years has
been recognised as an organ of great importance ; immediately in
front of this the optic nerves will be seen to decussate.

20 BRAIN AND BODY OF FISH

The functions and structure of these organs wiU form the material
tor a later chapter ; we have noticed that other authors dismiss the
lobi inferiores with a few scanty remarks, but we hope to be able
to point out the significance of this interesting but neglected
structure.

CHAPTER II
BRAIN PATTERN— Continued

Having grasped a general idea of the layout of the brain, it is now
necessary to consider its intimate structure, wliich is revealed by
stud\'ing its tissue by microscopic methods. These methods are
the same as those used in human anatomy, and involve an elaborate
process which requires the use of a great variety of stains, which
differentiate special cells and nerve fibres or neurons.

In this way certain cells can be distinguished, and we are entitled
to assume that a motor nerve cell recognised as such in human
histology, is a motor cell in a fish, and that a nerve cell of another
definite type in man has the same function in fish. The lobes are
thus found to consist of accumulations of nerve ceUs of different
types, and these are so arranged that some receive impulses or
messages from the sense organs of the body, both from the external
parts and from the viscera or internal parts, wliile others send out
impulses, wliich set the motor machinery in action. These messages
are carried to and from the brain by the nerve roots and neurons
as we have already mentioned. But, further, these groups of cells
must of necessity have relations with the other groups, so that
strands of nerve fibres, called tracts, carry messages from one
centre to another ; finally it is necessary that there should be a
centre to correlate all these various messages, so that there are
other tracts to carry messages to a higher centre which controls
the whole nervous system. We have described the brain of fish
as being an enlargement of a simple nervous tube which is known
as the spinal cord ; this also has tracts of fibres leading up to and
from the brain. Aromid the central canal there is an accumulation
of cells for the reception of sensory impulses and for the sending out
of motor impulses. In all vertebrate animals the nerves given
off from the sjnnal cord have two roots, and two columns of nerve
fibres pass up the cord to the brain.

It Mas established by Bell just a hundred years ago that the
anterior (in fish ventral or inferior) column of the spinal marrow
and the anterior roots of the spinal nerves were for motion, and that
the posterior (in fish dorsal) column and posterior roots were for

21

22 BRAIN AND BODY OF FISH

sensation. There is a similar division, though more complex, in
the brain, and so it is found that the lobes of sensation are found
on the dorsal or posterior aspect. It will be understood that the
term posterior in man becomes superior in fish owing to the upright
posture of man, contrasted with the horizontal poise of fish.

The reader must forgive this rather elementary description,
but it is necessary if he is to follow the more technical matter.

It may well be asked what means are to be adopted to understand
the use or function of the lobes which have been enumerated ?
Are you justified in comparing the brain of a fish with what is
known about the brain of man ? As to the lobes which receive
impressions from peripheral sense organs that have an exact counter-
part in man there is no difficulty.

For example, the eye of a fish has much the same structure
as in man, and is connected directly to the two optic lobes by large
nerve trunks which cross each other as they enter the brain. There
can be no doubt that this system is rightly assumed to be a visual
apparatus. The truth of this assumption is further established,
when the brains of two familiar flat-fish are compared. The plaice,
which feeds by day and largely by sight, has very large optic lobes
which tend to be convoluted, whereas the sole, which is nocturnal in
its habits and feeds largely by smell and touch, has very small
optic lobes.

It may be stated as a general rule that fish with nocturnal habits
have small optic lobes ; in the cod family this generalisation is
well shown. The , cod, w hiting, and pollack have large optic lobes
and are hunters by day ; on the other hand the ling, burbot and
rockling have small optic lobes.

According to Cimjiingham the ling is somewhat nocturnal and
the rockling entirely nocturnal in their habits, and according to
Tate Regan the burbot goes in active pm-suit of prey at night.

The above examples of the variations in the size of the lobes are
isimple and straightforward, and show well the relation of function
to the central nervous system ; they further establish a i^rinciple
which enables us to unravel the more complex areas of the brain
such as those concerned in the appreciation of taste which are
known as the gustatory centres. These centres are very important
in fish as the substances that give rise to the sensations of taste
are always in solution. Therefore it is not surprising to find that
fish, living entirely in a fluid medium, are furnished with an extensive
gustatory system. Taste is a chemical sense, and the organs of
taste enable the animal to recognise sweet, sour, salty and bitter
substances.

BRAIN PATTERN 23

By what means and by what structures is a fish enabled to
make use of its surroundings, ^liich must contain such a vast
number of different sapid substances ? The organs of taste in
man are represented by certain patche^^ or groups of pecuHarly
modified cells of the superficial layers of the skin, which are lodged
in the thickness of the surface layers of the tongue. These groups
of cells have a bud-like arrangement, and have therefore been
termed taste-buds. Without going into details at the j^resent
stage, it may be stated that they are found on the tongue, the throat,
and at the entrance to the windpipe. Leydig discovered in fishes
flask-shaped organs, similar to taste-buds, in certain parts of the
skin, and they also occur in the mucous membrane of the mouth
and throat in those animals.

We shall describe later certain other sites in which they are
found in fish.

Taste-buds have been compared in general form and appearance
to the leaf -buds of a plant, but the arrangement of their cells may
be also likened to the segments of an orange. They are flask-
shaped bodies, the base of the flask lying in the depths of the skin
or mucous membrane, and the neck projecting towards the free
surface. The cells are enclosed in a sort of adventitious capsule,
but the most superficial cells of this are perforated to allow of
access of the apex of the taste-buds to the free surface : the ceUs
that form the bud do not actually reach the surface, but from their
apices arise fine hairlets which project into an opening called the
gustatory pore. By this means the ceUs receive stimuli from the
various substances in solution, and from these cells sensory impulses
are carried to the centres in the brain.

The lobes that receive all these impulses are the vagal and
facial lobes. The vagal lobes are both sensory and motor in
function receiving sensory impulses from the gifls and pharynx,
and they have also a motor area known as the nucleus ambiguus
from which efferent fibres pass by the vagal motor root : the facial
lobes receive sensory fibres from the taste-buds situated on the
skin, mouth, lips and barbels, but the facial nerve has also a small
motor branch leading to the neck muscles.

A broad summary of the facts relating to these two lobes can
be made in Herrick's words " the vagal lobe for mouth-tastmg and
the facial lobe for skin-tasting are local erflargements of the visceral
sensory brain. All the taste-buds in the pharynx and back of
the mouth are supplied by the vagal and glossopharyngeal nerves,
those in front of the mouth the lips, the barbels, and outer skin from
the root of the facial nerve."

24 BRAIN AND BODY OF FISH

It may well be asked what is this new nerve, the glossopharyngeal,
which has not been mentioned hitherto. This nerve and the small
lobe into which it runs, which may be regarded as a forward extension
of the vagal lobe, supplies an anterior gill-arch so that for om:
present piu-pose it may be looked upon as functionally part of the
vagal lobe, supplying a small anterior part of the gill -arches and
neighbouring structures. But it is important to remember its
existence, as in man as we shall mention later, it supplies a particular
part of the pharynx and has a specialised function in the more
complex structure of the higher vertebrates.

We are now in a position to apply the same process of reasoning
to the vagal and facial lobes as was used in the interpretation of the
function of the optic lobes. For this purpose the examples will be
taken from three species of the large family of carps or Cyprinoids,
of which there are many members in British freshwaters. Let us
see what can be learnt from the comparison of the hind-brains of the
bream, the roach and the gudgeon, to be more concise their facial
and vagal lobes.

The habits of these fish are familiar to any coarse fisherman.
The bream is found on the bottom in muddy waters, where reeds
grow, and he sucks at the bait rather than bites, and lifting the bait
up, tilts the float and drags it slowly along the surfa'ce before finally
taking it into its mouth. It is able to extract food-stuffs from the
mud. The roach is a more lively fish and often bites freely, but
usually investigates the bait before actually closing its lij)s ; as a
rule it feeds near the bottom, but it also takes a fly and feeds at times
near the surface ; the gudgeon frequents gravelly bottoms and
searches for its food among the stones with its sensitive barbels.

When the brains of these fish are examined the difference in
size of their lobes is very evident. The bream has a very large vagal
lobe, the roach a small one, and the gudgeon one of moderate size ;
on the other hand the gudgeon has a very large facial lobe, the roach
a very small one and the bream a small one. It has been observed
that the vagal lobe is for mouth-tasting, so that the question arises
what is the reason for this great enlargement of the vagal lobe in the
bream, which is stiU more marked in the carp. The answer to this
question is that these fish have a speciaUsed organ on the palate,
that consists of ridges lined with taste-buds which enables the animal
to sift and sort, retain, or reject nutrient material from the mud and
decaying vegetable matter, on which it feeds : this organ with its
taste-buds is represented in the vagal lobe and is the cause of its
great development. The size of the facial lobe in the gudgeon is
stiU more easy to explain ; this fish has a pair of barbels riclily

BRAIN PATTERN 25

furnished witli taste-buds, with which it searches for shrimps,
small molluscs, Morms, and insect larva?, in the sand and gravel. It
has rather a curved snout and thick lips so that it is a typical " skin-
taster." and therefore we find a very large facial lobe. But this is
not the whole story of the facial nerve and lobe in the gudgeon. If
the facial nerve is traced into the lobe it will be found to divide into
two j)arts. an anterior bundle of fibres passes laterally into the front
of the facial lobe, and a posterior bundle passes backwards into the
hinder part of the lobe, thus dividing the lobe into two separate
areas. There are other fish which have a similar cU vision of the
facial nerve within the brain ; for example the barbel, which has
two pairs of barbels ; in this fish the bmidles are of unequal size,
the anterior is small, ^^■hile the posterior is larger and enters a large
division, forming the hinder part of the lobe.

The tench is also the possessor of a similar division of the facial
nerve, but it has only a short barbel on either side of the mouth ;
associated with this small barbel we find that there is a large lateral
bundle passing to the front of the facial lobe and a very small strand
passing to the hinder portion. We see, therefore, that the size of
the two parts of the lobe varies with the development of the barbels,
in the barbel a large posterior portion with many barbels, and in the
tench a small posterior portion with a single small barbel on either
side. These conchtions have also been found to occur in certain
Cyprinoid fish of the Madras tanks as noted by Bhimachar, who
describes a fish very like a tench with small barbels and a small
cUvision of the nerve passing posteriorly. The conclusion appears
to be justified that fish with barbels have a division of the facial
lobe, and that it is the posterior part of the lobe that receives the
gustatory fibres arising from the taste-buds of the barbels, at any
rate in the Cyprinoid family.

This observation concerning the differentiation of two bundles
of the facial nerve reminds the writer of the remark of tSir Charles
Bell, " he who discovers a new nerve or furnishes a more accurate
description of the chstribution of those already known, affords us
information in those points that are more likely to lead to an accurate
knowledge of the nervous system."

The brain of a roach is in sharp contrast with that of either
bream or gudgeon. Both vagal and facial lobes are small, but to
compensate the fish for this slight development of its gustatory
centres, it has large optic lobes, and is very sharp-eyed, so that very
fine gut must be used by the angler, if he wishes to be successful in
his sport. But it must not be considered that the roach is lacking
in the sense of taste, because every fisherman must have noticed

26 BRAIN AND BODY OF FISH

how discriminating this fish is as regards baits. It seems, therefore,
that the large size of the vagal in the bream is solely due to the large
palatal organ and the large size of the facial lobe in the gudgeon is
mostly due to the possession of a pair of barbels ; and that the
absence of these specialised structures in the roach is the cause of
the smallness of both lobes.

In the spinal cord of man there are considerable enlargements in
the cervical and lumbar regions and the mid-dorsal region is small
in comparison. This is due to the large nerve trunks that go to
the arms and legs respectively from these areas ; but as a rule in
fish the spinal cord is of uniform calibre throughout. But there is
one family of fish, the gurnards, that gives us an example of a new-
function arising in a fin, associated with the formation of globular
swellings of the cord. The gurnard appears to have adopted the
method of protection employed by the crustaceans, and to have
clothed its large head with a carapace like a lobster, and like a
lobster has found it necessary to provide itself with feelers or
antennae ; the three posterior rays of the pectoral fins have become
separate and form three slender processes on either side, which
possess extensive motion on a double row of joints, not connected
with the fins. These processes or fingers are supplied with
peculiar nerves, and are consequently in possession of special
functions.

That they are organs of feeling cannot be doubted ; but the fish
has also been seen, according to Couch, " when resting on the ground,
to close the pectoral fins and to creep by the help of these processes
as if they were organs of motion that could be employed without
exciting alarm to the prey which the motion of the fins might
possibly do." What interests us particularly is the fact that the
additional function which the anterior spinal nerves of the gurnard
have to jDerform in supplying the sensitive pectoral appendages
and their muscles has caused the development of a paired series
of globular swellings of the corresponding portion of the spinal
cord. On opening the spinal column it is easy to expose the cord,
which is then seen to have six pearl-like swellings on either side
immediately posterior to the medulla. The three posterior of
these are the central connections of the nerves from the tliree
sensory filaments, while the tlu-ee anterior receive by a common
trunk the nerves from the pectoral fin.

We see here in a simple way the principle, that is still further
developed in the brain, namely an increase in the size of a nerve
centre when it is called upon to deal with a new function. In the
gurnard as soon as certain rays of the pectoral fin took on the new

BRAIN PATTERN 27

function of tactile organs there was developed a specialised areas in
the spinal cord to receive the sensations of touch.

It must not be assumed that there' is no tactile sensation in the
pectoral fin, because the most posterior rays of the fin may be seen to
approach in structure the sensory filaments. But the main func-
tion, propulsion, of the fin has been lost or altered by the develop-
ment of the fingers, so that two functions can be distinguished,
propulsion in the anterior portion and tactile sensations in the pos-
terior, although according to Couch the fingers at times seem to
act as organs of locomotion. Other examples of the change of
functions in a fin will be described later, when it will be necessary
to discuss the habits of the rockhngs and their brain pattern.

CHAPTER III
THE CARPS

It is fortunate that the most suitable fish for the study of form and
function, and the relation of bodily structure to brain pattern, are
among the commonest fish in the rivers, lakes, and broads of our
native land. We refer to the carps or Cyprinoids, and for our
purpose we have examined the brains, both by the naked eye and
by microscopical methods, of the following fishes, the carp, goldfish,
bream, tench, roach, rudd, dace, minnow, chub, bleak, barbel,
gudgeon, and the loach ; the last, though not strictly a Cyprinoid, is
so nearly related, and so similar in many ways to the gudgeon,
that it is included in the list. British Cyprinoids subsist on a mixed
diet, some are almost entirely vegetarian, some feed on insects,
shrimps, worms, flies, and larvae and some on small shellfish. A
few are at times predacious, as the chub, and others, found mostly
in large lakes, are almost entirely surface feeders and might almost
be called plankton-feeders.

It will be found that these fish can be arranged into groups
according to their diet and habits, and that each group has a charac-
teristic brain-pattern ; and further it is found that these groups
correspond to the groups given by Tate Regan in his synopsis, based
on external characters only, of the British species of Cyprinoids.
It is necessary to supplement the naked eye description of the brains
of these fish by microscopic methods, but in the first place we will
be content to describe the naked-eye appearances ; but it may now
be mentioned that a lack of the study of serial sections has led to
many erroneous statements in the past. A study of the Plate 3
will enable the reader to see at a glance the striking difference in
the brain-pattern of the three groups ; but a short descriptive
account will serve to emphasise the more important details.

Group I. — Carp, Goldfish, Bream, and Tench.

The vagal lobes are large and oval or crescentic in shape. In the
carp the facial lobe is not overlapped by the vagal lobes, as it is in
the goldfish. In the bream the facial lobe is small and lies further
forward than in the other members of the group. In the tench

28

PLATE 3.
Group I. — Vagal Lobes, Large.

29

M U

J ^

f n /p

''^ ^

n

'^z i

■iK. '"

>

? o /H

■'S^ ^

5

[^ o

J 1 ~

O

Group II

O

S:

c

■.Og:

JJ

CD

o i

-Vagal and Facial Lobes, Small.

c
o

Group III. — Faclil Lobes, Large.

S

V.L. — Vagal lobe. F.L. — Facial lobe. Clm. — Cerebellum. A.T. — Acoustic tuber-
cles. OP.— Optic lobes. OK.— Olfactory lobe. OLF.B.— Olfactory bulb.
Pall, pallium. — The pallium is only shown in the Carp and Dace.

Note. — In the barbel, loach and goldfish the optic lobes appear large, as the tecta
optica are separated by the valvula cerebelli. The olfactory lobes in the third
group are large. The brain of the tench might almost be placed in Group III,
as the facial lobe is large as weU as the vagals.

30 BRAIN AND BODY OF FISH

the facial lobe is large and separates the anterior ends of the vagals.
In fact the tench has some of the characteristics of Group III, as
will be evident when we recall the facts given in the last chapter,
when describing the division of the facial nerve.

Group II. — Roach, Rudd, Dace, Chub.

The optic lobes are large when compared with the vagal and
facial lobes. The vagal lobes are small, and the facial very small,
but is really larger in depth than would appear from a superficial
examination ; this is evident when serial sections are examined.

Group III. — The Barbel, Gudgeon and Loach.

In these three fish the vagal lobes are well-marked, but the facial
lobe is very much enlarged, particularly so in the gudgeon and the
loach ; but it will be found that the facial lobe of the barbel has
a very considerable depth, and it widely separates the anterior ends
of the vagals. In this group it will be noticed that the optic lobes
are separated posteriorly, especially in the barbel. This is due to an
extension of the cerebellum, inserting itself between the tecta optica,
which form the roof of the optic lobes. This extension of the cere-
bellum is known as the " valvula cerebelli." As valvula has a very
different application in human anatomy, its use here is unfortunate.
The " valvula cerebelli " is peculiar to fish and its function is specu-
lative. It is small in the cod, larger in the carp and still larger in
Group III, reaching its maximum in the barbel among the Cyprinoids.
It is of immense size in a family of African fishes, known as the
Mormyridse, which we shall have to describe in a later chapter,
but we may note here the great development of the snout. The
question of its significance may some day be solved by the study
of its comparative anatomy ; for the present the only clue seems to
be, that it is most developed in ground-feeding fish with a snout-
like proboscis.

Mormyrus has a small mouth at the end of a more or less
elongated snout, so that it is sometimes called the elephant fish.
" The barbel has a rather long snout, with the uj^per profile decurved,
and the gudgeon is rather similar to the barbel in general form as
well as in the shape of the head " (Regan), and both grope and grub
for their food.

Although the feeding habits of British Cyprinoids are similar,
in that they subsist on a mixed diet, some are almost entirely
vegetarian, while other are predacious. They can however be
divided into three groups according to their diet.

THE CARPS 31

Group I, which contains the Scanie fish as appear in Group I of
the table, is characterised by their liabit of extracting nutrient
material from mud. Yarrell gives as the food of the carp the larvae
of insects, worms, and the softer parts of aquatic plants. Shrimps
are also eaten. Bream swim in slioals feeding on worms and other
soft-bodied animals with some vegetable substances (Yarrell).
Walton gives as baits for a bream " paste of brown bread and honey,
gentles or the brood of wasps that be young." There is at the root
of docks or flags or rushes in watery places " a worm like a maggot
at Avhich tench will bite freely." But for the carp or bream he
recommends " as big a red worm as you can find without a knot " ;
but this must be carefully cleaned with moss that must be changed
fresh every three or four days. But the dominant characteristic
of this group is the power of siftmg mud and extracting nutriment
from the organic matter it contains.

Group II. — This includes besides the fish, the brains of which
are figured in Group II of the plate, the minnow. All will take a
fly, and the rudd and the dace give good sport to the dry fly fisher-
man. The dace feeds on weeds, insects larvae, and flies. The food
of roach and rudd is very similar. But the chub is a predacious
fish ; he leaps at flies or feeds at the bottom on weeds, slirimps,
worms, or young frogs, and also preys on mimiows and gudgeon.

According to Walton the chub will take a grasshopper. He
recommends as baits, " a black with its belly slit to show its white
or a piece of short cheese. Nay, sometimes a worm, or any kind of
fly as the ant fly, the flesh fly, or wall fly, or the dor or beetle you
may find under cow dung ; or a lob which you will find in the same
place and in time will be a beetle ; it is a short white worm like to
but bigger than a gentle."

Group III. — The feeding habits of the gudgeon and barbel are
well described by Isaac Walton. The gudgeon frequents gravelly
bottoms, " the Germans call him the groundling by reason of his
feeding on the ground and on the gravel : and he there feasts him-
self in sharp streams. He and the barbel both feed so and do not
leap for flies at any time as most fishes do. He is easily taken with
a smaU red worm. The food of a barbel is partly of an animal and
partly of a vegetable nature. He does not disdain any sort of
vegetable matter, which he finds by rooting about on the bottom
or banks with his snout often turning over stones and using the
barbels as tasters in search of food." In general it may be said that
their diet is one of shrimps, small mulluscs, insect larvae, worms,
and the eggs or fry of other fish.

32 BRAIN AND BODY OF FISH

A comparison of the groups described above and shown in the
Plate with Tate Regan's Synopsis according to external charac-
teristics only can now be made.

Group I. — Dorsal fin long, anal fin short. Last simple ray more
or less spinous or serrated.

Carp with two barbels.
Crucian carp and goldfish with no barbels.
This corresponds with Group I according to brain pattern as
above.

Group II. — Dorsal and anal fins short.
A. — Mouth with barbels.

Barbel, gudgeon, and tench.
This compares with Group III of brain pattern.
B. — Mouth with no barbels, scales small.
Roach, rudd, chub, dace, etc.
These correspond with Group II of brain pattern.

Group III. — Dorsal fin short, anal fin long Abdomen compressed
and forms a sharp keel over which the scales do not pass.
Bream and bleak.

It is true that the bream has a brain of the carp type, but it is
the only fish that does not fall into line.

This forms a fourth group from the point of view of brain
pattern which will shortly be described. These relations of form to
brain pattern are facts that are as interesting as they are remark-
able. The apparent discrepancies are easily explained. As regards
the tench we have already noticed that it has large vagal lobes, but
also has a large facial lobe and that the facial nerve divides on enter-
ing the lobe ; my original view is, therefore, probably incorrect, and
the tench should have been placed in Group III of brain pattern.
The only fish that fails to fall in with my classification is the bream.
Its relations to the bleak requires more research as the bleak and
other fish with a keel have a very unique brain pattern and are,
moreover, plankton-feeders.

An examination of the brain of a Cyprinoid inhabiting the large
lakes of tropical Africa has furnished material wliich has enabled
us to describe the fourth group of Cyprinoid brain, in which it will
be found that the brain of the bleak can be included. In 1929, Mr.
Michael Graham made a report on a " Fishing Survey of the
Victoria Nyanza," and brought home a specimen of a Cyprinoid,
Engraulicypris argenteus, which he kindly placed at my disposal
for the purpose of an examination of its brain. An interesting

THE CARPS 33

fact noticed by Graliam is that this fish has ceased to behave as a
Cyprinoid, and lias become entirely a plankton-feeder. Its name
has an obvious reference to its habits and appearance, Engrauli- or
anchovy, Cypris or carp, that is to say, an anchovy-carp, that
recalls the habits of a Clupeoid fish, but is really one of the carp
family. To quote the Report, " so far as the evidence goes this fish
resembles the Clupeoid fishes in its pelagic or open water habitat,
as well as in its structure. We frequently observed Engraulicypris
apparently catching Copepods and other members of the
plankton near the surface. Their stomachs contained Cladocera
or Copepoda."

" Tliis is an interesting example of fish belonging to a typically
river family, the Cyprinoids, taking on a very different form, where
the conditions resemble the sea, especially in the abundance and
stability of a rich population of plankton, and adopting not only
a pelagic existence but the shape and form of a pelagic family."
The resemblance to marine fish does not end here ; " some floating
segmenting eggs in the plankton, which evidence seemed to prove
were those of Engraulicypris, were found," and this fact is com-
mented on " as so far as I know this is the first record of a floating
egg in a fresh- water fish."

On exposing the brain of this fish it was at once apparent that
no facial lobe was to be seen in the usual position, not even a small
one as is found in the roach group.

It was also found that the external appearance of the brain of
a bleak was very similar to that of Engraulicypris, and we shall be
able to show that these fishes have the same pattern of internal
brain structure when they are examined by microscojDical methods.
My first observations on the bleak were made during a visit to the
Lake Annecy. Both in this Lake and the Lake of Geneva the silvery
scales were used for the manufacture of artificial pearls. This
industry has almost died out as the Japanese culture of artificial
pearls has driven it out of the market. I noticed a mass of bleak
lying on a fishmonger's slab at Annecy, which was strikingly sug-
gestive of a number of smaU herrings lying in a mess of blood-
stained mucus with detached scales. In turning over the pages of
the " Complete Angler," I was struck with the following jjassage,
" The Bleak or freshwater Sprat, a fish that is ever in motion at the
top of the water, ought to be much valued though we want Allomot-
salt and the skill of the Italians to turn them into anchovies."
Both the sprat and the anchovy belong to the herring family so
that it was apparent to Isaac Walton that the bleak was similar
in appearance to the jDlankton-feeding Clupeoids.

34 BRAIN AND BODY OF FISH

Now, after all these years, a study of the brain of this group of
the carp family reveals that they have a brain pattern characteristic
of plankton-feeding fish in addition to many similar external
characters.

In the next chapter we propose to study the pattern of the brain
of Engraulicypris and bleak by the method of serial microscopic
sections, and we shall find the great importance of this method in
unravelling the intricacies of the central connections of the organs
of special sense and their relative size.

CHAPTER IV
THE CARPS— Continued

As sufficient information cannot be obtained of the various lobes of
Engraulicypris by the naked eye, it becomes necessary to employ
another method of observation ; this is the method of examining
the internal structiu-e of an organ by cutting serial sections. To
those Avho are unfamihar with this method, the following sketch
of the process may be useful. The tissue to be examined is first
" fixed," that is to say, put in a solution, which prevents chstortion
or destruction of the cellular elements, which might occur in the
later stages of the process. After being fixed, the tissue is
" hardened " by various reagents ; in the case of brain tissue, the
cheesy consistence is changed to that of soft leather. The specimen
is now embedded in paraffin of low melting point and allowed to
stand in an incubator until the paraffin has thoroughly permeated
the tissue. It is then taken out and allowed to cool. The sohd
paraffin block is then put in a proper position on a microtome and
very thin shces are cut in series, which come away from the micro-
tome in ribbons, lengths of which are placed on shdes, and numbered..
The paraffin is then dissolved out, and the sections stained in the^
appropriate manner to show up details. These are examined
seriatim and successive drawings made under the microscope.

Suppose the first drawing was made of a transverse section of
the hinder end of the medulla, each section, as you pass forwards,
is studied and, when a modification of the picture occurs, a fresh
drawing is made and so on, until you have examined and made
pictorial notes of the whole of the medulla. In this way it is possible
to make a mental picture and build up the structure of the various
lobes and plot them out. Also the intimate details of the tissue are
made clear, and the different types of cells recognised, and the course
of the connecting strands of nerve fibres foUowed. The latter often
form definite tracts to which terms are applied, signifying their
supposed functional significance.

To help the reader to follow the interpretation of the series of
transverse sections of the medulla oblongata and cerebellum of
the roach, we give a half-a-dozen diagrammatic drawings of this

35

36 BRAIN AND BODY OF FISH

area of the brain. The series begins at the bottom of the Plate 4,
and Fig. i is a section of the hinder end of the medulla across the
vagal lobes, where they are most prominent. These are separated
by a deep cleft, the rhomboid fossa, which is the opening out of the
fourth ventricle. At the base of the fossa on either side there is a
group of large motor cells, known as the nucleus ambiguus from
which pass the efferent fibres of the motor root of the Xth or vagal
nerve. External to this motor nucleus on either side is a large
bundle of longitudinal fibres cut transversely, which are a prominent
featiu"e of the medulla and can be traced in four of the sections of
the plate. This bundle is known as the great longitudinal secondary
gustatory tract, and in it can be distinguished three divisions, an
upper (that is most dorsal), known as the spinal root of the fifth
nerve, a middle known as the descending gustatory tract which
receives fibres from the facial lobe, and the lower the ascending
secondary gustatory tract. In the middle line just below the rhom-
boid fossa are a number of bundles of nerve fibres known as the
longitudinal median bundles (or the fasciculus medialis).

Fig. ii is a section somewhat anterior to the preceding and the
vagal lobes no longer appear. In the site of the rhomboid fossa is
now seen the facial lobe, and below it is the central canal or ventricle.
On either side of the central lobe we see the commencement of the
fifth lobes. These represent the skin areas, so are usually called the
somatic -sensory lobes. Descending fibres pass on either side into
the gustatory tract from the facial lobe, and crossing these descend-
ing fibres pass from the fifth lobes to cross each other tlu"ough the
median longitudinal bundle. The facial nerves are also to be seen
at the base of the facial lobe into which their fibres gradually pass.

Fig. iii is very similar to the section which has been just des-
cribed, but the fifth lobes are much more prominent and almost
entirely embrace the facial lobe, which is much diminished in size.

Fig. iv. — The facial lobes are replaced by the two trunks of the
facial nerves which are passing backwards to enter the lobe as was
seen in Fig. iii.

Fig. V. — Here the cerebellum first appears and is joined to the
medulla laterally by the acoustic tubercles. The facial nerves are
now cut longitudinally and appear to pass transversely towards the
ventricle ; the fifth also are seen on either side.

If we study a similar series of sections of the madulla oblongata
of Engraulcypris, a very different picture will be observed. We
have seen that the naked eye appearance of the medulla shows the
absense of a facial lobe as seen in the roach, and the vagal lobes are
not easily identified. c

PLATE 4

Sections of medulla of Roach.
S.G.— Stratum granulare. S.M. — Stratum moleculare. C.A.A.— Central acoustic
area. A.T. — Acoustic tubercle. L.L.N. — Lateral line nerve. VII. — Seventh
or facial nerve. Vth. — Fifth or somatic-sensory lobe or nerve. G.T. and
G.L.S.G.T.— Great lateral secondary gustatory tract. Xn. — Tenth or vagal
nerve. D.F.V. — Descending fibres of fifth lobe. D.F.VII. — Descending fibres
of facial lobe.

38

PLATE 5.

YS"*^

Sections of medulla of Bleak.

iS.G. — Stratum granulare. S.M. — Stratum moleculare. C.A.A. — Central acoustic
area. A.T. — Acoustic tubercle. L.L.N. — Lateral line nerve. VII. — Seventh
or facial nerve. Vth. — Fifth or somatic-sensory lobe or nerve. G.T. and
G.L.S.G.T. — Great lateral secondary gustatory tract. Xn. — Tenth or vagal
nerve. D.F.V. — Descending fibres of fifth lobe. D.F.VII. — Descending fibres
of facial lobe.

THE CARPS 39

The serial section Plate 5, Fig. i, shows the vagal lobes, which
are very small compared with the vagals of the roach, as figured in
Plate 4, Fig. i. The fii'st lobe to produce a definite prominence
dorsally in Engraulcypris is the Vtli or somatic- sensory lobe, and on
either side this lobe hides the vagal as seen in Plate 5, Fig. ii. The
vagals are here club-shaped in section, with the rounded dorsal
portions tending to meet in the middle Une. In Plate 5, Fig. iii,
the junction is complete., and the fifth lobe have also approached
the middle line. The presence of a small facial lobe is now recognised
by descending fibres passing from the central lobe, formed posteriorly
by the united vagals, but now formed by the medium fusion of two
very small facial lobes. Descending fibres passing laterally down-
wards and out"s\"ards into the great longitudinal secondary gustatory
tracts, an important bundle very obvious in the cjrprinoicl medulla.

This identification of the facial lobes is confirmed if the section
shown in Plate 5, Figs, v and vi are examined. In Fig. v, the Vllth or
facial nerve is seen cut in section, and in Fig. vi the same nerve is seen
cut horizontally as it passes in its usual course from the periphery
to the margin of the ventricle. To enable the reader to follow the
differences in the medullae of the roach and the plankton-feeding
type as illustrated by EngrauHcypris we will summarise the matter.

In the roach the naked eye examination shows two lateral pro-
minences posteriorly the vagal lobes, and these embrace anteriorly
a small facial lobe. The sections confirm this picture ; in Engrauli-
cj^pris neither vagal nor facial lobes are seen superficially and can
only be recognised microscopically.

We now come to that part of the brain wliich hes between the
medulla and the optic lobes and consists of the cerebellum and its
lateral supports connecting it with the medulla, which give rise to
the prominences known as the acoustic tubercles or the acoustico-
lateralis areas. These can be seen on either side of the base of the
cerebellum in Plates 4 and 5, Fig. vi. The acoustic tubercles
receive afferent fibres from the eighth nerve or auditory and from
the laterahs nerve which is the nerve of the organs of the lateral
Une of which we shall speak later. The function of the cerebellum
and acoustic tubercles has been the subject of a great deal of
theoretical speculation, but the former is usually held to be associated
with the perception of position in space, as recorded by the semi-
circular canals of the internal ear. The cerebellum may be simply
a globular protuberance or may be tongue-shaped, but it always has
a characteristic internal structure ; it has a core which consists of
darkly staining cells, called the stratum granulosum, which is
surrounded by a marginal layer of pecuhar cells, which lie in an

40 BRAIN AND BODY OF FISH

outer layer varying very much in thickness, which has a uniform
consistence, and stains with difficulty, called the stratum moleculare
{see Plates 4 and 5, Figs, v and vi). In Fig. v the cerebellum is
separated from medulla, but in the last section, Fig. vi, the cere-
bellum is no longer separated from the medulla and the tissue that
unites it has two large lateral prominences, the acoustic tubercles,
which have a granular structure rather finer than the granular layer
of the cerebellum, but also staining deeply. Below on either side
Plate 4, Fig. vi, are seen the lateral line nerves entering these
lobes.

We have deferred from speaking of the lateral line system till
now as it is a controversial subject. There is present in fishes a
system of small sensory canals widely distributed under the skin.
These contain sensory organs somewhat similar to those of the semi-
circular canals of the internal ear and their functions are probably
intermediate between those of the organs of touch in the skin and
those of the internal ear, responding to water vibrations of low
frequency, and probably in the orientation of the body in space.
These are the lateral line canals and we are all familiar with the
lateral line in fish, which is so clearly seen running along the side
of fish nearly midway between the dorsal and ventral margins and
often made more obvious by being pigmented. Similar canals are
also found on the head. These canals are supplied by special roots
of the following cranial nerves which aU finally enter the acoustic
tubercle or acoustico-lateralis lobe ; these are the seventh or facial
nerve, the ninth or glossopharyngeal nerve, and the tenth or vagal
nerve. These, together with the eighth or acoustic nerve, are asso-
ciated with the reception of vibatory sensations of a greater range
from those of hearing proper to slow vibrations such as are felt by
the skin of man and also by what is known as bone conduction.

It will be our aim to see whether the methods of comparative
anatomy may not help to unravel the central areas which must
presumably be associated with these various functions, and with
this object in view we must draw the attention of the reader to an
area of small cells interpersed with transverse nerve fibres which
appears at the base of the cerebellum just before its free portion
joins laterally with the medulla. This area is known as the central
acoustic area and at times forms a definite lobe. It is well shown
in Plates 4 and 5, Fig. v, and it will be observed that it is more
prominent in Engraulicypris than in the roach. In the bleak it
is still more prominent. Fibres pass to this area from the eighth
nerve, and there are also fibres passing from it to the acoustico-
lateralis area.

THE CARPS 41

Besides the lateral line organs there are diffusely scattered
sense organs in the skin also innervated by the lateralis system and
both arise from a " common rudiment in the epidermis of the embryo
in the position of the future auditory organ. This rudiment grows
baok^^•ards along the side of the body and also forwards." {Camb.
Nat. Hist.). This seems to be an important fact in the discussion
of the acoustic tubercles. There is very little known about the
acoustico-lateral area of fishes, except the fact that it receives all
the nerve fibres from the internal ear and from the several kinds of
lateral line organs. Herrick states " that the central terminations
of these different kinds of fibres are so intertwined within this area,
that it has not been possible hitherto to separate completely the
reflex centres of the many diverse functions, represented in this
complex system of peripheral sense organs. There is, however,
an incomplete separate localization within this area of several
specific functions, but the reflexes, served by all of the organs of
the acoustico-lateral comjDlex, are evidently in very close physio-
logical association. These reflexes fall into three groups : I. —
Postural and equilibrial, served chiefly by the semicircular canals of
the ear ; II. — Auchtory, served by the end-organs of the saccule ;
III. — Reactions following excitations of the lateral line organs by
slow water vibrations and by other agents as yet imperfectly
known."

But before we enter on further discussion of the function of the
acoustic tubercles it will be necessary to give a short description of
the organ of hearing in fish, and also mention the various accessory
organs of hearing w^hich are found in certain families. This sub-
ject deserves a special chapter, as it is rather a long story, and it
treats of many facts which have only been firmly estabUshed by
recent research.

icc^

f J

^ « ,. .j!v^^^_/p>ir^(^ -* '^ '

^^.3>^-*ff^-

( ^^Cim^A — Mice. \

M^

r|LDtrlcle.

otolith of Saccule

Otolltb

of Saccule

CHAPTER V
HEARING IN FISH

Hitherto, very little has been said about hearing in fish, and the
function of the eighth or auditory nerve ; most people will not be
surprised by these omissions, and the more knowledgeable among
them would say, fishes cannot hear they only perceive vibrations ;
this, however, is not correct, as accurate anatomical investiga-
tions supported by the most convincing physiological experi-
ments have proved that many fish have a wide range of hearing

/b.SC

Fig, i. — Two drawings of the ear of Herring from specimens prepared
by the author, and internal ear of man after Schafer.

S.C.C. — Semi-circular canals. H.S.C.C. — Horizontal semi-circular canal. AMP. —
Ampullae, c.c. — Canal of cochlea, e.s.c. — External Canal, s.e. — Saccus
endolymphaticus.

and that the minnow can recognise notes of as wide a range as can
the human ear. The ear of a fish can be best understood by a
reference to the two diagrams which compare the general structure
of the internal ear of a tyj)ical fish with the human ear.

The lettering which accompanies these diagrams wUl be sufficient
to make clear all that is essential for the reader to understand for
our present purpose. But there are many fish that have an accessory
organ, in addition to those parts of the organ of hearing, known as
the saccule and utricle with its three semicircular canals. We should
expect that these fishes would be provided with a more speciahsed
centre in the brain, and, therefore, before studying the central
connections of the auditory nerve, it would be wise to consider the
nature of these accessory organs. In pursuit of this idea we are led

42

HEARING IN FISH

43

to the study of the air-sacs or swim-bladder of fish, and it is found
that this organ is capable of presenting a great variety of methods
of increasing the auditory function in different families of fish.

PLATE 6.

BARBEL

TH£UMATlC r>UCT

'SULLET

Diagram of Fish to show relations
of swim- bladder and auditory
vesicle to internal ear and dis-
tribution of the taste-buds.

Distribution of taste-buds in Man
and relation of air-passages to
the middle-ear.

Black dots signify site of taste-buds.

The history of the swim-bladder is one of the most remarkable
tales in the development of present day fishes. Although the swim-
bladder was originally a lung, it is now^ largely used as a hydrostatic
organ or buoyancy tank ; nevertheless, there is often a vesicle or
secondary air-bag given off from its anterior end, which is connected
by various means with the internal ear. The very large family
known as the carps or Cyprinoids have a swim-bladder lying free
in the upper part of the abdominal cavity only attached to a small

44 BRAIN AND BODY OF FISH

plate of bone projecting downwards from the vertebral column to
which the anterior sac or air vesicle is not only attached by its outer
wall, but also by a very important muscle rising from an ossicle,
the first of the chain of small bones leading from the anterior sac
to the internal ear. These are known as the Weberian ossicles.
(•Plate 7.)

In the carps the posterior air-sac, which acts as a buoyancy tank,
has a tube connecting it with the gullet, and this duct, known as the
pneumatic, allows swallowed air to be not only introduced into the
bladder by means of the pneumatic bulb or pump, but also to be
discharged when necessary in accord with the hydrostatic require-
ments. Between the two sacs there is a communicating duct which
is kept closed by a sphincter muscle, controlled by a nervous gang-
lion ; this enables the gaseous pressure in the anterior vesicle to be
kept at a level most suitable for the reception of vibrations ; the
function of the anterior sac is that of a drum, which acts as a hydro-
phone. It may be said that this drum combines the functions of
a vibrating membrane and that of a middle ear, while the " ductus
communicans " has the same physiological use as the Eustachian
tube in the human ear. Vibrations received by the body wall of
the fish are communicated to the anterior sac directly and not by
any external ear, which does not exist in fish. There is another
important fact to be noted, namely, that the orifice of the pneumatic
duct as it enters the gullet is surrounded by a ring of taste-buds,
which act as sentinels protecting the orifice, and only allow bubbles
of swallowed air to pass into the posterior sac.

If the two diagrams (Plate 6 and 7) of the coimections of the
lung in man, and the swim-bladder in fish, with the auditory
organ are studied, a remarkable similarity in general arrangement
can be traced. In the human ear there is an external ear which
leads to a drum which forms an outer wall to the middle ear. To
this drum is attached a chain of ossicles or small bones which trans-
mit vibrations to the internal ear, and the gaseous pressure in the
middle ear is controlled by means of a tube, the Eustachian tube,
that leads into the upper part of the pharynx which may be looked
upon as an extension of the air-passages upwards.

In the carp, an external ear being non-existent, the anterior sac
or air-vesicle receives the sound vibrations directly through the
body walls and from its anterior end there arises a chain of ossicles
that transmit these vibrations to the internal ear. Weber, who first
described these ossicles, named them after the small bones of the
human ear, malleus, stapes and incus ; but those comparative
anatomists, who followed him, did not consider that these bones

PLATE 7.

45

5.CC

WORMYRUS

Accessory auditory organs of Carp, Gymnotus, Herring and Mormjrrus.
W. — Wall of auditory vesicle. S.B. — Swim-bladder or buoyancy tank. A.V. —
Auditory vesicle. T. — Tripus. P.S.A.V. — Posterior spherical air vesicle.
A.S. — Anterior spherical with transverse membrane. PYR. — Pear-shaped air
vesicle. P.D.— Pneumatic duct. C.C— Cross canal. G.— Gullet. S. — Saccule.
R. — Round vesicle. U. — Utricle and ampullae. S.C.C. — Semi-circular canals.
E.— Epidermis. T.F. — Tympanic fenestra. G.L. — Ganglion with papilla.
v.— Vent.

46

BRAIN AND BODY OF FISH

had any auditory function so that a new terminology was given to
them, the tripus, scaphoid, and claustrum. But we have now
returned to Weber and it has been proved that his views are correct.
At the hinder end of the anterior vesicle we have noted the
ductus communicans and here we have the representative of the
Eustachian tube as it controls the gaseous pressure in the middle
ear. The posterior sac as we have aheady stated was primitively

— Canalis communicans.

— Posterior cavity.

■ — Scaphoid and Claustrum.

— Intercalarium.
— Vertebral column.
— Tripus.
— Central plate.
— Tensor tripodis.

— Membrana flaccida.
£.C — External coat of A.S.

• — Anterior sac.
^^ ■ — Posterior attachment of A.S.
1)0. — Ductus communicans.
S — Sphincter.
^^ — Pneumatic duct.

— Posterior sac.

Fig. ii.
Dorsal view of swim-bladder and Weberian Ossicles of Carp.

a lung and this is connected with the gullet by the pneumatic duct,
which corresponds to the larynx and, just as the larynx is guarded
by the epiglottis with its taste-buds, so the pneumatic duct is
guarded by its ring of sentinel buds.

There are many other interesting details which make this com-
parison of the organs of hearing in man and fish very striking. We
may mention two : The tripus or malleus of a fish is kept tense
by a small muscle, by which it is attached to the central plate, and
in the human ear there is a small muscle that controls the malleus,
the tensor tympani. The outer wall of the anterior vesicle in fish

HEARING IN FISH 47

consists of a very friable membrane composed of fibres running in a,
criss-cross fashion, \vliile the inner wall, which is only loosely
adherent to the outer wall, can be rejuoved with its contained
gases without interfering with the attachments of the ossicles.
The membrana tympani of man has a very similar fibrous structure
to that of the external coat. Other minute details might be des-
cribed, and these can be studied in the more teclinical papers that
have appeared in the Proceedings of the Royal Society and the
Transactions of the Royal Society of Medicine (Evans).

Before we leave the subject of the accessory organ of hearing
in carps, it wiU be of great interest to study the modifications of
the swim-bladder and Weberian ossicles in a tropical air-breathing
fish. We were very fortunate to be given by Mr. Burne, the physio-
logical ciu-ator of the Royal College of Surgeons, a specimen of the
electric eel, which belongs to a family closely related to the carps.
The diagram (Plate 9) of its swim-bladder and the auditory con-
nections will enable the reader to follow these modifications, and to
compare them with the corresponding organs in a carp. The most
striking fact is the very large size of the posterior sac, or buoyancy
tank, which has a volume ten times larger than an average sized
carp. If the pneumatic duct is traced back to the gullet it will be
found to enter an ovoid chamber before it actually enters the
oesophagus. This chamber is the air-pump, which pumps the
swallowed air into the sac. From this chamber there also passes
a fine duct that runs forwards and enters a small air-vesicle which
lies in the anterior part of the abdomen. From the front of the air-
vesicle the small bones, or Weberian ossicles, pass forwards to end
in the duct that communicates with the internal ear or saccule.

The most striking point in this arrangement is the complete
separation of the hydrostatic sac from the auditory vesicle which
makes the similarity of the duct leading to the air vesicle, to the
Eustachian tube of man, very convincing.

In the course of our dissection of this fish other points were
observed of interest which hitherto had not been completely investi-
gated. The famous surgeon and scientist, John Hunter, noticed
and described a number of fohate projections in the floor and roof
of the mouth ; he did not, however, recognise the function of these
outgrowths. W^hen these foliate papillse were examined micros-
copically, it was found that they had a superficial layer of air spaces
like the alveoli of a lung, and there can be no doubt that, when the
mud flats are parched by the sun, the electric eel is able to obtain
air by this accessory mouth-breathing organ.

It is not necessary to go into the details of the connections of

48 BRAIN AND BODY OF FISH

the Weberian ossicles with the saccule by means of certain ducts
which lead to a space called the " atrium sinus imparls." But it
will be interesting to quote some conclusions arrived at by Sorensen :

" I. — The wall of the air-bladder is capable of vibrating syn-
chronously with rapidly recurring sound waves.

" II. — The tripus is thrown into vibrations when the wall of
the bladder is vibrating.

" III. — All movements, also, vibrations of the tripus are trans-
mitted, by means of the tight inter-ossicular ligament, to the rest
of the Weberian ossicles and in this way to the atriun sinus imparls.

" IV. — The tones of the air-bladder can be transmitted to the
water without losing much in strength, and if so, vice versa, sound
waves can be transmitted from without to the air-bladder."

Recently, Prof. K. von Frisch has shown by experiments that
have been conducted on the principle of Pavlov's conditioned reflexes
that the minnow has a range of hearing as wide as has the human
ear. But we are the more indebted to him for putting on an experi-
mental basis, the fact that the swim-bladder has an important
share in the hearing of those fish with a Weberian apparatus. He
tested the range of hearing in minnows both before and after removing
the swim-bladder and found that hearing remained in those fish
which had had the swim-bladder removed, but that it was iveakened.
" We see, therefore, that those fish in which the swim-bladder is
connected with the labyrinth by Weberian ossicles, have an apparatus
through which the acuteness of hearing is increased."

We are now able to tiu'n to the discussion of the methods of
feeding of the fourth group of Cyprinoids as typified by the African
fish Engraulicypris and the bleak, which, as far as we were able to
judge, were mostly dependent on sight in the search for food. When
the serial sections of the roach and Engraulicypris were described
in a previous chapter there was noted a central area of round cells
with interlacing transverse fibres at the base of the cerebellum
connected laterally with the acoustic tubercles ; this we termed the
central acoustic area. When we come to discuss the auditory organ
of the herring and its central representation in the brain it will be
possible to give more fully the arguments in favour of adopting this
term and of associating this with audition.

It has already been i)ointed out that Engraulicypris has a rudi-
mentary facial lobe, and that the gustatory function must be smaU ;
that it has large optic lobes and that it has a surface habitat and is a
jjlankton feeder. The study of its central acoustic area, however,
seems to indicate that hearing may be of importance in regard to its

HEARING IN FISH

49

PLATE 8.

BLEAK

Four transverse sections of the medulla oblongata of the Bleak.

I. — Posterior across the vagal lobes. II. — More anterior across the small facial
or Vllth lobe. III. — Across the central acoustic lobe. IV. — Most anterior
across the cerebellum, acoustic, tubercles and central acoustic area. Vag.L. —
Vagal lobe. V.L. — Fifth or somatic-sensory lobe. VII. L. — Facial or seventh
lobe. C.L.M. — Cerebellum. S.G. — Stratum granulosum. A.T. — Acoustic
tubercle. VII.N. — Facial nerve. V.N. — Fifth nerve. C.A.L. — Central acous-
tic lobe.

D

50 BRAIN AND BODY OF FISH

surface feeding habits, as this area is very highly developed ; this
is also the case in the bleak, which has a similar habitat and method
of feeding. In fact, in the bleak figure, the central acoustic area
becomes a definite lobe which projects backwards from the base
of the cerebellum. The central acoustic area is also seen in the roach,
but is less marked ; but in the bottom-feeding Cyprinoids it is only
feebly represented. These facts suggest that surface feeding fish
find it advantageous to have an increased power of hearing. It is of
great interest to find that these considerations receive strong support
from observations made by Bhimachar on the Cyprinoids of the
Madras tanks.

In a paper which was published in the Proceedings of the Royal
Society, he states, " The acoustic area or lobe is the terminal centre
in the brain of the auditory function. This area is very prominently
developed in all the sight feeders, and fairly well developed in the
ground feeding fish which come to the surface to take in air. But in
the purely ground feeding fish as Nemachilus, which is not exposed
to the influence of external sound waves this area is almost com-
pletely absent. Compared with the central acoustic area of the
British planliton feeding fishes, such as the bleak, the Indian forms
like rasbora, nauria, etc., have not only a larger central acoustic
area but also an extension of this area behind the cerebellum in
the form of a distinct central acoustic lobe. Tliis is evidently due
to a more perfect sight-feeding habit of the trojjical fishes and their
consequent exposure to the effects of external sounds. In the
accessory air-breathing Cyprinoids and Siluroids the extent to
which the central acoustic area or lobe is developed gives a strong
indication of the air-breathing habit."

This confirmation of the observation of a British observer by
an independent Indian naturalist makes the conclusion, that the
auditory function of the central acoustic area or lobe in siu-face
feeding Cyprinoids is justified, will be supported by further observa-
tions on several other families of fish as w^e shall have occasion to
describe in the following chapter.

We have now reviewed the relations of all the special senses to
the various lobes of the carp family and we arrive at a certain
conclusion which may be expressed as a law, "The pattern of the
brain of a bony fish is determined by the proportional representa-
tion of the special senses in its feeding or hunting equipment."

CHAPTER VI
ACCESSORY ORGANS OF HEARING

We heave seen in the carps that from the anterior end of the primitive
lung, represented by the posterior sac, there grows a vesicle which
lies in the abdominal ca^aty. In two important families of present
da}' fishes, namely, the herring family or Clupeidse, and an African
family the Mormyridse, we find a duct given off from the swim-
bladder anteriorly, which divides into two finer ducts, and these
end in a vesicle or vesicles connected with the internal ear. The

^ Ca/'e^nai l>lcJ-£

Fig. iji. — Head of Mormyrus Kannume (Worthington) to show
position of external auditory orifice.

Diameter of eye -9 cm. Distance eye to ant. foramen TS cm. X Exposed portion
of sac '5 cm. F Vertical diam. foramen "7 cm.

connection of the vesicle with the bladder is n(»t permanent in the
Mormyridge, but its rudiment remains ; whereas in the herring it
persists in the adult. In these famihes the connection with the
auditory organ is within the cranial cavity, so that there is no need
for a series of ossicles, as is present in the carps. We propose firstly
to describe the more simple and less controversial of these two
accessory organs of hearing.

The African elephant fish or Mormyrus (Fig. iii) has a small
fenestra in the lateral wall of the skull which is closed by a very
thin osseous membrane loosely attached to the margin of the win-
dow, except at one point. Immediately beneath this lies an ovoid

51

52

BRAIN AND BODY OF FISH

vesicle (Fig. iv), about the size of a small pea in its longest diameter,
which hes in a vertical plane ; this contains gas ; attached to its
base is an almond-shaped sac, containing an otolith ; leading from
this and connected to it by a short duct, is a round sac the diameter
of which is less than the long diameter of the other sac ; this also
contains an otolith. In the floor of the cranial cavity anterior to
the air-vesicle are three impressions which lodge the utricle and two
ampullse, the spherical terminations of the semicircular canals ; and
posterior to the vesicle is another impression for the ampulla of the

Fig. iv. — Air vesicle, ampullae and canals of Mormyrus — enlarged.

A.V. — Auditory vesicle, p.s.c.c, h.s.c.c. and a.s.c.c. — Poserior, horizontal and
anterior semi-circular canals. S. — Saccule, lag. — Round sac with otoliths.
U. — Utride. a. — Ampullae.

horizontal semicircular canal. These canals embrace the vesicle,
but do not communicate with it.

The almond-shaped sac must be regarded as the saccule, and
the round sac presumably represents the lagena, which is the term
applied to a specialised portion of the saccule, which is supposed to
represent the cochlea of higher vertebrates. This is the nearest
approach to the ear of an air-breathing vertebrate that is known in
fish. It is true there is no external ear, but there is a tympanic
membrane communicating chrectly with a middle ear represented
by the air- vesicle, and this is in dii'ect contact with the internal ear,
represented by the saccule and lagena and their otoliths. There is
another interesting fact that must be noted ; projecting from the

ACCESSORY ORGANS OF HEARING 53

base of the vesicle is a small blind rudimentary duct ; tliis has been
sho^^^l by a study of the development of Gymnarchus, one of the
jMormp-ida\ to be the vestigial remains of the original swim-bladder
coiuiection.

We shall find when we describe the spherical air vesicles of the
herring that the posterior vesicle is surrounded by the semicircular
canals, just as we have described in Mormyrus ; this seems to point
to the probability that the posterior vesicle is functionally similar
to the air-vesicle of the latter fish. It may be asked why has the
swim-bladder connection been lost in Gymnarchus but remains in
the herring ; the explanation is that the herring has a wide range
of movement, at times swimming rapidly from the depths to the
surface ; if there was no connection, the gas in the vesicles would
expand to such an extent as to put the mechanism out of gear
when the fish comes to the surface ; moreover, there is a special
orifice in the swim bladder of the herring which allows of a free dis-
charge of gas, directly near the vent. We know little of the habits
of the IMormyi'idse, except that they are bottom -feeding fish, and
keep approximately at the same depth, so that there is no necessity
for any contrivance to release the pressure in the air-vesicle.

The accessory organ of hearing in the herring, which we shall now
attempt to describe, has never been quite satisfactorily investigated.
In an endeavour to tlu-ow more light on this very complicated
mechanism, we have dissected a very large number of fresh specimens
and have been struck by the number of details that can be made by
simple anatomical methods. But the most important contribution
to the interpretation of its structiu*e has been made by the study of
a series of horizontal sections which were given us by Dr. Hillier,
who has made a detailed study of the bones of the cranium of this
fish. These sections are very beautiful and seem to solve many of
the difficulties that have arisen in the correct understanding of the
function of this organ.

In order to simplify our description, it is proposed to describe
this complex mechanism under tliree headings, the ear proper, the
swim-bladder, and the accessory air-vesicles by which the first two
organs are connected. The two drawings of the ear proper (Fig. i)
are made from the dissections of a fresh specimen, and show the
usual type of internal ear that is found in fish, namely, the utricle
with its three semicircular canals, and the saccule with its
otolith. There are two other facts or relations that do not
appear in the drawings : the bony cavity of the saccule has a specially
modified external wall which is known as the auditory fenestra ;
this is closed by a membrane which has the characteristics of a

54 BRAIN AND BODY OF FISH

tympanic membrane, as found in the liigher vertebrates ; this
allows the incompressible fluid in the cavity to vibrate. It must be
understood that the membranous wall of the saccule lies in the
central cavity of the bony labyrinth of the ear, known as the vesti-
bule ; in the herring an appendix of the vestibule meets the
anterior spherical air vesicle, which we shall shortly describe, and
comes in close contact with it. The long silvery looking tube that
we see in the upper part of the abdomen of a fresh herring is the
swim-bladder. Hillier gives a very concise description of this
organ as follows :

"It is a simple long sac, opening posteriorly by a fine canal
through a dense sphincter muscle, and anteriorly opening through

Fig. v.— a dissection of the lateral wall of the Herring (enlarged)
from within to show the accessory auditory apparatus.

The front wall of the anterior spherical air-vesicle has been chipped away so that
india-ink could be injected into the posterior air-vesicle through the canal
leading into the pear-shaped vesicle. The ink has also passed into the canal
leading into the swim-bladder. Below the pear-shaped vesicle is the outer
wall of the saccule closed by the auditory fenestra.

a cartilaginous arch, into the fine duct that leads into the interior
of the labyrinth. A canal from the stomach, the pneumatic duct,
enters the swim-bladder about its middle. So that the swim-
bladder is unique among fishes in having three openings into its
cavity. It is curious to note that the communication from the
stomach is actually in line with the pharynx, and that the passage
to the lab3T:'inth is through a ring of cartilage ; for in this respect
it suggests the similar condition of things in mammals, with the
Eustachian tube running from the naso -pharynx to the cavity of
the middle ear."

The accessory portion of the auditory organ commences by the
fine duct given off from the anterior end of the swim-bladder ;
this almost at once divides into two finer ducts, which pass forwards

ACCESSORY ORGANS OF HEARING

56

a o

t- ^

.3

rt ^

be

^ -5 ®

.2 1

^1

3

.23 g p

T* rt

-^ O g

"^ o o

C3 ^ ^

c o ■^

>

-d

m

0)

•'"'

<o

CJ

^3

C

0)

'3 a

a
-M to

s >

.2 o
^ =*

— ! R T3 ^ cs o ■:5

•^ .2

S :2 '5,:S -

3 .2

"S ^

>, "^ " 2
= 2 o o

.22 cs
o ;3 3
3 n 2

§..§2

r; ^ 'd «

o

.2

a '-+3

'V' '5
^ >
o •=!

<u -*^
<« o
^^
•-^ t(
-^ -2

^ a

H ^

m

.2 o

f-* '^ ^

i=l o

.2 c
2 «

S -2 ~

0)

to
§ .2
o tc

C -C

-to O
« f-l

<

O 1-3
0) Ph

O I

.2 <i

^ § -2

"^ !- O

tc o

o "d ;r2

-^ rt tc

C3 ^, •=*

S 2 to

O 3 ^

S ct CO

^ S

-t^ -»s

S " 'I

cS O '"
-^ 'to rt

>

<

56 BRAIN AND BODY OF FISH

in the bone just above the auditory fenestra of the saccule and enter
on either side a small pear-shaped cavity ; from this again a small
duct passes upwards and outwards into the posterior spherical air-
vesicle, while from its anterior end another duct leads forward
into the anterior spherical air- vesicle. Both of these membranous
extensions of the swim-bladder are enclosed in bony capsules ; in
the bony capsule of the posterior vesicle are embedded the semi-
circular canals which thus surround this vesicle just as we have seen
the semicircular canals of Mormyrus surround its air- vesicle. If the
anterior wall of the anterior air-vesicle is carefully chipped away,
the shiny coat of the anterior extension of the swim-bladder will be
seen to occupy the outer half of the bony capsule, which is divided
by a transverse membrane ; the inner half of the bony capsule
communicates by a slit-like opening with the perilymphatic space
and near the medial margin of the orifice there lies an end-organ
with a layer of hair-cells sm-mounted by an otolith ; from the base
of this organ nerve fibres pass into a ganghon connected with the
auditory nerve.

If we consider the position of the posterior air-vesicle and its
connections with the exterior it will be noted that it lies in the
pterotic bone near the surface of the cranium and that immediately
in front there is a temporal foramen occupied by a bay-like expan-
sion of a lateral-line canal, and that the inner wall of the lateral-line
canal belonging to the lateral wing of the frontal bone is absent.
The place of this wall is taken by a tense membrane, which is
attached posteriorly to the front of the bony capsule of the posterior
air-vesicle, and anteriorly to the bone which connects the anterior
capsule to the outer waU of the cranium. How are we to construct
a reasonable theory as to the physiological use of this complicated
mechanism. Looking at the position of the posterior air-vesicle,
its site in the temporal region, and its proximity to the external
surface, suggest the theory that its function is to receive vibrations
from the surrounding water, which may also be received through
vibrations received from the tense membrane which forms the inner
wall of the adjacent lateral-line sinus, as this membrane is attached
to its capsule. The analogy of its envelopment by the semicircular
canals, as is seen in the air-vesicle of Mormyrus, also supports this
view.

Vibrations received by the air in the posterior vesicle would,
according to this suggestion, be carried to the air in the outer seg-
ment of the anterior spherical air-vesicle, and so to the transverse
membrane which divides the bony capsule ; this membrane would
convey the vibrations to the perilymph in the inner division, and

ACCESSORY ORGANS OF HEARING 67

would pass out througli tlic slit-like aperture and be received by
the otolith Ij'ing on the hair-colls close to this orifice and so to the
auditory ganglion. The membrane tends to take a somewhat
spiral course in the anterior capsule, so that the relative size of the
two compartments vary ; the inner division being large at the base
and the outer division being large at the apex. If this view be
correct the anterior vesicle Mould appear to be constructed some-
what on the plan of a rudimentary cochlea.

We regard the connection with the swim-bladder as simply a
means of regulating the pressure of gases in the system of air-
vesicles. The theory of an auditory function attributed to this
mechanism is nothing new. This was the view expressed by most
of the older writers. Weber considered that the septum in the
anterior osseous capsule functioned as a tympanic membrane.
Another theory held that the mechanism is part of a reflex system
which through efferent \nsceral nerves transmits impulses to the
swim-bladder, in order to adjust it to changes in hydrostatic pressure.
Another modern and very careful anatomist does not consider the
posterior air vesicle of any importance ; he dismisses it with the
statement, " Since the posterior vesicle has no apparent relation to
the labyrinth or other structure outside the bony capsule it will not
be discussed further in this paper."

Other objections have been raised to the theory which has been
expounded above, namely, the slight variations in the posterior
vesicle found in the pilchard and sprat. But we consider these
variations do not invalidate our view and hold that the herring
presents the most perfect adaptation of the accessory auditory
system to the function of hearing to be found in the family of
Clupeoids.

CHAPTER VII
THE CENTRAL ACOUSTIC LOBE

Having described the accessory organ of the herring we can now
turn to the consideration of the brain pattern of this fish, a typical
plankton feeder, and see whether it throws any light on the central
acoustic area we have described in the surface -feeding carps. The
medulla of a herring is much concentrated and has a well-marked
median lobe, projecting from the back of the cerebellum. This
lies in the position of the facial lobe of a carp, but microscopical
examination shows that it is connected with the acoustic tubercles
and has no anatomical relation with the facial nerve. When serial
sections are traced beginning at the hinder margin of this lobe, we
find that, resting on the fifth lobe, are two wings of tissue that meet
dorsally. These consist of groups of round cells between which
nerve fibres run, to meet at the apex, while an interrupted layer of

round cells forms a cortex. As
the sections are followed for-
wards these wings become
thicker and finally have the
shape of a pear in section,
surrounded by a dorsal exten-
sion of the basal tissue, from
which it springs.

Further forward the com-
mencement of the cerebellum
is seen lying dorsal to this
central lobe, and at the lateral
margins the acoustic tubercles
commence to make their ap-
pearance ; nerve fibres are seen
passing from these transversely
towards the central lobe. The
eighth or auchtory nerve is seen
entering laterally and forming
a distinct bundle of fibres
which approaches the central

-Y(jo6e

V.

Fig. vi.
Section of Brain of Herring.

-Lobe (somatic-sensory lobe). F.N.
Facial nerve. F.L. — Facial lobe.

58

THE CENTRAL ACOUSTIC LOBE

59

PLATE 10.— Herring.

n:

^i^

'^^if^;m^

Three drawings of the acoustic lobe of Herring. Fig. I. — Commencement of lobe
resting on the anterior end of somatic-sensory lobe. Fig. II. — A little anterior
to Fig. I, the lobe is now partly surrounded by the crura cerebelli. Fig. III. —
Shows the commencement of the cerebellum with the stratum granulosum
surrounded by cells of Purkinje lying in the middle of st. moleculare. The
acoustic tubercles are prominent and fibres pass inwards to the central acoustic
lobe. The eighth nerves are also seen.

St. Mol. — Stratum IMoleculare. St. Gr. — Stratum granulosum. C. P. — Cells of
Purkinje. A. T. — Acoustic Tubercle.

60 BRAIN AND BODY OF FISH

lobe wliile rather more anteriorly the large lateral-line nerve enters
the acoustic tubercle. It is thus apparent that this lobe is part of the
acoustico-lateralis system. The small facial lobes (Fig. 17) can be
recognised by tracing the facial nerves which pass on either side into
a small area, triangular in section, which abuts the ventricle. From
these areas descending fibres pass outwards into the great longi-
tudinal gustatory tracts ; the course of these fibres and that of the
facial nerves prove that these triangular areas are the facial lobes ;
these lobes do not appear on the surface. This detailed description
is given, as the central lobe has been mistaken for a facial lobe.

Certain interesting facts concerning the lateral-line organs must
now be mentioned. The herring has no lateral-line. There are
lateral -line organs on the head, and, as we have seen, one of these
makes a deep bay in the temporal region and appears to form an
accessory part of the auditory organ. This being the case it is clear
that the large central lobe cannot be explained by an increased
functional importance of the lateral-line system. But as the herring
is a surface feeding animal and has an elaborate auditory mechanism
the most reasonable conclusion to be drawn is that this central
lobe has an acoustic function and should be termed the central
acoustic lobe.

The conclusion to be drawn from the above facts is that the
brain of the herring is characterised by large optic lobes, a large
central acoustic lobe, and a small facial lobe, not apparent by naked
eye observation. This type is also found with slight modifications
in the sprat and pilchard and as the latter has a central acoustic
lobe very similar to that of the bleak it may be assumed that it is
characteristic of plankton feeders, as far as our present knowledge
allows this generalisation.

If this pattern is compared with a typical ground feeder, such
as the loach, a near relative to the Cyprinoids, it will be at once
evident how marked is the difference. The facial lobe is very large,
with a lobulated surface, and overlaps the vagal lobes. The facial
nerves, also very large, divide at the inferior surface of the lobe
into an anterior branch which sphts into two bundles, one passing
centrally and the other laterally to the dorsal area of the lobe. The
posterior branch on either side passes backwards close to the ventricle
and enters a large posterior lobe. Large descending fibres from
the lobe pass into the middle division of the great longitudinal
secondary gustatory tracts. The central group of cells described
in the bleak and herring at the base of the cerebellum does not
appear, but a few cells forming a narrow band across from one side
to the other and become slightly more prominent at the outer margin.

THE CENTRAL ACOUSTIC LOBE

61

The acoustic tubercles are well developed. The optic lobes are
apparently large, but tiiis is due to the tecta optica being widely
separated by a large valvula, just as in the barbel. In this fish the
pursuit of food is obviously mostly by taste, touch, and smell.

What conclusions can be drawn from the study of the herring's
brain of the functions of the central acoustic lobe ? This problem

PLATE 11.— Herrmg.

— ^'^- f ' \y^^^^:^^!^''<.\' \

Base of cerebellum (high power) showing desussating fibres and round-celled tissue
of central acoustic lobe. One acoustic tubercle with lateral line nerve and
fibres connecting the lateral and central lobes. The eighth nerve is seen and
fibres to it from the central acoustic lobe.

can be attacked by the method of elimination. The positive facts
point to the conclusion that sight is an important fact, as witness
the large optic lobes, and that taste and the gustatory organs can
take but little part, as shown by the insignificance of the facial lobe.
It would be contrary to the general rule that appears in our
study of the carp family, if there was not some functional significance
in the presence of the large central lobe we have described. The
herring, in the first place, possesses a very elaborate organ of hearing

62 BRAIN AND BODY OF FISH

which communicates with the swim-bladder. It has an anterior
and posterior spherical vesicle, and we have shown that the posterior
vesicle has certain anatomical connections with a large lateral line
organ near the pterotic bone, which enables vibrations to be received
by the air in this vesicle, and these are conducted to the air in the
anterior vesicle. This vesicle is divided into two sections by a
transverse membrane which receives the vibrations and com-
municates them to the perilymph occupying the other section, from
whence they pass by a foramen into a space connected with the
saccule, and pass over a special endorgan which has a direct con-
nection with the medulla by a nerve ganghon. It seems probable
that this elaborate organ must have a special representation in the
medulla. It is highly improbable that the central lobe can be due
to any increased functional activity of the lateral line system,
because the herring is peculiar in that it possesses no lateral line.
A possible explanation of the large central area might be that it is
for the control of the swim-bladder, but this view is shown to be
untenable when we examine the functions of this organ in other
families as will be described later. The evidence so far seems
conclusive that the central area has an auditory function, and we
propose to call it the central acoustic lobe or area according to the
extent of development into a definite lobe or not.

The hind-brain of a sprat has a similar pattern to that of the
herring. The central acoustic lobe appears as a large area, pear-
shaped in section, and fibres pass outwards to the acoustic tubercles
and also a definite strand of fibres leads to the lobe as described in
the herring.

The examination of the hind brain of the pilchard provides a
sort of connecting link which binds the central acoustic lobe of a
Cyprinoid with the Clupeoid. It is hardly to be distinguished as
regards this lobe from the bleak, but the relative size is larger ; a
further point is the large size of the acoustic tubercles which coalesce
very early after their appearance with the lateral margins of the
granular area of the cerebellum. The special tract noted in connec-
tion with the central lobe in the herring and sprat is not easy to
distinguish.

It will be interesting now to review aU the above facts which
associate the surface-feeding carps with the herring family. The
general appearance of the former has given rise to certain names
which show that early observers had been struck by the similarity
of certain members of the two families as regards their form ; for
example Engraulicypris can be translated into the English language
as the anchovy carp, and Isaac Walton gives as an alternative name

THE CENTRAL ACOUSTIC LOBE

63

for the bleak, the fresh-water sprat. We have described the small
easily detached scales and the abdominal keel as being present in
these fish and that the eyes are large. It is very instructive to be
able to point out the similarity in the brain pattern in the fourth
group of carps and the Clupeoids.

We have described the large central acoustic lobe or area and
its association with a small type of facial lobe which in both families
does not appear on external observation. The optic lobes are also
well developed. The vagal lobes are also feebly developed. We

Fig. viiA.

Fig, viiB.

Fig. vHa. — The central acoustic lobe of the sprat. It has been cut rather obliquely.
This lobe occupies a large area of the cerebellum. The acoustic tubercles are
prominent and the lateral line nerve is seen externally passing into it. The
eighth nerve is also seen.

Fig. viiB. — Transverse section of cerebellum of PUchard to shew central acoustie
lobe, c.e.l. s.g. and s.m. — Strata granulosa and moleculare of cerebellum.
V lobe. — Somatic sensory lobe.

have here a striking example of the correspondence of general form
and habits of feeding with the pattern of the brain. Another point
also seems to be worth mentioning, namely, that the relative size
of the central acoustic area varies inversely with the relative size
of the facial lobes. In a later chapter we shall be able to show that
the central acoustic lobe appears to be well developed in other
families of fish in which the faculty of hearing is of considerable
importance in their life history.

If the dominant fresh-Mater fishes of the British Isles had been
the MormjTidae instead of the carps, the attitude of the scientist

64 BRAIN AND BODY OF FISH

and of the fishermen towards the question of hearing in fish would
not have been so sceptical, and we should have heard fewer dogmatic
denials of its existence. The explanation of this statement is
that this family possesses such an obvious ear and auditory apparatus
that there could be no doubt of its acoustic function. Bearing this
in mind, when the icthyologist came to examine the ear of a cj^ri-
noid, he would have at once recognised that the Weberian ossicles
were an integral part of an organ of hearing, connecting the swim-
bladder with the internal ear ; the whole subject of hearing in fishes
would not have remained in such confusion as it has until recent
times ; and we would not have had to wait for a correct solution
of the problem until the beginning of the twentieth century. But
this family has many other claims to attention from the natm-alist,
and historically we read how it was venerated by the Egyptians ;
and the Mormyrs of the Nile are said to have been frequently
represented on mural paintings and hieroglyphics. The reason
for this distinction is the quaint and unusual features of the
animal.

A prominent nose always seems to attract attention, and Mormy-
rus has been provided with a snout that even an elephant might
envy. The front of the head in most of the species is prolonged
and tends to curve downwards, while in some it is prolonged into a
regular trunk, and the lower lip in some is continued into a fleshy
appendage, no doubt of use in searching for food. In most, the
mouth is small and the teeth are small and few. The eyes are small
and may be covered over by a skin. Whether the mouth and lips
are provided with taste-buds is not stated. The shape of the trunk
varies from that of a torpedo to that of an eel ; and in this con-
nection it is interesting to note that its near relation, the Albula, has
a larval form, through which the young pass, analogous to that of the
eels. The most eel-like of the Mormyrids is Gymnarchus, which
propels itself entirely by means of its dorsal fin, and moves with
equal ease either forwards or backwards, nosing its way backwards
by using the tip of its tail as a feeler. This fish also makes a floating
nest which is jealously guarded by the male.

To add to this tale of eccentricities, the flsh is provided with an
electric organ situate in the caudal region and derived from the
muscular system. There is a strange analogy between Gymnarchus
and Gymnotus electrophorus ; both fish are specially adapted so
as to be eel-like in form, and in both the auditory organ is speciahsed ;
in both there is an electric organ caudally situated and derived
from the muscular system ; but the communication of the swim-
bladder with the ear remains throughout fife in Gymnotus, though

THE CENTRAL ACOUSTIC LOBE

65

indirectly through the i:)iieiimatic bulb, but becomes vestigial in
G^'ninarchus.

The conformation of the brain in the Mormyrids has been, and
still remains, a wonder to the naturalist, and a puzzle to the biologist
and neurologist, so that the literature on the subject is large and

ol/JuO-

Coi/i.

1<U\ <ffil\

/.

in, V.C

i'd.ujih^ v,C

'/xJSt. U)ih(> VC .

Fig. viii. — Brain of Mormyrus kannume after Burne.
V.C. — Valvula cerebelli. lob.impar. =lob. acustico-lateralis.

diffuse ; according to the Cambridge Natural History, the brain is so
large as to attain a weight which equals one-fifty-second to one-
eighty second of the total \veight of the fish, and this great increase
of size is due almost entirely to the hypertrophy of what is known
as the " valvula cerebelli."

66

BRAIN AND BODY OF FISH

We have drawn attention already to this unfortunate terminology
" valvular It is a valve only in the sense that it separates the two
halves of the tecta optica, and it does so particularly in those fish
that grub in the mud and search between stones, such as the barbel
and gudgeon among the Cyprinoids. The valvula is thus a forward
tongue starting from the anterior aspect of the base of the cerebellum.
It is this that has burgeoned forth, one might almost say, bubbled
over so as to make a thick cap that covers the whole brain just as the
cerebral hemispheres do in the human brain.

\lalvulii

Xx..

/<f1^.CU:.t^S/'''Co

Fig. ix.^ — Section of brain of Mormyrus. — Semi-diagrammatic to show
cerebellum acoustic and lateralis lobes enveloped by valvula cerebelli.

To make this description clear we reproduce a sketch (Fig. viii),
redrawn from R. H. Burne ; this has been also figured by Herrick,
who does not agree with the identification of the two ovoid masses
that form the medulla as the " lobus impar " as shown in Burne's
lettering. I have had the opportunity through the kindness of
Dr. Tate Regan, and more recently Dr. Norman, of examining three
species of Mormyrida?, and have examined them both by the naked
eye and by serial sections. After making a number of serial di'awings
and studying the literature of the microscopic structiu-e of the
medulla and cerbellum, I have become convinced that the so-called
lobus impar consists of a central acoustic area and the acustico
lateral lobe. I shall show two drawings of my sections across these
lobes, and it will be seen that the eighth nerve j)asses to the central
acoustic lobe and that more anteriorly the lateralis nerve enters the

THE CENTRAL ACOUSTIC LOBE
PLATE 12.

67

Sei lefch

is

'VW 'hchA'-e-.

31 neAjuL
nftu?

¥m^Sf3

'/,

JT

/'

'<.-£

XpVv^'ir-'J/'v-'/

Medulla oblongata of Mormjrrus and acoustic lobe seen in transverse section. The
oblong area outlined in the upper of the three drawings is shown enlarged under
higher powers in the lower drawings.

68 BRAIN AND BODY OF FISH

■granular layer of the acustico-lateralis lobe ; this section also shows
the cerebellum lying more dorsally and the stratum granulosum
joining the above-mentioned granular area. The details of the
facial lobe are not very clear in my sections but the motor nucleus
of the vagal is well marked. The section also shows the leaf -like
structure of the calvula which seems to be almost unique in the
neurological world and one is left in wonder at the artistic beauty
of the design and the inner hidden meaning of its elaborate arcliitec-
ture.

From the point of view of my argument, we have here a clear
relation of the very definite auditory organ with a central acoustic
lobe ; as regards the functional significance of the valvula one can
only throw out the suggestion that this is a central correlation organ
associated with the acoustic, lateralis and olfactory lobes, which
appear to be the most important sense organs in the hunting equip-
ment of this fish.

As regards the feeding of Mormyrus there is very little said by
the various authorities. But Mr. Michael Graham tells me that
Mormyrus kannume on Lake Victoria apparently feeds entirely
on insect larvse of 45/45 stomachs containing any food, thirty-three
contained blood-worms and there was one record oligochaete ; and
one caddis. Its habitat surprisingly included rock and stony
bottoms. It is interesting to note that Worthington's observations
in Lake Albert and Kioga and elsewhere are similar to those of
Graham.

CHAPTER VIII

THE SILENCE OF THE SEA AND THE VOICE
OF FISHES

In the preceding pages a number of references to hearing have
been made and the auditory organs of fish have been described.
It may well be asked what are the sounds to be heard in the sea,
and also in rivers and inland waters.

Although sounds are badly conducted from air to water, yet
sound is well conducted in water. During the Great War the locali-
sation of sounds produced by submarines was of outstanding im-
portance to our Navy, and a committee was set up to enquire into
the matter. I have heard it whispered that the scientists, which
included physicists and biologists, did not see eye to eye, or rather
ear to ear, in all the discussions that took place ; nevertheless, the
hydrophone was invented, and the experiments in the detection of
sounds in water were very successful and provided valuable aid
to the Admiralty in their efforts to combat the submarine menace.

It is difficult for man, not being an aquatic animal, to project
his mind into the brain of a fish, nevertheless, by means of the
physiological methods, known as " conditioned reflexes," initiated
by Pavlov and applied by Bull to experiments on hearing in fish,
the subject has been put on a scientific basis. The ordinary man.
can reacUly prove for himself how well sounds are conducted in water
by immersing himself in a bath and listening to the ticking of a
clock or an electric buzzer placed under the water.

To those who are unfamiliar with the subject, from the point of
view of a physicist, the simplest method of obtaining an outline of
the subject is to read the Christmas lectures, given at the Royal
Institution by Su- William Bragg, on the " World of Sound."

Carlyle in another connection made use of the plu"ase " the deep
sea of Nescience," and it is a fact that the depths of the sea are very
silent. In the deep sea there is very little movement of water,
near the shore there is the fiow of the tides, and where waves break
there is noise because of the churning of the water into foam.

The movements of fish are noiseless and even when sea-birds
dive and dart about under water no sound is audible, as experiments

69

70 BRAIN AND BODY OF FISH

have proved. Noise is made when for some reason or other bubbles
are formed, and when an animal dives into the water after its prey,
it leaves no air cavity to make a noise. However, when a body
moves so quickly through the water that it does leave a cavity
behind it, the water rushes into the vacuum with all the sudden-
ness of an explosion and is the cause of the noise made by moving
steamers. This effect, called cavitation, is well known to naval
engineers, as it has often a destructive effect on ships' propellers.

A fish moves with little fuss and very fast because it leaves no
whu'ls behind it. This is simply due to its lines, as a naval architect
would express it. There is a fisherman's saying that a cod's head
and mackerel tail make for speed. The writer was the happy pos-
sessor of a small yacht designed on these principles, and it is a
■constant source of surprise to his guests, how quietly she sails with
no wave or fuss with a fresh beam wind. The design of a modern
air liner which recently appeared in the press, recalled the form of a
gurnard, even to the downward slope of the prow-like head, and
showed that the body had been designed for the least air resistance.

A fish has, therefore, the right shape for moving through the
water without noise and without unnecessary effort. Sir William
Bragg concludes, " that as animals that move under water can do
so with such little noise, it is to be expected that they cannot or do
not listen for sounds." However, there is other evidence that can
be brought forward that suggests that this conclusion is open to
adverse criticism. Dr. Beatty, of the Department of Scientific
Research and Experiment under the Admiralty, also writes on the
silence of the sea. He says, " the noises that at rare intervals float
upwards from the hydrophone are of a somewhat bizarre character ;
in the vicinity of an oyster-bed, a continuous rattle is heard from the
closing sheUs ; whales produce, by some mechanism as yet unknown,
a noise like the tinl^ling of a spoon on a plate ; and there is the
phenomenon of the drumming fishes. In the surface layers there
are the splashes due to leaping fish ; there is the roar of" breakers "
and the " brabble " of the beach along the coast, and at river
mouths, and the swirl and rip of off-shore conflicting currents.

Some degree of hearing power for such sounds may be advan-
tageous. When herrings swim, that is, come to the surface, they
■discharge bubbles from their swim-bladder and these bubbles
bursting must produce some sound and it is significant that herrings
have an accessory organ of hearing. The question of drumming
must now be discussed fully, and we will deal in the first place with
the Maigre " Sciaena aquila," an inhabitant of the INIediterranean
Sea.. This fish is well known to possess a voice, the production of

THE SILENCE OF THE SEA 71

which, by a specical mechanism, is well cstahHshed. The drumming
sound is supposed to have given rise to the Greek legend of the the
song of the Sirens, which beguiled Ulysses. The voice is the voice
of a troubador, so is heard mainly in the breechng season. In this
family there is a special sound-producing muscle, which is not directly
attached to the swim-bladder, but leads to a tendon that lies above
this organ. The muscle contracts in a series of vibrations 24 per
second, and these cause the air-bladder to vibrate ; when the bladder
is removed or deflated the sound ceases, but it can be produced
again if an artificial bladder of indiarubber is introduced. The
sound is described as of a continuous booming character. All
Seiaenidse do not drum, but it has been found that the otoliths or
earstones in the saccule of the ear are larger in the drummers than
those that are dumb, so that a relation exists between voice and
hearing in this family.

Dr. Beatty also describes the grunt fishes, which include the
sea-robin (Prionotus) and the toadfish (Opsanus). " Here the
operating muscles are intrinsically connected with the air-bladder,
so that the bladder can be removed and caused to grunt by electric
stimulation of the muscles. These fishes have been a source of
considerable inconvenience to hydrographers engaged in charting
the floor of the Atlantic by the method of echo depth sounding."

Coming to home waters we find that the familiar family of Gur-
nards are kno^\Ti to utter obscure grunting sounds, when taken out
of the sea, and that they continue them at intervals as long as they
are alive. The common English name gurnard is derived from O.F.
" gornart " akin to the French " grogner," to grunt. Several popular
names of members of the family doubtless refer to the sounds pro-
duced : the piper (Trygla lyra) and the grey gurnard or crooner are
examples. To croon means, in Scotland, to make a dull croaking
sound. According to Couch, " the grey gm^nards, commonly keep
together in companies, and in fine weather assemble in large numbers
and mount to the surface over the deep water and, when thus
aloft, they move along at a slow pace, rising and sinking iia the
water for short distances, and uttering a short grunt as if in self-
gratification."

The personal experience of the A\Titer of the article on sound
production is described in the work, " Reptiles, Amphibians and
Fishes," edited by Cunningham : " the sapphirine gurnard (Trigla
hirundo) emits distinct sounds which may be described as a succession
of short grunts. The sounds are produced in the air-bladder, which
is divided by a transverse diaphragm perforated by a hole in the
centre." The diaphragm contains radiating and circular muscle

72 BRAIN AND BODY OF FISH

fibres and the bladder has thick strong external muscles supplied
by two large nerves from the anterior part of the spinal cord.

According to Moreau, the diaphragm is thrown into vibrations by
air being forced from one compartment to the other. In these
fishes, it has been shown that the somid is produced by intrinsic
muscles in the wall of the air-bladder, and a grunt is caused by a
single contraction of these muscles : it can be called forth by electric
stimulation even in a bladder which has been removed from the
fish. The sound was not produced when the bladder was empty,
but returned when a rubber bladder was put inside the natural one ;
this proves that a diaphragm is not essential for the production of
the grunt. The grey gurnard is a fish of considerable interest apart
from its grunting, there is the strange develoj^ment of its pectoral
fin rays into fingers, and the pearl-like enlargements of its spinal
cord, associated with this special adaptation. The maigre shows
by its large otoliths the connection of voice wdth hearing. It is
found, when the brain of the grey gurnard is examined, that it has
a highly developed central acoustic lobe very similar to that found
in the j)ilchard. This can apparently be associated with hearing,
as most fish with closed swim-bladders and a bottom -feeding habitat
do not show a central acoustic lobe, but only a rudimentary acoustic
area.

It will have been noticed, that in the two families, the Sciaenidse
and Triglidse, the production of sound is dependent on the air-
bladder, and that the vibrations originate in connection with a
gaseous medium, and are produced by the organ which is homologous
with the lung, by which the voice of terresterial vertebrates is
produced. There are other methods of sound production in aquatic
animals, as example, the stridulating organs of members of the
Siluroid family, as seen in certain Indian species. The WTiter of
the above-mentioned article, to whom we are indebted for many
of our facts, describes the production of sound by a trigger-fish
" Balistes," which is abundant round the Island of Ascension. He
observed " behind the pectoral fin an area of skin, resembling a
drum, a portion of the air-bladder lying just beneath it. When the
drumming sound was produced the pectoral fin was moved rapidly
to and fro and the membrane could be seen to vibrate. No sound
was produced when the fin was held stationary."

Hitherto, w^e have mentioned mostly dwellers in the sea that pro-
duce sounds, but we must not forget to mention that the air-bladder
functions as a sound producing organ in many fresh-water fishes.
The blind fish of tlie Mammoth Cave of Kentucky calls to its felloA\'s
by sounds, and the blind fish (Lucifuga) that lives in the subterranean

THE SILENCE OF THE SEA 73

waters of Cuba, comninnioates Avith its companions in the same way.
Doras and several other South American cat-fishes have a special
adaptation for the production of sound, Jcnown as the elastic spring
meclianism by which the wall of the swim-bladder is made to vibrate.

JMuscles pass from the skull to the springs which are attached to
the bladder and the contraction of these muscles causes the springs
and the \\alls of the bladder to vibrate and to produce sound. There
are other sounds of a very different nature produced by certain
fresh- water fishes. These are the breathing sounds or " bruits de
souffle." as described by Dufosse, and are said to occur in the eels,
the carj), several species of loach, and the European cat-fish Silurus
glanis.

According to Dufosse, quoted in the Cambridge Natural History,
'* these sounds originate in some cases from the expulsion of gas
from the air-bladder through the pneumatic duct and mouth, and
in others as in the loach Misgurnus fossilis, they are produced by
the rapid expulsion tlu'ough the anus of bubbles of air previously
taken in at the mouth.

In conclusion, we must add that it is believed that both the cod
and the haddock produce sounds through the medium of the swim-
bladder. This is certainly true of the rock-haddock as we shall see
later. The above examples of noises made by fishes clearly prove
that the silence of the sea is not complete, and also give good grounds
for attributing a sense of hearing to many fish hitherto considered
to be deaf. But there is other evidence pointing to the suscepti-
bility of fishes to noises, but the conditions in this case are artificial.
We allude to certain methods adopted by fishermen designed to drive
the fish into their nets ; the first example to be given is a method,
not well laiown, of catching herring. It has been described to the
writer by one of the most experienced and observant of the skippers
of steam-drifters fishing out of Lowestoft.

Observations on the reaction of the herring to loud noises prove
that the immediate response is to dive. The Whitby and Scar-
borough fishermen employ a purely local method of fishing called
" beating for herrings." In August the fish are found some ten
miles from the shore in thirteen fathoms of water ; late in the even-
ing the fishermen put out to sea and, keeping a close watch, they
wait till they see the surface of the water alter in colour and character
as the mass of herring gradually " swim " or rise to the surface.
They at once steam or row into the midst of this area, and shoot
their nets ; as they so do every bucket, tin can, or shovel available,
is used to make a din by beating against the sides of the vessel,
accompanied by shouts and yells.

74 BRAIN AND BODY OF FISH

It would appear that the vibrations of the screw and the dis-
turbance produced by the oars is not sufficient to frighten the fish,
but the noise produced by " beating " is so alarming that the fish
dive to the bottom and are then enmeshed in the nets, which extend
some seven fathoms down below the tliree fathom level, at which
the net commences ; the reaction of the fish to loud noises is to make
them dive ; if they did not dive they would swim safely above the
net. Nature does not visualise that the danger is below, and in
swimming from the noise they become entrapped in a danger in
the depths. Another Clupeoid, the pilchard, is readily alarmed by
noises ; and the firing of a heavy gun at a distance of twenty miles
has, according to Couch, been known to cause the fish to sink.

Another example of noise being used in fishing is in the method
of seining for grey mullet that has been practised in Chichester
harbour from time immemorial ; this fish is very difficult to catch
with a net ; when surrounded by the net, they endeavour to leap
the cork line, to obtain their freedom. The fishermen, to prevent
the fish leaping, beat the water frantically with an oar ; others wade
waist deep making the welkin ring with shouts and oaths ; here
again the fish fail to reach safety and dive to destruction.

This method of fishing for mullet is not confined to the fishermen
of our south coasts. We have been told by one of the naturalists
of the Ministry of Agriculture and Fishery, R. S. Winpenny, of
the fishing for mullet in Egypt in which similar methods are
employed

As a result of the above observations an examination of the brain
of the grey mullet was undertaken and the size and structure of the
acoustic tubercles and central acoustic area examined. Here is a
fish which has well defined reactions to noises, and it was to be
expected that this area of the brain would show some specialisation
in connection with the function of audition. It was, therefore, not
surprising to find that this fish has a very well-developed central
acoustic lobe.

CHAPTER IX
FLAT-FISHES

The size and abundance of flat-fishes and the flavour of their flesh
malces this family ver^'^ familiar and useful to man. Flat-fishes are
so-called from their strongly compressed, high and flat body. They
are unable to maintain their body in a vertical position and rest and
swim on one side of the body only. The side turned towards the
bottom is sometimes the left and sometimes the right, colourless
and termed the " blind " side : that turned towards the light is
pigmented and frec[uently has adaptive patterns. Both eyes are
on the coloured side, and this is the most marked sign of the asym-
metry of this family. In the larval stage these fish appear to be
symmetrical and the change in the position of the eyes has always
aroused great interest and also the expression of conflicting views.

In pursuit of our study of the relation of hunting and feeding
liabits to the form and pattern of the brain, it would appear probable
that this asjanmetry would seriously interfere with the study ;
however, the only parts of the brain that show any distortion are
the olfactory lobes and, of course, the optic nerves. The optic
lobes, the cerebellum, and the medulla oblongata are practically
symmetrical. It will be of interest to enquire how this has happened,
and to study the history of the migrating eye. According to Norman
in his recently published monograph on Heterosomata, " in the early
stages of the development of the larva, before the bones of the skull
have become ossified, there is a supraorbital bar, as it is called,
which lies in the way of the migrating eye ; this disappears or is
absorbed and the eye passes through the gap, until it reaches the
supraorbital bar of the other side. This bar of the ocular side
becomes twisted over towards that side of the head by the movement
of the two eyes into their final position. The supraorbital of the
ocular side is finally absorbed, and as soon as the shifting of the
eyes has been completed the frontal bones appear."

There is a genus of flat-fishes known as Psettodes which has
retained a more symmetrical form and in which the eyes are as
often found on the right side as on the left side. It is said to swim,
not infrequently, in a vertical position, and there are good reasons

75

76 BRAIN AND BODY OF FISH

for regarding this fish as the least speciaUsed member of the flat-
fishes (Heterosomata), and to have Percoid affinities. The usual
explanation of the origin of flat-fish is that at some time in the distant
past their ancestors acquired the habit of resting on their side on
the bottom, and even sleeping in that position. It has been observed
that Wrasses still have that habit. The view is held that some
particularly lazy or drowsy fish adopted this attitude as a pleasant
way of passing the time, and incidentally discovered that by lying
on the bottom it was less molested by enemies. But the chief
effect that resulted, was that either the sand was apt to get into the
bottom eye, or that a permanent squint developed.

However, the process of adaptation caused the bottom eye to
migrate to a dorsal position by passing through the front of the head
in the manner we have described above ; another effect of this
recumbent position was the loss of pigment on the bottom side,
which is apparently due to the absence of light, as young flounders
kept in a tank with a glass bottom on which light was projected,
in the course of time, have been found to form pigment on the white
side. Flat-fishes have a remarkable power of making the skin
pattern on their dorsal side conform to their surroundings and as
vision was necessary to review the background against which they
lay, a further adaptation occurred. A sac-like profusion of the
membranous wall of the orbit developed, which like the cavity of
the orbit is filJecl Avith fluid. This enables the eye to be protruded
above the surface of the head to a great extent, and aUows free
movement. It is signiflcant that this protrusion is very marked in
the plaice, a fish richly endowed with pigment cells, some of a
briUiant colour, and that the optic lobes are remarkably A\eU-
developed and almost bi-lobed in shape.

Flat-fishes or Heterosomata have recently been the subject of
a valuable monograph by Norman. In it he discusses the different
views as to the origin of flat-fishes, whether they have been derived
from a single stock of either the cod type, dory type, or perch type,
or from a number of stocks. Norman regards the theory that they
have arisen from a generalised percoid stock as the most probable.
He considers Psettodes to be the least specialised member of the
order, and that the soles and Cynoglossidae may have arisen from
another part of the percoid stem. After studying this monograph,
the last volume of the Journal of the Marine Biological Association
reached my hands, and it was with some interest that in an article
by Ford on the vertebrae of Teleostean fishes, the similarity of the
backbone of Psettodes with that of the generalised percoid type was
pointed out.

FLAT-FISHES 77

Tho follo^^•i^<:J table shows the position of the fishes according
to tlie latest classitication :

Family Sub-family Species

1. Pleuronectidoi . Pleuronectince . Hippoglosstis. Halibut.

Pleuronectes limanda. Dab.
,, Platessa. Plaice.
„ Microcephalus.

Lemon sole.
,, Cynoglossus. Witch.
,, Flesus. Flounder.

2. Bothidce . . ScophtJialmincc . Rhombus maximus. Tui-bot.

,, laevis. Brill.

Lepidorhombus ivhijf-

iagonis. Megrim.

3. Soleidce . . Solea solea. Sole.

We now propose to study the brain pattern of the Heterosomata
and we shall find that the members of this family fall into groups,
which, it is interesting to note, closely correspond with the four main
groups of British flat-fish, as described by Cunningham. We shall
take these groups in the following order :

Group I. — Species with eyes on the right side. Teeth only on
the blind side. A " beard " on the lower side of head. Example,
the sole, Solea vulgaris.

Group II. — Species with the eyes on the right side. Teeth most
developed on blind side. Examples, the plaice, the lemon sole,
the witch, the dab and the flounder.

Group III. — Species with eyes on the right side. Mouth large.
Jaws similar on two sides. Example, the halibut.

Group IV. — Species with the eyes on the left side. Teeth and
jaws equal on both sides. Examples, the tiurbot, briU and megrim.

The examples mentioned are those fish which we have had the
opportunity of examining both in the fresh state and by the study
of serial sections. The asymmetry of the head, the upward aspect
of the eyes and the crossing of the optic nerves have been fully
discussed by many authorities : but, as for practical purposes, those
parts of the brain, which concern us, are nearly symmetrical, we do
not propose to make any further reference to these characteristic
features.

78

BRAIN AND BODY OF FISH

THE SOLE

In Plate 13 is an outline drawing of the brain of the sole seen from
above. The olfactory lobes are large and show some asymmetry, the
left overlapping the right ; the primitive end-brain is very large
and considerably larger than the ojDtic lobes which are definitely
small. The somatic-sensory lobes are large, and viewed from above
their outline has the appearance of two clubs with the narrower
ends tapering posteriorly and meeting in the middle line. The
cerebellum is of moderate size. Posterior to the somatic-sensory
lobes the vagal lobes are seen to be narrow and elongated. There

PLATE 13.

Sole. Lemon -dab

p.e.b. — Primitive end-brain
Acoustic tubercles.

Plaice. Witch. Dab.

0.1. — Optic lobes. c.l.m. — Cerebellum, a.t.-
s.s.l. — somatic-sensory lobes, v. — Vagal lobes.

is no sign of a facial lobe, although it does exist, but is very small
and completely hidden by other structures.

When serial sections of the brain of a sole are examined it is
found that certain conclusions as to the significance of the lobes,
as would appear to be justified by naked-eye examination are not
confirmed. The medulla has been described as very long, and having
a pair of very prominent lobes (facial) upon its dorsal surface.
These are not facial lobes. Anterior to the vagal lobes, which are
narrow and not prominent, the facial lobes appear as small rounded
areas resting on the dorsal extremity of the vagals ; more anteriorly
fibres are seen passing ventrally and outwards from the facial lobes
into the gustatory tracts in the typical manner ; these lobes are
limited in extent both laterally and in an antero -posterior direction.
The facial nerves are small and do not enter the medulla until the
tip of the cerebellum appears in section, when they pass tranversely

FLAT-FISHES

79

towcards the middle line, and turning caudally run parallel to the
ventricle before entering the facial lobe, the identity of which is
thus established. The somatic-sensory lobes completely envelop
the facials and are readily recognised by tlieir histological structure ;
they extend far in a cephalic direction and increase in size so as to
form two prominent bulges which appear in the naked eye view.

PLATE 14.— Two Sections of Brain of Sole.

n amb

VII. — Facial nerve, c.l.m. — Cerebellum, c.a.l. — Central acoustic lobe, s.s.l. —
Somatic-sensory lobe. f.l. — Facial lobe, n.amb. — Nucleus ambiguus. g.l.s.g.t.
— Great longitudinal secondary gustatory tracts, d.f.f.l. — Descending fibres
of facial lobe.

The acoustic tubercles are large. At the base of the cerebellum and
at first separate from it is a broad bandof tissue roofing the ventricle
on the inferior margin of which are to be seen groups of round cells
Mith tranversely rurming fibres dividing them ; this is similar to
the areas which have been described in the cyprinoid brain and
which we have associated with an auditory function. This central
acoustic lobe has been seen in the herring and mormjTus, both of
which have an elaborate auditory organ ; it is well-marked in surface
feeding cyprinoids, but is rudimentary in bottom feeding species.
How are we to account for its presence in such a typical bottom-
feeder as the sole which, moreover, has no sound producing organ.

80 BRAIN AND BODY OF FISH

We must now mention the existence of a special organ, the well-
known papillary area on the lower or left side of the head, an area
covered with a number of villi. Bateson has stated that " contrary
to expectation these villi do not bear sense organs of the nature of
taste-buds," and my own observations confirm this conclusion,
although it has been denied by Cunningham ; but it accords with
the evidence given above, that the facial lobes are minute. The
presence of these villi explain the large size of the somatic-sensory
lobes which receive the tactile sensations from these sensory
filaments.

" When searching for food the presence of which it recognises by
smell, the sole glides gently about over the sand, tapping with the
lower side of the head, in order to bring this sensory area in play.
The sole is one of the flat-fishes most addicted to burying themselves
in the sand or mud, leaving only the eyes exposed."

It is commonly held by fishermen in the North Sea that soles
not only cover themselves when light is directed upon them but
bury themselves during periods of intense cold. Russell states
that they take no notice of a worm dangled above them : but Bateson
observed " that they perceive objects approaching them, for they
will bury themselves if a stroke is made with a landing net," but this
might also be exjjlained by vibrations produced by the rapid motion
of the net. But soles, eels, and rocklings have a clear appreciation
of light and darkness, always being buried during the day and swim-
ming about in their tank at night.

The olfactory system is highly developed in the sole, as will be
described later, and the eyes of the sole are small. Not only are
the eyes of the sole small but their optic lobes are small in marked
contrast with the large optic lobes of the plaice. When we review
the above facts it seems possible to formulate a probable theory of
the significance of the central acoustic area in the sole. As a general
rule a central acoustic area is weU-marked when the facial lobe is
small and vice versa, and there is a large amount of evidence avail-
able in support of the view that it is associated with an auditory
function. How can we explain its presence in the sole ? We
suggest that an auditory centre would be of value to the sole, and
at least it is a reasonable assumjjtion that this centre is associated
with the perception of vibrations.

The tapping of the sand so characteristic of the sole's method of
hunting, reminds one of the thrush feeding on our lawns, tapping
and listening for the hidden worm. Sea-birds, such as certain gulls
and the sheldrake, tap for worms, in the same way, on our shores
and in estuaries.

FLAT-FISHES 81

If tliis eoiu'Iusioii is accepted it would appear that the sole feeds
by tiuicll, touch, and hearing, represented centrally in olfactory,
somatic-sensory, and central acoustic lobes, all of which are remark-
ably developed.

AVe have noted the very small facial lobes of the sole and the
slight development of the gustatory system. The olfactory organ
must now be described in detail. The striking similarity of the
olfactory organ of the sole to those of the conger and the eel has been
described by Bateson, " the plates are arranged in two rows on each
side of a central raphe upon which the two rows are folded longi-
tudinally, so as to form a lining to the olfactory tube. The raphe
in the sole is depressed so as to form a groove from which the plates
rise up."

Associated with this similarity we find that the olfactory bulbs
and lobes of the conger, eel, and sole present a very similar pattern.
" In front of the medulla of the eel, Anguilla vulgaris, the several
regions of the brain are of approximately the same size and each is
more or less clearly bilobed, the brain appears to consist of four
pairs of rounded equal-sized nodules. The anterior pair (olfactory
bulbs) are slightly pointed in front, and give off two large olfactory
nerves. The thalamencephalon is remarkably long for a teleostean
forming a narrow neck between the cerebrum and the optic lobes, a
descrijition which applies almost equally to the condition found in the
sole." The diet of the sole must now be referred to and its relation
to its brain pattern considered. The sole is nocturnal in its habits and
is caught more frequently at night than by day : and in the daytime
more frequently when the water is cloudy than when it is clea
The sole seeks its food more by sense of smell and touch than bj
sight and in the Irish Sea prefers worms (nereids) to any other form
of chet. (Jenkins, 1925.) According to Cunningham, " the sole
is a distinctly shoal-water fish. The stomachs of 36 soles were
examined and in 18 the food consisted of marine worms. Small
fragments of the shells of bivalves were found, but these seemed to
be, in most cases, to have been attached to the tubes of tube-
building worms, although small bivalves were sometimes found
entire. The throat-teeth are pointed and slender and cannot serve
for crushing shells as do those of a plaice.

" Twenty-five per cent, of the stomachs contained echinoderms,
mostly sand-stars, Crustacea were found in two per cent. In
soles caught off the west coast of Ireland it was found that worms
were most frequent, then echinoderms, then molluscs, then crusta-
ceans, and lastly fish in a few cases. The echinoderms were mostly
brittle-stars or sand-stars : among the moUuscs were small speci-

F

82 BRAIN AND BODY OF FISH

mens of the razor-shell ; the crustaceans were small specimens of
sand-hoppers and shrimps ; the fish small sand-eels."

The above details of the diet of this family may be tedious and
seem trivial, but they serve to emphasise the peculiar epicurean
diet of the sole, which among a large concourse of bottom-feeding
fish chooses to specialise in worms and despises shell-fish. It is
interesting to speculate on the causes of this discrimination and to
see if any of the facts we have noted may help to solve the problem.

In the first place, the presence of a central acoustic lobe suggests
that the ancestors of the sole were surface feeders and, the absence
of a marked gustatory system, that they were not bottom -feeders ;
while the large somatic-sensory lobes remind one of the similar
lobes of the whiting or ling and point to an inherited predacious
habit of feeding. But there is other evidence pointing to their
ancestors having been pelagic fish, namely, the presence of a rather
large swim-bladder which is entirely wanting in the full-grown sole,
but persists until it is half an inch long and until the left eye is very
near the edge of the head, as is pictured in a drawing by Cunning-
ham. The same observer notes the eggs of the sole when fertilised
are buoyant and float in water. It is distinguished from the eggs
of other fishes in British Seas by the fact that the outer layer of
the yolk is divided into separate segments and that there is an
immense number of minute oil-globules arranged in irregular
patches at the surface of the yolk. These considerations suggest
that the sole, originally a predacious fish, has at some time found
itself in a shallow habitat and adapted itself to a diet of worms, and
that its brain pattern has remained fairly constant in accordance
with its sensory equipment and that its acoustic centre has become
of advantage in its new habitat, and its habit of tapping the sand.

The conclusion that can be drawn from this study of the brain-
pattern of the sole can now be reviewed. If it is assumed that the
sensory activities of a hunting fish depend, as a rule, on three domi-
nant senses, these in the sole appear to be : I. — Olfactory, as indicated
by an elaborate olfactory organ and large olfactory lobes ; II. —
Auditory as suggested by the large central acoustic lobe ; and III. —
Tactile as is evidenced by the large somatic-sensory lobes ; so that
smell, hearing, and touch are the dominant senses ; while on the
other hand, sight and taste are poorly developed as indicated by
the small optic lobes, and the minute facial lobes, by which all
gustatory impressions are received.

As regards the central acoustic lobe, it will be remembered that
the bleak, a surface feeding cyprinoid, the pilchard, a plankton-feed-
ing clupeoid, the gurnard and the mullet, all have a central acoustic

FLAT-FISHES 83

lobe of tlio same type. It has also been noted that the surface-
feeding iishos have small facial lobes. It has been suggested by
some writers that a large central acoustic lobe may be associated
with the regulation of the gaseous pressm-e in the swim-bladder,
but inasmuch as the sole has no swim-bladder in the adult this
theory receives no support in its case. Moreover, the gurnard,
a bottom -feeding fish with a closed swim-bladder, does not need
any special control of the pressure in this organ.

Evidence has been given associating the large central acoustic
lobe in both the giu'nard and mullet with the reception of sounds,
the former with noises produced by the fish, and the latter with,
noises from external sources. The large somatic-sensory lobes,,
similar to those seen in the turbot, and the presence of a large swim-
bladder in the larval stage of the sole suggest that originally the-
sole had pelagic habits at an early evolutionary stage, and that the;
bottom habitat is a recent acquisition.

CHAPTER X
FLAT -FISHES— Continued

In this chapter the other groups of the flat-fishes (Heretosomata)
will be considered in relation to their feeding habits and their brain
pattern ; it will be found that these groups conform to the recent
classification of Norman and are supported by the work of Ford on
the vertebral column.

The second of our groups consists of certain species of the family
Pleuronectidse and are as follows :

The plaice, Pleuronectes platessa ; the flounder, P. flesus ;
the sand dab, Hippoglossoides ; the lemon sole, P. microcephalus ;
the witch, P. cynoglossus.

Plate 13, Figs. 2-5 contains outline drawings of the naked-eye
appearance of the brain of lemon sole, plaice, witch, and dab. The
characters of this group are best exemplified by the plaice. When
the drawing of the plaice is compared with that of the sole, the
most striking difference to be noted is the large size of optic lobes
each of which is bi-lobed in an antero-posterior plane. The primi-
tive end-brain is, on the other hand, much smaller. The somatic-
sensory lobes, instead of having an anterior bulge, and appearing
like two semi-colons facing each other, as in the sole, are narrow
and elongated, and separated by a deep cleft.

It is now necessary to see what is the evidence relating to the
acoustic and gustatory functions in the plaice. This can be fur-
nished by an examination of serial sections ; but before doing so
the sections of the optic lobes should be examined and compared
with similar sections of the sole.

The two large scale drawings of sections across the middle of
the optic lobes show the very great prominence of the tectum
opticum of the plaice compared with that of the sole ; these drawings
.also show the details of the lobi inferiores to which we shall refer later.

In the plaice the facial nerves are of considerable size and
pass transversely inwards before turning caudally. The facial
lobes are large, and form a slight projection into the walls of
the ventricle ; they are capped by the fifth lobes and do not
appear in a naked-eye view ; their descending fibres form a strong

84

FLxVr-F18HES 85

band and jiass ventrally and outwards into the great lateral secondary
gustatory tracts. JNloro caudally the motor nucleus of the tenth nerve
or the nucleus ambiguus appears, but the vagal lobes are of a very
moderate size. The great difference in size between the facial lobes
of the i)laice and those of the sole is no doubt to be associated with
the ])resence of taste-buds in the plaice, which are numerous, as
their description by Bateson makes clear.

PLATE 15.

V-

\/ /

•i^li.

Transverse sections across the optic lobes of (left) Sole, and (right) Plaice.

Note the greater size of the tecta optica in the Plaice and its convoluted form-
l.i. — Lobus inferior, n.r. — Nucleus rotundus. s.v. — Saccus vasculosus.

There is no central acoustic lobe or area. Bateson, in writing on
the olfactory organ of Pleuronectes, states that the plaice, flounder,
lemon sole, and dab, as well as the halibut, " have only one row of
olfactory plates arranged in a single series in a direction parallel
to the long axis of the body and not transversely to it." This is in
sharp contrast with the condition that is found in the sole.

The lemon sole has a brain of similar external appearance to
that of the plaice. The primitive end- brain is rather larger than
in the plaice and the optic lobes are large ; they do not, however,
possess the bi-lobed character of those of the plaice, nevertheless,
there is a distinct lateral bulge anteriorly on either side. The

S6 BRAIN AND BODY OF FISH

■cerebellum is moderate in size. The somatic-sensory lobes are more
prominent than in the plaice, in fact they approach the form
described in the sole. The vagal lobes are small. In the witch,
the primitive end-brain is slightly twisted, and the optic lobes are
smaller than in the plaice and lemon sole. The somatic -sensory
lobes and vagals are similar to those of the plaice, but sections show
that the former project more dorsally : the facial lobes are very
similar to those of the plaice, but are smaller and produce no bulge
into the wall of the rhomboid fossa. The optic lobes are well
developed.

The brain of the dab is like that of the witch but smaller. The
most striking feature of the brain pattern of this group is the size
of the optic lobes. These, in the plaice and lemon sole, have also
their peculiar characters. This is no doubt due to the importance
of sight in the hunting habits of these fishes. But it must be remem-
bered that this group is well known for its mastery of the art of
camouflage. It is true that the protective coloration in fish is
controlled by vision, which acts on the pigment cells, caUed chroma-
tophores, through the medium of the network of nerve fibres that
surround these ceUs. Flat-fishes have also the power of protruding
their eyes to a remarkable extent, particularly is this the case in
the fish we are now considering. It is assumed that it is necessary
for the fish to survey the ground before it can be able to simulate its
colour and pattern. Flat-fishes are peculiar in possessing an accesory
organ the recessus orbitalis, the function of which is to make the
protrusion possible.

The second feature to be discussed in Group II is the presence
of a well-marked facial lobe which is associated with the presence
of taste-buds on the Hps and, therefore, indicates an important
gustatory mechanism. At the same time it must be noticed that
there is no central acoustic area in those fishes of the group that we
liave examined by serial sections. We are now in a position to
learn what are the dominant special senses that make the hunting
equipment of the group. They are, I, sight ; II, taste or gustation ;
And lastly, smell. Audition takes no part. The marked difference
of this picture from that of the sole is striking and it is not siu'prising
that the sole is placed by systematists in a different genus. But
there are other differences that may be mentioned namely, the
character of the eggs. The eggs of the great majority of flat-fishes
are buoyant and pelagic. Those of the sole have a number of small
oil-globules at the surface of the yolk, but in aU the members of the
families, Bothidae and Pleuronectid£e, the oil-globule is either single
or absent altogether, while the plaice has an undivided yolk without

FLAT-FISHES

87

Eh

<

•S .£P

c » ■

S o

CO 'X^

0; ^

.5 ^i

c ^ I. -8
- -5 a ^

OJ

>

t/:

u^ yi c

-^ ° o

"S "2 -'^

<* o ^

.2 S J

cs o -r

S t:

>

X J3 o

^ 3 OJ

s

O

"d

C3

W)

o
3

a

!0

y

CO

o

ft

O

cS

^

d

_rt

aj

'a

C3

6

to

2

^C
"^

O

o^
b

1

g

o

aj

C

a

'■+3

<u

to

to

o

o

to

1
o

72

o

-»^

1

o

f

■"^

cj

■s

c3

s

«

g

12;

a:

o

>

o

1

S

[o

m

^

'-(3
o

O
>

to

S

Jj

a

o

(O

ci

to

to

88 BRAIN AND BODY OF FISH

an oil-globule. It should be remarked that the dentition of the
plaice is well adapted for crushing shellfish : this is in great con-
trast with the teeth of the sole.

Writing on the bottom fauna and food of fishes, Stevens describes
the feeding behaviour of the lemon sole, which depends on eyesight
to discover its prey, and forages principally by day. At Plymouth
its diet is solely marine worms and annelids, " Tubiculous annelids,
unless hunted discreetly, disappear to safety down their holes. The
lemon sole is always on the move. It comes to rest in a character-
istic attitude with the head and fore part of the body raised well off
the bottom. Remaining perfectly still in this position, it scans the
ground with its very prominent and movable eyes. Should it then
spy a worm, cautiously emerging from its burrow, it pounces upon
it with a kind of forward leap, bringing its mouth down almost
vertically upon its victim by a strong arching of the anterior part
of the body." The plaice and dab behave in a similar manner
when searching for food, but they do not raise the head quite so
high before they pounce.

The plaice feeds chiefly on bi valve shell-fish, whose shell it is
easily able to crush by means of the strong blunt teeth in its throat,
but it also eats sea-worms. In the Firth of Forth, the most common
food was a small species of bivalve called Scrobicularia, and next in
frequency was the razor shell, Solen ; cockles and clams were also
present. Of the worms all kinds were devoured, sea-mouse, lug-
worms, rag-worms and tube- worms. Sand stars were sometimes
eaten, and an occasional sMmp, but fish very rarely. The witch
favours marine worms ; they were present in 72 per cent, of the
stomachs examined ; Crustacea, 30 per cent. ; molluscs in 14 per cent,
and echinoderms in only 3 per cent. Dabs feed mostly on Crustacea.
These occurred in 48 per cent, of the stomachs. Next in abundance
were echinoderms, namely, sand-stars and brittle stars, which
amounted to 21 per cent. Molluscs were less common and worms
occurred in only 16 per cent, of the specimens ; sand-eels and small
herrings in only 5 per cent.

The members of this group are, therefore, bottom-feeders, and
their type of brain accords with their mode of hunting. The
olfactory lobes are but moderately developed, but the optic lobes
are large as also the facial lobes, the somatic-sensory lobes are less
prominent as one would expect in fish which are not predacious.

The third group of flat-fishes belong to the family known as the
Scophthalminse, according to Norman's classification, and the com-
mon British flat-fishes belonging to this family, that will now be
described, are three, two belong to the genus Scophthalmus, namely,

FLAT-FISHES 89

the turbot, Scoplitlialmus maxiimis, and the brill, S. rhombus, and
one to the genus J^epidorhonibus the Megrim or sail-fhike, Lei)idor-
hombus whifif-iagonis, hitherto known fis Pleuronectes megastoma.
This new terminology is rather harassing as it was a useful memoria
technica to associate " meg " with megrim and megastoma, the
latter word reminding one of the characteristic large mouth of this
fish. These species have tlie following characters, eyes on the left
side of the head, mouth large, and teeth and jaws equal on both
sides.

In describing the various species of fish, the brain pattern of
which has been discussed, our attention has been drawn only to
bottom-feeding fish, and those which feed on the surface. We have
not, hitherto, encountered predacious fish, although the chub almost
deserves that description.

It will not be unexpected, if it be found that fish with predacious
habits present a very different type of brain pattern. This pattern
is ^^■ell exemplified in the cod family as we shall have occasion to
describe in the following chapter, and it is also well shown in this
third group of flat-fishes. Before going into details, it may be
stated at once how different are the habits of feeding of this group
from those of the groups above described ; Group I feeds solely
on worms, and Group II feeds on worms and shell-fish, while Group
III has almost exclusively a fish diet. The characteristic pattern
of a predacious fish is the large size of the optic lobes and the somatic-
sensory lobes, the small size of the facial lobes and the absence of a
central acoustic area or lobe.

The turbot, brill and megrim are bottom-feeding fish, and do
not as a rule frequent depths greater than twenty fathoms. The
swim-bladder is found in the post -larval stages and remains quite
late in this transitional stage in the turbot, and specimens have
been described as large as an inch long, in which it is still present.
The transitional stages of flounder, plaice, and dab have been ob-
tained near the bottom, and as they have no swim-bladder, they
can only swim up from the bottom by active movements of the fiins.
But the young turbot has a neutral buoyancy and can rise or sink
in the water without any active swimming. Cumiingham has
pointed out the advantage to the young fish of this power and habit
of swimming at the surface. Even at these early stages the turbot
and brill feed on other fishes, smaller than themselves ; and in
summer time the young of all kinds of fishes are more plentiful
on the surface than near the bottom.

We will now describe the brain of the turbot, which may be
regarded as the type of Group III, and the Plate 17 will enable

90

BRAIN AND BODY OF FISH

the reader to contrast the pattern of the three members of this
group. The medulla of the turbot is remarkably broad anteriorly,
and this feature is particularly evident in the serial sections ; the
somatic-sensory lobes are large and project more prominently as

Brill

PLATE 17.

Turbot.

Halibut.

p.e.b. — Primitive end-brain, op.l. — Optic lobe, c.l.m. — Cerebellum, a.t. — Acoustic
tubercle, s.s.l. — Somatic-sensory lobe. v.l. — Vagal lobe. l.i. — Lobus inferior.
s.v. — Saccus vasculosus. p.g. — Pituitary gland. o.n. — Optic nerves,
decussating.

one passes caudaUy when they become separated from each other
and round a somewhat quadrilateral rhomboid fossa ; the lateral
margins of which bulge somewhat through the inward projection
of the small facial lobes. The motor nuclei of the vagal or the nuclei
ambigui are very well-developed. When the lining of the ventricles
is closely examined the epithelium is seen to be much thickened and
the lower angle of the ventricle is V-shaped and occupied by a fan-

FLAT-FISHES 91

shcaped area of tissue, the free edge of which is lined with cihated
e})itheliuin beneatli wliich are small pear-shaped cells, and from
these fibres pass do^\■n^^•ards to enter a 'tine nervous meshwork.

The acoustic tubercles are not prominent and there is no central
acoustic area.

The cerebelhnn is characterized by the great extent of the stratum
granulosum which leaves only a small area of stratum moleculare.
When Me compare the similar sections of the megrim, we note that
the medulla is not so broad, but this is compensated by its greater
depth. It has well-marked somatic-sensory lobes and smaU facial
lobes A\liicli do not project so much into the wall of the rhomboid
fossa ; there are small acoustic tubercles and no central acoustic
area. The cerebellum is of the same type as that of the turbot, with
a very large area occupied by the stratum granulosum. If we com-
pare this picture with that of the second group it is seen that the
somatic-sensory lobes are more developed and that the facial lobes
are much smaller wliile the vagal lobes are more prominent.

The food of the turbot, according to Cunningham, "consists almost
entirely of other fish. It is a predacious creature ; on the west
coast of Ireland the principal fish found in turbots stomachs were
sprats and sand-eels ; occasionally a dab, a sole, or a. pout." On
the trawling grounds of the south-west of England he found " in
the turbots stomachs the boar-fish known as the cuckoo, pilchards,
Mhiting, sea-bream and pout, never anything besides fish."

The diet of the megrim, as of the turbot and brill, is fish, namely,
sprats, sand-eels, whiting, gobies, etc. We find here a group of
predacious fish, and the pattern of the brain is just what would be
expected. The optic lobes are large, the sensory-somatic lobes very
prominent and the facial lobes small — a condition comparable to
that found in the pollack, with this difference, that the facial lobe
is not apparent on a superficial examination, as they are completely
concealed by the overlapping of the somatic-sensory lobes.

THE HALIBUT, Hippoglossus.

Although the halibut belongs to the family Pleuronectidte and
has its eyes on the right side of the head, it is of a very different
build from the plaice, and has a very different habitat and diet.
Just as we have seen in the carps, that fish, though belonging to the
same family, may present a great variety of brain pattern, so in
the Pleuronectidse we shall find great differences in the pattern of
the medulla oblongata, according to their methods of feeding.
This is well shown when we contrast the brains of the hahbut and
plaice.

92 BRAIN AND BODY OF FISH

The halibut grows to a very large size and has small smooth
scales. The eyes are separated by a fairly wide space. The mouth
is large with strong jaws, and pointed teeth equally developed on
both sides of the head. There is no air bladder. The upper side
is of an olive brown to black colour and the blind side is white.
The body is both longer and plumper than the majority of flat-fish.
On the other hand, the plaice has a shorter, deeper, and thinner
body, with a small mouth twisted over towards the blind side, so
that the jaws and teeth are more developed on that side, and the
teeth are adapted for crushing.

We have already described the hunting methods of the plaice ;
it is only to be expected that the form of the halibut would indicate
a different mode of life, and we find, although it sometimes lies flat
on the bottom in wait for its prey, it is a powerful swimmer and
often goes in active pursuit of other fishes. It is a predacious fish,
and is found in depths up to 200 fathoms. Its diet is piincijDally
fish, but also feeds on crabs and shellfish, and is often found to
frequent grounds similar to those occupied by cod.

The halibut is remarkable in its power to adapt its coloration
to its surroundings and for the rapidity with which the pigmentary
changes take place. Norman states that the fish does not consciously
imitate the suiToundings, the whole process partaking of the nature
of a series of reflex processes and accomplished in less than a second.

The pattern of the brain of the hahbut can be readily surmised
by a consideration of its predatory habits and its deep sea habit.
The optic lobes are large, and the somatic sensory lobes very large ;
while the facial are small and no central acoustic area can be
demonstrated. The optic lobes are particularly enlarged in
a dorsoventral direction, and appear to bulge ventraUy so as to
give it a bi-lobed shape ; this, as in the plaice, may be associated
with its facility for protective coloration.

A more difficult problem for investigation is the cerebellum ;
this is prominent and its lateral margins are extended ventrally
so as to overlap the margins of the somatic sensory lobes for some
distance posteriorly ; these lateral extensions are not only composed
of the stratum moleculare, but the stratum granulosum extends
downwards, so as to join the granular matter of the acustico-
lateralis lobes, which are not very prominent ; where these two
structures meet there is an accumulation of large ceUs, similar to
those known as the cells of Purkinje. This unusual arrangement
of stratum granulosum and acoustic tubercle may be explained
by the deep-sea habitat ; we shall have occasion to describe the
modifications that occur in the acoustic tubercles of such fish as

FLAT-FISHES 93

the hake and the scabbard fishes in association with the great
depths of the sea they inhabit ; in the hatter fish we shall note,
in addition to the large size of the acoustic tubercles, that the
granular portion extends inwards in the form of a pyramid (in
section) and the apices of this extension consist of accumulations
of large cells.

There is reason to connect this development with direction
finding in the darkness of the depths where sight can be of little
avail ; we suggest that, as the halibut descends to two hundred
fathoms, this modification of the acustico-lateral area may also be
associated with its hunting in conditions where light does not
penetrate.

CHAPTER XI
THE COD FAMILY

Before comparing the brain pattern of the different species
of Gadoids it may be well to make a few general remarks about this
family. As Cunningham has pointed out, like the flat-fish, all
the members of this family are without spiny fin-rays and generally
without spines or bony armour ; they carry no weapons of offence
or defence except their powerful Jaws and teeth. Some, like the
cod, are powerful swimmers and the form of their body makes for
speed which has given rise to the fishermen's saying " a cod's head
and a mackerel tail," in describing the best lines for a fast vessel.

The fin rays are soft and flexible and the scales are small, smooth,
and thin and easily detachable. The dorsal and ventral fins extend
a great part of the length of the body, the tail fin is separate. The
eyes are large, except in the nocturnal feeders and in many kinds
there are barbels on the lips and chin. The majority feed in the
day time, hunting their prey by sight, and being voracious and
rapacious fish have large mouths. Some, besides a fish diet, feed
on Crustacea and mollusca so that there is a great variety in dentition
but all have powerful pharyngeal teeth.

The following is a classification after Cmmingham of British
species ; but we include only those fish which have been studied in
relation to their brain pattern.

I. Species with three dorsal fins and two ventral fins :

i. The cod. — Grows to a large size ; body not much elongated ;
colour greenish-yellow with numerous small black spots,
lateral line, white, and there is a barbel on the chin,
ii. The haddock. — Smaller, but similar, has a black blotch on

the shoulder and a black lateral line. Barbel small.
ill. The ivhiting. — Sides silvery, no barbel ; body more slender
and better fitted for active motion as seeking their prey in
higher regions than do the two preceding,
iv. The coal-fish. — Differs from the preceding in having a
lower jaw longer than the upper. It has a small barbel.
Colour dark slate blue, almost black on sides ; lateral fine
white.

94

THE COD FAMILY 95

V. The pollack. — Like the preceding but with no barbel and
the lower jaw longer. Colour a dark green.

II. Species with two dorsal fins, the hinder long, the front one
short ; one ventral fin :

vi. The ling. — A barbel on the chin. Body elongated, scales

minute, fins narrow, very flexible,
vii. The Mediterranean ling. — Molva elongata, more elongate,

and more eel-like than the common ling, molva vulgaris,
viii. The burbot. — Closely resembles the ling, but the body is

broader behind, the head and the scales are larger and the

fish is much smaller.
ix. The fork-beard. — A barbel on the chin. The pelvic fin

forms a long filament extending behind the vent.

III. Species in which there is no separate first dorsal, but the
front of the single dorsal is a narrow fringe kept vibrating dming
Hfe:

X. The three-bearded rockling,
xi. The five-bearded rockling.

MERLUCCIIDAE

The hake. — This has been omitted from Cunningham's classi-
fication, as although it is in many ways like a gadoid, several features
have made it clear that it belongs to a separate genus. Its external
features are, no barbel, scales larger than the ling, fins broader and
stifEer. Teeth larger and mouth black.

When we attempt to study the relationship of feeding-habits
to the brain pattern in the cod family, we find some different
features from those we have described in the Cyprinoids. In the
latter family we w^ere able to describe three groups, each with its
characteristic brain pattern, and the third group could be further
sub-divided into two sub-groups. In the cods or gadidse we shall
find that the different species can be graded into a greater number
of groups, and that in the first of these groups further sub-divisions
can be made according to their diet, whether this is mostly of Crus-
tacea and shell-fish, or is mostly fish ; between these two extremes
we find intermediate types, for example, a diet largely of Crustacea
but also fish or a diet mostly fish, but occasionally Crustacea. When
we examine the pattern of the medulla oblongata we find a gradation
corresponding to the types which vary between the two extremes.

Some authorities have been unable to identify the facial Jobe in
the cod family. We have made a detailed study of the brain of
the whiting and have clearly established its existence ; in those

96 BRAIN AND BODY OF FISH

fish that feed on Crustacea and shell-fish it is well-developed ; but
in those fish of predatory habits these lobes are less well marked,
but on the other hand the somatic-sensory lobes, or fifth lobes, are
highly developed, and in some cases almost envelope the gustatory
lobes.

Plate 18 gives drawings of the dorsal aspect of the species of
gadoids, all common and familiar fishes in British waters, except
the forkbeards and the deep sea ling, Molva Elongata.

In the first figure a, which represents the dorsal aspect of the
brain of the haddock or " shell-fish," as the Germans call it, the
somatic-sensory lobe is of moderate size, but the facial oi gusta-
tory lobe, v.f.L, is large. The haddock is the type of a ground
feeding gadoid with a diet of Crustacea, molluscs, echinoderms and
worms. If we now compare this drawing with Fig. d, which repre-
sents the brain of the pollack, a typical predatory fish, the somatic-
sensory lobes are very prominent and spread backwards on either
side so as to embrace the minute facial lobes which are com-
jjletely surrounded by the tapering ends of the lobes which meet
posteriorly in the middle line. The pollack is very predacious and
its diet is fish and rarely anything else, when fully grown. Between
these extremes we have the cod, whiting, and ling, and the size of
the anterior (SS) somatic-sensory areas increases according to the
importance of fish in their diet, and at the same time the size of the
facial or gustatory lobes diminishes as the crustacean and shell-
fish element in their diet figures to a less extent ; so that we get
a, series, starting from the haddock, of the cod, feeding on Crustacea
mollusca and fish ; the whiting, feeding on shrimps but mostly fish,
and the ling, feeding almost entirely on fish. The hake has been
purposely left out of the series ; the reason for this omission will be
clear from the drawing ; the brain pattern is most " uncodlike."

The cerebellum of the hake is not tongue-shaped as in the
gadidse, but is sessile, and is prolonged laterally into the two
large acoustic tubercles, from which lateral eminences extend
backwards as narrow ridges which meet medially and enclose a
tiny lobe, presumably representing the gustatory lobe. We gather
from this description that the hake has no business to be included
in the gadidse, but should be placed in a separate family, the
merluciidae, as recent writers have already done. As both the life
history of this fish and its brain pattern are unique we propose to
discuss the many interesting facts that arise in a separate chapter.
But there are two other features in the gadidse which are of im-
portance, the presence or absence of barbels and the relative size
of the optic lobes.

THE COD FAMILY

97

PLATE 18.

The Cod family or -Gadidae.

Ca2 9<,e:-/|-jU7A'^;5

'PKyCiS |.Kyc-<i MOTiTLLA

Group !.—(«) Haddock, (b) Cod, (c)

Whiting, (fZ) Pollack.
Group II.— (^) Ling, (/) Burbot, (e-)

Deep-sea Ling.
Group III. — (g) Forkbeard.

(h) Greater Forkbeard.
Group IV. — (i) Rockling.

p.e.b. — Primitive end-brain,
o.l. — Optic lobes,
c.l.m. — Cerebellum,
a.t. — acoustic tubercle.
s.s. — Somatic-sensory lobe,
v.f. — Facial lobes.

98 BRAIN AND BODY OF FISH

Taking the barbels first, we note that the haddock, cod, and hng,
have barbels. But it will at once be asked why has the ling a barbel,
when it is not a bottom-feeding fish, and has a minute gustatory-
lobe ; it can be readily understood that barbels are of use to the
haddock and cod in search for food on the bottom. It has, we believe,
been shown that the barbels of the ling do not have taste -buds, and
are only tactile in function and the optic lobes give a further clue
to the problem. The optic lobes of the ling are small while the optic
lobes of the pollack and haddock are large, and associated with
this condition, it is well known that the ling is a typical deep sea
fish and, moreover, is a night feeder.

Another point can be made out by comparing the drawing of
the brain of the ling with the pollack. The primitive end-brain of
the ling is definitely larger than that of the pollack, and this is
associated with an increased olfactory function doubtless of import-
ance in hunting at night. There are several facts worth noting in
the dentition of the fish we have been considering ; it would natur-
ally be expected that those fish that feed on Crustacea and shell-
fish would be provided with crushing teeth, and this proves to be the
case. We give in Plate 21 six drawings of the upper jaws of
the fish under consideration. We do not propose to give a detailed
account of the various dentitions, but just to point out the general
characters presented in each species. The drawing of the haddock
shows the lower jaw divided and held apart so as to expose the
upper jaw ; this is seen to have a marginal area of small sessile
teeth, 3 mm. in width in the middle line and a small toothed
area on the vomer. There are similar small teeth on the edge of
the lower jaw. These teeth are adapted for crushing.

The next drawing is of the upper jaw of the cod, a much more
formidable organ. There are three rows of sharp teeth, the anterior
row of teeth being the longest, and the widest central portion is
6 mm. deep ; within this area on the vomer, is a V-shaped area with
a rough tuberculated surface. So that there is present an anterior
dentition useful for hunting fish and also a crushing area for shell-
fish, etc.

The drawing of the whiting is readily comparable with that of
the haddock, as the lower jaw has been divided and separated in the
same way. There is a single row of sharp pigmented teeth on the
edge of the upper jaw, they are widely separated, and the longest
may be 3 mm. in length. The lower jaw has similar teeth, but smaller.
There is a V-shaped area on the vomer, as in the cod, but the teeth
are small and sharp ; this dentition clearly indicates a predacious
habit.

THE COD FAMILY . 99

The ling lias a very elaborate dentition, as is well shown in the
drawing. TJiere is a broad band of small teeth, throe deep extending
fmther back than in the cod and witliin this on the vomer is a
raised horse-shoe shaped area bearing four teeth on either side,
and a single median tooth ; these are conical in form, strong, and
sharp. The dentition of the pollack is simpler ; there is a band of
small teeth in two rows, but mesially in four rows ; there is a V-
shaped area on the vomer with small sharp curved teeth in two
rows. Both the dentary formula of ling and pollack point to a well-
marked predatory habit of feeding. The dentition of the hake will
be described later in the chapter devoted to this interesting fish.

Reviewing the above descriptions it is clear that the haddock,
cod, whiting, and pollack have teeth suitable to their diet, and we
note a more or less gradual change in type as the crustacean and
shell-fish diet gives place to a purely fish diet. The ling has a unique
dentition and this is most evident in the vomerine teeth. It may be
that this dentition is of some particular value in its nocturnal
habits of hunting, but what precise use the conical teeth are to the
fish and why they have been evolved is difficult to imagine.

In the descriptive letterpress to the drawings of gadoid brains,
there is given a short summary of the chet of each fish. We propose
to amplify the facts there stated, as there has been a vast amount
of research on the stomach contents in recent years by the naturalists
of the Ministry of Agriculture and Fisheries. These reports we
shall quote and also make free use of the textbooks of Cunningham
and Travers Jenkins. Starting with the haddock, it is found that
crustaceans and molluscs are found most frequently. The actual
percentages vary, but if we pool the observations of all the observers
they are present in about equal jDroportions. Polychaete worms and
echinoderms are also found, but fishes rarely, and never more than
three to five per cent, of the contents. Herring spawn may occur,
whence the name, " spawny haddocks," apphed to the fish at certain
periods. The haddock is, therefore, with reason, called by the
Germans " the shell-fish."

The cod, gadus morrhua, is the next in the series and is found
to be not only a feeder on the bottom fauna, but is also predacious
and hunts other fish. Like the fishermen who go to sea in a drifter-
trawler, when herring are about, the pursuit of fish is the means of
livelihood, and at other times the bottom of the sea provides the
harvest. But further complications arise in the feeding habits of
the cod, as with advancing size the diet of this fish changes, it
might almost be said becomes reversed. The cod is a voracious
feeder on other fish and also has a marked appetite for Crustacea.

100 BRAIN AND BODY OF FISH

Their diet is of herring and whiting, also hermit crabs, various other
crustaceans such as Norway lobster, molluscs and Aphrodite. But
if the stomach contents of this fish are classified according to the
size of the fish, it is found that those of 15 cm. have no fish in their
stomachs, but when the length of 30-60 cm. is reached 50 to 67 per
cent, of the contents consists of fish. But when we look at the
records for Crustacea it is found that the small fish contain Crustacea
up to 100 per cent. ; but as the size reaches 60 cm. the j^ercentage
falls to 63 per cent. The small number of polychaetes and mollusca
remains constant.

The whiting, the next in our series, has a diet which might
appear to vary according to the observer ; but this is not really so,
as when one studies the sites from which the specimens were caught,
it will be found that stomach contents give very conflicting results.
Some authorities say that the food of this fish is mainly whiting and
herring, and shrimps in large quantities. Cunningham found that
in the Firth of Forth their principle food was fish and Crustacea,
65 per cent, of the former and 37 per cent, of the latter. On the
other hand, Todd, in the report of the Food of Fishes (North Sea)
gives Crustacea 67 per cent, and fishes 40 per cent., which is the
reverse of Cunningham's data. But if we analyse Todd's statistics,
we will find that on certain banks as the Leman Ground, that fish
were 91 per cent, and, if herring were about, 88 per cent.

The ling is a voracious animal of prey, and devours all varieties
of cod, whiting, mackerel, megrims, dabs and haddock. All writers
agree that it lives almost entirely on other fish and is a nocturnal
feeder.

The pollack is another voracious feeder on herring, sprats,
pilchards and mackerel. It is purely a fish eater.

It has been suggested above, that the pattern of the medulla
in the cods shows a gradual transformation as the species changes
from a purely crustacean and molluscan diet to a purely fish diet.
When we consider the facts, just described, it would seem more
accurate to describe the gadoid brain as being of three types : I. —
With well-marked facial lobes, as exemplified by the haddock ;
II. — Facial lobes small, but large somatic -sensory lobes, as in the
cod and whiting, which are both molluscan and crustacean in their
diet and also fish eaters. III. — Large somatic-sensory lobes,
embracing very small facial lobes, as in the ling and pollack ; and
as we have seen, also in the turbot ; these are predacious fish, feeding
solely on fish. In relation to its shell-fish tastes it may be mentioned
that in both cod and pouting taste-buds have been described on the
lips, barbel, pectoral fins and the body.

THE COD FAMILY

101

PLATE 19.
Drawings C'f the Brains of five Gadoids and of Merlucius.

■h vX

y

Top row.— ('?) Haddock. (6) Cod. (c) Whiting.
/. — Facial lobe diminishes in size from a to c. s.s. — Somatic-sensory increases from
a to c. Diet of Haddock. — Crustacea mollusca and worms. Diet of Cod. —
Crustacea Mollusca and fish. Diet of Whiting. — Mostly fish, also shrimps.

Bottom row. — [d) Ling, (e) Pollock. (/) Hake.
/. — Facial lobes very small in cl to e. s.s. — Somatic-sensory large and embrace th©
facials especially in e. Hake atypical. Diet of Ling. — Fish almost entirely.
Diet of Pollack. — Fish rarely anything else. Diet of Hake. — Fish only.

CHAPTER XII
THE COD FAMILY—Continued

In the previous chapter we have described the habits and brain
structure of the fishes, the brains of which are figured in cohimn I
of Plate 18, and it may have been noticed, that with these fishes
we included the ling, Molva molva or vulgaris, which stands at the
top of Column He. Its position there is due to the fact that it is
typical of a group of gadidse which have elongate bodies with small
scales, and among these we include the blue and Mediterranean
lings, and the only fresh-water cod, the burbot or eel-j)out, Lota
vulgaris. The Mediterranean ling or molva elongata is much
more eel-like than the other lings, and is caught in deep water
south-west of the British Isles. It is silvery in colour, but greyish
towards the dorsum. The jaws are long and narrow with a pro-
truding lower jaw. There are large teeth on the palate arranged
in the form of a horseshoe, as we have seen in M. vulgaris, and the
teeth of the lower jaw are sharp. The pectoral fin is prolonged into
a long filament. There is a barbel on the chin and the first three
rays of the ventral fin are elongated.

According to Hickling, the range in depth of this fish is 220-520
fathoms, and as they usually come to the surface with stomachs
everted they probably feed in the lower layers. The eyes are large,
the diameter of the orbit being approximately one inch. Another
interesting feature is the large size of the swim-bladder, which is
remarkable for the extent of its rete mirabile ; this in a small
specimen measiued four inches in length. When we compare the
pattern of the brain of molva elongata with that of the common ling
we note that the cerebellum is in both of moderate size and the optic
lobes are not large. The acoustic lobes are very much larger in M.
elongata than in the common ling, and the somatic-sensory lobes
a,re also very large and project beyond the margins of the cerebellum.
The prominence of the somatic -sensory lobes is well shown in a
lateral view, wliich depicts the cerebellum making a luckle-like curve
over these lobes. The facts revealed by this examination, namely,
the very prominent acoustico-lateral lobes, similar to those of the
hake, and the large somatic-sensory lobes confirm the suggestion

102

THE COD FAMILY

]03

PLATE 20.

A\V

ALL

C.LM
SSL.

Top row. — The brains of three deep water fish, left to right — Hake, Ling, Molva
elongata. Below. — Three drawings to scale of M. elongata. To left — Dorsal
and lateral view natural size. To right. — Twice natural size. Note the very
prominent acoustic tubercles in M. elongata. P.E.B. — ^Primitive end-brain.
O.L. — Optic lobes. C.L.M. — Cerebellum. A.T. — Acoustic tubercles. S.S.L.
— Somatic-sensory lobes. A.L.L. — Acustico-lateralis lobe. L.INF. — Lobus
inferior.

104

BRAIN AND BODY OF FISH

that the former are highly developed in order to give some functional
advantage to a predacious fish hunting in the depths which, in the
case of the deep sea lings, must be associated with almost complete
darkness.

PLATE 21.

Haddock

Pollack

Cod

Ling*.

Hake

Dentition of the Gadoids figured in Plate XX to show relation of teeth to feeding-
habits. The drawings are of the upper jaw from below, u.j. — Upper jaw.
l.j. — Lower jaw. v.t. — Vomerine teeth.

The drawing of the burbot (Plate 18, Fig./.) in column II is on
a bigger scale than the drawings of the two lings in the same column,
and the cerebellum has been removed at its base to expose the
medulla-oblongata. This fish is the only gadoid that has a fresh-
Avater habitat. The burbot or eel-pout has, as its name suggests,
some of the characteristics of an eel, as well as those of a cod. It is

THE COD FAMILY 105

more or less cylindrical in shape with a flattened head. The scales
are small and the body is covered Avitli a thick slime. Its move-
ments are sinuous, and, like an eel, it is nocturnal in its habits, and
can live a long time out of water. In addition to the barbel on the
chin, it has developed small barbels on each anterior nostril, which
recalls a similar condition found in the eel. The dentition is very
interesting on account of its similarity to that of the ling, which, as
Travers Jenkins points out, is its nearest relative. According to
Tate Regan, the mouth is wide with bands of small pointed teeth
in the jaws, and a crescentic band of similar teeth on the vomer ;
it may be added that the latter are longer than those on the jaws
and are recurved and pigmented. This vomerine dentition corres-
ponds very closely with the condition all ready described in the Ung.
To quote Tate Regan once more, " the burbot is a gluttonous
fish and eats great quantities of the eggs of other species. It is
very destructive to other fish. In the daytime it pounces on small
fish which come near its lurking place, whilst at night it goes in
active pursuit of prey." The brain of the burbot is a very interest-
ing example of brain-pattern giving a clue to the habits of a fish,
and also its relationships. The medulla oblongata recalls at once
that of the ling. The somatic-sensory lobes are large and globular
and are prolonged posteriorly by two extensions which meet medially
and enclose two small facial lobes which are deeply situated. The
optic lobes are small, but the primitive end-brain is very prominent
and the olfactory organs are large. The picture is, therefore, that
of a night -feeder which hunts mostly by smell and its feelers ; its
predatory nature is weU shown by its large somatic-sensory lobes,
as we have already noted in the common ling.

FORKBEARDS {Phycis phycis)

In the preceding pages we have considered the variations in
the pattern of the medulla oblongata of those Gadoids which fre-
quent the seas that surround the British Isles. A jom-ney to the
islands of sub-tropical seas enables the observer to examine species
of different fish, which, we shall show, tlirow light on various pro-
blems which still puzzle the investigator ; moreover, such a journey
provides further examples of the eccentric adaptations which occur
in the cod family. Frequenting the shores of Madeira is a small
member of the Gadidae known locally as " Abrotia " among the
Portuguese ; its scientific name is Phycis phycis ; there is another
fish seen also on their markets bearing the same name, but this
inhabits deep water. It is the former that is of considerable
interest, as it shows how a shore habitat may contain members of

106 BRAIN AND BODY OF FISH

the same family living on a very similar diet and yet presenting a
very different method of hunting their prey. The general aspect
of Phycis is that of a typical cod, and a further examination proves
that it has the characteristics of this family in the details of its jaws
and dentition, its fins, scales and so on.

Its dentition is very similar to that of gadus morrhua with a
well-marked V-shaped area on the vomer and this indicates a
shell-fish diet, so that it is not surprising that the German name for
this fish is " shell-fish," which, it will be remembered, is also given to
our haddock. The head is shorter and broader than that of the
haddock and suggests that of a cat ; this is due apparently to the
large space occupied by the otohths, which are exceptionally large.
There is a small barbel under the chin, and on either side is a long
finger-hke feeler which is prolonged distaUy by two filaments ;
this is merely a modified fin-ray and does not bear any taste-buds.
These also recall the cat and its whiskers. But when in the course
of removing the roof of the skull to expose the brain, the firm grasp
on the trunk caused the animal to purr, the name of cat-fish seemed
most appropriate.

On opening the abdomen to investigate the cause of this strange
noise, to my astonishment I found that the swim-bladder was not
attached to the dorsal wall of the cavity as in most cods, but that
there was a triple-expansion gas-bag loosely attached, except at
its anterior end. The three compartments of the sac were separated
by narrow constrictions ; the jJosterior compartment was the largest,
of the size of the posterior sac of a Cyprinoid of similar weight. It
had a tough external coat and a thin internal coat. The middle
division was smaller and lying in the front part its wall contained
two small retia mirabilia, vascular bodies covered with a specialised
epithehum, which enables the animal to secrete gas ; the anterior
sac was partially divided into two so as to produce two auricles on
either side, which were capped by a thick mass of muscular tissue.
At the base of each auricle there was also a small rete mirabile.

It was the pressure on this sac that presumably caused the
queer noise which was heard when I pressed the body of the fish,
and there can be little doubt that the contraction of the muscles
of the auricles of the anterior chamber are able to produce sound
by diiving gas from one sac to the other ; in fact, act as a kind of
bag-pipe. How strange it seems to find a sound producing organ in
a cod when the cod of our waters has only an air-bladder which
divides anteriorly into a pair of caecal prolongations which extend
forwards to the head and are often ciu-iously coiled. This condition
of the swim-bladder is associated with the auditory organ.

THE COD FAMILY 107

To quote the Cambridge Natural History " these caecal pro-
longations from the air-bladder have tlieir extremities apposed
to the outer surfaces of the fibrous membranes which close a pair
of vacuities in the outer walls of the peri-otic capsules, the inner
surfaces being bathed by the perilymph surrounding the auditory
organs. It remains somewhat of a mystery or an unsolved problem
why this sub-tropical species should possess a sound-producing
organ.

In our waters, sound-production is known in few species, and the
gurnards are the most familiar ; it is interesting that the gurnards
are also characterised by the presence of tactile processes which
are really specialised fin-rays. It is, however, a fact that in the
warm waters of the Atlantic there is a host of sound-producing
fishes, such as the maigre and other sciaenoids and roncadors. It
would appear that Phycis has followed the local fashion. These
remarks on the internal anatomy lead one to the description of a
remarkable brain-pattern in the species Phycis. The large size of
the anterior lobes of the brain is very noticeable as they are as large,
if not somewhat larger, than the optic lobes. These anterior lobes
are not simply due to the primitive end-brain. The primitive end-
brain is represented by two small globular median lobes which are
surrounded laterally and anteriorly by large olfactory lobes. The
cerebellum is very long and tongue-shaped, longer and narrower
than that of gadus morrhua which it otherwise resembles. The
acoustic tubercles are small. When the cerebellum is lifted to
expose the medulla the somatic-sensory lobes are seen to be of only
moderate size and the facial lobes not clearly distinguishable.

It appears, therefore, that we have here a shell-fish eating fish,
not unlike the haddock, which goes about its job of getting its dinner,
not by gustatory organs but by smell and touch, as evidenced by its
large olfactory organs and by its sensory tentacles and also by the
size of the somatic-sensory lobes ; as we have already stated above,
we have no theory as to the value to the animal of its elaborate swim-
bladder, but it should be noted that its otoliths are of noticeable
size and in this respect resemble the maigre.

When we contrast this brain with that of the haddock we notice
that the optic lobes are much larger than the end brain in our had-
dock and that the cerebellum is not so long, and that the facial lobes
are as prominent as the somatic-sensory.

The rock-haddock, Phycis blennioides, is said by Gunther to be
an occasional visitor to the British Isles. To-day it is landed in
large quantities at ]\Iilford Haven from the trawlers fishing out of
that port. This species, together with Phycis phycis and P. Mediter-

108 BRAIN AND BODY OF FISH

ranean, are characterised not only by their external features, and by
their internal structure, but also by the pattern of their brains. It
was fortunate that my acquaintance with P. phycis preceded my
knowledge of P. blennioides, as several adaptations rudimentary in
the latter are further specialised in the former. The head is larger
in P. phycis and the eyes more prominent. In both species the
pelvic fin is prolonged on either side by a long filamentous and bifid
tentacle ; this seems to be longer in P. blennioides than in P. phycis.
Couch describes this fish under the name Greater Forkbeard, or
Hake's Dame. It would be interesting to know the origin of this
quaint appellation. He says that " these tendrils from their
structure may be judged to be endowed with lively powers of sensa-
tion ; they have joints along their course and are well supplied with
nerves from what may be termed an axillary plexus, situate in the
axilla of the fin, one branch of which passes along the course of the
fin-rays and sends a branch to penetrate through it, while the other,
which anastomoses with the first branch in the axilla, is carried along
the posterior margin. Their special function is shown by their
proceeding from the spinal cord to their termination, without com-
municating with any other nerve ; " this is another example of the
careful meticulous detail of the old school of anatomists.

An interesting point that requires investigation is why the gur-
nard should have a paired series of globular swellings in that portion
of the cord which receives the nerves from their sensitive filaments,
and that the forkbeard has no similar enlargements in its spinal cord.
P. blennioides has a range in deiDth of 200-500 fathoms and has been
caught in 140 fathoms. According to Hickling it feeds on Crustacea
which almost certainly belong to a bottom-fauna, such as crabs,
Geryon and Gonoplax, and the prawns, Munidia and Nephrops ;
so that there is strong evidence of a bathypelagic habit in this
deep-sea fish.

We have compared the brain of several specimens of P. blen-
nioides landed at Milford Haven and compared them with those of
P. phycis landed at Funchal. The type in both species is the same.
The olfactory lobe and bulb are in close opposition and form a lobe
about the same size as the optic lobes. The cerebellum is long
but not so long as that of P. phycis ; the medulla oblongata is
smallish, and the general appearance is not unlike that of the ling
and burbot. The acoustic tubercles are small ; this is of some
interest as it suggests that the depth at which this fish lives has not
been the cause of any great development of this lobe. We have
seen a great enlargement in the acoustic lobes of the deep-sea ling,
and it might be thought that its bathypelagic habitat was the cause

THE COD FAINIILY 109

of this adaptation ; but tlie above observations of the forklieards
does not support this theory. The above type of brain, togetlier
with the known habits of feeding of B. blennioides, points to the
importance of smell and tactile sensation in their hunting equij)-
ment. 8ight seems to be of secondary importance and the small
facial lobe does not point to a gustatory function of their tentacles.
The diversity of form of the swim-bladders of the two forkbeards,
we have studied, is remarkable.

We have already mentioned the tripartite bladder of P. phycis.
The swim-bladder of P. blennioides is a thick-walled sac, the speci-
men that I have before me is 13 cm. in length, but the anterior
lateral prolongations end in a tendon which adds another 1.5 cm.
to the length. There are two slight constructions, suggesting a
resemblance to the bladder of P. phycis, which divide it into an
anterior portion 3 cm. in length, a middle division 4 cm. in length,
and a posterior 6 cm. long. The diameters of the three divisions are
4 cm., 3 1 cm. and 2| cm. from behind forwards ; there are retia
mirabilia in the anterior and middle sections, but the unique feature
is the row of appendages on either side about the size of a small pea,
about 25 in all on each side. These communicate with the interior,
and there are grooves crossing transversely from one to the other.
The function of this adaptation of the bladder is unknown, but must
obviously not be a sound-producing organ like that of its near
relative, P. phycis.

The study of the lings and the forkbeards suggests an explanation
of the function of the acustico-lateral areas. The deep-sea ling
differs from M. vulgaris in the size of the acoustic tubercles. The
deep-sea blemiy shows no enlargement of these lobes. The deep-
sea ling and vulgaris have large somatic-sensory lobes, but these in
the P. blennioides are not large. We may argue from this that large
acoustic tubercles are of some functional importance to a pelagic
predacious fish hunting in the depths in darkness, but are of no
value to the sedentary fish hunting by touch.

The rocklings are a very interesting group of gadoids and com-
prise three species, in which there is no separate first dorsal fin, but
the front part of the single dorsal is a narrow fringe kept vibrating
during life. There are three well-known species. The three-
bearded rockling, the four-bearded rockling and the five-bearded
rockling. Recently, a fourth species has been identified at the
Plymouth Laboratory. As their name implies, the species are
distinguished by the number of their barbels, three, four or five,
situated on the cliin and upper lip. In the five -bearded rockling
there is one barbel on the chin, two on the upper lip, and two on

110 BRAIN AND BODY OF FISH

the margin of the anterior nostril. These barbels have been des-
cribed as possessing taste-buds.

The most important and unique characteristic of this fish is the
adaptation of the dorsal fin to subserve the function of taste. One
cannot do better than quote an early writer on this organ, for instance
Couch (1866) says, " the organ often represented as the first dorsal
fin lies in a chink from which it projects when the fish is in the water.
It is formed of a membrane from the edge of which rises a tliickly
placed row of threads, the foremost of which is the stoutest and
most prominent. This organ is in continuous and rapid motion.
It is well furnished with nerves which render it acutely sensible to
impressions." (It is now known to be well provided with taste-
buds.) " This dorsal organ is supphed by a special nerve which
reaches it directly from the brain, a branch of this nerve also goes-
to the pectoral and ventral fins which are thus endowed with
particular sensation (these are also provided with taste -buds), in
addition to those of action, the last-named faculty being controlled
by branches of the intercostal nerves."

We quote the above as showing the accurate anatomical observa-
tions of the older authors. The following description of the brain of
a five-bearded rockling is from a specimen which had been specially
hardened for the purpose of cutting serial sections.

We reproduce a drawing of the naked eye appearance (see
Plate 18, Fig. iv) of the brain after the cerebellum had been
removed to expose the medulla oblongata. The primitive end-
brain, as in the burbot, is large, approaching the optic lobes in
length. The optic lobes are of moderate size compared with
those of other gadidse. The acoustic tubercles are prominent
and prolonged posteriorly into the somewhat globular somatic-
sensory lobes. These are prolonged caudally by extensions
which would be similar to those of the burbot, if they were not
displaced and separated by the globular eminences of the enlarged
facial lobes, which meet but do not fuse in the middle line. The
vagal lobes are very prominent more caudally and join in the middle
line.

The study of serial sections will show the dominance of the facial
lobe in the sensory elements of the medulla. We give a series of
sections of the medulla oblongata, which have been drawn under
low power by means of a projector apparatus (Plate 22). The first
drawing is a section of across the somatic-sensory lobes near the
base of the cerebellum, and it cuts across both the facial nerves which
have a transverse course just below the fifth lobes. These nerves
are remarkable for their size, no doubt due to the branches of the

THE COD FMIILY

111

PLATE 22.

Four transverse sections of the medulla oblongata of the Rockling. I. — The most
anterior shows the large facial nerves passing madially to enter (in II)
the facial lobes which are separated thereby into inner and outer divisions ;
in III their descending fibres are seen entering the great lateral secondary
gustatory tracts (G.L.S.G.T.). In IV the vagal lobes (X) appear and the
nuclei ambigui.

112 BRAIN AND BODY OF FISH

facial, somatic in origin, from the fins and the barbels, and dorsal
organ. As the sections are traced caudally the facial nerves turn
sharply backwards and are seen as a thick trunk entering the facial
lobes. This trunk at first lies superficial to each lobe, which seems
to be divided into a mesial and lateral portion by some of the fibres
of the nerve, the lateral portion being the larger.

As the nerve loses itself in the substance of the facial lobe this
is seen to increase greatly in size, and each lobe, appears in section as
a somewhat pear-shaped mass bulging laterally, and ventrally a
stalk, composed of descending fibres from the lobe, passes downwards
and outwards to enter the great longitudinal secondary gustatory
tracts. More posterior still the large mass of the facial lobes appears
to be again separated into two divisions, the internal portion ex-
tending further caudally than the lateral. In the more posterior
sections the motor nuclei of the vagal lobes appear, and dorsal to
this the sensory portion of this lobe is seen to pass upwards to occupy
the site previously taken by the inner division of the facial lobe.

We see here a somewhat similar condition to that found in the
gudgeon loach and barbel, a division of the facial lobe on either side,
and also of the seventh nerve after it enters the brain, corresponding
to the elaboration of the gustatory function which results from the
varied stimuli received from the different groups of taste-buds. The
course of the nerves supplying the dorsal organ have now been
traced to the medulla, and it may be assumed to a specialised area
of the facial lobe ; there can be little doubt that the vibratile mem-
brane of the dorsal fin causes a stream of water to pass along the
chink with its collection of taste-buds, and the taint of any prey or
food would be recognised by these organs.

The rockhngs are entirely nocturnal in their habits and remain
under stones or in crevices among rocks except when they dart out
to seize their prey, which they recognise by smell, but more particu-
larly by taste and to a much less extent by sight. The large primitive
end-brain points to a well-developed olfactory function and the
smajl optic lobes are in keeping with their nocturnal habits.

CHAPTER XIII

THE HAKE, THE SCABBARD FISH,
PROMETHEUS AND NESIARCHUS

Until quite recently the hake was considered to belong to the
cod family, in more teclmical language it was classified among the
gadidse. In Cunningham's standard work on " Marketable Marine
Fishes," the hake was included with the ling and the forkbeards,
among the species with two dorsal fins, the hinder long and the front
one short. But the hake has no barbel, which is jiresent in the ling.
The scales are larger than in the ling, and the fins broader and stiffer.

The term hake (Ger. hecht. pike) is now applied to several species
constituting the genus Merluccius, allied to the cod, but now often
made a separate family, Merluciidse.

M. Merluccius is the common European form. In America,
however, the term is applied to certain marine gadoid fish, having
narrow filamentous pelvic fins, placed under the tln^oat. It is
important to remember this fact when consulting American authors.
For example, Herrick notes that " in the gadoids, particularly the
tom-cod and hake, the free filiform rays of the pelvic fin function
in a similar fashion to the barblets of Ameiurus, and are hkewise
provided with end-organs of both touch and taste." Now the tom-
cod (compare the French " tucaud," whiting pout) is a small gadoid
which resembles the common cod-fish except in size, but the hake,
described by Herrick, is not the same fish as that frequenting
British waters, but is allied to the forkbeards.

In the course of a study from a comparative point of view of
the brains of gadoids, the very great dissimilarity of the brain of
the hake from the brains of other gadoids was very evident. Its
brain pattern will, therefore, be described, and then its character-
istics, habitat and diet.

The brain is rather smaller that the average brain of a gadoid,
and the most obvious difference is to be seen in the cerebellum,
which instead of being tongue-like and extending backwards so as
to overlap the medulla oblongata is only a median globular eminence,
which projects above two lateral lobes which are the very enlarged
acoustic tubercles, as is demonstrated when serial sections are
H 113

114 BRAIN AND BODY OF FISH

examined. These acoustic tubercles, or acustico-lateral lobes,
extend backwards and diminish in size so as to form two lateral
ridges, which border the rhomboid fossa and merge into the somatic-
sensory lobes. These lobes do not produce any globular swellings
as in the ordinary gadoid medulla, but they gradually approach
the middle line posteriorly, so as to enclose a triangular area within
the apex of which appears a minute lobe divided medially by a
depression ; this our serial sections prove to be a small facial lobe.

To return to the acoustic lobes, it is found that their extent is
not only great laterally but also in a dorso-ventral direction. It is
also found when serial sections are examined, that the large-celled
granular layer of the acoustic tubercles is both wide and deep, and
that anteriorly it merges directly into the stratum granulosum of the
cerebellum. There is no central acoustic area. The facial lobes are
small and are completely overlapped by the somatic-sensory lobes,
but the nucleus ambiguus, the motor root of the vagal lobe, is large
and extends some distance in a caudo-cephalic direction. It is
quite clear that this pictiu-e of the brain of the hake is quite
atypical.

According to Gunther Merluccius has some striking features,
" The neural spines of all the abdominal vertebrae are extremely
strong, dilated, and wedged into one another. The parapophyses
of the third to sixth vertebrae are slender whilst those of all the
following abdominal vertebrae are very long and broad, convex on
the upper and concave on the lower surface ; the two or tliree
anterior pairs are as it were inflated. The whole forms a strong
roof for the air-bladder and reminds us of a similar structure in
Kurtus."

If any of my readers should be interested in the remarkable
structural differences of the backbone and vertebral spines of Mer-
luccius compared with the generalised type of the gadoids, they are
referred to a recent account with beautiful illustrations in the
Journal of the Marine Biological Association, 1937 (vol. XXII, No 1),
by E. Ford.

Considering the great depths at which the hake is found, it would
seem reasonable that this bony canopy has been evolved to give
support to the swim-bladder which must be subject to great jDres-
sures. The hake is a deep-water fish, having been taken at depths
of 400 fathoms, that is to say, approximately half a mile ; it is also
oceanic, only approaching the coast occasionally. It is more abun-
dant on the south coast of England and Ireland than on the other
coasts of the British Isles. It is not caught on the east coast of
England and the neighbomdng parts of the North Sea.

THE HAKE AND THE SCABBARD FISH 115

The feeding habits of lialce, and its Hfe history, have been studied
witli great skill and perseverance by Hickling. and he has made the
subject accessible to the general reader by an excellent small mono-
graph in the Buckland series. The following are extracts from this
work, " The food of large hake consists of fishes and fantails or
cephalopods. The most important item is the blue or bastard
whiting (gatUis poutassou), then come small hake, horse-mackerel,
mackerel, fantails and herring, in that order. During winter and
spring, when most of the hake are Hving at depths of 90 to 300
fathoms or more, their food is almost entirely of blue whiting,
smaller hake and fantails. When the hake move into shallower
waters in summer and autumn they find themselves among herring,
mackerel and horse-mackerel, and feed heavily on these fish, as welt
as on smaller hake. An interesting fact about the food of hake is
that bottom -living fish are rarely eaten. All the fishes and fantails,
i.e. small cephalopods, M'liich form the main food of the hake, are
creatures that live in the middle depths of the sea, such as the blue
whiting, which can be caught near the surface in suitable nets in
places Avhere the depth is more than 1,000 fathoms, herrings and
mackerel, wliich are usually caught in drift nets, and fast-swimming
fantails whose relatives are the giant squids. Since the hake feeds
so much on mid-water creatures, it must usually swim in mid-
water to get its food. It must be, therefore, usually out of reach of
the trawl. But the hake feeds mostly at night and lies quietly on
or near the sea bottom during the day. Diminished catches of
hake at night are as notable in very deep water (300 fathoms or
more) as in shallower water ; it can scarcely, therefore, be a simple
matter of light and darkness. When the hake are feeding at night
they probably scatter about at all levels of depth in their hunt for
food. When we study the food of hake a step farther back we find
that all these fishes and the fantails feed on a group of small mid-
water fishes and sluimps, of which easily the most important are
the krill. Krill are shrimps specially adapted for swimming, and
have a definite habit of swimming upwards towards the siu-face at
night and sinking to the bottom by day."

The teeth of hake are sharp and spaced like the whiting, but they
are movable with elastic hinges which bend back. They are pig-
mented and so also is the mouth ; there is a small V-shaped area
of dentition on the vomer. Some fishermen say that the bite is
poisonous. It is not wise to speculate too boldly, but we feel there
is some ground for suggesting that the skeletal changes and the
luiusual conformation of the acoustic tubercles may be associated
with the great range of depth which is so characteristic of its habitat.

116 BRAIN AND BODY OF FISH

THE ACOUSTIC TUBERCLES IN DEEP-WATER FISHES

In the previous pages the unusual pattern of the cerebellum and
acoustic tubercles of the hake has been figured and described, and a
tentative suggestion was made that the large development of the
acoustic tubercles might be due to the great depths (300 to 400
fathoms) in which this fish has been found. An opportunity haa
since arisen of investigating the pattern of the brain of fishes which
descend to great depths, even a mile down, and we propose to des-
cribe the remarkable confirmation of the theory that was advanced
to explain the brain -pattern of the hake.

The warm waters of the Atlantic provide the Madeira fishermen
with a rich harvest in the spring of " Espada," the black scabbard-
fish (Aphanopus carbo), which is often so plentiful, that a fish
several feet long can be bought for sixpence. Incidentally, it may
be recorded that my ignorance was so great that I paid a sliilling
for a head, to the great joy of both merchant and my friends. This
fish is caught at depths of 600 to 1,600 metres, and always is brought
to the surface dead.

Lying on the slabs of the fish-market these fish appear sheathed
in a skin of black patent leather, shiny and without scales, elongated
like an eel, with the body compressed rather than round ; the dorsal
fin extends as a continuous fringe from head to tail, the eyes are
very large, as big in circumference as a half-a-crown, while a fear-
some looking head gives support to large and long jaws, both upper
and lower of which bear large and prominent teeth. A bite from
these fish is accompanied with very severe bleeding, and the blood
from the gills is said to produce inflammation, so that the after-
results are marked on the fisherman's disfigured hands.

The caudal fin is very small, and there is a narrow constriction
separating this diminutive forked tail from the body. The penalty
for its large eyes and small tail, is for its dead body to be carried
home with the tail tlu-eaded through the empty sockets, according
to the cook's technique with the whiting. Both fish are very good
eating.

This fish, and the others I shall shortly mention, belong to the
cutlass-fishes or hair-tails (Family : Trichiurida?), and curiously
enough are distantly related to the mackerel and tunny, the forked
tail being the only superficial resemblance that is left. But it is
interesting to note that on the same day, in the same market, are
to be seen large numbers of tunny and mackerel, and also horse-
mackerel and other carangida^, together with rows and rows of
Espada.

THE HAKE AND THE SCABBARD FISH

117

On removing the roof of the bony cranium, it is found that the
brain is well supported by a thick tenacious nnicus, so that it is ad-
visable to harden the brain, before endeavouring to separate it from
the enveloping mucus. The brains of all the Trichiuridse hitherto
examined, Plate 23, which include the white scabbard fish, the rabbit-
fish (Promethicthys prometheus), and Nesiarchus nasutus, conform
to the same type, and recall the type seen in the hake. The optic
lobes are relatively not large, the cerebellum small, and not tongue-
shaped, bearing on its lateral margins globular projections, which
are the acoustic tubercles, more prominent than in the hake. The
somatic -sensory lobes are well-marked, but the medulla oblongata
does not show any differentiation into lobes posterior to these
prominences.

PLATE 23.

Aphanopus Lepidopus Promethichthys Nesiarchus

carbo. caudatus. prometheus. nasutus.

PEB— Primitive end-brain. OL— Optic lobe. CLM— Cerebellum.
ALL — Acoustico-lateral lobe. SSL — Somatic-sensory lobe.

The primitive end-brain is globular, of moderate size, and is
prolonged anteriorly into smaller olfactory bulbs. Another deep
sea fish which I was able to examine was the sea-bream, Brama rail,
which also belongs to the Scombriformes, a stout fish with a deep
body which descends to over 200 fathoms. On removing its brain
it was at once evident that further confirmation of the view was
present, that excessive prominence of the acoustic tubercle was
associated with a deep water habitat.

The brain of this fish is characterised by a larger optic lobe, and a
larger cerebellum than in the scabbard fishes, while the acoustic
tubercles though prominent are not so globular. Brama is a pelagic
and predatory fish and is often to be seen in the Maderia fish market.

The white-scabbard fish is of similar shape as the black, but is
remarkable for its uniform silvery appearance ; it may grow to a
length of six feet. According to Norman this is a surface fish, but
my informant, ]Mr. Gunther Mohl, the curator of the Museum at
Funchal, told me that it is sometimes caught at 100 fathoms, and

118 BRAIN AND BODY OF FISH

Gunther describes it as a deep-water fish ; it seems probable that
this fish has a wide range of movement just as the hake. The teeth
are smaller than those of the black scabbard fish, but it has three or
four large teeth at the anterior end of the upjDer jaw. This fish is also
known in the seas around New Zealand, and it is very sensitive to
cold ; it sometimes swims ashore in its thousands on frosty nights,
apparently " in a state of temporary insanity " (Norman). This
view does not commend itself to the writer, to whom it suggests
some movement very similar to the mass immolations of lemmings
and springbok.

The coelho or rabbit fish (Promethicthys prometheus) differs
from the preceding, by having a large forked tail like a mackerel ;
it is silvery on the dorsal part of the body, while below the lateral
fine, which is very curved, the tint is greyish ; the dorsal fin has
spinous rays and is black in colour. The head is not so elongated
as in the two scabbard fishes and the jaws are shorter, but the canine
teeth of the upper jaw are curved and prominent. The depth to
which this fish is known to descend is from 100 to 300 or 400 fathoms.

It will be wise to consider in some detail the lateral line organs
which are supphed by the laterahs nerve ; this, together with the
acoustic or eighth nerve ends in the acoustic tubercles or acustico-
lateralis lobes. They are pits developed in the skin and run along a
line on the sides of the fish ; each has a sensory organ of the same type
as is present in the ampullae of the semi-circular canals of man.
The linings of these pits are more sensitive than other parts of the
body to effects jsroduced by the flow of water or any change of
pressure. They can detect pressure produced by their own move-
ments and, though a fish does not produce vibrations rapid enough
to cause sound, yet it cannot move without causing movements of
water, and when water moves the pressure changes from place to
place and time to time. Special changes occur when it is passed
by another fish.

Fish are, therefore, sensitive to both pressure and to imjDulses
from outside. They must also be sensitive to the pressure of the
column of water, which in the fishes which we are now considering,
must be very great. The effect of great pressure on physical
phenomena must be very difficult to estimate ; but we know from
the experience obtained in diving operations that talking in com-
pressed air gives the voice a high pitch and sounds squeaky.

There are many other physical facts to be considered in relation
to the waves produced by the movement of a body in water and these
can be investigated in a ripple tank. To quote an article by Sir
Wilham Bragg, " If the water in such a tank is touched circular

THE HAKE AND THE SCABBARD FISH 119

ripples spread away and the ripples fade away as the circles become
wider. If two places are touched at the same time two sets of
interlacing ripples start out. Each set goes on its own way right
through the other set, as if it were not there." A little imagination
will make one realise how important these facts are when we en-
deavour to understand the function of the organs of the lateral line.

In a discussion on the significance of the enlarged acoustico-
lateralis lobes of Aphanopus carbo there are several interesting
facts pointing to a correct interpretation. If we compare the brain
of the hake with that of a pollack, both predacious fishes, we note
the smallness of the optic lobes in the former (see drawing) ; if sight
is of so little importance to the hake in hunting, some other special
sense will probably be found to take its place.

Can the lateral line organs take the place of eyes, and of sight ?
Hickling is of opinion that the lateral line organs are direction finders,
that is to say, are able to interpret the position of a moving object,
producing waves at a distance, as these waves would affect differently
a lateral organ situate in the front of the body, from one situate pos-
teriorly. Incidentally, it may be noticed that the hake is elongated
and, therefore, there is a greater length of lateral line and a longer
base for recording waves. This may be expressed in another, more
nautical, way, by sa;)^ng the lateral line organs are able to take
cross bearings of a distant fish, from which undulations of water
replace the vibrations of light or the waves of photons. The func-
tion of these organs have recently been investigated by Dr. Sands
at Pljanouth ; he has shown that the effect of movements of water
in the lateral canals, when recorded by the electric variations in an
isolated nerve fibre prove that the lateral line organs are receptors of
vibrations.

These researches by Dr. A. Sand, published in the Proceedings
of the Royal Society, have been summarised in the Report of the
Council of the Marine Biological Association. His method of investi-
gation has shown the great sensitivity of the sense organs to move-
ment of fluid in the canal, and has established the exact proportion-
ahty between the rate of flow and the frequency of impulse discharge.
The behaviour of the sensory system with reference to displacement
of fluid in the lateral line canals proves that they record and perceive
vibrations coming from a distance ; and the way in which the im-
pulse discharges follow the frequency of low tones produced in the
neighbourhood of the fish justifies the description of the lateral-line
system as a form of auditory organ. We have already shown how
the experiments of Bragg prove that waves of low frequency are
produced by the movements of a fish, but that they are not of

120 BRAIN AND BODY OF FISH

sufficient frequency to produce sound. These waves are, therefore,
perceived by the lateral line organs and as they are distributed
along a long base the wave would be received in varying intensity
by organs situated at a distance from each other. In this way a
series of observations made by a predatory fish of the waves pro-
duced by its prey would enable its position to be plotted, in the
central area of the brain, which we have given reasons to suggest,
is the acustico-laterahs lobe.

These facts relating to the hake are of still greater importance
in the hunting equipment of the scabbard fish and of other deep sea
fish ; it is true that these fish have very large eyes but the optic
lobes are comparatively small ; however, the acoustic tubercles are
very large. The eyes are large in order to obtain the maximum
of light in the darkness of the deep, and the optic lobes are of little
use until the last act of the drama and the seizing of the prey. The
lateral line organs may, therefore, be the means of guiding the
attacking fish in pursuit of prey, and this accounts for the functional
enlargement of the acoustic tubercles. It must also be noted that
the scabbard fish is of great length, and this gives it a long base for
their observations and increases the efficiency of the lateral line
organs.

CHAPTER XIV
THE EEL. ANGUILLA VULGARIS

In this chapter we propose to consider the feeding habits of the
common eel and its brain pattern, without any reference to the
other species M'hich spend their entire hfe in the sea, such as the
conger and the Moray or Muraena ; although we shall make some
reference to the olfactory lobe and organ of the conger. There is a
good description of the brain of the common eel by Burne in the
Catalogue of the Museum of the Royal College of Surgeons, from
which we quote. Looking at the brain from the dorsal aspect we
notice that " in front of the medulla the several regions of the brain
are approximately of the same size, and as each is more or less
clearly bilobed, the brain appears to consist of four pairs of rounded
equal-sized nodules, situated one behind the other. The anterior
pair (olfactory bulbs) are slightly pointed in front, and give off twa
large olfactory nerves. The optic lobes and cerebellum are divided
down the middle by a shallow groove. The cerebellum is consider-
ably broader than long. The medulla is small and much shortened."
But if the cerebellum is removed, it will be seen that acoustic
tubercles are not prominent and that they pass obliquely into two
elongated and narrow lobes which are the somatic-sensory lobes ;
these are continued caudally into the narrow vagal lobes. There
is no sign externally of any facial lobes.

We must now refer to the same writer's description of the brain
of the conger, "it is remarkable for the large size of its olfactory
centres. It shows a decided right-handed rotation of its anterior
lobe. The olfactory bulbs are of great size. The right bulb lies
partly below the left, much as in the sole. The thalamencephalon
is remarkably long for a teleostean, forming a narrow neck between
the cerebrum and the moderately developed optic lobes," a descrip-
tion which applies almost equally well to the condition found in the
sole.

We must make a chversion for a moment to point out that not
only are the olfactory bulbs of great size, but the olfactory organs
themselves are highly specialised. These have been described by
Bateson as foUows, " the olfactory plates lying within the organ are

121

122 BRAIN AND BODY OF FISH

arranged in two rows on each side of a central raphe, upon which
the two rows are folded longitudinally so as to form a lining to the
■olfactory tube." It is interesting to note when we consider the
important part the olfactory system plays in the feechng habits of
both eels and soles, that the sole has a similar arrangement of its
olfactory plates, but the raphe in the sole is depressed so as to form
a, groove from which the plates rise up. It will be noted how
intimate the functions of smell and taste are in the eels.

In anguilla, taste-buds are present on the tongue and lips and on
the skin of the anterior tubular nostril. A section across the medulla
of the eel will show that the large facial nerve passes in a transverse
direction to end in a large facial lobe on either side, and this oval-
shaped area makes a slight bulge into the rhomboid fossa ; however,
it does not appear on the siu'face, being completely covered by the
overlying somatic-sensory lobes. Nevertheless, the facial lobes
extend a considerable distance in an antero-posterior direction.

It may be recalled that the sole has a minute facial lobe and that
taste-buds are absent ; but there is a well-marked central acoustic
lobe. In describing the complex cerebellum of the eel, we shall
have occasion to refer to its acoustic area. As an earlier study of
the cerebellum of the eel disclosed several interesting and suggestive
details in connection with its relations to the acoustico-lateralis
lobes, I have re-examined the serial sections and made a series of
drawings under low jDower by the help of a Leitz projector. Plate 24
is a section across the commencement of the acoustic tubercles, and
the cerebellum is seen overhanging the prominent fifth lobes of
the medulla. Fig. ii is a section across the most prominent
portion of the acoustic tubercles which are here both broad and
deep ; the superior extremity of the large-celled area on either side
curves medially and merges with the stratum granulosum of the
cerebellum, which does not extend across the middle of the cere-
bellum as is usual among the teleosteans, but is divided medially ;
the result is that two sigmoid areas are formed, facing each other,
and formed by the large-celled tissue of acoustic tubercles and the
stratum granulosum of the cerebellum. The stratum gi'anulosum
extends right up to the lateral margins of the cerebellum and in
the middle line there is a narrow band of stratum moleculare
separating the two sigmoid areas. In this section there can be
seen a small group of cells on either side representing the central
acoustic area of other fish ; it is necessary to note this, as it is said
by some observers that an eel can grunt, and therefore we would
expect some indication of an acoustic area. Fig. iii is a section still
further forward where the acoustic tubercles have ceased to be

PLATE 24.

n

123

i^M^^^#-..»

Four transverse sections of the cerebellum and acoustic tubercles of the eel Anguilla
vulgaris. Fig. I. — The most posterior shows the large acoustic tubercles
T.A. on either side of the stratum moleculare of the cerebellum. S.S.L. — The
Somatic-sensory lobes give off descending fibres and between these and the
ventricle V is seen the beginning of the facial lobes. G.L.S.G.T. — Great
lateral secondary gustatory tracts. Fig. II.— More anterior. The acoustic
tubercles form a sigmoid curve with the strata granulosa. Fig. III. — The next
in an anterior direction shows the large divided strata granulosa and Fig. IV
the stratum moleculare dips down between them so as to produce an inverted
pear-shaped section.

124 BRAIN AND BODY OF FISH

joined with the cerebellum. The stratum granulosum occupies the
greater part of the ventral portion of the lobe extending from one
lateral margin to the other but divided in the middle hne by an
extension of the ventricle, so that the cerebellum would be divided
into two, it if were not for a narrow area of stratum moleculare
which unites it dorsally. The acoustic tubercles are still present
in this section.

In the last figure (iv) the cerebellum has become considerably
narrower and it is no longer divided medially. The stratum
granulosum now occupies the lateral margins of the ventral portion
of the lobe, while the greater part of the lobe is formed by a central
area of stratum moleculare which expands dorsally into a hemis-
pherical prominence. This description may seem tedious but we
shall now suggest the possible functional significance of this unusual
type of cerebellum.

We are now in a position to review the relative importance of
the various sense-organs represented in the brain of the eel, and
certain conclusions seem to be clearly indicated. The olfactory
system is obviously very important, as shown by the size of the
olfactory bulbs and the complexity of the olfactory organ. The
insignificant size of the optic lobes reflects the secondary importance
of sight in the feeding habits of a nocturnal animal. It is true that
the eyes of the silver eel become enlarged, previous to the trans-
atlantic migration, but this phenomenon is probably due to the
obscurity of the depths into which it will enter. The gustatory
system is next in importance, as shown by the large size of the facial
lobes, which receive all the afferent fibres from the taste-buds of
the lips, tongue, and the marginal barblet of the nostrils. This
picture is in marked contrast to the condition found in another
nocturnal feeder, the sole, which has poorly developed facial lobes
but a highly developed central acoustic lobe ; we have just seen
how poorly developed the central acoustic area is in the eel ; this
seems to be but another example of the fact that the facial lobe
varies inversely in size with the acoustic lobe and vice versa.

The question which presents some difficulty is the significance
of the large size of the acoustic tubercles and their relations with the
cerebellum. Here there are three points to be considered, first, the
width and depth of the large-celled tissue of the acoustic tubercles,
secondly, the fusion of this tissue with the mesially divided stratum
granulosum of the cerebellum, and thirdly, the division of the cere-
bellum into halves, connected only by a dorsal band of the stratum
moleculare. The only fish that we have had an opportunity to
examine that has an equally large acoustic tubercle is the hake and

THE EEL 126

the scabbard fish ; tliose fish arc known to frequent depths of over
400 fathoms. During its migration the eel swims at great depths,
and it is difficult to say what depth it descends when it enters the
final stages of its rejH'oductive activity. The division of the cere-
bellum and the union of the tissue of the acoustic tubercles with
the divided stratum granulosum must be considered together. It
"would seem almost inevitable that the acoustic-lateralis area should
join up "with the inner ends of the stratum granulosum as this tissue
extends peripherally to the lateral margins of the cerebellum ; so
that tliis condition may be said to be the result of the divided cere-
bellinn ; further tliis division might reasonably be expected to be
of some value in direction finding, and the direction-memory that
sets the course to the Atlantic abyss and tropical trysting place may-
be helped and furthered by this adaptation.

The diet of the eel varies very much with its age and the old
eels are not only predatory but almost omnivorous. According to
Tata Regan, the diet of the yellow^ eel is chiefly worms, small fish,
crayfish, etc. Like the sole it is a nocturnal feeder. As is well
known the eel ceases to feed when about to undertake its migration
to its breeding ground. There are other interesting facts which
aft'ect the brain, in association with this phenomenon, but we must
defer describing them until we have mentioned some recent observa-
tions on the pituitary gland in fish generally.

CHAPTER XV
THE ANATOMY OF GUSTATION

G. H. Parker, of Harvard, in his monograph on smell, taste, and
alHed senses, says : " there are no separate gustatory nerves in the
vertebrates as there are olfactory nerves and optic nerves. Gusta-
tory fibres occur in several cranial nerves and it is by means of these
that the taste-buds of various regions are provided with those nervous
connections that have been described." The result of our study of
the distribution of the taste-buds in a large number of fish of differ-
ent species makes it clear that taste-buds, although as regards their
microscopic anatomy they are apparently similar, yet have functions
that vary accorchng to their central representation in the medulla
oblongata. Taste-buds supplied by the facial nerve have a true
gustatory function ; those supplied by the glossopharyngeal
nerves on the soft palate, etc., have a sentinel function, connected
with swallowing, and those supplied by the vagus have a sentinel
function designed to protect the lungs from food-stuffs, a directional
function.

Taste, like smell, is a chemical sense, and the gustatory organs
enable the animal to recognise sweet, sour, salty, and bitter sub-
stances. Fish living entii'ely in a fluid medium must necessarily
be furnished with an extensive gustatory system and so we find
that certain families not only possess taste-buds over the whole
surface of the body, but have a peculiar comb-like organ with ridges
covered with taste-buds in the mouth known as the palatal organ.

Recently the late Prof. A. E. Boycott, when examining the orifice
of the pneumatic duct which leads from the swim-bladder into the
gullet, noticed that there was a ring of taste-buds, and that these
guarded the entrance to an enlargement of the duct of an unusual
structure. This is known as the pneumatic bulb.

This observation has been confirmed and described by myself in
the tench, and the significance of the existence of taste-buds in this
situation pointed out. Recent experiments by Damant and Evans
have proved conclusively that under normal conchtions fish with
a pneumatic duct, opening into the gullet, fill their swim-bladder by
coming to the surface and swallowing air, which is pumped into the

126

THE ANATOMY OF GUSTATION 127

bladder by tlie ])uounuitic bulb. The taste-buds, it Avould appear,
act as sentinels and discriminate between the food and the swallowed
froth, which can be observed lying at the orifice of the pneumatic
duct. "When it is borne in mind that in man taste-buds are known
to be present not only on the tongue but also on the anterior pillars
of the fauces, the under surface of the palate, the pharynx, and the
epiglottis, it becomes clear that in man also these organs must at
times have a function other than true taste and can also be regarded
as sentinels, particularly those situated on the epiglottis, which is an
important part of the mechanism which prevents food " going the
wrong way."

The barbels of fish are usually well supplied with taste-buds
and they have been described on the margin of the anterior nostril ;
but perhaps the most unusual position in which they have been
found is in the site of the dorsal fin in the rockling.

It is clear from the facts above, given as to the site of taste-buds,
that these organs can perform different functions, and we can give,
as an example, the exercise of these functions as they occur in a
carp. In the process of feeding the fish first of all recognise the
presence of food by smell, which may be called a distance receptor ;
the food is then examined by the gustatory organs, the taste-buds
on the lips and barbels, and, if considered satisfactory, it is passed
on to the palatal organ which, in mud feeding fish, separates the
inorganic particles from the proteins and carbohydrates present in
the mud before passing the nutrient material into the gullet ; the
imiervation of the first stage of taste is by the facial nerve, and the
central connection in the brain is the facial lobe ; the innervation
of the second stage is by the vagal nerve and the central connection
is the vagal lobes. Again, when the carp swallows air, the air must
pass the taste-buds in the lips and after being swallowed passes with
a certain amount of mucus into the gullet and thus reaches the
orifice of the pneumatic duct where it is recognised by the taste-buds
there present ; the pneumatic bulb then sucks in the air bubbles and
drives the gas into the swim-bladder. The innervation of these
taste-buds is also by means of the vagal nerve and the central
connection is the vagal lobes.

It would appear, therefore, that the facial lobe is concerned with
purely gustatory sensations, whereas, the vagal lobes have a dis-
criminatory function allied to taste which enables the animal to
sort and sift, retain or reject, by means of the palatal organ and
also to separate air bubbles from solid matter by means of the
sentinel ring of taste-buds guarding the orifice of the pneumatic
duct. These considerations have also their counterpart in the

128 BRAIN AND BODY OF FISH

gustatory mechanism in man ; here the anterior part of the tongue
is supphed by the chorda tympani, which represents the branch of the
facial nerve passing along the inner surface of the mandible in fish
and the taste-buds have a true gustatory function.

The anatomy of fish and its nomenclature is very much obscured
by the fact that the detailed knowledge of human anatomy supphes
the terms which are used in describing simpler, but similar, structures
in a lower vertebrate, such as a fish. We propose to approach our
problem in the reverse direction and follow the simple nervous
structure of a fish upwards to the higher vertebrates. If we take
the facial lobe of a carp, a typical teleostean fish, we note that the
facial lobe hes anteriorly to the vagals, and can be looked upon as a
forward extension of the latter, although in this fish the bilobed
character of the facial is lost. It receives all the gustatory fibres
by the Vllth or facial nerve from the lips, tongue, barbels and body.
This nerve, the Vllth, or as it is called by T. H. Huxley, the portio
dura, has been described by this great anatomist, in the pike.
" The portio dura of the pike, which leaves the skull by a special
foramen in the pro-otic bone, traverses the hyomandibular bone
and then divides into two branches, one of which runs backwards
to the hyoidean arch, while the other is directed forwards and
dowTiwards and passes to the inner surface of the lower jaw, along
which it runs to the extremity of the ramus." This last branch is
represented in Man by the chorda tympani. But, as has already
been stated, the facial muscles so important and largely developed
in man are not represented at all in fish. Hence, in the latter we
might expect to find, only mandibular, and hyoidean branches of
the portio dura corresponding with the chorda tympani on the one
hand, and the stylo-hyoidean and digastric branches on the other
in man. And this is really the case, as the following pages will show.

The facial nerve in man at once raises the picture of a lop-sided
face, due to Bell's paralysis, which is caused by disease affecting this
nerve. Sir Charles Bell, the first nem-ologist, insists that this
nerve is the respiratory nerve of the face and points out its diminished
importance in fish as being due to the absence of facial muscles in
these animals. The correctness of this assumption is borne out by
the study of the facial lobe in Teleosts, which we have seen is an
anterior extension of the vagal lobe. Huxley helps us to under-
stand the change in the function of tliis nerve. He points out that
the portio dura of the seventh nerve in man perforates the petrous
bone, and after skirting the inner wall of the tympanum leaves the
skull by the stylo-mastoid foramen. Before it does so it gives off a
recurrent branch, the chorda tympani, which takes a very singular

THE ANATOMY OF GUSTATION 129

course, entering the tympanum, crossing the auditory ossicles, to
make its way out at the front wall of the tympanum, then uniting
^^ith the gustatory division of the trigeminal, or Vth nerve, and,
passing along the inner side of the ramus of the mandible with it,
until eventually it leaves it to become connected with the sub-
maxillary ganglion. The principal portion of the portia dura, on
the other hand, makes its way out by the stylo-mastoid foramen
and is distributed to the facial muscles, some comparatively insignifi-
cant branches only being furnished to the levators of the hyoid and
depressors of the lower jaw.

We can now refer to the studies of Gushing and others which
prove that the Vth nerve takes no part in the innervation of taste-
buds in man. According to the diagram after Gushing, the taste-
buds of the tongue are innervated by the IX or glossopharyngeal
nerve and fibres pass along the lingual nerve which are continuous
with the chorda tympani and end in the geniculate ganglion. It is
established, therefore, that the true courses of the nerves of the
taste-buds of the tongue are through the VII and IX nerves.

An interesting and difficult problem arises. How are we to
explain the fact that a taste-bud seems to be associated with a
particular function according to its situation, and that there is no
anatomical difference in its structure wherever it may be situated.
It would seem that taste-buds have varying functions, or rather
that the afferent impulses from a taste-bud have a different effect
according to the part of the brain, in fish a lobe, which receives the
stimulus. If we assume that the facial lobe receives impressions
recognised as of a true gustatory type and the vagal lobe receives
impressions recognised as of a warning type we explain the varying
functions of the taste-buds as due to their central nervous repre-
sentation.

The facts remain clear, however, that the facial nerve in fish is
primarily a sensory nerve carrying gustatory stimuli and has only
a small motor element. The facial in man is an important motor
nerve supphing the face muscles and also has a stylohyoid and
digastric branch ; its small sensory element appears as the chorda
t\nnpani and this is the nerve that carries true gustatory stimuli
from the anterior part of the tongue and the circumvallate papillae.
The warning and discriminatory impulses in fish are carried by the
glossopharjTigeal and vagal nerves from the branchial region and
palatal organ to the vagal and glossopharyngeal lobes and the vagal
receives also sentinel impulses from the taste-buds surrounding the
orifice of the pneumatic duct.

In man the taste-buds of the palate and pillars of the fauces are
I

130 BRAIN AND BODY OF FISH

innervated by the glossopharyngeal nerve and here food is prevented
from passing into the post-nasal space ; the sentinel taste-buds of
epiglottis and pharynx are innervated by the vagus and thereby a
mechanism is called into play that prevents food entering the glottis.
It would appear that the central connections in the brain interpret
the sensations received in accordance with the area from which
the afferent impulses arise.

In concluding these details of the facial lobe, it is necessary to
mention the large lateral nerve that conveys the gustatory impression
from the skin. The facial nerve has a ganglion called the facial
or geniculate ganghon ; into tliis the lateralis portion of the facial
enters ; it has the following branches each of which may have a
ganglion at its root a superficial opthalmic, a buccal and an external
mandibular branch.

As regards the names of the prominent lobes of the medulla it
may be well to mention that the large bilateral lobes have been
called the lobus posterior and the pneumogastric lobe, while the
central lobe or anterior have been called the mediam lobe of Baudelot
or the " lobule mediane du Bulbe." The terms now in common use
are, the vagal lobe for the former and the facial lobe for the latter.
Herrick has made a detailed examination of the skin of a small
cat-fish and has prepared a diagram reproduced in many textbooks
of the distribution of the taste-buds over the external surface and
of the cuteneous branches of the seventh nerve which supply them.
A broad summary of the above facts relating to the lobes of the
medulla can be made in Herrick's words, " The vagal lobes for mouth-
tasting and facial lobes for skin-tasting are local enlargements of
the visceral sensory brain as is illustrated in the dog-fish. All the
taste-buds in the pharynx and back part of the mouth are supplied
by the vagal and glossopharyngeal nerves — those in front of the
mouth lips and outer skin from a root of the facial nerve which
apparently corresponds with the portio intermedia of human
anatomy."

The whole matter is best summed in the words of Sir Charles
Bell, " an animal may have a particular organ developed and with
the external apparatus there is a corresponding or adjusted condition
of the appropriated nerve."

THE ANATOMY OF GUSTATION 131

APPENDIX I
THE IX NERVE

In our description of the vagal lobes and facial lobes in fish, we have,
liitherto, made little reference to the glossopharyngeal lobe and
nerve. This lobe lies just anterior to the vagal lobe and is only to be
defined by the study of serial sections. The ninth or glossopharyn-
geal nerve is, according to the Cambridge Natural History (Fishes),
perhaps the most typical of all the branchial nerves ; it has a pre-
and post -branchial branch which enclose the hyo-brancliial cleft.
It has a palatine branch which extends forwards and anastomoses
with the corresponding branch of the seventh.

The latest authoritative description of the taste-buds in the
adult human being says that " they have been identified on the
dorsal surface of the tongue except the mid-dorsal region on both
the anterior and posterior surfaces of the epiglottis on the inner
siu-face of the ar;yi}renoid process of the larynx on the soft palate
above the uvula, on the anterior pillars of the fauces, and on the
posterior wall of the pharynx. In fishes the taste-buds of the gill
region are supphed by the vagus and the glossopharyngeal. In
man certain parts of the vagus are distributed to the larynx and to
the epiglottis as welJ as to the most posterior part of the tongue and
innervate the taste-buds of these regions.

The glossophar3ntigeal supplies the posterior third of the tongue,
including the foliate and vallate papillae and adjacent part. The
anterior part of the tongue is supplied by the chorda tympani of the
facial as we have seen above. The view that each area of taste -
buds has a different function accorcUng to the central connections
of the nerve which carries the afferent impulses to the brain is
strongly supported by these anatomical details. It would appear
that, in man, there are three areas to be considered, the true taste
area supplied by the facial nerve, an area on the anterior pillars of
the fauces and uvula supplied by the glossopharyngeal nerve which
controls the act of swallowing and prevents food entering the pos-
terior nares and an area presided over by the vagus on the posterior
wall of the pharynx and epiglottis which prevents food entering the
windpipe.

CHAPTER XVI

HEARING, EQUILIBRIUM AND THE CEREBELLAR

FUNCTIONS

It has been the aim of this comparative study of the cerebellum
and acoustic tubercles, or more correctly the acoustico-laterahs
lobes, to disentangle the various functions of the nervous components
of this complex area.

If it is found that certain areas in the cerebellum and acoustic
tubercles are enlarged in a particular fish, endowed with certain
habits or mode of life, and diminished in a fish that has not these
habits or mode of life ; it seems reasonable to associate this area with
a certain function, to which the habit or habitat gives the clue.

For instance, if it is found that a surface-feeding fish has an
accessory air-breathing organ, and that this fish has an enlarged
area at the base of the cerebellum which is rudimentary or absent in
other members of the same family of fish, it is reasonable to associate
these two facts with a common factor, namely, the atmosphere, in
the one case the air is the medium for supplying oxygen and in the
other the vehicle for the perception of vibrations or sound ; in other
words respiration and audition ; and this is the case in certain
Indian fishes, as has been shown by Bhimachar, in which the
central acoustic lobe is well marked in proportion to the develop-
ment of an accessory air-breathing organ. When we examine these
facts closer we find that in jilankton-feeding animals such as the
herring, bleak, etc., there is a similar development of the central
acoustic area which in some fishes becomes a separate lobe.

But before we proceed with this discussion of the acoustic area,
we propose to describe the microscopic structure of the acoustic
tubercles of the scabbard fish, a bathysmal fish, living at a depth of
600 to 1,G00 metres ; the naked eye appearance of these tubercles
has already been described, and they are typical of all the bathysmal
fishes we have had an opportunity to examine.

If a section across the most prominent part of the large acoustic
tubercles be examined (Plate 26), it will be seen that the lateral
convexities consist of the typical small-celled tissue, so characteristic
of tliis area, in which the jDeripheral cells are found in small groups.

132

HEARING AND CEREBELLAR FUNCTIONS

133

This greatly enlarged area appears to diminish gradually in extent
towards the middle line, so as to give the appearance of a p\Tamid
(in section), the ai)ex of which nearly meets that of the oj)j)osite
side. The small-celled tissue at the apex becomes gradually
replaced by elongated pear-shaped cells, which bridge the inter-

PLATE 25.

CAA

A transverse section of the brain of the Black Scabbard fish (Aphanopus carbo)
across the commencement of the acoustic. A.T. — tubercles and the cere-
bellum. The central acoustic area C.A.A. is well seen and fibres passing from
it to the auditory nucleus. S.G. — Stratum granulosum and S.M. — Stratum
moleculare of C.L.M. — Cerebellum.

vening gap. The cerebellum in this section is completely separated
from the acoustico-lateral areas, and has its typical conformation
of strata granulosum and moleculare. But continuous and inferior
to the small-celled area we have just described is a narrow area of
transverse fibres interspersed with rows of small cells, which we
recognise as similar to the central acoustic area we have described
in the carps. The remaining portion of the lobe consists of a tissue
similar to that of the stratum moleculare with a few^ large cells

134 BRAIN AND BODY OF FISH

scattered here and there. If we now examine a section posterior
to the above (Plate 25), we find that the special tissue of the acoustic
tubercle occupies but a small part of the periphery of the lobe and
that the lobe is made up of tissues of the central-acoustic type, fibres
which pass downwards and outwards to enter a gangloinic mass of
cells before joining the fibres of the eighth nerve. At this level the
tip only of the cerebellum appears, while below the somatic sensory
lobes are seen to meet in the middle line, although anteriorly these
lobes are separated, and form a widening of the rhomboid fossa.

The importance of this special adaptation of the acoustic tubercles
must be emphasised. Here we find a typical section of a cerebellum,
separated from a highly differentated acoustic tubercle, the granular
areas of which meet medially and are joined up by a group of large
nerve cells.

This area is again completely separated posteriorly from a
central acoustic area, in which there is no intermmghng of lateralis
nerve fibres. These three areas differing so markedly in structure
must clearly be endowed with separate functions, and their relation
to function can only be assessed by comparison with other teleostean
types.

If we refer to the drawing of the brain of the herring, we see that
the central acoustic lobe forms a definite projection mesially and
that some fibres from it pass outwards to the acoustic tubercles, but
most of the fibres pass downwards and outwards to a ganglionic
mass before joining the eighth nerve ; in this family the central
acoustic lobe is the largest and most differentiated element in the
acustico-lateralis complex. If we now consider the minnow a fish,
which in the laboratory of Prof, von Friscli has revealed a nacute
perception of sounds, by means of a modified Pavlov technique of
conditioned reflexes, we find that it has a well-developed central
acoustic area, in which both cells and nerve fibres form a characteris-
tic pattern ; this we look upon as the central representation of the
auditory function.

It may be noted in passing that the gurnard, a fish with a sound-
producing mechanism, has a similar central acoustic area to that of
the minnow. Von Frisch has also shown experimentally that the
saccule in the minnow is the peripheral sense organ for the perception
of sound waves. We must, therefore, associate the similar area in
the scabbard fish with audition.

The work of certain Japanese investigators seem to confirm the
usually accepted view that the cerebellum proper is concerned with
equilibration. By a process of exclusion, therefore, we reach the con-
clusion that the enlarged lateralis area in the scabbard fish must be

HEARING AND CEREBELLAR FUNCTIONS

135

associated with the organs of the hiteral Hue which as Dr. Sands
has shown are the organs for the reception of waves of low
frequency.

Accortling to Sand, Hoagland was the first to discover the spon-
taneous activity of the lateral -line receptors and the former has

PLATE 26.

A transverse section a little anterior to that shown in the previous Plate. The cere-
bellum is larger. The acoustic tubercles are very prominent and have a tri-
angular shape in section with a curved base. Where the apices approach each
other the round celled tissue is replaced by scattered large cells. A small
area of central acoustic tissue is still seen. S.P. — Cells of Purkinje. S.S.L. —
Somatic-sensory lobes.

shown that the lateral-line system is exceedingly sensitive to low-
frequency \dbrations. " Direct mechanical stimulation of the lateral
line, as well as vibrations propagated from a distance, are effective
in exciting the reception, on account of the movements they cause
to occur in the endolymph of the canals." Let us consider what
theoretical advantage to a deep-sea fish there would be in the in-
creased development of the acoustic tubercles.

136 BRAIN AND BODY OF FISH

It must be remembered that the lateral Line of the black scabbard
fish may extend for from four to five feet or more and, therefore, a
vibration reaching the fish when in an obhque position as regards
the object, from wliich the chsturbance proceeds, would excite the
anterior sense organs with a vibration of greater amplitude than
that wliich would reach the organs situated at the posterior extremity.
This, it is clear, would enable the lateral line to be used as a direction
finder. When the fish becomes directly head on to the object pro-
ducing the waves, the lateral organs of either side would receive an
equal stimulation. The fish would now go all out in the direction
of equal stimulation and on sighting the prey, when almost on top
of it, would open its capacious mouth, and the final stage of the
tragedy would end in the capture and swallowing of the victim.
It is difficult to conceive any other system of direction fincUng than
a series of observation points situated along a base of several feet,
and the proved capabihty of the lateral fine organs to receive such
vibrations, as would be made by a moving object, makes this
explanation seem most probable.

When we add to these considerations the facts of the great
enlargement of the acoustic tubercles in those bathysmal fishes,
which can find no use for sight in hunting for their daily bread, and
also recall the fact that the eyes of these fish, though large, are
associated with quite smaU optic lobes, there seems no doubt
that this explanation of their methods of hunting is liighly
probable.

We have mentioned the more recent work on the function of
the lateral-line organs, but have not described the work of G. H.
Parker, who cut the nerves of the lateral line and also the fifth and
seventh nerves supplying the similar organs on the head of fish ; he
then compared the behaviour of fishes so treated with normal
specimens. The result of these experiments was to suggest that the
chief use of the lateral-line organs is to perceive the movements of
other fish in the water.

Cumiingham, in commenting on Parker's work, says, " we do not
know at present to what distance vibrations due to such movements
would extend, but it is evident that if a fish is affected by a body
falhng in the water, it must also perceive the movement of another
fish or animal in the water. The above account of the various
methods of hunting that are employed by fishes has proved that some
fish hunt by sight, some by smell, and some by taste, but experiments
have so far not been made to ascertain ho w^ far a fish A\ithout sight
or hearing would be able to preceive an enemy in its neighbourhood
or detect the exact position of the fish it preys upon.

HEARING AND CEREBELLAR FUNCTIONS 137

In conclusion, we must refer to the observations in a previous
chapter, on tlie acoustico-laterahs lobes of the eel. It will be remem-
bered that these lobes were very developed and had an unusual
relation to the stratum granulosum of the cerebellum ; we are
inclined to think that this adaptation may be the result of its deep-
sea habitat in the later stages of its life journey.

CHAPTER XVII
VALVULA CEREBELLI

We purpose in this chapter to refer in a little more detail to the
valvula cerebelli, wliich has been mentioned on several occasions in
connection with mormyrus and the barbel. The anatomical position
of this organ does not render it accessible to experimental biological
research ; if efforts were made to ablate tliis organ in the living
animal it would be necessary after exposing the tecta optica of the
optic lobes to remove them and then to dissect out the valvula from
its posterior attachment to the body of the cerebellum. This might
be possible in a hardened brain in the dead animal, but it would
involve very serious damage to other parts if attempted in the living
animal, and the results of such an ablation, as shown by the behaviour
of the fish, would be open to justifiably adverse criticism. This
being the case it seems that an attempt at understanding the phy-
siology of the valvula might be undertaken by a series of anatomical
observations on groups of fish closely related, but yet differing in
a marked way in their habits and behaviour.

There has been very little done in this direction, but it may be
of some value if the facts already known about this organ be men-
tioned, and they may suggest an attack on this problem by other
workers. In the cod, Gadus morrhua, the hinder part of the optic
ventricle is occupied by a forwardly projecting process of the cere-
bellum, the valvula. " This structure is formed by the invagination
of the anterior parts of the cerebellum into the cavity of the mid-
brain, and thus in sagittal section shows two super-imposed layers —
the lower one continuous behind with the cerebellum, and passing
in front by reduplication into the dorsal layer. The latter is closely
applied to the first and is continuous posteriorly with the liinder
margin of the tectum opticum." This description is taken from the
physiological catalogue of the Royal College of Surgeons and,
accompanying this, are drawings which demonstrate the relations
very clearly (page 19).

If we dissect the brain of a mackerel we find a very different
picture. The valvula cerebelli occupies the major part of the
cavity of the optic lobes and causes them to appear very large.

138

VALVULA CEREBELLI 130

There is a central portion continuous with the cereljelhini and
two hiteral sausage-like lobes which are bent backwards on each
other so that their terniiuations turn rn wards towards their attach-
ments to the central lobe. The Trichiuridse Avhich include the
scabbard fishes are nearly related to the Scombriformes so it is not
sur])rising that they have also a special development of the valvula,
as A\e have demonstrated in serial sections of the Black scabbard
fish. When we come to the carps we find that in the ground-
feeding members of the family the tecta optica are more or less
widely separated by the protrusion of a large valvula, so that the
o])tic lobes are superficially very large and this is most marked in the
Barbel.

As we have already noticed, MormjTus has a brain remarkable
for the immense development of the valvula which bursts through
the tectum opticum and forms a canopy over the whole brain.
To quote the R.C.S. catalogue, " this unusual relation of the valvula
cerebelli to the tectum opticum appears to be a further extension
of some such process as that seen in the carp, in which the lateral
parts of the tectum are divaricated and the central area much thinned
out, but without extrusion of the valvula. The wings of the valvula
are called anterior lateral and posterior. Their deep surface is
occupied by a layer of small cells (nuclear layer) covered over super-
ficially by a number of parallel ridges, each composed of molecular,
nuclear, intermediate, and fibrous layers." The result is of a very
beautiful design.

We will now mention a fact of great importance, as will be seen
later, that the valvula is connected by tracts to the lobi inferiores, a
structiu"e which will be the subject of the next part of this chapter.
As to the function of the lobi inferiores, we have the authority
of Herrick for the statement that "the ascencUng secondary
gustatory tract of both facial and vagal lobes terminate in the
superior secondary gustatory nucleus, situate in the lateral wall
of the isthmus. The third chief gustatory tract arises from the
cortical layers of the above ; this tract is somewhat obscm*e, but
it has been proved that fibres pass from it ventrally to the lateral
lobule of the lobus inferior." He also gives a diagram showing
the relations of the olfactory and gustatory paths in teleosts in
which an olfactory conduction path terminates in the lobus inferior,
" this is the chief centre of correlation of olfactory and other
higher senses."

140 BRAIN AND BODY OF FISH

THE LOBI INFERIORES, SACCUS VASCULOSUS
AND INFUNDIBULUM

There is an area of the brain that receives in teleostean fisli
even less consideration than the cerebelhim does in human anatomy
— we refer to the ventral surface of the brain which lies beneath
the optic lobes, posterior to the crossing of the optic nerves, and
anterior to the medulla oblongata. This area is illustrated in the
drawing of the brain of the halibut. In the right hand figure the
lobi inferiores appear partially obscuring the large optic lobes.
They are large bean-shaped bodies with their median surfaces in
apposition. In the middle line in front of these is the spherical
pituitary body. The apposition of the pituitary body and lobi
inferiores is not complete, and a triangular space is left, by which
there emerges, on the ventral surface of the brain, the red thin-
walled saccus vasculosus.

Anterior to the lobi inferiores, the most prominent area of grey
matter is the lobus infundibuli. From the third ventricle, which
lies in front of the level of the optic lobes, there passes a hollow
passage in the body of the lobus infundibuli. This passage is
prolonged downwards and somewhat backwards and communicates
ventrally with the cavity of the pituitary body. Whereas the ven-
tricle is lined with ciliary epithelium the infundibulum has a lining
of ependyma, which in part consists of large multipolar nerve
cells. Before terminating in the cavity of the pituitary, that
is to say more dorsally, the infundibulum communicates with
large cavities within the lobi inferiores, and finally it is prolonged
into the saccus vasculosus ; this is a thin-walled glandular sac and
is very vascular ; its function would appear to supply the ventricles
with their fluid content. If the drawing of the transverse section
of the optic lobes of the plaice be referred to the saccus vasculus
will be seen and its communication with the infundibulum.

The ventricles of the lobi inferiores are also to be seen, and dorsal
to them on either side are two very characteristic oval areas of
nerve cells, these are the nuclei rotundi. The minute anatomy of
these nuclei is as follows : a number of round or oval islets of grey
matter lie in a granular matrix interspersed with nerve fibres.
Nerve fibres pass from the ventral part of the lobi inferiores into
the nuclei and some smround it. Dorsally, a large strand of
nerve fibres passes from the nucleus to the thalamus.

Before considering the very important subject of the pituitary
body, we propose to mention the relations of the lobus infundibuli
to the lobi inferiores. We have made a special study of the course

VALVULA CEREBELLI 141

of the nerve connections in the Megrim as well as the sole, plaice,
and other Hattish. If we examine sections of the lobus infundibuli,
we note that anteriorly beyond the attachment of the pituitary
body, fibres pass ventrally from either side of the infundibulum,
and miite to form a median bundle of nerve fibres ; a little fm-ther
caudally, a meshwork of fibres appears on the lateral margins of the
mescnc'ojihalon. and the median bundle becomes dumb-bell shaped.
Further back still the median bundle divides into two trunks which
move dorsally. The pituitary is here attached to the margins of
the infundibulum, and a ganglion appears on either side of the
attachment, from wliich fibres pass dorsally along the margin to
the meshwork before mentioned. Finally nerve fibres pass from
this meshwork to join the fibres of the nervus infundibuli, and the
combined trunk enters the ventral end of the nuclei rotundi. It
thus appears that the lobi inferiores are the co-ordinating centres,
not only for the olfactory and hypothalmic impulses but also for
the optic and gustatory nerves, and that the nuclei rotundi also
receive fibres from the infundibulum and from the lateral nuclei
at the attachment of the pituitary body. As we have already
seen fibres pass to these lobes from the valvula cerebelH, so that
there is every reason on anatomical grounds to consider the lobi
inferiores are an important organ for regulating the various messages
received from both sensory organs and hormone producing organs
as the pituitary body and possibly also the pineal gland.

CHAPTER XVIII
THE PITUITARY BODY

The late Professor Haldane once said that anatomy, if studied
with inteUigence, can be as important as physiology in the investi-
gation of function. In this chapter I hope to be able to convince
my readers that this is possible ; if the pituitary body is studied
with a definite object in view, results can be obtained without the
aid of experimental biology, which are quite as convincing as many
physiological experiments. The only drawback to this method is
the length of time involved to get the required answer to the
problem set before the observer. Before describing the pituitary
body of the eel, which is the fish we have chosen to study, we may
state that it is known that in vertebrates, this gland produces
hormones (chemical messengers) that have a great influence on the
generative organs, and these are known as gonadotropic hormones ;
it is further known that certain changes take place in the gland,
as shown by the staining reactions of certain cells, when these
hormones are producing their effect.

Now the eel, as is well known, puts on its wedding dress, which
is evident when it changes from the yellow eel to the silver eel,
and its eyes become larger and its flesh becomes firmer, when it is
about to migrate and start on its marvellous journey to its breeding
ground. These changes do not take place in the British Isles as a
rule until the end of the summer or the beginning of autumn.
Now it occurred to the writer, as the result of certain findings as
totheappearanceof the gland when sections were examined, which
did not tally with textbook descriptions, that the differences noted
might be due to the different periods of the year, at which the fish
had been taken. This proved to be the case and it was decided to
make a systematic series of examinations of the pituitary of the
eel, commencing in the spring and going on month by month until
the end of October, or the beginning of November. These ob-
servations were carried out during a period of two years, and it was
very definitely shown that marked changes took place in the micros-
copic appearance of the gland in September and October ; and it
was further shown that these changes took place before the ovaries

142

THE PITUITARY BODY 14S

and sperm began to develop and could be clearly recognised. The
conclusion drawn was that we had demonstrated a connection be-
tween the hornionic activity of the gland and the physiological m-ge
to migrate.

The pituitary body of fish is a small body resting just in front
of the lobi inferiores and is attached to the ventral walls of the
infundibuhim either cUrectly or by a stalk. In the eel it is ovoid
in shape, and Mlien divided sagittally in the middle line, it is some-
what kidney-shaped, with the hilum facing the infundibulum.
The pituitary body in the eel consists of two lobes anterior and
posterior ; they are entirely different, both in their structm'e and
development. The posterior lobe is developed as a hollow down-
growth of the part of the embryonic brain w^hich afterw^ards becomes
the third ventricle. In fishes the cells wliich compose its walls
become converted into nerve-cells and fibres, and, as the lobus
infundibuli, become an integral part of the brain. The anterior
lobe is developed as a tubular prolongation from the epiblast of the
lining epithelium of the cavity of the mouth, with which it is there-
fore originally in connection. In the adult it is constituted of a
large number of tubules or alveoli like those of a secreting gland,
and in like manner Uned by epithelium. It is very vascular and has
a number of sinusoidal spaces. Vesicles are sometimes present,
and these may be filled with a colloid substance similar to that
found in the thyroid vesicles.

The posterior lobe of the eel consists of two parts, closely related,
the pars intermedia and pars nervosa. In comparison with a
mammalian gland, the pars nervosa takes a very small share in
the make-up ; a number of nerve fibres spread out fan-like and
loose themselves in pockets of the pars intermedia. This consists
of a large number of acini, the peripheral part of which is, in the
resting stage, composed of tlu-ee or four layers of cells, while in the
centre is a circle of ependyma cells connected with the neuroglia
of the pars nervosa. In the active stage, that is to say towards the
autumn, a very different picture is seen. The peripheral cells are
almost entirely replaced by a yellow-staining secretion. In the
centre the lumen is surrounded by empty ovoid cells. In some
acini the granular secretion is converted into an ovoid refractile
mass of brown colloid.

Surrounding each alveolus is a meshwork of fine capillaries, and
witliin them can be clearly seen small masses of brown colloid,,
together with a few red blood corpuscles. The blood vessels arise
from small peripheral arteries which can be seen passing inwards
to end in the capillary anastomosis.

144 BRAIN AND BODY OF FISH

The capillaries from the acini converge to form three groups
from which large vessels arise, two lateral and one central, and these
unite towards the hilum to form one main trunk which passes
backwards and enters the substance of the lobus infundibuli. The
evidence as to the direction of the circulation in this system, which
appears to start from the fine capillaries, is as follows ; there can
be seen in the course of the branching nerve fibres of the pars
nervosa, as it enters the pockets of the pars intermedia, a number
of small roimd globules of colloid which are homogeneous and
refractile. It will also be noted that the hollows into which the
nerves dip are bordered by blood vessels, from which it seems that
the globules have escaped. These globules have been described
in the mammalian pituitary by Miss Una Fielding, who has looked
over my sections and given me much help in reference to the volum-
inous literature on the subject. She points out that the direction
of the blood flow can be gauged by their colloid accompaniment.
As colloid (whatever it may be) is formed in the pituitary, and not
in the brain, its association with vessels may be taken as incUcating
that the direction of the flow is certainly hypophysio-hypothalmic
(that is from the pituitary to the part of the brain known as the
hypothalamus).

Sometimes the colloid globules occur in the lumen of the vessels
and sometimes in the inter-vascular tissue of the neural portion
of the stalk. Our very deflnite observation of the colloid material
in the network of capillaries that surround the acini seems to confirm
the suggestion of de Beer that products other than those of the
anterior lobe may be blood borne.

It has been established by Hogben that the normal rhythm of
colour change in amphibia is controlled by the secretion of the
pituitary gland. He points out that " the pituitary gland is a speci-
fically vertebrate structure, from which no active substance has
been isolated in pure form. Extracts of the pituitary of mammals,
birds, reptiles, and fishes (teleostean and elasmobranch) have a
powerful excitatory action on the mammalian uterus and upon
mammary secretion. Extracts of all classes of Amniota, Amphibia,
and Teleostei exert a pressure diuretic action on the mammal. In
fact the posterior lobe of the mammal while the storehouse of
probably several substances of very great activity has not yet been
proved conclusively to have any functional significance." There is,
on the other hand, the clearest evidence for regarding the pituitary
secretion as the main factor in co-ordinating the pigmentary res-
ponses of amphibia to the changing conditions of its environment.

The pigmentary changes in the eel when about to migrate are

THE PITUITARY BODY 145

well kno^nl. Before the beginning of puberty, the eels, male and
female alike, have a green and yellow coloration, dark brownish
green on the back, and bright yellow o*n the belly. These colours
now change ; the green takes on a shade of purple and the yellow
of the belly fades away and becomes a pearly white. There is also
noted an apparent increase in the size of the eyes, they protrude
more, in fact there is a condition allied to exophthalmos. Associated
with these external appearances we have noted a great increase in
size of the jDosterior lobe and also a gradual replacement of the
peripheral cells of the alveoli by yellow colloid secretion which also
appears in the surrounding capillaries. It seems, therefore, that the
theory that these colour changes are connected with the increased
size of the lobe and the secretory activity receives some support
from the analogy of the pigmentary reponse in amphibia as des-
cribed by Hogben. Before referring in detail to other changes in
the pituitary gland at the time of approaching migration it will
be necessary to say something about the microscopic appearances
of the anterior lobe.

Compared to the posterior lobe, the anterior is very vascular.
It consists of groups and columns of epithelial cells with dilated
sinusoids. The cells are either cliromophil or chromophobe. The
chromophil are either acidophil or basophil. The acidophil cells
are granular and stain pink with eosin, while the basophil cells
stain with haematoxylin and are larger than the eosin-staining,
though their granules are smaller. The sinusoids vary much in size,
and in that part of the anterior lobe, which faces the opening of the
infmidibulum, there is a wide depression occupied by what appears
to be a stringy secretion, and deep pits dip into the epithelial portion
of the gland. Anterior to the infimdibulum the stalk of the pituitary
descends from the margin of the lobus infundibuH. In addition
to the secretion above mentioned it is important to notice the
small globular refractile bodies described in the mammalian pituitary
by Poppa and Fielding. These globules are seen to be present in
considerable numbers in the hilum facing the infundibulum and
I have counted twenty globules of various sizes in two consecutive
serial sections and in a neighbouring section, two vessels have been
cut in a longitudinal direction containing numerous globules which
give the apjDearance of a string of beads and others are seen lying
free.

These vessels communicate with the capillaries of the pia mater
clothing, the ventral margin of the lobus infundibuH and from
these vessels ascending branches supply the tissues of the stalk.
The above is the appearance of a section of the gland in the autumn,

K

146 BRAIN AND BODY OF FISH

in \\ liicli there is very marked enlargement and a great increase of
eosinophil cells. Newton in Evans's Advances in Physiology says,
" In spite of the presence of only three types of cell in the anterior
pituitary it is nevertheless the birthplace of a number of hormones.
These act on the general bodily growth, the gonads, the mammary
glands, the thyroid and carbohydrate metabolism, and other tissues
and glands." What is of interest to us in the study of the eel is
obviously not the mammary glands. But bodily growth, the gonads,
the thyroid, and general metabolism all enter into the picture of
the complicated functions of the anterior pituitary. The earliest
knowledge of the functions of the pituitary were connected with
growth in certain pathological conditions. In mammals the anterior
lobe secretes a growth hormone which jDroduces, if in excess, gigan-
tism in children, and in adults acromegaly.

The relationship between the gonads and the anterior pituitary
is based on accurate observation. The secretions of the ovary
are dependent on the pituitary and occur in a cyclic fashion. These
fimctions of the gonads are not automatic, for removal of the
pituitary in immature animals prevents their apj^earance, and in
mature animals leads to their arrest. Implantation of anterior
pituitary substance, or the injection of suitable extracts before
puberty, causes precocious sexual matm-ity. It is not surprising,
therefore, to find in the eel about to migrate a great enlargement of
the anterior pituitary. It has already been stated that the source
of the gonadotropic hormones is in doubt, but the anterior pituitarj-
is increased in nearly all species during jDregnancy. This, in
mammals, is said to be due to chromophobe cells, but in man
some eosinojihile cells are produced and are kno^vn as pregnancy
cells. It is clear that so far as my observations are concerned,
Avhich extend over a period of over tw^o years, that the enlargement
of the anterior pituitary in the eel is due to a gi'eat increase of
acido|)hile cells. It is interesting to refer to some earher work on
the stimulus for a breeding migration, especially in the eel.

Walter Heape suggested that the immediate stimulus came
from the periodic increase of the internal secretions, elaborated by
the gonads. John Hammond (1925) explains his observations on
growth and reproduction, by assuming that the anterior pituitary
hormone is responsible at different times for stimulating both
processes, its physiological effect being transferred from one function
to the other in the j^rocess of individual fife. Heape w^as very
definite in his view that the imimlsc to migrate is induced by changes
in the functional activity of certain organs included in the re-
jM'oductive system, and suggests " that the details of the problem

THE PITUITARY BODY 147

must be left in the hands of the student of the reproductive
system."

Recading Heape's work carefully leaves us with the impression
that he recognised the switching off of the endocrines from a growth
producing function to a gonadotropic function ; but the work of
Parkes and others on the gonadotropic function of the anterior lobe
of the pituitary was after his time. Nevertheless, on the eel he says,
" comparison with a great variety of other animals and especially
of fish induces me to hold a strong opinion that development of the
reproductive glands and cells is due to a switching off of energy
which was previously engaged in those metabolic processes which
result in growth."

Our earliest knowledge of the functions of the pituitary were
connected \\ith. growth in certain pathological conditions, and it
is now known that in mammals the anterior lobe secretes a growth
hormone wliich produces, if in excess, gigantism in chilch'en, and
in adults acromegaly. Extracts of the pars distalis in mammals
causes growth to take place when it has ceased as the result of the
removal of the pituitary.

It is interesting to note the condition analogous to gigantism
in those eels that have been prevented from making the gametic
migration. To quote Roule, " not all the eels set out on this
hazardous journey. Some at the period of puberty cannot from
the nature of the locality leave the lake or the pool in which they
are too closely imprisoned. Like the sterile trout of the deep
lakes, and for the same reason, their ovaries, for these eels are
usually females, change and atrophy. Rendered sterile and
natm-ally castrated like pullets they grow fat in the same way,
but go on living and growing larger. Eunuch eels of this sort have
been kept in captivity for twenty or tliirty years."

I am informed from a keen and scientific fisherman that he has
seen eels of this kind in New Zealand tliree feet or more in length
with the girth of a man's thigh. We have made arrangements
so that at the earliest opportunity, the pituitary of a " eunuch "
eel will be examined, and we look forward to some interesting
findings.

It is now possible to review the colour and other changes that
precede the migration of the eel in relation to the activity of the
pituitary body. The colour changes have been shown to occur
some time before the actual migration and at the same time the
posterior lobe enters on a stage of activity as shown by the increase
of colloid secretion.

The anterior lobe may be the cause of the exophthalmos, as it is

148 BRAIN AND BODY OF FISH

known to cause an increase of thyroid activity ; we have more
direct evidence of the effect of the anterior lobe on migration as the
increase in size, the increase of the acidophile cells immediately
precedes migration, the onset of which coincides with the appearance
of both ova and sperm. We hope that another year of monthly
examinations may add further confirmation of these observations
and conclusions.

CHAPTER XIX
THE PROBLEM OF PAIN IN FISHES

Do fish feel pain ? This is not an easy question to answer. For
obvious reasons we can have no real knowledge of a fish's feelings
or sensations, and there is only the answer obtained by analogy,
b}^ anatomy, and by the reactions of fish to certain stimuli or their
behaviour. I suppose we all instinctively recoil when we see a
fisherman skin a sole alive, and shudder at the living shambles in
the fish-markets of many Continental countries, and most of us
prefer to knock a caught fish on the head rather than allow it to
prolong its death agonies by flapping on the floor of the boat. But
we cannot estimate the pain or sufferings of a fish by its facial
expression as it has no facial muscles, and though certain fish, as the
gurnard, grunt when taken out of the sea we cannot say whether
these sounds indicate pain or discomfort.

There is often some confusion of thought when we use the term
cold-blooded ; the original meaning of cold-blooded is having cold
blood, and is used in relation to those vertebrates whose body
temperature varies with that of the water in wliich they live ; but
it is also used in a figurative sense as lacking in sensibihty of feeling,
and thus the conclusion is reached, quite unjustifiably, that a fish
is devoid of feeling. Some light can be throA\Ti on the question if
we approach the subject from an anatomical standpoint.

An analysis has been made in man of cutaneous sensibihty by
certain scientists, who have allowed themselves to become their
own laboratory animals, and have stuched the various sensory
functioas, and their recovery after experimental division of their
cutaneous nerves. Accorchng to some, all forms of cutaneous
sensibility, touch, temperature and pain, are grouped in two series,
one, kno^^'n as protopathic sensibility, subserves general diffuse
sensibility of a primitive form. The sense organs are arranged in
definite spots, yet the spot stimulated cannot be accurately localised.
There are separate spots for touch, heat and cold, and be it noted
" pain spots." The other series knowTi as epicritic sensibility is
more discriminatory and is probably of a later evolutionary origin.

We need not discuss the latter further as pain sensibility is not

149

150 BRAIN AND BODY OF FISH

included in this type, pain being wholly protopathic. Some
physiologists state that there are definite nerve-endings for pain,
while others regard pain as a quality which may be present in any
sense and not as itself a true sensation.

The free nerve-endings among the cells of the epidermis are
probably the pain receptors, because these endings alone are present
in some parts of the body where susceptibility to pain is the only
sense quality usually present, such as the dentine and pulp of the
teeth and the cornea of the eye. A section of the tooth of a fish
shows the outer layer of dentine and fine nerve terminals entering
the cavity, and ending free.

Again in a section of the cornea there are various plexuses in the
connective tissue and sub-epithelial layers which end in an intra-
epithehal plexus terminating in naked fibrils, often varicose, amongst
the superficial cells. Similar endings are found tliroughout the
epidermis. Skin-pain is, therefore, probably received by free
nerve endings lying between the superficial cells and is wholly of
the protopathic type. But it must be borne in mind that excessive
stimulation of other sense organs may give rise to pain, although in
such cases the pain is accompanied by the usual sensation of the
particular organ. For example, a very loud noise may cause acute
pain, but it retains, nevertheless, the character of sound.

The epidermis of fish is, of course, characterised by its scales,
which is in sharp contrast to the appendages of the human skin,
hairs and their sebaceous glands, and the sweat glands. Before
mentioning the special types of sense organs peculiar to the epidermis
of fish we must refer to the end-organs of chemical sensibihty found
in man only on moist epithelial surfaces, but in fishes they may be
present over the entire surface of the body. But the most character-
istic of the sensory organs of fish are taste-buds and the organs of
the lateral line. The former are composed of cells arranged like
the segments of an orange which are enclosed in a sheath which
leads to the surface by a pore so as to allow access to the tips of
the sensory cells. They are most in number on the li23s and barbels,
but are found all over the body and on the fins.

These important sense organs are also associated with tactile
sensation, as such fishes as the barbel, gudgeon, and loach have
prominent barbels, which they use in hunting for food in gravel
and stony bottoms, and it is noteworthy that in these fish the nerves
divide within the brain, so as to lead to special areas of the facial
lobe, suggesting a differentiation of function. Pit-organs is the
term appHed to a type of sensory organ found either isolated or in
a long connected linear tube, called the lateral line, often clearly

THE PROBLEM OF PAIN IN FISHES ir,l

defined by pigment, as in the haddock. These organs are very
similar to the sense orgaiLs in the ain])unae of the semi-circular
canals of the fish's ear and their function is to register waves of low
frequency.

Finally we come to an organ peculiar to fish, the function of
^^•hich has been investigated recently by Dr. Sands at the Plymouth
^Marine Laboratory ; by means of an oscillograph he has made
tracings of the action currents in these mucus canals, named from
their discoverer, Lorenzini's ampullae, and finds that they record
changes of temperature. To review the functions of the epidermis
of fish we find Touch, Chemical Sense, Taste, Vibrations and Tem-
perature all represented ; we have no means of recognising pain-
spots, but we may assume their existence as they are a primitive
form, and even if they do not exist we have seen that excessive
stimulation of any sense organ can produce a painful sensation.

There is one more piece of evidence pointing to the perception
of 25ain in fish. IMany fish are provided with poisonous spines, and
the most marked effect of the puncture by these spines is acnte
pain, in man of an excruciating nature : these organs were not
evolved for protection against man. Fish have been seen to attack
small fry with these spines, and experimental observations show
that acute symptoms follow the subcutaneous injection of the
venom of these organs. In conclusion we may say that there is
no evidence that fish do not feel pain, and there is no reason to
suppose that they are devoid oi' a protective mechanism common
to most vertebrates.

CHAPTER XX
RETROSPECT AND CONCLUSIONS

The interesting connections of brain pattern with the diet and
habitat of fish can now be reviewed.

In cyprinoids the association of diet with brain pattern is very
clearly seen in the enlarged vagal lobes and facial lobes of the
bottom feeding fish, and the differences of size of these lobes accord-
ing to the habit of extracting nutriet material from mud, or groping
and grubbing on stony or gravelly bottoms. There is reason to
beheve that the enlargement of the vagals is due to the palatal
organ, which is an adaptation for sorting, sifting, retaining, or
rejecting food particles ; there is also strong evidence that the
enlargement of the facial lobes is due to the presence of barbels
which are supplied with taste-buds, as the facial nerve divides
within the brain into two branches leading to anterior and posterior
areas of the lobe of varying size, according the imjDortance of
the barbels : when we come to the next group which are surface
feeders we find that sight is a very important sense in hunting and
therefore the optic lobes are prominent, while the vagals and facials
are small. But in the highly speciahsed planlvton feeders, like the
bleak and engraulicypris, the naked eye appearance of the brain is
remarkably altered in type, as the facial lobes do not appear on the
sm-face and can only be demonstrated in serial sections, when it is
found to be completely overlapped by the somatic-sensory lobes ;
and, further, a special area at the base of the cerebellum becomes
very prominent, and being associated with audition we have ven-
tured to call it the central acoustic lobe. This lobe is present to a
lesser extent in the fish belonging to Group II, but in the purely
ground feeders it is rudimentary. Bhimachar has also shown
that cyprionodont fish, living at the surface of the water, have
the medulla oblongata narrow, without prominent vagal and facial
lobes, and that there has been developed a conspicuous central
acoustic area.

THE BRAIN PATTERN IN GADOIDS

In this family the relationship of brain pattern to diet and
habitat is more complex. In the first group, according my classi-

152

RETROSPECT AND CONCLUSIONS 153

fication, there is a great similarity of brain pattern, but the relative
size of the somatic-sensory and facial or vago-facial lobes varies
as the diet passes from a shell-tish diet, through a mixed diet, to
a purely fish diet.

A division of this group may be made by separating the night
feeders, the ling and burbot, from haddock, cod, and so on, as in
these fish the olfactory lobes become prominent, and the optic lobes
less so in accordance with the jiredominance of smell in their search
for food. The third cUvision includes two species of an interesting
genus physics, or the forkbeards, phycis from the shallow waters of
the bays of Madeira, and phycis blemiioides landed at Milford
Haven.

The brain of both fish differs from the usual gadoid type in
the largeness of the primitive end-brain and the smallness of the
optic lobes ; the former has also a characteristic division. The
cerebellum is narrow and very elongated, and the medulla small,
with fairly large somatic-sensory lobes. Tliis bottom feeding
animal A^liich inhabits rocky ground has its ventral fin on either side
converted into a long bifid feeler, and it also has a barbel. Smell
and touch wotdd appear to be its chief hunting equipment. Phycis
is known as the " schellfisch " by the Germans, and so also is the
haddock and phycis blemiioides is also known as the rock haddock ;
this is no doubt due to the similarity of diet, but the habitat seems
to have given rise to a different adaptation in the forkbeards, as
the true haddock, hunts through its taste-buds, and phycis by
olfaction and touch, probably as a result of their rocky habitat ;
therefore diet has been over-ridden by habitat in determining the
method of hunting and brain pattern.

This seems a favourable opportunity to mention other variations
in the primitive end-brain of gadoids ; tliis lobe is always closely
connected with the olfactory lobe and varies in size with the im-
portance of smell in the hmiting equipment ; the ling and bm-bot
emphasise this fact ; when we compare their brains with other
gadoids such as the pollack and cod we notice that these lobes in
the ling are large and almost hemispherical, approaching the optic
lobes in size ; the burbot shows the same relative size of these
lobes, as is seen in the ling ; it has been described above that these
fish are both night feeders. The primitive end-brain in flat-fish
also shows well-marked variations in size. This has been noted
in the sole, and the drawing of the brain of this fish shows that it
exceeds the optic lobe in diameter and gives rise to olfactory bulbs,
wliich tend to overlap. But the members of the groups of flat-fish
II and III, in my classification, show throughout a very definite

154 BRAIN AND BODY OF FISH

similarity in size of their primitive end-brains, thus in Group II,
the end-brain of the lemon-dab, plaice, witch, and dab are larger
than those of Group III, the brill, turbot, and megrim. If we
compare the drawings of the lemon-dab and brill it will be seen
that both olfactory bulbs and primitive end-brain are more de-
veloped in the former than the latter. As Group II are bottom
feeders and feed largely by smell as well as taste, both elements of
the anterior lobes are functionally well develoj)ed.

An examination of the plate giving the dorsal view of the three
types of Cyprinoid brain will reveal the same information as to
feeding habits as we have just noted in plaice and lemon-dab.
The barbel, gudgeon, loach, and tench have an elongated primitive
end-brain of considerable size, and these fish all feed largely by
taste and no doubt also by smell. The sight-feeding group as
typified by the roach have small end-brains, while the carp group
have end-brains intermediate between the above groups.

Passing backwards we reach the mid-brain which superficially
appears as the optic lobes. These lobes are not so simply con-
stituted as would appear ; if the transverse section of the optic
lobes of the plaice be referred to, it will be seen that there is a
broad cellular layer of concentric bands of nerve tissue and fibres,
which is known as the tectum opticum, and that this forms a roof
to a cavity into which protrudes the base of the lobi inferiores and
the nerves passing dorsaUy from the nuclei rotundi. The optic
nerves passing from the internal surface of the tectum also
occupy the anterior part of the cavity on its lateral aspects. But
posteriorly as can be seen in the drawing of the brain of the cod
there is another protrusion or rather insertion of another structiu-e,
the valvula cerebelli, which not infrequently causes a wide separation
of the anterior ends of the tecta and exceptionally pushes its
way right through and envelops the optic lobes as in Mormyrus.

We would recall in illustration of tliis condition, the optic lobes
of the various groups of carps ; it has been shown that those that
feed by sight have big optic lobes as the roach, but the barbel a
ground feeder has also a large optic lobe, but the wide separation of
the tecta optica by the valvula is the cause, as is very obvious in the
drawing, which illustrates these types of cyprinoid brain. The
loach is another good example of an apparently large optic lobe.
Nevertheless the fact remains true that fish that hunt in the dark,
as the sole, have small optic lobes, wliile those that hunt by day,
as the plaice, have larger lobes. But large eyes do not necessarily
mean that a fish will have well developed optic lobes. This fact
is observed in certain fish that frequent great depths, as the Black

RETROSPECT AND CONCLUSIONS 155

Scabbard fisli, known as Espada, and the Rabbit fish or Prome-
tliit'lithys pioinotluMis. which have eyes as Large in circumference
as a two-shilling ])iece, but yet have optic lobes of average size.
At depths of 300 to 400 fathoms, approximately half a mile down,
there can be very little light and the large eyes are adapted to receive
as many photons as possible.

If the eyes of a plaice are removed they will be found to lie in
a membranous covering, filled with fluid, and this communicates
with another sac situated ventrally which also contains fluid.
By means of a muscle external to the lower sac, the fluid is driven
into the sac lining the orbit and thus the eyeball is protruded.
It would appear that this accessory organ is evolved as an aid to
the phenomenon of protective coloration. We can daily observe
that in addition to the black and yellow cliromatophores which
produce the various markings in the plaice that there are bright
red markings. The eyes of this fish are also well known for their
marked prominence. It is a matter for thought and consideration
whether the bi-lobed optic lobes of the plaice may not be associated
with the function of protective coloration. It has been noted how
that in the gudgeon, the development of barbels with their taste-
buds has produced a division of the facial lobe into anterior and
posterior lobes and that this anatomical concUtion is associated with
the grubbing habits peculiar to the species. On the same principle
there may have developed a special area in the tectum opticum
connected with the power of self- concealment by colour changes ;
tliis would explain the bi-lobed character of the plaice's optic lobes,
and also the bulge of the lemon sole's optic lobe.

The fact that these fishes feed by sight does not seem to explain
the condition, as many sight feeding fishes exist which have large
eyes and well developed optic lobes, but show no modifications of
these lobes of a similar nature. If the eyes of a sole are dissected
the minute size is very impressive and when the dissection is carried
fiu-ther it is difficult to make out any recessus orbitalis although
there is a slight projection usually of the cavity of the orbit.

The more the phenomenon of protective coloration in flat-
fishes is considered, the more complex and mysterious is the mechan-
ism of its evolution.

The various physiological and anatomical devices that have
been evolved, to perfect these fishes in adaptation to their sur-
roundings, are very complex. To the naturalist that probes
into the details of the modifications of structm-e, that complete
the picture, a feeling of wonder at the results of random variation
must arise.

156 BRAIN AND BODY OF FISH

The flat fishes have adopted a recumbent posture wliich gives
them protection from enemies and conceahnent from their prey ;
at the same time this protection is dependent on adaptive coloration
-and conformity to their background. To enable the fish to do this,
the background must be surveyed and, by some unfathomed optical
reflex, the fish becomes more skilled than the leopard, and changes
its spots, and thus produces the desired design. The great develop-
ment of the eyes and optic lobes in the plaice have been
described, but the higher centres, which presumably carry out the
artistic effects, are unknown ; but there must be some central
nervous mechanism present to control the nerve plexuses that
surround each pigment cell in the skin. The reflex artistry of the
plaice, which with natural pigments reproduces on its own canvas
a picture of its background, by what must, under the best of
circumstances, be an oblique view is one of the astounding pheno-
mena of nature.

The observations on the protrusion of the eyes and the recessus
orbitahs are most remarkable. The migration of the eye, and the
anatomical details that accompany this change of outlook, have been
described ; but a further adaptation became necessary after the
eye had reached its comparatively exalted position, as the range
of vision was not sufficiently extensive to obtain a view of the
surrounding ground ; this apparently was the case with the upper
eye. It was necessary that some mechanism should be evolved
to allow the eye to be protruded ; so it appears that a hydi'aulic
sac was develojied in connection with the membranous wall of the
orbit, and this allowed the eyeball to be protruded to a remarkable
extent. Now the protective coloration ensures the survival of the
fish, but before this protection could be obtained an extensive view
of the surroundings was essential and this was not possible until
this mechanism had been evolved ; therefore the recessus orbitalis
must have been formed before any protective colour change could
have become efficient. It would seem that the greater the power
of protrusion of the eye, the more efficient is the power of colour
adaptation, as is exemplified by a study of the plaice.

Therefore the history of the development of a flat-fish recounts
not only the adoption of the recumbent lateral posture, the migration
of the under eye to an upward site, and the loss of pigment on the
under side of the body, but the development of a high j)ower pro-
jecting optical apparatus with reflex colour-photographic control
of the chromatophores. An interesting fact has also been established
in regard to colour changes, that the time required by a particular
individual to copy the ground could be decreased by repetition,

RETROSPECT AND CONCLUSIONS 157

which clearly indicates the activity of a higher centre in the
nervous system controlling the response received from the visual
field.

We have elaborated the subject of protective coloration in
flat-fishes, because it forms a good illustration of the relation of one
of the special senses to the brain pattern. Both the naked-eye
and microscopic appearances of the optic lobes of plaice and lemon-
sole show a very striking develoiiment, and this is clearly related to
the great importance of sight both in pursuit of their prey and in
provichng protection from their enemies through adaptation of their
markings to their surroundings.

We approach the subject of the cerebellum with some trepidation
tempered, however, with some confidence and conviction, that the
methods of comparative anatomy which have been so valuable in
the study of the medulla oblongata will tlu'ow some light on a
hitherto very obscure subject.

It will be convenient to treat the cerebellum by compartments,
remembering that there is a great overlapping of the areas devoted
to their different functions. I, consists of the central lobe which
may be globular or ovoid and at times is prolonged posteriorly
into a long tongue-shaped extension wliich overlaps the medulla ;
this lobe has a molecular layer enclosing a granular layer with its
superficial ring of scattered Purkinjecells. II, on either side of the
base of the cerebelkmi are the acoustic tubercles or acustico-lateralis
areas, which some text-books still refer to as the restiform bodies ;
these are described as a pair of prominent cord-like masses of nerve
tissue on the dorsal surface of the medulla oblongata, forming
lateral boundaries of the fourth ventricle and are continued up-
wards as the posterior peduncles of the cerebellum. Ill, posteriorly
at the base of the cerebellum, lying medially between these ped-
uncles is an area of small-celled dark-staining tissue, interspersed
with transverse fibres, wliich at times forms a definite lobe, pro-
jecting backwards as in the herring, but in the carps either forms
a transverse bulge or a well-defined area, known as the central
acoustic area. IV, anteriorly the cerebellum gives off a protrusion
from its base, which projects into the cavity enclosed anteriorly
by the tecta-optica, known as the valvula ; this we have recently
discussed, in describing the optic lobes, as sometimes causing a
wide separation of the tecta optica.

It will now be our object to consider the functions of each of
these four departments as they can with some accuracy be called.
Recently a distinguished surgeon and neurologist remarked to the
writer how little was known about the functions of the cerebellum

158 BRAIN AND BODY OF FISH

and an eminent physiologist told how he had removed the cere-
bellum of fish without any effect on its equilibrium.

In reference to these remarks we find, that the classic text-
book on physiology by Michael Foster devotes 32 pages to the
cerebrum and only one to the cerebellum. It is usually taught
that the cerebellum is associated with the co-ordination of move-
ments, and this view is based on the effects produced by extirpation
of portions of the cerebellum ; it has also been observed that lateral
lesions and incisions produce a greater effect than medial or sym-
metrical ones. But clinical evidence in man is conflicting, as cases
have been recorded, in which extensive disease has existed without
any obvious disturbance of the co-ordination of movements.

It must be remembered that the acoustico-lateralis system is
only developed fully in fish, and consists of three branches connected
with roots of the seventh nerve, the eighth or auditory nerve, and
the vagus lateral-line branch, and that the fibres of all these nerves
are derived from nerve-cells in the acoustico-lateralist area of the
medulla. We shall consider first the central acoustic area III.

It was observed in the coiu"se of a systematic examination of
the brains of Cyprinoids that Engraulicypris and Bleak had minute
facial lobes only evident by examining serial sections, and that they
possessed a well-marked central acoustic lobe or area. Associated
with these facts, it is to be noted that they are surface feeders ;
in fact, Engraulcypris behaves like a clupeoid fish and as its name
impHes is an anchovy-carp. The next quest was the investigation
of the facial lobe of the herring, a typical plankton feeder. Tliis
has a well-marked central lobe, in the position of a facial lobe, but
really due to a median backward projection of the cerebellum and
its microscopic structure is like that of an invaginated central
acoustic area ; the eighth nerve can be traced to its lateral margins.
The evidence seemed to point to an auditory function. It did not
appear likely that this specialised area could be connected with the
lateral line organs as the herring has no lateral line, but only a few
sense organs on the head, one in particular being closely connected
with the posterior and anterior spherical air-vesicles which form the
accessory auditory organ of this fish.

It was argued that if a central acoustic area was really auditory
in function those fish which are known to produce sounds, such as
the gurnards, would have this area well developed; sound pro-
duction without auchtion would appear to be a useless accomplish-
ment. Certain fish, such as the mullet, are very susceptible to
sounds, and fishermen make use of this characteristic reaction in
their methods of fishing, and it is interesting to note that fishermen

RETROSPECT AND CONCLUSIONS 159

both in the south of England and as far away as Egypt apply the
same methods.

To return to the carps it is found tliat the surface feeding mem-
bers of this family have the central acoustic area well developed,
A\hile the ground feeders have this area very small. The same
obstu'vation has been made on the fish which abound in the Macb-as
tanks as described by Bhimachar, and he adds further that those
fish which have a prominent central acoustic lobe have also an
accessory air-breathing organ. It maj' be mentioned here that the
recent work of Bull ]\Iamiing, Fritsch and others have put the
question of hearing in fish beyond dispute, and the minnow is noA\'
known to have a range of hearing as wide as that possessed by the
hunuin ear.

We now come to the acoustic tubercles and shall endeavom-
to unravel its functions. Presumably if the central area is acoustic
in function, the lateral areas must be either connected with the
orientation of the body and so with the semi-circular canals, or
\\-ith the perception of waves received from the lateral-line organs
through the movement of the fluid in their canals. Recently a
Japanese observer has described experiments in which the effect
of ablation of the cerebellum was observed. These experiments
supported the view that the main body of the cerebellum is con-
nected with the registration of the position of the body in
space.

The acoustic tubercles must, therefore, be chiefly receptors
for the sense organs of the lateral line. The question now arises
are there any examples of an exceptional development of the
acoustic tubercles and is such a condition associated with any
special habitat ?

The examination of the brain of gadoids revealed the possibility
of such associated conditions being jDresent. The pattern of the
brain of gadoids is always of the same generalised type, the dif-
ferences being most noticeable in the lobes of the medulla oblongata
and in the size of the optic lobes. We have seen how enormous
is the size of the facial lobe in the rockling, and have associated this
development with the peculiar adaptation of the dorsal fin. But
this is not the only gadoid that has a marked and unusual develop-
ment of a sensory area.

Molva elongata, one of the deep-sea lings, has exceptionally
large acoustic tubercles, as is shown in the plate of the gadoid
brains, and this development is associated with a bathysmal habitat,
and also with a predacious habit of feeding. If we consider what
we have described as the pattern of the brain of the deep-sea gadoid

160 BRAIN AND BODY OF FISH

phycis blennioides of a sedentary habit, and not a predacious fish,
it will be clear that it is not only a deep sea habitat that is responsible
for large acoustic tubercles. Molva elongata hunts not only at
great depths, but is essentially a predatory fish so that two concUtions
must be present to cause the great enlargement of the acoustic
tubercles. These considerations are confirmed by what we have
noted in the pattern of the hake's brain. The cerebellum is small
and on either side are two globular projections in the position
of the acoustico-lateraUs areas, from which two ridges pass posteriori}-
and then converge and meet at the caudal end of the medulla.

It must be mentioned that the hake is a near relative of the cods,
and strictly is a species of the Merlucciidae. Its habits we have
described ; it is a mid-water feeder, but descends to from 300 to
400 fathoms in the winter. It is suggested that the hake, like
molva elongata, has this great development of the acoustic tubercles
in association with its bathysmal and predatory habits. A still
more remarkable confirmation of these suggestions is the pattern
of brain in four deep-sea fish of marked predatory habits, all members
of the Trichiuridae, Athanopus carbo, Lepidopus caudatus, Pro-
methichthys and Nesiarchus nasutus. In these fish the acoustic
tubercles are enormous and the somatic-sensory lobes also large.

It is known that in the deep sea at a depth of two hundred
fathoms light does not penetrate. According to Cunningham
it is certain that fish, that habitually live beyond this depth, show
enlargement of the dermal sensory tubes. We have seen that the
deep sea fish such as the scabbard fish and the hake have a special
enlargement of the acoustico-lateral area so as to produce a definite
lobe, and it is a justifiable assumption to associate this condition
with the enlargment of the sensory canals. How are the fish in
complete darkness able to find their prey ? They may be able to
detect its presence by smell, but it is doubtful at what distance
this would be possible, and whether smell could give any exact idea
of the position of the prey. It has been pointed out that the
lateral-line organs, situated at intervals along the side of an elongated
body, would receive different impressions from a moving body,
that was the source of waves, according to the angle these un-
dulations hit the side ; and it is obvious that when the pursuer
was heading for the prey the waves would have a very different
effect than if the undulations were received obliquely. It is also
known that the lateral-line organs perceive vibrations of water, and
react to them in the same way that the similar organs of the semi-
circular canals do. The conclusion is therefore drawn that the
enlarged acoustic tubercles of deep sea fish are organs for direction

RETROSPECT AND CONCLUSIONS 101

finding of a moving j)rey, and are stimulated by the impressions
received from the hxteral-line organs.

We must now refer to the valvuhv cerebelli. We are not quite
in the dark even about the functions of tliis difficult area. There
is no doubt that those fish like the barbel, the gudgeon and the
loaches which grub among stones and gravel for their food have the
tecta optica widely separated by the prominence of the valvula
and a freak development of a very highly elaborated valvula occurs
in a group of fishes with strongly curved snouts, the Mormyridae :
this suggests that tliis part of the cerebellum may be associated
Mith the grubbing habit, correlating the tactile, olfactory, and
gustatory sensations. In this connection it is well to recall that
fibres from the lobi inferiores and also gustatory fibres enter an
area at the base of the valvula.

Herrick is our authority for some evidence as to the termination
of the ascending secondarj^ gustatory tract of both facial and vagal
lobes ^^•hicll is stated to be in a nucleus, situated in the lateral wall
of the isthmus. The chief gustatory tract arises from it and some
of these fibres pass to the lobi inferiores. He also gives a diagram,
in which an olfactory conduction path terminates in the lobus
inferior. It would indeed be surprising if we could fit in everything
in this nem"ological puzzle ; but an attempt has been made by the
comparative method to supplement the knowledge that has been
gained by other methods ; there can be little doubt that more
observations ^^'ill be made and some conclusions confirmed, and
others revised or extended.

To sum up the new views that have been suggested by the
methods of comparative anatomy : the cerebellum is the centre
whereby the position of the animal in space is recognised and the
semi-circular canals are in close cormection with its main body ; the
central, acoustic area is the auditory centre and most of the fibres of
the eighth nerve enter it ; the lateral-acoustic areas are associated
with direction, and direction in a vertical plane is well shown by
a study of their enlargement in abysmal fishes ; the valvula is
associated with habits of grubbing, and the taste and olfactory
functions, and may also be a correlation centre of these functions.

The posterior crura of the cerebellum or the continuation of
the acoustic-lateralis area into the medulla oblongata leads into
those parts of the medulla bounding the anterior portion of the fourth
ventricle ; this area is the skin area and is often enlarged and globular,
forming the somatic-sensory lobes or fifth lobes ; these lobes are
usually much developed in predatory fishes. This is well shown
in a study of the gadoid brain in which we find such fish as the

162 BRAIN AND BODY OF FISH

cod, ling and pollack, with large fifth lobes and, among flatfish,
as turbot and particularly the halibut, the same enlargement is
present.

Our task is nearly finished and we return with relief to the
seventh or facial nerve and taste-buds. As has been pointed out,
these organs apparently subserve different kind of sensory stimuh,
which we have called gustatory and "sentinel." The taste-buds,
with gustatory function are usually in the front of the mouth and
on the lips, and the sentinel in the branchial region and at the
entrance of the pneumatic duct into the gullet.

These groups are supplied by the facial, the glossopharyngeal
and the vagal. When taste-buds are also situated on barbels
it is found that the nerve fibres from the barbels enter the facial
lobe with the fibres from the lips and body, but separate within
the lobe, the barbel fibres passing to the posterior part of the lobe.
The facial lobe only undergoes great enlargement when there are
barbels present.

In the carps the mud-eating members of the family have a
specialised area, carrying taste-buds called the palatal organ,
which sifts the mud from the nutriment ; tliis area is supplied by
the IX and vagal nerves and is the cause of the great enlargement
of the vagal lobes. It is interesting to note how taste-buds may
form special taste areas as is seen in the rocklings, which have
adapted the dorsal fin to a taste organ. The dorsal fin is adapted
in other fish for other purposes besides propulsion and gustation,
it becomes a weapon of defence in the weever and a lure in the
angler fish ; in the rocklings, however, it causes a great enlargement
of the facial lobe.

The question has not yet been answered, is the adaptation of
an organ to a specialised function as the palatal organ, the cause
of the increased size of the vagal lobe, or vice versa, does the neces-
sity of providing a sorting organ, cause an enlarged vagal to stimulate
the formation of a palatal organ ; also how far random variation
is responsible for the existence of the palatal organ, and the en-
larged vagal.

It is a matter of common knowledge that, in the liigher verteb-
rates, the importance of the special senses vary to a considerable
extent.

In man we all realise how little the sense of smell is developed
compared with the conchtion found in the lower animals. This
fact is even borne out in the study of language. Logan Pearsall
Smith in the appendix to his book, " Words and Idioms," gives
a hst of " corporeal " idioms, and points out that the human head

RETROSPECT AND CONCLUSIONS 163

with its eyes and ears, nose and month, is the source of more than
two hundi'cd idioms. If a list of the idioms, rehiting to the eye,
ear and nose (as an olfactory organ and not a feature) is made
it A\ill be fomid that there are forty-tln-ee idioms related to the
e5'"e, twenty-two relating to the car, and nine relating to the nose,
so that roughly the relative importance of these three senses is
in man five — three — one. We cannot apply this test to a dog,
but ])robably the relative importance of these senses in a dog Mould
be smell, as indicated by its marvellous scent, then hearing and
lastly sight ; but we imagine the proportional significance of each
sense would not vary to the same extent, that would appear to
be the case with. man. As regards hearing in the dog, it is known
that it can be called by a Galton whistle when the note sounded
is beyond the range of human hearing, and it is usually held that
a dog is coloiu" blind.

Contrary to general opinion the range of hearing in a fish,
such as the mimiOAv, is as wide as in man. and there is good reason
to believe that fish are not colom'-blind.

Fortunately the brain of a fish is so constituted that their is no
difficulty in estimating the relative importance of the different
senses in the make-up of the brain, as we have already had occasion
to note, in describing the various lobes and their different sizes
in the four groups of the carp family. There will, therefore, be no
difficulty in coming to some definite conclusions as regards the
brain of the sole, and these are, as we have already indicated, ol-
factory, auditory and tactile, as shown by the size of the olfactory
central acoustic and somatic-sensory lobes : the facial lobe is
insignificant, as there are no taste-buds on the papillary area.
Why the gustatory system should be neghgible in the sole and
so important in other bottom feeding fish is difficult to explain,
and we have no answer available.

The fact that the sole is able to live in brakish water makes it
seem probable that it may be in an early evolutionary stage towards
becoming anadromous, like the flounder. The sole likes warmth
and has been known to live in fresh water, and this accounts for
its occasional wanderings into estuaries.

This raises the question of choice. Recently Professor F.
Balfour-BroMne gave an interesting Presidential Address on the
species problem in relation to water-beetles, and the conclusions
he drew may help us to understand the liistory of the sole. If
choice can be attributed to ^atcr-beetles, it can also be attributed
to flat-fishes, winch are far higher up the ladder of evolution. He
pointed out that although the struggle for existence undoubtedly

164 BRAIN AND BODY OF FISH

occurs, choice also plays a part in the formation of different com-
munities associated with different tj^pes of habitat and that even
locahsation under particular climatic conditions may be due to
choice. Choice of a particular food by certain individuals of a
species may give rise to biological races.

THE LONDON AND NORWICH PRESS, LIMITED, ST. GILES WORKS, NORWICH