Full text of "Comparative physiology of the brain and comparative psychology"

See other formats

Digitized by the Internet Archive

in 2007 with funding from

IVIicrosoft Corporation

http://www.archive.org/details/comparativephysiOOIoebrich

Ube Science Series

EDITED BY

ptoUseov 3. »c*een Cattell, flD.B., t>b,

AND

COMPARATIVE PHYSIOLOGY OF THE BRAIN
AND COMPARATIVE PSYCHOLOGY

Comparative Physiology of the

Brain and Comparative

Psychology

Jacques [Loeb, M.D.

Professor of Physiology in the University of Chicago

Illustrated

New York

G. P. Putnam's Sons

London: John Murray

I002

G. P. PUTNAM'S SONS

Ube 'Rnicfierbocliec presf, flew ^tft

PROFESSOR ERNST MACH

HHi:)l

PREFACE

It is the purpose of this book to serve as a short
introduction to the comp3.rative physiology of the
brain and of the central nervous system.

Physiology has thus far been essentially the physi-
ology of Vertebrates. I am convinced, however, that
for the establishment of the laws of life-phenomena
a broader basis is necessary. Such a basis can be
furnished only by a comparative physiology which
includes all classes of the animal kingdom. My ex-
perience in the course on comparative physiology
at Wood's HoU seems to indicate that the transition
from the old to the comparative physiology can be
most readily accomplished through the physiology of
the central nervous system. ^

The physiology of the brain has been rendered
unnecessarily difficult through the fact that meta-
physicians have at all times concerned themselves
with the interpretation of brain functions and have
introduced such metaphysical conceptions as soul,
consciousness, will, etc. One part of the work of
the physiologist must consist in the substitution of
real physiological processes for these inadequate con-
ceptions. Professor Ernst Mach, of Vienna, to whom

vi PREFACE

this book is dedicated, was the first to establish the
general principles of an antimetaphysical science.

I have added at the end of each chapter a list of
the chief papers of which I have made use. Although
far from complete, this may serve the beginner as a
guide for the further study of the subjects touched
upon.

The book appeared first in German and was trans-
lated by Anne Leonard Loeb. As a number of new
facts have been found since the German edition ap-
peared, and as it seemed desirable to formulate my
antimetaphysical standpoint more precisely, I have
made extensive alterations.

My thanks are due to a number of friends who
have offered suggestions, — most of all to my pupil,
Miss Anne Moore.

The University of Chicago,
October i. 1900.

CONTENTS

CHAPTER I.

PAGB

Some Fundamental Facts and Conceptions Concern-
ing THE Comparative Physiology of the Cen-
tral Nervous System i

CHAPTER II.
The Central Nervous System of Medusae. Experi-
ments ON Spontaneity and Coordination . . i6

CHAPTER III.
The Central Nervous System of Ascidians and its

Significance in the Mechanism of Reflexes . 35

CHAPTER IV.

Experiments on Actinians 48

CHAPTER V.

Experiments on Echinoderms 61

CHAPTER VI.

Experiments on Worms 72

CHAPTER VII.

Experiments on Arthropods loi

CHAPTER VIII.
Experiments on Mollusks 128

vii

via

CONTENTS

CHAPTER IX.
The Segmental Theory in Vertebrates

CHAPTER X.
Semidecussation of Fibres and Forced Movements

CHAPTER XI.
Relations between the Orientation and Function of
Certain Elements of the Segmental Ganglia

CHAPTER XII.
Experiments on the Cerebellum ....

CHAPTER XIII.
On the Theory of Animal Instincts

CHAPTER XIV.
The Central Nervous System and Heredity

CHAPTER XV.

The Distribution of Associative Memory in the Ani
MAL Kingdom

CHAPTER XVI.

Cerebral Hemispheres and Associative Memory .

CHAPTER XVII.
Anatomical and Psychic Localisation .

CHAPTER XVIII.
Disturbances of Associative Memory .

CHAPTER XIX.

On Some Starting-Points for a Future Analysis of
THE Mechanics of Associative Memory

Index

PAGS

160

213
236

277
289

ILLUSTRATIONS IN THE TEXT

FIGURE PACK

1. Hydromedusa (Gonionemus Vertens) 17

2. Diagram of the Bell of Aurelia Aurita, (After Claus.) . 18

3. Experiment in Dividing a Hydromedusa 19

4. Arrangement for Producing Automatically Pulsating Air-

Bubbles 21

5. Dr. Hargitt's Experiment 27

6. Diagram of the Ascidian Heart . . . . . .28

7. Localising Reflex in Tiaropsis Indicans 31

8. Diagram for Explaining the Localising Reflex in Medusa 32

9. ClONA Intestinalis 36

10. The Ability of the Actinians to Discriminate ... 49

11. Continuation of the Experiment in Fig. 10 .... 50

12. An Actinian (Cerianthus) with a Normal Head and with

an Artificially Produced Head 52

13. An Actinian (Cerianthus) that had been Placed in a Test-

TuBE, Head down, Regaining its Normal Orientation . 55

14. Cerianthus Regaining its Normal Orientation . . .57

15. Actinian that has been Forced by Gravitation to Push

itself through a Wire Net Three Times .... 59

16. Nervous System of a Starfish .... . . 61

17. Mechanism of the Turning of a Starfish that has been Laid

ON its Back 62

18. The Same Experiment on a Starfish whose Nerve-Ring has

been Severed in Two Places 63

19. Geotrgpic Reaction OF CucuM ARIA Cucumis . . . .67

X ILLUSTRATIONS IN THE TEXT

FIGURE PAGE

20. ThysanozoOn Brocchii, a Marine Planarian .... 72

21. Thysanozoon Divided Transversely 73

22. ThysanozoOn with Transverse Incision 75

23. Fresh- Water Planarian (Planaria Torva) .... 77

24. Two-Headed Planarian Produced Artificially. (After

van Duyne.) 81

25. Planarian with Two Heads that are Attempting to Move

IN Opposite Directions, and in so Doing are Tearing
the Common Body. (After van Duyne.) .... 82

26. The Brain and a Series of Segmental Ganglia of an An-

nelid (Nereis) 83

27. Dorsal View of the Central Nervous System of an Earth-

worm 84

28. Side View of the Central Nervous System of the Earth-

worm 85

29. A Group of Nereis whose Brains have been Removed. They

AT Last Collect in a Corner of the Aquarium and Perish
in the Vain Attempt to Go through the Glass. (After
Maxwell.) 93

30. Head of Nereis. (After Quatrefages.) 98

31. LiMULUs Polyphemus with the Central Nervous System

Exposed 103

32. Lobster with Central Nervous System Exposed . . .115

33. Diagrammatic Representation of the Central Nervous

System of a Snail (Paludina Vivipara) .... 128

34. Brain of Sepia 129

35. The Frog's Brain 139

36. Position of the Appendages of Limulus after Destruction

OF the Right Half of the Brain 156

37. Attitude of an Amblystoma under the Influence of a Gal-

vanic Current Passing from Head to Tail . . . 160

38. Attitude of an Amblystoma when the Galvanic Current

Passes from Tail to Head 160

39. Cerebral Hemispheres of a Dog 260

COMPARATIVE PHYSIOLOGY OF THE BRAIN
AND COMPARATIVE PSYCHOLOGY

INTRODUCTION TO

HE COMPARATIVE PHYSIOLOGY
OF THE BRAIN

CHAPTER I

iOME FUNDAMENTAL FACTS AND CONCEPTIONS
CONCERNING THE COMPARATIVE PHYSIO-
LOGY OF THE CENTRAL NERVOUS SYSTEM

I. The understanding of complicated phenomena
lepends upon an analysis by which they are resolved
into their simple elementary components. If we ask
^hat the elementary components are in the physio-
logy of the central nervous system, our attention is
iirected to a class of processes which are called re-
lexes. A reflex is a reaction which is caused by an
[external stimulus, and which results in a coordinated
lovement, the closing of the eyelid, for example,
^hen the conjunctiva is touched by a foreign body,
►r the narrowing of the pupil under the influence of
light. In each of these cases, changes in the sensory

2 COMPARATIVE PHYSIOLOGY OF THE BRAIN

nerve-endings are produced which bring about a
change of condition in the nerves. This change
travels to the central nervous system, passes from
there to the motor nerves, and terminates in the
muscle-fibres, producing there a contraction. This
passage from the stimulated part to the central
nervous system, and back again to the peripheral
muscles, is called a reflex. There has been a growing
tendency in physiology to make reflexes the basis of
the analysis of the functions of the central nervous
system, consequently much importance has been at-
tached to the underlying processes and the necessary
mechanisms.

The name reflex suggests a comparison between
the spinal cord and a mirror. Sensory stimuli were
supposed to be reflected from the spinal cord to the
muscles ; destruction of the spinal cord would, ac-
cording to this, make the reflex impossible, just as
the breaking of the mirror prevents the reflection of
light. This comparison, however, of the reflex pro-
cess in the central nervous system with the reflection
of light has, long since, become meaningless, and at
present few physiologists in using the term reflex
think of its original significance. Instead of this,
another feature in the conception of the term reflex
has gained prominence, namely, the purposeful char-
acter of many reflex movements. The closing of the
eyelid and the narrowing of the pupil are eminently
purposeful, for the cornea is protected from hurtful
contact with foreign bodies, and the retina from the

FUNDAMENTAL FACTS %

injurious effects of strong light. Another striking
characteristic in such reflexes has also been empha-
sised. The movements which are produced are so
well planned and coordinated that it seems as though
some intelligence were at work either in devising or
in carrying them out. The fact, however, that a de-
capitated frog will brush a drop of acetic acid from
its skin, suggests that some other explanation is
necessary. A prominent psychologist has maintained
that reflexes are to be considered as the mechanicaP^
effects of acts of volition of past generations. The j
ganglion-cell seems the only place where such me-
chanical effects could be stored up. It has there-
fore been considered the most essential element of
the reflex mechanism, the nerve-fibres being regarded,
and probably correctly, merely as conductors.

Both the authors who emphasise the purposeful-
ness of the reflex act, and those who see in it only a
physical process, have invariably looked upon the
ganglion-cell as the principal bearer of the structures
for the complex coordinated movements in reflex
action.

I should have been as little inclined as any other
physiologist to doubt the correctness of this concep-
tion had not the establishment of the identity of the
reactions of animals and plants to light proved the
untenability of this view and at the same time offered
a different conception of reflexes. The flight of
the moth into the flame is a typical reflex process.
The light stimulates the peripheral sense organs, the

4 COMPARATIVE PHYSIOLOGY OF THE BRAIN

stimulus passes to the central nervous system, and
from there to the muscles of the wings, and the moth
is caused to fly into the flame. This reflex process
agrees in every point with the heliotropic effects of
light on plant organs. Since plants possess no nerves,
this identity of animal with plant heliotropism can
offer but one inference — these heliotropic effects must
depend upon conditions which are common to both
animals and plants. At the end of my book on helio-
tropism I expressed this view in the following words :
** We have seen that, in the case of animals which
possess nerves, the movements of orientation toward
light are governed by exactly the same external con-
ditions, and depend in the same way upon the external
form of the body, as in the case of plants which possess
no nerves. These heliotropic phenomena, conse-
quently, cannot depend upon specific qualities of the
central nervous system (i)." On the other hand,
the objection has been raised that destruction of the
ganglion-cells interrupts the reflex process. This
argument, however, is not sound, for the nervous
reflex arc in higher animals forms the only protoplas-
mic bridge between the sensory organs of the surface
of the body and the muscles. If we destroy the gan-
glion-cells or the central nervous system, we interrupt
the continuity of the protoplasmic conduction between
the surface of the body and the muscles, and a reflex
is no longer possible. Since the axis-cylinders of the
nerves and the ganglion-cells are nothing more than
protoplasmic formations, we are justified in seeking

FUNDAMENTAL FACTS 5

in them only general protoplasmic qualities, unless we
find that the phenomena cannot be explained by
means of the latter alone.

2. A further objection has been raised, that al-
though these reflexes occur in plants possessing no
nervous system, yet in animals where ganglion-cells
are present the very existence of ganglion-cells neces-
sitates the presence in them of special reflex mechan-
isms. It was therefore necessary to find out if there
were not animals in which coordinated reflexes still ^
continued to exist after the destruction of the central
nervous system. Such a phenomenon could be ex-
pected only in forms in which a direct transmission of
stimuli from the skin to the muscle is possible, in
addition to the transmission through the reflex arc.
This is the case, for instance, in worms and in Ascidi-
ans. I succeeded in demonstrating in Ciona intesti-
nalis that the complicated reflexes still continue after
removal of the central nervous system (2).

A study, then, of comparative physiology brings
out the fact that irritability and conductibility are the I
only qualities essential to reflexes, and these are both/
common qualities of all protoplasm. The irritable\
structures at the surface of the body, and the arrange- )
ment of the muscles, determine the character of the I
reflex act. The assumption that the central nervous
system or the ganglion-cells are the bearers of reflex
mechanisms cannot hold. But have we now to con-
clude that the nerves are superfluous and a waste ?
Certainly not. Their value lies in the fact that they

6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

are quicker and more sensitive conductors than undif-
ferentiated protoplasm. Because of these quaHties of
the nerves, an animal is better able to adapt itself to
changing conditions than it possibly could if it had
no nerves. Such power of adaptation is absolutely
necessary for free animals.

3. While some authors explain all reflexes on a
psychical basis, the majority of investigators explain in
this way only a certain group of reflexes — the so-
called instincts. Instincts are defined in various ways,
but no matter how the definition is phrased the mean-
ing seems to be that they are inherited reflexes so
purposeful and so complicated in character that no-
thing short of intelligence and experience could have
produced them. To this class of reflexes belongs
the habit possessed by certain insects of laying their
eggs on the material which the larvae will afterwards
require for food. When we consider that the female
fly pays no attention to her eggs after laying them, we
cannot cease to wonder at the seeming care which
nature takes for the preservation of the species. How
can the action of such an insect be determined if not
by mysterious structures which can only be contained
in the ganglion-cells ? How can we explain the in-
heritance of such instincts if we believe it to be a
fact that the ganglion-cells are only the conductors
of stimuli ? It was impossible either to develop a
mechanics of instincts or to explain their inheritance
in a simple way from the old standpoint, but our con-
ception makes an explanation possible. Among the

FUNDAMENTAL FACTS 7

elements which compose these complicated instincts,
the tropisms (heliotropism, chemotropism, geotropism,
stereotropism) play an important part. These trop-
isms are identical for animals and plants. The
explanation of them depends first upon the specific
irritability of certain elements of the body-surface,
and, second, upon the relations of symmetry of the
body. Symmetrical elements at the surface of the
body have the same irritability ; unsymmetrical ele-
ments have a different irritability. Those nearer the
oral pole possess an irritability greater than that of
those near the aboral pole. These circumstances
force an animal to orient itself toward a source of
stimulation in such a way that symmetrical points on
the surface of the body are stimulated equally. In
this way the animals are led without will of their own
either toward the source of the stimulus or away
from it. Thus there remains nothing for the ganglion-
cell to do but to conduct the stimulus, and this may
be accomplished by protoplasm in any form. For
the inheritance of instincts it is only necessary that
the ^gg contain certain substances — which will de-
termine the different tropisms — and the conditions
for producing bilateral symmetry of the embryo. The
mystery with which the ganglion-cell has been sur-
rounded has led not only to no definite insight into
these processes, but has proved rather a hindrance in
the attempt to find the explanation of them.

It is evident that there is no sharp line of demarc-
ation between reflexes and instincts. We find that

8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

authors prefer to speak of reflexes in cases where
the reaction of single parts or organs of an animal
to external stimuli is concerned ; while they speak
of instincts where the reaction of the animal as a
whole is involved (as is the case in tropisms).

4. If the mechanics of a number of instincts is
explained by means of the tropisms common to ani-
mals and plants, and if the significance of the gan-
glion-cells is confined, as in all reflex processes, to their
power of conducting stimuli, we are forced to ask
what circumstances determine the coordinated move-
ments in reflexes, especially in the more complicated
ones. The assumption of complicated but unknown
and perhaps unknowable structures in the ganglion-
cells served formerly as a convenient terminus for all
thought in this direction. In giving up this assump-
tion, we are called upon to show what conditions are
able to determine the coordinated character of reflex
movements. Experiments on galvanotropism of ani-
mals have proved that a simple relation must exist
between the orientation of certain motor elements in
the central nervous system and the direction of the
movements of the body which is called forth by the
activity of these elements. This perhaps creates a
rational basis for the further investigation of coordi-
nated movements.

5. We must also deprive the ganglion-cells of all
specific significance in spontaneous movements, just
as we have done in the case of simple reflexes and
instincts. By spontaneous movements we mean

FUNDAMENTAL FACTS 9

movements which are apparently determined by inter-
nal conditions of the living system. Strictly speaking,
no movements of animals are exclusively determined
by internal conditions, for the atmospheric oxygen
and a certain temperature or certain limits of tem-
perature are always necessary in order to preserve
the activity beyond a short period of time.

We must discriminate between simple and conscious
spontaneity. In simple spontaneity we must consider
two kinds of processes, namely, aperiodic spontaneous
processes and rhythmically spontaneous or automatic
processes. The rhythmical processes are of import-
ance for our consideration. Respiration and the
heart-beat belong to this category. The respiratory
movements prove without possible doubt that auto-
matic activity can arise in the ganglion-cells, and
from this the conclusion has been drawn that all
automatic movements are due to specific structures
of the ganglion-cells. Recent investigations, how-
ever, have transferred the problem of rhythmical
spontaneous contractions from the field of morphology
into that of physical chemistry. The peculiar quali-
ties of each tissue are partly due to the fact that it
contains Ions (Na, K, Ca, and others) In definite
proportions. By changing these proportions, we can
impart to a tissue properties which it does not ord-
inarily possess. If In the muscles of the skeleton
the Na ions be increased and the Ca Ions be reduced,
the muscles are able to contract rhythmically, like the
heart. It Is only the presence of Ca ions in the

10 COMPARATIVE PHYSIOLOGY OF THE BRAIN

blood which prevents the muscles of our skeleton
from beating rhythmically in our body. As the mus-
cles contain no ganglion-cells, it is certain that
the power of rhythmical spontaneous contractions
is not due to the specific morphological character
of the ganglion-cells, but to definite chemical con-
ditions which are not necessarily confined to gang-
lion-cells (3).

The coordinated character of automatic movements
has often been explained by a *' centre of coordina-
tion," which is supposed to keep a kind of police
watch on the different elements and see that they
move in the right order. Observations in lower
animals, however, show that the coordination of
automatic movements is caused by the fact that
that element which beats most quickly forces the
others to beat in its own rhythm. Aperiodic
spontaneity is still less a specific function of the gang-
lion-cell than rhythmical spontaneity. The swarm-
spores of algae, which possess no ganglion-cells,
show spontaneity equal to that of animals having
ganglion-cells.

6. Thus far we have not touched upon the most
important problem in physiology, namely, which
mechanisms give rise to that complex of phenomena
which are called psychic or conscious. Our method
of procedure must be the same as in the case of in-
stincts and reflexes. We must find out the ele-
mentary physiological processes which underlie the
complicated phenomena of consciousness. Some

FUNDAMENTAL FACTS ii

physiologists and psychologists consider the purpose-
fulness of the psychic action as the essential element.
If an animal or an organ reacts as a rational man
would do under the same circumstances, these authors
declare that we are dealing with a phenomenon of
consciousness. In this way many reflexes, the in-
stincts especially, are looked upon as psychic func-
tions. Consciousness has been ascribed even to the
spinal cord, because many of its functions are pur-
poseful. We shall see in the following chapters
that many of these reactions are merely tropisms
which may occur in exactly the same form in plants.
Plants must therefore have a psychic life, and, follow-
ing the argument, we must ascribe it to machines
also, for the tropisms depend only on simple mechan-
ical arrangements. In the last analysis, then, we
would arrive at molecules and atoms endowed with
mental qualities.

We can dispose of this view by the mere fact that
the phenomena of embryological development and of
organisation in general show a degree of purposeful-
ness which may even surpass that of any reflex or
instinctive or conscious act. And yet we do not
consider the phenomena of development to be depend-
ent upon consciousness.

On the other hand, physiologists who have appre-
ciated the untenable character of such metaphysical
speculations have held that the only alternative is
to drop the search for the mechanisms underlying
consciousness and study exclusively the results of

12 COMPARATIVE PHYSIOLOGY OF THE BRAIN

operations on the brain. This would be throw-
ing out the wheat with the chaff. The mis-
take made by metaphysicians is not that they
devote themselves to fundamental problems, but
that they employ the wrong methods of invest-
igation and substitute a play on words for ex-
planation by means of facts. If brain-physiology
gives up its fundamental problem, namely, the dis-
covery of those elementary processes which make
consciousness possible, it abandons its best possi-
bilities. But to obtain results, the errors of the
metaphysician must be avoided and explanations
must rest upon facts, not words. The method should
be the same for animal psychology that it is for
brain-physiology. It should consist in the right
understanding of the fundamental process which re-
curs in all psychic phenomena as the elemental com-
ponent. This process, according to my opinion, is the
activity of the associative memory, or of association.
Consciousness is only a metaphysical term for
phenomena which are determined by associative
memory. By associative memory I mean that
mechanism by which a stimulus brings about not
only the effects which its nature and the specific
structure of the irritable organ call for, but by which
it brings about also the effects of other stimuli which
formerly acted upon the organism almost or quite
simultaneously with the stimulus in question (4). If
an animal can be trained, if it can learn, it possesses
associative memory. By means of this criterion it

FUNDAMENTAL FACTS

can be shown that Infusoria, Coelenterates, and
worms do not possess a trace of associative memory.
Among certain classes of insects (for instance,
wasps), the existence of associative memory can be
proved. It is a comparatively easy task to find out
which representatives of the various classes of ani-
mals possess, and which do not possess, associative
memory. Our criterion therefore might be of
great assistance in the development of comparative
psychology.

7. Our criterion puts an end to the metaphysical
ideas that all matter, and hence the whole animal
world, possesses consciousness. We are brought to
the theory that only certain species of animals possess
associative memory and have consciousness, and that
it appears in them only after they have reached
a certain stage in their ontogenetic development.
This is apparent from the fact that associative
memory depends upon mechanical arrangements
which are present only in certain animals, and
present in these only after a certain develop-
ment has been reached. The fact that certain ver-
tebrates lose all power of associative memory after
the destruction of the cerebral hemispheres, and
the fact that vertebrates in which the associative
memory either is not developed at all or only slightly
developed {e. g., the shark or frog) do not differ, or
differ but slightly, in their reactions after losing the
cerebral hemispheres, support this view. The fact
that only certain animals possess the necessary

14 COMPARATIVE PHYSIOLOGY OF THE BRAIN

mechanical arrangements for associative memory,
and therefore for metaphysical consciousness, is not
stranger than the fact that only certain animals
possess the mechanical arrangements for uniting the
rays from a luminous point in one point on the
retina. The liquefaction of gases is an example of a
sudden change of condition which may be produced
when one variable is changed ; it is not surprising
that there should be sudden changes in the onto-
genetic and phylogenetic development of organisms
when there are so many variables subject to change,
and when we consider that colloids easily change
their state of matter.

It becomes evident that the unravelling of the
mechanism of associative memory is the great dis-
covery to be made in the field of brain-physiology
and psychology. But at the same time it is evident
that this mechanism cannot be unravelled by histo-
logical methods, or by operations on the brain, or by
measuring reaction times. We have to remember
that all life phenomena are ultimately due to motions
or changes occurring in colloidal substances. The
question is. Which peculiarities of the colloidal sub-
stances can make the phenomenon of associative
memory possible ? For the solution of this problem
the experience of physical chemistry and of the
physiology of the protoplasm must be combined.
From the same sources we must expect the solution
of the other fundamental problems of brain-physio-
logy, namely, the process of conduction of stimuli.

FUNDAMENTAL FACTS 15

Bibliography.

1. LoEB, J. Der Heliotropismus der Thiere und seine Ueberein-
stimmung mit dem Heliotropismus der Pflanzen. WUrzburg,
1890. A preliminary note on ihese experiments appeared
January, 1888.

2. LoEB, J. Uniersuchungen zur physiologischen Morphologie
der Thiere II. WUrzburg, 1892.

3. LoEB, J. American Journal of Physiology ^ vol. iii., p. 327
and p. 383, 1900.

4. LoEB, J. Bettrdge zur Gehirnphysiologie der Wiirmer,
P finger's ArchiVy Band Ivi., 1894.

CHAPTER II

THE CENTRAL NERVOUS SYSTEM OF MEDUSAE.
EXPERIMENTS ON SPONTANEITY AND CO-
ORDINATION

I. Experiments on Medusae or jelly-fish afford
us an excellent opportunity for analysing the con-
ditions for spontaneity and coordination, and for
deciding whether or not these phenomena are depend-
ent upon ganglion-cells. The subumbrella of the
Medusae has a very thin layer of muscle-fibres which
contract rhythmically. The contraction diminishes
the size of the swimming-bell, and forces the water
out. By means of the recoil the animal moves for-
ward. In regard to the nervous system, we must
discriminate between two classes of Medusae : first,
the Hydromedusae (Hydroidea, Fig. i), and, second,
the Acalephae, one representative of which (Aurelza
aurita, Fig. 2) is familiar to many laymen. The
nervous system of the Hydromedusae consists of a
double nerve-ring along the margin of the umbrella
{d, Fig. i). The upper nerve-ring forms a flat layer
in the ectoderm, and consists of thin fibres and gan-
glion-cells. The lower nerve-ring has thicker fibres

EXPERIMENTS ON MEDUSAE

and more ganglion-cells, and is connected with the
upper ring by nerve-fibres. In addition to this ring,
which is called the central nervous system, there is
also a peripheral nerv-
ous system, a plexus,
consisting of nerves
and scattered ganglion
cells, spread out over
the whole subumbrella
{b, Fig. i), between
the epithelium and
the muscle-layer. The
convex surface of the
umbrella consists of a
non-contractile, gelat-
inous mass, and no
nervous elements are
to be found in it.

Acalephse (Fig. 2)
have no continuous
nerve-ring, but a row of separate nerve-centres {S, Fig.
2) extends around the margin of t^e umbrella, lying in
the ectoderm, which covers the basis of the marginal
bodies (sense organs). The number of these centres
corresponds, at least in Aurelia aurita^ with the num-
ber of sense organs. This nervous system contains
no ganglion-cells, but processes called nerve-fibres go
out from special epithelial cells. The muscle-layer
of the umbrella also is said to contain a peripheral
nervous plexus (i).

Fig. I.

(Gonionemus

Hydromedusa.
vertens.)

a, umbrella ; b^ subumbrella with muscles ; c, man-
ubrium ; d^ margin of the swimming-bell with the
nerve-ring.

i8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Our first question is : Is the spontaneous locomo-
tion of the Medusae, or the rhythmical contraction of
their swimming-bell, a function of the ganglion-cells ?

©Romanes found
w that if the margin of
the bell of a Hydro-
. . * * medusa {b, Fig. 3) be

cut off, the rhythmical
contraction of the
centre of the bell {a,
Fig. 3) ceases, while

Fig. 2. Diagram of the Bell of Aurelia \\\^ marp"in b which
AuRiTA, WITH Eight Sense Organs.

(After Claus.)

contains the nerve-
ring, continues to ex-
ecute rhythmical contractions (2). The wound does
not even cause a decrease in the number or in the
strength of the marginal contractions. The exper-
iment has been repeated by other authors with the
same result. Any sort of wound can be made in
the umbrella without disturbing the rhythmical con-
tractions so long as the nerve-ring remains intact.
Thus Romanes concluded that these rhythmical con-
tractions of Hydromedusse originate in the nerve-
ring or its ganglia. I have found recently that this
whole problem is not so much a morphological problem
as a problem of physical chemistry. The osmotic
pressure of the sea-water is about equal to that of a
-| n NaCl solution. I found that if the centre of a
swimming-bell be put into a f n NaCl or -| n NaBr
solution it goes on beating rhythmically. But if a

EXPERIMENTS ON MEDUSA

small quantity of CaCl^ or KCl, or both, be added,
the centre stops beating. The centre would beat in
sea-water were it not for the presence there of Ca, K,
and possibly other ions (3).
The centre contains some
scattered ganglion-cells. It
might be argued that the
presence of these cells makes
the rhythmical contractions
in a pure NaCl solution pos-
sible. It is easy to prove
that such is not the case.
The striped skeletal muscles
of a frog do not contract
rhythmically in blood or
serum. I have shown that
this is due to the presence
of Ca ions in these liquids.
If the muscle be put into a pure NaCl or NaBr solution
of the same osmotic pressure as the blood, the muscles
contract rhythmically (4). Yet these muscles contain
no ganglion-cells. Hence it is not the presence or absence
of ganglion - cells which determines the spontaneous
rhythmical contractions, but the presence or absence of
certai7i ions, Na ions start or increase the rate of
spontaneous rhyth^nical contractions ; Ca ions diminish
the rate or inhibit such contractions altogether. How
can these ions have such an influence? In order to
explain this we must go back to the fundamental
character of protoplasmic motion. Protoplasmic

Fig. 3. Experiment in Divid-
ing A Hydromedusa.

The amputated margin continues
to contract rhythmically, while
the bell no longer contracts.

20 COMPARATIVE PHYSIOLOGY OF THE BRAIN

motions are due to changes in the physical character
of the colloidal material in the protoplasm. These
changes may consist in changes in the state of matter
or in the absorption of water by these colloids, or in
secondary changes derived from those before men-
tioned. We know that the physical qualities of the
colloids are influenced greatly by the nature and
osmotic pressure of the ions in the surrounding solu-
tion. For that labile equilibrium of the colloids
which is required for spontaneous rhythmical contrac-
tions, the Na, Ca, and K ions must be present in
definite proportions in the tissues. This proportion
must be different for the centre and the margin of a
Hydromedusa. While for the margin the proportion
in which these three ions exist in the sea-water is
adequate, for the centre of a Hydromedusa more
Na ions and less Ca ions are required. Hence, if we
put a centre without the margin into normal sea-water
it does not beat, but it will beat when put into a pure
NaCl or NaBr solution of the same osmotic pressure
as sea-water. In the pure NaCl solution Na ions of
the solution will enter into the tissues and take the
place of some of the Ca ions. This will give the col-
loids of the muscles those qualities which allow rhyth-
mical contractions. If too many Na ions enter the
tissues of the centre it will lose its irritability. The
latter will, in this case, be restored again by adding a
trace of CaCl^ to the solution. It thus happens that
the problem of spontaneous activity is no longer a
question of the presence or absence of the ganglion-

EXPERIMENTS ON MEDUSA

cells, but of the physical qualities of the colloidal sub-
stances in the tissues. But must we conclude from
this that the Na ions are the cause of the spontaneous
rhythmical contractions of the Medusa? I think
not. The ions only bring about a certain labile equi-

V^J^ff^^7f,/f^/JJJ?/J^?}JJ/77^f

v/y/^//y///////?W//f////mf/,jju>)>nj,>,,,i,jjji^jj,,j^j^,,,„„,,,,,fJ^^

Fig. 4. Arrangement for Producing Automatically Pulsating
Air-Bubbles. (See text.)

librium in the condition of the colloids of the con-
tractile tissue which allows the true cause of the
contractions to be effective. But what is this
cause ?

J. Rosenthal seems to have been the first to call
attention to the fact that it is in no way essential for
a rhythmical phenomenon to have a rhythmical cause,
and that constant conditions can lead to rhythmical
effects. If a small, constant stream of water flows
into a pipette, it will pass out rhythmically in drops.
The weight of the drop must be greater than the

22 COMPARATIVE PHYSIOLOGY OF THE BRAIN

surface-tension in the periphery of the opening of the
outlet before the drop can break off. As long as the
quantity of water running into the pipette, in the unit
of time, remains below a certain limit, it will be some
time before the drop will be heavy enough to fall.
Quincke has given a simple and elegant method by
which it is easy to produce rhythmical contractions in
air bubbles (5). I will describe the experiment as shown
in my lectures. A glass plate P (Fig. 4) is placed in
a dish B, filled with water. The lower, narrow end of
the thermometer tube T is under and at the middle
of the air-bubble, while the upper end rests in a dish
A filled with 95 per cent, alcohol. The alcohol
rises in a fine stream toward the centre of the bubble.
As soon as the alcohol comes in contact with the
bubble, the alcohol spreads out on the limit between
the air and the water, because the sum of the surface-
tensions between air and alcohol and alcohol and
water is less than the surface-tension between air and
water. By the decrease in the surface-tension the
bubble becomes flatter and broader. In consequence
of the vortex movements in the water that are pro-
duced by the spreading, the flow of the alcohol to the
bubble is interrupted. The layer of alcohol around
the bubble diffuses rapidly into the surrounding water,
and the bubble becomes again higher and narrower.
The alcohol can flow to the bubble again now that
the vortex-movements have ceased, and the flattening
of the bubble again takes place, and so on. Under
the above-mentioned conditions I obtained about

EXPERIMENTS ON MEDUSA 23

eighty pulsations per minute — /. e., about the period-
icity of the heart.

Now, as regards the origin of the rhythmical activ-
ity of Medusae, of the heart, and of respiratory activity,
we can imagine that a constant fermentative produc-
tion of certain compounds in the automatically active
cell corresponds to the constant flow of alcohol in
Quincke's experiment. These substances may be of
such a nature that they occasion spreading-phenomena
or some other physical change in the colloids of the
muscle. But a certain quantity of these substances
must be present before this change occurs, hence the
periodicity of the contractions. But whether it be a
constant fermentative production of some substance
or not, the ultimate constant cause for the production
is the heat or the intensity factor of the same — the
temperature. It now can no longer surprise us that
Romanes found that the centre of an Acalepha is able
to beat rhythmically in normal sea-water if severed
from the margin. As long as we assume that the
ganglion-cells are the essential element in spontaneity,
this experience on Acalephse would be difficult to ex-
plain. As it is, we are only obliged to conclude that
in Acalephae there is less difference between the col-
loidal substances of the margin and centre than in
Hydromedusae.

2. Not only the spontaneous character of locomo-
tions is commonly considered to be due to ganglion-
cells, but the coordinated character of these motions
as well. Let us see how far this notion is correct.

24 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Romanes found that if the whole margin of the
umbrella of a Hydromedusa be cut off, and only a tiny
piece left, this is sufficient to keep up the spontane-
ous activity of the jelly-fish in sea-water. From this
it would appear that any element of the margin may
be considered a centre for the rhythmical contractions
of the whole Medusa. But if this be the case, how
does it happen that the whole umbrella contracts sim-
ultaneously, and why do we not find one part of the
margin in systole and the other in diastole ? This
coordination is by no means to be taken for granted.
It is present only in healthy specimens, and is wanting
in injured or dying specimens, a fact to which Roma-
nes called attention. The problem of the mechanism
of this coordination has been dismissed by many au-
thors by the assumption of a " coordinating centre "
that is supposed to control this coordination. We
shall shortly be in a position to decide whether coor-
dination in lower animals is controlled by a special
** centre of coordination," or whether it is not rather
the result of simple laws of stimulation and conduction.

Romanes found in Acalephae that coordination
ceases when all direct connection between the nervous
centres has been interrupted by radial incisions in the
umbrella, the various sectors no longer contracting
simultaneously. The same thing results in Hydrome-
dusse, if conduction through the nerve-ring is inter-
rupted. In such cases, the radial incision must reach
well toward the centre of the bell. If, however, such
incisions are made in the umbrella without injuring

EXPERIMENTS ON MEDUSJE

the margin and the nerve-ring, no disturbance of
coordination ensues. It seems that the continuity of
the structures located in the marginal portions of the
umbrella is necessary for the coordinated activity.
Now how does it happen that so long as the continuity
is preserved all the elements act synchronically, while
the synchronism disappears if the continuity is inter-
rupted?^ In order to answer this question, we must
turn our attention to an organ which shows the phe-
nomena of coordinated rhythmical activity in a strik-
ing manner — namely, the heart. If the heart of a frog
be divided into several pieces, they will all be rhythm-
ically active, but the number of contractions will vary
in the different pieces. The sinus venosus beats most
rapidly, and the number of its contractions in a unit
of time equals that of the heart before it was divided.
Thus we see that the whole heart beats in the rhythm
of the part that has the maximum nuTuber of contrac-
tions per m^inute. From this we must assume that the
coordination of the heart's activity is due to the fact
that the part which contracts most frequently, forces
the other parts to contract in the same rhythm. They
will be forced to do this if the activity of the sinus
venosus acts as a stimulus upon the other parts. A
centre of coordination is therefore entirely unneces-
sary.

Porter succeeded by an ingenious method in causing

' It should be emphasised that incisions through the margin alone do not
interfere with coSrdination in Gonionemus, but that it is necessary to continue
the incisions to the centre of the swimming-bell. But even under such circum-
stances the animal may still contract in a coordinated way.

26 COMPARATIVE PHYSIOLOGY OF THE BRAIN

strips of a mammalian heart to beat. He also draws
from his observations the conclusion that there
is no reason for assuming the existence of a centre of
a coordination (6). In Medusae, also, a synchronical
contraction of all the parts takes place if the stimulus
from the portion first active can travel rapidly enough
to the rest of the margin. This is only possible when
the margin is uninjured. It is evident, however, that
the neighbouring tissue as well as the nerve-ring is in-
volved, because the radial incision must reach well
toward the centre of the bell if we wish to stop the
coordination. In this case the wave of stimulation
must pass around the incisions, a process which in-
volves so much time that the separate parts are able
to contract independently, and the synchronism is
lost. In injured or dying Medusae, where the contact
of the cells is less close, uncoordinated, rhythmical
activity occurs.

In order to test this idea further, I proposed to
Dr. Hargitt, who was w^orking in my laboratory, that
he attempt to graft two Hydromedusae, and observe
whether they continue to contract synchronically or
independently after healing. For this purpose it was
necessary to remove the margin of the Medusae.
Two of them were then placed with their wounded
surfaces in contact, and kept in this position. Figure
5 shows two Gonionemi grafted in this way. They
grew together along the entire line of contact with the
exception of a small part at O. New tentacles would
probably have developed there in time had we not

EXPERIMENTS ON MEDUSA

killed the animals in order to preserve them. In
other experiments, the two animals did not heal
together so completely. It happened in the case
where the animals had grown
together most completely, as
represented in Figure 5, that
they contracted synchronically
like one animal two days after
the operation. The animals,
on the other hand, that had not
grown together to such an ex-
tent did not contract synchro-
nically. I believe that if one
could succeed in healing two
hearts together completely,
they would also beat synchro-
nically.

The assumption of a ** centre
of coordination " situated in
the ganglia of the margin of a
Medusa thus becomes un-
necessary. In the frog's heart,

the sinus venosus beats faster than the auricle, ven-
tricle, and bulbus aortae. Hence, each contraction of
the sinus venosus acts as a stimulus, which causes a
contraction of the auricles, and the contraction of the
latter is the stimulus which causes the contraction of
the ventricle and bulbus aortae. It would follow from
this that if we could cause the bulbus aortae in the
frog's heart to beat as fast as the sinus venosus we

Fig. 5.

Dr. Hargitt's Ex-
periment.

Two Gonionemi grafted to-
gether. Two days after the
operation synchronous con-
tractions of both animals
were observed.

28 COMPARATIVE PHYSIOLOGY OF THE BRAIN

might see a reversal of the heart-beat. Nature has
made this experiment for us on a large scale in the
Ascidian's heart (Fig. 6).

The latter has the peculiarity that the waves of

contraction do not
spread out con-
stantly in one di-
rection, as in the
hearts of other an-
imals, but perist-

^ ^ _^ ^ „ altic and antiperi-

FiG. 6. Diagram of the Ascidian Heart. . ^

T .v A J- u _ . .• r .• staltic waves of

In the Ascidian heart, contractions occur for a time

in the direction from a to b, and then from b to Contraction alter-

a. If the heart be cut open at c, the left half nate in it If for
contracts only in the direction from a to f , the , . ,

right half only in the direction from b \.o c. example, it haS

contracted five
hundred times in succession from left to right, sending
the blood to the right, this activity is followed by per-
haps three hundred pulsations from right to left,
which cause the blood to flow through the blood-
vessels in the opposite direction. These contractions
are followed again by a large number of pulsations
from left to right, etc. Mr. Lingle made the follow-
ing experiments on the Ascidian's heart at Wood's
Holl in 1892. \{ a b (Fig. 6) be an Ascidian's heart
and it be divided at ^, both pieces, a c^ and b c, con-
tract uninterruptedly in a constant direction, the
former in the direction from a to c, and the latter in
the direction from b to c. Mr. Lingle found, further-
more, that the source of the automatic activity is

EXPERIMENTS ON MEDUSA 29

confined to two small regions {a and b, Fig. 6) which
correspond to the sinus venosus and the bulbus aortae
of the frog's heart. When we excise these two pieces
from the heart they continue to beat without inter-
ruption, while the long part between the two pieces
no longer pulsates (in sea-water at least). These ex-
periments, it seems to me, leave no room for doubt
that the change in the direction of the contraction in
the Ascidian's heart is determined by each of the two
ends getting the upper hand alternately, and forcing
the other centre to act in its rhythm for a time. This
*' getting the upper hand " might possibly mean no-
thing more than that one end gains the time in which
to send off a wave of contraction before the other
end begins to contract. For this it is only necessary
that a single heart-beat of the leading end be delayed
or fail entirely, a phenomenon that also appears oc-
casionally in the human heart. In this way the other
end of the heart gains time in which to send out a
wave of contraction, and its automatic activity will
continue to be the stimulus for the activity of the
first end until a delay occurs in one beat or until one
beat is skipped, thus allowing the first end time again
to become automatically active, and so on.

Last year I asked the members of the class in gen-
eral physiology at Wood's Holl to find out whether
the latter view was correct. Their observations were
as follows : Suppose at a certain time a to be the
active and b the passive end of the heart. After a
short time a begins to beat more slowly or ceases to

30 COMPARATIVE PHYSIOLOGY OF THE BRAIN

beat altogether. During the pause, the end b suc-
ceeds in sending out a wave of contraction which
reaches a before it has had time to send out a wave
of its own. One sees occasionally at the time of a
reversal that at first both ends send out contraction-
waves which may meet in the middle of the heart.
At the next heart-beat, the end which is about to stop
delays the sending out of the wave a little more, and
at the next heart-beat the wave starting from the
other end can pass over the whole heart without being
blocked.

Hence the coordination of movements In Medusae
(or in the heart) is not due to a hypothetical centre
of coordination situated in the ganglion-cells, but to
the fact that the element which is first active acts
as a stimulus upon its next element, and so on.

3. It may be shown that even more specialised
forms of coordination do not depend upon the pre-
sence or interference of ganglia. When the back of
a frog Is touched with acetic acid, the frog wipes off
the acid with Its foot. If one leg Is tied. It uses the
other for this purpose. The turtle acts in a similar
manner when acetic acid is applied to the back of its
shell. It cannot reach the stimulated spot, but the
legs move dorsally under the shell as far as possible
towards it. Physiology has contented Itself In regard
to these phenomena by pointing to the complicated
nature and Impenetrable structural secrets of the
central nervous system. Yet the same reactions oc-
cur In a Hydromedusa, in which case the term

EXPERIMENTS ON MEDUSAE 31

*' central nervous system" has only a conventional
significance. Romanes found that if we stimulate a
spot a (Fig. 7) on the concave side of the umbrella
of a Tiaropsis indicans with a needle, the manubrium

is broug^ht to the

stimulated spot y^ ^\^ ..••'

(Fig. 7), as though / ^^ -^

the animal wished / /^ | \

to remove the / / I f \ I

stimulating object / / \ \ \ /

(2). This move- (/L^s:^====^^^^^^v?^ 1/

ment takes place 1^^^-^^^ ^^3f'

as follows: A "'''''liil^^

bending of the

m a n u b r i u m as Tr ,v • . *u • • .• 1 * ^ *a.

If the point a on the margin is stimulated, the
well as 01 the bell manubrium is brought to the stimulated spot,
ensues in that mer- somewhat as a decapitated frog tries to wipe off
... ft * ^op of acetic acid with its foot.

idian of the um-
brella which passes through the stimulated point a.
It seems as though all the muscle-fibres cooperated
in bringing the manubrium to the stimulated spot.
The central nervous system has nothing to do with
this reaction, for Romanes found that it continued
after excision of the whole margin with the nerve-
ring. On the other hand, if we make an incision
in the umbrella parallel to the margin and stimulate
a spot below the line of incision, movements of
the manubrium, although not pronounced ones, ap-
pear in the direction of the quadrant where the
stimulated spot is located, but an exact localisation

Fig. 7. Localising Reflex in Tiaropsis
Indicans.

32 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Fig. 8. Diagram for Explain-
ing THE Localising Reflex in
Medusae. (See text.)

is impossible. Romanes concludes from this that
there are radial lines of differentiated tissue pass-
ing through all parts of the bell and that it is
their function to transmit impressions to the manu-
brium. He assumes that
this tissue is of a nervous
character. I believe that
the whole phenomenon can
be explained without the as-
sumption of a special differ-
entiation of nervous tissue
in radial directions. It seems
to me that the following as-
sumption is possible : Every localised stimulus leads
to an increase in the muscular tension on all sides,
which is most intense near the stimulated spot. Now
if we decompose each of the lines of increase of tension
{aa' ab' ad ad' ae\ Fig. 8) radiating from the stimul-
ated spot, into a meridional component aa' dd' bb\ etc.,
and an equatorial component, it is evident that the lat-
ter can have no influence on the manubrium. Only
the meridional components can have an influence, and
of these the one passing through the stimulated spot is
the largest. This fact must necessarily cause a bend-
ing of the manubrium toward the stimulated spot.
It also shows why an incision parallel to the mar-
gin of the umbrella makes an exact localisation impos-
sible and only allows uncertain movements towards
the stimulated quadrant.

I hardly believe that the mechanisms for the

EXPERMIENTS ON MEDUSA 33

analogous reflex in a frog or turtle are of a more com-
plicated character. Nature works with very simple
tools. The tool employed in the reflex of localisa-
tion is the curvature produced by stimulation, — con-
tact, for instance. We meet with this in its simplest
form in plants, in which the side that comes in con-
tact with a solid body becomes concave. Plants cer-
tainly possess no central nervous system containing
mysterious reflex structures. In their case, irritabil-
ity and conductibility suffice as an explanation. In
Medusae the method appears more complicated only
in so far as in them the contractile tissue is real mus-
cle-fibre. . In the frog, the only further complication
is the fact that the conduction takes place through a
special kind of tissue — namely, nerve-tissue. In its
first anlage, this central nervous system is of a very
simple segmental character. I believe that the cent-
ral nervous system preserves this simple character
better than any other tissue. The muscles undergo
considerable displacement during the development,
but the changes occurring in the central nervous sys-
^tem by no means equal those occurring in the mus-
clar system.

It seems thus possible to explain the above-men-

:ioned phenomena of coordination in Medusae by

leans of the simple facts of irritability and conduct-

[ivity without attributing any other functions to the

;anglion-cell except those which occur in all conduct-

[ing protoplasm.

COMPARATIVE PHYSIOLOGY OF THE BRAIN

Bibliography

1. O, u. R. Hertwig. Das Nervensystem und die Sinnesorgane
der Medusen.

2. Romanes, G. J. Jellyfish^ Starfish and Sea Urchins.
The International Science Series, 1893.

3. LoEB, J. On the Different Effects of Ions upon Myogenic and
Neurogenic Rhythmical Contractions^ etc., American Journal of
Physiology^ vol. iii., 1900.

4. LoEB, J. Ueber lonen, welche rhythmische Zuckungen der
Skelettmuskeln hervorrufen. Festschrift fur Fick. Braunschweig,
1899.

5. Quincke. Ueder periodische Ausbreitung an Fliissigkeits-
oberfldchen^ etc. Sitzungsberichte der Berliner Akademie der Wis-
sensch.y 1888, ii., S. 791.

6. Porter, W. T. The Coordination of the Ventricle, The
American Journal of Physiology ^ vol. ii., 1899.

CHAPTER III

THE CENTRAL NERVOUS SYSTEM OF ASCIDIANS
AND ITS SIGNIFICANCE IN THE MECHANISM
OF REFLEXES

I. If we wished to observe the order of the natural
system in this book, we should not let the Ascidians
follow the Medusae. We consider it more profitable,
however, to discuss simple cases before taking up the
more complicated ones. Having reached the con-
clusion, at the end of the preceding chapter, that the
spontaneous coordinated activities in Medusae are not
due to specific morphological structures of the gan-
glion-cells, we will now attempt to find out whether
the reflex actions of animals depend upon the struct-
ure of the central nervous system or of the peripheral
parts. In Ascidians the central nervous system con-
sists of a single ganglion {d, Fig. 9). This ganglion
is situated between the oral and aboral tubes {a and
b, Fig. 9).

Ciona intestinalis (Fig. 9), a large, transparent
Ascidian, possesses a very characteristic reflex. If
either the oral or aboral opening be touched, both
openings close, and the whole animal contracts so

^6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

that it becomes small and round. This reflex is de-
termined by two groups of muscles, first by ring-
muscles in the oral and aboral openings, second by

longitudinal muscles,
which run lengthwise
through the animal. By
the contraction of these
muscles the animal is
protected from the en-
trance of foreign bodies
into the body cavity.
This reaction is a typ-
ical reflex act, and is
eminently purposeful.
According to the pre-
vailing ideas concern-
ing the decisive role
that the ganglion plays
in reflexes, the pro-
cedure is as follows :
If the oral or aboral
opening be touched, the stimulation is conducted
through the peripheral nerves to the ganglion, where
a mysterious reflex mechanism is brought into play,
which gives the muscles the command to contract in
a manner corresponding to the nature of the stimulus.
Ferrier, for instance, in his text-book, mentions the
one ganglion of the Ascidians as illustrative of the
significance of the ganglion in reflexes.

I removed the ganglion from a number of Cionse.

Fig. 9. CioNA Intestinalis.

n oral, b aboral opening ; r , foot, d, location of
ganglion.

EXPERIMENTS ON ASCIDIANS 37

For some time after the operation, in most cases for
about twenty-four hours, the animals remained con-
tracted. At the end of this period they began to re-
lax again. To my great surprise, I found that the
typical reflex continued. If we let a drop of water
fall on such an animal, the typical reflex act is pro-
duced just as in the normal animal. Hence the reflex
cannot be determined by specific structures of the
ganglion. But what does determine the reflexes, and
what is the function of the ganglion ?

The answer to the first question must be that the
reflex is determined by the structure and arrange-
ment of the peripheral parts, especially the muscles.
The mechanical stimulus throws the muscles directly
into activity, and the stimulation is transmitted from
muscle-element to muscle-element directly, as in the
heart or the ureter. But is the central nervous system
superfluous in this animal ? We get the answer to
this question if we determine the threshold of stimul-
ation. The threshold of stimulation for producing
this reflex is higher in animals which have been
operated upon than in normal animals. As the
source of the stimulus, I used the kinetic energy of
drops of water, which fell from a pipette upon the
animal. Since the weight of the falling drop in the
pipette is always the same, the minimum of the height
from which a falling drop can produce a contraction is
a convenient measure of the irritability ; the latter is
of course equal to the reciprocal value of the thresh-
old of stimulation. In one case there were in an

38 COMPARATIVE PHYSIOLOGY OF THE BRAIN

aquarium (equally near the surface), a Ciona freshly
operated upon and a normal Clona. The minimum
height from which a contraction could be produced
was as follows for the normal animal (a) and the
animal operated upon {b) :

a (normal) b (operated)

8 mm 65 mm

4 mm 75 mm

10 m.m 80 mm

80 mm
In two other animals used for the experiment I
obtained the following values :

a (normal) b (operated)

6 mm. 22 m.m.

8 m^m 20 mm.

It seems to me that the difference in the irritability
arises from the fact that in the normal Ascidian the
stimulation is conducted through the nerves and the
ganglion, in which case less energy is required. In
the Ascidian operated upon, however, the muscles are
stimulated directly, and the conduction of the stim-
ulation probably takes place from muscle-cell to
muscle-cell, just as in the heart. We know, more-
over, that the direct irritability of muscle-fibres is not
so great as that of the nerves. Hence the nerves and
the ganglion only play the part of a more sensitive and
quicker conductor for the stimulus (i).

2. It may seem as though no conclusions could
be drawn from these cases in regard to the '' reflex

EXPERIMENTS ON ASCIDIANS 39

centres " of higher animals. It is frequently stated
that in higher animals the ganglia have assumed func-
tions which in lower animals can be performed by the
peripheral organs. It is similarly stated that the higher
the animal ranks in the natural system, the more the
functions ''migrate" toward the cerebral hemispheres.
But how such an upward migration of functions is
conceivable, none of these authors attempt to explain.
It can easily be shown, however, that conditions are
the same in higher and lower animals. We must only
be careful to homologise a lower form with a single
organ or segment of a higher animal. When the in-
tensity of the light is suddenly increased, the pupil of
our eye becomes narrower. The sphincter of the iris
contracts, and the rays of light are excluded just as
foreign bodies are shut out by the contraction of the
sphincters in the Ascidians. In the eye, just as in the
Ascidian, we have to deal with a typical reflex act.
The increased intensity of the light stimulates the
retina. The stimulation passes through the optic
nerve to its centres, and is carried from there by
means of the oculomotorius nerve to the sphincter of
the iris, which contracts. It would nevertheless be
wrong to assume that the centre for the pupillary
reflex plays any other part in this process than that of
a protoplasmic connection between the retina and the
iris. It has been shown by Arnold, and later by
Brown-Sequard and Budge, that even in the excised
iris the pupil still contracts when the light strikes the
former. I myself have often observed in sharks,

40 COMPARATIVE PHYSIOLOGY OF THE BRAIN

whose brain I had removed, that Hght caused the
pupil to contract several hours after death, when
signs of decomposition had already begun to appear.
Steinach has proved that in this case the muscle-
elements in the iris are stimulated directly by the light
(3). This reflex is therefore determined by the mus-
cles of the iris, and the nervous connections serve
only as quicker and more sensitive conductors. Thus
we see that the eyeball behaves toward light just as
the Ascidian behaves toward mechanical stimuli.

Some physiologists seem to doubt that the muscles
can be stimulated directly by light without the inter-
vention of the ganglion-cells. But we know that
phenomena of contraction are also produced by the
light in the unicellular swarmspores of algae, which
certainly contain no ganglia. Furthermore, no one
doubts that muscles without ganglion-cells can also be
stimulated chemically or mechanically. Why should
there not also be muscle-fibres that can be stimulated
directly by light ? There is no reason for assuming
that all muscles must behave exactly like the muscles
of the frog's leg, simply because the experiments on
it have by chance furnished the prevailing views con-
cerning muscles.

The reader may believe that the pupillary reflex is
an exceptional case, but this is not true. Defaecation
and urination in higher animals may be considered as
reflex phenomena of the spinal cord. The pressure of
the faeces or of the urine acts as a stimulus, which
affects the centres for the activity of the muscles of

EXPERIMENTS ON ASCIDIANS 41

these organs, and this stimulation is said to cause the
contracted sphincters to relax. Goltz and Ewald
have found, however, that after extirpation of the en-
tire spinal cord up to the cervical part, defaecation and
urination still occur normally (4). Only for a time after
the operation the sphincters are relaxed. Later on
everything again becomes normal. These phenomena
probably belong to the same class as the one already
described in the Ascidian. The processes in the
normal evacuations of the bladder and rectum are not
determined by the morphological structure of the so-
called reflex centre, but by the muscles of the bladder
and of the rectum themselves. The spinal cord
serves only as a more sensitive and quicker conductor
for the stimulus. Goltz and Ewald are inclined, it is
true, to assume that, after all, ganglion-cells or un-
known nervous structures determine these results.
But the fact that the muscles of the skeleton can be
caused to contract rhythmically when put in the right
solution, makes this assumption unnecessary ; more-
over, the facts of comparative physiology must also be
taken into consideration. The Actinia mesembryan-
themum of the East Sea and the Mediterranean per-
haps show fewer differences morphologically than the
sphincter ani and the gastrocnemius, and yet the
Actinia mesembryanthemum of the Mediterranean
shows a form of irritability which the Actinian of the
same name from the East Sea does not show, namely,
negative geotropism. I mention this illustration, to
which many others might be added, in order to show

42 COMPARATIVE PHYSIOLOGY OF THE BRAIN

that forms which are morphologically alike need not
necessarily be alike in all their reactions. Experiments
on fermentation show that a small stereochemical dif-
ference of a carbohydrate or proteid can produce an
entirely different physiological effect.

The possibility, of course, remains that scattered
ganglion-cells exist in Ascidians under the epidermis
just as in Medusae. Mr. Hunter, who has studied the
nervous system of Ascidians, informs us that he has
found cells in certain places under the epidermis of As-
cidians which he believes to be ganglion-cells. But
after all that has been said about the scattered gan-
glion-cells in Hydromedusae (see page 19) and their
role in rhythmical contractions, it is not necessary to
consider the importance of scattered ganglion-cells for
reflexes. Schaper has recently made an observation
which makes it seem as though in the young larvae of
Amphibians conditions similar to those in Ascidians
exist. He amputated the brain of the larva of a frog
during the first days of development, and saw that the
animal was still able to move spontaneously seven days
after the operation. When sections of the animal were
made, it was found that the spinal cord had also per-
ished (2). This observation should be repeated and
enlarged upon. It is quite possible that during the
first days of development a direct transmission of the
waves of stimulation may take place from the skin to
the muscles in the larva of the frog, without the in-
tervention of the central nervous system, as happens
in the Ascidians.

EXPERIMENTS ON ASCIDIANS 43

3. The objection might now be raised that the
bladder and rectum are minor organs of the body.
But what has been said above concerning them
also holds good for larger and more important
groups of organs, namely for the blood-vessels.
These are able to adapt their width to external con-
ditions ; the vessels of the skin become dilated when
a loss of heat is desirable, and they contract in the
cold when the loss of heat should be reduced. It is
assumed that the mechanisms for these purposeful
reflexes are contained in the central nervous system.
Goltz and Ewald (4) have found, however, that dogs
which have lost the spinal cord almost up to the me-
dulla oblongata live for years. This alone proves
that the blood-vessels can adapt themselves to the
external temperature, independently of the central
nervous system. Goltz had already proved that the
blood-vessels regain their tonus if all the nerves of a
limb be severed, the limb being connected with the
animal only by means of the blood-vessels. The
same thing occurs after extirpation of the spinal cord.
The temperature of the hind-paws of animals whose
spinal cord has been destroyed up to the thoracic
part becomes normal again after the operation — that
IS to say, the hind-paws have the same temperature
as the fore-paws which remain connected with the
central nervous system. If we hold the hand in
snow for a time, we observe as a local after-effect a
relaxation of the muscles of the blood-vessels and an
increase in the temperature of the hand. Goltz and

44 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Ewald were able to show that the same phenomena
may also be observed when the hind-legs of dogs
whose spinal cord has been destroyed are packed for
a time in snow.

From the standpoint of human physiology these
results seem strange, but from that of comparative
physiology they are readily understood. The various
reactions of plants to external stimuli are just as pur-
poseful as those of animals. Why should it not be
possible, then, for single organs and tissues of higher
animals to react purposefully to external stimuli, and
is there any reason why the purposeful character of a
reaction should be dependent upon the structure of
the central nervous system ?

We have been able to rid ourselves of erroneous
views concerning the significance of the ganglia of
the central nervous system in higher animals through
the help of the Ascidians ; they also help us further to
determine the true role of the nervous system. Al-
though the dogs experimented upon by Goltz and
Ewald were able to adapt the width of their blood-
vessels to the variations of temperature, it was neces-
sary to shield them much more carefully from sudden
changes of temperature than is necessary in the case
of normal animals. The threshold of stimulation was
raised and probably the rapidity of the conduction
decreased. For this reason, dogs whose spinal cord
is destroyed are no longer fit to live out-of-doors.
As regards regulation of temperature, they are like
an intoxicated person, and would perish in the cold

EXPERIMENTS ON ASCIDIANS 45

much sooner than a normal animal. Hence the
nervous system does not contain any regulating me-
chanisms, but it serves as a quicker conductor, and
allows the peripheral organs to work with greater
precision.

4. Bethe has recently made a difficult experiment
on Carcinus mcBuas, which, however, was successful
in only two cases. If this experiment is correct, it
proves that, in the conduction of a reflex in the cent-
ral nervous system, the process of conduction does
not of necessity pass through the ganglion-cell itself
(5). An anatomical observation caused Bethe to
perforni this operation. ** Almost all the ganglion-
cells of Carcinus are unipolar, and often the axis-cyl-
inder of the cell runs for a long distance before it
gives off the first dendrites and sends out the peri-
pheral fibre. It seemed very strange to me that a
stimulus entering through the sensory nerves into the
central organ should go through the dendrites to the
far-distant motor-ganglion cells, and travel the great
part of the same path before entering the peripheral
motor fibre, instead of going directly to the motor
fibre. It was easy to decide this question by sep-
arating the ganglion-cells with their axis-cylinder
process from the motor neurons without injuring the
neuropiles. If the ganglion-cell were absolutely es-
sential for the reflex, the muscles involved should
become paralysed immediately after the operation ; if
it were not essential, no paralysis should occur, at least
for some time, and the stimulus could go across

46 COMPARATIVE PHYSIOLOGY OF THE BRAIN

directly from the dendrites to the peripheral fibre." It
was possible to perform the operation in Carcinus
on the ganglion-cells which innervate the muscles of
the second antenna. The cutting of the peripheral
nerves {Antennarius secundus) that go to these gan-
glion-cells immediately causes a complete paralysis of
the antennae, a proof that the fibres of these nerves
are the only conductors of the stimulus which can
call forth a reflex movement of these antennae. But
when Bethe removed the ganglion-cells, without in-
juring the neuropile of the second antenna, ''the
second antenna retained its tonus and its reflex irrit-
ability. It does not hang down limp, but remains
stiff and in the normal position. When stimulated,
it is withdrawn, but is stretched out again when the
stimulation ceases. From this it is evident that the
ganglion-cells are not necessary for reflexes. The re-
flex arc either does not pass through the ganglion-
cells or does not need to pass through them. It is
further apparent that the ganglion-cell has nothing to
do with the tonus of the muscles, and that the per-
manent influence which the central nervous system
exercises upon the tension of the muscles is not pro-
duced in the ganglion-cells (6)."

This experiment, even if it be correct, adds no-
thing of importance to our conclusions. If the reflex
arc acts only as a quick protoplasmic conductor, the
question whether the stimulus has to pass through the
ganglion itself or not becomes of secondary import-
ance.

EXPERIMENTS ON ASCIDIANS 47

Bibliography.

1. LoEB, J. Untersuchungen zur physiologischen Morphologie
der Thiere. II. Wiirzburg, 1892, S. 37.

2. ScHAPER, A. Experimentelle Studien an Amphibienlarven.
Archiv fiir Entwicklungsmechanik^ Bd. vi., 1898.

3. Steinach, E. Untersuchungen zur vergleichenden Physiolo-
gie der Iris, Pfluger's Archiv, Bd. lii., 1892.

4. GoLTZ und Ewald. Der Hund mit verkUrztem RUcken-
mark. Pflilgers Arch.^ Bd. Ixiii., 1896.

5. Bethe, a. Das Centralnervensystem von Carcinus manas.

I. Theil, II. Mittheilung. Archiv f. mikroskop. Anatomie und
EntwicklungsgeschichtCy Bd. 1., 1897.

6. Bethe, A. Das Centralnervensystem von Carcinus mcenas.

II. Theil. Arch. f. mikroskop. Anatomie und Entwicklungs-
geschichte, Bd. li., 1898.

CHAPTER IV

EXPERIMENTS ON ACTINIANS

I. The two preceding chapters have furnished
proof of the fact that the phenomena of purposeful
reflex action, of spontaneity, and of coordination are
determined, not by specific characters of the ganghon-
cells, but by general peculiarities common to all pro-
toplasm. These peculiarities are irritability and the
power of conducting stimuli, both of which will find
their explanation in the physics of colloidal sub-
stances.

In this chapter we wish to put the foregoing con-
clusions to a test by showing that a group of animals
without any true central nervous system are able to
show reactions complex as those in higher animals.
Without such a parallel we should be more than
ready, in the case of higher animals, to attribute such
reactions to the specific structure of the ganglia or
the ganglion-cells.

We cannot speak of a central nervous system in
Actinians in the same sense as in Ascidians. Under
the ectoderm there are elements which are interpreted
by some authors as ganglion-cells and nerve-fibres.

EXPERIMENTS ON ACTINIANS 49

The unreliability of this interpretation is apparent,
however, from the fact that Claus considers it uncer-
tain. He mentions the possibility of a conduction of
stimuli as one of the conditions that speak for the ex-
istence of a nervous system in Actinians. But a con-

FiG. 10. The Ability of the Actinians to Discriminate.

The tentacles press the meat a into the mouth, while they drop the water-soaked
paper b.

duction of stimuli also occurs in plants. During the
year 1888 in Kiel, and 1889-90 in Naples, I made in-
vestigations on the reactions of Actinians, which show
how little reason we have for concluding that compli-
cated reactions need depend upon similarly compli-
cated reflex centres (i). It is very obvious from these
experiments that the structure and irritability of the
peripheral organs determine the reactions. We will
begin with the description of experiments on the Ac-
tmia equina {mesembryanthemurn) of the East Sea.

50 COMPARATIVE PHYSIOLOGY OF THE BRAIN

If a wad of paper soaked In sea-water be placed on
the mouth of one of these Actinians it Is refused, while
a piece of crab-meat, which to us does not differ in
taste from the wad of paper, is usually accepted with-
out delay. I tied one end of a short thread around a

Fig. II. Continuation of the Experiment in Fig. id.

paper wad and the other end around a piece of meat,
and threw both on the outstretched tentacles of a
starved Actinian. The tentacles that came in contact
with the meat {a, Fig. lo) reacted at once by bend-
ing in such a way as to bring the meat to the mouth,
while the tentacles that were in contact with the pa-
per did not react. I withdrew the thread and placed
it on the oral disc in such a way that the paper rested
on the tentacles where the meat had rested before,
and vice versa. The meat was then drawn into the
mouth and the string with It, but the paper remained
outside the oral opening (Fig. 1 1). During the next
twenty-four hours no change took place ; later on, the
thread was ejected without the meat. The latter was

EXPERIMENTS ON ACTINIANS 51

probably digested. I have often repeated the experi-
ment, always obtaining the same result, except that
occasionally the string was ejected sooner, in which
case the meat remained on the string, partially or en-
tirely undigested. These phenomena have the same
explanation as the behaviour of insect-eating plants.
The chemical substances diffusing from the meat, to-
gether with the tactile stimuli exerted by it, cause a
bending of the tentacles that are touched in such a
way that they become concave and carry the meat to-
ward the oral opening. The contact of the meat
with the mouth causes the sphincter of the oral open-
ing to relax ; the pressure of the tentacles, together
with the activity of the oral disc, then pushes the meat
into the interior of the digestive tract. But if these
specific chemical stimuli are wanting, if we give the
animal, for instance, water-soaked filter-paper, the
contractions of those muscles which carry the tenta-
cles to the mouth are not produced. The tentacles
remain relaxed or relax still more under the stimulus,
and this fact, together with the ciliary movement,
causes the paper wad to fall off.

2. It is said that the nerve-elements are much more
numerous in the vicinity of the mouth than in any other
part of the animal. One might think that this con-
centration of nerve-elements determined the reflex
mechanism for these reactions. For this reason, I
have made use of results obtained while carrying
on investigations concerning heteromorphosis. I had
found that in an Actinian of the Mediterranean,

52 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Cerianthus membranaceus, new tentacles could be pro-
duced by a lateral incision in the body of the animal.
But in some of these cases no mouth is formed. Fig.

12 shows such an ani-
mal ; a is the normal,
b the new head. If the
incision was very small,
only single tentacles
were formed, without
the oral disc. These
new tentacles behave
toward food exactly
like the tentacles of the
old mouth. If we offer
such a new head a
piece of meat, the tent-
acles seize and press it
against the centre of
the oral disc, where the
mouth should be. After pressing in vain for some min-
utes the tentacles relax and the meat falls off. This
experiment could be repeated for months, in fact as
long as I observed the animal (2). In other cases the
second head was so near the old one that it was easy
to stimulate the tentacles of both simultaneously with
the same piece of meat In this case a fight arose
between the two tentacle systems, each attempting to
draw the meat toward its own oral disc. Parker has
lately shown that even a single tentacle, after being
severed from the animal, grasps a piece of meat and

Fig. 12. AcTiNiAN (Cerianthus) with
A Normal Head {a) and an Arti-
ficially Produced Head {b).

Although the latter has no oral opening the ten-
tacles carry the meat to the place where the
mouth ought to be.

EXPERIMENTS ON ACTINIANS 53

bends with it toward the place where, in relation to
itself, the mouth ought to be (3).

If we look at these facts without prejudice, we
must conclude that the reaction of the tentacles is
determined only by the irritability of the tentacle-
elements themselves, and by the arrangement of their
contractile elements. The following observations
may also be considered in support of this conclusion.

3. If an Actinia equina be divided transversely, the
oral piece, which we will call the head-piece, has the
normal head, with mouth and tentacles on its oral end ;
on its aboral end the body-cavity is open to the
exterior, and food may pass through the opening in
either direction. The old mouth of a head-piece was
as particular as usual in regard to the selection of its
food, while the aboral end readily swallowed pieces of
paper. The old mouth often refused meat, but the
aboral mouth was almost always ready to accept it,'
even when it would refuse paper.
I I laid a piece of an Actinian that took food in at
both ends on its side, and tried to find out whether
both mouths would take food simultaneously. I first
placed a piece of meat on the aboral mouth, in order
to cause it to open. As soon as this happened and
the meat was being taken into the mouth I offered
the oral mouth also a piece, and this was likewise
accepted. The act of swallowing in the other mouth
was interrupted at once by the contraction of the ring-
muscles. After a few moments, however, when the
meat in the oral mouth had been swallowed, the

54 COMPARATIVE PHYSIOLOGY OF THE BRAIN

muscles of the aboral end relaxed and the meat taken
in before by this mouth fell out. When I fed the
mouths in succession, the mouth that was fed first
ejected the food as soon as the other began to eat.
It is obvious from this that a peristaltic wave is
started from the end which takes up food.

Thus far we have considered only the head-piece.
If we turn our attention now to the foot-piece, we
find that on the oral end a new oral disc with tentacles
soon begins to form. Before this has occurred, how-
ever, the mouth takes pieces of meat and swallows
them. It seemed to me as though this new mouth,
even before the regeneration of the oral disc, re-
sembled the normal mouth more than the aboral
mouth in the head-piece, for it did not accept paper
wads and grains of sand, while It swallowed meat
well.

4. In the foot of the Actinians the contact-irrita-
bility is of special interest. The foot of a normal
Actinia equina attaches itself to the surface of solid
bodies. The character of the surface is of great im-
portance for producing these processes of attachment.
If it finds no other body, the Actinian attaches itself
to the glass of the aquarium, and glides about on it.
If, however, the shell of a Mytilus is placed in the
aquarium and the animal comes in contact with it
while moving about, it immediately attaches itself to
the shell, and remains there, whether the shell is
empty or inhabited. The surface of an ulva leaf has
the same effect. While the animal upon contact with

I J

S-5

•«l

'^ s

Z
^

6 '^

^•s

* c^

» «

«««

'??

T3 O

ffi

2"H

. ^

« 73

'^ CM

.S"5

JJ >

■<

•S<a

u ^

(1^

^ t;

P^-5

5 >

«^

o ^

> ^

Irt^

B a

V V

ll<

■<

?l^

pti

«3

ube sho
the bot
rientatic

1^ <N

•H

s 6

•S 8 **

"2 -o **

§^3

111

ot,

•ill

56 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the ulva leaf will at any time leave the glass and
attach itself to the leaf, the reverse is not liable to
happen. This contact-irritability of the foot does not
change if the head or the greater oral part of the ani-
mal be amputated. The mechanisms for the discharge
of these reactions must, therefore, be located in the
foot itself and not in the ganglion-cells of the oral disc.
5. In higher animals we recognise a tendency to give
the body a certain orientation in space. We usually
call such an orientation in a higher animal its position
of equilibrium. Certain Actinians also show such phe-
nomena. If we put a Cerianthus Into a test-tube
filled with sea-water, and place the glass so that the
head of the animal is down, the foot up, and the long-
itudinal axis vertical, the tip of the foot will begin
after a few moments to bend downward vertically.
In Fig. 13 the course of such an experiment is given
from life. Some minutes before 12 o'clock the ani-
mal was placed in the test-tube in the manner de-
scribed above. At 12 o'clock the foot had begun to
bend downward (Fig. 13, ci)\ in the next thirteen
minutes the bending toward the head had progressed
{U) ; five minutes later the foot had reached the bot-
tom of the tube {c). The bending progressed steadily
to new elements lying near the head ; and since the
foot now stood upon the bottom of the tube, the
farther advance of the curvature toward the head
resulted in the lifting of the latter (Fig. 13, ^and ^),
whereupon the animal raised itself bodily, and at i
o'clock had the position /. The process of righting

EXPERIMENTS ON ACTINIANS

required one hour. The animal remained in this
position for two days, and then crawled out of the
glass.

In analysing the conditions that determine the
righting of the Cerianthus in this case, two circum-

FiG. 14. Cerianthus Regaining its Normal Orientation.

It was placed on the net horizontally, and within half an hour had regained its normal
vertical position, by pushing itself through the meshes of the net.

stances must be taken into consideration, namely,
gravitation and the contact-stimuli. It can be easily
shown that gravitation alone is able to produce the
above-mentioned reaction of the Cerianthus. A wire
net, whose meshes are so fine that the body of a Ceri-
anthus can only be drawn through them by force,
is laid horizontally upon a glass standing in the

58 COMPARATIVE PHYSIOLOGY OF THE BRAIN

aquarium. A Cerianthus Is laid on the wire net. After
a few minutes the foot of the animal begins to bend
downward and to work its way through one of the
meshes of the net. In the oral pole no change takes
place except that the tentacles lay themselves close to-
gether, so that they look like a brush whose handle
is formed by the body of the animal. The animal
forces its body farther and farther through the meshes
until it is at last able to keep itself in a vertical posi-
tion as represented in Fig. 14. This orientation can
be reached in half an hour. If we turn the wire net
over as soon as the animal has reached the position
represented in Fig. 14, so that the foot is up, it does
not pull Itself out of the net again, but the foot near
the tip begins to bend downward vertically. The
bending then progresses from element to element of
the body, from the foot toward the head, until the tip
of the foot reaches the wire net, when it again pushes
itself through as far as possible. If the wire net be
turned over again, the process Is repeated. Thus the
animal can be forced, simply with the aid of gravit-
ation, to weave itself in and out of the net. Fig. 15
shows a Cerianthus that has been forced to push itself
through three times in this manner. The drawing
was made from life. In these experiments we have
an example of a geotropic irritability, — in other words,
of positive geotropism. As this kind of irritability is
very common in the roots of plants, it follows that for
the mechanism of these reactions no specific qualities
of the ganglion-cells are necessary. If a transverse

EXPERIMENTS ON AC TIN I AN S

Fig. 15. AcTiNiAN that has been
Forced by Gravitation to Push

ITSELF THROUGH THE NeT THREE

Times (a, b, and c). See text.

incision be made in the middle of a Cerianthus which
almost but not quite separates the two halves, and the
animal be placed immediately after the operation on
a wire net, the foot works
itself into one of the
meshes up as far as the
incision and assumes a
vertical position. The
oral piece, on the con-
trary, from the place of
incision to the head, usu-
ally remains lying hori-
zontally on the net. This
shows that the foot pos-
sesses geotropic irritabil-
ity. But if the Actinian be divided transversely we
see that the head-piece as well as the tail-piece pushes
itself through the meshes, although not so frequently.
While an Actinian that is suspended vertically in a
test-tube or in a mesh of a wire net seldom retains this
position longer than two days, it remains indefinitely
in the sand after burrowing. In addition to gravita-
tion, some other stimulus must hold it there. I be-
lieve that it is the contact-stimulus of the sand. I
called this kind of irritability stereotropism, and have
shown that in a series of animals it determines their
habits. Positive geotropism and positive stereotrop-
ism cause the Cerianthi to burrow in the sand vert-
ically, and the positive stereotropism keeps them
permanently in the burrow.

6o COMPARATIVE PHYSIOLOGY OF THE BRAIN

We see from this that quite complicated reactions
occur in these animals although they do not possess a
central nervous system like that in higher animals.
Were we to come across these same reactions in
higher animals we should be inclined a priori to as-
cribe them to the complicated structure of the central
nervous system. The experiments on Actinians will
perhaps prevent us from drawing such a conclusion
before we have forcible reasons for so doing. A high
degree of complication in the reactions of animals can
be reached where no central nervous system exists, or
where it serves only as a sensitive and quick proto-
plasmic conductor. The cause of complicated reac-
tions lies, therefore, in the irritabilities and structures
of the peripheral organs.

Bibliography.

1. LoEB, J. Untersuchungen zur physiologischen Morphologie
der Thiere^ I., 1891. Wiirzburg, G. Hertz.

2. LoEB, J. Zur Physiologie und Psychologic der Aktinien.
Pflugers Archiv, Bd. lix., 1895.

3. Parker, G. H. The Reactions of Metridium to Food and
Other Substances. Bulletin of the Museum of Comparative Zoology
at Harvard College^ vol. xxix,, 1896.

4. Pollock, W, H. On Indications of the Sense of Smell in
ActinicB. Jour. Linnean Soc, London, vol. xvi., 1882.

5. Nagel, W. Experimentelle sinnesphysiologische Untersuch-
ungen an Coelenteraten. PJlUger's ArchiVy Bd. Ivii., 1894.

CHAPTER V

EXPERIMENTS ON ECHINODERMS

I. The nervous system of the starfish consists, first,
of a central nerve-ring around the mouth (Fig. i6),
and, second, of the peripheral nerves radiating from
this ring into each of the arms.

It is a well-known
fact that if such an ani-
mal be laid on its back
it soon rights itself. In
species like that repre-
sented in Fig. 1 6 the
ambulacral feet found
on the ventral surface
in great numbers ex-
ecute the righting.
These little feet are
muscular tubes, which
end in a plate. By
means of this plate the
foot, like the sucker of
the leech, can cling to solid bodies. If a starfish
be laid on its back, the tube-feet of all the arms

Fig. i6. Mervous System of a
Starfish.

rt, central nerve-ring that surrounds the mouth,
by peripheral nerves of the arms.

62 COMPARATIVE PHYSIOLOGY OF THE BRAIN

are stretched out at once and are moved hither and
thither as if feeHng for something, and soon the
tips of one or more arms turn over and touch the un-
derlying surface with their ventral side (Fig. 17).
The tube-feet of these arms attach themselves to this

Fig. 17. Mechanism of the Turning of a Starfish that has
BEEN Laid on its Back.

The tube-feet of the three arms at the left are pulling while the other two arms are quiet.
This causes the animal to turn a somersault toward the left which brings it again into
the natural position.

surface, and the animal is then able to turn a somer-
sault and regain its normal position. For this
result, it is essential that all five arms do not attempt
simultaneously to bring the animal into the ventral
position. Should the tips of all five, or even four,
arms tug simultaneously, it would be impossible for
the animal to turn over. In normal starfish having
five arms, not more than three begin the act of turn-
ing ; the other two remain quiet. If we, however,

EXPERIMENTS ON ECHINODERMS 63

destroy the nervous connection between the arms, for
instance, by making two incisions at a and b, Fig. 18,
this cooperation of the arms ceases. The normal
starfish requires but a few minutes to turn over, but
the specimen represented in Fig. 18 remained on its

Fig. 18. The Same Experiment on a Starfish whose Nerve-Ring
HAS been Severed in two Places (a and b).

The right and left arms are consequently no longer connected nervously. If such an ani-
mal is laid on its back, the tube-feet of four or even all the arms in most cases tug
simultaneously. This prevents the animal from righting itself.

back the whole afternoon, although the arms were
struggling constantly to right it. The experiments
seem to indicate that in a normal starfish the stimulus
produced by the pulling of two or three arms in the
same direction has an inhibitory effect on the other
arms. This inhibition ceases when the nervous con-
nection between the single arms is broken. Romanes

64 COMPARATIVE PHYSIOLOGY OF THE BRAIN

found that a single arm containing only the peripheral
arm-nerve rights itself when laid on its back. Hence
the central nerve-ring acts only as a conductor and
not as a " centre " for this reaction (i).

2. In analysing this righting reflex of the starfish,
there are two possibilities to be considered. Either
gravity forces the starfish to turn the ventral side
toward the centre of the earth, or contact-irritability,
i. e., stereotropism, forces the animal to bring the vent-
ral side in contact with solid bodies. The fact that
the animals leave the horizontal bottom of an aqua-
rium and attach themselves to the vertical sides shows
that gravity is not the cause. Preyer made an ex-
periment from which he concluded that the righting
of the starfish is due to their being forced to have
the ventral side down. He suspended a starfish in
the middle of the aquarium by fastening each of its
arms by threads to a cork that floated on the surface
of the aquarium. If suspended with its back down,
Preyer noticed that the starfish turned over. This
might suggest the idea that the righting of the star-
fish is a geotropic phenomenon. I have repeated
Preyer s experiment and confirmed his observation (2).
At the same time, however, I made a control experi-
ment which Preyer omitted. In the beginning I
fastened the starfish to the cork-plate in such a way
that the ventral side was turned toward the bottom.
But the starfish even then turned over. This shows
that the suspension makes it restless and causes it to
perform all sorts of turning movements. I believe

EXPERIMENTS ON ECHINODERMS 65

that the ventral side of the starfish is positively ste-
reotroplc, or, in other words, that the starfish becomes
restless if Its ambulacral feet are not in contact with
solid bodies.

3. Preyer accredits the starfish with possessing
*' intelligence." He placed one arm of an Ophiuris
In a piece of rubber tubing in order to see whether
the animal would be clever enough to rid itself of this
impediment to its movements. He found that after
a time the arm " freed itself " from the tube. I have
repeated the experiment in these animals and found
that the Ophiuris pays no attention to the rubber
tube. The animal of course loses it after a time un-
less it fits too closely, but it Is always purely a matter
of chance, and there is no more intelligence involved
than the clothes-line displays when the clothes are
blown from it by the wind. Romanes found that
when one arm of a starfish is stimulated the animal
moves in a direction opposite to the stimulated arm.
This also looks like intelligence, for the animal seems
to be able to avoid a danger. The late Professor
Norman called my attention to the fact that when
one arm of a starfish Is stimulated the feet of this
arm are drawn in and the arm becomes inactive.
This is, however, only true of the stimulated arm ;
the others remain active. Therefore, according to the
parallelogram of forces, a movement away from the
point of stimulation will take place. Intelligence
plays no part in this phenomenon.

4. The tendency to crawl upwards on vertical

66 COMPARATIVE PHYSIOLOGY OF THE BRAIN

surfaces is a pronounced reaction of Echinoderms and
is quite common in other animals, for instance in the
Actinia fnesembryanthemum of the Mediterranean,
and in the CoccinelH. This tendency is also present
in plant-organisms — for example, Plasmodia, — and
here Sachs has traced it back to negative geotrop-
ism. I will repeat here the description which I have
already given in a former publication of the phenome-
non as it appears among Echinoderms (3).

No one who observes the animals on rocks or posts
near the surface of the ocean when the water is quiet
can fail to notice the relatively large number of
Echinoderms. Many of these — for example, the Cu-
cumaria cucumiSy which is very common in the Bay
of Naples — always live near the surface, not beyond a
depth of about 30 m. It can be shown that Cucuma-
ria, like Plasmodia or CoccinelH, are forced, when on
vertical surfaces, to crawl upward. Cucumaria has a
slender pentagonal body, 10 cm. or more in length,
with radial, branching tentacles on its oral end.
There are five ridges on the body, and in these are
situated longitudinal rows of tube-feet, by means of
which the animal crawls upward, even on smooth
glass walls. If placed in an aquarium, it crawls about
on the bottom until it comes to a vertical side ; it then
crawls upward and remains on the highest point, if
possible just below the surface of the water. This
position then usually becomes permanent, and the
animal is converted into a sessile organism.

If a Cucumaria is allowed to attach itself to a

EXPERIMENTS ON ECHINODERMS

vertical glass which can be revolved around a hori-
zontal axis in the aquarium, it will crawl upward
whenever the glass is turned. This is not a compen-
satory movement produced by the centrifugal force,
for during the rotation of the glass the animal re-

vyyyyyyyyyyyyyyyyyyyy/'yyyyyyyyyyy/:>Z7zy.^

Fig. 19. Geotropic Reaction of Cucumaria Cucumis.

The animals are in a battery jar (a, ^, f , <f ). It is filled with water and rests on the bridge
B B in the aquarium A A. Running water is supplied through the tube ^ at o. They
collect at the highest point (^, d) of the glass.

mains quiet, and not until a quarter or half an hour
after the rotation does it begin to migrate upward.
Neither is the upward migration caused by the light
falling in from above. If the animals are placed in
an aquarium in which light is allowed to enter only
from the side or from below, they will still crawl up-
ward on the vertical sides. In a dark room they be-
have just as they do in the light.

One might believe that the need of oxygen determ-
ined the upward migration of the Cucumarise to the

68 COMPARATIVE PHYSIOLOGY OF THE BRAIN

surface of the water. It can be shown, however, that
this is not the case. If a large beaker filled with
water be placed inverted in the aquarium, the Cucu-
mariae that are under the beaker begin to creep up to
the bottom of the glass. They also do so when the
experiment is made in the manner represented in Fig.
19. A bridge B B is placed in the aquarium A A,
the horizontal part of the bridge B B being below
the surface of the water of the aquarium. The hori-
zontal part has a round opening o over which the in-
verted beaker abed filled with water Is placed. Fresh
water is supplied at a low pressure at 0 through a
glass tube^, which has been properly bent. The Cu-
cumarise nevertheless go away from 0 and remain at
the highest point c d, or near cd on the vertical sides
(Fig. 19), where they ultimately die.

Experiments on the centrifugal machine yielded no
result, for the animals did not move during the rota-
tion. Gravity is the only condition which can account
for the phenomenon, and I imagine the influence
which gravity exercises to be in a manner similar to
that observed among insects — for example, in butter-
flies which have just emerged from the chrysalis.
The wings of the butterfly do not unfold immediately,
and it runs about restlessly until it comes to a vertical
surface. When this is reached, the butterfly creeps
upon it and remains there for some time with its head
up. After the wings are spread, other conditions
cause the animal to be restless again.

Because of this dependence on gravity, the Cu-

EXPERIMENTS ON ECHINODERMS 69

cumarlae are of necessity inhabitants of the surface-
regions of the ocean. If a larva were carried down
to a great depth, its negative geotropism would force
it to migrate upward until the highest point was
reached or until death put an end to its upward
journey.

Certain starfish — for instance Asterina gibbosUy
which also lives near the surface of the water — behave
like Cucumaria. All the experiments I have made
on Cucumaria can likewise be successfully performed
on Asterina gibbosa, but with the difference that the
exceptionally voracious Asterina does not remain per-
manently at the highest point of the vertical surface.
In two days, or sometimes even sooner, it begins to
move or drops down.

Positive heliotropism naturally has the same effect
as negative geotropism. Asterina tenuispina, like As-
terina gibbosa, lives at the surface of the sea. It is
not, however, geotropically irritable ; but it is posi-
tively heliotropic. I put a large number of specimens
of both species in a heap in an aquarium, into which
rays of light from one side only fell nearly horizon-
tally. In a short time the two species had parted,
the Tenuispinse crawling off on the floor toward the
source of light. The Gibbosse, scattered about on
the bottom of the aquarium in every direction,
crawled up the vertical sides without being influ-
enced at all by the light in their movements. In the
ocean, where the vertical rays of daylight are chiefly
concerned in the orientation of animals, positive

*]o COMPARATIVE PHYSIOLOGY OF THE BRAIN

heliotropism must drive Asterina tenuispina to the
surface of the ocean, just as Asterina gibbosa is driven
there by negative geotropism.

Preyer mentions briefly in his extensive work on
The Movements of the Starfish the '' tendency of these
animals to move upwards." " The strong tendency of
starfish and brittlestars to go upward cannot be traced
back to lack of air, lack of food, changes in tempera-
ture or current, or to a desire for light, for -they climb
up just the same when these conditions are eliminated.
Probably some peculiarity of the bottom, or of just
that part of the bottom where the animal is, makes it
unsuitable for the suction of the tube-feet. The ani-
mals can remain there no longer, so they move up-
wards. But it is possible that parasites, which I have
often found in the ambulacral furrows, may cause this
upward migration, for as the stimuli produced by
them come from below, they might seem to belong to
the bottom."

The first sentence in this generalisation is wrong ;
the light attracts Asterina tenuispina upwards. Sec-
ond, the character of the bottom does not determine
the phenomenon. If Asterina gibbosa be placed in a
cubical box with glass sides, the animals leave the
basal horizontal side and crawl up the vertical sides.
If the box then be turned 90° around a horizontal
axis, the side which is now basal is deserted by the
animals. They crawl up and remain on the side
which, while horizontal, they had left. Finally, if
Preyer believed that parasites force the animals to

EXPERIMENTS ON ECHINODERMS 71

crawl upward, it is difficult to see why they should
not drive the animals down from the vertical side.
As a fact, however, Asterina gibbosa^ as well as Cucu-
maria ctccufnis, remains on the highest point of the
vertical side. I believe it is much nearer the truth to
ascribe the vertical upward movements of certain
starfish to an action of the force of gravity.

Bibliography.

1. Romanes, G. J. Jelly-fish^ Starfish and Sea Urchins. New
York, 1893.

2. Preyer, W. Ueber die Bewegung der Seesterne. Mit-
theilungen aus der zoologischen Station zu Neapel^ Bd. vii, S. 96.

3. LoEB, J, Ueber Geotropismus bet Thieren. Pfliiger's Ar»
chiv^ Bd. xlix, 1891.

CHAPTER VI

EXPERIMENTS ON WORMS

I. We shall consider separately in this chapter two
kinds of worms : first, those in which the ganglia are

all crowded together in

rN^ ^ the head end — e. g.,

Planarians ; and second,
those with a series of
segmental ganglia — e.
g., Annelids.

Sea- and fresh - water
Planarians differ little
structurally, yet they
may show different re-
actions upon losing the
oral ganglion.

Thysanozoon (Broc-
chii). Fig. 20, a marine
Planarian, is very com-
mon in the Bay of Na-
ples. It is from i to 3
cm. long and nearly as broad. The oral end of
the body, which can be recognised by two tentacles

Fig

ThysanozoOn Brocchii,
Marine Planarian.

^, brain ; w, mouth ; «, longitudinal nerve.
(Diagrammatic after Lang.)

EXPERIMENTS ON WORMS

(^, Fig. 20), contains the brain of the animal. This
consists of two connected gangHa, from which a series
of nerves, containing single ganglion-cells, go out ;
among the latter, the
two large longitudinal
nerves running length-
wise throughout the
animal {n, Fig. 20) are
conspicuous. In the
periphery a plexus is
formed (i). The central
nervous system consists
of the double ganglion
in the forward end.
Like all Planarians,
Thysanozoon crawls on
the side of the aquarium
or on the surface film
of the water. It differs
from the fresh - water
Planarians in being able
to perform, in addition, genuine swimming move-
ments. With the sides of its body it makes vibra-
tions similar to those made by the wings of a but-
terfly. If while a Thysanozoon is gliding about on
the surface of the water it be divided transversely
with a pair of scissors, the posterior or aboral half
{b, Fig. 21) at once falls to the bottom, while the
oral piece {a. Fig. 21) containing the brain creeps
on undisturbed. If the division be made with a

Fig. 21.

Thysanozoon Divided
Transversely.

The anterior piece a, containing the brain, shows
spontaneity ; the posterior piece b, none.

74 COMPARATIVE PHYSIOLOGY OF THE BRAIN

sharp knife while the Planarian is crawling on a
glass plate, the oral piece (a, Fig. 21) crawls on un-
disturbed, while the progressive movement of the
posterior piece ceases entirely. The spontaneity of
the progressive movement of the Thysanozoon is
then a function of the part of the body containing
the brain (2).

In a Thysanozoon that has been divided, both pieces
live and regenerate the lacking parts. The oral piece,
however, regenerates more rapidly than the aboral
piece, which has to form a head. I have not investi-
gated whether the latter also forms a new brain. I
kept such pieces alive for four months. The spon-
taneity of the posterior piece never returned ; the
spontaneity of the anterior piece remained.

If we put a normal Thysanozoon on its back it soon
rights itself. The question now arises whether, like
the progressive movements, these righting movements
are a function of the brain. This is not the case. A
Thysanozoon from which the brain has been removed
rights itself if laid on its back, only the reaction pro-
ceeds more slowly than in the normal animal, or even
in a piece of an animal if this piece contains the brain.
We see here again that the nervous system only
serves to bring about a quicker reaction.

If, instead of dividing the Thysanozoon completely,
we leave the two parts on one side connected by a
thin piece of tissue in such a way that (Fig. 22) the
posterior piece can receive no direct innervations
from the brain through the longitudinal nerves.

EXPERIMENTS ON WORMS

conduction of stimuli would still be possible through
the side nerve-plexus.

Such an animal was placed after the operation on
the bottom of the aquari-
um ; the anterior piece im-
mediately began to move,
while the posterior piece
attempted to attach itself
to the bottom. The latter
soon yielded to the tugging
of the oral piece, however,
and took part in its pro-
gressive movements in an
entirely coordinated man-
ner, as though no incision
had been made. After a
time the oral piece turned
around and crawled over
the back of the posterior
piece, which was dragged behind passively, and was
turned on its back. It righted itself immediately and
moved off actively in the same direction as the oral
piece. Changes of direction originated only in the
piece containing the brain and were never transmitted
directly to the posterior piece. But if the oral piece
continued for a time to move in the same direction
and with the same rapidity, the same movement would
soon take place in the posterior piece. Hence the
posterior piece did not behave entirely like a piece
from which the brain had been removed, for it made

Fig. 22. ThysanozoOn with
Transverse Incision.

76 COMPARATIVE PHYSIOLOGY OF THE BRAIN

progressive movements, nor yet like a normal Thy-
sanozoon for it had lost its spontaneity. This becomes
still more apparent from the following observation :
I threw a Thysanozoon similarly operated upon into a
tank of water. Both pieces performed synchronic
swimming movements. The oral piece soon reached
the vertical side of the aquarium and began to creep
upwards. As a result of the change of direction in
the anterior piece, the tissue connecting the two parts
became twisted and the back of the posterior piece
came in contact with the glass, while the ventral side
was turned toward the water. It then made swim-
ming movements and in this way followed the crawl
ing movements of the oral piece. The posterior
piece therefore is not simply dragged behind passively,
but takes an active part in the progressive movement
when the movements are continuous. This is also
evident from the fact that it would often crawl along
on the back of the oral piece, especially if the latter
suddenly began to move more slowly.

These experiments show that a Thysanozoon from
which the brain has been removed no longer moves
spontaneously, nor is it possible to produce progress-
ive movements in it by any external stimulus. If
touched, local contractions, only, result.

2. The brain and nervous system of the fresh-water
Planarians (Fig. 23, from Jijima) are so similar to those
of the marine Planarians that for our purpose it is un-
necessary to give a special description of them. The
principal difference is probably that the two longitu-

EXPERIMENTS ON WORMS

dinal nerves contain a greater number of ganglion-
cells, so that they almost form segmental aggregations.
From this similarity we should Infer that the brain-
functions of the fresh-water Plana-
rlans would be analogous to those
of the Polyclads. However, such
is not the case. If we divide a
fresh-water Planarian, for instance
Planarta torva, transversely, the
posterior half, that has no brain,
crawls just as well as the oral half.
Spontaneity In Planarta torva is,
therefore, by no means a function
of the brain. Every piece of the
animal that Is not too small pos-
sesses spontaneity. The decap-
itated animals crawl with the
anterior end in front like normal
animals (2).

The question now arises as to
how it happens that in Thysano-
zoon spontaneous movements cease
if the head be amputated, while in
fresh-water Planarians this opera-
tion does not have such a result,
to account for the difference by the fact that the
fresh-water Planarians have more ganglion-cells
throughout the longitudinal nerves than the Thysano-
zoon. With the aid of comparative physiology it is
possible to show that such a view Is untenable. In

Fig. 23. Fresh- Water
Planarian (Plana-
RiA Torva).

G^ brain, m, longitudinal nerve.
(After Jijima.)

One is tempted

78 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the crayfish, the suboesophageal ganglion with the
ventral ganglion chain represents a much more highly
developed ganglion-system than the longitudinal
nerves in Planaria torva. We shall see, nevertheless,
that a crayfish, which possesses these ganglia, but
has lost the supracesophageal ganglion, no longer
moves spontaneously. We shall see, furthermore,
that a frog that has lost the cerebral hemispheres and
thalamus opticus does not move spontaneously, al-
though it possesses many more ganglia in the spinal
cord than Planaria torva. The same frog, however,
moves spontaneously again if, in addition, the optic
lobes and the pars commissuralis of the medulla
oblongata be removed.

Spontaneous progressive movements are not a
specific function of ganglia or of ganglion-cells ; we
observe them even in the swarmspores of algse and
in bacteria. Why the decapitated Thysanozoon no
longer performs progressive movements, and a decap-
itated fresh-water Planarian continues to move spon-
taneously, we are not yet prepared to say. It is
possible that the difference between fresh-water and
marine Planarians is somewhat of the same character
as that between Hydromedusse and Acalephse. In
the latter, both parts, margin and centre, beat rhyth-
mically in sea-water, while in the Hydromedusse only
the margin with the nerve-ring is able to do so. But
we were able to show that this difference between the
two classes of Medusae is not so much due to mor-
phological differences as to chemical or physical

EXPERIMENTS ON WORMS 79

differences. A reduction in the amount of Ca ions
in the sea-water allowed the centre of a Hydromedusa
to beat spontaneously. The case of marine Planarians
may be similar, and further experiments may yield the
result that with a change in the constitution of the
sea-water the posterior half of a Thysanozoon will
be able to show spontaneous locomotion.

The behaviour of Planaria torva toward light is of
special interest. The animal is especially sensitive to
changes in the intensity of light. If brought from
the dark into the light suddenly, it begins to move.
At first the direction of the movements seems to be
influenced by the light, for the animals move away
from the source of light as if they were negatively
heliotropic. However, they do not collect at the point
farthest from the source, as do negatively heliotropic
animals, but they scatter in all directions and come to
rest at last in a place where the light is comparatively
weakest. From this it would seem that an increase
in the intensity of light causes them to move, while
a decrease in the intensity of light causes them to
rest. This would account for the fact that we find
them by day always under stones or in relatively dark
places. I suspect that they begin to move about in
the night, and that they come to rest when day
returns. I have repeatedly tried the experiment of
covering in the morning one-half of the dish with
black paper. During the day no change takes place,
but the next morning all the animals are found under
the covered portion of the dish. The only possible

8o COMPARATIVE PHYSIOLOGY OF THE BRAIN

explanation for this behaviour is that they crawl about
in the dish during the night and in the morning stop
in the darkest place. These animals have at their
oral pole not only a brain but also comparatively well-
developed eyes. I resolved to try whether a decapit-
ated Planarian, in spite of the loss of brain and eyes,
would still show the same reactions toward light as
the normal animals. This is the case to a most sur-
prising extent. In the evening, about sixty specimens
of Planaria torva were cut transversely just behind
the brain and eyes. All the pieces were put into a
dish with vertical sides which was half covered with
black paper. The next morning nearly all the pieces,
posterior as well as anterior, were found in the covered
portion of the aquarium, where they were scattered
about pretty uniformly. In the uncovered portion of
the dish I found a few pieces, anterior, however, as
well as posterior ones, crowded together in a corner
where the intensity of the light was a comparative
minimum. Upon repeating this experiment with nor-
mal Planarians, the same result was obtained. When
the decapitated animals were at rest in the covered
portion of the dish, their rest was soon disturbed if,
without jarring the aquarium, the dark paper was
removed suddenly. At first they crawled about on
the side away from the light, then they collected
again where the intensity of light was a relative mini-
mum. This reaction occurred just as in normal ani-
mals, except that the reaction-time of the brainless
animals was greater than in normal animals. In the

EXPERIMENTS ON WORMS

pieces containing brain and eyes, the reaction be-
gan about one minute after exposure to light ; in the
pieces without brain, after about ^m^ minutes. In this
experiment, only diffused daylight was employed as a
stimulus. In a round dish with vertical sides, the
Planarians do not collect, like strictly heliotropic ani-
mals, on the window- or room-side of the dish, but on
the right and left sides. Decapitated Planarians be-
have in the same way. All these reactions occur the
day after the operation. Fresh
material should always be used for
these experiments.

After what has been said, it is
hardly necessary to mention that
pieces of Planar ia torva from which
the brain has been removed right
themselves as well as normal animals.

According to some authors, the
starfish represents a colony of as
many individuals as it has arms.
We have seen that these react har-
moniously as long as the nervous fig. 24. two-headed
connection is uninterrupted. In
Actinians that have been made two-
headed artificially, this harmony no
longer exists ; for instance, in taking food both heads
struggle for the same piece of meat. At my suggest-
ion. Dr. van Duyne tried to produce multi-headed
Planarians artificially. He succeeded in making
them with as many as six heads. Fig. 24 shows a

Planarian Produced
Artificially. (After
van Duyne.)

82 COMPARA TIVE PHYSIOLOG V OF THE BRAIN

two-headed specimen. If the heads were far enough
apart, they no longer moved synchronically in the

same direction,
in which case
the pulHng in
opposite direc-
tions (Fig. 25)
was so stronof
that the animal
was torn asun-

FiG. 25. Planarian with two heads that are , , .

Attempting to Move in Opposite Directions, ^^^ v /*

AND IN so Doing are Tearing the Common %, In the An-

BODY. (After van Duyne.) ^^^jj^^ ^^ ^^^

a segmental arrangement of the central nervous sys-
tem. This type of structure is also found in Arthro-
pods and in Vertebrates. It will perhaps make our
task easier if we conceive the segmented animal to be
a colony of as many individuals or animals as there
are segments (or ganglia) present in the body. Each
segment is then comparable to an Ascidian in which
the central nervous system consists of but one gan-
glion. The fibres and cells of each ganglion form
for the corresponding segment a protoplasmic bridge
between the skin and muscles. A stimulation be-
ginning, however, in one segment is not confined to
that segment, for the single ganglia of the various
segments are connected with each other by means of
nerve-fibres, the so called longitudinal commissures.
By means of these, a stimulation which originates in
one segment is transmitted also to the neighbouring

EXPERIMENTS ON WORMS

/ _^— .u.

ganglia and from these to those farther away, until at
last it reaches the end of the animal.

The central nervous sys-
tem of Annelids corre-
sponds to the spinal cord
of Vertebrates and consists
simply of a chain of ganglia.
These lie entirely on the
ventral side of the animal,
with the exception of the
most anterior (supraoeso-
phageal) ganglion (Fig.
26), which lies above the
oesophagus on the dorsal
side. This is connected
with the suboesophageal
ganglion by a double com-
missure, which forms a loop
through which the oesopha-
gus passes. It may be
called the brain, although
the small analogy exist-
ing between Vertebrates
and worms makes the use
of the term purely arbi-
trary.

A question of funda-
mental interest to us arises
at this point : Is the brain simply a segmental gan-
glion, or is it an organ of a higher order which

Fig. 26. The Brain and a Series
OF Segmental Ganglia of an
Annelid (Nereis).

Oy supraoesophageal ganglion or brain ; c,
commissure ; «, suboesophageal ganglion.
(After Claparede.)

S4 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Ge-

regulates and guides the activity of the other
gangha ?

In our analysis of the nerve-functions we will begin
with the earthworm. We will consider first its pro-
gressive movements, and will attempt to answer the
question, Does coordinated progressive movement, in

which all the segments of
the body participate, de-
pend upon the brain (^,
Figs. 27 and 28) ? The
locomotion of the earth-
worm is a very simple
process. The setae play
an important role, although
they are not visible to the
naked eye ; they act like
locomotor appendages and
give the animal a hold on
Fig. 27. Dorsal View of the Cen- the ground. The real mus-
TRAL Nervous System of an ^|^g ^£ locomotion, how-
Earthworm. . ' .

ever, are contamed m the

Oy supraoesophageal ganglion ; c, commissure ;

«, suboesophageal gangUon ; 6-, pharynx ; CUtaUeOUS mUSCle - layer.
G, ganglia of the ventral cord. , . . - . _ .

This consists 01 ring-nbres
and longitudinal fibres. When the worm begins to
move, the ring-fibres contract first, causing the worm
to become longer and thinner. The bristles are
turned backward and, because of the resistance of
the ground, prevent the animal from moving back-
ward. In this way the head is pushed forward. As
soon as the maximum elongation has been reached,

EXPERIMENTS ON WORMS

the longitudinal muscles contract and the worm
becomes shorter. As the bristles are still turned
backward, the shortening can only be accomplished
by the approach of the posterior end toward the
head. The entire worm is therefore compelled to
move forward. What happens if ^

we divide the ganglion-chain of /

the animal in the middle of the
body, or if we remove some
ganglia from that region ? Will
the forward piece move inde-
pendently of the posterior piece ?
Benedict Friedlander has made
this experiment and found that
the coordination continues in spite
of the division of the central ^^Z~
nervous system (4). If the for-
ward piece begins to move, the
aboral piece will also move in the
same direction and at the same
rate. This overthrows the idea
that coordination in these animals
is determined by a special centre '^ '"fpraoesophageai ganguon or

J ^ brain ; «, subcesophageal

of coordination which is located in ganglion; a intestine; c,

. T-» 1 1 1 ganglia of the ventral cord.

the bram. But how, then, does
the coordination take place ? When the forward
piece elongates and attempts to shorten itself by
contracting the longitudinal muscles, the skin of the
aboral piece is stretched. This pulling probably
acts as a stimulus which causes the longitudinal

Fig. 28. Side View OF THE
Central Nervous Sys-
tem OF THE Earth-
worm.

86 COMPARATIVE PHYSIOLOGY OF THE BRAIN

muscles of the aboral piece to contract refiexly or
perhaps directly. In this way, therefore, coordina-
tion between the oral and aboral piece is possible in
spite of the interruption of the nervous connection.
Friedlander obtained further proof of this by divid-
ing worms completely and connecting the two halves
by strings. Even then he found that the aboral piece
moved with the anterior piece in a perfectly coordin-
ated way. These facts prove that the brain has no
leading role in the coordination of the progressive
motions of the earthworm.

What part, then, does the central nervous system
play in the coordination ? It serves only as a quick
conductor for the stimuli. Friedlander has shown
that the quick motions which an earthworm shows
upon a sudden stimulus are no longer transmitted
to the posterior part of the body if the ganglion-
chain be severed. If the nervous connection be
broken so that stimuli cannot be conducted through
the nerves, the peripheral structures suffice to make
coordinated movement possible.

One might suppose the coordination in the pro-
gressive movements of higher animals to be of an
entirely different nature from that of worms. An ob-
servation made by Goltz, however, shows that in
dogs, at least, this is not the case. When a dog with
divided spinal cord is lifted up by its fore-legs, so that
the back part of the body hangs down perpendicu-
larly, a remarkable phenomenon may be observed.
The hind-legs perform pendulum motions which

EXPERIMENTS ON WORMS 87

resemble locomotion. These motions are presumably
produced by the passive stretching of the skin on the
ventral side of the hip-joint by the weight of the legs.
These motions are comparable to the reflex contrac-
tion of the longitudinal muscles of the earthworm,
which is due to the stretching of the skin. Because
of this reflex, coordinated locomotion would be quite
possible in a dog with divided spinal cord, if the dog
only could remain standing on its hind-legs. The
walking movement of the fore-legs would cause the
stretching which is necessary in order to bring about
the walking movement of the hind-legs. The differ-
ence in the behaviour of a dog with divided spinal
cord and of an earthworm with divided ventral nerve
cord in regard to coordinated progressive movements,
is not caused so much by differences in the functions
of the central organs as by differences in the develop-
ment of the peripheral organs of the skin and of the
organs of locomotion. If the dog had short stumps
instead of its long, jointed legs, we should have, after
dividing the spinal cord, the same phenomenon of
progressive movements that we have in the earth-
worm. The irritability of various parts of the peri-
pheral organs and the simple segmental arrangement
of the nervous elements suffice to preserve the loco-
motion when it has once been started. The correct-
ness of this conclusion is confirmed by experiments
on Nereis, which were made in my laboratory by S.
S. Maxwell (5). In these animals, the coordination
of the movements of the oral and aboral pieces is

88 COMPARATIVE PHYSIOLOGY OF THE BRAIN

practically destroyed by dividing the ganglion chain,
for the deep incisions between the single segments
prevent the entire cutaneous muscle layer from being
stretched equally. The structure of the ventral nerve-
cord in Nereis is so similar to that of Lumbricus that
we should not be justified in seeking in it for the
conditions which cause difference of behaviour. In
earthworms, Maxwell succeeded in confirming Fried-
lander's observation. I obtained similar, although
not as marked, results on leeches (2).

If an earthworm be divided, the posterior, brainless
piece continues to perform progressive movements.
This fact confirms the opinion that the brain has no
controlling part in progressive movements.

4. The question now arises. Are the remaining char-
acteristic functions of the earthworm brain-functions
or segmental functions ? If we place Lumbricus fceti-
dus in a transparent closed vessel, the animals appear
to be positively stereotropic. As soon as they reach
an angle in the aquarium, they remain there, crawling
along where the glass can touch them on two sides.
They are also sensitive to the differences in the intens-
ity of light, remaining in those places where the
light is weakest. It seems, too, when one or more
animals settle down anywhere, that the others stop
more readily in that place. This is an illustration of
"sociability" among lower animals. It is probably
an instance of chemotropic irritability. The surface
secretions emanating from the worm's body have a
quieting influence on other worms of the same kind ;

EXPERIMENTS ON WORMS 89

for this reason they become quiet when in contact
with a worm of the same species. These chemical
stimuli act as a trap, just as the comparative minimum
in the intensity of the light acts. It should be noted,
in this connection, that when animals are sensitive to
differences in the intensity of light, the less refractive
rays which pass through red glass have less effect
upon them than the more refractive rays which pass
through blue glass. The earthworms become quiet
under red glass sooner than under blue glass.

How do decapitated earthworms act ? Decapitated
Lumbrici foetidi show the same stereotropism that
normal worms show. When they reach the concave
angle of a vessel, they have no inclination to leave it
again. They also show the same response to light.
They rest in those places where the intensity of the
light is relatively weakest, and they move when the
intensity of the light is increased. It can also be
shown that light passing through blue glass acts like
light of greater intensity, while light passing through
red glass has the effect of light of weaker intensity (2).

In all these experiments the decapitated pieces
crawl about with either the tail or the anterior end
in front.

It is an interesting fact that the reaction-time when
light is the stimulus is not appreciably greater in de-
capitated than in normal earthworms. The animals
used for the experiment were in a box in which they
could be exposed to diffused daylight suddenly with-
out being jarred. In from three to eighteen seconds

90 COMPARATIVE PHYSIOLOGY OF THE BRAIN

after being exposed to the light, the decapitated worms
made the first movements. The interval was about
the same In normal worms.

Ltimbrici fostidi live In the decaying compost of
stables, and probably the chemical nature of certain
substances contained In the compost holds them
there. When one-half of the bottom of the box is
covered with moist white blotting-paper, the other
half with a thin layer of compost, all the normal worms
that are placed on the paper soon gather on the com-
post. The aboral pieces of divided worms behave in
the same way. When placed upon the blotting-paper,
they are not attracted directly by the odours of the
compost, but as soon as they come In contact with It
in moving about, they crawl on It and do not leave It
again. After a short time all the brainless worms are
on the compost. When placed on a heap of compost,
most of them crawl into it within a short time. This
is not due solely to the light, as the same reaction
also takes place in the dark (2).

Thus we see that in decapitated earthworms all the
reactions shown by the normal worms are retained.
Hence the brain (supraoesophageal ganglion) has in
this case no leading r6le.

We cannot be too careful in drawing conclusions
in regard to the principal function of a ganglion.
Nereis, a much more highly developed Annelid than
the earthworm, burrows in the sand ; if decapitated,
this function ceases. One might suspect that this was
due to the loss of the brain, but such Is not the case.

EXPERIMENTS ON WORMS 91

Earlier experiments had led me to suspect that the
" spontaneous " or " instinctive" burrowing was only
a reflex produced by the contact-stimuli of the sand.
I then attempted to find out whether it were not pos-
sible under special conditions to produce the same
reflex in brainless pieces. I placed such a piece of a
Nereis on the sand ; as usual it remained quiet. I
then gradually covered the forward end with sand.
The rest of the animal immediately began to make
the typical movements which the animal makes in
forcing Its way into the sand. At the same time the
glands began to secrete the sticky substance which
cements the particles of sand together, forming the
wall of the burrow-hole. This secretion-phenomenon
regularly accompanies the burrowing of these animals ;
it is the same secretion that in other animals leads to
the formation of a case.

But why does the Nereis not burrow when deprived
of its brain ? For the simple reason that it makes
use of the organs of the mouth in burrowing, and these
are amputated with the head. Hence it is the loss of
a peripheral head-organ which keeps the decapitated
Nereis from burrowing, and not the loss of the brain.
The brain in this case merely performs the function
of a segmental ganglion — that is, it acts as the ganglion
of that segment to which the peripheral head-organ
belongs.

5. We will now turn our attention to the brain-
functions of Nereis.

After a Nereis has burrowed in the sand it lives in

92 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the same case for a long time. If the supraoesopha-
geal ganglion (o, Fig. 26) be removed, the animal be-
comes restless, as S. S. Maxwell has found. It crawls
about on the sand unceasingly, making no attempt to
burrow. This restlessness is marked by one feature
which we find in higher animals after certain injuries
to the brain — namely, the Nereis does not withdraw
from obstacles but attempts to force its way through
them.

If normal Nereis are in a square aquarium the bot-
tom of which is covered with sand, they will crawl
about, if undisturbed, on the sides of the glass. This
is the result of stereotropism. A Nereis that has lost
the supraoesophageal ganglion will behave in the
same way, except that when it reaches a corner it
does not turn out but attempts to go through the
glass. If there are several animals which have been
operated upon in a vessel, they will assume the po-
sition represented in Fig. 29. The worms remained
like this for many hours at a time, and then died in
consequence of their vain attempt to go forward.
Those reactions are wanting which in the normal
Nereis result from the application of contact-stimuli
to the oral end. The reader who is familiar with
brain-physiology may already have been reminded in
this connection of the dogs from which Goltz re-
moved the anterior half of the cerebral hemispheres.

If glass tubes 20 cm. long, with a bore a little
larger than the diameter of the worm, are placed in
an aquarium without sand, the normal Nereis will

EXPERIMENTS ON WORMS

crawl into them and will not leave them again. This
is due to stereotropism. If there are, for instance,
six such tubes in a vessel and six normal Nereis are
put into it, we may be sure that after a few hours
every Nereis will have established itself in a tube. It

Fig. 29. A Group of Nereis whose Brains have been Removed. They at
LAST Collect in a Corner of the Aquarium and Perish in their Vain
Attempt to Go through the Glass. (After Maxwell.)

frequently occurs that a Nereis goes into a tube that
already has an occupant. In that case the new-comer
withdraws with a start as soon as it touches the old
occupant. As long as the new-comer is in pos-
session of its brain, it leaves the tube under such

94 COMPARATIVE PHYSIOLOGY OF THE BRAIN

circumstances, but if it has lost the supraoesophageal
ganghon, the presence of the other worm in the tube
has no inhibitory effect. It tries to force its way into
the tube even if it perishes in the attempt. If both
worms have lost the supraoesophageal ganglion, they
rub their heads together until they are sore. If we
wish to keep them alive, they must be separated by
breaking the tube. If we compare the conduct of a
Nereis whose brain has been amputated, with that of
a normal worm, the difference seems to be of the
same nature as that between an insane and a rational
human being. It would be erroneous, however, to
conclude that the normal, brain-endowed Nereis pos-
sesses reason or intelligence. The peculiar irritability
by means of which the Nereis draws its head back and
moves backward out of the tube depends upon organs
which are located in the forward end of the body and
whose sensory nerves go to the supraoesophageal
ganglion. Hence, if the supraoesophageal ganglion
is extirpated, the connection between these organs
and the rest of the body is interrupted, and the stimuli
which affect the forward part of the body can no
longer produce backward movements in the posterior
portion of the animal.

This does not, however, explain the change of
character, the restlessness of the Nereis which has
been deprived of its brain. It is maintained that, if
the spontaneous activity or the reflex irritability of
an animal is increased after the loss of a part of the
brain, that part is an inhibitory mechanism. Nothing

EXPERIMENTS ON WORMS 95

is gained, however, by making such a statement. We
wish to know how the supraoesophageal gangHon can
inhibit movements, and how its absence can increase
spontaneity.

It is not possible to offer at present more than a
suggestion. We can increase and decrease the loco-
motor activity of a jelly-fish at desire by changing
the constitution of the sea-water. If we increase the
number of Na ions in the sea-water, the rate of rhyth-
mical contractions in Gonionemus increases and the
animal becomes restless. If the number of Ca ions
be increased, the animal becomes quiet. It is, more-
over, a fact that the different parts of a Gonionemus
are affected somewhat differently by the same ions,
inasmuch as the margin is more immune against the
effects of Ca ions than the centre. I think it possible
that there is a similar difference between the segments
belonging to the supraoesophageal and suboesopha-
geal ganglion. It might be possible that the ions
(or some other substance) of the blood influence the
supraoesophageal ganglion or its segments in such a
way as to cause a decrease in the locomotions, while
the Game constituents of the blood do not have such
an effect upon the subcesophageal ganglion or its
segment. But the blood is not the only agency
which is to be considered in this connection. The
supraoesophageal ganglion of Annelids is connected
with the alimentary canal by nerves. The processes
which go on in the intestine — that is, the chemical pro-
cesses of secretion and digestion — can only affect the

96 COMPARATIVE PHYSIOLOGY OF THE BRAIN

whole animal nervously as long as the supraoesopha-
geal ganglion Is Intact. If It Is removed, the whole
Influence of this so-called sympathetic nervous system
ceases. It is possible that the stimuli which pass
from the sympathetic Into the central nervous system
may condition the alternation of rest and activity
which characterises the normal animal, and that the
removal of this stimulation may remove the necessity
of resting.

Maxwell has found that a Nereis which has lost the
subcesophageal ganglion becomes quiet. Such ani-
mals make no attempt at burrowing. The reason for
this Is that the motor nerves of the oesophageal mus-
cles originate in the subcesophageal ganglion, so that
removal of this ganglion causes paresis or paralysis of
these muscles. The pharynx plays a great r6le In bor-
ing the hole. It Is due to this same paralysis or paresis
of the oesophageal muscles that a Nereis no longer
eats after losing the subcesophageal ganglion (5).

We wish to mention here, however, that removal
of the subcesophageal ganglion does not bring about
disturbances In taking up food In all Annelids. Max-
well found that the leech Is still able to suck itself full
of blood after losing the ganglion. McCaskill dis-
covered, however, that in the leech the motor nerves
of the sucking apparatus originate in the supra-
cesophageal ganglion. The subcesophageal ganglion in
the leech behaves like the first link In the ganglion-
chain.

As regards the restlessness of Nereis after removal

EXPERIMENTS ON WORMS 97

of the supraoesophageal ganglion and the repose after
the removal of the suboesophageal ganglion, we wish
to emphasise the fact that they have nothing to do
with the wound. Maxwell's observations were made
on animals whose wounds were healed. If we make
a wound like that made in removing the ganglion,
only with the difference that the ganglion is left intact,
none of the above-mentioned disturbances occur.
Immediately after the operation the worm burrows
again in spite of the wound.

Differences like those found between the behaviour
of normal Nereis and of Nereis from which the brain
has been removed do not appear in earthworms under
the same conditions. What causes this difference?
Is the supraoesophageal ganglion in the earthworm
a segmental ganglion, while in Nereis it is a "control-
ling ganglion," a brain in the sense of the anthropo-
morphic nerve-physiology ?

I am inclined to believe that we have to deal with
differences of the same character as those found
between Acalephse and Hydromedusse. In addition
it should be said that there is a much higher degree
of differentiation of the head-organs in Nereis than in
the earthworm. We have already seen in preceding
chapters that the apparent functions of the brain or of
the ganglia are chiefly determined by the peripheral
organs. In Nereis the differentiation of the head-
segments is carried much farther than that of the
other segments (Fig. 30). In the earthworm, on the
other hand, the difference is much less (Fig. 2 7).

98 COMPARATIVE PHYSIOLOGY OF THE BRAIN

In Vertebrates the head contains special sense-
organs, mouth-organs, which are lacking in the other
segments. In judging of the relation of the brain-
ganglia to the other segmental ganglia of the body
this fact should not be overlooked. Not infrequently
physiologists have ascribed to a ganglion what in

reality was due to the
higher differentiation of
the peripheral organs of
the segment.

We desire now to
touch briefly upon the
behaviour of the muscles
after extirpation of the
ganglion, for the phe-
nomena will occupy our
attention repeatedly.

In the case of loss or
a congenital lack of a
piece of the spinal cord,
the skeletal muscles belonging to the corresponding
segment atrophy. Nothing of the kind occurs in
leeches or earthworms from whose ventral chain
a piece has been removed. I believe that the dif-
ference is determined as follows : In worms direct
impulses flow from the neighbouring muscles to the
muscles that have been deprived of their ganglion,
while in Vertebrates, as soon as the spinal cord is
destroyed, the protoplasmic connection between the
skeletal muscles and the rest of the body is destroyed

Fig. 30.

Head of Nereis.
Quatrefages.)

(After

EXPERIMENTS ON WORMS 99

and it is not possible for stimuli to be transmitted.
In the muscles of the blood-vessels of the Vertebrates,
however, a conduction of the stimulation from element
to element is possible. For this reason their reactions
remain intact even in higher animals after destruction
of the corresponding segments of the spinal cord.
In Nereis, after division of the ganglion-chain, a
phenomenon may be observed that reminds one of
the Brondgeestian phenomenon. The back part of
the body becomes more flat, while the part that is
connected with the brain remains round. This indi-
cates a relaxation of the ring-muscles in the part of
the animal that is situated behind the point of
division.

The results of our physiological analysis of the
functions of the central nervous system in Annelids are
in perfect harmony with Professor C. O. Whitman's
investigations on the morphological structure of the
central nervous system in Annelids. He arrives at the
conclusion that the brain of Annelids is not of a higher
order than the other segmental ganglia (7).

Bibliography.

1. Lang, A. Untersuchungen zur vergleichenden Anatomie und
Histologic des Nervensystems der Plathelminthen. Mittheil. aus der
Zoolog. Station zu Neapel^ Bd. I.

2. LoEB, J. Beitrdge zur Gehirnphysiologie der Wurmer^
Pfluger's Archiv, Bd. 56, 1894 ; and Ueber kUnstliche Umwand-
lung positiv helistropischer Thiere in negativ helioiropische und um-

\gekehrt, Pfliigers Archiv, Bd. 54, 1893.

3. Graber. Grundlinien zur Erforschung des Helligkeits-und

loo COMPARATIVE PHYSIOLOGY OF THE BRAIN

Farbensinns der Thiere. Prag und Leipzig, 1884. Verlag von
Tempsky & Freitag.

4. Friedlander, Benedict. Ueber das Kriechen der Regen-
wiirmer^ Biologisches Centralblatt^ Bd. 8 ; and Zur Beurtheilung
und Erforschung der thierischen Bewegungen^ Biolog. Centralblatt,
Bd. II ; and Beitrdge zur Physiologie des Centralnervensy stems und
des Bewegungsmechanismus der Regenwurmer^ Pfluger*s ArchiVy Bd.

58-

5. Maxwell, S. S. Beitrdge zur Gehirnphysiologie der An-
neliden. Pflilger's Archiv^ Bd. 67, 1897.

6. VAN DuYNE, John. Ueber Heteromorphose bei Planarien.
Pflilger's ArchiVy Bd. 64, 1897.

7. Whitman, C. O. The Metamerism of Clepsine. Festschrift
fur Leuckart, Leipzig, 1892.

CHAPTER VII

EXPERIMENTS ON ARTHROPODS

I. Experiments on the lowest animal forms have
taught us that the peculiar reactions of animals are
determined, first, by the different forms of irrita-
bility of the elements composing the tissues, and,
second, by the arrangement of the muscle-fibres.
The central nervous system does not control response
to stimulation : it merely serves as a conductor from
the point of stimulation to the muscle through which
weaker stimuli may pass, and pass more rapidly than
would be possible if the muscle were stimulated
directly.

In the Annelids each ganglion is the relay station
for the sensory and motor nerves of the correspond-
ing segment. If the head exercises a stronger in-
fluence upon the behaviour of the animal than any
other segment, as in Nereis, for instance, I believe
it is due to the fact that in the oral end more kinds
of irritability are present and more peripheral organs
are differentiated (sense-organs, mouth, etc.) than in
the other segments. The fact that in this case the
sympathetic nervous system takes its origin from the

lOI

I02 COMPARATIVE PHYSIOLOGY OF THE BRAIN

supraoesophageal ganglion also helps to increase
the predominance of the head-segments. Hence it
is not the presence of the supraoesophageal ganglion
which determines the greater number of reactions
and their more complicated nature in the oral seg-
ments of some Annelids, but it is the presence of the
greater number of irritabilities and the greater number
of specific organs in the forward end of the body. In
addition there may exist chemical differences between
the various segments of an animal.

We shall now see that this conception of the
central nervous system also holds good for the Ar-
thropods. We will begin the analysis of the brain-
functions of these forms with Limulus polyphemus

(Fig- 30-

Zoologists maintain that Limulus is a very old form.

If tenacity of life favours the age of the species as it
does the age of individuals, this assertion can be readily
understood, for it is difficult to conceive of a tougher
animal. At my suggestion. Miss Ida Hyde made ex-
periments on the functions of the single parts of the
central nervous system of Limulus polyphemus, with
special attention to the respiratory centres (i). Con-
cerning these centres, Faivre had made assertions
which did not harmonise with the apparent segmental
arrangement of the central nervous system. He as-
sumes that the suboesophageal ganglion which is
located in the head has a coordinating influence
on the respiratory movements, but in forms like these
with the respiratory organs (gills) in the abdomen,

EXPERIMENTS ON ARTHROPODS

103

the respiratory nerves must originate In the abdomi-
nal ganglion-chain. The conditions existing in Verte-

I.Afri.

fl. AU

III.AU

Fig. 31.

LiMULUS Polyphemus with the Central Nervous

System Exposed.

Oy supraoesophageal ganglion; c, commissure; «, suboesophageal ganglion ; I-IV{V and
yi Abd.\ abdominal ganglia of the respiratory segments.

brates evidently gave rise to the idea of a coordinating
ganglion located in the head. In Vertebrates the

I04 COMPARATIVE PHYSIOLOGY OF THE BRAIN

diaphragm, the chief respiratory muscle, is far sepa-
rated from its nervous segment, but this is due merely
to a shifting during development. The Anlage is
really located near the head. Such displacements
during growth do not take place to this extent in
Arthropods. Faivre seems to have been entirely
under the influence of current views of Vertebrate
physiology, especially those of Flourens, and so he
was led into making incorrect statements regarding
the physiology of the Invertebrates.

The central nervous system of LImulus consists of
the following parts (Fig. 31) : A supracesophageal
ganglion 0, which Is usually called the brain, an
oesophageal ring {c, Fig. 31) which encloses the oe-
sophagus and consists of fibres and ganglia, a sub-
oesophageal ganglion u, and the ventral chain with six
abdominal ganglia. These parts send out a series
of peripheral nerves. In LImulus the situation of
the nerve-centres is schematically developed : every
peripheral organ has Its nerve-centre in that part
of the nervous system which belongs to its segment.
Perhaps this can be best seen from the following ex-
periment made by Miss Hyde : The whole central
nervous system of a LImulus was removed with the
exception of a little piece of the oesophagus-ring {c.
Fig. 31) and the abdominal ganglia (I-VI Abd, Fig.
31). No connection remained between these two
pieces. The piece of the oesophageal ring lay at the
same height with the three mouth-appendages that
are used for taking in food. These three mouth-

EXPERIMENTS ON ARTHROPODS 105

appendages retained their function, and eating move-
ments were performed reflexly when meat was placed
on the appendages. The rest of the appendages
were entirely paralysed, with the exception of the
gills on the ventral side of the abdomen. The animal
was reduced to a mere eating and breathing machine.
It was fed artificially and so kept alive.

Patten has shown further that each feeding-appen-
dage continues to take food normally and carries it
to the mouth if the piece of the oesophagus-ring
from which its nerves take their origin is preserved.
These feeding-appendages discriminate the chemical
and tactile nature of the food that is offered them,
just as the tentacles of the Actinians do — they refuse
to accept it unless the substances offered fulfil cer-
tain chemical and mechanical conditions. As regards
the conception of these phenomena and their mechan-
ics, no difference, of course, exists between the be-
haviour of the tentacles of the Actinians and the
behaviour of the mouth-appendages of the Limuli ex-
cept that determined by the skeletal relations.

If we remove one half (for instance, the right half)
of the supraoesophageal ganglion in Limulus {0, Fig.
31), the animal usually moves no longer straight
ahead, but in a circle with more or less of a curvature
toward the uninjured (left) side. This is an instance
of the well-known circus-motions. We shall return
to the mechanics of such motions in a later chapter.

If the whole supraoesophageal ganglion be re-
moved, the animal is able to take food placed upon

io6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the mouth-organs, but loses its spontaneity in so far
as this is expressed by progressive movements. It
will even retain abnormal postures in which it is
placed. The operations were performed during the
period of heat. Male Limuli that had lost the supra-
oesophageal ganglion no longer noticed the females.
On the other hand, the legs attempted to remove an
irritating object from the surface of the body. De-
capitated frogs act in the same way.

In the cases mentioned above, the Limuli had en-
dured the operation well and their wounds were
entirely healed. If the oesophageal commissure {c.
Fig. 31) be severed on one side, circus-movements
will appear in the direction of the injured side, but
these only last until the wound is healed. The circus-
motions which ensue upon extirpation of one-half of
the brain also disappear after a time. If ganglia be
removed from the oesophagus-ring, the appendages
corresponding to the extirpated ganglia are perma-
nently paralysed.

2. After extirpation of the subcesophageal ganglion
(u. Fig. 31) the animal remains extremely quiet, and
often lies on the same spot for days. But its respira-
tion continues normal, and this proves the erroneous-
ness of Faivre's opposed assertion. Except for its
immobility and the fact that the extensors of the
joint between the thorax and abdomen are paralysed
as a result of nerve-injuries, the animal appears
normal.

The four or six ganglia of the abdomen (Fig. 31)

EXPERIMENTS ON ARTHROPODS 107

innervate the five gills which are located on the
abdomen of the animal. If the whole central nervous
system with the exception of these ganglia be removed,
the rhythmical respiratory activity continues un-
changed. Immediately after the operation, which is
accompanied by a great loss of blood, the respiration
may be interrupted for an hour or more. If we touch
the gill-plates during this time, the stimulation occa-
sions a series of rhythmical respiratory movements
which, however, soon cease. After a time, the gills
begin their respiratory activity again spontaneously,
and are only interrupted by an occasional cramp.
This interruption of the respiratory movements is also
found occasionally in the normal Limulus, where, if it
remains quiet, the respiratory movements may cease
for an hour or more. At this place we will not go
into details concerning this phenomenon.

The abdominal ganglia are thus centres for the
automatic movements of the abdominal gill-plates.
All the gills move in the same phase. It is probable
that the inspiration begins with the first gill and ex-
tends to the following gills in succession, but rapidly
enough to make the whole appear simultaneous. Ac-
cording to the prevailing opinions, we should be
obliged to assume from this either that only one, for
instance the first, of the four abdominal ganglia, is
automatically active, and that the rest are stimulated
from this, or that, if each of the four ganglia is
rhythmically active, a common centre of coordination
exists somewhere in the four ganglia. If we sever

io8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the ventral cord between two ganglia, for instance
between the second and third, we find that in spite of
the division all the gills continue to breathe. Any
ganglion may be entirely isolated — that is, the commis-
sure before and behind it may be severed, and the
corresponding gill continue to make respiratory
motions. This proves that every ganglion is the seat
of an automatic periodic activity. But how does it
happen that all the gills move simultaneously as long
as their ganglia are connected ? The number of the
respirations produced is the same even when the
abdominal ganglia are isolated. This is probably due
to the fact that the number of respirations is de-
termined by the temperature and the chemical nature
of the blood. The amount of carbon dioxide and
certain other substances, especially those formed in
the muscles (Zuntz and Geppert), controls the number
of respirations. The phase of the movement, on the
other hand, is not the same in the various segments
where the ganglia are isolated. The gills that are
situated anterior to the place of incision may be in-
spiring while those behind the incision are expiring.
These phenomena lead me to believe that in the nor-
mal animal coordination is regulated in the same way
that it is regulated in the activity of the heart and in the
movements of Medusae. The ganglion that acts first,
that is to say, the ganglion that acts quickest, stimulates
those connected with it nervously and so determines
the correspondence of phase. This view is supported
by the fact that no matter how the ganglia may be

EXPERIMENTS ON ARTHROPODS 109

separated from each other, those that are connected
nervously always keep their gills in the same phase of
activity. Were there a centre of coordination in any
ganglion, a group of ganglia separated from this centre
would be active in an uncoordinated manner, but
such is never the case.

3. In higher animals, the conditions controlling re-
spiration scarcely differ from those in Limulus. There
is a series of segmental ganglia in the thoracic portion
of the spinal cord which sends nerves to the thoracic
respiratory muscles of the respective segments. These
ganglia extend into the cervical portion of the spinal
cord, and the fourth, third, and fifth pairs of spinal
nerves give rise to the fibres of the phrenic nerve
which goes to the diaphragm. The diaphragm in
reality belongs to the corresponding segments of the
neck portion, and has attained its present position only
through a shifting of position during growth. One
would expect in text-books of physiology to find the
phenomena of respiration explained as follows :
Chemical changes which are continually going on in
the body, or in these segmental ganglia, under the
influence of heat (the temperature of the body), pro-
duce a periodic activity in these ganglia and conse-
quently in the respiratory muscles. The segmental
connection existing between the ganglia and the
muscles would bring about coordination just as it
does in Limulus. But in the majority of text-books
we find statements of the following character : The
automatic activity of the respiratory muscles is pro-

no COMPARATIVE PHYSIOLOGY OF THE BRAIN

duced much higher up, at a certain point of the
medulla oblongata near the place where the vagus
enters, which Flourens called the noeud vital. This
place is supposed to be the respiratory centre. This
view is justified by two facts : first, the destruction of
the nceud vital causes a cessation of respiration ;
and, second, severing the spinal cord between the
noeud vital and the origin of the phrenic nerve like-
wise causes respiration to cease. These facts do
not justify the conclusion that Le Gallois, Flourens,
and with him the majority of modern physiologists,
have drawn — namely, that the automatic activity of
respiration is located not in the segmental ganglia, but
higher up in the noeud vital. We should have just as
much right to assume that in Limulus the rhythmical
respiratory activity was produced higher up, in the
suboesophageal ganglion for instance, for in this animal,
too, respiration ceases for a time immediately after the
removal of the suboesophageal ganglion. We have
seen in this case, however, that the cessation is only
temporary, and is due to the shock, for respiratory
activity can go on again even when the whole central
nervous system, with the exception of the abdominal
ganglia, has been removed. Neither is the cessation
of respiration in Vertebrates permanent after removal
of the noeud vital or division of the spinal cord be-
tween the noeud vital and the third cervical vertebra.
Langendorff has made the important discovery that
decapitated Vertebrates which have lost the noeud
vital are still able to perform independent respiratory

EXPERIMENTS ON AR THRO PODS 1 1 1

movements (2). It was necessary to make these exper-
iments on young or new-born Vertebrates, as on them
the effect of the shock does not last so long. If one
succeeds in keeping these animals alive by introducing
artificial respiration until the effect of the shock result-
ing from the operation has passed off, spontaneous
respiration begins again. I consider it possible that,
if we could keep an adult Vertebrate alive without the
noeud vital for some time, the respiratory motions
would be resumed again. But why does respiration
stop temporarily after the isolation of the segmental re-
spiratory ganglia from the higher parts of the central
nervous system ? An answer to this question would be
in part an explanation of the mystery of the shock-
effects. It might be possible that something has to be
supplied constantly by certain nerve-elements in the
subcesophageal ganglion or the medulla to the seg-
mental respiratory ganglia, which enables the latter to
be active automatically. In destroying the noeud vital
we perhaps destroy the pathway along which these
constant impulses are carried to the segmental re-
spiratory ganglia in the spinal cord. But where do
these impulses come from and what is their character ?
In watching the respiratory motions of a Limulus, I
received the impression that the operculum always
moves first, and that the respiratory motions of the
lower segments follow successively. In the lower
Vertebrates, e. g, the frog, we have a mouth respir-
ation, whose segmental ganglia are situated in the
medulla. Likewise the segmental ganglia for the

112 COMPARATIVE PHYSIOLOGY OF THE BRAIN

respiratory activity of the gills in fish are situated in
the medulla. Could it not be possible that in Mam-
malians the segmental ganglia for the gill-respiration
continue to be active, although the gills or the oral
respiration have disappeared ? If this were so, we can
understand that the segmental ganglia for gill-respi-
ration in the medulla begin to be active first. Their
activity is the stimulus for the activity of the next
lower segmental ganglion, and so on.

If we cut the cord between medulla and phrenic
nerve, respiration must stop. But if we could keep
such an animal alive long enough, the lower segmen-
tal ganglia would be altered in such a way as to
breathe automatically again.

That the shock-effect after such an operation can-
not be due to an exhaustion of the phrenic ganglia is
made obvious by the following experiment : W. T.
Porter made hemisections of the spinal cord between
the medulla oblongata and the origin of the phrenic
nerves (3). If one half, for instance the left half, of
the medulla be cut, the left half of the diaphragm no
longer partakes in the respiratory movements, while
the respiratory motions of the right half continue.
But if the right phrenic be cut, the left half of the
diaphragm begins its rhythmical motion again, while
the right half of the diaphragm stops breathing. It
is of course, at present, just as impossible to explain
why the cutting of the right phrenic nerve causes the
left half of the diaphragm to breathe again, as it is to
explain why a frog that had lost its spontaneity after

EXPERIMENTS ON AR THRO PODS 1 1 3

an operation In the thalamus opticus begins to move
spontaneously again if the optic lobes and the pars
commissuralis of the medulla are removed.

In Limulus an anterior and a posterior nerve
originate from every ganglion of the ventral chain.
It was interesting to determine whether these nerves
have functional differences like those of the anterior
and posterior roots of the spinal cord of Vertebrates.
It has been maintained that Arthropods are Verte-
brates that walk on their backs. Faivre has stated
that there is not only a separation of the motor and
sensory roots in Arthropods, corresponding to Bell's
law, but that also in Arthropods, in contrast with Ver-
tebrates, the ventral side of the ganglia is sensory, the
dorsal motor. Now this Is not true of the nerve-
roots which start from the ganglia in Limulus. If the
posterior nerve be severed and Its peripheral stump
stimulated, we get Inspiratory movements of the half
of the gills to which this nerve goes. All the other
gills are unaffected. Hence this nerve contains
motor fibres. If the ventral stump be stimulated,
the whole animal becomes much excited. From this
we see that the posterior nerve also contains sensory
fibres. If the anterior nerve be severed, stimulation
of the peripheral stump has no effect. Stimulation
of the central stump excites the entire animal. Hence
the anterior nerve is purely sensory. Limulus Is
better adapted for deciding this question than the
smaller Arthropods. The conditions in the latter are
probably the same as in the former, for Vulplan (4)

114 COMPARATIVE PHYSIOLOGY OF THE BRAIN

and latterly Bethe (5) energetically reject the Idea
that dorso-ventrally the gangllon-chain of Arthropods
is the reverse of the spinal cord of Vertebrates.

4. We will now turn our attention to the crayfish
as the next representative of the Arthropods whose
braln-physlology has been carefully Investigated. Fig.
32 gives a diagram of the central nervous system of
the lobster, which Is almost Identical with that of the
crayfish, o Is the supracesophageal ganglion with
the nerves for the eyes and antennae. In addition It
gives off the sympathetic nervous system which goes to
the Intestine. Both oesophageal commissures, c, go
backwards to the subcesophageal ganglion, u. The
latter is seemingly one ganglion, but it supplies six
pairs of segmental organs, namely, the mouth-ap-
pendages. The microscopical examination shows that
this subcesophageal ganglion In reality consists of six
separate ganglia. We often meet with a fusion of
ganglia, and consequently an apparent lack of clear-
ness In the segmental arrangement. It is due to this
fact that in the brain-physiology of Vertebrates the
segmental arrangement of the central nervous system
has been left entirely out of consideration. Next
after the subcesophageal ganglion come the five
thoracic ganglia (I-V T, Fig. 32) belonging to the
segments of the forceps and the four pairs of loco-
motor appendages. In addition to these, there are
the five ganglia of the abdomen (I-V Abd,, Fig. 32)
that Innervate the swimmerets, and the tail, which
serves as a swimming-organ. The best experiments

Fig. 32. Lobster with Central Nervous System
Exposed.

.,supra«sophageal ganglion (brain); ..commissure; «, ^f oesophageal
ganglion; /-Fr, five thoracic ganglia ; /-K^ 3^., first five abdominal

ganglia.

ii6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

on the central nervous system of these animals have
unquestionably been made by Bethe, and we shall in
the main follow his presentation. Many of the facts
which Bethe describes from the animals used in his
experiments are familiar to me from personal observ-
ation, and I am convinced that the picture he gives
is correct.

If in a crayfish both the commissures {c. Fig. 32)
which connect the supraoesophageal ganglion 0 with
the rest of the brain be severed, the behaviour of the
animal is no longer controlled by the brain 0, It does
not make spontaneous progressive movements. When
stimulated it begins to move, but after having gone
about 20 cm. it stops. This lack of spontaneous pro-
gressive movements agrees with the description given
by Flourens of the Vertebrate from which the cere-
bral hemispheres had been removed. Flourens's repre-
sentation was wrong, however, for a dog operated
upon in this way shows increased spontaneity in its
progressive movements.

Annelids and Arthropods are closely related as re-
gards the central nervous system. However, Nereis
shows an excess of progressive movements after
removal of the supraoesophageal ganglion, while As-
tacus no longer moves spontaneously. I believe that
the difference depends only upon circumstances of
minor importance. Ward has already found — and
Bethe has confirmed the fact — that in brainless cray-
fish the legs are unceasingly active, either cleaning
each other or performing pendulum -movements.

EXPERIMENTS ON AR THRO FOBS 1 1 7

They, however, make no progressive movements. I
beheve this is due possibly to a secondary effect of
the extirpation of the supraoesophageal ganglion.
The legs of such an animal have an abnormal posi-
tion, being more strongly flexed at the joints nearest
the body than they are normally. The tension of
the extensors probably suffered severely from the
operation. Such mechanical disturbances might easily
cause difficulty in locomotion, while simple pendulum-
movements of the legs, which require practically no
labor, could still be performed. The fact that after re-
moval of the brain of crayfish the tension of the flex-
ors predominates in certain joints is of interest, as we
meet with the same phenomenon in dogs that have
lost the anterior region of the cerebral hemispheres,
and as it also comes to our attention in man after
apoplexies which result in the paralysis of an arm.

Bethe concludes from these pendulum-movements
that the brain is an organ of inhibition. As regards
this, the remarks hold good that have already been
made in this connection on annelids (see p. 94).

The weakening of the muscles in the crayfish whose
brain has been extirpated shows itself also in the fact
that the forceps no longer pinch as hard as those of
normal animals.

After what has been said concerning the segmental
character of the central nervous system, it is to be
expected in the crayfish that, since the segmental
ganglia of the organs of mastication are located in
the suboesophageal ganglion, extirpation of the supra-

ii8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

oesophageal ganglion will not interfere with the nor-
mal character of its eating movements. I give the
description in Bethe's words : ** The animal devoid of
the supraoesophageal ganglion is able to eat and
selects its food. It is true that pebbles, small
pieces of wood, etc., are seized by the forceps of the
front pairs of legs, but when brought near the mouth
they are rejected. A piece of meat, however, is
always taken into the mouth and masticated. The
swallowing is difficult, just as in the case of Carcinus.
The piece often remains for a long time between the
maxillipedes without being swallowed, and at last falls
to the ground. Pieces of paper that have been satu-
rated with meat-juice are treated in the same way.
Stones that have been covered with meat-juice are
also brought to the mouth, but no attempt is made to
masticate them. They are usually dropped as soon
as they come in contact with the maxillipedes." Thus
we see that the nature of the stimulus determines the
results just as in the case of Actinians. The brain of
the crayfish has nothing to do with these reactions.
The central nervous system is in this case to be con-
sidered only as an organ for the conduction of stimuli,
a function that could just as well be performed by plant
protoplasm or muscle-tissue as by nerve-protoplasm.
In the crayfish, the original segmental arrangement
of the nervous elements is so well preserved that re-
moval of the brain does not interrupt the proto-
plasmic nervous connection between the surface of the
mouth and the muscles of the thoracic appendages.

EXPERIMENTS ON AR THRO PODS 1 1 9

When these animals from which the brain has been
removed are laid on their backs they return to the
ventral position.

The observations made by Bethe In crayfish in
which he had severed the oesophageal commissures
on only one side are interesting. The division was
made on the right side. If he touched the left side of
the head of such an animal, first the forceps of the
stimulated side reached toward the stimulated spot
and then those of the other side followed with accur-
acy. At the same time, the animal attempted to escape
backwards. If the same stimulus was applied to the
right side of the head, the forceps did not react. Even
with a strong stimulus no reaction followed. Hence
the stimulus that produces the localising reflex can only
be transmitted through the longitudinal commissure of
the same side to the appendages. This seems to hold
good generally for the Arthropods, since Bethe was able
to prove it in Carcinus, Squilla, and Hydrophylus. It
seems to hold good not only for the conduction through
the oesophageal commissures but through all longitu-
dinal commissures. After division of an oesophageal
commissure, circus-motions often but not always occur
toward the normal side. The animal is also able to
move straight ahead, but this requires some effort.

If the right oesophageal commissure be severed, the
tonus of the muscles on the right side (the injured
side) of the abdomen is diminished, and as a result
the abdomen is curved toward the left and becomes
concave on that side.

I20 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Division of the brain in the middle Hne — that is,
separation of the two halves of the brain — destroys the
geotropic reactions of the eyestalks. It is still more
remarkable that such animals no longer prefer to re-
main in the dark like normal animals.

If the longitudinal commissures be divided between
the mouth-ganglia (subcesophageal ganglion) and the
ganglion of the chelae, all locomotor movements be-
come impossible, although the legs are not paralysed.
This is strange, because the subcesophageal ganglion
contains the segmental nerve-elements of the oral ap-
pendages but not those of the locomotor appendages.
In other Crustaceans, extirpation of the subcesophageal
ganglion has no such paralysing efifect on the loco-
motor movements. It is impossible to tell at present
what causes the exceptional behaviour of Astacus in
this regard. I do not believe we are obliged to as-
sume that this is an instance of a deviation from the
laws of the segmental arrangement of the nerve-ele-
ments (centres) of the limbs. This is shown by the
fact that the legs of such an animal are not paralysed,
but are unceasingly occupied in cleaning the abdo-
men, the pedes spurii, or each other. Indeed, more
than that, *' if we give one of the forceps of a loco-
motor appendage a piece of meat or paper, other
legs approach immediately, seize the meat, and carry
it to the mouth," in spite of the fact that all nervous
connection between the nerves of the mouth-organs
and the legs has been severed. It is true that the ap-
pendages around the mouth often refuse to accept

EXPERIMENTS ON ARTHROPODS 121

and forward the pieces of meat that are thus offered
them by the legs.

As regards the further Isolation of the ganglia that
are located posterior to the suboesophageal ganglion,
the facts which have been described in LImulus are
in general true. As long as the ganglion of a seg-
ment remains connected with the segmental organs,
the functions of that segment remain unimpaired.
Bethe has found single exceptions to this rule, but it
is conceivable that these exceptions are shock-effects
resulting from the operation.

We will now report more briefly concerning Bethe's
experiments on some other Arthropods.

5. Squilla no longer swims spontaneously after the
supraoesophageal ganglion has been isolated (that is,
after division of the commissure between the supraoe-
sophageal ganglion and the mouth-ganglion). The
spontaneous progressive movements usually seem to
be destroyed. When stimulated, however, the ani-
mal moves normally. The nervous mechanism for
the locomotor reflexes is localised in the three ganglia
of the locomotor appendages, that is, these append-
ages still move normally, even though the connection
with the ganglia lying in front has been interrupted.

In grasshoppers (JPachytylus cinerascens) isolation
of the supraoesophageal ganglion causes the spontan-
eous progressive movements to cease. These animals,
after the operation, clean their antennae with their
fore-legs like a normal animal. According to Bethe
these localised reflexes of the legs are produced by

122 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the stimulus of the operation. Abnormal positions
of the legs, resulting from the operation, occur just as
in Astacus and Squilla.

If the supra- and subcesophageal ganglia be re-
moved by decapitating the animals, they are still able
to perform some walking movements, and especially
hopping movements, when stimulated. This agrees
with the idea of the preservation of the purely seg-
mental arrangement of the nerve-connections. Yer-
sin s experiments on crickets are very significant in
this regard. I take them from Bethe's paper. They
deal with crickets from which both longitudinal com-
missures had been removed between the subcesophag-
eal ganglion and the first thoracic ganglion. Yersin
kept these animals alive for weeks. "■ When laid on
their backs they were able to turn over. When stim-
ulated they moved forward a few steps, or to the side,
according to the point of stimulation. In doing so
they occasionally tumbled over. When stimulated,
they still made attempts at flying without being able
to lift themselves from the ground." Yersin observed
that a male and a female, both of which had been op-
erated upon in this way, were able to pair. Of course
it was necessary to place the male on the female in
this case. The male on which he made this observ-
ation had already given off a spermatophore.

Bees lived only a short time after the extirpation
of the supracesophageal ganglion. The bee shows
the same restlessness that was noticed in Astacus.

Bees whose brains were divided lengthwise in

EXPERIMENTS ON ARTHROPODS 123

symmetrical halves showed a perceptible functional
disorder only in their behaviour toward the hive. *' If
carried back to it they crawl about on the board be-
fore the entrance but make no attempt to enter, and
they pay no attention to their companions."

If bees be decapitated, the supra- and suboesopha-
geal ganglion being thus removed, they are still able
to walk, although awkwardly. When laid on their
backs they turn over with the help of their legs.
" When stimulated on the ventral side they grasp the
object (pencil) with their legs, pull it toward them,
bend the abdomen, and attempt to sting it." But not
all anirnals give such favourable results. In brain-
physiology only those animals operated upon which
show the slightest disorders can be considered, be-
cause the exhaustion may render the rest of the cen-
tral nervous system pathological.

As was to be expected a priori on the basis of the
segmental theory, the stinging-reflex is possible as long
as the abdominal ganglion is preserved. Bethe showed
that the abdomen, when severed from the body, still
bends if stimulated on the ventral side and reaches
the stimulated spot with the outstretched sting. At
the same time poison is ejected. The reflex also
continues when all the abdominal segments with the
exception of the last one have been amputated.

If the supraoesophageal ganglion in a water-beetle
(Hydrophilus) be extirpated, the progressive loco-
motor movements are not only not interrupted, but the
animal goes about almost unceasingly, showing only

124 COMPARATIVE PHYSIOLOGY OF THE BRAIN

modifications which suggest that the relation in the ten-
sion of the antagonistic muscles of its legs is changed.
The animal turns out for obstacles that come in its
way. If a beetle that has lost the supracesophageal
ganglion is thrown into water it swims off, drawing in
the first pair of legs. The normal water-beetle rests
quietly under dark objects. The water-beetle whose
supracesophageal ganglion has been divided by a
longitudinal incision no longer shows these reactions,
although the light is still able to produce other effects
in the animal. If suddenly exposed to strong light
or a dark shadow — when in motion, namely — it ceases
to move.

An animal whose right oesophageal commissure has
been severed does not brace itself against obstacles as
strongly with the legs of the right side as with those
of the left side. It seems to me this shows that the
extensors of the right side are weakened, as it is they
that have to perform the task of bracing. Further-
more, the right legs are moved constantly. It may
be possible that these two facts are in some way con-
nected. The decreased opposition of the extensors
renders the pendulum-movements of the legs easier.
The same explanation may hold for Astacus, bees, etc.
If the supra- and suboesophageal ganglia be extirpated,
the animal only makes progressive movements when
stimulated. But the ability to perform coordinated
progressive movements is not destroyed. When laid
on its back the animal still tries to regain the ventral
position, but the efforts made by the legs are vain.

EXPERIMENTS ON ARTHROPODS

125

If put under water it still makes swimming move-
ments, but these do not help it forward.

The experiments on the other ganglia of this ani-
mal performed by Bethe do not concern us in this
book. We will only quote that result of Bethe's
which is of most importance for our purpose : '* Nei-
ther the suboesophageal nor the prothoracic ganglion
is the seat of the reflex for righting the animal when
turned on its back, nor of the coordination of the mus-
cles of locomotion, walking, or swimming, as Faivre
maintains. It would seem as though these reflexes
were located rather in each thoracic gayiglion for the cor-
responding segment T This last sentence expresses the
principal truth for all complicated central nervous sys-
tems. Each segment of a segmented animal may be
regarded as a simple reflex animal, comparable to the
Ascidian, and the analysis of the reflexes depends
upon the same principles and leads to the same re-
sults in both cases. The complication that appears
in segmented animals consists in the fact that when a
process of stimulation takes place in one segment it
is communicated to the neighbouring ganglia, and
these ganglia produce processes of the same kind. It
is possible that the nature of the stimulation also
helps to determine the nature of the movement. The
assumption of special centres of coordination is
superfluous. One other fact is of importance in exper-
iments in extirpating and severing nervous connec-
tions, namely, that the division may bring about in
those parts which are protoplasmically connected with

126 COMPARATIVE PHYSIOLOGY OF THE BRAIN

the place of operation a change which is sometimes
transitory, sometimes permanent — the so-called shock-
effects. The highest degree of these shock-effects is
attained in case of degeneration. It is a remarkable
fact that, in an operation on the central nervous sys-
tem, the effect of the shock is much greater in the
part posterior to the place of operation than in the
anterior part, toward the head. This may indicate
that there is a constant current of impulses or influ-
ences in the direction from the brain to the posterior
parts of the central nervous system. The interrup-
tion of these influences may be responsible for the
condition which we call shock-effects and which may
be transitory. These shock-effects are incomparably
less strong in cold-blooded than in warm-blooded ani-
mals. We do not possess enough facts to enable us
to give an explanation of the shock-effects.

Bibliography.

1. Hyde, Ida H. The Nervous Mechanism of the Respiratory
Movements of Limulus Polyphemus, journal of Morphology^ vol.
ix., 1894.

2. Langendorff, O. Studien iiber die Innervation der Athembe-
wegungen. I. Mittheilung. Archiv f. Physiologie, 1880.

3. Porter, W. T. The Path of the Respiratory Impulse from
the Bulb to the Phrenic Nuclei, journal of Physiology, vol. xvii.,
1894-95.

4. VuLPiAN. Lefons sur la Physiologic ginirale et comparee du
Systlme JSferveux. Paris, 1866.

5. Bethe, A. Vergleichende Ui^tersuchungen ilber die Eunctionen
des Centralnervensystems der Arthropoden. Pfliiger's Archiv^ Bd.
Ixviii., 1897.

EXPERIMENTS ON ARTHROPODS 127

6. Bethe, a. Das Centralnervensystem von Carcinus mcenas.
Archiv f. microskop. Anatomie und Entwicklungsgeschichte^ Bd. 1.,
1897 ; Bd. li., 1898.

7. Steiner, J. Die Functionen des Centralnervensystems und
ihre Phylogenese. III. Abtheilung. Die wirbellosen Thiere.
Braunschweig, 1898.

CHAPTER VIII

EXPERIMENTS ON MOLLUSKS

The literature on the functions of the central
nervous system of Mollusks is extremely meagre.
It is nevertheless valuable, as it furnishes us with
further proofs of the theory that the simple and
rhythmical spontaneity, as well as reflex processes, do
not depend upon the brain or specific peculiarities of
the ganglia. A Gastropod whose brain (^, Fig. 33)

has been removed
continues to move
spontaneously. Stei-
ner has observed
this in a transparent
pelagic species of
snail, Pterotrachea,
that is about 10 cm.
long (i). The foot
of this snail has been
transformed into a
swimming organ. Neither one- nor two-sided de-
struction of the supraoesophageal ganglion has the
slightest influence upon the character and the

128

Fig. 33. Schematic Representation of
THE Central Nervous System of a
Snail (Paludina Vivipara).

J", brain ; /', pedal ganglion. (Modified after Leydig.)

EXPERIMENTS ON MOLLUSKS

129

quantity of the spontaneous progressive movements.
Destruction of the pedal gangUon, on the other hand,
puts an end to all locomotion. Steiner concludes,
therefore, that "the
pedal ganglion alone
has control of the entire
locomotion of the ani-
mal." This anthropo-
morphic conclusion
goes too far. The only
conclusion we are justi-
fied in drawing from
this observation is, that
the protoplasmic con-
necting fibres between
the skin and the foot-
muscle of the animal
pass through the gan-
glion. Steiner further attempted to see if he could
produce circus-motions by means of a one-sided divi-
sion of the oesophageal commissure in other Mollusks,
Pleurobranchia and Aplysia. He succeeded no better
than in Pterotrachea. One-sided destruction of the
pedal ganglion in Cymbulia, however, caused paralysis
of one-half of the locomotor organ. The animal
naturally moved in a circle, for only one wing served
as an oar.

The Cephalopods have an extremely complicated
brain (Fig. 34). It consists of a dorsal and a ventral
mass, each of which is composed of several ganglia.

Fig. 34. Brain of Sepia.

Cg^ cerebral ganglion ; Spg^ supraoesophageal
ganglion ; Bg^ buccal ganglion ; Tg^ ganglia
of the tentacles. (After Glaus.)

I30 COMPARATIVE PHYSIOLOGY OF THE BRAIN

The dorsal and ventral ganglia are connected by
commissures. In addition, they possess a series of
peripheral ganglia, the tentacle-ganglia {Tg, Fig. 34),
for instance. It is of significance for the segmental
theory that the tentacle-ganglia suffice to produce ten-
tacle-reflexes, as V. Uexkull has shown In Eledone (2).
It has been inferred from experiments on Vertebrates
that peripheral ganglia cannot transmit reflexes.

Now, as regards experiments on the brain of Ceph-
alopods, Steiner reports as follows concerning Oc-
topus vulgaris: ''If the dorsal ganglion on one
side be removed, or both commissures of one side be
severed, not the slightest change is visible in the life-
processes of the animal, for It moves spontaneously as
before, attacks Its prey {Carcinus mcenas^ cleverly,
and devours It. But the picture changes if the dorsal
ganglion be entirely removed. To be sure the two
forms of locomotion are preserved, for the animal
creeps with the aid of its arms, or shoots like an arrow
through the waves, when water is forced out of the
mantle-cavity rhythmically. These movements are,
however, no longer spontaneous^ for they occur only
when the animal Is stimulated, neither does it take
Its food spontaneously. The normal octopus, which
is endowed with marked intelligence [ ? ], is wont to
observe its surroundings most attentively, but now It
sits indifferent to Its surroundings, as though idiotic,
and only its regular breathing gives evidence that it
still lives. Vision Is unimpaired, for it draws back
when a stick Is brought toward its eye." V. Uexkull's

EXPERIMENTS ON MOLLUSKS 131

article on Eledone is more exhaustive than Steiner s.
One of his observations, describing the extraordinarily
excited condition of an animal whose cerebral gan-
glion had been removed, is worthy of mention. " All
the reflexes seemed increased. When anyone ap-
proached the basin the Eledone that had undergone
this operation swam off, while the normal animals
remained quiet. There was an incessant play of
colors. During the second night, in spite of the pro-
tecting net, it escaped and died on the floor of the
laboratory." V. Uexkull concludes from this that there
are inhibitory centres in the cerebral ganglion. We
have seen that Bethe arrived at a similar conclusion
in regard to the supraoesophageal ganglion of the
Arthropods. We have discussed this possibility in
connection with Maxwell's experiments on Nereis.

The arm-nerves originate in the pedal ganglion.
But the latter is connected with the supraoesophageal
ganglion directly by means of the anterior commis-
sures and indirectly by means of the posterior commis-
sures. Now it is of interest to know that the influence
which the anterior part of the supraoesophageal gan-
glion exerts on the arm-movements when stimulated
is exactly the opposite of that exerted by the posterior
part ; if the entire supraoesophageal mass between
both pairs of commissures be separated by a frontal
incision and both stumps be stimulated down deep,
where the central ganglia are located, according to v.
Uexkull, we obtain the following results : Stimulation
of the anterior stump causes the cup-like suckers to

132 COMPARATIVE PHYSIOLOGY OF THE BRAIN

take hold strongly ; stimulation of the posterior stump
causes the suckers to let go and the arms to be with-
drawn. Thus the antagonistic activities of the arms
depend upon two different parts of the central nervous
system. " An animal whose supra oesophageal mass
has been divided in the vicinity of the first central
ganglion behaves like an animal that is only able to
take hold of objects. It grasps every object firmly
and liberates itself again only with difficulty. It
usually retains its hold and sits with extended arms,
or crawls forward with the greatest difficulty. Such
an animal placed on the back of a torpedo seizes it
firmly with the arms, and no shocks of the electric
organs are of avail to rid the fish of its burdensome
rider. On the other hand, it is evident that the Ele-
done only participates in the ride involuntarily from
the fact that it becomes dark brown and throws ink.
If a normal Octopus by mistake grasps after a tor-
pedo, it never remains in so dangerous a neighbor-
hood more than a few seconds [I have observed
this in Octopus, never in Eledone]." It seems to me
that the conclusion to be drawn from these facts is,
that the anterior and posterior parts of the supra-
oesophageal ganglion are connected with antagonistic
muscle-groups. This relation is of interest in view of
galvanotropic experiments, which we shall discuss
later on. It is furthermore probable from v. Uex-
kiill's experiments that the act of eating depends upon
the integrity of the first central ganglion, while the
second and third central ganglia are necessary for all

EXPERIMENTS ON MOLLUSKS

133

the remaining functions of the arms, for instance, loco-
motion and steering.

The fact discovered by v. Uexkull that the basal
ganglion, when no longer connected with the central
nervous system, produces coordinated chewing move-
ments when stimulated is of great importance for the
segmental theory. The skin and muscles are in this
case connected by nerve-fibres which do not pass
through the central nervous system but through a
peripheral ganglion that v. Uexkull terms the bucco-
intestinal ganglion. This is another fact that speaks
for the idea that the ganglia are only to be considered
as organs of transmission, that is, as connecting pro-
toplasmic threads for reflexes, and not as bearers of
mysterious reflex mechanisms.

BIBLIOGRAPHY.

1. Steiner, J. Die Eunctionen des Centralnervensy stems der
wirbellosen Thiere. Sitzungsberichte der Berliner Academie der
Wissenschaften^ 1890? i-* P- 32.

2. V. Uexkull. Physiologische Untersuchungen an Eledone
moschata. Zeitsch. f. Biologie^ Bd. xxxi., 1895.

3. Steiner, J. Die Eunctionen des Centralnervensy stems und
ihre Phylogenese. III. Abtheilung. Die wirbellosen Thierey Braun-
schweig, 1898.

CHAPTER IX

THE SEGMENTAL THEORY IN VERTEBRATES

I. The segmental arrangement of the central nerv-
ous system of Vertebrates is suggested by the ar-
rangement of the spinal nerves. The number of
segmental ganglia present in the head exceeds the
number of cranial nerves. The auditory nerve and
the vagus, for instance, originate from more than one
segment each. Dohrn, Locy, and others have shown
this. Locy states that there are originally fourteen
segments in the head of the embryo of the shark,
while there are only twelve cranial nerves. Physi-
ology is more interested in the decision of this quest-
ion than morphology, because upon it depends the
theory of coordinated movements. The question
of segmentation may also be of importance indi-
rectly in connection with the idea of localisation in
the cerebral hemispheres, for the so-called centres of
the cerebral cortex are merely the places where the
fibres from single segments of the central nervous
system enter.

The spinal nerves originate in the spinal cord, and,

134

EXPERIMENTS ON VERTEBRATES 135

as has been said above, suggest externally its segmental
character. Each has a ventral motor and a dorsal
sensory root, and we desire to call attention to the
fact that the dorsal root passes through a ganglion
(the spinal ganglion). If the ventral root be severed,
paralysis of the muscles of the corresponding segment
occurs. If the dorsal root be severed, the correspond-
ing segment becomes insensible, or, more properly
speaking, the transmission of impulses which proceed
from the periphery to the muscles of this segment and
to the remaining segments becomes impossible. The
operation itself, however, has still another influence
on the tension of the muscles of the same segment
(perhaps also of other segments). The amount of the
muscle-tension under normal conditions varies (prob-
ably with the chemical conditions of the muscles). If
a muscle be stretched with a certain weight it attains a
certain length. But if the posterior nerve-roots be
severed while the muscle is still nervously connected
with its segment the muscle lengthens (E. v. Cyon).
The operation causes a shock, in other words, prob-
ably a chemical change in the muscle. The nature of
this change is as yet unknown. This influence of the
posterior roots on the muscles shows itself also in the
movements of an animal in which the posterior roots
of the hind-legs have been severed : the movements of
the legs are disturbed. It is known that the nerves
of the brain are also of segmental origin, only in this
case the inequalities of growth obliterate externally
the segmental relations. From the fact that the chiefly

136 COMPARATIVE PHYSIOLOGY OF THE BRAIN

sensory trigeminus, which may be considered as the
posterior root of the faciaHs, possesses a peripheral
gangHon (gangHon Gasseri), while the chiefly motor
facialis has no peripheral ganglion, Bell concluded
that the posterior roots of the spinal cord, which pos-
sess peripheral ganglia, are sensory, while the anterior
roots, which possess no ganglia, are motor. Bell
found (by means of vivisection) that division of the
trigeminus produces disturbances in eating in those
animals that take their food with the lips : these
disturbances are caused, naturally, by the weakness of
the corresponding muscles.

We will add a word here concerning the import-
ance of the ganglion-cells for the preservation of the
axis cylinder. The axis cylinder may be regarded as
a protoplasmic extension of a ganglion-cell, which
lives only as long as it is connected with the cell.
Now the ganglion-cells of the dorsal roots are located
in the spinal ganglion, those of the ventral roots in
the ventral horns of the spinal cord. If the posterior
roots be severed, that part of the fibres which is con-
nected with the spinal cord degenerates, while the
part that is connected with the spinal ganglion is
preserved, and grows or regenerates. If the ventral
roots be severed the peripheral stump degenerates,
while the stump that is still connected with the spinal
cord is preserved and grows. We may mention here
briefly that the nerve-fibres of the posterior roots,
according to Golgi's school, are not fused with the
ganglion-cells of the posterior horns in the spinal

EXPERIMENTS ON VERTEBRATES 137

cord, but are only in contact with them/ For the
transmission of the impulse this fact is of no import-
ance ; it is not necessary in either case that the gan-
glion-cells of the posterior horn and the sensory
nerves be grown together, they need only to be in
sufficiently close contact. Engelmann called atten-
tion to these relations long ago in his excellent article
on conduction in the ureter.

2. Sufficient data exist for proving the segmental
localisation of reflexes in the spinal cord. In a dog
whose spinal cord has been severed somewhere in the
thoracic region the posterior part is entirely separated
from the anterior part as far as the motor and sensory
functions are concerned. Immediately after the oper-
ation severe shock-effects appear, but these are only
temporary, and we shall return to this subject later.
The interruption of the continuity is permanent, for
in the central nervous system of higher animals no re-
generation has been observed, but only a healing
together of the cut surfaces by means of connective
tissue. In such an animal the part located behind the
point of division shows all the reactions which are
possible in the corresponding segments. Goltz has
proved this for dogs. Rubbing of the skin produces
scratching movements of the hind-legs ; erection of
the penis and urination can be produced by stimulat-
ing the foreskin. The reflexes of the rectum and
bladder and the vasomotor reactions are intact. We

' Apathy's publications arouse suspicion as regards the results obtained by
Golgi's methods.

138 COMPARATIVE PHYSIOLOGY OF THE BRAIN

have already called attention to the fact that the re-
spiratory movements are segmental processes. Goltz
has shown that those reflexes in which the muscles of
the arms are active are also segmental (2). During
the period of heat the male frog clasps the female
with his fore-limbs. If the head and back part of the
body of a male frog be amputated during this time,
so that only a piece consisting of the arms and the
segmental piece of the spinal cord belonging with the
arms remains, rubbing the skin on the ventral side of
this piece suffices to produce the clasping reflex.

3. In considering the brain of Vertebrates we are
obliged to deal with the brain of the cold-blooded
animals, for the simple reason that in warm-blooded
animals we cannot well perform brain-operations in
the vicinity of the medulla oblongata without having
the respiration cease. In cold-blooded animals the
shock-effects are not so great. We have selected the
brain of the frog as a type because it has been worked
out the most carefully. It consists chiefly of the
cerebral hemispheres {GH, Fig. 35), thalamus op-
ticus {Th., O), optic lobe, cerebellum {KH), and
medulla oblongata. The diagram (Fig. 35) gives the
origin of the nerves of the brain (V-XI). It is our
aim to show in this chapter that the individual activi-
ties of the frog are dependent upon the segmental
ganglia and that we have no right to speak of ** cen-
tres" for the single activities unless the word centre Is
synonymous with the expression segmental ganglion.

We will first consider the coordinated progressive

EXPERIMENTS ON VERTEBRATES

139

movements. It was for a long time a dogma that
progressive locomotive movements could only be per-
formed by frogs that were still in possession of their
cerebral hemispheres.
This statement was
made by Flourens. He
observed that frogs
devoid of the cerebral
hemispheres no longer
move spontaneously
(3). Later on Schra-
der showed that this
observation was not
correct ; that this lack
of spontaneity only
occurs when the thai-
ami optici are injured
(4). Are we to con-
clude from this that
the power of spontan-
eous locomotion is lo-
cated in the thalami
optici ? This would be wrong, for if the whole brain of
a frog including the pars commissuralis of the medulla
oblongata be removed, it seems '' possessed of an irre-
sistible desire to move ; it creeps about untiringly in
an entirely coordinated manner and does not rest until
it comes to a corner of the enclosure" (Schrader).
It behaves like the Nereis in Maxwell's experiments
which was deprived of its brain. Flourens made his

Fig. 35. The Frog's Brain.

GH^ cerebral hemispheres; Th.Oy thalamus opticus;
Lob. opt^ lobi optici ; KH^ cerebellum ; V-XIy
origin of the 5th to nth brain-nerves. (After
Wiedersheim.)

140 COMPARATIVE PHYSIOLOGY OF THE BRAIN

localisation too high up. We wish to emphasise the
fact that the frogs from which Schrader removed the
whole brain, including the pars commissuralis, not
only moved but were still able to climb. The con-
dition of rest which appears after injury to the thala-
mus is thus not due to the loss of spontaneity. Steiner
has also attempted to localise locomotion in a ** cen-
tre." He found that frogs after losing the pars com-
missuralis of the medulla made no more progressive
movements, and concluded from this that the sole
and undivided control over all locomotions of the body
belongs to this part (5 and 6). Schrader's contradic-
tory results overthrow Steiner's conclusions. The
latter author evidently made his observations on mor-
ibund animals, for his frogs survived the operation
only a week at the most, while Schrader's lived many
months and entirely recovered from the operation.

According to the segmental theory, on the other
hand, it is to be expected that only those parts of the
central nervous system are necessary for locomotion,
which correspond to the segments of the muscles of
the arms and legs. Thus it must be possible to ob-
tain coordinated locomotion as long as the segmental
ganglia of the muscles of the arms and legs are intact.
This agrees with the result obtained by Schrader
that after extirpating the whole brain, including the
pars commissuralis, coordinated locomotion still oc-
curs. We can go still farther and extirpate the whole
medulla as far as the tip of the calamus scriptorius
and still obtain coordinated locomotion. ** Disturb-

EXPERIMENTS ON VERTEBRATES 141

ance of the coordination in movements first begins with
the apparent decrease in the ability to use the fore-
legs, which becomes more and more apparent the
nearer the incision approaches the origin of the bra-
chial plexus from the tip of the calamus scriptorius.
When this is reached the animal falls flat on its belly ;
the fore-legs are no longer able to carry the body. If
the animal be stimulated in the middle-line, for in-
stance at the anus, the hind-legs throw the body for-
ward. The forward extremities participate still with
* alternating ' but insufficient and peculiar trembling
movements. A really coordinated progressive move-
ment no longer takes place." We see again in this
case the entire validity of the segmental theory : in-
juries of the spinal cord in the vicinity of the brachial
plexus interfere with the walking movements only in
so far as the cooperation of the fore-legs comes into
consideration. The hind-legs, on the other hand,
continue to function normally. Similar phenomena
may be observed in fishes. They also cease to move
about when the brain is removed up as far as the
medulla oblongata. It would be quite wrong, how-
ever, to conclude from this that the centre of locomo-
tion is located in the medulla oblongata. If the head
of a shark be amputated the body swims about spon-
taneously. This experiment was made by Steiner.
From the standpoint of the segmental theory this re-
sult was to be expected. The tail is the organ of
locomotion for the shark, and only the corresponding
segmental ganglia of the spinal cord are required for

142 COMPARATIVE PHYSIOLOGY OF THE BRAIN

its activity. If the spinal cord of a young salamander
be severed the swimming movements of the anterior
and posterior parts are so well coordinated that it
hardly seems credible that an operation has been per-
formed. The same is true of the eel (8). The con-
ditions are about the same as in the earthworm.

Rubbing the back of the frog causes it to croak, and
in a frog whose brain has been removed as far as
the medulla oblongata, this sound, as Goltz has
found, can be produced with machine-like regularity
(2). Viewed from the segmental standpoint this reflex
is naturally conditioned by the integrity of the me-
dulla, since it is there that the motor nerves for the
production of the voice originate. The centre-theory
had found a supposed ** centre " for this reflex higher
up in the brain.

4. The instinct for food and self-preservation was,
like all the instincts, located in the cerebral hemi-
spheres. An analysis of this instinct shows that it is
composed of several reflexes, which are discharged
successively. The first is a visual reflex ; the frog
catches only objects (flies, for instance) that are in
motion. The opticus ends in the thalamus opticus,
hence it is to be expected that the loss of the cerebral
hemispheres would not prevent the frog from catch-
ing flies. Schrader found this to be the case. If
previous authors believed their experiments to prove
that the cerebral hemispheres are necessary for see-
ing, they were misled by the shock-effects of the
operation, and in this way made the localisation too

EXPERIMENTS ON VERTEBRATES 143

high. Goltz had already shown, moreover, that the
frog deprived of its cerebral hemispheres avoids ob-
stacles. The same holds good for the fish that has
lost the cerebral hemispheres. The first act in taking
food thus consists in an optical reflex. As soon as
the food comes in contact with the palate it arouses
swallowing reflexes. These reflexes are completed
by means of the vagus group. According to the
segmental theory these reflexes should still be possi-
ble even when all the parts of the brain lying in front
of the nuclei of the vagus have been removed. Such
is the case. As long as the medulla oblongata is
preserved the frog swallows the food that is put into
its mouth.

The respiration of frogs is chiefly mouth- and neck-
respiration. The corresponding nervous segments
for these parts of the body lie in the medulla oblon-
gata and in the beginning of the spinal cord. If the
latter be severed behind the nceud vital (calamus
scriptorius), as Schrader found, all the muscles whose
nerves originate behind the place of division continue
to participate in the respiration coordinately.

It was formerly assumed that the compensatory
movements of frogs were dependent on organs of the
mid-brain. Schrader found, however, that frogs
whose brain had been extirpated as far as the medulla
oblongata (the origin of the acusticus) still showed
compensatory movements. The earlier physiologists
were deceived by accidental effects of the operation.
For the sake of completeness it should be mentioned,

144 COMPARATIVE PHYSIOLOGY OF THE BRAIN

further, that the reflexes of wiping and warding off
objects from the body of the frog are of a purely
segmental character.

We have now given a review of the principal reac-
tions of the frog and have found that no localisation
of functions exists either in the brain or the spinal
cord, that these are only segmental reflexes, just as in
the Annelids and Arthropods. This conception was
natural after results obtained from experiments on
lower animals. That Schrader had foreseen it before
the experiments reported here were known is proved
by the closing sentence of his article on the frog's
brain : "" The series of experiments we have given
teaches us that the central nervous system of the frog
can be divided into a series of sections, each of which
is capable of performing an independent function. It
brings the central nervous system of the frog into
closer relation with the central nervous system of the
lower forms, which consists of a series of distinct gan-
glia that are connected by commissures. It speaks
against the absolute monarchy of a single central ap-
paratus and against the existence of different kinds of
centres, and invites us to seek for the centralisation in
a many-sided coupling of relatively independent sta-
tions." The question might be raised as to whether
the activities of the frog which we have considered
include all the reactions of this animal. The more
complicated instincts are for the most part nothing
more than series of segmental reflexes. I am inclined
to recommend using the word chain-reflexes, whereby

EXPERIMENTS ON VERTEBRATES 145

the performance of one reflex acts at the same time as
the stimulus for setting free a second reflex. The tak-
ing of food may serve as an illustration of such a chain-
reflex. The optic reflex of the moving fly produces
the snapping reflex ; the contact of the mouth-epi-
thelium with the fly produces the swallowing reflex.
Each of these reflexes is purely segmental. By tak-
ing into account the act of transmission, complicated
acts can thus be resolved into a few segmental
reflexes. A second fact must be taken into consider-
ation if we wish to trace back the reactions of a frog
to segmental reflexes, namely, that the irritability of
the organs of its body changes. In the chapter on
instincts we shall find how chemical conditions, espe-
cially, affect the form of the irritability of the animal,
and how all conditions which bring about chemical
changes in the body (temperature, food, sexual pro-
ducts) also modify its irritability. We shall then un-
derstand why the frog burrows at the beginning of the
cold weather in autumn and puts in an appearance
again with the awakening of spring, or, strictly speak-
ing, with the beginning of the warm weather. The
segmental reflex in the frog is, however, determined
also by the irritability of the peripheral organs and the
arrangement of the muscles. The segmental ganglion
acts, in the main, simply as the protoplasmic connec-
tion between the surface of the body and the muscles.
The experiments of Goltz and of Goltz together
with Ewald on the spinal cord of dogs prove that this
law of segmental reflexes is also correct for dogs.

146 COMPARATIVE PHYSIOLOGY OF THE BRAIN

However, In warm-blooded animals every operation in
the vicinity of the medulla oblongata is accompanied
by such severe shocks to the segmental respiratory
ganglia that the experimental proof is still wanting
for the ganglia of the medulla in higher Vertebrates.
It has been attempted with electrical stimuli, but
such experiments only show that some kind of proto-
plasmic connection exists between the stimulated spot
and the segmental ganglia of the active muscles. The
fact, for instance, that the respiratory movements are
affected by stimulation of the third ventricle only
proves that there are fibres at that place which go to
the segmental respiratory ganglia. The conclusion,
however, cannot legitimately be drawn from this that
respiratory ganglia or ** respiratory centres" are located
in the third ventricle. Two facts have combined to
hinder the development of the segmental theory.
First, comparative physiology of the brain and embry-
ology have never been duly considered. Because the
brain of Vertebrates only reveals its segmental char-
acter in the earliest embryological condition, only a
small number of physiologists have thus far seriously
believed that the segmental character of the central
nervous system would furnish the key for comprehend-
ing its functions. The second fact is disregard of the
shock-effects upon those parts of the central nervous
system situated behind the seat of the operation.
It Is possible that certain impulses flow constantly
from the cephalic to the lower parts of the central
nervous system. The stopping of these Influences

EXPERIMENTS ON VERTEBRATES 147

causes a change in the conditions of the segmental
ganglia behind the seat of the operation. This change
is the shock-effect.

Finally, the most important difference between the
segmental conception of the central nervous system
and the centre-theory may be pointed out. Accord-
ing to the latter theory, the central nervous system
consists of a series of centres for as many different
''functions." Each ** function" is determined by the
structure of its ** centre." According to the seg-
mental theory, there are only indifferent segmental
ganglia in the central nervous system, and the dif-
ferent reactions or reflexes are due to the different
peripheral organs and the arrangement of muscles.
The centre-theory must remain satisfied with the
mere problem of localising the apparent " seat " of a
" function " without being able to give the dynamics
of the reactions of an animal, as the latter depend
in reality upon the peripheral structures, and not on
the structures of the ganglia. For this reason the
segmental theory alone will be able to lead to a
dynamical conception of the functions of the central
nervous system.

This difference may be made more apparent by
comparison of these functions with those of the re-
tina. The optical perception of forms consists in
the power of single elements to determine, accord-
ing to their position on the retina, different space-
sensations. One retinal element may aid in bringing
about many different pictures. Viewed from the

148 COMPARATIVE PHYSIOLOGY OF THE BRAIN

segmental standpoint, we imagine the r6le of the
central nervous system to be similar to this : the
various elements or ganglia take the place of the re-
tinal elements in the perception of forms. The same
elements or ganglia participate in many " functions."
Every element shares in the result according to the
location of the segment, and other general or special
qualities. But if we attempt to make clear to our-
selves how the retina should act according to the
centre-theory, we find that every retinal element
would have to serve for the perception of one image
only, that we could see only as many different images
as we have retinal elements (for instance, rods). We
do the centre-theory no injustice in making this com-
parison : its consistent representatives really assume
that each image of memory is deposited in a special
cell, that the number of the cells of the brain de-
termines the number of the images of memory which
are possible.

I wish to call the attention of the reader to the fact
that Dr. A. Meyer has arrived, independently, at simi-
lar conclusions concerning the segmental character of
the central nervous system of Vertebrates as those set
forth in this chapter (9).

Bibliography.

1. GOLTZ. Ueber die Functionen des Lendenmarks des Hundes.
Pfliiger's Archiv, Bd. viii., 1874.

2. GoLTZ. Beitrdge zur Lehre von den Nervencentren des
Frosches. Berlin, 1868. Verlag von Hirschwald.

EXPERIMENTS ON VERTEBRATES 149

3. Flourens, p. Recherches expirimentales sur les Propriitis et
les Eonctions du Systeme Nerveux, etc. 2. edit. Paris, 1842.

4. SCHRADER, Max E. G. Zur Physiologie des Eroschgehirns.
Pflugers Archiv, Bd. xli., 1887.

5. Steiner, J. Die Eunctionen des Centralnervensy stems und
ihre Phylogenese. Erste Abtheilung : Untersuchungen iiber die
Physiologie des Eroschhirns. Braunschweig, 1885.

6. Steiner, J. Die Eunctionen des Centralnervensystems und
ihre Phylogenese. II. Abtheilung, Die Eische, Braunschweig,
1888.

7. GoLTZ, Fr., and Ewald, J. R. Der Hund mit verkiirztem
Rilckenmark. Pfluger's ArchiVy Bd. Ixiii. Bonn, 1896.

8. BiCKEL. Beitrdge zur Riickenmarksphysiologie des Aales.
Pfliiger's Archiv, Bd. Ixviii.

9. Meyer, Adolf. Critical Review of the Data and General
Methods and Deductions of Modern Neurology. Journal of Com-
parative Neurology y vol. viii., 1898.

CHAPTER X

SEMIDECUSSATION OF FIBRES AND FORCED
MOVEMENTS

It is apparent from the foregoing that in the central
nervous system of Vertebrates only segmental gan-
glia and only segmental reflexes appear. Superior
centres, a "centre of coordination" for instance, can-
not exist. Irritability and conductivity suffice to pro-
duce coordination in Medusae, in the heart, in the
respiration of Limulus, and in the movements of the
earthworm and the salamander. Schrader has rightly
expressed it : the nature of the nervous connections
alone determines the cooperation of different seg-
ments in a common activity. Some of these connec-
tions require special mention, for instance, those in
which decussation and semidecussation of fibres ap-
pear. Possibly the most familiar example of semi-
decussation is found in the optic nerves. Here the
fibres cross, so that while each eye has its own special
nerve each tract contains fibres from both eyes. The
fibres of the temporal half of the retinae pass through
the chiasma uncrossed (that is, remain on the same
side of the head and brain), the fibres which come

150

FORCED MOVEMENTS 151

from the nasal side of the retinae undergo a decussa-
tion, i. e., they cross to the other side of the head and
brain. The left optic tract contains fibres from the
temporal side of the left eye, and from the nasal,
or internal, side of the right eye. If the left tract be
cut, the left sides of both retinae become blind, and
the patient recognises nothing more in the right half
of the field of vision. This is a case of hemianopia.

A similar semidecussation also occurs in the motor
nerves of the eye. For the time being we may con-
sider the various muscles of each eye as a unit. In
the lateral movements of our eyes the rectus externus
of one eye and the rectus internus of the other co-
operate. If we assume an inherited connection be-
tween the retinal elements and the movements of the
eyes, the right externus and the left internus must be
innervated by the left half of the brain. The nerve-
fibres of the externi must thus be crossed, those of
the interni not crossed. The semidecussation in this
case naturally occurs in the brain, and not peripherally.
The pathological expression of this motor semidecus-
sation is the deviation conjugde^ which is a motor
affection, corresponding to the sensory affection,
hemianopia. We can only expect to find these
semidecussations where symmetrical organs always
receive equal innervations, as in the case of our eyes.
Our arms and legs can move independently of each
other, but in lower Vertebrates the case is different.
The symmetrical fins of the fish receive equal innerv-
ations. I have shown that associated changes of

152 COMPARATIVE PHYSIOLOGY OF THE BRAIN

position of the eyes and fins can be produced by de-
struction of an ear or the acoustic nerve (i). If we
destroy in a shark the left auditory nerve or the left
side of the medulla, where the auditory nerve enters,
the left eye of the animal looks down, the right up.
This change of position of both eyes suggests that
the relative tension between the muscles that raise
the eyes has changed. In the left eye the tension of
the lowering muscles predominates over that of their
antagonists ; in the right eye the reverse is the case.
The fins, likewise, show associated changes of posi-
tion. The left fin is raised dorsally, the right is bent
ventrally. While it can be said that both eyes are
rolled about the longitudinal axis of the animal toward
the left, the fins are rolled about the same axis to the
right. Although the pectoral fins show the associated
changes of position most clearly, these changes also
exist in all the remaining fins, only with the difference
that the amount of the change of position decreases
the farther the segment is removed from the point of
operation. The influence of the operation must de-
crease as the distance of a ganglion increases. The
resistance to the transmission of the change increases
with the distance.

These observations enable us to draw a conclusion
concerning the connection of the muscles with the
right and left halves of the corresponding ganglia.
We may assume that a permanent decrease, but not a
permanent increase in the tension of the muscles can
result from the destruction of one part of the brain.

FORCED MOVEMENTS 153

It is thus the muscles directly or indirectly connected
with the left side of the medulla oblongata (in the
acoustic segment) which show a decrease in tension
after the destruction of the left ear. Accordingly, the
left side of the medulla is connected with the raising
muscles of the left eye and the lowering muscles of
the right eye, as well as with the lowering muscles of
the left pectoral fin and the raising muscles of the
right pectoral fin. If we start with the idea that all
the muscles of an eye or a fin form a common whole,
a kind of semidecussation is present. It is, however,
not only the muscles of the fins that undergo such
changes of tension, but probably also the muscles of
the spinal column.

If the symmetrical muscles of the organs of loco-
motion possess different tension, the usual stimuli for
locomotion must naturally lead to unsymmetrical in-
stead of symmetrical movements. When the lower-
ing muscles predominate in the right pectoral fin and
the raising muscles in the left, the animal, when these
fins are used, will come under the influence of a couple
of forces which must produce a rolling movement
around the longitudinal axis of its body toward the
left. As long as the animal swims slowly, rolling mo-
tions do not occur, for they are compensated for.
The friction of the fish in the water will suffice to
destroy a slight rolling motion. But if the animal
attempts to swim rapidly, e, g., if it be excited, it be-
gins to roll. These rolling motions are called forced
movements, a poorly selected term. The same move-

154 COMPARATIVE PHYSIOLOGY OF THE BRAIN

ments have also been noticed in dogs and rabbits
after an operation on one side of the medulla.

If a fish whose progressive movements are deter-
mined by the sculling motions of the tail turns to the
right, the tail moves with greater force toward the
right than toward the left. This condition might be
made permanent if it were possible to weaken the
muscles on the left side of the spinal column. This
occurs when the right side of the acoustic segment of
the medulla is destroyed. The fish moves in a circle
toward the right. We also obtain circus-motions to-
ward the right if we destroy the ventral portion of the
left optic lobe. Hence, fibres must pass from the
ventral portion of the left optic lobe to the right
acoustic segment of the medulla. After such an
operation, an increase in the tension of the skeletal
muscles occasionally shows itself, for the fish may lie
permanently bent into a circle without being able to
straighten itself out again.^ Such a fish can no longer
swim straight ahead. The difference in the tension
of the muscles on the two sides of the animal is, how-
ever, usually not so great, in which case the circus-
motions will appear only spasmodically, for ex;ample,
when the animal is excited.

One-sided division of the spinal cord and of the
medulla behind the acoustic segment produces no
forced movements (2 and 3). On the other hand, roll-

* If such a fish be decapitated the curvature of the body remains. It may
even remain after death. We have to deal with an organic change in the
muscles, caused by the operation.

FORCED MOVEMENTS 155

ing and circus-motions may occur after injury to the
brain in front of this segment, wherever places are
met with which are directly or indirectly connected
with the acoustic segment. This is the case, for in-
stance, after a one-sided lesion of the pons or the
cerebral hemispheres in rabbits and dogs. In both
animals circus-motions occur after destruction of the
cerebral hemispheres, in rabbits toward the intact side,
in dogs toward the injured side. All the facts prove
that the semidecussations take place in the vicinity of
the acoustic segment and not farther down. In man,
so far as I know, circus-motions have never been ob-
served ; this is probably due to his upright walk. It
would be interesting to make the experiment of having
patients afflicted with certain diseases (for instance,
diseases of the inner ear) walk on all fours (with
closed eyes) and to observe whether circus-motions
occur.

As we have already mentioned, it is a well-known
fact that in Arthropods after destruction of one half
of the supracesophageal ganglion circus-motions can
occur. That they need not occur has been shown by
Miss. Hyde and also by Bethe (4 and 5). According
to the investigations of the latter author these circus-
motions in Invertebrates are called forth by very dif-
ferent disturbances in the muscle-tension. It is often
due simply to a disturbance in the muscle-tension of
the extremities of one side, the other side being ap-
parently normal. In Crustaceans, associated changes
of position of the extremities can also occasionally

156 COMPARATIVE PHYSIOLOGY OF THE BRAIN

occur after destruction of one half of the supraoesoph-
ageal ganglion. In Cephalopods von Uexkiill has

observed forced
movements after
lesion of an ear.
Fig. 36 shows the
predominance of
the flexors of the
legs over the ex-
tensors on the left
side of the body of
a Limulus. In this
animal the right
half of the brain
had been de -
stroyed and it
showed circus-mo-
tions toward the
left.

Steiner has made
a peculiar applica-
tion of the facts of
forced movements.
He imagines that
the ability to move
forward is a spe-
cific "function" of
the brain, and has believed that it would be possible by
means of this criterion to decide whether or not a gan-
glion of a lower animal should be called a brain. The

Fig. 36.

FORCED MOVEMENTS 157

facts of comparative physiology do not favour this con-
ception. Spontaneous progressive movements exist
in Infusoria which possess no nervous system, and even
in plant organisms, for example, in the swarmspores
of algae. It is an important principle of physiological
epistemology that a phenomenon which occurs gener-
ally, cannot possibly be the specific function of an
organ which is peculiar to a few forms only. Steiner
soon found a fact that showed the erroneousness of
his theory, the fact that the decapitated shark con-
tinues to swim about in the tank. Schrader had like-
wise found that the frog without a brain is still able to
perform spontaneous progressive movements. Steiner
maintains further that *' the brain is defined by the
general centre of movement in connection with the
action of at least one of the higher sensory nerves."
'' In addition to its great simplicity this definition has
still another advantage, namely, that it is satisfied by a
single experiment ; because of the two elements of
which the definition is made up, one element is always
given anatomically. This is the higher sensory nerve,
whose presence also vouches for its function. The
one experiment that it is necessary to make has to
prove that in addition to the sensory apparatus the gen-
eral centre of movement also exists. The proof is then
complete if the one-sided removal of the central nerv-
ous part so changes the direction of the movements
of the animals that a circus-motion, which is generally
known by the name forced movement, takes the place
of a forward movement" (Steiner). This idea is

158 COMPARATIVE PHYSIOLOGY OF THE BRAIN

likewise erroneous and easily leads to absurdity. One-
sided destruction of the cerebral hemispheres in man
produces no forced movements. Thus, according to
Steiner, the cerebral hemispheres should not be-
long to the brain. Second, according to Steiner,
the ear must be a brain. One-sided lesion of the
ear is sure to produce forced movements in a series
of animals, and, moreover, the auditory nerve is a
higher sensory nerve. I have mentioned this sub-
ject at this place because it is a typical illustration of
what plays on words in physiology lead to. It is
not our task to find a definition for the word brain,
but to gain an insight into the functions of the central
nervous system. It is of minor importance what
name we give to the different parts of the central
nervous system.

In connection with this chapter we wish to call at-
tention to the more recent experiments of Sherring-
ton and H. E. Hering, from which it seems to follow
that with the innervation of a muscle the relaxation of
its antagonist results simultaneously.

Bibliography.

1. LoEB, J. Ueber Geotropismus bei Thieren. Pfluger's Archiv^
Bd. xlix., 189 1.

2. LoEB, J. Ueber den Antheil des Hornerven an den nach
Gehirnverletzung auftretenden Zwangsbewegungen, Zwangslagen
und associirten Stellungsdnderungen der Bulbi und Extremitdten.
Pfluger's ArchiVy Bd. 1., 189 1.

3. Steiner. Die Functionen des Centralnervensystems und ihre
Phylogenese. II. Die Fische. Braunschweig, 1888.

FORCED MOVEMENTS 159

4. Bet HE, A. Vergleichende Untersuchungen uber die Func-
tionen des Centralnervensystems der Arthropoden. P Auger's Ar-
chiVy Bd. Ixviii., 1897.

5. Bethe, a. Das Centralnervensystetn von Carcinus mcenas.
II. Mittheil. Arch. f. mikroskop. Anatomie^ Bd. 1., 1897.

6. Steiner, J. Die Functionen des Centralnervensystems wir-
belloser Thiere. Sitzungsberichte der Berliner Akademie der Wis-
senschaften. 1890, I. S., 39.

CHAPTER XI

RELATIONS BETWEEN THE ORIENTATION AND
E UNCTION OF CERTAIN ELEMENTS OF THE
SEGMENTAL GANGLIA

The results of some investigations carried on by
Garrey and myself showed that if a constant current
be sent through a trough in which are larvae of
Amblystoma, peculiar changes may be observed in
the postures of the animals (i). If the current passes
through them longitudinally from head to tail (Fig.
37) the back becomes convex and the ventral side

Fig. 37. Attitude of an Amblystoma under the Influence of a
Galvanic Current Passing from Head to Tail.

Fig. 38. Attitude of an Amblystoma when the Galvanic Cur-
rent Passes from Tail to Head.

concave. This change of position is occasioned by
the muscles of the ventral side (the flexors of the
spinal column) becoming more tense than the dorsal

160

ORIENTATION AND FUNCTION i6i

muscles (the extensors of the spinal column) from the
passage of the current. On the other hand, if the
current goes through the animal in the direction from
tail to head both head (Fig. 38), and tail are raised.
The body becomes more concave on the dorsal side
and convex on the ventral side. The extensors of the
spinal column become more tense than the ventral
muscles. A pronounced opisthotonus exists. In
order to show the phenomenon clearly the animal
must be brought into the current gradually. If
we continue to raise the intensity of the current,
changes of position also take place in the legs. The
changes in the hind-legs are more easily described
than those in the fore-legs. If the current passes from
head to tail the hind-legs are braced backward (Fig.
-^f), making the forward movement (to the anode)
easier. If the current passes from tail to head the
hind-legs are braced forward (Fig. 38), making the
backward movement (to the anode) easier. How can
these phenomena be explained ? The current has two
kinds of effects. A conduction of the current takes
place through ions. Wherever the progress of ions is
blocked in the central nervous system, an increase in
their concentration will occur and this must be followed
by physical or chemical alterations of the colloids. The
progress of ions may be blocked by semipermeable
membranes at the external limit of neurons or some-
where inside the neurons. Wherever anions are
blocked different effects (anelectrotonus) will be pro-
duced than at places where the progress of kations is

i62 COMPARATIVE PHYSIOLOGY OF THE BRAIN

blocked (katelectrotonus). The action of the various
ions on nerve-elements is as yet unknown. The other
effect of the current may consist in the migration of
certain colloids in one direction and of water in the
other direction.

If in the larvse of Amblystoma the tension of the
flexors of the spinal column predominates in a de-
scending current (from head to tail) and the tension
of the extensors of the spinal column predominates in
an ascending current, this proves that the nervous ele-
ments of the flexors and extensors in the central 7ier-
vous system, which are afl^ected by the ions, possess an
opposite orie7itation. Maxwell and I have developed
more definite ideas concerning the orientation of these
elements, but such details are for the time being of
minor importance. I only wish to state that the rela-
tive orientation of these elements must be the same
in every segment of the spinal cord ; for when the
spinal cord in the larvse of Amblystoma is severed or
the whole animal cut into several pieces the effects
of the current remain the same. Since the article
mentioned above was published I have found that
crayfish (young and small specimens were used for
these experiments) behave toward the current like
Amblystoma larvae. If the median-plane of the
crayfish is in the direction of the lines of the current
(which are all straight and parallel in these experi-
ments) and the head is turned toward the anode, the
flexors of the body contract and the crayfish rolls it-
self into a complete ring, provided that the density of

ORIENT A TION AND F UNCTION 163

the current is exactly right. The back is convex, the
ventral side concave. But if the current passes from
tail to head, the back becomes entirely straight, the
extensors being contracted to the utmost limit.^ The
dorsal side cannot become concave because the exo-
skeleton of the crayfish does not allow it. Thus in
the body of the crayfish the motor elements of the
extensors and flexors which are affected by the cur-
rent must have the same orientation as in the body of
Vertebrates. This holds good not only for the flexors
and extensors of the body but generally, as we shall
at once see.

We have already mentioned the fact that a con-
stant current passing through Amblystoma larvae in
the longitudinal direction affects not only the tension
of the flexors and extensors of the body, but also
the muscles of the extremities. The tension too is
changed in such a way, as has been already intimated,
that it renders movement toward the anode easy,
movement toward the kathode difficult. If, for in-
stance, the current passes through the animal from
head to tail the hind-legs are braced backward and
the position of the fore-legs is changed correspond-
ingly, so that the progressive movement of the ani-
mal is made easy, the backward movement dif-
ficult. If, however, the current passes from tail to
head the hind-legs are braced forward and the posi-
tion of the fore-legs is changed correspondingly ; the

' It is necessary that in this experiment the intensity of the current be
increased very slowly.

i64 COMPARATIVE PHYSIOLOGY OF THE BRAIN

animal can go backwards easily, while forward move-
ment is difficult. In fact it is apparent that if the
animals attempt to move in any direction while
the current is passing through them they go toward
the anode. ^

From this fact it follows that, for the nervous appa-
ratus of the progressive movements of Amblystoma, a
close relation must exist between the orientation of
the determinative elements of the motor nerves and
their function : in fact the nervous elements which
cause the progressive movements must have with re-
gard to the longitudinal direction of the animal an
orientation which is opposite to the orientation of
those elements which cause the backward movement.
Garrey and I had called attention to the fact that the
observations of Blasius and Schweitzer show that
other Vertebrates, for instance, young eels, behave
like Amblystoma. The same also holds good for the
shrimp. This last assertion is based upon a series of
experiments made by Maxwell and myself (2). These
experiments were made chiefly on Palaemonetes. This
Crustacean uses the third, fourth, and fifth pairs of legs
for its locomotion. The third pair pulls in the for-
ward movement and the fifth pair pushes. The fourth
pair generally acts like the fifth and requires no further
attention. If a current be sent through the animal
longitudinally, from head to tail, and the strength in-
creased gradually, a change soon takes place in the

^ This explains the galvanotropic gatherings observed by Hermann, Blasius
and Schweitzer, and others.

ORIENTATION AND FUNCTION 165

position of the legs. In the third pair the tension
of the flexors predominates, in the fifth the tension of
the extensors. The animal can thus move off easily
with the pulling of the third and the pushing of the
fifth pairs of legs, that is to say, the current changes
the tension of the muscles in such a way that the
forward movement is rendered easy, the backward
difficult. Hence it can easily go tow^ard the anode,
but only with difficulty toward the kathode. If a
current be sent through the animal in the opposite di-
rection, namely, from tail to head, the third pair of legs
is extended, the fifth pair bent ; that is, the third pair
can push and the fifth pair pull. The animal will
thus go backward easily and forward with difficulty.
We see here again that the nervous elements of the
central nervous system which bring about the for-
ward movements have the opposite orientation as re-
gards the longitudinal axis of the animal from the
nervous elements which bring about the backward
movement. But we can go still further in the devel-
opment of this law. Palaemonetes can not only walk,
but is also a good swimmer, and it can swim back-
wards as well as forwards. In swimming forwards
the swimming appendages, among which the tail fin
must be counted, push backwards forcibly and for-
wards gently ; in swimming backwards the opposite
occurs. If the current be sent through Palaemonetes
in the direction from head to tail ^ the swimming

' The tail fin behaves toward the current like the abdominal swimming ap-
pendages and not like the body. This must be taken into consideration in
galvanotropic experiments.

i66 COMPARATIVE PHYSIOLOGY OF THE BRAIN

appendages, and the tail also, are stretched backward
or dorsad to their fullest extent. This proves that the
tension of the muscles that move those organs back-
wards is greater than that of their antagonists. The
shrimp can thus swim forwards toward the anode
easily under the influence of such a current, but back-
wards only with difficulty. If the current passes
through in the opposite direction (from tail to head),
however, the tail and the ventral appendages are
turned forward. The tension and the development
of energy now predominate in those muscles which
move the swimming appendages backwards. In this
way the animal can swim backwards easily, while it is
difficult or impossible for it to swim forwards. Hence
the nervous elements, which determine the forward-
swimming, must also have the same orientation in re-
gard to the longitudinal axis of the animal as those
elements which determine the walking forwards, while
the nervous elements for swimming backwards have
the opposite orientation.

The fact yet remains to be considered that Palae-
monetes, like many other Crustacea, can also move
sideways. This movement is.produced by the pulling
of the legs on the side toward which the animal is
moving (contraction of the flexors), while the legs of
the other side push (contraction of the extensors).
If a current be sent transversely, say from right to
left, through the animal, the legs of the right side
assume the flexor-position, those of the left side the
extensor-position. In the legs of the right side the

ORIENTATION AND FUNCTION 167

flexors produce more energy, in the legs of the left
side the extensors. The transverse current assists the
animal in moving to the right toward the anode and
prevents it from moving to the left toward the kath-
ode. Hence the nervous elements which produce
the sidewise movement of the crayfish toward the
right must have the opposite orientation in regard to
the longitudinal axis from the nervous elements which
produce the sidewise movement to the left.

Maxwell and I had attempted to give a picture of
the arrangement of those elements on the assumption
that they are the motor neurons. No reason exists,
however, for regarding the ganglion-cell in toto as the
point of action for the chemical effect of the ions set
free. It may be any element inside of the ganglion-
cells, or even of the fibre itself ; it is not even neces-
sary that this element should be especially noticeable
histologically. Life-phenomena are determined by
physical and chemical conditions which are outside the
realm of histology. But whatever it is, it is certain
that those determinative elements in the central ner-
vous system whose activity produces movements of the
body^ have a fixed orientation in the body, which evi-
dently stands in some simple relation to the direction of
the movem^ent which is produced by it.

The idea of such a simple relation between the
orientation of nervous elements and the direction of
motion produced by them is no more strange than
the facts observed in the stimulation of the horizon-
tal canal of the labyrinth. If this canal be slightly

i68 COMPARATIVE PHYSIOLOGY OF THE BRAIN

touched motions of the eyes or head occur in the plane
of this canal. In this case we have to deal with a sim-
ple relation between the orientation of the canal and
the plane in which the organs or the whole body of
the animal move. This fact is just as mysterious
as the more general facts mentioned in this chapter. I
am inclined to assume that the peculiar relation be-
tween the semicircular horizontal canal and the motions
produced by the stimulation of this canal finds its
explanation through the facts mentioned in this chap-
ter. It is possible that the central endings of the
nerve of the horizontal canal are connected with the
motor elements in the medulla whose activity pro-
duces motions in the plane of the horizontal canal.
When Flourens made his experiments on the semi-
circular canals he found that there was a striking
resemblance between the effects of a destruction of
the canals and the sectioning of the crura cerebelli.
He came to the conclusion that there must be a sim-
ple relation between the direction of the fibres of the
crura cerebelli and the motion produced by them (3).
His observations are not in all points correct ; yet
with some modification his fundamental idea remains
true. The next chapter, on the cerebellum, will give
us some more data about his observations.

It is possible for us to conceive from this how it
happens that the same optic stimulus or the same
space-perception is able to direct our eyes toward a
certain point, to turn our head in that direction, to
guide our finger thither, or to bring our legs into such

ORIENT A TION AND FUNCTION 169

activity that our body arrives at that place. It is
possible that the elements of the central nervous system
which become active in this way all have the same ori-
entation in each segment^ and what we call an innerva-
tion may be a process in which the orientation of the
elements plays a role. The effect of the electric cur-
rent might be an example of such a process. This
problem of the physiology of coordinated movement
which we touch upon here has always seemed to me
the most mysterious in the whole physiology of the
central nervous system, and the way offered here of
reaching a simple solution seems to me worthy of
mention. The whole conception can easily be classified
under the segmental conception of the central nervous
system. Movements of the eyes, head, arms, and legs
depend upon as many different segmental ganglia.
Each of these ganglia has some features in common
with every other ganglion, for instance the orientation
and arrangement of the elements (neurons?). If a
process of such a nature that it can only stimulate
elements oriented in a certain way in each ganglion
spreads through the segmental ganglia, it must pro-
duce a movement in exactly the same direction in the
appendages of each segment. This does away with
the necessity of imagining artificial connections of the
neurons which would be able to produce such a series of
coordinated motions in different limbs and segments.

If the question be raised, however, as to how it
happens that a simple relation exists between the
orientation of the motor nerve-elements and the

170 COMPARATIVE PHYSIOLOGY OF THE BRAIN

movement or progressive movement produced by
them, we must again refer to the simple segmental
relations of the first embryonic formation that re-
mains better preserved in the central nervous system
than in the muscles. The problem with which we have
to deal here is ultimately a problem of embryology.

Bibliography.

1. LoEB, J., and Walter E. Garrey. Zur Theorie des Gal-
vanotropismus. II. Mittheilung. Versuche an Wirbelthieren. Pflil-
ger's Archiv^ Bd. Ixv., 1896.

2. LoEB, J., and S. S. Maxwell. Zur Theorie des Galvanotro-
pismus. Pfluger's Archiv, Bd. Ixiii., 1896.

3. Flourens, p. Fonctions du Syst^me nerveux. Paris, 1842.

[

CHAPTER XII

EXPERIMENTS ON THE CEREBELLUM

The experiments on the cerebellum support to a
certain extent the observations mentioned in the
preceding chapter.

The cerebellum, like the cerebral hemispheres, is
a structure which clearly expresses inequality of
growth. Both may be considered as evaginations
and appendages of the segmental nervous system.
The cerebellum is connected with the central nervous
system by three crura, the crura cerebelli ad medullam
oblongatam, the crura cerebelli ad pontem, and the
crura cerebelli ad corpora quadrigemina. The latter
extend forward in a pretty straight line, the first ex-
tend backward, and the peduncles to the pons at right
angles to both. Magendie discovered and Flourens
confirmed the fact that lesion of these tracts possess-
ing so characteristic an orientation to the chief axes
of the body produces ** forced " movements whose di-
rection bears a simple relation to the orientation of
the severed peduncle. If a peduncle of the pons
be severed on one side the animal rolls about its
longitudinal axis. If the crura cerebelli that extend
forward be severed the ^nimal rushes forward with

171

172 COMPARATIVE PHYSIOLOGY OF THE BRAIN

great force ; if the crura cerebelli ad medullam ob-
longatam be severed the animal goes backwards or
shows a tendency to turn somersaults backwards.
'* La direction des mouvements produits par la section
des fibres de I'encephale est done toujours determinee
par la direction de ces fibres" (Flourens).

Flourens called attention to the analogy of these
phenomena with those he observed after lesion of the
semicircular canals. This analogy, however, does not
exist just as he states it. He compares the effect of
the one-sided division of the pons with the division
of a horizontal canal. This is not correct. So far
as I know, such a lesion does not produce rolling mo-
tions about the longitudinal axis in any animal. On
the other hand, destruction of a whole ear, probably
in most cases, causes rolling motions. Flourens
states further that after destruction of the anterior
canals an animal turns somersaults forwards, after de-
struction of the posterior canals backwards. Flourens
assumes that the nerves of the three canals continue
into the corresponding peduncles of the cerebellum,
and that this origin of the nerves is the cause of the
phenomena that we observe after lesion of the single
semicircular canals (3). But this is probably not cor-
rect, since the auditory nerve ends in the medulla. It
is possible, however, that the cerebellum is connected
with the same motor elements in the medulla with
which the acoustic nerve is connected. The cerebel-
lum might thus appear as an appendage of the acous-
tic segments.

EXPERIMENTS ON THE CEREBELLUM 173

This harmonises with the results Ferrier obtained
from his stimulation experiments (i). He found that
stimulation of the different parts of the cerebellum
causes associated movements of the eyes, and that
the direction of the movement changes with the posi-
tion of the electrodes. The head also moves in the
same direction as the eyes. Movements of the limbs
were also observed, but it could not be determined
whether or not they were associated with the move-
ments of the head. From this we may conclude that
possibly or probably the movements which are pro-
duced by stimulation of the cerebellum are somewhat
related to those movements which are produced by
stimulation or injury of the semicircular canals, only
that the stimulation experiments on the cerebellum,
according to Ferrier, often yield no results.

Extirpation of the cerebellum leaves the sensory
and psychic functions of the animal undisturbed. It
is only in the movements that peculiar disturbances
appear, which are described differently by the differ-
ent authors. The motions of the animals resemble
somewhat those of a patient suffering from St. Vitus's
dance, inasmuch as they do not reach the intended
aim and are often excessive in character. It is neces-
sary that the head of a dog whose cerebellum has been
injured be held in the dish when it eats, for if not held
every effort sends the head so much too far that the
animal is not able to get its food. This disturbance
is most pronounced immediately after the operation
and may disappear more or less after a certain time.

174 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Such disturbances also show themselves in the
limbs. The animal often staggers about like an in-
toxicated person and finds difficulty in keeping itself
on its legs. All these peculiarities perhaps point ul-
timately to a decrease in the tension of the skeletal
muscles. The measured movements of the normal
animal are only possible if the tension of the antag-
onistic muscles is so great that excessive movements
cannot take place. But if the muscles of the spinal
column are relaxed in the dog without the cerebellum,
as has been maintained (and apparently with good
reason), every intended movement can go wide of its
aim.

According to the results of Luciani's numerous
experiments, weakness or relaxation of the muscles
seems to form the most constant factor in the effects of
operations on the cerebellum. The affected groups
of muscles, however, seem to vary with the position
of the part of the cerebellum which is destroyed.
Flourens, who attributed special functions to each sec-
tion of the brain, maintained that the cerebellum was
the general centre of coordination, because lesions of
the cerebellum bring about the above-mentioned dis-
turbances. Luciani (2) has shown, however, that
some of the dogs that had lost the cerebellum were
still able to perform coordinated swimming motions
in the water and even coordinated walking motions.
The weakness of all or of certain groups of muscles
may lead to ataxic disturbances, but in some cases
these may be very slight. Thus we see that Flourens's

EXPERIMENTS ON THE CEREBELLUM 175

theory that the cerebellum is an organ of coordination
is not correct. It may be too that a part of the dis-
turbances observed after lesion of the cerebellum are
due to secondary effects of the operation on the med-
ulla or the corpora quadrigemina.

This latter conception is supported by comparative
physiology. In fishes and frogs, in which the shock-
effects are slight, the cerebellum can be removed
without producing any disturbance in the behaviour
of the animals (Vulpian, Steiner). In sharks, whose
cerebellum is strongly developed, I myself have made
numerous division-experiments and numerous experi-
ments on partial or total extirpation of the cerebel-
lum, and no change whatever took place in the
behaviour of the anirnals. It is impossible and un-
justifiable in this case to talk of a definite " function "
of the cerebellum.

It may be well, in consideration of what has been
said in the preceding chapter and of observations
which will be discussed later on concerning the results
of lesions of the cerebral hemispheres, to remind the
reader of a hypothesis made by Magendie. He saw
animals walk or fly backwards permanently after les-
ion of a certain part of the medulla oblongata. He
saw further that a lesion of the corpora striata pro-
duces an impulse to run forwards. Finally he ob-
served the rolling motions of the animals about their
longitudinal axis after one-sided lesion of the pons.
He makes the following remark in this connection :
*' Comme notre esprit a besoin de s'arr^ter ^ certaines

176 COMPARATIVE PHYSIOLOGY OF THE BRAIN

images je dirai qu'il existe dans le cerveau quatre im-
pulsions spontanees ou quatre forces qui seraient pla-
cees aux extremites de deux lignes, qui se couperaient
a angle droit, Tune pousserait en avant, la deuxieme en
arriere, la troisieme de droit a gauche en faisant rouler
le corps, la quatrieme de gauche a droite en faisant
executer un mouvement semblable de rotation. Dans
les diverses experiences d'ou je tire ces consequences
les animaux deviennent des especes d'automates mon-
tes pour executer tels ou tels mouvements et incap-
ables d'en produire aucun autre." The last statement
goes too far, but Magendie's main thought deserves
more consideration than it has heretofore received
from physiologists. The galvanotropic facts men-
tioned in Chapter XI. show most conclusively that in
Crustaceans and Vertebrates there exists a relation
between the orientation and function of certain motor
elements, and a similar relation also finds expression in
the experiments on the horizontal semicircular canal.^

' Dr. Lyon has shown that only the stimulation of the horizontal canal gives
rise to motions in the plane of this canal, while the same result cannot be ob-
tained with any degree of certainty through a stimulation of the two other
canals (4).

Bibliography.

1. Ferrier. The Functions of the Brain. New York, 1886.

2. LuciANi, LuiGi. Das Kleinhirn. Leipzig, 1893.

3. Flourens, p. Fonctions du Systlme nerveux. Paris, 1842.

4. Lyon, E. P. A Contribution to the Comparative Physiology
of Compensatory Motions. The American Journal of Physiology^
vol. iii., 1900.

CHAPTER XIII

ON THE THEORY OE ANIMAL INSTINCTS

I. The discrimination between reflex and instinctive
actions is chiefly conventional. In both cases we have
to deal with reactions to external stimuli or conditions.
But while we speak of reflex actions when only a
single organ or a group of organs react to an external
stimulus, we generally speak of instincts when the
animal as a whole reacts. In such cases the reactions
of the animal, although unconscious, seem often to be
directed towards a certain end. A fly acts instinct-
ively when it lays its ^gg on objects which serve
the hatching larvae as food. We call the periodical
migrations of animals instinctive. We call it instinct-
ive when certain animals conceal themselves in cracks
and crevices where they are safe from persecution.
But the purposeful character of instincts cannot be
used to distinguish them from reflexes, as a great
many of the reflexes are also purposeful, for instance,
the closing of the eyelid if the conjunctiva be touched
or the wiping off of acetic acid that is put on the
skin of a decapitated frog. On the other hand, it
cannot be said that every instinctive action is purpose-

178 COMPARATIVE PHYSIOLOGY OF THE BRAIN

ful, for instance, the flying of the moth into the
flame.

In many cases the greater complication of instinctive
actions compared with simple reflex actions is due to the
fact that in the instinctive actions we have to deal
with a chain of reflexes in which the first reflex be-
comes at the same time the cause which calls forth
the second reflex. The taking up of food by the
frog is a good illustration of this. The motion of the
fly causes an optical reflex which results in the snap-
ping motion. The contact of the fly with the mucous
membrane of the pharynx sets free a second reflex, the
swallowing reflex, which brings the fly into the oeso-
phagus. If it be true that the instincts belong to the
same class of processes as the reflexes, their relation to
the central nervous system should be the same. We
have seen that as far as reflexes are concerned, the
nervous system only acts as a protoplasmic conductor
between the periphery (sense organs) and the muscles.
I think it is possible to show that this is also true for
instincts. In order to prove this, we shall have to go
into an analysis of the instincts. We shall select for
our analysis such simple cases of instincts as depend
upon tropisms.

2. We have seen that when certain Crustaceans, for
instance Palaemonetes, are subjected to the effect of a
galvanic current such changes of tension take place in
the muscles of the appendages that movement toward
the anode becomes easier, and toward the kathode
more difficult. The result is that if the current is

THEORY OF INSTINCTS 179

continued long enough, all the animals collect at the
positive pole. When this process is observed with-
out a careful analysis, it seems as though these Crus-
taceans possessed the instinct to move toward the
anode, just as the moths possess the instinct to move
into the flame. The flight of the moth into the flame
is in reality only the result of a tropism, — heliotropism,
which differs from galvanotropism chiefly in that the
rays of light take the place of the curves of the current.
The reader knows that certain plants when exposed
to the light on one side, for instance, when cultivated
at a window, bend their tip toward the window until
the tip of the stem is in the direction of the rays
of light. The tip then continues to grow in the
direction of the rays. We call this dependence of
orientation on light heliotropism. We speak of
positive heliotropism when the organ bends towards
the source of light, of negative heliotropism when
the organ bends away from it. It is generally
assumed that the light has a chemical effect in these
cases.

The relations of symmetry in plants and animals
play an important part in these phenomena. We
will take, by way of illustration, the stem of a
hydroid, Eudendrium that is being raised near a win-
dow. I have found that it bends toward the window
like a positively heliotropic plant under the same'
conditions. The process may be described as follows :
The light strikes the Eudendrium-stem from the
side. A contraction of the protoplasm on that side

i8o COMPARATIVE PHYSIOLOGY OF THE BRAIN

ensues and a greater resistance is thus offered to the
increase in length on this side than on the opposite
side. The result is that the stem bends and becomes
concave on the side toward the light. As soon, how-
ever, as the bending has progressed so far that the
stem comes into the direction of the rays of light, all
the symmetrical elements are struck by the light at
the same angle. The intensity of light is thus equal
at symmetrical points, and there is no longer occasion
for the stem to leave this direction. It thus con-
tinues to grow in the direction of the rays of light.
Negatively heliotropic elements, roots, for instance,
differ from positively heliotropic elements in that the
light produces a relaxation of the protoplasm. Hence
when the light comes from one side, the resistance to
the growth on that side will be less than on the op-
posite side, and the tip will bend away from the
source of light. As soon as the tip comes into the
direction of the rays of light and the symmetrical
points are all struck by them at the same angle, the
intensity of the light on both sides is the same, and
every cause for leaving this direction is removed. It
has been known for a long time that many animals
are " attracted " by the light and fly into the flame.
This was considered a special instinct. It was said
that these animals loved the light, that curiosity
drove them into it. I have shown in a series of
articles, the first of which appeared in January, 1888,
that all these actions are only instances of those phe-
nomena which were known in plants as heliotropism.

THEORY OF INSTINCTS i8i

It was possible to show that the heliotropism of
animals agreed in every point with that of plants.
If a moth be struck by the light on one side, those
muscles which turn the head toward the light become
more active than those of the opposite side, and
correspondingly the head of the animal is turned
toward the source of light. As soon as the head of
the animal has this orientation and the median-plane
(or plane of symmetry) comes into the direction of
the rays of light, the symmetrical points of the sur-
face of the body are struck by the rays of light at the
same angle. The intensity of light is the same on
both sides, and there is no more reason why the
animal should turn to the right or left, away from the
direction of the rays of light. Thus it is led to
the source of the light. Animals that move rapidly
(like the moth) get into the flame before the heat of
the flame has time to check them in their flight.
Animals that move slowly are affected by the increas-
ing heat as they approach the flame ; the high tem-
perature checks their progressive movement and they
walk or fly slowly about the flame. The more re-
fractive rays are the most effective in animals just as
in plants (i).

Hence the " instinct" that drives animals into the
light is nothing more than the chemical — and indirect-
ly the mechanical — effect of light, an effect similar to
that which forces the stem of the plant at the window
to bend toward the source of light, or which forces
Palaemonetes to collect at the anode. The moth

1 82 COMPARATIVE PHYSIOLOGY OF THE BRAIN

does not fly into the flame out of '* curiosity," neither
is it ** attracted " by the Hght ; it is only oriented by it
and in such a manner that its median-plane is brought
into the direction of the rays and its head directed
toward the source of light. In consequence of this
orientation its progressive movements must lead it to
the source of light.

We now come to the most important question in
this chapter, namely, the relation of the central
nervous system to the instincts. As long as such
apparently complex things as the instincts are not
analysed but treated as entities, it is easy to believe
that they are based upon very mysterious nervous
structures. It would harmonise with the centre-
theory to assume for the moth a '*flying-into-the-
flame centre," ^ and to seek for its localisation in the
central nervous system. The fact that the flying of
the moth into the flame is nothing but positive helio-
tropism, and the fact that the positive heliotropism of
animals is identical with the positive heliotropism of
plants, proves that this reaction must depend upon
conditions which are common to animals and plants.
Plants, however, possess no central nervous system,
therefore I believe that it is impossible for the helio-
tropic reactions of animals to depend upon specific
structures of the central nervous system. It is much

*Steiner tries indeed to "explain" the righting motions of the starfish by
the assumption of a "righting centre " in the central nervous system. He
does not consider the possibility that contact stimuli and the irritable structures
at the periphery may be sufficient for this reaction, and that the nerves act
only as protoplasmic conductors between the skin and the muscles.

THEORY OF INSTINCTS 183

more probable that they are determined by properties
which are common to animals and plants. From what
has been said above it is easy to infer what these
properties are : First, heliotropic animals as well as
heliotropic plants must contain a substance on their
surfaces which undergoes a chemical change when
subjected to the influence of the light, and this change
must be able to produce changes of tension in the
contractile tissue. Second, heliotropic animals as
well as heliotropic plants possess symmetry of form
and a corresponding distribution of the irritabilities.
These two groups of conditions determine the helio-
tropic reaction unequivocally. But what has the
central nervous system to do with this ** instinct " of
the moth to fly into the flame, or, as we may now
say, with its heliotropism ? I believe nothing more
than that the nervous system contains a series of
segmental ganglia which establish the protoplasmic
connection between the skin and muscles. If we
destroy the central nervous system, the heliotropic
reactions in many animals cease, but mainly for the
reason that the connection between the skin, or the
eyes, which are affected by the light, and the muscles,
is interrupted. Hence it would be just as wrong to
assume a specific centre for the flight of the moth
into the flame as it would to assume a specific centre
for the going of Palsemonetes to the anode.

3. We will select another instinct, namely, the
habit many animals have of crawling into cracks and
crevices. This " instinct " is very prevalent in the

i84 COMPARATIVE PHYSIOLOGY OF THE BRAIN

animal kingdom, especially among insects, worms, etc.
This is called an instinct of self-preservation, and it
is assumed that the animal thus escapes from its
pursuers. The centre theory would assume a special
centre for this instinct. This is, however, only an-
other instance of a simple tropism. Many plants
and animals are forced to orient their bodies in a cer-
tain way toward solid bodies with which they come in
contact. I have given this kind of irritability the
name stereotropism. Like the positive and negative
heliotropism and geotropism, there is also a positive
and negative stereotropism, and there are also stereo-
tropic curvations. I have found, for instance, that
when a Tubularia is brought in contact with a solid
body, the polyp and the growing tip bend away from
the body while the stolon sticks to it. The polyp
is negatively stereotropic and the stolon positively
stereotropic. Stereotropism plays a very important
part in the processes of pairing and the formation of
organs. The tendency of many animals to creep
into cracks and crevices has nothing to do with self-
concealment, but only with the necessity of bringing
the body on every side in contact with solid bodies.
I have proved this, for instance, in a peculiar species
of butterfly, Amphipyra, that is a fast runner. As
soon as free, it runs about until it finds a corner or a
crack into which it can creep. I placed some of these
animals in a box, one half of which was covered with
a non-transparent body, the other half with glass. I
covered the bottom of the box with small glass

THEORY OF INSTINCTS 185

plates which rested on small blocks, and were raised
just enough from the bottom to allow an Amphipyra
to get under them. Then the Amphipyra collected
under the little glass plates, where their bodies were
in contact with solid bodies on every side, not in the
dark corner where they would have been concealed
from their enemies. They even did this when in so
doing they were exposed to direct sunlight. This re-
action also occurred when the whole box was dark. It
was then impossible for anything but the stereotropic
stimuli to produce the reaction. The same phenom-
enon may be observed in worms, for instance, in
Nereis. If an equal number of Nereis and glass tubes
be placed in a dish of sea- water, we may be sure
that after a time we shall find a worm in each tube.
This even occurs when the tubes are exposed to the
direct rays of the sun, which kill the worms. This is
also a reaction which is common to plants, hydroids,
and animals possessing a central nervous system,
which must therefore depend upon circumstances
which have nothing directly to do with the central
nervous system. These circumstances are apparently
chemical effects in the skin, which are produced in
these forms by the contact with solid bodies. This is
another instance where the central nervous system
only plays the part of a protoplasmic conductor. It
would be entirely wrong to attempt to look for a
'* centre of self-concealment " in these animals. This
is confirmed by experiments on worms that have been
cut into pieces.

1 86 COMPARATIVE PHYSIOLOGY OF THE BRAIN

4. We will now turn our attention to the consider-
ation of some more complicated instincts. It always
seemed to me one of the most wonderful arrange-
ments in nature that, in many species, the female lays
her eggs in places where the newly born larvae find
just the kind of food they require. The fly lays its
eggs on decaying meat, cheese, or similar material,
and it is on these substances that the young larvae
feed. I have often placed pieces of lean meat and
pieces of fat from the same animal side by side on
the window-sill, but the fly never failed to lay its eggs
on the meat and not on the fat. I further tried to
raise the larvae on fat. As was to be expected, they
did not grow, but soon died. It was possible to dis-
cover the mechanics of the peculiar instincts of the
mothers through experiments on the young larvae.
The larvae are oriented by certain substances which
radiate from a centre, and this orientation takes place
in the same way as in the orientation of heliotropic ani-
mals by the light. The centre of diffusion takes the
place of the source of light, and the lines of diffusion
(that is the straight lines along which the molecules
move from the centre of diffusion into the surrounding
medium — i. e., the air) the place of the rays of light.
The chemical effects of the diffusing molecules on
certain elements of the skin influence the tension of
the muscles, as the rays of light influence the tension
of the muscles in heliotropic animals. The orienta-
tion of an organism by diffusing molecules is termed
chemotropism, and we speak of positive chemotropism

THEORY OF INSTINCTS 187

when the animal Is forced to bring Its axis of symme-
try into the direction of the lines of diffusion and to
turn its head toward the centre of diffusion. In
such an orientation every pair of symmetrical points
on the surface of the animal is met by the lines of
diffusion at the same angle. It can easily be shown
that larvae of the fly are positively chemotropic toward
certain chemical substances which are formed, for
Instance, in decaying meat and cheese, but which are
not contained in fat. The substances in question
are probably volatile nitrogenous compounds. The
young larvae are probably led by those substances to
the centre of diffusion in the same way as the moth
into the flame. The female fly possesses the same
positive chemotropism for these substances as the
larvae, and is accordingly led to the meat. As soon
as the fly is seated on the meat, chemical stimuli seem
to throw into activity the muscles of the sexual or-
gans, and the eggs are deposited on the meat. It
may also be possible that at the time when the fly Is
ready to deposit Its eggs the positive chemotropism
is especially strongly developed. It is only certain
that neither experience nor volition plays any part in
these processes. If the question be raised as to what
is necessary in order to produce these reactions, the
answer is, first, the presence of a substance in the skin
or certain parts of the skin (sense-organs) of the ani-
mal which is altered by the above-mentioned volatile
substances contained in the decaying meat, and
second, the bilateral symmetry of the body. The

i88 COMPARATIVE PHYSIOLOGY OF THE BRAIN

central nervous system plays no other role in this than
that it forms the protoplasmic bridge for the conduc-
tion from the skin to the muscles. In organisms in
which this conduction is possible without a central
nervous system, in plants, for instance, we also find
the same reactions.

5. We find another instance of a preservative in-
stinct in the young caterpillars of many butterflies.
The larvae of Porthesia chrysorrhoea creep out of the
eggs in the autumn and winter in colonies in a nest on
trees or shrubs. The warm spring sun drives them
out of the nest and they crawl up on the branches of
the tree or shrub to the tip, where they find their first
food. After having eaten the tips, they crawl about
until they find new buds or leaves, which in the mean-
time have come out in great numbers. It is evident
that the instinct of the caterpillars to crawl upwards,
as soon as they awake from the winter sleep, saves
their lives. Were they not guided by such an in-
stinct, those that crawled downwards would die of
starvation. What r6le does the central nervous sys-
tem play in these instincts ?

I have found that the young caterpillars of Por-
thesia are oriented by the light. Until they have
taken food they are positively heliotropic. This
positive heliotropism leads them to the tips of the
branches where they find their food. During the win-
ter they are stiff and do not move. The higher tem-
perature of the spring brings about chemical changes in
their bodies, and these chemical processes cause them

THEORY OF INSTINCTS 189

to move. But the direction of their movements is
determined by the light. Out-of-doors, where the
diffused light strikes the animal on all sides, every ray
of light can be resolved into a horizontal and a verti-
cal component. The horizontal components destroy
each other, and only the effect of the vertical compo-
nents remains. Hence the animals are forced, as a re-
sult of their positive heliotropism, to crawl upwards
until they reach the tip of a branch. They are held
there by the light. The chemical stimuli which are
transmitted to the animal by the young buds produce
the eating movements. In this instinct, which is
necessary for the preservation of life, we have an-
other instance of simple positive heliotropism, and the
central nervous system plays only the r6le of a proto-
plasmic connection between the skin and contractile
tissue, which in plants is performed just as success-
fully by undifferentiated protoplasm.

We have seen, however, that these same caterpil-
lars leave the tips of the branches as soon as they
have eaten and crawl downward. Why does the
light not hold them on the highest point permanently ?
My experiments showed that these caterpillars are
only positively heliotropic as long as they remain un-
fed ; after having eaten they lose their positive helio-
tropism. This is not the only instance of this kind,
for I have found a series of facts which show that
chemical changes influence the irritability of the ani-
mal toward light. We can imagine that the taking
up of food leads to the destruction of the substances

I90 COMPARATIVE PHYSIOLOGY OF THE BRAIN

in the skin of the animal which are sensitive to Hght,
upon which substances the heHotropism depends, or
that through the consumption of food the action of
these substances is indirectly prevented.

6. The analysis of other instincts, for instance, the
migratory instinct of animals, leads to the same result
as the analysis of the protective instincts. These in-
stincts are not functions of certain localised *' centres,"
but of irritabilities of certain peripheral structures and
of the connection of the same with the muscles, whereby
the central nervous system only serves as a protoplas-
mic connection. It would naturally be more inter-
esting to select for our discussion the migrations of
birds, but it is difficult to make laboratory experi-
ments on this subject, and without laboratory experi-
ments we cannot easily obtain reliable results. For
this reason I have made use of another class of
periodic migrations, namely, the periodic depth-migra-
tions of pelagic animals. A great number of these
animals begin a vertical upward migration toward the
surface of the ocean in the evening, while in the
morning they migrate downwards. It is a remarkable
fact that these forms never go below a depth of four
hundred metres in their downward migrations. This
fact suggests that the light is the controlling power
in these depth-migrations. Water absorbs the light,
and the thicker the layer of water the more the light
is absorbed. It has been found that at a depth of
four hundred metres a photographic plate is no longer
affected. My investigations show that the movable

THEORY OF INSTINCTS 191

animals living at the surface of the ocean are all per-
manently or transitorily positively heliotropic (and
also often negatively geotropic). Those among them
that carry out the daily depth-migrations described
above have some other peculiarities which wo can
only understand if we go somewhat deeper into the
theory of animal heliotropism. We have already
mentioned that there is a negative as well as a posi-
tive heliotropism : negatively heliotropic animals bring
their median-plane into the direction of the rays of
light, but turn their aboral pole toward the source of
light. The difference in the behaviour of negatively
and positively heliotropic animals is as follows : If
light strikes one side of a positively heliotropic animal,
an increase takes place in the tension of those mus-
cles which turn the head to the source of light, while
in the negatively heliotropic animal under the influ-
ence of one-sided illumination a decrease takes place
in the tension of the same muscles. The result is
that the negatively heliotropic animal is forced to
move away from the source of light. Perhaps still
another possibility should be considered here, namely,
that the light aids the progressive movement when it
strikes the oral end of a positively heliotropic animal,
while it inhibits the progressive movement when it
strikes the aboral end. The opposite may be true of
negatively heliotropic animals. This would suggest
a further analogy between heliotropism and galvano-
tropism.

Groom and I performed experiments on the larvae

192 COMPARATIVE PHYSIOLOGY OF THE BRAIN

of Balanus perforatus which were known to make
periodic depth-migrations (2). As one of our results,
we found that these animals are sometimes negatively,
sometimes positively heliotropic, and that we were
able to make them positively or negatively heliotropic
at desire. By weak light, especially gaslight, which
contains comparatively few of the heliotropically effect-
ive blue rays, they became and remained positively
heliotropic, while in strong light they soon became
negatively heliotropic. This circumstance determines
the periodic depth-migration of these animals. When
they are near the surface of the ocean in the morning,
the strong light makes them negatively heliotropic and
forces them to go downwards vertically, because in
the open sea only the vertical components of the re-
flected skylight have any effect. As soon as they
approach a depth where the light is sufficiently weak,
they become positively heliotropic. They must then
begin to migrate upward again, but cannot reach the
surface, because they soon come to a region where the
light is so strong that they again become negatively
heliotropic. Hence during the day they are held at a
certain depth, which is, however, less than four hun-
dred metres. But as soon as it becomes dark and the
intensity of the light decreases more and more, they
are forced to rise to higher regions on account of their
positive heliotropism, until during the night, while the
intensity of the light is weak, they are held at the sur-
face of the water. Toward morning, when it begins
to dawn, they again become negatively heliotropic and

THEORY OF INSTINCTS 193

once more begin their downward migration. But the
pelagic animals also show another depth-movement
of a greater period, which corresponds more nearly
with the migration of birds of passage. Chun has
found that in the Bay of Naples during summer cer-
tain forms also remain at a greater depth during the
night, never coming to the surface. This is probably
due to the higher temperature which the surface of
the water has in summer. I have found that certain
animals, for instance, the larvae of Polygordius, are
positively heliotropic in a low temperature, while
in a higher temperature they become negatively
heliotropic (4).

I have also mentioned that geotropism also plays a
part in these depth-migrations. The same circum-
stances which make the animals negatively helio-
tropic also make them positively geotropic, and vice
versa. Thus I was able to show that in a low tem-
perature the larvae of Polygordius are also negatively
geotropic, while in a high temperature they are posi-
tively geotropic (4). By means of this geotropism
they are also forced in the dark to go to the surface
when the temperature of the water is low. It is also
probable that in many forms internal conditions simi-
lar to the nyctitropic phenomena in plants are in-
fluential in causing periodic depth-migrations.^ We
thus find that the migratory instinct, as far as it is

* This may account for the periodic migrations of certain animals (Medusae)
in polar regions. In such animals, changes in the specific gravity may take
the place of heliotropic reactions.

194 COMPARATIVE PHYSIOLOGY OF THE BRAIN

expressed in the depth-migration of pelagic animals, is
frequently determined by the presence of substances
in the surfaces of the animal which are sensitive to
light. These substances, however, produce different
effects according to the intensity of the light or of
the temperature (or perhaps according to internal
conditions). They are further determined by the re-
lations of symmetry of the animals. The central
nervous system has nothing further to do with these
phenomena than that it furnishes the protoplasmic
connection between the skin and muscles. This dis-
agrees with the centre theory of these instincts, but
agrees with the segmental theory.

7. One might think that these ideas held good only
for Invertebrates. Goltz has, however, made a re-
markable discovery which seems to confirm the opin-
ion that in Vertebrates the conditions are practically
the same. A female dog that has given birth to a
young one bites off the navel cord, licks the young, is
very affectionate towards it, and allows no stranger to
touch it. These motherly instincts are inherited, and
there is no doubt that with the act of giving birth and
the resulting processes in the sexual organs changes
take place in the animal which make these instincts
possible. One might think, especially in this case, of
centres in the central nervous system which are stim-
ulated directly through the nerves of the uterus. Now
Goltz found that these instincts are also fully devel-
oped in dogs whose spinal cord is severed so far up
that the stimuli from the uterus cannot reach the

THEORY OF INSTINCTS 195

brain (6). It is probable that certain substances
which are developed during the pregnancy, birth, and
lactation influence the character of the animal, just
as certain poisons, for instance, alcohol, tobacco, or
morphine, influence the reactions of a human being.
It is of course possible that the sympathetic plays a
part here, although this has been rendered improbable
through the more recent experiments of Goltz and
Ewald and of Ribbert.^

8. We have confined our attention to the simplest
instincts, for these are best adapted for a complete an-
alysis. Should we attempt a complete enumeration
and discussion of instincts, we should have to devote
several volumes to that subject alone. We should
like to call attention to the conditions which are re-
sponsible for the fact that many instincts are difficult
to analyse. One source of complication lies in the
fact already mentioned, that changes in the condition
of the blood, for example, those produced by metabol-
ism, may change the forms of irritability and reaction.
The young caterpillar of Porthesia is only heliotropic
so long as it is starving, while it becomes indifferent
to light as soon as it is fed. In plant-lice, the helio-
tropic irritability is connected with the growth of
wings. The wingless forms may or may not show
positive heliotropism ; if we produce wings (by
lowering the temperature or by letting the plant on
which it lives dry out), the animal becomes energetic-
ally positively heliotropic. In ants heliotropism is more

* See next chapter.

196 COMPARATIVE PHYSIOLOGY OF THE BRAIN

intimately connected with the sexual development. I
have never found true heliotropism in the workers,
while the sexually mature males and females are de-
cidedly positively heliotropic. Wherever these tran-
sitory changes of irritability are present, it requires
experimental work to succeed in the analysis of the
instinct.

A second series of difficulties arises from the influ-
ence of associative memory in many cases of instincts.
The periodic depth-migration of marine animals is a
simple case of instinctive migrations, while the migra-
tions of birds or the accomplishments of the carrier-
pigeon seem to be complicated by memory. It seems
to be certain that the carrier-pigeon finds its way back
by its visual memory of the locality from which it
started. In the same way the migration of birds may
be determined, if it is true that migrating birds return
to their old nest. In the case of the birds, there is
present in addition a purely inherited, instinctive
element which causes restlessness at the time of
migration. This restlessness and, perhaps to a cer-
tain extent, the direction of its flight are susceptible
of a purely physiological analysis. The element of
memory complicates many instinctive actions of wasps.
I have had a chance to observe solitary wasps and am
convinced that they find the way to their nest by means
of the visual memory of the locality where it is situ-
ated. The same is apparently true of bees and pos-
sibly of ants. (See Chapter XV.)

9. The analysis of instincts from a purely physio-

THEORY OF INSTINCTS 197

logical point of view will ultimately furnish the data
for a scientific ethics. Human happiness is based up-
on the possibility of a natural and harmonious satis-
faction of the instincts.^ One of the most important
instincts is usually not even recognised as such,
namely, the instinct of workmanship.''^ Lawyers,
criminologists, and philosophers frequently imagine
that only want makes man work. This is an errone-
ous view. We are instinctively forced to be active
in the same way as ants or bees. The instinct of
workmanship would be the greatest source of happi-
ness if it were not for the fact that our present social
and economic organisation allows only a few to sat-
isfy this instinct. Robert Mayer has pointed out that
any successful display or setting free of energy is a
source of pleasure to us. This is the reason why the
satisfaction of the instinct of workmanship is of such
importance in the economy of life, for the play and
learning of the child, as well as for the scientific or
commercial work of the man.

10. We have finally to defend our physiological
analysis of instincts against the reproach that it ignores
the theory of evolution. In other words, it has been

' It is rather remarkable that we should still be under the influence of an
ethics which considers the human instincts in themselves low and their gratifi-
cation vicious. That such an ethics must have had a comforting effect upon
the Orientals, whose instincts were inhibited or warped through the combined
effects of an enervating climate, despotism, and miserable economic conditions,
is intelligible, and it is perhaps due to a continuation of the unsatisfactory eco-
nomic conditions that this ethics still prevails to some extent.

^ I take this name from Veblen's book on The Theory of the Leisure Class^
New York, 1899.

198 COMPARATIVE PHYSIOLOGY OF THE BRAIN

urged against us that instincts should be explained
historically and not physiologically or causally. It
seems to me that living organisms are machines and
that their reactions can only be explained according
to the same principles which are used by the physicist.
Our ultimate aim in the analysis of instincts is to find
out by which physical and chemical properties of pro-
toplasm they are determined. Of course the physicist
finds it useful to illustrate the mechanism of compli-
cated machines by the comparison with simpler or
older machines of the same kind. We have made use
of this same method and heuristic principle by utilis-
ing in this book the reactions of simpler forms for the
analysis of more complicated forms. Even if we
were ill possession of a scientific phylogeny instead of
the fairy tales that go by that name at present, it
would not relieve us of the task of explaining the
instincts on the basis of the physical and chemical
qualities of protoplasm.

II. At first sight it may seem a hopeless task to
find a connection between the instinctive actions of
animals and the properties of their protoplasm. And
yet the task is not so great if we choose the right
method. This method, in my opinion, consists in
varying the instincts of an animal at desire. If we
succeed in this we are able to find out how the physi-
cal qualities of protoplasm may affect the instincts. I
have tried this in one case. A number of marine ani-
mals (Copepods, larvse of Polygordius) which go away
from the light can be forced to go to the light in two

THEORY OF INSTINCTS 199

ways, first by lowering the temperature, and second,
by increasing the concentration of the sea-water
(whereby the cells of the animals lose water). This
instinct can again be reversed by raising the tempera-
ture or by lowering the concentration of the sea-
water. Hence these instincts must depend upon
such reversible changes in the material of the proto-
plasm as can be brought about by a loss of water or
by a reduction of temperature. What these changes
are can only be determined by further experiments.
We find other instances where decrease in tempera-
ture has the same physiological effects as a loss of
water. Plant-lice exist in wingless and in winged
forms. We can at any time cause the growth of
wings in the wingless forms by lowering the tempera-
ture or by letting the plant dry out (whereby the
amount of water in the cells of plant-lice is reduced).^

Bibliography.

1. LoEB, J. Der Heliotropismus der Thiere und seine Ueber-
einstimmung mit dem Heliotropismus der Pflanzen. Wiirzburg
1890.

2. Groom and Loeb. Der Heliotropismus der Nauplien von
Balanus perforatus und die periodischen Tiefenwanderungen pelagis-
cher Thiere. Biologisches Centralblatt^ Bd. x., 1890.

3. Loeb, J. Ueber den Instinct und Willen der Thiere,
P Auger's ArchiVy Bd. xlvii., p. 407, 1890.

' I have found repeatedly that by the same conditions by which phenomena of
growth and organisation can be controlled the instincts are controlled also.
This indicates that there is a common basis for both classes of life phenomena.
This common basis is the physical and chemical character of the mixture of
substances which we call protoplasm.

200 COMPARATIVE PHYSIOLOGY OF THE BRAIN

4. LoEB, J. Ueber kunstliche Umwandlung positiv heliotropis-
cher Thiere in negativ heliotropische und umgekehrt. Pfliiger's
Archiv, Bd. liv., 1893.

5. LoEB, J. On Egg Structure and the Heredity of Instincts.
The Monist, July, 1897.

6. GoLTZ, F. Ueber den Einfluss des Nervensystems auf die
Vorgdnge wdhrend der Schwangerschaft und des Geburtsaktes.
PflUger's Archiv, Bd. ix., 1874.

CHAPTER XIV

THE CENTRAL NERVOUS SYSTEM AND
HEREDITY

I. The question as to how far the central nervous
system comes into consideration for the processes of
heredity is of great importance in educational prob-
lems. If we could hope that, as a result of the activ-
ity of a generation, its descendants would be born
with a talent for this special activity, there would be a
fertile field for the improvement of the human race.
In order to decide this question, we must first turn our
attention to those peculiarities which we know to be
hereditary — namely, the form of the body and the in-
stincts. The analysis of the instincts given in the
previous chapter places us in a position to answer the
question as to how they can be transmitted through
the ^^g. All hereditary qualities of form, instincts,
and reflexes must be transmitted through the sex-
ual cells. The difficulty that appears is this : How
can the sexual cells, which only represent a liquid
mass enclosed in solid membranes, be the bearers of
such apparently complicated structures as the forms
that originate from them with their instincts and

20I

202 COMPARATIVE PHYSIOLOGY OF THE BRAIN

reflexes ? Either the apparent sImpHcity of the struct-
ure of the egg is only an illusion, and in reality the
structure of the ^gg is no less complicated than the
full-grown animal, or the sum of the elements which
we call the form and instincts of the full-grown ani-
mal is only the resultant of a few simpler elements
which can readily be transmitted through the ^gg
without its possessing a complicated structure. The
discussion of the mechanics of instincts in the last
chapter shows the latter to be the case. Let us con-
sider those instincts that depend on heliotropic reac-
tions— for instance, the flying of the moth into the
flame. This instinct is unequivocally determined, first,
by the presence of a substance in the surface of the
animal which is sensitive to light, and second, by the
symmetrical structure of the animal. For the trans-
mission of a substance which is sensitive to light
through the ^gg no complicated mysterious structure
is necessary. Neither is a complicated structure
necessary for the ^gg in order that it may transmit
the symmetrical relations of the animal.

For the inheritance of form the conditions are not
very different. The ^gg is not the bearer of the form
of the full-grown animal, but of certain chemical sub-
stances, especially of ferments. According to the
stereochemical configuration of the latter, the products
of assimilation, and with these the materials of the
body, turn out differently. The process of develop-
ment is not only a morphological but a chemical dif-
ferentiation, and new combinations of substances are

NERVOUS SYSTEM AND HEREDITY 203

continually formed from the original raw material.
A further differentiation of the form may be and often
is connected with every metabolic differentiation of
the substance of the body. The results of experi-
mental morphology harmonise entirely with this con-
ception which was originated by Jaeger and Sachs,
and which I have tried to develop in a series of
articles. I will only mention the experiment in which
the ^^'g of the sea-urchin (Arbacia) was given the
form of a double sphere, whereby each sphere de-
veloped into a complete sea-urchin. In this case it
makes no difference whether the transformation of
the sea-urchin into a double sphere takes place in the
freshly fertilised ^gg or after the ^gg has already
reached the 16- or 32-cell stage. These facts can only
be understood if we think of the ^gg as nothing more
than the bearer of certain chemical swhstdinc^s and not
of mysterious morphological structures of a nature as
complicated as that of the full-grown animal ; and if
we regard the morphological process of development
only as a result or accompanying phenomenon of
corresponding chemical transformations and physical
changes. We may mention further in this connec-
tion that the processes of heteromorphosis — that is,
the transformation or substitution of one organ for
a morphologically different one by means of certain
external influences — force us to the same view.

2. Tornier has developed a theory of the inherit-
ance of acquired characters on the assumption of a
new role of the central nervous system. According

204 COMPARATIVE PHYSIOLOGY OF THE BRAIN

to this theory, every change that takes place in the
body is said to be accompanied by a corresponding
change in the central nervous system. The changes
in the central nervous system are then said to bring
about a corresponding change in the ^^g. Thus, ac-
cording to this theory, just as close a relation must
exist between the central nervous system and the
morphogenetic processes as between the central nerv-
ous system and the motor and sensory functions.
It can readily be shown, however, that this assumption
of Tornier goes much too far. When the larva of
Amblystoma transforms itself into a sexually mature
animal, it loses the gills which are located on the head
and the tail-fins that are on the tail. Both organs
disappear simultaneously. In a series of Amblystoma
larvae I severed the spinal cord in the vicinity of the
shoulder-girdle. The parts of the animal before and
behind the place of division are, as regards motor and
sensory functions, like two separate animals. If the
morphogenetic processes were as closely related to
the central nervous system as the sensory and motor
functions — as Tornier's theory demands — we should
have expected that the gills and tail-fins would no
longer be absorbed simultaneously, but at different
times, just as in two different animals. Without ex-
ception, in these animals with severed spinal cord the
absorption of the head- and tail-organs occurred simul-
taneously (i). In some of the animals operated upon,
the change took place in a few days after the division,
in others a longer interval elapsed. There can thus

NERVOUS SYSTEM AND HEREDITY 205

be no doubt that the connection between the morpho-
genetlc functions and the central nervous system is
much more sHght than between this organ and the
sensory and motor functions.

I am incHned to believe that the simultaneous dis-
appearance of gills and tail-fins is due to some change
occurring in the blood, — e.g., the appearance of certain
enzymes, or possibly changes in the number of red
blood corpuscles, etc.

It has been urged that in this experiment the sympa-
thetic system transmitted the connection between the
two halves of the animal. The sympathetic has
always been used as a bridge across the gulf between
preconceived notions and facts. I am pretty certain
that at least in a number of my Amblystomas the sym-
pathetic was cut. But as I did not make sure of this
I will not urge this point. But I may at least point
out the true reliability of this bridge. It had been a
generally accepted belief that the secretory activity
of the milk glands during and after pregnancy was
caused by the stimulation of the nerve-endings of the
uterus. Goltz severed the spinal cord in the pectoral
region of a female dog which afterwards became preg-
nant and gave birth to young ones. It turned out
that the mammary glands in front and behind the place
of the section began to secrete milk equally well.
Goltz concluded that the secretion was not due to a
nervous influence. As was to be expected, those who
try to explain everything by the omnipotence of the
central nervous system at once pointed out that the

2o6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

sympathetic connected the two halves of the spinal
cord in Goltz's dog. Recently Ribbert made an ex-
periment which, if correct, does away with these
mysterious sympathetic influences (8). He trans-
planted a milk gland to the ear of a guinea-pig. The
guinea-pig became pregnant and the gland on the ear
began to secrete. It is evident that a change in the
blood or lymph must be responsible for the secretion
of milk glands during pregnancy, possibly the appear-
ance of certain enzymes.

Schaper has added an experiment that speaks for
the lack of dependence of the morphogenetic devel-
opment on the central nervous system. In a tadpole
six mm. long he extirpated the brain and the medulla
oblongata. When the animal was killed seven days
later, the spinal cord seemed to have vanished.
Nevertheless the healing of the wound, growth, and
development continued during the seven days (2).
In face of the fact that the first processes of de-
velopment precede the formation of the central ner-
vous system in every animal, these results need not
surprise us. They suffice, however, to convince us
that the processes of development and the formation
of organs are less closely connected with the central
nervous system than the sensory and motor processes.
For this reason we cannot well decide in favour of the
assumption that every impression on the central nerv-
ous system must impart itself to the ^^^, with which
it is, moreover, not connected.

3. But how shall we make the fact that certain

NERVOUS SYSTEM AND HEREDITY 207

mental diseases are hereditary harmonise with this
view? It is, perhaps, not impossible that those men-
tal diseases that are hereditary are, in reality, chem-
ical diseases caused by poisons that are formed in the
body just as special substances, for instance, alcohol,
hashish, and other intoxicating substances, produce
temporary mental diseases (3). The delirium of fever
as well as certain other mental diseases may owe their
origin to poisons which are formed in the body. It
is quite possible that these poisons are also formed
in the normal body. It is only necessary that they be
formed in somewhat larger quantities or destroyed
in somewhat smaller quantities in the body of the in-
sane than in the normal man. It is further not at all
necessary that these hypothetical poisons which cause
mental diseases be formed in the central nervous sys-
tem. They may be formed in any organ of the body.
It is only necessary that they affect the central nervous
system — in other words, that they be nerve-poisons.
Nothing is better qualified to make this view clear
than the result which the destruction of the thyroid
gland has on the mental and physical development of
children. We know that in case of degeneration of the
thyroid gland the growth and mental development of
children are retarded. Idiocy may result from the de-
struction of the thyroid gland. It has been found
that an improvement or even a cure can be attained
by feeding patients afflicted with this trouble with
the thyroid substance of animals. Baumann found
that the thyroid gland contains an element which is

2o8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

contained In no other organ of the body, namely, Io-
dine. It Is thus conceivable that hereditary mental
diseases are chemical diseases. The germ-cells may
in these diseases also be influenced by the poisons
circulating In the blood.

4. If we thus deny the immediate Influence of the
central nervous system on the germ, and assume a
chemical theory of heredity, it might still be possible
that the central nervous system could influence hered-
ity indirectly, in so far as it can affect the chemical
processes of the body. As illustrations of a chemical
effect of the nerves, the fact is mentioned that stimu-
lation of the nerves of certain glands produces a secre-
tion. Mathews has shown, however, that in cases
where stimulation of the sympathetic produces a secre-
tion, the glands contain muscular fibres which contract
when stimulated, and in this way press a liquid out of
the ducts (4). (Conditions seem to be different in the
case of the secretion produced by stimulation of the
chorda, but it is also possible that in this case the secre-
tion is only an indirect effect of the stimulation caused
by changes In the circulation.) There are other
cases of an apparent chemical effect of the nerves.
The fact that herpes zoster follows the nerves has
led many to assume that this disease Is caused by a
trophical influence of the nerves. But we know that
in the case of rabies the micro-organism or the poison
creeps along the nerves. Goltz has found that ulcer-
ations and suppurations occur on the skin behind the
cut after division of the spinal cord, which are so sym-

NERVOUS SYSTEM AND HEREDITY 209

metrical that it is impossible to attribute them solely
to external injuries. They occur only during the first
weeks after the operation, disappearing later on (5).
It is conceivable that the cause of these phenomena is
to be sought in abnormal chemical processes which
are perhaps caused by the vasomotor nerves in so far
as disturbances in the supply of oxygen, etc., are de-
termined by them. These disturbances occasionally
fail to appear. Physicians are familiar with these
phenomena of bed-sores which ensue after lesion of
the spinal cord. One fact that Goltz and Ewald found
is especially interesting for the theory of these pro-
cesses. When they severed the spinal cord of animals,
these phenomena of ulceration of the skin were very
pronounced. But if they afterward operated on the
spinal cord behind the cut, the disturbances were much
less severe or failed to appear. Thus the separation
of a part of the spinal cord from the brain is accom-
panied by more serious consequences than the sub-
sequent destruction of the spinal cord itself (5).

An inflammation of the cornea occurs generally
after the division of the trigeminus of the same side.
This inflammation is naturally caused by bacteria, but
the fact that these bacteria affect the cornea whose
sensory nerve is severed might have two causes :
either the animal on account of the lack of sensibility
might not notice the foreign bodies (dust, etc.) that
get into the eye and cause a wound, or as a result of
the division of the nerve changes take place in the
cornea which render it more susceptible to inflam-

2IO COMPARATIVE PHYSIOLOGY OF THE BRAIN

matlon. The latter might be the case if Gaule's
statement is correct, namely, that histological changes
can be shown in the cornea ten minutes after
the division of the trigeminus (6, 7). In this
case it can only be that the power of resistance
or, more accurately speaking, the chemical nature
of the tissue is changed as a result of the lesion
of the nerve. If this be true, it does not force
us to the assumption of specific trophic nerves ; if
it is true that the influence of every nervous im-
pulse on the affected tissue is chemical, all nerves are
in one sense trophic, and it would be quite erroneous
to maintain that certain nerves serve trophic func-
tions exclusively while others are sensory and motor.
There are no specifically trophic nerves, but it is pos-
sible that many nerves produce indirectly (for instance,
through disturbances of the circulation and limitation
of the supply of oxygen) such extensive chemical
changes that morphological changes of the tissue
ensue.

If this is in reality the case, a possibility still exists
that the central nervous system also affects the sex-
ual cells indirectly, in so far as disturbances of circu-
lation and hence chemical changes are produced, which
may modify the sexual cells contained in the testes
and ovaries chemically. Thus there might be a very
remote chance that brain-activity of one generation
might lead to the formation of chemical substances
which affect the sexual cells. It is difficult to under-
stand, however, what should cause these sexual cells

NERVOUS SYSTEM AND HEREDITY 211

to produce descendants with greater intellect. The
intellect is not proportional to chemical changes, like
muscular activity. In the brain of an idiot and of a
genius the same chemical changes may occur. The
difference between the two, however, is that the idiot
fails to notice valuable associations of ideas while the
brain of the genius retains them. We arrive thus at
the conclusion that a transmission of hereditary char-
acteristics through the ^^g is only possible in the
form of specific chemical substances, and that the
central nervous system could only influence heredity,
if it could bring about the formation of special sub-
stances in the ^g^ (by influencing metabolism). It
would, of course, first have to be proved that the cen-
tral nervous system has such an influence upon the
sexual cells, and this is extremely doubtful. For this
reason we should not be justified in maintaining that
the activity of a generation can produce an heredi-
tary increase of the ability and tendencies in the same
direction. Herbert Spencer gives as a proof of this last
possibility the fact that the circles of touch in the tip
of our tongue are the smallest. He believes that
this is due to the fact that from time immemorial man
had the tendency to examine the spaces between the
teeth with the tongue, and this is supposed to have
caused an hereditary increase in the nerve-endings of
the tongue. Spencer overlooks the fact that in the tip
of the nose the circles of touch are also a comparative
minimum, and it is certain that this organ has not
been used for such a purpose since time immemorial.

212 COMPARATIVE PHYSIOLOGY OF THE BRAIN

It is more probable that the relative number of the
nerve-endings or, more correctly speaking, the rela-
tive size of the circles of touch in the tip of the
tongue and the tip of the nose is determined by the
relatively small radius of curvature or the minimal
areal growth of these tips.

Bibliography.

1. LoEB, J. Hat das Central nervensy stem einen Einfluss auf
die Vorgdnge der Larvenmetamorphosel Archiv fiir Entwickelungs-
mechanik, Bd. iv., 1896.

2. ScHAPER, A. Experimental Studies on the Influence of the
Central Nervous System upon the Development of the Embryo.
Journal of the Boston Soc. of Medical Science^ Jan., 1898.

3. Meyer, Adolf. A Short Sketch of the Problems of Psy-
chiatry. Am. Jour, of Insanity, vol. liii., 1897.

4. Mathews, A. The Physiology of Secretion. Annals N.
Y. Acad, of Science, vol. xi., No. 14, 1898.

5. GoLTZ and Ewald. P>er Hund mit verkiirztem Riicken-
mark. Pflilger's Arch,, Bd. Ixiii., 1896.

6. Gaule, J. Der Einfluss des Trigeminus auf die Hornhaut.
Physiologisches Centralblatt, Bd. v., 1891.

7. Gaule, J. Wie beherrscht der Trigeminus die Erndhrung
der Hornhaut. Physiologisches Centralblatt, Bd. vi., 1892.

8. RiBBERT, H. Ueber Transplantation von Ovarium, Hoden
und Mamma. Arch. f. Entwickelungsmechanik, vol. vii., 1898.

CHAPTER XV

THE DISTRIBUTION OF ASSOCIATIVE MEMORY
IN THE ANIMAL KINGDOM

I. The most important problem in the physiology
of the central nervous system is the analysis of the
mechanisms which give rise to the so-called psychic
phenomena. The latter appear, invariably, as a func-
tion of an elementary process, namely, the activity of
the associative memory. By associative memory I
mean the two following peculiarities of our central
nervous system : First, that processes which occur
there leave an impression or trace by which they can
be reproduced even under different circumstances
than those under which they originated. This pecu-
liarity can be imitated by machines like the phono-
graph. Of course, we have no right to assume that
the traces of processes in the central nervous system
are analogous to those in the phonograph. The sec-
ond peculiarity is, that two processes which occur
simultaneously or in quick succession will leave traces
which fuse together, so that if later one of the pro-
cesses is repeated, the other will necessarily be re-
peated also. The odour of a rose will at the same

213

214 COMPARATIVE PHYSIOLOGY OF THE BRAIN

time reproduce its visual image in our memory, or,
even more than that, it will reproduce the recollection
of scenes or persons who were present when the same
odour made its first strong impression on us. By
associative memory we mean, therefore, that mechan-
ism by means of which a stimulus produces not only
the effects which correspond to its nature and the
specific structure of the stimulated organ, but which
produces, in addition, such effects of other causes as
at some former time may have attacked the organism
almost or quite simultaneously with the given stimu-
lus (2). The chief problem of the physiology of the
brain is, then, evidently this : What is the physical
character of the mechanism of associative memory ?
As we said in the first chapter, the answer to this
question will probably be found in the field of physi-
cal chemistry.

I think it can be shown that what the metaphys-
ician calls consciousness are phenomena determined
by the mechanism of associative memory. Mach has
pointed out that the consciousness of self or the ego
is simply a phrase for the fact that certain constitu-
ents of memory are constantly or more frequently
produced than others (i, 11). The complex of
these elements of memory is the " ego " or the ** soul,"
or the personality of the metaphysicians. To a cer-
tain extent we are able to enumerate these con-
stituents. They are the visual image of the body so
far as it lies in the field of vision, certain sensations
of touch which are repeated very frequently, the

DISTRIBUTION OF MEMORY 215

sound of our own voice, certain interests and cares, a
certain feeling of comfort or discomfort according to
temperament or state of health, etc. (i, 11).

An inventory of all the memory-constituents of the
ego-complex of different persons would show that the
consciousness of self is not a definite unit, but, as
Mach maintains, merely an artificial separation of
those constituents of memory which occur most fre-
quently in our perceptions. These will necessarily
be subject to considerable variation in the same per-
son in the different periods of life.

If we speak of loss or an interruption of conscious-
ness, we mean a loss or an interruption of the activity
of associative memory. If a faint is caused directly
by lack of oxygen or indirectly by a disturbance in the
circulatory system, the activity of associative memory
ceases. This was proved by Speck's experiments on
the effects of a low pressure of oxygen. When he
breathed air with less than eight per cent, of oxygen,
he soon fainted. In these experiments, he had to
count the number of respirations. Before he fainted,
he became confused in his counting and forgot what
happened. When this disturbance in counting began
to appear, he knew it was time to discontinue the ex-
periment. When a loss of consciousness is produced
by narcotics or anaesthetics, we have again to deal
with an interruption in the activity of the associative
memory. It is the same in the case of a deep sleep.

The metaphysician speaks of conscious sensations
and conscious will. That the will is only a function

2i6 COMPARATIVE PHYSIOLOGY OF THE BRAIN

of the mechanism of the associative memory can be
proved. We speak of conscious volition if an idea
of the resulting final complex of sensations is present
before the movements causing it have taken place or
have ceased. In volition three processes occur. The
one is an innervation of some kind which may be
caused directly or indirectly by an external stimulus.
This process of innervation produces two kinds of
effects. The one effect is the activity of the associa-
tive memory which produces the sensations that in
former cases accompanied or followed the same inner-
vation. The second effect is a coordinated muscular
activity. It happens that in such cases the reaction-
time for the memory-effect of the innervation is
shorter than the time for the muscular effect. When
some internal process causes us to open the window,
the activity of the associative memory produces the
idea of sensations which will follow or accompany
the opening of the window sooner than the act of
opening really occurs. As we do not realise this
any more than we realise the inverted character of
the retina-image, we consider the memory-effect of the
innervation as the cause of the muscular effect. The
common cause of both effects, the innervating pro-
cess, escapes our immediate observation as our senses
do not perceive it. The will of the metaphysicians
is then clearly the outcome of an illusion due to the
necessary incompleteness of self-observation. Our
conception of will harmonises with Miinsterberg's
and James's views on this subject (6, 12). I think

DISTRIBUTION OF MEMORY 217

that we are justified in substituting the term activity
of associative memory for the phrase consciousness
used by the metaphysicians.

2. We have spoken of associative memory because
the word memory is often appHed in quite a different
sense scientifically, namely, to signify any after-effect
of external circumstances. For instance, the term
memory has been used to account for the fact that a
plant which had been cultivated in the tropics will
often not endure low temperatures so well as a plant
of the same species which was raised in the north. It
is true in this case that preceding conditions influence
the ability of the plant to react, but the process differs
from the one which we have called associative memory
in the lack of associative processes. No definite stim-
ulus produces in a plant, in addition to its own effects,
those of another entirely different stimulus which at
some former time occurred simultaneously with the
given stimulus. It is probable that the tropical plant
is somewhat different chemically from the plant raised
in the north. This would account for its smaller
power of resistance. Further illustrations of a differ-
ent use of the word memory can easily be given.

Many moths sleep during the day and wake in the
evening when it becomes dark. If kept for days in
a dark room, they will continue at first to do the same
thing. The same is true of certain plants. One
might also say in this case that the moth or the plant
" remembers " the difference between day and night.
It is probable, however, that internal changes take

2i8 COMPARATIVE PHYSIOLOGY OF THE BRAIN

place in the organism, corresponding to the periodic
change of day and night, and that these changes con-
tinue for a time in the same periodicity, when the ani-
mal is kept in the dark.

3. We will then consider the extent of associative
memory in the animal kingdom instead of the extent
of consciousness among animals. How can we deter-
mine whether an animal possesses the mechanism
necessary for associative memory ? The criteria for the
existence of associative memory must form the basis
of a future comparative psychology. It will require
more observations than we have made at present to
give absolutely unequivocal criteria. For the present,
we can say that if an animal can learn, that is, if it
can be trained to react in a desired way upon certain
stimuli (signs), it must possess associative memory.
The only fault with this criterion lies in the fact that
an animal may be able to remember (and to associate)
and yet may not yield to our attempts to train it. In
this case other experiments must be substituted which
will prove that the animal does associate or remember.

We may conclude that associative memory is pre-
sent when an animal responds upon hearing its name
called, or when it can be trained upon hearing a
certain sound to go to the place where it is usually
fed. The optical stimulus of the place where the
food is to be found and the sensations of hunger and
satiety are not qualitatively the same, but they occur
simultaneously in the animal. The fusion or growing
together of heterogeneous but by chance simultaneous

DISTRIBUTION OF MEMORY 219

processes is a sure criterion for the existence of as-
sociative memory (2).

Associative memory probably exists in most mam-
mals. The dog which comes when its name is called,
which runs away from the whip, which welcomes its
master joyfully, has associative memory. In birds, it
is likewise present. The parrot learns to talk ; the
dove finds its way home. In lower Vertebrates, mem-
ory is also occasionally found. Tree-frogs, for ex-
ample, can be trained, upon hearing a sound, to go to
a certain place for food. In other frogs, Rana escu-
lenta, for instance, no reaction is as yet known which
proves the existence of associative memory. Some
fishes evidently possess memory ; in sharks, however,
its existence is doubtful. With regard to the Inver-
tebrates, the question is difificult to determine. The
statements of enthusiasts who discover consciousness
and resemblance to man on every side should not be
too readily accepted.

4. In my experiments on the tropismsof animals, it
became clear to me how easy it is for an observer
who is inclined to think anthropomorphically to re-
gard machine-like effects of external stimuli on lower
animals as the expression of intelligence. He needs
only to neglect the analysis of the external stimuli. I
have protested against the anthropomorphisms of
Romanes, Eimer, Preyer, and others in a series of ar-
ticles (2, 3). Bethe has recently published a paper on
the psychic qualities of ants and bees in which he
took special pains not to fall into the gross anthropo-

220 COMPARATIVE PHYSIOLOGY OF THE BRAIN

morphisms which have characterised this field here-
tofore (4). But I am afraid that he went too far
and that he overlooked the fact that bees and ants
possess associative memory. Bethe assumes associa-
tive memory as the criterion for the existence of con-
sciousness, as I had done before. (He has evidently
overlooked, or at least does not mention, my work on
this subject.) According to him: '* An animal that
is able to do the same things the first day of its exist-
ence which it can do at the end of its life, that learns
nothing, that always reacts in the same way upon
the same stimulus, possesses no consciousness." This
statement is not sufficient. It is possible that an ani-
mal at birth, or just after hatching, may not be fully
developed. In this case it may be able later to per-
form actions which would have been impossible
on the first day, without possessing associative mem-
ory. Yet according to Bethe's definition such actions
would indicate associative memory.

It is a well-known fact that if an ant be removed
from a nest and afterwards put back it will not be
attacked, while almost invariably an ant belonging to
another nest will be attacked. It has been customary
to use the words memory, enmity, friendship, in de-
scribing this fact. Now Bethe made the following
experiment. An ant was placed in the liquids (blood
and lymph) squeezed out from the bodies of nest
companions and was then put back into its nest ; it
was not attacked. It was then put in the juice taken
from the inmates of a *' hostile " nest and was at once

DISTRIB UTION OF MEM OR Y 221

attacked and killed. Hence chemical stimuli of
certain volatile substances will excite the ants. In
this case we do not need to assume intelligence any
more than we do in the case of the tentacles of Ac-
tinians which, as we have seen, will immediately carry
a piece of filter paper soaked in meat-juice to the mouth
while they ignore a piece of paper soaked in sea-water.
The assumption of machine-like irritable structures is
quite sufficient here to explain the reaction. Mem-
ory is quite unnecessary. Possibly the behaviour of
the ant may be explained in the same way. Bethe
was able to prove by special experiments that these
reactions of ants are not learned by experience,
but 'are inherited. The "knowing "of *' friend and
foe " among ants is thus reduced to different reactions,
depending upon the nature of the chemical stimulus
and in no way depending upon memory.

Memory and intellect are supposed to be responsible
for the fact that an ant is able to find its way back to
the nest and that when " foragers " have discovered
honey or sugar the other ants of the nest soon go to
it in great numbers. The ability to communicate in-
formation was assumed in this case. Bethe, however,
was able to determine by means of ingenious experi-
ments that an ant, when taking a new direction from
the nest for the first time, always returns by the same
path. This shows that some trace must be left be-
hind which serves as a guide back to the nest. If
the ant returning by this path bear no spoils, Bethe
found that no other ants try this direction. But if it

222 COMPARATIVE PHYSIOLOGY OF THE BRAIN

bring back honey or sugar, other ants are sure to try
the path. Hence something of the substances carried
over this path by the ants must remain on the path.
These substances must be strong enough to affect the
ants chemically. I can prove by the following obser-
vation, which must surely have been made before me
by many breeders of butterflies, that Bethe is justified
in the assumption that insects are affected by ex-
tremely weak chemical stimuli. I placed a female
butterfly of a certain species in a cigar-box, and closed
the box. The box was then suspended half way be-
tween the ceiling and floor of the room and then the
window was opened. At first no butterfly of this
species was visible far or near. I n less than half an hour
a male butterfly of the same species appeared on the
street. When it reached the height of the window,
its flight was retarded and it came gradually toward
the window. It flew into the room and soon up to
the cigar-box, upon which it perched. During the
afternoon, two other males of the same species came
to the box. Thus we see that butterflies and certainly
many other insects possess a delicacy of chemical
irritability which, if possible, is finer than that of the
best blood-hound. Plateau maintains that insects are
attracted to the flowers by the odour rather than by the
colour and marking. The dioptric apparatus of insects
is very inferior to that of the human eye, while their
chemical irritability is much superior to that of our
olfactory epithelium. I believe that both odour and
colour may influence insects.

DISTRIBUTION OF MEMORY 223

One of the most remarkable conclusions of Bethe
is the assumption that the roads of the ants have two
paths which differ chemically from each other, one
leading from and one toward the nest. Bethe tried
to prove this by experiments that had been undertaken
before by Lubbock, who obtained no definite results.
Bethe arranged a broad ant-street so that it led over
a turn-bridge. He revolved this bridge 180°, when
the ants were passing to and from the nest, and found
that it was impossible for the two armies to continue
on their way. He again turned the bridge 180° so
that the tracks had the original orientation. The
ants continued in the direction they were pursuing
when disturbed. An observation made by Forel also
agrees with this : " An ant that is picked up from
the path while moving and then put down again is al-
most sure to take the same direction, no matter what
orientation is given to its body." This, however,
only holds good for a street which is often travelled.
A weak track which leads in one direction is qualified
to lead in the opposite direction, as is shown by the
fact that an ant which has found a new supply returns
to the nest the same way that it came. It is evidently
the load and lack of load which determine which path
the ant will take (that is, to or from the nest). The
load causes the ant to go to the nest reflexly ; the lack
of a load causes it to go from the nest. Bethe comes
to the conclusion that the reactions of ants, which
have always been considered psychic phenomena, are
merely reflex processes comparable to the tropisms.

224 COMPARATIVE PHYSIOLOGY OF THE BRAIN

5. Although I heartily sympathise with Bethe's re-
action against the anthropomorphic conception of
animal instincts, I yet believe that he is mistaken in
denying the existence of associative memory in ants
or bees. The fact that bees find their way home
through the air cannot depend upon any trace left in
their path. It can only depend upon memory and, as
I believe, upon visual memory. If the bee-hive be
removed while the bees have swarmed out, they will
return to and gather at the spot where the entrance
to the hive used to be. Bethe is not willing to admit
that this indicates the existence of a visual image of
memory of the locality of the nest, professing to con-
sider it possible that unknown forces guide the bee
reflexly.

I have recently had a chance to observe the activity
of solitary wasps and have come to the conclusion
that these animals are guided back to their nest by
their memory.

My observations were made on Ammophila, a spe-
cies of wasps, whose habits have been carefully stud-
ied and described by Mr. and Mrs. Peckham (7).
Ammophila makes a small hole in the ground and
then goes out to hunt for a caterpillar, which, when
found, it paralyses by one or several stings. The
wasp carries the caterpillar back to the nest, puts
it into the hole, and covers it with sand. Before this
is done. It deposits its ^g<g and the caterpillar serves
the young larva as food.

I will describe one observation on the means these

DISTRIBUTION OF MEMORY 225

wasps employ of finding their way to the nest, which
absolutely excludes the assumption that they are
guided refiexly by known or unknown stimuli, and
which indicates that they find their way through
memory. An Ammophila had a hole in a flower-bed
in my front yard. The wasp, of course, left the yard
flying. Towards noon I saw an Ammophila running
on the sidewalk of the street in front of the yard
carrying a caterpillar in its mouth. The weight of
the caterpillar prevented it from flying. The yard
is separated from the street by a cemented stone
wall. I noticed that the wasp repeatedly made an
attempt to climb upon the wall, but kept falling
down. Suspecting that it might have its nest in
the yard I was curious to see whether and how it
would find the nest.

It followed the wall until it reached the neighbour-
ing yard, which had no wall. It now left the street
and crept into this yard. Then crawling through
the fence which separated the two yards, it dropped
the caterpillar near the foot of a tree, and flew
away. After a short zigzag flight it alighted on
a flower-bed in which I noticed two holes. It soon
left the bed and flew back to the tree, not in a
straight line but in three stages, stopping twice on its
way. At the third stop it landed at the place where
the caterpillar lay. The caterpillar was then dragged
to the hole, pulled into it, and covered with sand.

As the wasp only walks to the hole when carrying
a caterpillar, it is impossible to say that it followed a

226 COMPARATIVE PHYSIOLOGY OF THE BRAIN

trace and was guided reflexly when It carried the cat-
erpillar to the nest. The repeated attempts to climb
the wall of the yard which first attracted my atten-
tion indicate that the wasp remembered the location
of the nest. The fact that it returned to fetch the
caterpillar indicates that it remembered having
dropped it, and also where it had been dropped. The
zigzag character of its flight shows that it was not
guided reflexly.

While these animals without doubt possess asso-
ciative memory they possess little '* intelligence."

I mentioned that the Ammophila covered the hole
in which it had buried the caterpillar. In order to
cover it, the wasp had to pick up little grains of
sand in the neighbourhood of the hole and carry them
in its mandibles to the hole. Once, while it had its
back turned to the nest and was picking up a grain of
sand, I covered the hole with a clover blossom. The
wasp was no longer able to find the hole. It ran and
flew about in the most excited manner, returning each
time to the place where the hole had been, without
being able to discover it. I finally removed the
flower, and the wasp immediately found the hole and
continued covering it with sand. The blossom with
which I covered the hole weighed considerably less
than the caterpillar which the wasp carried with such
ease between its mandibles. The fact that the wasp
kept returning to the spot where the hole was, indi-
cates again the existence of memory in these animals.

Bethe's conclusions have been criticised by Was-

DISTRIBUTION OF MEMORY 227

mann (8) as far as ants, and by von Buttel-Reepen
(9) as far as bees, are concerned. I think that bees
and ants possess associative memory. In their re-
actions, however, reflex or instinctive elements and
memory elements are mixed together. The task re-
mains to discover how much of a r6le associative
memory plays in the various habits of bees, ants, and
wasps.

6. The possibility of associative memory must be
conceded in the case of spiders, certain Crustacea, and
Cephalopods, but it is in all probability wanting in
!oelenterates and in worms. We saw that Act-
inians refuse water-soaked paper wads and take
meat, though our organs of taste cannot distinguish
between the two. Some authors would have called
this an expression of intelligence because the Actinian
can "discriminate" and **make a selection." Accord-
ing to this, consciousness and intelligence should be
attributed to the chemical elements, for they unite
only with certain other elements. The term " power
of discrimination " is often merely an ill-chosen ex-
pression for the fact that different causes have differ-
ent effects. This difference of the effects may in some
cases depend on associative memory, but in order to
find out these cases we must first prove that the
forms under consideration have associative memory.
In Actinians, however, all attempts to prove the exist-
ence of associative memory have been fruitless. This
is shown in the experiments on Cerianthus mentioned
above, in which I succeeded in producing, below the

228 COMPARATIVE PHYSIOLOGY OF THE BRAIN

normal head, a second head, which had an oral disc
and tentacles but no mouth (Fig. 12, p. 52). The ten-
tacles never learned that no mouth was present, but
continued when meat was offered to make the attempt
to force it into an opening that did not exist.

Some reactions of lower animals cannot be repeated
indefinitely. We must not conclude, however, that
this is due to processes of association and that the
animal has learned certain effects. It is a well-known
fact that many worms that live in cases suddenly
withdraw into their cases when a shadow is cast on
them. I analysed this process and showed that the
shadow has nothing to do with the phenomenon. It
is due to a reaction against negative variations in the
intensity of the light, comparable to the '' break-con-
traction " of a muscle. The experiment does not suc-
ceed if repeated an indefinite number of times in
succession. Nagel concludes from this that these
worms possess " the ability to judge." " The animal
recognises that the shadow cast so frequently does
not signify the approach of an enemy or of any other
danger" (Nagel). In reality these reactions are in-
herited forms of irritability that have nothing to do
with experience. The reason that the reaction ceases
if repeated frequently is due to a simple after-effect
of the stimulus, a case that we often meet in the physi-
ology of both animals and plants. The assumption
that such low animals as eyeless worms and snails
possess ideas or even the one idea of *' an approach-
ing enemy or other impending danger" is entirely

DISTRIBUTION OF MEMORY 229

arbitrary. Graber also maintained that animals that
go to the light do so because they love it, and another
author thought that animals fly into the flame out of
curiosity. It is not worth while to follow up such an-
thropomorphisms in the biological literature. Biol-
ogy is as much justified in ignoring them as modern
physics is in ignoring the fact that savages explain
the locomotive by supposing a horse to be concealed
within it. On the contrary, biology should concern
itself with a systematic investigation of the differ-
ent animals in regard to the existence of associative
memory. The total results of such an investigation
will furnish the material for a future comparative
psychology.

7. Our conception meets with an apparent difficulty
in the fact that stimuli which call forth sensations of
pain in us produce also reactions in lower animals
which have no memory. These reactions are natur-
ally regarded as the expression of sensations of pain.
The injured worm writhes and wriggles, and it is diffi-
cult to rid ourselves of the impression that these
movements are the expression of severe pain. Yet
W. W. Norman proved that this conclusion is by no
means justified (5, 10). He found that if an earth-
worm is divided transversely, only the posterior piece
makes these writhing movements, while the anterior
piece crawls off as if nothing had happened. It
would, of course, be absurd to assume that the pos-
terior piece alone is capable of a sensation of pain,
while the anterior piece, which contains the brain,

22,0 COMPARATIVE PHYSIOLOGY OF THE BRAIN

experiences no such sensation. If we continue the ex-
periment and divide the posterior piece in the middle,
the anterior part crawls off calmly while the posterior
part again makes writhing movements. We obtain
the same results if we divide the anterior piece. No
matter how the worm is divided, the piece in front
of the place of division shows coordinated crawling
movements, while the piece behind the place of divis-
ion makes writhing movements. It is not even neces-
sary to cut the worm. If we only touch it with the
point of a pencil the posterior part wriggles, the ante-
rior part elongates. The only conclusion that can be
drawn is that the stimulus of cutting produces a dif-
ferent effect when it extends forward through the
worm, from the effect which it produces when it ex-
tends backward. The movements do not indicate
that the animal possesses sensations of pain.

Similar observations can be made in other Annelids.
In Planarians I had already observed that they give
no evidence of pain when they are divided trans-
versely. The forward piece crawls or swims as if
nothing had happened, occasionally merely hasten-
ing its movements.

But even in insects and Crustaceans pieces can be
cut off without any reaction from the animal which
might be interpreted as the expression of a pain-
sensation.

Janet has observed that the abdomen of a bee can be
cut off while the bee is sucking honey without causing
any interruption in its occupation. In 1888 I noticed

DISTRIBUTION OF MEMORY 231

something similar in a small Crustacean, Gammarus,
during copulation. The abdomen of the male can be
cut off while it is seated on the female without caus-
ing it to release the female. In fact, unless my mem-
ory deceives me, these males without abdomen, when
torn away from the female, were ready to hold another
as soon as they could find one. Norman has added
a great many similar observations on insects and
Crustaceans (10). The result of all these observa-
tions is that either these Invertebrates do not react
to injury in a way which indicates the existence of
pain-sensation, or that, if there seem to be such re-
actions, they do not justify the assumption of the
existence of pain-sensations.

We cannot be surprised that among those repre-
sentatives of the lower Vertebrates which have no
associative memory, or only traces of it, similar con-
ditions exist.

Hermann and other physiologists maintain that
the reactions of lower Vertebrates under the influence
of an ascending current are due to pain-sensations,
while the descending current is said to have a sooth-
ing effect. Garrey and I came to the conclusion that,
in both cases, different sets of muscles were thrown
into activity (see Chapter XL). In order to test
Hermann's view, we experimented on larvae of Am-
blystoma whose spinal cord had been cut between the
head and the tail-end of the body. We found that
in the ascending current only the tail-end of the ani-
mal showed those reactions which Hermann and the

232 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Other physiologists had considered as the expression
of pain-sensations. I may mention further that when
the motions following the stimulation of the semicir-
cular canals were first observed they were considered
by some authors as the expression of pain-sensations.

Norman observed that sharks and flounders react
in no way against very severe operations, e. g., the
laying bare of the semicircular canals, provided that
respiration was not interfered with (lo). As soon
as the water-supply to the mouth was cut off, they
made violent motions, which are characteristic for the
condition of beginning asphyxiation and which have
nothing in common with conscious acts. Sharks and
flounders belong to that class of Vertebrates which
have practically no associative memory.

It therefore seems to me that our experience con-
cerning the pain-sensations of animals does not con-
tradict our view regarding the limits of associative
memory or the consciousness of the metaphysicians.

Of course I do not expect to convince the senti-
mentalists and Darwinians. The former will say that
their " feeling " tells them that an earthworm is
capable of pain-sensations. My reply to these is
that the burden of proof rests upon them. If a per-
son maintains that there is a gaseous Vertebrate in
the air it is plainly his duty to prove its existence, and
not the duty of all the other scientists to disprove it.
Otherwise we might be called upon to waste our lives
in disproving the statements of any insane person or
impostor. The Darwinians will doubt the possibility

DISTRIBUTION OF MEMORY 233

that pain-sensations or any definite characters should
appear in certain forms without existing (although in
a rudimentary form) in the whole animal kingdom.
To these we shall reply in the next chapter (p. 251).
8. At the end of the chapter on instincts we
mentioned that in those animals which possess asso-
ciative memory the instinctive reactions may be
modified or complicated by the influence of the
associations. This influence can be so powerful that
the instincts are warped or suppressed altogether.
By education and experience the memory of man is
filled with a number of associations which can inhibit
any reflex or instinctive motor process. To a certain
extent these inhibitory associations are necessary for
the preservation of the life of the individual. More-
over, it is necessary to provide the child with associa-
tions which prevent ''dissipation," e.g., the cultivation
of one or a few instincts at the expense of others.
The greatest happiness in life can be obtained only
if all the instincts — that of workmanship included —
can be maintained at a certain optimal intensity. But
while it is certain that the individual can ruin or di-
minish the value of its life by a one-sided develop-
ment of its instincts — e. g., dissipation, — it is at the
same time true that the economic and social condi-
tions can ruin or diminish the value of life for a great
number of individuals.^

* It is no doubt true that in our present social and economic condition more
than ninety per cent, of human beings lead an existence whose value is far be-
low what it should be. They are compelled by want ta sacrifice a number of
instincts, especially the most valuable among them, that of workmanship, in

234 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Although we recognise no metaphysical free-will,
we do not deny personal responsibility. We can fill
the memory of the young generation with such asso-
ciations as will prevent wrongdoing or dissipation.
If in a human being such associations are lacking, it
points to an organic deficiency or to an insui^cient
education, for which in some cases the parents, in the
majority of cases, our present social conditions, are
responsible.

Punishment is, perhaps, justifiable in so far as it
may bring about inhibitory associations or may be
able to strengthen the inhibitory associations of
weaker members of society. Inhibitions to be effect-
ive, however, must be cultivated in youth, as the time
at which the penal code is enforced is usually too late
for any lasting benefit. Cruelty in the penal code
and the tendency to exaggerate punishments are sure
signs of a low civilisation and of an imperfect
educational system.

order to save the lowest and most imperative, that of eating. If those who
amass immense fortunes could possibly intensify their own lives with their abun-
dance, it might perhaps be rational to let many suffer in order to have a few
cases of true happiness. But for an increase of happiness only that amount of
money is of service which can be used for the harmonious development and
satisfaction of inherited instincts. For this comparatively little is necessary.
The rest is of no more use to a man than the surplus of oxygen in the atmos-
phere. As a matter of fact, the only true satisfaction a multi-millionaire can
possibly get from increasing his fortune, is the satisfaction of the instinct of
workmanship, or the pleasure that is connected with a successful display
of energy. The scientist gets this satisfaction without diminishing the value
of life of his fellow-beings, and the same should be true for the business man.

DISTRIBUTION OF MEMORY 235

Bibliography.

1. Mach, E. Contributions to the Analysis of the Sensations.
The Open Court Publishing Co., Chicago, 1897,

2. LoEB, J, Beitrdge zur Gehirnphysiologie der Wiirmer.
Fflugers Archiv^ Bd. Ivi., 1894. Zur Psychologie und Physiologic
der Aktinien. PJluger's Archiv^ Bd. lix., 1896. Zur Theorie der
physiologischen Licht- und Schwerkraftwirkungen. Pfluger's
Archiv, Bd. Ixvi., 1897.

3. LoEB, J. Weitere Bemerkungen ilber den Heliotropismus der
Thiere und seine Uebereinstitnmung mit dent Heliotropismus der
Pflanzen. PlUger s Archiv, Bd. xlvii.

4. Bethe, a. Diirfen wir den Ameisen und den Bienen psychische
Qualitdten zuschreiben? Pfliiger's Archiv^ Bd. Ixx., 1898.

5. Norman, W. W. Diirfen wir aus den Reactionen niederer
Thiere auf Schmerzempfindungen derselben schliessen? Pfiilger's
Archil^ Bd. Ixvii., 1897.

6. MuNSTERBERG, H. Die Willenshandlung. Freiburg, 1888.

7. Peckham, G. W. and E. G. On the Instincts and Habits of
the Solitary Wasps. Wisconsin Geological and Natural History
Survey^ 1898.

8. Wasmann, E. Die psychischen Fdhigkeiten der Ameisen.
Zoologica^ vol. xi., 1899.

9. V. Buttel-Reepen. Sind die Bienen Reflexmaschinen?
Biologisches Centralblatt^ vol. xx., 1900.

10. Norman, W. W. Do the Reactions of the Lower Animals
against Injury Indicate Pain- Sensations ? The American J^ourn. of
Physiology^ vol. iii., 1900.

1 1. Mach, E. Die Analyse der Empfindungen und das Verhdlt-
niss des Physischen zum Psychischen. Jena, 1900.

12. James, W. The Principles of Psychology. New York, 1890.

CHAPTER XVI

CEREBRAL HEMISPHERES AND ASSOCIATIVE

MEMORY

I. The view that consciousness is only a meta-
physical term for the phenomena determined by the
mechanisms of associative memory finds support in
the results of experiments on higher animals. Extir-
pation of the cerebral hemispheres causes complete
loss of associative memory. After this operation, no-
thing remains that could possibly be interpreted by
the metaphysicians as a phenomenon of consciousness.

If the cerebral hemispheres of a Rana esculenta or
temporaria be extirpated, the frog seems on the whole
to be unchanged. This has been proved beyond
question by Schrader (i). Such a frog catches flies,
buries itself in the mud when the cold season comes,
and changes its habitation from the land to the water,
like a normal frog. None of these processes, how-
ever, are functions of associative memory ; they de-
pend upon inherited structures. The frog either has
no associative memory or it is so insignificant that it
does not in any way affect the behaviour of the frog.
This explains the fact that the loss of the cerebral

236

CEREBRAL HEMISPHERES AND MEMORY 237

hemispheres, which produces so great a change in the
personality of a higher animal, has much less effect on
a frog. In the shark, nothing in the habits or reac-
tions of the normal animal shows the existence of
associative memory. Most of its reactions are inher-
ited and composed of segmental reflexes. We find,
correspondingly, that it shows very little change after
the extirpation of the cerebral hemispheres, for in
spite of their loss the segmental reflexes are pre-
served.

It would be a mistake to assume that the loss of the
cerebral hemispheres in no way affects the animal.
Its loss has a certain effect upon the segmental re-
flexes. Nereis has no associative memory, yet it shows
a certain lack of inhibition after the loss of the supra-
cesophageal ganglion (see Chapter VI.). Something
similar is noticeable in lower Vertebrates whose cere-
bral hemispheres are removed. For instance, in ad-
ders all segmental reflexes are preserved after loss of
the cerebral hemispheres. Schrader found, however,
that such animals no longer show any " fear " — it was
not possible to frighten them although all the opticus-
reflexes still functioned (2). From this we must con-
clude that the effects of those stimuli which extend
from the opticus-segment into the central nervous
system are different, so long as the cerebral hemi-
spheres exist, from what they are when the hemi-
spheres have been extirpated. Something of this
kind also shows itself in the frog. Goltz has found
that the frog without cerebral hemispheres is better

238 COMPARATIVE PHYSIOLOGY OF THE BRAIN

suited for demonstrating reflexes than the frog with
cerebral hemispheres. If the skin on the back of
a normal frog is touched, it may or may not croak.
Goltz showed that this croaking reflex never fails in a
frog whose cerebral hemispheres have been excised
(3). In the normal frog, however, touching the skin
of the back produces, in addition, another reflex : it
shows a tendency to leap away. The normal frog as
well as the frog without cerebral hemispheres is a re-
flex animal — that is, its reactions are chiefly segmen-
tal reflexes. But there is this difference between the
two : In the animal with cerebral hemispheres the
same stimulus can produce more than a single reflex,
and this fact adds to the greater complication and
capriciousness of the reactions of the animal. On the
other hand, the cerebral hemispheres can also restrict
the play of the segmental reflexes. The clasping re-
flex of the male frog in the act of copulation is a
segmental reflex of the arm-segments during the
period of heat. It seems that sexual substances de-
termine this reflex, since it cannot be shown to exist
in animals that are castrated before the period of
heat. Now male frogs that have lost the cerebral
hemispheres are much more indifferent in the choice
of the object they clasp during the period of heat
than animals with cerebral hemispheres.

2. In birds the conditions are different from those
which exist in frogs and sharks. We are indebted to
Schrader for an exact and, in many respects, classic
investigation of the effect of the extirpation of the

CEREBRAL HEMISPHERES AND MEMORY 239

cerebral hemispheres on birds (4). The work of this
investigator, Goltz's article on the dog without cere-
bral hemispheres, and Goltz's and Ewald's article on
the dog with shortened spinal cord are among the
best on the physiology of the central nervous system.
Until their work appeared, it was a dogma (and is
still in many text-books) that animals which have lost
the cerebral hemispheres can no longer move spon-
taneously. Flourens is responsible for the statement.
Schrader first disproved it in regard to the frog and
then succeeded in disproving it in the case of birds.
** None of the animals under observation [pigeons]
showed a sleep-like condition longer than three to
four days [after excision of the cerebral hemispheres].
According to Rolando and Flourens, animals which
have undergone this operation, except when certain
stimuli are applied to the skin, remain absolutely
quiet. At first, this is true. The pigeons remain
standing, where they are placed, with ruffled feathers,
the head drawn in, the eyes closed, and often on one
leg. Occasionally they shake themselves, clean their
feathers with their beak, stretch sleepily, and in the
act of defsecation take a few steps. If left undis-
turbed, nothing else is to be observed. When thrown
up into the air, they fly down diagonally, strike ob-
stacles, and fall rather than alight on the floor, where
they at once sink back into their stupor again. If
the skin is stimulated, they take a few steps, but in so
doing are liable to run into obstacles" (Schrader,
loc, cit.).

240 COMPARATIVE PHYSIOLOGY OF THE BRAIN

The difference between Flourens's and Schrader's
observations lies in the fact that Flourens considered
this condition permanent, while Schrader showed that
it lasts only a few days or until the '' shock-effect " of
the operation has passed off. The objection might
be raised that Schrader did not entirely remove the
hemispheres, but this was not the case. Schrader's
experiments are masterpieces in regard to the perfec-
tion of the mode of operation. The contradiction in
the statements of the two authors is due to the fact,
as it so often is in brain-physiology, that the minor
effects of the operation in one case were strong, in
the other slight, or that one author based his opinion
upon the most severe disturbances, the other upon
the slightest disturbances. The latter is the only re-
liable method in physiology of the brain, because in
addition to the disturbances caused by the loss of
part of the brain, the shock-effects on the rest of the
central nervous system also appear in the mosaic of
symptoms. Schrader's experiments are models in
regard to technique, but this cannot be said of
Flourens's experiment, to which fact the excellent
investigator Magendie vainly called attention.

In Schrader's experiments, a few days after the
operation, spontaneity not only returns^ but is even in-
creased. The animal wanders about in the room un-
tiringly the greater part of the day. It is not blind,
for its movements are determined by visual impres-
sions. Like the frog without cerebral hemispheres, it
turns out to avoid obstacles. '' Dusty window-glass,

CEREBRAL HEMISPHERES AND MEMORY 241

transparent bell-jars placed in their way were avoided
just as much as chairs and table legs, or boards of
different colours." It is evident that optic space-per-
ception still continues, even when the cerebral hemi-
spheres (and with them the associative memory) have
disappeared entirely. If such a pigeon is placed in
an uncomfortable position, it flies to another place
with perfectly coordinated movements. Schrader
gives the following description : ** We place our
pigeon on the cloth-covered stopper of a large bottle.
The stopper is large enough to support the animal on
both feet, and Is placed in the middle of a large, empty
room, so that the pigeon is one to two metres above
the floor. For some minutes the pigeon sits with its
head drawn in, its feathers ruffled, in a condition of
sleep or inhibition ; then it shakes itself and begins to
turn around and look about ; finally it stoops and with
an exertion looks down on the floor as if it wished to
measure the height. It makes preparations to fly
down, stops again, however, turns about once more,
and again directs its attention to the floor. The dura-
tion of this play varies, but at last it flies down in a
slight curve and alights easily on the floor. If a
cross-bar is placed at the same height, one or two
metres from the bottle, the pigeon flies determinedly
to the bar and seats itself there. If a chair be used
instead of a bar, the pigeon is seated on the arm " (4).
These experiments show that these pigeons are able
to measure distance by visual impressions also.

Schrader s observation is also of importance for the

242 COMPARATIVE PHYSIOLOGY OF THE BRAIN

solving of the problem as to whether sensations of
space are purely a matter of memory, as Helmholtz,
among others, assumes, or whether they are deter-
mined by inherited structural conditions, as Hering,
for instance, maintains. The question is of great im-
portance for the further investigation of the mechanics
of the brain, and for this reason we mention it in
passing. It has been assumed that space-sensations
are acquired because the new-born infant does not
immediately show signs of orientation in space. The
fact is overlooked that the new-born infant comes into
the world incomplete — that is to say, certain structures
become complete during the first year or even later.
The same erroneous conclusion was drawn in regard
to walking. The child was supposed to *' learn " to
walk. The fact that the chick can walk when it
comes out of the ^'g^ would have sufficed to prevent
this error on the part of the empiricists, if physiolo-
gists had earlier appreciated the importance of com-
parative physiology. The difference between the
chick and the human suckling consists in the fact that
the structural development of the former is more
complete at the moment of hatching than the struct-
ural development of the latter at the time of its
birth. The child can begin to walk only when the
nerves, muscles, etc., have reached the required de-
gree of development. The same is true of visual
space-perception. The newly hatched chick has vis-
ual perception ; that is, it picks at points that differ
from their environment in colour and intensity of light.

CEREBRAL HEMISPHERES AND MEMORY 243

It does not learn this reaction any more than a plant
learns its heliotropic reactions, and it is no more
necessary for the suckling than for the chick to learn
space-reactions. They come *' of themselves " as soon
as the embryonic development of the suckling has ad-
vanced far enough. This conception, to which com-
parative physiology forces us, is further supported
most effectually by Schrader's observation (and by
those of earlier authors, for instance, Longet) that
visual space-perception in birds continues after the
cerebral hemispheres have been removed. The pos-
sibility that this holds good for birds and not for
mammals is refuted by a statement of Christiani in
regard to rabbits. The fact, however, that space-
reactions can be modified by the memory, that we
can ** learn " to shave before a mirror, for instance,
or can *' learn " to grasp things in spite of prismatic
glasses, does not contradict this conception any more
than the acquired accomplishments of the dancer con-
tradict the fact that normal walking is not a matter
of memory. The fact that coordinated progressive
movements on the turn-table occur in the direction of
the plane of the rotation, and those produced by a gal-
vanic current occur in the direction of the curves of
the current, also speaks for this nativistic conception.
From this digression we will now return to Schra-
der's experiments. The pigeon described above as
wandering about the room all day, sleeps at night.
Sleep has nothing to do with consciousness and mem-
ory, for it occurs in plants. It is not surprising then

244 COMPARATIVE PHYSIOLOGY OF THE BRAIN

that the animal without cerebral hemispheres shows
the difference between sleeping and waking.

The g-reat difference between the normal male
pigeon and the pigeon that has lost its cerebral hemi-
spheres is shown forcibly by the following facts :
During the period of heat the male pigeon courts the
female with cooing, but if a female pigeon is placed
before the cooing male whose hemispheres have been
removed, it remains unheeded. This entire lack of
memory is the chief point in which the animal without
cerebrum differs from the normal animal. *' For the
former everything is only a mass in space, it moves
aside for every pigeon or attempts to climb over it,
just as it would in the case of a stone. All authors
agree in the statement that to these animals all objects
are alike. They have no enemies and no friends.
They live like hermits no matter in how large a com-
pany they find themselves. The languishing coo of
the male makes as little impression upon the female
deprived of its cerebrum as the rattling of peas or the
whistle which formerly made it hasten to its feeding
place. Neither does the female show interest in its
young. The young ones that have just learned to fly
pursue the mother, crying unceasingly for food, but
they might as well beg food of a stone " (4).

Taking all the reactions of the pigeon without
cerebral hemispheres together, it seems to me that
the conclusion may be drawn that loss of the cerebral
hemispheres causes the loss of the associative mem-
ory. Inherited reactions remain after the loss of the

CEREBRAL HEMISPHERES AND MEMORY 245

cerebrum, but that which is acquired by the activity
of memory during the Hfe of the individual is lost
forever.

In order to emphasise this loss of memory after ex-
tirpation of the hemispheres, we will quote the follow-
ing observation made by Schrader on a falcon. The
falcon, as everyone knows, is a good hunter. Schrader
placed some mice, and a falcon from which the hemi-
spheres had been removed in the same cage. Every
time that a mouse moved the falcon jumped on it and
caught it in its claws, if the movement occurred with-
in its field of vision. The normal falcon in such cases
devours the mouse, but for the falcon without cerebral
hemispheres the matter was at an end when the mouse
was caught. The activity of the associative memory
was lacking and the mouse was forgotten as soon as it
ceased to move. When the falcon moved, the mouse
escaped, but if the mouse moved again the process was
repeated. Any inanimate object that moved would, of
course, be caught in the same way. The falcon and
mice remained together until one day the mouse de-
voured the back of the living falcon. Deprived of its
memory the falcon was entirely defenceless (2).

One disturbance takes place in animals that have
lost the cerebrum which does not belong in the same
class with disturbances of memory, namely, the inabil-
ity to take food unassisted. In frogs and, according
to Steiner's observations, also in fishes (5), the ability
to take food independently continues to exist after
excision of the cerebral hemispheres. Birds without

246 COMPARATIVE PHYSIOLOGY OF THE BRAIN

cerebral hemispheres starve unless they are fed.
Schrader came to the conclusion that this is due to a
disturbance of the motor innervation, for they are un-
able to swallow a pea placed in the front part of the
beak ; to be swallowed it must be placed well back
toward the throat From the results of these experi-
ments on frogs, I believe that we might go one step
farther than Schrader and conclude that, in this case,
the tension is decreased in certain groups of muscles
which are necessary for taking food independently.
We shall again meet with such a decrease in the ten-
sion of certain muscles after lesion of the cerebral
hemispheres. This decrease in tension is, however, a
secondary effect of the operation on the remaining
segmental tracts of the central nervous system, and is
not determined by loss of the cerebrum. It is very
probable too that if Schrader's experiments are con-
tinued birds may be found in which disturbances in
eating will not occur.

3. The bold attempt to remove both hemispheres
entirely from a full-grown dog and then to keep it
alive not only for months but for years has been
attempted and carried on successfully by Goltz (6).
The results of his experiments in a few words are as
follows : In such a dog all those reactions in which the
associative memory plays a rdle are lacking perma-
nently, while the simple reactions that only depend on
inherited conditions remain just as in pigeons and in
other animals.

The dog without cerebral hemispheres sleeps and

CEREBRAL HEMISPHERES AND MEMORY 247

wakes. It moves spontaneously — that is, without visi-
ble external stimulus. The only abnormal feature in
the progressive movements of the dog without cere-
bral hemispheres was its extreme restlessness. When
not asleep it moved about in the cage unceasingly,
and this perhaps accounts for the fact that such ani-
mals show a tendency to lose flesh. The postures
peculiar to dogs in urinating and defaecation were still
assumed by these dogs. The reactions to sensory
stimuli were normal in so far as no associative mem-
ory was necessary. Meat and milk were devoured
greedily, but if made bitter with quinine they were
ejected. The dog growled and snapped if its paw was
pinched. If its foot was placed in cold water it was
removed at once. If one paw was injured the dog
was still able to go on three legs. If it was asleep it
could be waked by blowing a horn in the next room.
If in a dark room it closed its eyes when a strong
light was suddenly allowed to strike it. It seemed
more wide-awake and restless when it was hungry and
more quiet after it had been fed. In regard to eating,
the dog without cerebral hemispheres was more nor-
mal than Schrader's doves. To make the dog eat, it
was only necessary to hold the plate up to its nose, so
that the nose came in contact with the meat. The
facts that motor disturbances exist and that such
dogs do not turn out for obstacles, behaving in this
regard like blind dogs, may be regarded as shock-
effects on the brachial and optic segments, produced by
the operation. The dog could still bark and howl.

248 COMPARATIVE PHYSIOLOGY OF THE BRAIN

But everything requiring associative memory was
gone. The dog was not able to seek its food. It
recognised neither its master nor its playmates. It
could hear but could not discriminate between scold-
ing and petting. It was impossible for it to get itself
out of any uncomfortable situation. The period of
heat was no longer noticeable. The effects are simi-
lar to those upon pigeons, with the difference that the
secondary effects of the operation on the remaining
parts of the central nervous system are greater in
dogs. The reasons for this may be purely technical
or anatomical, or may be due to a greater sensitive-
ness of the central nervous system in dogs. We may
mention in this connection that hemorrhages in the
human cerebral hemispheres are often accompanied
by a complete paralysis of the extremities, while this
is never the case in dogs.

The fact that in animals which normally possess no
memory, loss of the hemispheres occasions little dis-
turbance, and the fact that in animals possessing
memory, the latter disappears upon destruction of the
hemispheres, prove that the hemispheres are an essen-
tial organ for the phenomena of associative memory.

4. Pfluger expressed the opinion many years ago
that an animal that has lost its brain still possesses
consciousness (7). He drew this conclusion from the
reactions of decapitated animals. If the tail of a de-
capitated eel be rubbed gently on one side the tail
presses itself against the finger, but if touched with a
burning match it is turned away. From these and

CEREBRAL HEMISPHERES AND MEMORY 249

similar observations, which are doubtless correct,
Pfluger concluded that the spinal cord possesses con.
sciousness. Pfluger's statements gave rise to a lively-
discussion. His opponents could not refute his con-
clusions entirely, but they advanced arguments to show
that the spinal cord does not possess consciousness.
Goltz's ingenious experiments deserve special mention
in this connection (3). They show that the decapi-
tated frog is not able to rescue itself from an unpleas-
ant situation. A blinded but otherwise normal frog
and a frog without cerebral hemispheres were placed
together in a trough filled with water and the water
heated gradually. When the temperature of the water
rose, the blinded frog became restless, jumped about,
and attempted to escape from the trough. The frog
without cerebral hemispheres, on the other hand, re-
mained quiet and the heat rigor overcame it in the
attitude it assumed when put into the trough. This
of course speaks against the presence of consciousness
in the spinal cord. But since this did not directly
prove the erroneousness of Pfluger's conclusions, opin-
ions remained divided. I believe we are now in a
position to prove that Pfluger's observations not only
allow but demand an entirely different explanation, and
that it is wrong to make them a criterion for the exist-
ence of consciousness. The experiment with the tail
of the eel is a case of tropism. The eel is positively
stereotropic. It is forced to bring every part of its
body as far as possible in contact with solid bodies,
like Nereis, many insects, the stolons of Hydroids,

250 COMPARATIVE PHYSIOLOGY OF THE BRAIN

and the roots of many plants. It lives chiefly in
cracks. This is no more a process of consciousness
than the boring of a root in the sand. It exists in
every segment of the eel, and if touched on one side
with the finger positively stereotropic curvations
toward the finger ensue. The stimulus of rubbing
increases the tension of the muscles on the stimulated
side. But while it is positively stereotropic it is not
positively thermotropic. If a burning object is applied
it produces a relaxation of those muscles which move
the body toward the stimulated side.^ The body is
thus moved toward the opposite side. In this case
too consciousness plays no more part than it does in
the tropic reactions of a plant. The whole discussion
of the *' spinal-cord-soul " was needless, and might have
been avoided if Pfliiger had realised that those phe-
nomena which the metaphysician calls consciousness
are a function of the mechanism of associative mem-
ory. In that case the question would have been : —
Does the decapitated animal still possess associative
memory, or are its reactions all due to inherited struct-
ures and irritabilities ? With the aid of comparative
physiology it would have been found that all the reac-
tions of such an animal may occur in forms which
possess no associative memory. The mechanisms
which allow an associative memory in Vertebrates
seem to be located in the cerebral hemispheres. In

' If Pfliiger had made his experiments on decapitated snakes he would have
obtained different results. Exner mentions that such animals press their body
against a fiery coal just as well as against the finger (13).

CEREBRAL HEMISPHERES AND MEMORY 251

Invertebrates they will probably be found in the
supraoesophageal ganglion.

5. The spinal-cord-soul is not the only instance in
which biologists have been led astray by their blind
acceptance of metaphysical notions. A second and
perhaps more general instance is the assumption that
consciousness exists in every animal and is present to
a certain degree even in the ^^^. Many authors ob-
ject to the idea that a thing like consciousness or the
soul should get into the body suddenly at a certain
stage of development. What they consider true for
the ontogenetic development they also assume for the
phylogenetic development, and they are led to believe
that each animal possesses consciousness. All these
speculations collapse as soon as we free ourselves
from the influence of metaphysics, and realise that
the term consciousness or soul is applied by meta-
physicians to phenomena of associative memory, and
that the latter depends upon a physical mechanism
which must be just as definite as, for example, the
dioptrical apparatus of our eye. I do not think that
anybody maintains that every animal must have an
apparatus which unites the rays of light emanating
from a luminous point to an image point on the sur-
face of its body, simply because certain animals have
such an apparatus. Moreover, I do not believe that
even our biological metaphysicians assume that this
dioptrical apparatus exists already in the human ^<gg,
and that the latter is already capable of visual space-
perception, because it would be too awkward to as-

252 COMPARATIVE PHYSIOLOGY OF THE BRAIN

sume that visual space-perception should begin at a
definite period of embryonic or post-embryonic de-
velopment. And yet the matter is in no way different
for psychic phenomena, if we realise that what we call
psychic is only a metaphysical term for functions of
associative memory. Just as our dioptrical apparatus
can only begin to function after the eye has reached
a certain stage of development, the mechanism of as-
sociative memory can only begin to function after the
brain has reached a certain stage of development.
And just as only certain animals are provided with
apparatus for visual space-perception, only certain ani-
mals are provided with the mechanism necessary for
associative memory.^ I think it is time for us to re-
alise that some of the phenomena of embryonic de-
velopment are not continuous processes but decidedly
discontinuous. This is, of course, less obvious if we
limit our study of the organism to methods of stain-
ing and sectioning, but it becomes very striking if
we add some physiological methods. A pure \ n
NaCl solution is extremely poisonous to the eggs of
Fundulus during the first twelve hours. After that
it is decidedly less harmful. There is then a discon-
tinuity in the physical or chemical conditions of that
embryo at about twelve hours after fertilisation. An-
other discontinuity is connected with the beginning
of circulation. Before circulation begins a f n KCl
solution is no more harmful to the embryo of Fundulus

^ These considerations dispose also of the conception of consciousness in
plants or of the barbarous notion of consciousness in molecules and atoms.

CEREBRAL HEMISPHERES AND MEMORY 253

than a f n NaCl solution. As soon as the heart be-
gins to beat, the KCl becomes much more poisonous
than the NaCl solution. A similar discontinuity is
noticed if we try the effects of lack of oxygen. As
soon as the circulation begins, the Fundulus embryo
becomes quite suddenly much more sensitive to a lack
of oxygen. The functional changes in the embryo
itself are sudden and not gradual or continuous. The
heart-beat, for example, starts at a certain time, sud-
denly, after a certain stage of development has been
reached.

The idea of a steady^ continuous development is in-
consistent with the general physical qualities of proto-
plasm or colloidal material. The colloidal substances
in our protoplasm possess critical points. If we in-
crease the pressure of a gas below a certain tempera-
ture, at a certain critical point the gas becomes
liquid. The colloids change their state very easily,
and a number of conditions — temperature, ions, en-
zymes— are able to bring about a change in their state.
Such material lends itself very readily to a discon-
tinuous series of changes, while a gradual steady
development, such as most Darwinians assume, is
practically excluded.

We, of course, concede that the associative memory
shows different degrees of development or perfection
in different animals. These different degrees are
mainly differences in capacity and resonance. By
difference in capacity I mean a difference in the
number of associations of which the brain is capable.

254 COMPARATIVE PHYSIOLOGY OF THE BRAIN

By difference in the resonance I mean the ease with
which associations are produced. It is necessary, for
example, in the case of a great complex of sensations,
that the images of memory which correspond to cer-
tain constituents of that complex are easily repro-
duced, and in the case of a very elementary sensation
greater images of memory, which contain that ele-
mentary sensation as a constituent, should be repro-
duced. The quality of resonance is perhaps the more
important, as long as the capacity does not fall below
the average. The intelligent man differs from the
stupid man, among other things, in the ease with which
by means of the associative memory he makes the
analysis or synthesis of the complexes of sensation :
that is, in the slow or stupid man only such images of
memory are called up associatively as were con-
nected before with the entire stimulating complex ;
while in the quick thinker complexes of memory are
also produced associatively which are connected with
single elements of the stimulating complex.

6. After what has been said, it is clear that the ab-
solute mass of the brain cannot be the principal factor
in determining intelligence (lo). In different races
of dogs, for instance, the brain varies just as much as
the weight of the body. Dogs of a small breed may,
however, be more intelligent than dogs of a large
breed. It also follows from this that the relation of
mental activity to the metabolism of the central
nervous system is totally different from that of mus-
cular activity to the metabolism of the muscle. The

CEREBRAL HEMISPHERES AND MEMORY 255

power-rate of activity of the muscles is proportional
to their mass, and something similar may be true of
the glands. The results obtained by weighing the
brain of man have proved, conclusively, that the mass
of the cerebrum, unless it falls below a certain mini-
mum, in no way affects the degree of intelligence.
The same facts prove that the number of ganglion-
cells bears no direct relation to the degree of intelli-
gence. The small dog has fewer ganglion:cells than
the large dog, inasmuch as the size of the cells varies
comparatively little in dogs of different size.

Speck, who has called attention to this difference
between muscles and brain (8), has also made another
important discovery, namely, that in case of lack of
oxygen associative memory first disappears. He in-
haled air deficient in oxygen from a gasometer, and
counted during his experiments. As soon as the
partial pressure of the oxygen of the air fell below
8 ^ of one atmosphere, he forgot to count very soon
and then fainted, although the other functions of his
body showed no change. Speck concludes from this
that the cerebral hemispheres are most sensitive to a
lack of oxygen. It is not absolutely necessary to
conclude from this that the cerebral hemispheres
have relatively the greatest metabolism of all the
organs. It is possible that lack of oxygen affects the
physical qualities of colloids in the brain in such a
way as to make the functioning of the mechanisms of
associative memory impossible. I have shown that
lack of oxygen leads to a liquefaction of the cell-walls

256 COMPARATIVE PHYSIOLOGY OF THE BRAIN

in certain forms, and it seems to be pretty generally
true that the formation of solid cell-walls becomes
impossible under such conditions (n). It is possible
that in the case of lack of oxygen, physical changes
in the state of certain constituents of the brain are
prevented which are necessary for the activity of
memory.

Some physiologists seem to be of the opinion that
when the brain contains a good deal of blood the
body has a special feeling of happiness. I recall a
popular lecture by a prominent psychiatrist in which
he maintains that when the cerebral hemispheres con-
tain a great deal of blood the proprietor of this brain
enjoys the absolute happiness (?) of an intoxication
from champagne. This psychiatrist evidently ima-
gines that the greater the supply of blood is, the
better the brain is nourished, and that with the in-
creasing nourishment of the brain the feeling of
happiness increases. Among the food-substances
which are offered to the brain in large quantities by
the dilatation of the arteries oxygen takes the first
place. It was formerly assumed that the oxygen-
supply determined the metabolism, but we now know
definitely that internal processes in the tissues
determine the consumption of oxygen, probably
processes of fermentation. If a certain quantity of
oxygen is present in the brain, the superfluous oxygen
has no effect. The same is probably true of all the
other food-constituents. Under normal conditions
the oxygen-supply in the brain is sufificient as long as

CEREBRAL HEMISPHERES AND MEMORY 257

the circulation is normal. It harmonises with these
facts that mental activity does not influence the
phenomena of oxidation, as Speck has proved by very
careful experiments. But from this we must not con-
clude that the activity of the brain takes place with-
out chemical changes, only that the chemical changes
which are determined by mental activity are too
slight to be recognised. The statement that dilatation
of the blood-vessels of the brain produces a sensation
of happiness is not based upon any fact that has
been proved scientifically.

7. The amoeboid changes in the ganglion-cells have
been utilised to account for the phenomena of asso-
ciation. As far as normal processes of association
are concerned, these amoeboid changes cannot play
any r6le, as they are much too slow. We notice
migrations of the cones and the pigment in the
retina, yet the idea that these protoplasmic motions
play any rdle for space- or colour-perception has to
be abandoned for the same reason.

Other authors hold that conditions of incomplete
association, as in the case of dreams, or interruption
of association, as in the case of deep sleep or
narcotics, are due to a partial or complete discon-
nection of the ganglion-cells by a shortening of the
processes. It does not seem to me that the obser-
vations which we thus far possess prove anything of
that character (9, 12).

258 COMPARATIVE PHYSIOLOGY OF THE BRAIN

Bibliography.

1. SCHRADER, Max E. G. Zur Physiologie des Froschhirns.
P finger's Archiv, Bd. xli., 1887.

2. ScHRADER. Die Stellung des Grosshirns im Reflextnechan-
ismus. Archiv fur experiment. Pathologie und Pharmakologiey
Bd. xxix., 1892.

3. GoLTZ, F. Beitrdge zur Lehre von den Nervencentren des
Frosches. Berlin, 1868.

4. ScHRADER, Max E. G. Zur Physiologie des Vogelgehirns.
Pfiugers Archiv, Bd. xliv., 1889.

5. Steiner, J. Die Funciionen des Centralnervensy stems und
ihre Phylogenese. II. Abth. : Die Fische. Braunschweig, 1885.

6. GoLTZ. F. Der Hund ohne Grosshirn. Pfliiger's Archiv ,
Bd. li., 1892.

7. Pfluger, E. Die sensorischen Functionen des Rilckenmarks.
Berlin, 1853.

8. Speck. Physiologie des menschlichen Athmens. Leipzig,
1892.

9. Duval, M. Thiorie histologique du sommeil. C. R. Soc. de
Biol., 1895.

10. Donaldson, H. H. The Growth of the Brain. London,

1895-

11. LoEB, J. Untersuchungen Uber die physiologischen Wir-
kungen des Sauer staff mangels. Pfiiiger's Archiv, vol. Ixii., 1895.

12. Bawden, H. H. a Digest and a Ci'iticism of the Data
upon which is Based the Theory of Amoeboid Movements of the
Neuron. The Journal of Comparative Neurology, vol. x., 1900.

13. ExNER, S. Entwurf zu einer physiologischen Erkldrung der
psychischen Erscheinungen, Leipzig and Wien, 1894, p. 85.

CHAPTER XVII

ANATOMICAL AND PSYCHIC LOCALISATION

I. It follows from the facts of the preceding chap-
ter that the cerebral hemispheres are a necessary
organ for the phenomena of associative memory.
We are not quite justified in saying that they are the
specific organ for this function. It may be possible,
although not probable, that other parts of the brain
are also required for this purpose. It is certain that
the spinal cord is not needed for this function, for
animals whose spinal cord is severed, or from whom
the greater part of it has been removed, show no de-
ficiency in the process of associative memory.

The cerebral hemispheres form an appendage of
the segmental central nervous system. They are con-
nected with at least some of the segmental ganglia by
special nerve-fibres. As these different bundles of
fibres enter the cortex at different places, it is obvious
that if we stimulate the various spots of the surface of
the cortex with electric currents of the smallest in-
tensity necessary to produce a reaction, we must
notice different effects. If, for instance, a current of
minimal intensity be sent through the spot D (Fig.

259

26o COMPARATIVE PHYSIOLOGY OF THE BRAIN

39), where the fibres from and to the brachial segment
of the cord of a dog enter the cortex, contractions of

certain muscles of the fore-
leg must follow. If we
stimulate the region A (Fig.
39), which is connected with
the sensory or motor ganglia
of the eyes, motions of the
latter must be produced. It
is, moreover, evident that if
we injure the spot D in the
cortex we must get some-
what different after-effects
from those produced when
A is injured. In the former
case we must expect motor
disturbances in the use of

Fig. 39. Cerebral Hemispheres the fore-leg, in the latter dis-
OF A Dog. ^ , r • •

A, optical region; D, brachial region; tUrbaUCeS of VlSlOn.
G, region of the hind-leg. (After J|- jg q£ cOUrSe, nOt tO

Munk.) ' ' ^ ^

be expected that the distri-
bution of segmental fibres on the cortex follows min-
utely the arrangement of the ganglia in the spinal
cord. Displacements of elements occur in the cerebral
hemispheres during the process of growth. This is
indicated by the formation of folds formed on the
surface. It is possible that not all the segmental
ganglia send fibres directly to the hemispheres, and it
is possible that certain ganglia are connected with the
cortex at more than one spot or region. From the

ANATOMICAL AND PSYCHIC LOCALISATION 261

fact that the different bundles of fibres from the vari-
ous segmental ganglia enter at different spots in the
cortex, some authors have drawn the conclusion that
there is not only an anatomical localisation of fibres ^
but also a psychic localisation of functions. They
assume that the various psychic functions take place
in different regions of the cortex. The occipital
region, where the fibres from the segmental ganglia
of the optic nerve enter, is considered by these authors
as the seat of visual consciousness. At the spot D
(where the brachial fibres enter or leave) the " con-
sciousness of the fore-leg" is said to be localised.
These assumptions are contradicted by the plain facts
of associative memory. Simultaneous processes in
different sense-organs are fused in our memory. The
odour of a rose recalls its visual image. This could
not be possible if the visual processes were confined
to one region of the cerebral hemispheres ; they must
spread to the olfactory region, and vice versa. The
same can be said of other kinds of stimulation and of
combinations of more than two stimuli. Moreover,
we remember not only simultaneous sense-impressions,
but we remember a whole series dependent upon suc-
cessive stimuli of different character, if only the first
constituent of such a series has been aroused. This
indicates that even the after-effects of a stimulus must
spread all over the cerebral hemispheres, so that they
may fuse with the successive processes going on in
the brain. It is thus obvious that the assumption of
a localisation of psychic functions in the cortex is

262 COMPARATIVE PHYSIOLOGY OF THE BRAIN

opposed to the elementary facts of associative memory
or consciousness.

2. Experiments on the brain indicate that while
there exists to a certain extent an anatomical localisa-
tion in the cortex, the assumption of a psychical
localisation is contradicted by the facts. The occipi-
tal region of the cerebral hemispheres is said to be
the seat of visual processes, the temporal lobes the
seat of auditory processes. If the occipital regions
are removed, only the visual processes are said to
cease, while if the temporal regions are removed, only
the auditory processes are said to disappear. We
know that persons who were born blind and deaf have
shown a normal or even a superior intellect (Laura
Bridgman). If the theory of psychic localisation
were correct, we should expect that an animal from
whose hemispheres the occipital and temporal regions
are removed would become blind and deaf, but would
remain normal in other directions. But Goltz has
shown that such an animal (dog) becomes hopelessly
idiotic (2, v.). The processes of association even of the
other senses are no longer normal. This agrees with
the idea that in processes of association the cerebral
hemispheres act as a whole, and not as a mosaic of a
number of independent parts.

Goltz has proved that if we remove one whole
hemisphere in a dog the personality of the animal or,
in other words, the sum-total of its associations remains
the same. The dog recognises its friends and all the
other objects it has ever known, and it reacts in such

ANATOMICAL AND PSYCHIC LOCALISATION 263

a way as to indicate that its associative memory has
not suffered through the operation. But if the an-
terior parts of both hemispheres be removed, the dog
is no longer normal, but idiotic. It no longer reacts
in the same way it did before, and it is obvious that
its associative memory has suffered. The same is true
if both posterior halves of the cerebral hemispheres be
removed (2, V.).

If we ask at present what determines this difference,
we are at a loss to give an answer. We might point
out that the right and left hemispheres are practically
symmetrical, while the anterior and posterior parts
are not symmetrical. If the form or orientation of
the elements be of importance, we might conceive of
the possibility that in a brain with only one cerebral
hemisphere all the processes could occur in approxi-
mately the same form, while in the brain with both
posterior or both anterior halves of the hemispheres
gone, the processes of association could not be re-
peated in the same, but in a mutilated form. Hence
the idiocy which follows such operations. We might
illustrate this by an analogous experience in the phy-
siology of sound. Each vowel is determined by a
sound of a certain pitch. If a singer sings in a pitch
higher than that of the determinant, the vowel becomes
indistinct. It is possible that in the brains of the
above-mentioned dogs the associations are rendered
impossible or difficult, because certain elemental pro-
cesses are no longer possible.

3. In this connection I may mention that the bo-

264 COMPARATIVE PHYSIOLOGY OF THE BRAIN

servations of Goltz indicate a connection of the main
regions of the cerebral hemispheres with certain
regions in the medulla oblongata. A dog that has
lost the anterior halves of both cerebral hemispheres
has a tendency to run with its head bent down. A
dog which has lost the posterior halves of both hemi-
spheres shows the opposite tendency. It moves very
little and its head is carried high in the air. Its an-
terior legs are stiff and often stretched forward. The
difference in the position and progressive motions of
these two animals seems to be somewhat similar to the
difference in the attitude of an Amblystoma when
stimulated by constant currents. The position of
the dog in which the anterior halves of the cerebral
hemispheres are removed resembles that of an Am-
blystoma in a descending current, while the attitude
of the dog without the occipital halves of the hemi-
spheres is like that of an Amblystoma in an ascending
current. (See Chapter XI.) If in a dog one cerebral
hemisphere be removed, while the other is intact, the
dog makes circus-motions toward the injured side.
There is an unmistakable analogy between these ob-
servations and the older experiments of Magendie
and Flourens on the sectioning of the crura cerebelli.
While dogs after the loss of the anterior halves of
the cerebral hemispheres often become irritable and
ugly, dogs which lose the occipital halves of both
hemispheres invariably become good-natured and
harmless. This indicates a connection of the cerebral
hemispheres with organs of the body for which with

ANATOMICAL AND PSYCHIC LOCALISATION 265

our present knowledge of anatomical localisation we
cannot yet account.

4. Those who believe in a psychic localisation in
the cerebral hemispheres base their claims chiefly on
the effects of small lesions. If our point of view is
correct, we should expect that small lesions either
make no noticeable functional disturbance at all, or
cause disturbances which are no more psychic than
those following the cutting of a peripheral nerve.
Hitzig and Fritsch were the first to destroy the cortex
of the centre of the fore-leg (D, Fig. 39) in one of
the hemispheres of a dog (i). When this centre was
destroyed in the left hemisphere, the right leg showed
the following disturbance : ** In running, the animals
did not use the right fore-paw to advantage. It was
turned in or out too much and did not furnish a proper
support. This never happened with the other paws.
Movement did not fail entirely, but in the right leg the
movement of adduction was somewhat weaker. In
standing, the dorsal side of the paw was often used in-
stead of the sole " (i)- ^^ ^he paw was placed in abnor-
mal positions, no attention was paid to it by the dog.
Hitzig and Fritsch draw the following conclusions from
these observations : " The animals evidently had only
an imperfect consciotisness of the condition of this limb ;
ihey had lost the ability to form perfect ideas concern-
ing ity In the opinion of Hitzig we have to deal
with a psychical disturbance, or, as we should say, a
disturbance of associative memory. This disturbance
of associative memory is, however, confined to such

266 COMPARATIVE PHYSIOLOGY OF THE BRAIN

processes as involve the right fore-leg. In our opinion,
the phenomena observed by Hitzig are the outcome of
a weakening of certain groups of muscles and a diminu-
tion of the sensibility in the right leg. Such disturb-
ances could just as well be produced by a pressure
upon certain peripheral nerve-fibres.

That Hitzig's psychological interpretation of his
observation is wrong has been proved by Goltz. If
Hitzig's idea were correct, we ought to assume that, if
the centre of the right fore-leg were removed, a dog
should no longer be able to use the right paw as a
hand, where such a use is based upon the activity of
associative memory. Goltz not only removed the
centre but the entire left hemisphere of a dog that
had been taught to dig its food out of a heap of
pebbles. This dog showed all the disturbances of
the right leg which Hitzig described. Yet it con-
tinued to dig its food (pieces of meat) out of the
heap of pebbles with the right fore-paw. It preferred
to use the left paw for this purpose. But when this
was forbidden it used the right paw with success.
This experiment proves that the conscious or psy-
chical character of the motions of the fore-leg is not
affected by the removal of its cortical centre. A close
observation of the way the dog uses this paw shows
that certain muscle-groups must have suffered by
the operation, and a closer analysis of these purely
muscular disturbances explains the anomalies which
Hitzig had mistaken to be of a psychical character.
Removal of the fore-leg centre causes a decrease in

ANA TOMICAL AND PSYCHIC LOCALISA TION 267

the tension of the extensors of the leg (and perhaps
also of other groups of muscles). For this reason
the leg slips easily and bends in the ankle-joint so that
the animal sometimes steps on the back instead of the
sole of the foot. It does not notice if the leg is
placed in an abnormal position. This is partly due to
the diminution in resistance caused by the weakening
of certain muscles and partly due to a reduction in the
sensibility of the skin. Goltz has proved that it re-
quires a greater pressure on the skin of this leg to
cause the dog to withdraw it than on any of the other
legs. This explains also why the dog does not notice
if the foot whose cortical centre has been removed is
placed in cold water. There are, then, changes in the
tension of certain muscles and a reduction in the
sensibility of the skin which suffice to explain all
the disturbances observed by Hitzig, but there is no
loss of muscular ''consciousness" as Hitzig assumes.
To a certain extent, similar effects can be produced by
dividing the posterior roots of the arm-nerves. It
would hardly occur to anyone to maintain for this
reason that the psychic centre of the arm-movements
is localised in the posterior roots. Further proof that
these disturbances described by Hitzig are due to a
decrease in the tension of the extensors is furnished
by the fact that in man, when an arm becomes para-
lysed after a local disease in the cerebral hemispheres,
a contraction producing a flexed position of the arm
takes place after a time. Not all the muscles of the
arm are completely paralysed as a result of the disease

268 COMPARATIVE PHYSIOLOGY OF THE BRAIN

in the hemispheres ; but the tension of the extensors
has decreased, and as a result the tension of the flexors
alone determines the position of the arm.

In Goltz's experiment the centre of the right fore-
leg alone had been removed. It has been said that
the centre of the left fore-leg situated in the other
hemisphere performed the psychic functions for both
legs after the operation. I made an experiment to
which this objection is not possible. A dog was
taught to walk on its hind-legs when it wanted to be
fed. Then the hind-leg centres were removed (G, Fig.
39) in both hemispheres. In spite of this loss the
dog was still able to walk on its hind-legs. When-
ever I offered it food or whenever it expected to be
fed it rose voluntarily on its hind-feet. The conscious
actions or associatio7ts for the use of the hind-legs had
not suffered, but there was decidedly a muscular dis-
turbance inasmuch as the dog was not able to stand
so long on its hind-legs as it could before the opera-
tion. I showed this dog at the naturalists' meeting in
Berlin in 1886. The day after the demonstration I
showed the brain of the animal that had been killed
in the meantime. The hind-leg centres had been
removed completely.

It must, however, be explicitly stated that not every
limited lesion in the motor centres leads to a disturb-
ance. This is not only of importance from a theoretical
but also from a practical point of view. A physician
need not be surprised if a post-mortem examination
shows a circumscribed lesion in the cortex which had

ANATOMICAL AND PSYCHIC LOCALISATION 269

not caused any clinical symptoms. It is obvious that
certain organs are more easily disturbed by a lesion
in the cortex than others. An operation in the centre
of the fore-leg produces disturbances more easily than
an operation in the centre of the hind-leg. There
are certain parts of the body in which no disturbances
can be produced by the extirpation of their so-called
centres in the cortex. Nobody has thus far been able
to produce a paresis or paralysis of the upper eyelid
in a dog or to produce loss of sensibility in the cornea
by an operation in the cortex. It must, moreover,
not be overlooked that all the disturbances which
follow small lesions of the cortex in dogs are only
transitory.

5. Not only the motor but also the sensory disturb-
ances which follow an operation in the cerebral
hemispheres have been interpreted as psychic dis-
turbances. We know that a lesion of the surface of
the occipital lobes causes visual disturbances. Munk
has interpreted these disturbances which follow a
small lesion of one of the visual spheres as psychic
(4). There is a small region (Aj Fig. 39) in each of
the occipital lobes the destruction of which, according
to Munk, causes a psychic blindness in the opposite
eye. By psychic blindness Munk means the fact that
the dog does not recognise what it sees, although it is
by no means blind. If the cortex be removed at the
region A, in the left hemisphere, the dog shows psychic
blindness in the right eye. Such a dog, for instance,
is no longer afraid of a burning match or of the whip.

270 COMPARATIVE PHYSIOLOGY OF THE BRAIN

provided its left eye be closed. Munk assumes that
the image of memory of the whip or the burning
match had been deposited in the region Aj and was
lost with the loss of this place. It can be shown that
Munk is as much mistaken concerning the psychic
character of the visual disturbance following the de-
struction of a small region in the occipital lobes, as
Hitzig was in regard to the psychic character of the
motor disturbances following the destruction of a
centre of the fore-leg. In the majority of cases the
removal of the region A ^ in one hemisphere produces
no visual disturbance. In the cases where a visual
disturbance is produced it is only temporary. I
noticed indeed that such dogs may no longer recog-
nise objects in the opposite eye, but the reason for
this is altogether different from that assumed by
Munk.

It is known that in man the destruction of the
visual sphere in one hemisphere causes the same dis-
turbance as the destruction of the optic tract of the
same side — namely, a hemianopia of the opposite half
of the visual field. This disturbance is not psychic
but purely physiological, inasmuch as it results in
a loss of irritability on one side of each retina, but not
in a loss in the processes of association. The same
occurs in a dog whose visual sphere has been injured
in one spot, with this difference, however, that the
loss of irritability is not complete. Thus if the left
occipital region be injured in a man, a hemianopia of
the left sides of both retinae follows, and the patient

ANATOMICAL AND PSYCHIC LOCALISATION 271

sees nothing in the right half of his visual field. If
the same operation be performed in a dog, it causes
not a complete hemianopia but a hemiamblyopia (5).
The dog is not blind for the right half of its visual
field, but has only a reduced power of vision. It be-
haves like an animal that pays less attention to that
half of its visual field, or whose threshold for this half
is reduced. If we stand before such a dog and hold
two pieces of meat in front of it, simultaneously, one
piece in each hand, the dog invariably chooses the
piece at its left. It almost seems as though it did
not see the piece at the right. Now we know that a
moving object acts as a stronger optical stimulus than
a stationary object. If the two pieces of meat are
again held before the dog in the manner described
above, only with the difference that the piece that is
in the right half of the field of vision is moved, the dog
jumps at the latter (6). This proves that in the dog
the threshold of stimulation for optical stimuli has
been raised in the right half of the field of vision.
But how could Munk mistake the hemiamblyopia for
a psychic disturbance? In a dog, the divergence of
the optical axes is greater than in man. Hence the
right half of the visual field is controlled more by the
right eye than by the left. If we produce the hemi-
amblyopia or the hemianopia in a dog, the eye oppo-
site the injured hemisphere is blind or injured for
considerably more than one half of its retina. If the
other eye of such a dog be closed, its field of vision is
reduced to a very small area, and the dog does not

272 COMPARATIVE PHYSIOLOGY OF THE BRAIN

recognise the objects, although It is not entirely blind.
What Munk mistook for psychic blindness is, in real-
ity, only hemiamblyopia or hemianopia (5, 6).

6. One of the main arguments which Munk used
for his assumption of a psychic character of the visual
disturbances caused by the effects of a unilateral lesion
of a visual sphere was the fact that these disturbances
disappear in about six weeks. On the basis of this
fact he constructed the following hypothesis : The
visual images of memory are deposited each in a
single ganglion-cell or a group of cells in the region
A^ of the opposite hemisphere. If this region be re-
moved, the dog loses all its images of memory. But
new images of memory can be deposited in the sur-
rounding parts A. This will be done after the loss of
the region Aj and the dog becomes normal again after
six weeks. If this hypothesis of Munk were correct,
visual disturbances of such a dog should not disap-
pear if it were kept in the dark, where it would have
no chance to acquire new visual images of memory.
I made that experiment. In dogs which possessed
only the right eye, the region Aj in the left cerebral
hemisphere was destroyed. In the majority of these
dogs, the operation produced no effect. In a few,
hemiamblyopia occurred. Of these several were put
in an absolutely dark room for the following six weeks.
As soon as they were taken out they were entirely
normal. This proves that their recovery was not due
to the acquisition of new visual images of memory,
but to the fact that a purely physiological effect upon

ANATOMICAL AND PSYCHIC LOCALISATION 273

the irritability of the optical apparatus caused by the
operation wears off after a certain time.

We then come to the conclusion that the apparent
psychic blindness which follows the destruction of the
region Aj in the opposite hemisphere is exclusively a
hemianopia or hemiamblyopia. This disturbance is
no psychic disturbance inasmuch as it can be pro-
duced by an injury to a peripheral nerve, the optic
tract.

7. We must seek an explanation for the temporary
character of the disturbance which follows small
lesions. If the lesion covers a large area, the dis-
turbance is more permanent. Goltz assumes that
these transitory effects are shock-effects due to the
operation. He was led to this assumption through
his experiments on the spinal cord. If the spinal cord
be cut in a dog, no segmental reflexes occur during
the first days or weeks after the operation in the part
of the animal below the cut. " Pressing on the hind-
feet produces no reaction. In the male dog, erection
of the penis cannot be aroused reflexly. The urine
collects in the relaxed bladder. The anus gaps. In
brief, the whole posterior part of the body seems un-
irritable. A few days later the apparently dead spi-
nal cord may have recovered almost entirely. The
posterior part of the animal then offers a large num-
ber of reflex phenomena. No one will assume that
that piece of the spinal cord which is separated from
the brain in so short a time acquires entirely new

powers as a reflex organ ; we must assume that these
18

274 COMPARATIVE PHYSIOLOGY OF THE BRAIN

powers were only suppressed or inhibited temporarily
by the lesion of the spinal cord." The same is true
in regard to the vasomotors. Division of the spinal
cord reduces the tonus of the blood-vessels of the
posterior legs. After a time the blood-vessels recover
and become normal again. Now if the sciatic nerve
in the same animal be severed, a new temporary para-
lysis of the vasomotors follows. This proves that
the vasomotor paralysis in the hind-legs that occurs
after the division of the spinal cord is due to a
shock-effect of the operation. What the nature of
this shock-effect is we do not know. Perhaps v.
Cyon's experiment throws light on this : v. Cyon
showed, namely, that the tension of the muscles de-
creases after the division of the posterior root of their
segment (3).

8. We conclude from all these observations on dogs
that small lesions do not cause any disturbances in the
processes of associative memory, and that Hitzig and
Munk are wrong in interpreting the disturbances fol-
lowing the excision of a small piece of the cortex as
psychic disturbances. In the majority of cases such
slight lesions cause no disturbance, and where any
is caused it is of such a character as could be pro-
duced by the lesion of a peripheral nerve. If we wish
to produce psychic disturbances by a lesion of the
brain, we must destroy extensive parts of both hemi-
spheres. Operations in one hemisphere alone, and
even the destruction of an entire hemisphere, have no
such effect.

ANATOMICAL AND PSYCHIC LOCALISATION 275

It has been claimed that the intellect is the func-
tion of special parts of the brain. Hitzig and others
assumed that the frontal lobes of the cerebral hemi-
spheres are the organs of attention. I have repeat-
edly removed both frontal lobes in dogs (6). It was
impossible to notice the slightest difference in the
mental functions of the dog. There is perhaps no
operation which is so harmless for a dog as the re-
moval of the frontal lobes. Flechsig thinks that it is
not only the frontal lobe but the cortex of certain
other regions which is responsible for mental activity,
inasmuch as it is the seat of "centres of association."
I have removed the cortex of Flechsig's ** centres of
association " in dogs without having noticed anything
that justifies Flechsig's hypothesis. The assumption
of " centres of association" is just as erroneous as the
assumption of a centre of coordination in the heart.
Association is, like coordination, a dynamical effect
determined by the conductivity of the protoplasm.
Associative processes occur everywhere in the hemi-
spheres (and possibly in other parts of the brain), just
as coordination occurs wherever the connection be-
tween two protoplasmic pieces is sufficient. It is just
as anthropomorphic to invent special centres of associ-
ation as it is to invent special centres of coordination.

Bibliography.

1. Hitzig, E. Uniersuchungen ilber das Gehirtiy Berlin, 1874 ;
and Reicherfs und Du Bois-Reymond's ArchiVy 1870.

2. GoLTZ, F. Ueber die Verrichtungen des Grosshirns.

((

{(

xiv., 1877.

XX., 1879.

xxvi., 1881.

xxxiv., 1884.

276 COMPARATIVE PHYSIOLOGY OF THE BRAIN

I. Abhandlung P finger's ArchiVy Bd. xiii., 1876.

II.
III.
IV.

3. V. Cyon, E. Gesammelte Physiologische Arbeitetiy p. 197 a. f.
Berlin, 1888.

4. MuNK, H. Ueber die Functionen der Grosshirnrinde. Ber-
lin, 1881.

5. LoEB, J. Die Sehstorungen nach Verletzungen der Gross-
hirnrinde. P finger's Archiv, Bd. xxxiv., 1884.

6. LoEB, J. Beitrdge zur Physiologie des Gross hirns. PfiUger's
ArchiVy Bd. xxxix., 1886.

CHAPTER XVIII

DISTURBANCES OF ASSOCIATIVE MEMORY

I. We have mentioned the hypothesis that each
image of memory is localised in a special ganglion-cell
or a group of ganglion-cells. As soon as a new image
of memory arrives, it is, according to this hypothesis,
deposited in one of the empty cells. Who deposits it
and who finds out which cell is empty and which oc-
cupied is a question the originators of such hypotheses
do not ask. This conception treats the image of
memory as if it were something substantial, u e.,
something characterised by mass.^ Munk has as-
serted the possibility of proving that in a dog the sin-
gle visual images of memory are localised in isolated
cells, or groups of cells, at the part A^ (Fig. 39). He
gives as proof two experiments, ** in which extirpation
of the part Aj caused the loss of all but one of the
visual images of memory. One single visual image

' This peculiar hybrid between metaphysics and anatomy owes its origin
largely to Gall. Gall was an industrious worker in the anatomy of the brain
and at the same time a huge fraud. The anatomy of the brain was not suffi-
ciently sensational for him, so he enlivened things somewhat by grafting upon
his anatomy the worst metaphysics he could possibly get hold of. The various
nooks and corners of the brain became the seat of soul-powers of his invention.
This artificial connection between metaphysics and brain-anatomy or histology
has since become traditional.

277

278 COMPARATIVE PHYSIOLOGY OF THE BRAIN

of memory in each case was found to be preserved and
unimpaired : in one case the image of the pail, out of
which the dog was accustomed to drink, remained ; in
the other, that of the motion of the hand, which be-
fore the operation had been the signal for the dog to
give its paw." It was this statement of Munk that
led me as a student to make experiments on the
brain. I hoped that a road to an exact psychology
had been opened. I began my experiments as a con-
firmed supporter of Munk. The more experiments
I made the more it became apparent that many of
Munk's statements were incorrect, especially his mea-
gre statements concerning the supposed localisation
of single images of memory. It is my opinion that
these histological or corpuscular hypotheses of the
images of memory must be supplanted by dynamical
conceptions. The dynamics of the process of associ-
ation is the true problem of brain-physiology. Even if
the hypotheses of psychic localisation were not contra-
dicted by all the facts, the pointing out of the centres
would not be a solution of the dynamical problem.
By merely showing a student the location of a power-
plant, we do not explain to him the dynamics of
electric motors.

I have mentioned above the possibility that pro-
cesses of association will become abnormal if certain
elemental constituents are mutilated or impossible.
I selected as an example our ability to recognise
a vowel. If the vowel is sung at a pitch which
excludes its specific formative sound, it becomes

DISTURBANCES OF MEMORY 279

indistinct. A study of patients afflicted with amnesia
seems to support this analogy. It is not possible to
use all the reports of such cases. I think that the
majority of practitioners have neither the training nor
the time to analyse them. I will confine myself to
two cases from the Clinic of Professor Rieger in
Wurzburg, one of which was analysed by himself (i),
and the other by his assistant, Dr. Wolff (2). In the
first case the patient had suffered a concussion of the
brain in a railroad accident. Among a number of
other disturbances, his memory showed peculiar gaps.
The patient was able to recognise only the numbers
I, 2, and 3. The corpuscular theory of the images of
memory would assume that all numbers which the pa-
tient had originally possessed had been located each
in a special cell, and that these cells had all perished
with the exception of the cells which contained the
first three numbers. This at once seems strange ,and
becomes still stranger when taken in connection with
the following observation. In every case it took the
patient some time to find the word one when the fig-
ure I was held before him. The reaction-time for
naming a 2 was considerably longer, and for naming a
3 was still longer. He was able to reckon with these
three numbers, but when a 3 occurred he required more
time than when a i or a 2 occurred. The determin-
ation of the reaction-time furnishes the explanation of
the fact that all numbers beyond 3 were wanting. All
of Rieger's experiments on this patient showed that if
he did not succeed in finding the name of an object

28o COMPARATIVE PHYSIOLOGY OF THE BRAIN

within a certain time (about eighteen seconds) it was
impossible for him to do so at all. Now for finding
the word three when he was shown the figure 3 he
required almost eighteen seconds, and in fact he even
failed occasionally to find it. The first three numbers
are the ones that a child first learns, and are also
those used most frequently during life. We know
that the words we use least are the ones most liable
to vanish from our memory (for instance, the vocabu-
lary of a foreign language). It is possible that in the
brain of the patient the processes were partly mu-
tilated or rendered more difficult. The numbers
used most frequently could cross the threshold ; those
used less frequently could not. This conception is
further confirmed by the fact that by touching the
edges the patient was able to distinguish a ten- from
a fifty-pfennig piece, although the numbers ten and
fifty were otherwise gone, and as stamped on the
coins were only hieroglyphics to him. The money-
conception of the ten- and fifty-pfennig piece had
formed more associations and clung more tenaciously
to the memory of this man, who had to struggle for
his existence, than the abstract conceptions ten and
fifty, which had existed in his head only as a scholas-
tic luxury. Hence any adequate idea of the nature of
the disease of this man must be a dynamic one. In
the injured brain of this patient certain processes
were able to take place as before, except that they
were less intense or incomplete. Those innervations
forming constituents of relatively many or important

DISTURBANCES OF MEMORY 281

associations were still possible, or occurred in a more
normal form, while other innervations became impos-
sible or were mutilated. In this case it would be just
as erroneous to assume that the single conceptions or
letters are all localised in single cells, and that the
corresponding cells in the patient had perished, as it
would be erroneous to conclude in a case of interfer-
ence of sounds that the source of vibration was
removed.^

2. The second case mentioned is still clearer (2).
The disturbance of associative memory was also
caused by an accident. When the patient was asked
the colour of the leaves of a tree, he was unable to
answer the question unless he was allowed to go to
the window and look at a tree. In this case he an-
swered correctly. As long as he could not see a tree
it was impossible for him to tell the colour of a leaf.
Pieces of green, red, and blue paper were put before
him, and he was asked which the leaves looked like,
but he was unable to tell. If asked whether the trees
were blue, he answered that this was possible. Only
when looking at a tree was he able to remember that
the leaves were green. When asked how many legs
a horse has, he went to the window and waited until a
horse passed by. This enabled him to find the word
four. Only in winter was he able to tell the colour of
snow. In summer he admitted the possibility that

' Conditions similar to those that existed in this patient can be artificially
produced by the dynamometer-experiments which will be described in the next
chapter.

282 COMPARATIVE PHYSIOLOGY OF THE BRAIN

snow is black. He was once asked the colour of the
blood. He opened a little pustule on his hand, and
as soon as a drop of blood came out he gave the
answer, red.

It is obvious from these facts that the patient un-
derstood every question and was sufficiently intelli-
gent to secure those impressions which allowed him
to answer the question. He could tell the colour of
sugar if allowed to look at it, but this did not help
him to tell whether or not sugar tastes bitter. In
order to do so he had to put the sugar into his mouth.
When a smooth piece of glass was shown to him
he could not tell that it was smooth until he had
touched it.

Two things are evident — first, the patient was not
able to remember any perceptible quality of an ob-
ject unless the object was under his immediate per-
ception ; and second, he remembered the various
qualities only if the specific senses for these qualities
were affected. In a normal being the word sugar or
the sight of sugar suffices to produce the association
of its sweet taste. In this patient only contact with
the tongue suggested the word sweet, although he
was intelligent enough to know how to arouse the
correct association.

The names of a great many objects may be sug-
gested to a normal person through any of several of
the senses. For example, we find the word violin if
we see the object as well as if we hear it played with-
out seeing it. The patient in this case was a violin

DISTURBANCES OF MEMORY 283

player, but it was necessary for him to see the instru-
ment before he could name it. When a key was put
into his pocket and he was allowed to touch it he
could not say what it was. He could, however, find
the word if he could see the key in the door. When
his hand was put to his ear, he could not tell what he
touched unless he looked at the doctor's ear. When
the doctor covered his own ear, the patient was un-
able to find the word. It is obvious that in his case
the visual perception was, on the whole, more effect-
ive than any other sensation. A sense-perception
was necessary to call forth the association of concrete
objects, and of the many possible sense-perceptions,
which in normal cases might have brought about the
word, the strongest alone in him sufficed. The word
umbrella was only suggested when the umbrella was
opened. From this we might imagine that a change
in the machinery of association had taken place, which
allowed only the processes having a maximal intens-
ity or amplitude to arouse an associative process, the
others remaining without any effect.

The process was the same in regard to abstract as-
sociations. The patient complained that his annuity
was too small. He remonstrated against the doctor's
insinuation that he had murdered his wife or that he
was a scamp. But whether a beggar is a wealthy
or a poor man, or whether God lives in hell or in
heaven, were problems which he was not able to de-
cide, although he was a believer.

There was another peculiarity in his mechanism of

284 COMPARATIVE PHYSIOLOGY OF THE BRAIN

association which is in line with the necessity for
sense-impressions for the remembrance of words.
Before he was able to pronounce a word he had to
go through the motion of writing it. When asked
the colour of the leaves, he had to go to the window
and look at a tree, and then he had to go through
the motion of writing the word green with his finger
before he could give the correct answer. When not
allowed to use his fingers for this purpose, he used his
toes, and when this was forbidden, he made the writing
motions in his mouth with his tongue. When all
three motions were forbidden, he was not able to find
the word.^ He did not write phonetically, but ortho-
graphically. It would be absurd to think for a mo-
ment that in this case one single centre, or one single
tract between two centres, was injured. The whole
apparatus was equally affected. I believe that the
associative mechanism of the patient differed only
in degree from the associative mechanism of a
normal being. Wolff pointed out that for each act
of remembering there is one association more power-
ful than the rest. But for a normal being the weaker
associations are sufficient for the reproduction, while
in our patient only the strongest one sufificed. One
may ask how it happens that we so seldom hear of
such simple, clear cases as those published by Rieger
and Wolff. I believe the majority of physicians who
deal with such patients have neither the scientific

^ It was not necessary for him to see what he wrote or to actually write ; it
was sufi&cient to go through the motions of writing.

DISTURBANCES OF MEMORY 285

training nor sufficient time to make an exhaustive
analysis of the case. Wolff's patient had been in the
hands of half a dozen specialists, and they discovered
only the peculiar writing motions that the patient used.
This of course led them to false conclusions. If the
analysis in such a case is incomplete, the results must
be misleading.

3. It is worth while to compare the mental con-
dition of these patients with that of lower animals.
The two patients mentioned above forgot immedi-
ately what was said to them. If the correct associa-
tion did not occur to them after a short time, the
question had to be repeated. There was, however,
one exception. Objects or occurrences which were
intimately connected with their instincts they remem-
bered— for instance, money matters. We can imagine
that conditions maybe similar in lower animals — e. g.,
wasps, which either forget easily or only seem to re-
member certain things which are intimately connected
with their instincts — e. g., the location of the nest.

A qualitative difference has been supposed to exist
between the associative memory of man and that of
animals. These patients may help us to arrive at a
decision in regard to this question. When the patient
was asked the colour of blood, the question aroused
associations which caused him to provide the visual
impression of blood. If we compare with this the
fact that a wasp is no longer able to find its nest
when the latter is covered with a small blossom, we
might imagine that there is a qualitative difference

286 COMPARATIVE PHYSIOLOGY OF THE BRAIN

between the associative memory of the wasp and of
man. It might be argued that man possessed the
power of creating new associations, i. e.^ the abiHty
of substituting or changing the existing conditions, in
order to make a new process of association possible.
But this abiHty is not entirely lacking in animals.
When in Thorndike's experiment a cat goes volunta-
rily into a certain cage and waits there to be offered a
fish, we have to deal with the same apparent ability
of creating new associations. On the other hand, the
superiority of man in this direction can be accounted
for by the fact that his capacity for forming and re-
taining new associations is very much greater than
that of animals.

The question. What is the colour of blood ? pro-
duces not only one association — the word red — but
a number of other associations, for instance, the asso-
ciation of a wound and the association of the produc-
tion of a wound. If at that time the sense-impression
of a pustule occurs, the association arises that the
opening of the pustule causes the appearance of
blood. All experiments point to the fact that this
overwhelming abundance of associations which even
a disabled human brain can form is lacking in animals.
One impression may arouse only a very limited num-
ber of associations. This is evident from Thorndike's
experiments on dogs and cats (3), and from Whit-
man's observations on pigeons (4). This small capacity
for associations makes the reactions of animals appear
machine-like and less intelligent. I think that the

DISTURBANCES OF MEMORY 287

greater capacity of the human brain for associations
and the greater celerity with which these associations
are formed and retained are sufficient to explain why
mankind has been able to control nature, while animals
remain at its mercy.

In a pamphlet on Instinct and Intelligence, Father
E. Wasmann, S.J., a well-known entomologist, has
raised the question as to whether or not animals pos-
sess intelligence (5). The answer to such questions
varies with the definition of the word intelligence, and
hence such discussions result in a discussion of words
and definitions. Such scholastic discussions are very
serviceable for the defence of a dogma or an opinion.
Wasmann's pamphlet belongs in this category. But
we cannot overlook the fact that the steady progress
of science dates from the day when Galileo freed sci-
ence from the yoke of this sterile scholastic method.
The aim of modern biology is no longer word-discus-
sions, but the control of life-phenomena. Accordingly
we do not raise and discuss the question as to whether
or not animals possess intelligence, but we consider it
our aim to work out the dynamics of the processes of
association, and find out the physical or chemical con-
ditions which determine the variation in the capacity
of memory in the various organisms.

Bibliography.

I. RiEGER, K. B esc hrei bung einer Intelligenzstorung in Folge
einer Hirnverletzungy etc. Verhandl. der Wurzburger physikalisch
medicinischen Gesellschaft, Bd. xxii. and xxiii., 1889 and 1890.

288 COMPARATIVE PHYSIOLOGY OF THE BRAIN

2. Wolff, Gustav. Ueber krankhafte Dissoziation der Vor-
stellungen. Zeitschrift filr Psychologie und Physiologie der Sinnes-
organe, Bd. xv., 1897.

3. Thorndike, E. L. Animal Intelligence. The Psychological
Review, vol. ii., 1898.

4. Whitman, C. O. Animal Behaviour. Biological Lectures
Delivered at Wood's Holl, 1898. Boston, Ginn & Co.

5. Wasmann, E. Instinct und Intelligenz im Thierreich,
Freiburg, 1897.

CHAPTER XIX

ON SOME STARTING-POINTS FOR A FUTURE

ANALYSIS OF THE MECHANICS OF ASSO-

CIA TIVE MEM OR V

I. The facts have thus far shown that the reflexes
are determined chiefly by the structure of the sense-
organs, or of the surface of the body, and the arrange-
ment of the muscles. The central nervous system
participates in these functions only as a conductor.
The true problem with which the physiology of the
reflexes is concerned is the mechanics of protoplasmic
conductivity. This problem is no longer a biological
problem but a problem of physical chemistry.

The only specific function of the brain, or certain
parts of it, which we have been able to find is the
activity of associative memory. There is at present a
tendency to consider the anatomical and histological
investigation of the brain as the most promising line
for the analysis of these functions. It seems to me
that we can no more expect to unravel the mechan-
ism of associative memory by histological or morpho-
logical methods than we can expect to unravel the
dynamics of electrical phenomena by a microscopic

study of cross-sections through a telegraph wire or by
19 289

290 COMPARATIVE PHYSIOLOGY OF THE BRAIN

counting and locating the telephone connections in a
big city.

If we are anxious to develop a dynamics of the
various life-phenomena, we must remember that the
colloidal substances are the machines which produce
the life-phenomena. But the physics of these sub-
stances is still a science of the future. The new
methods and conceptions created by physical chemis-
try give us the hope that a physics of the colloidal
substances may be looked for in the near future. At
present we can only consider data of secondary im-
portance for the mechanics of associative memory.
The first group of these data is furnished by the
study of the functions of the sense-organs.

Helmholtz emphasised the fact that our senses only
furnish us symbols of the external world. Every
physical process that affects a sense-organ produces
changes in the organ. These changes are determined
by the peripheral structure or by the specific " energy "
of the sense-organs, as physiologists since Johannes
Muller call it. Whether a blow, an electric current,
or ether-vibrations of about 0.0008-0.0004 mm. wave-
length stimulate the retina, the sensation is always a
specific one, namely, light, while a blow or an electric
current produces sensations of sound in the ear.
This so-called law of the specific energy of the sense-
organs is not peculiar to the sense-organs ; it applies,
as was emphasised by Sachs, to all living matter ; it
even holds good for machines. It is in reality only
another expression for the fact that the eye, the ear,

FUTURE ANALYSIS OF MEMORY 291

and every living organ are able to convert energy in
but one definite form — that is, that they are special
machines. The determination of the way in which
this transformation of energy occurs in the various
organs would be the explanation of the specific energy
of the various senses.

Physiology gives us no answer to the latter ques-
tion. The idea of specific energy has always been
regarded as the terminus for the investigation of the
sense-organs. All the more credit is due Mach and
Hering for first having advanced beyond that limit
with their chemical theory of colour-sensations. Mach
has recently expressed the opinion that chemical
conditions lie at the foundation of sensations in
general (i).

For the eye we may consider it as probable that
light produces chemical effects. Various substances
are formed and decomposed in the retina, and the
chemical processes of the formation and decomposi-
tion of these substances determine the light- and
colour-sensations. The ether-vibrations of certain
wave-lengths influence these decompositions in a de-
finite manner. The electro-magnetic theory of light
will probably in this case lead to further discoveries.
Effects similar to those produced by light are also
brought about by the electric current. The current
itself can pass through the retina only by means of
electrolysis, and it may be that the increase in the
concentration of ions (wherever their progress is
blocked) brings about the light- and colour-sensations

292 COMPARATIVE PHYSIOLOGY OF THE BRAIN

caused by the current. It is not impossible that the
so-called visual substances — that is, the photo-sensi-
tive substances — are electrolytes. We can thus under-
stand how the electric current produces sensations of
light and colour in the eye. But it is more difficult to
account for the fact that pressure or a blow on the
eyeball produces the sensation of a flash. Carey Lea
has found that on photographic plates pressure pro-
duces changes of the same character as weak light.

The specific energy of the eye would accordingly
amount to nothing more than the fact that an increase
in the concentration of ions or certain other chemical
substances in the retina causes the sensation of light
and colour, no matter whether the changes are caused
by vibrations of the ether, by the electric current, or
by a blow on the eye. The stimuli which are trans-
mitted to the brain from the eye will hence show ex-
actly the variety and peculiarities which correspond
to the variety and peculiarities of the chemical pro-
cesses in the retina.

The same holds good for the stimuli which are
transmitted to the brain from the organs of taste and
from the nose. The chemical nature of the causes
that produce the sensations of smell and taste is so
apparent as to require no proof.

We find greater difficulty in dealing with the sense-
organs of the skin. Yet it is conceivable that a chem-
ical basis may also exist for the activity of these senses.
This idea finds support in a train of thought, by which
I attempted to explain the peculiar influence of grav-

FUTURE ANALYSIS OF MEMORY 293

ity on the orientation of animals and plants and their
formation of organs (2). In these cases, a change
in the orientation of the organs produces a change in
the chemical condition. If the chemical processes in
these instances consist in fermentative processes, the
amount of the chemical change in the unit of time
must be a function of the number of the ferment-mole-
cules and of the fermentable molecules that come in
contact. If we assume that both are present in dif-
ferent morphological constituents of the living cells,
that the ferment, for instance, is present in the nu-
cleus, the fermentable substance in the protoplasm of
the cell, it is apparent that a change in the position
of the cell or a pressure upon it will bring new mole-
cules of the protoplasm in contact with the nucleus.
In this way the metabolic activity may be increased.
Such changes in the peripheral nerve-endings of the
skin might result in innervations and reflexes. But
this is all so vague that it only indicates the possibil-
ity of the chemical character of the process. It seems
forced, if not altogether impossible, to apply this
theory to the cochlea of the ear. We could imagine
that the vibrations of sound produce corresponding
vibrations in the endings of the auditory nerve, by
which new molecules are brought into contact with
each other. But I cannot see how this assumption
could account for the different pitches or the phenom-
ena of consonance. While a chemical theory is pos-
sible or probable for certain sensations, e. g., light,
taste, and smell, it is very doubtful whether such a

294 COMPARATIVE PHYSIOLOGY OF THE BRAIN

theory can be applied to the other sense-organs. But
it is certain that if we wish to make any progress in
this direction we must follow the lead of Mach and
Herinof, and must cease to consider the so-called
law of the specific energy of the sense-organs as the
terminus of our investigation of the processes of
sensation.

2. If we wish to find out the dynamics of associa-
tion we must study the effects which simultaneous
processes have upon each other. Let us consider
periodic and aperiodic processes.

If we turn a wheel with one hand without thinking
of the manner or velocity of the rotation, and at the
same time repeat a poem to ourselves without moving
the lips, the number of the revolutions shows a simple
numerical relation to the number of the arses of the
verses. In German, where the arsis is pronounced
with greater emphasis than the thesis, the number of
the rotations of the wheel generally equals the num-
ber of the arses. Briicke first called attention to this
relation. Thirteen years ago I made a large number
of experiments (not published) concerning this sub-
ject that yielded the same result. But I found, fur-
ther, that if one intentionally turns the wheel rapidly
and recites slowly, the number of rotations is a simple
multiple of the arses. Two, three, or even more re-
volutions are made in the interval of one arsis. If one
recites very rapidly and turns the wheel very slowly,
the number of the arses becomes a simple multiple of
the number of revolutions. In the latter case, the

FUTURE ANALYSIS OF MEMORY 295

number of revolutions is often the same as the num-
ber of verses. If we assume that in thinking the
poem the respiratory innervations which follow the
rhythm can be represented as harmonic curves, and
that the same holds good for the innervations which
are responsible for the turning of the wheel, it follows
from these facts that harmonic processes of innervation
occurring simulta?ieously affect each other in such a
way that the periods of both processes are either equal
or in the ratio of simple m^ultiples of each other. It
requires great determination to withstand this law. I
consider it possible that where this succeeds the de-
viation from the law is only apparent, not real. In
reality it might be possible that one of the two har-
monic processes was stopped temporarily. The facts,
however, suffice to show that two harmonic processes
of innervation for different parts of the body, occur-
ring simultg.neously, influence each other and are most
liable to form processes of equal period.

The same is true not only for two or more simulta-
neous processes of motor innervation, but also for
simultaneous sensory processes and motor innerva-
tions, as is proved by dancing. The rhythm of the
music and the period of the motor innervations of the
legs and body coincide.

3. A priori it would follow from these facts that
two simultaneous aperiodic processes will in general
interfere with or inhibit each other. That this is to
a certain extent true is shown by the experience that
we cannot do two thines well at the same time. We

296 COMPARATIVE PHYSIOLOGY OF THE BRAIN

must add, however, the provision that the two things
are aperiodic. If they be periodic, the opposite is
true. We cannot solve an equation while jumping
over a broad ditch. According to Fechner's inter-
pretation of this fact the brain at any time has only a
certain amount of energy at its disposal. In jump-
ing over a ditch all the energy is supposed to pass
into the muscles and nothing is left for the process
of thinking. I showed fourteen years ago that Fech-
ner's conception was not correct. The inhibition of
a process of thought by simultaneous muscular ac-
tivity is greater when we innervate one arm than
when we innervate both arms simultaneously. Ac-
cording to Fechner, however, the greater the number
of muscle-groups that were innervated the more
energy must be consumed in the brain. In these ex-
periments I measured the maximal pressure which
the flexors of the hand afe able to exert on a dynamo-
meter. This pressure does not decrease when the
other hand or all the muscles are innervated simul-
taneously, but even increases (3). The further appli-
cation of this method explained the fact that we
cannot well be mentally and physically active at the
same time.

If we begin by solving a moderately difficult prob-
lem in mental number work, and if we attempt when
in the midst of the task to attain the highest dynamo-
metrical pressure with the hand, the pressure re-
mains from 20-30^ below the maximum that we
otherwise attain when we devote our attention to the

FUTURE ANALYSIS OF MEMORY 297

pressure alone (3, 4). It often occurs, however, that
the maximal pressure is obtained while reckoning.
In this case the experimenter certainly interrupted
his reckoning while pressing. This is shown by the
fact that in this case either the task is not solved cor-
rectly or the problem is entirely forgotten by the
subject experimented upon. It was a great excep-
tion if the maximal pressure was attained and the
task also solved correctly. The experiments result
quite differently, however, when the experimenter
first begins by pressing, and the problem is given
when the maximal pressure has already been reached,
so that it is only necessary to keep up the pressure.
In this case I noticed no, or only a very slight, in-
fluence of both activities : the person could reckon
correctly, although with effort, and the curve either
did not descend, or descended but little lower than
without the reckoning.

Thus we see that a simultaneous, static innerva-
tion, no matter how strong, does not prevent the
reckoning ; that, on the other hand, a rapidly increas-
ing maximal motor innervation disturbs the process
of reckoning perceptibly. I have attempted to dis-
cover whether a sudden stoppage of motor innerv-
ation— that is, a sudden relaxation of the statically
contracted muscles — disturbs the process of reckon-
ing. This is not the case.

Whatever may be the explanation of these phe-
nomena, we see that two simultaneous, maximal,
aperiodic processes of innervation which require an

298 COMPARATIVE PHYSIOLOGY OF THE BRAIN

effort disturb each other. On the other hand, if they
have not a maximal intensity they can take place simul-
taneously. I fottnd that easy tasks 07"- the reproduc-
tion of simple matters of m^emory did not lower the
maxiTfium of the pressure.

4. These experiments recall the disturbances of
associative memory which were discussed in the
preceding chapter. By causing powerful processes
of motor innervation to go on, we interfere with
all associations, except those which have occurred
very frequently. This was the characteristic of the
patients mentioned in the preceding chapter. But at
the same time it does not exhaust the case. The
processes of innervation in the brain of these patients
were possibly mutilated not only in intensity but also
in other dimensions or directions.

Perhaps the cases of the inhibition of reflexes also
belong in this same category of phenomena. We
have mentioned that a dog with severed spinal cord
shows pendulum-movements of the hind-legs when
they are allowed to hang down. But if we press the
skin of the tail gently the pendulum-movements of
the legs at once cease (Goltz). Some authors seem
to be under the impression that a shock-effect must
consist in the exhaustion of the parts under the in-
fluence of the shock. This is not necessarily true.
The shock-effect may be due to a phenomenon of
interference or to a comparatively slight physical
change which results in a mutilation of the processes
of innervation.

FUTURE ANALYSIS OF MEMORY 299

The r6le which the intensity plays in the case of
two simultaneous processes of innervation recalls the
influence two simultaneous wave-motions have upon
each other. A superposition of two waves is only
possible as long as the amplitude is not too great.
It looks as if two processes can occur simultane-
ously in our brain only when their intensity is weak
enough to allow a superposition.

It is perhaps allowable to pursue this possible
analogy of the processes of innervation with wave-
motions a step farther and apply it to the process of
association. A process remains associated with those
processes in our brain which occur quite or almost
simultaneously. Let us imagine that every process
in our central nervous system has a definite form in
so far as it can be represented by a curve in which
the time-elements are represented by the abscissas
and the intensity of the processes by the ordinates.
If two processes take place simultaneously and their
intensity is not too strong, they superpose each other.
The traces which this process leaves in our central
nervous system correspond to the curve which is de-
termined by the superposition of both elementary
curves. If one of the processes takes place later on,
the other process also is reproduced by resonance.
On the other hand, a very complicated process may
reproduce simpler processes, which are contained
in the former as constituents and have already oc-
curred once before in their simple form. Our ex-
perience concerning sound-sensations shows indeed

300 COMPARATIVE PHYSIOLOGY OF THE BRAIN

that simultaneous simple harmonic motions may in
our sensations fuse to a compound sound of definite
character {Klangfarbe). Moreover, a trained ear is
able to decompose a compound Klang into its simple
harmonic constituents. The mechanism of associa-
tive memory must share these peculiarities of our
organs for the perception of sounds. I believe that
what we commonly call intelligence depends partly
upon the development of this power of resonance of
the mechanism of association.

The existence of phenomena of resonance in our
nervous processes may account for the fact that stim-
ulation' of the same organ yields entirely different
results if we change the character or rhythm of stim-
ulation. Only certain sounds cause a dog to howl ;
only a certain way of rubbing the skin of a frog
causes the animal to croak. The so-called law of
the specific energy of the sense-organs has pushed
these important facts into the background and has
tried to convey the idea that the character of the
stimulation was something indifferent. Although
it is true that a blow on the eye gives rise to a
sensation of light, nobody would for one instant mis-
take this light-sensation for one caused by ether-vi-
brations. It is of course impossible to throw light
on this subject from the anatomy or histology of the
brain. But our experiences in regard to sound-sensa-
tions promise the possibility of an analysis of these
phenomena. Hermann and Mach come to the con-
clusion that the physical resonance-theory of Helm-

FUTURE ANALYSIS OF MEMORY 301

holtz is no longer tenable, and that it may have to be
substituted by a physiological resonance-theory (6, 7).
According to Hermann, we may assume that the
nervous end-organs themselves are especially sens-
itive for stimuli of a definite period (7). A gen-
eralisation of this assumption would lead to an
understanding of the above-mentioned fact. The
motor organs of the larynx may be considered as
resonators, and this would explain why only cer-
tain tones cause a dog to howl and why only fric-
tion of a certain character — i. e., periods — causes a
frog to croak. The fact of the easy transmission of
sounds into innervations to the larynx in human
beings and parrots or song-birds would depend on
the same principle. But in theories of this character
we must leave some leeway for the influence of chem-
ical processes. The phenomena of correlation which
we notice in many animals during the period of heat
may be determined by substances circulating in the
blood during that period (internal secretion). This
may account for the change in the irritability during
that period.

5. Our space-sensations are varieties of three dimen-
sions. The main coordinates show a definite relation
to the main axes of our body. This leads us to a
consideration of the possibility whether certain struct-
ural conditions of our body determine the main coor-
dinates of our system of space-sensations. Hering
has shown that the motor innervations of our eyes
may be reduced to three kinds, corresponding to the

302 COMPARATIVE PHYSIOLOGY OF THE BRAIN

main axes of our body : (i) innervations to move our
eyes from right to left, or vice versa ; (2) innervations
to move them up and down ; and (3) move them
from a near to a far object, or vice versa (8). The
first motion takes place along the transverse axis, the
second along the longitudinal, and the third along
the dorso-ventral axis. The experiments on the hori-
zontal semicircular canals of the ear show that the
stimulation of this canal produces motions of the eyes,
head, or even of the whole animal in the plane of this
canal. Our experiments on galvanotropism indicate
the existence of a simple relation between the orien-
tation of certain motor elements in the central nerv-
ous system and the direction of the motions produced
by their activity. This is supported to a certain
extent by the experiments on the crura cerebelli. It
thus seems possible that simple geometrical relations
of structure are responsible for the fact that all our
innervations may be reduced to three classes deter-
mined by the main axes of our body. On the other
hand, Mach furnished the proof that the will or the
process of innervation for a motion is of the same
character as the process of space-sensations (5, 6).
The will to move our eyes to a certain point and
the space-sensation itself can be added algebraically.
The experiences derived from space illusions caused
by imperfect motility of the eyes or hands agree with
this view. Moreover, Mach has proved that we recog-
nise the geometrical symmetry of two figures very
easily only when the axis of symmetry coincides with

FUTURE ANALYSIS OF MEMORY 303

that of our body (5, 6). All these facts indicate that
the main coordinates of the physiological space are
determined by structural peculiarities of our body.
The ultimate structural elements in this case are not
necessarily of a morphological or histological order.
They may be, as Mach has intimated in another con-
nection, the stereochemical configuration of certain
molecules (i). As it is not my intention to enter
here upon a discussion of the nature of the space-
sensations, but to indicate what place they may oc-
cupy in a future mechanics of the activity of the brain,
these hints may suffice.

Bibliography.

1. Mach, E. DiePrinciptender Wdrmelehre^^. ^60. Leipzig,
1896.

2. LoEB, J. Zur Theorie der physiologischen Licht- und Schwer-
kraft-Wirkungen. Pfluger's Archiv^ Bd. Ixvi., 1897.

3. LoEB, J. Muskelthdtigkeit als Maass psychischer Thdtigkeit.
Pflugers Archiv^ Bd. xxxix., 1886.

4. Welch, J. C. On the Measurement of Mental Activity through
Muscular Activity^ etc. The American Journal of Physiology^
vol. i., 1898.

5. Mach, E. Contributions to the Analysis of the Sensations.
Chicago, 1897.

6. Mach, E. Die Analyse der Empfindungen und das Verhdlt-
niss des Physischen zum Psychischen. Jena, 1900.

7. Hermann, L. Beitrdge zur Lehre von der Klangwahr-
nehmung. Pfliiger's Archiv^ Bd. Ivi., 1894.

8. Hering, E. Die Lehre vom binocularen Sehen, Leipzig, 1 868.

INDEX

Acalephae, 16-34, 97
Acoustic nerve. See Auditory Nerve.
Actinians, 41, 48-60, 221, 227, 228
Amblystoma, i6<>-i62, 204, 205, 231,

264
Amnesia, 277-287
Amphipyra, 184
Annelids, 82-100
Anterior roots, 113, 135
Ants, 195, 220-224
Apathy, 137
Aphasia. See Amnesia.
Arbacia, 203
Arnold, 39
Arthropods, 101-127
Ascidians, 35-47

— heart of, 28-30
Association, 9

— corpuscular theory of, 277

— distribution of, 213, etc.

— disturbances of, 277, etc.

— dynamical theory of, 278, etc.,

294

— relation to cerebral hemispheres,

236, etc., 262, etc.

— relation to consciousness, 12-14,

214, etc., 236, etc.

— relation to instincts, 196
Associative memory. See Associa-
tion.

Astacus, See Crayfish.

Asterina, 69, 70

Auditory nerve, 134, 152, 155, 158,

172
Aurelia aurita, 17
Automatic processes, 9, 10, 16-30,

106, etc.

Balanus perforatus, 192
Baumann, 207

Bawden, 258

Bed-sores, 209

Bee, 122, 123, 224, 227, 230

Bell, 136

Bethe, 45-49, "4-127, 131, I55, I59.

219-224, 226, 235
Bickel, 149
Blasius, 164

Blood-vessels, 43, 44, 274
Brown-Sequard, 39
Briicke, 294
Budge, 39

Burrowing of worms, 91, 96
Buttel-Reepen, v., 227, 235
Butterflies, 182, 222

— larvae of, 188

Carcinus maenas, 45, 119

Centres, in spinal cord, 134-149, 273

— in cerebral hemispheres, 259-276.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Comparative physiology of the brain and comparative psychology"

See other formats