Full text of "Comparative physiology"

See other formats

Marine Biological Laboratory Library

Woods Hole, Massachusetts

Gift of F. R. Lillie estate - 1977

TEXT-BOOKS OF ANIMAL BIOLOGY

Edited by Julian S. Huxley, M.A.

Professor of Zoology, King's College, London

COMPARATIVE PHYSIOLOGY

TEXT-BOOKS OF ANIMAL
BIOLOGY

Edited by Professor Julian S. Huxley
Other volumes in preparation.

ANIMAL ECOLOGY. By C. S. Elton.

VERTEBRATE MORPHOLOGY. By
G. R. DE Beer.

EXPERIMENTAL ZOOLOGY. By Julian
S. Huxley.

ANIMAL MORPHOLOGY, WITH ESPE-
CIAL REFERENCE TO THE IN-
VERTEBRATA. By W. Garstang.

^-^:

COMPARATIVE
PHYSIOLOGY

LANCELOT T. HOGBEN

M.A.(Cantab.), D.Sc.(Lond.)
Assistant Professor in Zoology, McGili» UNiVERiiTY

LONDON

SIDGWICK & JACKSON, LTD.
1926

PRINTED IN GREAT BRITAIN BY

WILLIAM CLOWES AND SONS, LIMITED

LONDON AND BECCLES.

E. A. S. S.

AUTHOR'S PREFACE

There is, so far as I know, no work in English which aims at
giving an account of the physiology of the lower organisms.
Few of those who are aware of the existence of Winterstein's
monumental work are likely to find the time to obtain from its
encyclopaedic pages a bird's-eye view of the ground already
traversed and the fields that lie ripe for the research worker
armed with sufficient familiarity with animal life and under-
standing of physiological methods of inquiry. Winterstein's
Vergleichende Physiologic meets the needs of the research worker
who is in search not of problems to tackle so much as detailed
information of previous inquiries on similar lines to those
with which he is concerned. There seems nothing to supply
any encouragement to those who are not sufficiently advanced
in their studies to distinguish between lines of inquiry that are
practicable as well as profitable, to realise as yet what materials
are available for the solution of the problem in which interest
has already been quickened, or to have gained much insight
into the methods at our disposal for extending our knowledge
of the physiology of the lower organisms.

I am well aware that to attempt to supply this need within
the limits of space at my disposal would be a sufficiently
embarrassing task for an author reassured with a far more
exalted sense of his own equipment for the task than I can
boast. If I have succeeded in stimulating twenty-one years of
age (or thereabouts) to dip into an immense and at present
scattered literature and find some fruitful fields of inquiry and
sources from which more precise information can be obtained,
I shall have accomplished precisely what I set out to do.

These chapters represent the materials of a course of
lectures delivered first in the Zoology Department and later in

viii AUTHOR'S PREFACE

the Department of Physiology in Edinburgh University to
medical students having completed a course of elementary
physiology and to science students taking an honours course
in zoology. I have in mind the same mixed audience as
readers : the advanced student in zoology who knov^^s very
little physiology, and the student v^ho, having passed through
a course in physiology designed to equip him for the pursuit
of the medical profession, may wish to acquire information
about branches of the subject that have at present no such
remunerative value. In doing this one has the feeling of falling
between two stools. The physiological critic will object to
dealing with topics which practitioners do not regard as the
business of the physiologist ; while zoologists will protest
against omission of reference to experimental work which seems
to them to be as important as much that has been treated as
physiological in the pages which follow.

Since the objective of physiological inquiry is the quantita-
tive study of the relation between processes characteristic of
living organisms and properties of inanimate matter, in
attempting to treat the subject with reference to a coherent
theme one of two courses is open : to illustrate the known
properties of non-living matter by reference to their operation
in the processes of living organisms, or to consider what are
the characteristic properties of animate systems and inquire
how far it is possible to interpret each in terms of knovm
physico-chemical laws.

Against the former course, it is sufficient to point out :
first, that this method of treatment has been adopted as
successfully and comprehensively as possible in the existing
state of knowledge in such works as those of Bayliss and of
Hoeber ; secondly, that if carried out consistently it necessitates
the elimination of all reference to some of the most character-
istic properties which distinguish living systems. Generally
speaking, those who restrict the scope of physiology to pheno-
mena for which ready-made physico-chemical explanations are
at hand, make an exception for the treatment of reflex action.
By some obscure convention this grace is rarely extended to
the phenomena of reproduction. I shall make no apology for

AUTHOR'S PREFACE ix

regarding the building-up of a new animate system as a proper
field for physiological inquiry.

On the other hand, I do not regard the terms '' experi-
mental " and " quantitative " in the sense employed above
as co-extensive. For this reason no reference is made in the
last chapter to the large body of work on implantation of
organs and regeneration, much of which is of great importance,
but like the too familiar descriptions of tracts in the mammalian
spinal cord, not susceptible as yet to treatment in relation to
the fruits of inquiries based on the use of physiological methods
as ordinarily understood.

Generally speaking, I have borne in mind the fact that
Winterstein's Handbuch makes the literature of comparative
physiology accessible to those who care to consult it up to 191 2.
I have therefore aimed at familiarising the reader with what
has been done during the last ten or fifteen years. Where
references cannot be obtained by consulting monographs, the
particulars of the journals in which they are found are given.
The completeness of Winterstein's bibliography makes any
attempt to give further assistance to the student a work of
supererogation.

In the selection of materials, one is naturally expressing
one's individual judgment ; and it is hoped that the reader
will appreciate that the author puts forward no claim to be
authoritative or encyclopaedic. The material selected has been
chosen to help the student of zoology to appreciate what is being
achieved by the application of physiological methods to the
study of the lower animals, and to widen the horizon of the
student of physiology who has not been brought into touch with
the diversity of problems which are suggested by a considera-
tion of function in a wider range of animals than those with
which he has been accustomed to deal in the course of his
medical studies.

My thanks are due to Dr. A. D. Macdonald and Mr. A. D.
Hobson, who read the MS., and to Professor Julian Huxley foe
valuable suggestions.

LANCELOT T. HOGBEN.

March i, 1925.

CONTENTS

CHAPTERS I— III

PAGE

Response — the Manifestations of Vital Activity

I. Muscular Contraction ....... i

II. Ciliary Activity, Amoeboid Motion and Colour Response . 23

III. Secretion ......... 47

CHAPTERS IV— VI

Metabolism— THE Sources of Vital Energy
IV. Respiration ......... 64

V. Nutrition 85

VI. The Circulation of Body Fluids . . . . , loi

CHAPTERS VII- IX

Co-ordination — the Integration of Vital Activities

VII. Endocrine Co-ordination 118

VIII. The Mechanism of Nervous Conduction and Excitation . 134
XI. The Analysis of Behaviour in Animals . . . • 151

CHAPTERS X— XII

Reproduction — the Building up of a New
Animate Unit

X. The Fertilisation of the Egg ...... 169

XI. Inheritance ......... 182

XII. The Physiology of Development ..... 200

LIST OF ILLUSTRATIONS

1. Diagram illustrating liberation of potential energy in muscular

contraction ........ 6

2. Scheme to illustrate modern work on muscular contraction . 9

3. Time relations of the electrical variation in the isometric

response to two successive stimuli . . . '13

4. Mechanical heat and electro-cardiogram of heart of Homarus 14

5. Effect of excess of calcium on the perfused heart of Homarus 17

6. Effect of removal of magnesium on the perfused heart ot

Pecten . . . • • • • • . j8

7. Effect of removal of calcium on the perfused heart of Pecten 19

8. Diagram of ciliary motion .....

9. Relation of hydrogen ion concentration to ciliary movement

10. Relation of oxygen consumption and mechanical activity of

ciha to temperature ......

11. Relation of hydrogen ion concentration to the velocity of

amoeboid movement ......

12. Effect of temperature on velocity of amoeboid movement

13. Melanophores of Fundulus .....

14. Melanophores of Fundulus : effects of adrenaline.

15. Results of variation of CO2 tension of inspired air of insect

16. Dissociation curves of mammalian blood

17. Dissociation curves of Arenicola blood

18. Dissociation[]curves of Hamocyanin in Crustacean Blood

19-20. Absorption spectra ......

21. Relation of amount of COj taken up by mammahan blood to
CO2 pressure .......

24
27

32
35
38

68
72

74
76
78

XIV

LIST OF ILLUSTRATIONS

22. Relation of amount of COg taken up by crustacean and

cephalopod blood to CO2 pressure

23. Ciliary currents on Lamelli branch gill

24. Crystalline style of bivalve mollusc

25. Innervation of heart of Palinurus

26. Heart and nerves of Limulus .

27. Inhibition of heart-beat of Limulus by electric stimulation of

brain .....

28. Heart of Cephalopod

29. Circulatory system of Ascidian .

30. Action of adrenaline on heart of Pecten

31. Action of adrenaline on crop of Aplysia

32. 33. Function of pituitary gland in coloration of frogs

To face page

34. Excitability to second stimulus in sciatic gastrocnemius of

frog

35. Diagram of Adrian's experiment

36. Diagram illustrating electrical conditions

resting neurone ....

37. Excitability by single or double stimuli of

claw of Astacus ....

38. Diagram of simple reflex arc

39. Diagram of pedal ganglion of Razor-shell

40. Effect of unequal illumination on insect

41. Genetic segregation

42. Sex-linked inheritance in Drosophila

43. Crossing in the gypsy-moth Lymantria

44. Oxygen consumption of chick embryos

in excited and
abductor nerve of

COMPARATIVE PHYSIOLOGY

CHAPTER I

MUSCULAR CONTRACTION

Physiological science is concerned with describing those
properties which distinguish Hving beings from inorganic
objects, and relating the processes specially characteristic of
the former to the more familiar and accessible phenomena of
which we have exact knowledge in the realm of inanimate
matter. It is not legitimate to be dogmatic regarding the
extent to which similar principles will be found to hold good
both in biological and physical science. But the onus of proof
lies on those who discourage the attempt to further this end.
There have always been those who wish to set limits upon the
extent to which the mechanistic approach to vital phenomena
can continue to yield profitable results. On the very eve of
Wohler's synthesis of urea, Henry wrote with reference to the
artificial production of organic compounds, " It is not probable
that we shall ever attain the power of imitating nature in these
operations. For in the functions of a living plant a directing
principle appears to be concerned, pecuHar to animated bodies,
and superior to and differing from the cause which has been
termed chemical affinity." It may be said, however, that the
validity of a mechanistic outlook stands quite apart from the
possibility of manufacturing animate systems, just as truly as
the justifiability of interpreting the movements of the heavenly
bodies in terms of the dynamical relations of immediate ex-
perience is independent of the likelihood that we shall ever
succeed in bringing into existence a new satellite for Jupiter.

^ COMPARATIVE PHYSIOLOGY

For our present purpose we shall wherever possible seek
to relate the properties of living matter to those of inanimate
nature ; where this cannot be done in the present state of
knowledge, we must proceed with the task of recording our
observations, as in the physical sciences, in quantitative
terms.

In so doing we shall consider first the characteristic
activities which living organisms display ; second, the sources
of energy which lie behind these activities ; third, the way in
which the activities of an organism are co-ordinated with the
changing conditions of the external world ; and finally, the
means by which a new animate unit is brought into being.

Organisms respond to their surroundings by movements of
various kinds — muscular y ciliary ^ amoeboid ; by the elaboration
of material secretions ; by the production of light, electrical
discharge ; and by changes in bodily colour. Structures which
carry out these responses in Metazoa are collectively referred to
as effector organs. The first type of effector which will be dealt
with is muscle. Muscular activity is a ubiquitous phenomenon
in metazoan organisms ; and therefore cannot be excluded from
the present survey of the physiology of the lower animals,
although our knowledge of muscular mechanism is largely
derived from the study of vertebrate animals.

Of no form of response in organisms is our knowledge more
extensive than in the case of muscular contraction. The
greater part of this knowledge is based upon the study of
amphibian skeletal muscle. Before considering the quantita-
tive treatment of the energy changes associated with excitation
in muscle, a brief sketch must be given of those elementary
phenomena which can be demonstrated when an excited muscle
is allowed to lift a weighted lever whose movement is recorded
on the surface of a revolving drum. By this method (isotonic
contraction) we can arrive at some preliminary insight into the
sequence of events in the contraction cycle.

When a muscle is excited by a single electrical stimulus the
curve recorded in this way shows three distinct phases : (i) a
period of latency intervening between the applicadon of the
stimulus and the beginning of response ; (ii) a period in which

{

MUSCULAR CONTRACTION 3

the mechanical response rises to a maximum ; and (iii) a period
in which the mechanical response falls off, i.e. the muscle
relaxes. If a sufficient interval elapses between the completion
of relaxation and the application of a second stimulus, the
contraction curve traced out by the response to the latter will
be identical with the first. If, however, we record the response
to a succession of stimuli of equal strength sent in successively
at equal intervals very soon after relaxation is complete, the
period of relaxation becomes progressively more prolonged, and
the height to which the lever rises at each contraction gradually
falls off. Thus we must add to the above a fourth phase, the
recovery period, during which the muscle is restored to its
original condition. When a second stimulus is applied to a
muscle before it has completed the process of relaxation, the
contraction due to the second starts from the level at which the
previous one happens to be when the new one comes into
operation ; if the process is repeated a summation results so
that the height of the combined contraction is greater than that
of a single twitch. But each stimulus produces less increase
than its predecessor, a Hmit being soon attained when further
succession of stimuli only permits a maximal level to be main-
tained. This maximum depends partly on the frequency with
which the stimuH are sent in, i.e. how far the contraction due
to one stimulus has progressed before that due to the next
starts. Prolonged contractions of this kind correspond to the
dehberate and sustained movements of everyday life, and are
described by the term tetanus.

In seeking for light on the way in which the muscular
mechanism works it will be best to consider separately the
energy changes which occur when a muscle is excited. There
are, in addition to the mechanical response itself (which can
only occur if the muscle is permitted to shorten), chemical,
thermal, and electrical phenomena which appear whenever the
muscle is excited with an adequate stimulus. Since we are
ultimately concerned to explain the origin of the mechanical
response, it is of the utmost importance to have in the first
place some means of treating the mechanical energy of con-
traction with quantitative accuracy. Though this might at

4 COMPARATIVE PHYSIOLOGY

first sight appear a simple issue, it is really a very complex one,
and it is only within recent years that the work of A. V. Hill has
placed the question on a satisfactory basis.

{a) Mechanical Response in Muscle.— In attempting to
grasp the significance of the chemical and thermal aspects of
contraction, it is necessary to measure the potential mechanical
energy of contraction. The inevitable Umitations of laboratory
equipment tend to give the beginner a distorted idea of the
significance to be attached to records of work done by a muscle
in Hfting a lever. Since the tension of the muscle is not the
same at every stage in the contraction, there are two obvious
difficulties in the interpretation of obsei-vations on isotonic
contractions : either the muscle is too heavily weighted and
cannot contract fully, or it is insufficiently weighted at the
beginning and cannot exert its maximal energy. A more
subtle difficulty lies in the fact that the muscle is an elastic
body. When a resting muscle is stretched by virtue of a load
it possesses potential energy like that of an extended spring ;
thus it does not follow that the work it may be made to perform
when its tension is increased during excitation is entirely the
result of the energy so liberated. An analogy given by A. V.
Hill will make this clear. Imagine a balance with two scale-
pans equally balanced when empty. Suppose that a weight of
one kilo is placed in one pan, the other being held in its original
position by a spring. The spring is now exerting a tension of
one kilo. If a small weight— say, lo grams— is placed on the
empty scale-pan it will sink, let us say, i cm., and the other
pan will be raised a corresponding height, thereby doing work
proportional to the product of the weight and the distance
through which it is raised. Clearly the lo-gram weight has
not contributed more than one-hundredth of this energy ; the
remainder is derived from the potential energy of the spring.
In just the same way, the energy which we get out of a weighted
muscle is not simply the amount of energy liberated by the
contractile mechanism sensu stricto.

The way in which we may calculate the maximal amount of
work which the muscle is capable of doing in virtue of the
energy freed in contraction will present no difficulty to the

MUSCULAR CONTRACTION 5

student who is familiar with the expression for maximum work
done by a gas in changing its volume. The condition for
maximal work done by a gas in expanding from v to v' is first,
that the internal pressure P should be opposed at each stage by
an external pressure F—dp differing from it by an infinitesimal
amount. Then, if the process be carried out so slowly that the
gas is not allowed to gather momentum and dissipate part of
its energy as heat

W= r FJv

In order to realise the total amount of potential energy liberated
in contraction, one must imagine that the tension of the muscle
does work at every stage against a load differing by an in-
definitely small amount from the tension it exerts. Since the
energy released at each infinitesimal step is the product of the
force into the distance, the total energy is the sum of a series of
products Tdl ; expressed analytically :

W= r T.dl

(E being the extended length and C contracted length of the
muscle.)

In this form it is not possible to evaluate W directly, since
we do not know what function T is of L. But the integral
formula at once suggests that the potential energy of the
contractile mechanism is represented by the area of a curve
expressing the relation of tension to length in the unstretched
muscle. We have therefore to construct a tension-length
indicator diagram for the muscular machine analogous to the
familiar pressure- volume indicator of the heat engine.

The accompanying diagram (Fig. i) will explain to the
reader unfamiliar with the notation of the calculus the way in
which W is calculated. OC is the contracted, OE the extended,
length of an unstretched muscle. Ti is the initial tension of
the muscle, T2, T3, T4, etc., the tensions exerted when the
muscle has been allowed to shorten by equal steps A^- The
average tension between Ti and T2, T2 and T3, etc., are
represented by ^1, toy ^3, etc. In contracting through the
first step A^ the work done (force X distance) is hAh in

6 COMPARATIVE PHYSIOLOGY

contracting through the second step the work done is t^l^l.
The total work is the sum of a series of products repre-
sented by the rectangular areas which are together equivalent
to the area T,EC.

If TjC is curvilinear, the area enclosed by the curve may be
made as near as we like to the sum of these rectangles by
making A^ sufficiently small.

To construct such a tension-length curve the muscle is
excited isometrically, i.e. an arrangement is used by which the
tension is recorded by a spring lever without appreciable

Fig. I.

movement of the muscle itself. In general the records so
obtained resemble those of isotonic twitches. The procedure
adopted is as follows. The muscle is held rigidly by a clamp
which can be screwed up and down a graduated scale. At
the beginning of the experiment the clamp and muscle are
screwed down until the cord attaching the other extremity of
the latter to the tension lever is just tight, but not sufficiently
so to stretch the muscle appreciably. A stimulus is now given,
and the tension recorded. The screw of the clamp is now
turned so as to raise it i mm., thereby slackening the cord.

MUSCULAR CONTRACTION 7

The tension developed when the muscle is stimulated is then
recorded. This represents the force the muscle exerts when
it has contracted through i mm. In a similar way the tension
is observed in stages of i mm. till the muscle is so slack that
it exerts no tension on the lever, i.e. till the clamp has been
raised through a distance corresponding to that through which
the muscle contracts in a single twitch.

The experiment is only valuable as a means of arriving at
a simple expression for ETdl. in terms of easily determined
quantities. Now the potential energy of an elastic body
stretched to a length x from the unextended condition ex is
JTa:(i— rx), and we might therefore anticipate that the energy
of contraction would be of the general form K.T/, where K
is a constant, T the initial tension and / the normal length.
According to Hill's determinations the area of the tension-
length curve satisfies the relation K=i/6 approximately.

Thus the potential energy liberated in a single twitch is
T//6. In practice all this energy is not realised, because while
it has been found possible to devise an apparatus by which the
force opposed to the muscle is balanced against the actual
tension throughout the act of contraction, we cannot carry
out the process so slowly as to avoid degradation of energy
through internal friction {i.e. protoplasmic viscosity). For
an understanding of the mechanics of muscle we are only
concerned with the theoretical value of W, i.e. Tljb.

(b) Chemical Phenomena in Contraction of Muscle.— The
essential facts as regards skeletal muscle are two : first, that
the muscle can contract and relax in the total absence of oxygen,
though the presence of oxygen delays the onset of fatigue ;
secondly, that lactic acid is produced in the process of con-
traction, rapidly disappearing in the presence of oxygen, but
accumulating if oxygen is excluded. The elucidation of this
aspect of the contractile processes is due in the first place
to the work of Fletcher and Hopkins. The conclusions which
can be drawn from their observations are that the production
of lactic acid without utilisation of oxygen is the salient event
associated with the tension which results in muscular con-
traction ; that the accumulation of lactic acid underlies the

8 COMPARATIVE PHYSIOLOGY

phenomenon of fatigue, in which the muscle is unable to
recover its original reactivity between successive stimulation ;
and that the oxidative removal of lactic acid is an essential
feature of the recovery process by which the state previous to
excitation is restored. More recently Meyerhoff has shown
that the appearance of lactic acid in muscle is correlated v/ith
the disappearance of a corresponding amount of glycogen, a
hexose diphosphate being an intermediate compound in the
transformation. The appearance of the lactate ion in the
contraction process must not be taken to imply that there is
any observable increase in hydrogen ion concentration in a
single twitch or short tetanus ; Ritchie finds that there is not.
The presence of free lactic acid must ordinarily be instan-
taneous, as one would expect in a buffered system such as that
which exists in tissues. Since the mechanical relaxation of
muscle occurs as well in the absence of oxygen, this part of the
recovery may be assumed to correspond with the immediate
neutralisation of the lactic acid set free at excitation. Further
light can be obtained on this question by considering heat
production in muscular contraction.

(c) Heat Production in Muscle.— That the temperature of
a muscle rises during contraction is easily demonstrated by
stimulating living and dead muscle in contact with the metallic
junctions of a thermopyle placed in circuit with a sensitive
galvanometer. The extension of our knowledge of the heat
production of muscle in recent years is chiefly due to the work
of A. V. Hill. Two questions are of pre-eminent interest as
throwing light on the mechanism of muscular contraction,
namely, the relation of the heat produced to the chemical events
of the contraction cycle, and the relation of heat-production
to the potential energy for mechanical work set free at
excitation.

Hill has formulated the first of these two issues in the
following way : Is the heat-production of muscle given out
in some process by which the tension of muscle is increased,
or in the recovery process by which the mechanism is restored
to its original condition ? A satisfactory answer to this
question is obtained by studying the time-relations of heat

MUSCULAR CONTRACTION 9

production and comparing the effects of stimulation upon heat-
production in conditions promoting or impeding the recovery
process.

According to Hill's data three definite conclusions can be
dra\vn. Firstly, when a muscle is excited directly or indirectly
in oxygen by either a single shock or a short tetanus, the
liberation of heat continues for some time after the mechanical
response is over. Secondly, the amount of heat so liberated
after mechanical response is over is, in oxygen, at least as great
as that evolved in the contraction and relaxation itself. But

Heat production in Ng

Heat production in O.^
Fig. 2. — Scheme to illustrate modern work on muscular contraction.
I. Latent period — concentration of ions at surface of excitation (elec-
trical change) . 2. Development of tension — associated with disappearance
of glycogen and appearance of lactic acid. 3. Relaxation — lactic acid
neutralised by buffer action of muscle proteins. 4. Recovery — lactate ions
in part oxidised : some glycogen reappears, (a) Heat of formation of
lactic acid from glycogen, {b) Heat of dissociation of muscle proteins.
(c) Heat of combustion of part of lactic acid.

thirdly, when the muscle is stimulated in nitrogen, there is
hardly any heat-production after contraction is over, though
the normal quantity is evolved when oxygen is again admitted.
It seems therefore justifiable to infer that oxygen is used up,
and " delayed " heat liberated, in that part of the recovery
process which occurs after relaxation is complete. As we have
already seen, this part of the recovery process is dependent on
the presence of oxygen and cannot occur in pure nitrogen.

10 COMPARATIVE PHYSIOLOGY

Thus part of the heat generated in muscular contraction arises
in the process which results in tension, part in the process by
which the substance liberated at excitation is reinstated or
removed.

When the muscle is stimulated isometrically no work is
done. All the energy must disappear as heat. By subtracting
the heat-production of isometric contraction in nitrogen from
the heat-production of isometric contraction in oxygen, that
of the oxidative recovery process is obtained. The heat-
production in pure nitrogen represents, on the hypothesis
advanced above, at least two processes : the liberation of lactic
acid and its subsequent neutralisation during relaxation.
The formation of one gram of lactic acid from glycogen in
vitro is accompanied by the liberation of 190 calories. The
appearance of one gram of lactic acid in muscle is associated
with the evolution of total energy equivalent to nearly 400 cals.
The heat of neutralisation by bicarbonates or phosphates is far
too small to account for the excess. Meyerhof suggests that
the remainder may be due to the heat of dissociation of the
muscle proteins. This has recently received some experimental
confirmation from the work of Hartree and Hill, who have
shown that to keep the hydrogen ion concentration inside the
muscle within reasonable limits, we must assume the existence
of some buffer in it more effective than a bicarbonate or
phosphate solution. As in the case of blood, this is presum-
ably brought about by alkali proteinates capable of forming
neutral salts and undissociated protein.

The heat-production of oxidative recovery, on the other
hand, only accounts for a small fraction of the lactic acid which
disappears. Hartree and Hill estimated that in oxygen the
delayed heat-production is one and a half times the initial
heat-production, which represents about 400 cals., as stated,
per I grm. lactic acid. The combustion of i grm. of lactic acid
is accompanied by an evolution of 3661 calories in vitro. Since
now, according to Meyerhof, the disappearance of lactic acid
is associated with increase in glycogen in the muscle, it is
suggested that the lactic acid disappears partly by oxidation,
the energy so liberated appearing in part as heat and being

MUSCULAR CONTRACTION ii

utilised in part for the resynthesis of the remaining lactic
acid. About one-fifth of the lactic acid removed appears to
be oxidised. The sequence of phenomena is represented
diagrammatically in Fig. 2.

{d) The Nature of the Mechanism. — In considering the
heat-production of muscle in relation to the chemical events
of the contraction cycle, the muscle has been assumed to
respond under conditions in which no mechanical work is
done, so that all the energy set free at excitation is measured in
heat units. It has been shown, however, that we can determine
how much of this energy is available for the performance of
mechanical work in appropriate circumstances. The potential
energy available for the performance of mechanical work in a
single twitch is T//6. When the value of H for initial heat-
production (heat-production without recovery) in isometric
contraction is reduced to the same units, the ratio T//6H
expresses the absolute mechanical efficiency of muscle ; and
its value is found to approximate to unity (as high as 0-91 in a
series of A. V. Hill's experiments) in a suitable muscle such as
the frog's sartorius. This implies that the whole of the energy
liberated at excitation is free energy, i.e. capable, unlike heat,
of being transformed entirely into mechanical work.

Heat, as implied in the Second Law of Thermodynamics,
cannot be transferred into mechanical work without waste.
No heat engine can be more efficient than a reversible engine
working between the same temperature-limits, i.e. with a
maximal efficiency of T— T'/T on the gas thermometer scale.
Thus, in order that a frog's sartorius may have an efficiency of
0-25 (which is actually realisable), it would be necessary, on
the assumption that the energy was derived from the heat
produced, for the muscle to be raised to a temperature above
the boiling-point of water. The muscular machine is not
therefore a heat engine ; and the fact that all the energy set
free at excitation is available for doing mechanical work signifies
that the exciting substance must he liberated at the seat of tension ;
if it were not so, a large part of the energy liberated would be
irreversibly degraded into heat.

It is not difficult to construct a mechanical model to show

12 COMPARATIVE PHYSIOLOGY

how the appUcation of acid to a colloidal system may give rise
to a considerable quantity of mechanical energy. Thin strands
of catgut immersed in acid undergo quick and extensive
shortening ; the process is completely reversible when the
acid is removed, and can be repeated indefinitely. Shortening
in the length of a muscle-fibre could occur without a change in
volume by increased curvature of the surface ; as regards the
muscle as a whole, this as a matter of fact is what does happen.
Now Hartridge and Peters have shown that increase of hydrogen
ion concentration increases the surface tension at an oil-water
interface, and Neuerschloss finds that analogous phenomena
occur with lecithin sols. There is very good reason, based
on the penetration of dyes into living cells, to bear out the
conclusion that lipoid substances accumulate at the cell surfaces
and interfaces. Hill has shown that, provided that the muscle
is allowed to shorten before maximum tension has been
developed, the heat-production of a muscle which is released
after stimulation so as to shorten (without lifting a weight)
is less than in a rigidly isometric twitch ; this is so if the muscle
shortens while the processes which give rise to tension are
still at work, but if the muscle is liberated after maximum
tension has been attained there is no diminution of heat-
production. Were the heat-production uniformly distributed
through the substance of the muscle one can see no reason why
this should occur, since the muscle does not change its volume.
But if the heat production is located at definite interfaces, it
must tend to become smaller if the area of these interfaces is
reduced, as it presumably must be, when the muscle shortens.
(e) Electrical Phenomena in Muscle. — When the cut end of a
muscle is connected with one lead of a delicate galvanometer
and the uninjured surface with the other, there is found to be
a difference of potential between the two surfaces, the cut end
being negative to the uninjured surface of the muscle. When
the latter is stimulated, there is a diminution of this potential.
This diminution or '' negative variation " is referred to as
current of action^ and flows in the opposite direction to the
normal or demarcation current which is from the cut to the
uninjured surface of the resting muscle. It is not a specific

MUSCULAR CONTRACTION

attribute of the muscular mechanism, but is a phenomenon
shared by other excitable tissues. It has been the subject of
extensive research partly because of its practical application
to the diagnosis of heart disease (electrocardiography), partly
because it provides a very delicate means of detecting the
existence of an excitatory process, where mechanical appUances
fail, and to some extent because it has been customary in the
past to seek for an explanation of colloidal behaviour, and
therefore the processes which occur in the living organism, in

Fig. 3. — Time relations of the electrical variation in the isometric
response to two successive stimuli (after Fulton).

Responses of the intact gastrocnemius of a decerebrate frog at 16 5°.
Time indicated above, o"02 sec. The horizontal shadows from above
downwards are : the signal denoting the moment of the second stimulus
(make) ; the myograph ; the string of the galvanometer ; signal for the
first stimulus (break) ; line of zero tension, 29 mm. movement of the
myograph vertically being equal to 500 gms. tension. String tension 5 mm.
per m. V. , the magnification being 285. Stimuli delivered to the cut nerve,
the cathode being at a point I'g cm. from the entry of the nerve into the
muscle; stimuli just-maximal induction shocks. Initial tension 90 gms.
Frequency of the myograph 460 per sec.

terms of electrical phenomena rather than the stoichiometrical
relations which form the subject-matter of traditional
chemistry.

There is Httle doubt that the demarcation current arises
from the distribution of electrolytes in the muscle system ; its
relation to temperature, according to the work of Bernstein,
follows the thermodynamical relation which applies to the

COMPARATIVE PHYSIOLOGY

Sees.

concentration cell. The action-current occurs in denervated
(Adrian) as well as normal muscle. It travels as a wave along
the length of the muscle like the mechanical response itself.
Any excited region becomes momentarily electronegative to an
unexcited part. The existence of this potential indicates a
redistribution of ions within the system in which the potential
difference is developed. One explanation, offered by Mines,
is that the sudden concentration of hydrogen ions at the
sensitive surfaces of the fibres sets up there a condition which
may be likened to that of a concentration battery. There are,
however, two very cogent objections of a general nature to this

proposal. One is that the
electrical change in striped
muscle (see Fig. 3) is
practically complete before
the mechanical response
begins ; the negative varia-
tion occurs in the latent
period and begins con-
temporaneously with the
application of the stimulus.
It is thus probable that
tro-cardragramreToVherrtof Ho^ the electrical phenomenon

is associated with events
antecedent to those with which Hill's hypothesis is concerned,
since Hill has shown that the production of acid continues
up to the point at which maximum tension is developed. This
conclusion is strengthened by the close similarity (Chapter VIII)
between the electrical accompaniments and physical conditions
which are associated with the initiation of the excited state in
nerve and muscle. Therefore the electrical response may be
considered more conveniently in relation to excitation in the
more restricted sense (Chapter VIII). It is of interest in
connection with Mines' hypothesis, however, to note that the
phase of negative variation is often succeeded by a second
phase which is developed at the beginning of the mechanical
response. Possibly the origin of the electrical phenomena is
itself complex. It is difficult to believe that a concentration

MUSCULAR CONTRACTION 15

of hydrogen ions, such as Hill's hypothesis demands, produces
no measurable bio- electric effect.

(/) The Phenomenon of Tonus. — ^An aspect of muscular
contraction which is still very obscure is exemplified in a rather
striking manner by certain contractile mechanisms of inverte-
brates. In the living animal a skeletal muscle is not normally
relaxed completely, as in the isolated nerve-muscle preparation.
The partial contraction of skeletal muscle in situ is maintained
by the C.N.S. The opposite is the case with smooth muscle
which when isolated from the body remains in a state of tonus,
and in the intact animal is subject almost universally to control
by nerves whose action is inhibitory. Smooth muscle is
capable of maintaining this state of tonus for very long periods
without any appreciable sign of fatigue. Thus Parnas (19 10)
found that the adductor of the mollusc Dioxinia can keep its
shell closed for twenty or thirty days at a stretch against a
tension per sq. cm. of muscle attachment of about two and a
half million dynes — a tension exerted in virtue of the elastic
cushion which causes the shell valves to fly apart when the
muscle is relaxed. This can be brought about, in the freshwater
mussel Anodon at least, by stimulating the inhibitory nerves
from the pallial ganglion ; section of the nerve supply of the
adductors does not lead to relaxation. In Pecten a rather
curious phenomenon results from the co-operation of two
separate constituents of the adductor muscle, one composed
of striated fibres and the other of smooth muscle fibres. When
scallops (Pecten) are taken out of the water they usually give
one or two flaps of the shell- valves and then close tightly ; if
a solid object is interposed between the valves they close on it
Hke a vice. But if the foreign body is then made to slide out
of position, the valves remain set fast at the same degree of
closure. Uexkull (19 12) has shown that on cutting away the
smooth muscle, the remainder can be excited to contraction
by nervous stimulation ; but the contraction of the striated
fibres of the adductor persists only so long as stimulation lasts.
The motor portion (striated fibres) of the adductor serves to
bring the valves together rapidly, while the more slowly
reacting *' catch muscle " (smooth fibres) keeps the valves closed

1 6 COMPARATIVE PHYSIOLOGY

by its sustained tonus. A double neuromuscular mechanism
probably of the same type exists in the adductor of Astacus,
Homarus, Carcinus and other decapod Crustacea (Lapicque,
Keith Lucas). The economy of such an arrangement, which
combines the rapidity of action of striped muscle with the low
energy output of the tonus mechanism, is evident. Tonus is
not associated with increased metabolism, and cannot there-
fore be of the same nature as a low-grade tetanus.

The tonus of a catch muscle may be looked upon as a natural
form of isometric response. We are probably dealii(ig here not
with a change in the intensity factor but with the capacity factor
of the surface energy of the cell. It is possible to think of a
mechanism of isometric response by which a mechanical stress
sets up in some part of the protoplasmic system a change in
phase relations of the colloidal constituents such as to oppose it
by a virtual tension co- extensive with the maintenance of the
external force. The coagulation of liquid silk in stretching is
possibly analogous.

(g) Relation of the Muscle Cell to Electrolytes.— Certain
phenomena, notably those associated with the maintenance of
tone which has just been discussed, raise difficulty in the way
of any attempt to extend to plain muscle the conclusions
respecting the contractile mechanism in striped muscle. This
is also the case in considering the role of the hydrogen and
calcium ions respectively in the contractile process. The
effect of adding acid to a saline medium in which a preparation
of striped muscle is immersed is to produce contraction ; on
the other hand, it is found both for the cardiac muscle of the
vertebrate heart (frog, dogfish, skate), and the unstriped muscle
of the molluscan heart (Pecten) (Mines, 19 13), as well as for
various forms of mammalian plain muscle which have been
recently studied with great care by Lovatt Evans (1923), that
increased hydrogen ion concentration produces arrest, at first
reversibly and beyond a certain point irreversibly, in the relaxed
condition. The action of calcium on striped muscle has been
carefully studied by Overton (1904) and Mines (1912) ; in a
medium containing sodium ions but no calcium the muscle
displays rhythmical spontaneous twitching ; the addition of a

MUSCULAR CONTRACTION

small trace of calcium suffices to prevent these spontaneous
movements and bring about stoppage in the relaxed condition.
Similarly, removal of calcium brings about systolic stoppage
of the crustacean heart (Hogben). Mines also showed that
addition of Ca diminishes excitability of muscle towards
electrical currents of long duration. On the other hand, the
effect of increased calcium is to diminish relaxation in the plain
muscle of the crop of the fowl (Fienga) and the pharynx of
Aplysia (Hogben), while complete absence of calcium produces
diastolic arrest of the heart in Pecten (Mines), Raia, Scyllium

Fig. 5. — Effect of excess of calcium on the perfused heart of the
lobster, Homarus (Hogben).

and Rana (Mines) ; on the other hand, Lovatt Evans describes
diastolic arrest in the heart of Helix by excess of calcium.

So far we have little information with respect to the specific
role of anions, though certain of these have very characteristic
effects, as illustrated in the universal action of cyanides in
depressing oxidative processes. The fact that proteins appear
to behave as amphoteric electrolytes and are found in the cell
in general on the alkaline side of their isoelectric point, thus
existing as metallic proteinates, is a valid reason for attach-
ing special significance to the relation of kations to biological

i8 COMPARATIVE PHYSIOLOGY

processes. On the basis of a prolonged series of experiments
on the cardiac and striped muscle of the frog, the heart muscle
of Pecten and several species of elasmobranch fishes, Mines
has attempted the classification of kations under three headings,
combining, nomadic, and polarising ions. Mines postulated
(i) that the normal activity of the muscular apparatus depends
on the maintenance of a certain degree of permeability at some
cell-surface ; (2) that the permeability of the cell-membrane
depends partly on the chemical composition determined by the
combination {inter alia) of Ca (and Sr under experimental
conditions) with some constituent of the cell ; and partly on
the electrical potential between the two sides of the membrane
itself. This latter is supposed to be modified by {a) the ability
of certain ions (Na and K), the nomadic ions, to pass through

Fig. 6. — Effect of removal of magnesium (b — a) on the perfused heart
of Pecten (Mines, Journ. Physiol. 43, 1912).

it selectively ; and {h) the adsorption of certain other ions,
e.g, Mg, La . . . and Ce . . . to the surface itself, thereby
reducing or reversing its normal charge. The hydrogen ion
is regarded as acting sometimes in one way, sometimes in the
other.

On the whole the indications of recent work are distinctly
favourable to the role which Mines postulates for calcium.
The effects of the other polyvalent ions are more obscure, since
Mg . . , Ce . . . , La . . . , etc., agree with Ca in depressing the
striped muscle of the vertebrate, and that of the crustacean
heart ; while their action is opposite to that of Ca in a large
number of instances in the case of cardiac and plain muscle.
Thus in Pecten, removal of Mg, which is apparently essential
to the heart-beat, produces systolic arrest, while removal of

MUSCULAR CONTRACTION 19

Ca produces diastolic stoppage of the heart. In his conception
of the action of the " polarising " ions Mines was unduly
influenced by the work of Schulze and of Linder and Picton
on precipitation of arsenious sulphide sols. The relation
of valency to coagulation phenomena in hydrophile sols of
sulphur (Oden), lecithin (Neuerschloss), and gelatin (Loeb)
differs very considerably from that described by the Schulze-
Linder and Picton law, and gives evidence of antagonism
between divalent and monovalent ions which is suggestive in
relation to certain vital phenomena described below. Loeb's
work makes it very doubtful if the action of hydrion and trivalent
ions on membrane potential depend on the same mechanism,
and the degree of reversibility of the effects of one or the other
on the vertebrate heart reinforce this conclusion. As regards

^ ■__^mi*"'

,00 riqzo J\!u C/00 Cai ryzof^f^'^^

Fig. 7.— Effect of removal of calcium on the perfused heart of Pecten
(Mines, loc. cit.).

the ** nomadic '* ions. Mines' hypothesis gives no explanation
of the opposed action of Na and K in certain cases and their
relative potency in others ; moreover, the extent to which these
ions are actually capable of penetrating the muscle cell is an
open question.

Thecombining nature of theCa ion,on the other hand, finds
considerable confirmation in the work of Clark (19 12) on the
frog's heart, and is reinforced by very diverse lines of inquiry
into cell physiology which will be described in connexion with
ciliary activity. Clark showed that increase in the Ca/Na+K
ratio revives the amplitude of the frog's heart when it has
become diminished by prolonged perfusion, this loss being
associated with the removal of a substance of lipoid nature.
He suggests that Ca which precipitates lecithin, a normal
constituent of the cell membrane, maintains the normal semi-

20 COMPARATIVE PHYSIOLOGY

permeability of the muscle cell by its action on the colloidal
lipoid constituents of the surface layer. The work of Clowes
affords a crude model of how this effect might be produced.
In an oil-water emulsion addition of Ca salts produces a reversal
of phase so that the water becomes the internal and the oil the
continuous phase, i.e. so that the system as a whole becomes
impenetrable to water soluble substances. For a more realistic
conception the work of Neuerschloss on lecithin sols should
be consulted. If Ca determines the condition of the lipoid
constituents of the cell surfaces, the assumption that Na and K
are nomadic ions in Mines' sense provides for a ready explana-
tion of the widespread antagonism of Ca to Na as illustrated
by the work of Mines on striped muscle and that of Clark on the
vertebrate heart. Such antagonism might be regarded simply
as the opposition to the penetration of Na ions set up by surface
action of calcium.

An example drawn from the field of invertebrate physiolog}^
is afforded by the work of LiUie (1909) on the larvae of Areni-
cola. The larvae of this polychaete in isotonic solutions of the
chlorides of Na.,K., Li., and NH4 undergo contraction to about
half their normal length in a few seconds, and simultaneously
the yellow pigment enclosed in the cells of the organism
diffuses out into the medium, showing that the normal semi-
permeability of the cells has been suspended. Solutions of
the chlorides of Mg. . and Ca. . do not have this action either
as regards the contracture or discharge of colouring matter ;
and addition of a small quantity of Ca. . ions to a solution of
isotonic sodium chloride prevents the contracture and pigment
extrusion produced in a pure solution of the latter. On the
other hand lipoid-solvent anaesthetics such as chloroform
produce contracture and discharge even in pure magnesium
chloride. This antagonism is by no means confined to muscular
phenomena. Thus Loeb (19 12) showed that if eggs of the
Atlantic minnow (Fundulus) are placed in hypertonic sea- water
they remain alive for days, floating on the surface in an
apparently impermeable condition. If for sea- water hyper-
tonic NaCl is substituted, they sink to the bottom, undergo
shrinkage and rapidly perish. The addition of a small quantity

MUSCULAR CONTRACTION 21

of calcium to the sodium chloride solution prevents the
untoward effects of the latter alone and maintains the normal
impermeability of the egg-membrane. The work of Osterhout
on the electrical conductivity of plant tissues affords interesting
parallels.

Clowes has in fact constructed a model in which the inter-
stices of a partition of filter-paper fixed by rubber rings in a
U-tube are filled with an emulsion consisting of oil and a
saline medium containing sodium, potassium, and calcium
chlorides in roughly the same proportions as they occur in
living tissues. The conductivity of the artificial membrane
varies in the presence of electrolytes in the medium in a manner
closely analogous with the conductivity of protoplasmic sur-
faces in plant cells, as in Osterhout 's experiments. In pure
NaCl conductivity increases ; in pure CaCl2 it decreases, owing
presumably to reversible changes of phase in the interstices
of the filter paper.

Some light is thrown on the relation of electrolytes to the
contraction of muscle by a study of the concomitant electrical
phenomena. Thus in vertebrate heart- muscle the absence
of calcium does not prevent electrical changes after mechanical
response has ceased. To have a clear appreciation of the
relation of muscle to electrolytes it is essential to recognise
not only that muscle is the seat of a state of tension which in
appropriate circumstances results in mechanical work being
done, but that it is also the seat of a propagated disturbance
leading up to the events which condition this state of tension.
The foregoing treatment of the contractile mechanism has been
focussed on the events which succeed the explosive breakdown
of some substance intermediate between glycogen and lactic
acid ; but, as we have seen, the electrical phenomena must be
referred to phenomena independent of the nature of the
contraction process itself. The disturbance which travels
along the muscle initiating the breakdown of glycogen has so
many points in common with excitation and conduction in
nerve that there is excellent justification for extending to the
action-current of muscle an interpretation analogous to that
suggested for the corresponding phenomenon of nerve discussed

22 COMPARATIVE PHYSIOLOGY

in a later chapter. Excitation in nerve depends on surface
phenomena which are characteristically sensitive to changes in
the ionic composition of the external medium ; it is highly
probable therefore that the effect of any given ion may be
independently related to the initial and final stages in the
contraction cycle ; and we should therefore be wrong in
drawing from the different behaviour of striped and plain
muscle to electrolytes the conclusion that the contractile
mechanism is fundamentally different. This is clearly the
case with the calcium ion. Diastolic arrest of the heart of the
frog (Daly and Clark) in the absence of Ca ions is not due to
any failure of the excitatory mechanism : the electrical response
continues. On the other hand, diastolic arrest of the heart of
the lobster (Hogben) by excess of calcium ions is associated
with cessation of the electrical change, and is probably due to
the depressant action of calcium on the excitatory process.

Further Reading

Hill (191 3). The Absolute Mechanical Efficiency of Muscle. Journ.
Physiol. 46.

(19 14). The Heat Production in Prolonged Contraction. Ibid. 47,

(1922). The Mechanism of Muscular Contraction. Physiol. Reviews,

Hartree and Hill (1923). The Anaerobic Processes involved in Muscular
Activity. Journ. Physiol. 58.

Hogben (1925). Studies on the Comparative Physiology of Contractile
Tissues I. Quart. Journ. Exp. Physiol. 15 (for bibliography).

Evans and Underhill (1923). Studies on the Physiology of Plain Muscle.
Journ. Physiol. 58.

Mines (191 2). On the Relations to Electrolytes of the Hearts of different
Species of Animals. Journ. Physiol. 45.

(19 14). On Functional Analysis by the Action of Electrolytes. Ibid.

47.

CHAPTER II

CILIARY ACTIVITY, AMCEBOID MOTION AND COLOUR RESPONSE

In the previous chapter we have considered a form of response
which has several advantages for experimental treatment. The
activity of skeletal muscle is not spontaneous, it can be made
accessible at will. Further, muscular tissue can be obtained in
sufficient bulk to facilitate very considerably the measurement
of the energy changes which accompany its activity. On the
other hand; all that we know of the muscular mechanism is
of a statistical nature. In the case of ciliary, amoeboid, and
chromatophore movements, however, it is possible to make
direct observations on individual cellular units. To these
forms of response, which are more conveniently studied in the
lower organisms, we shall now turn, taking first of all ciliary
activity. Of this, our knowledge has been lately advanced by
the extensive investigations of Gray (i 922-1 924), which will
form the basis of the present treatment.

Ciliary Motion. — For a clear appreciation of the issues
raised by a consideration of the mechanism of ciliary motion,
a few remarks must be introduced concerning the contractile
rhythm of the ciliated cell. Favourable material for observa-
tions of this kind are afforded by the gills of the common
mussel Mytilns. The face and sides of each filament are
respectively lined with frontal and lateral ciliated cells, whose
movement maintains an efficient stream of water and mucus
(see Chapter V) easily detected by the naked eye if a little car-
mine is added to the medium. Microscopic observation shows
that the maintenance of these currents in a definite direction
is due to the characteristic manner in which the cilium moves.
The movement of a single cilium is divisible into two phases,
a very rapid forward or effective stroke, and a slower backward

COMPARATIVE PHYSIOLOGY

Forward or Effective Beat

Backward or Recovery Beat

or recovery stroke. The form of the beat suggests that it is
during the rapid effective stroke that the cilium performs v^ork
on the surrounding medium. At the conclusion of the forward
movement, it can be seen that the ciHum, which at the beginning
of the effective stroke is a more or less rigid rod moving forward
on a pivot at its base, becomes limp ; a stress is set up w^hich
starts at its base and is transmitted thence to its free end. For
the purpose of forming a working hypothesis, it may be assumed,
as suggested by Gray, '' that the energy which is expended by
the cilium is stored as tension energy." We are entitled to

surmise that this energy has
its origin in some chemical
compound either in the cilium
itself or in the cell to which
it is attached. The problem
to be faced is the elucidation
of the sequence of events by
which chemical energy is con-
verted into kinetic energy ;
or — as implied above — the
chemical processes by which
the state of tension in the
cilium is relieved.
We shall here accept Gray's hypothesis as a basis for
discussion. The results of the foregoing survey of muscular
contraction have led us to conclude that the physical changes
which result in contraction are associated with the production
of lactic acid from carbohydrate without intake of oxygen, and
that oxygen is employed in the recovery process to restore the
mechanism to its original condition. Though ciliated
epithelium is structurally very different from muscle, there are
two sets of considerations which suggest the possibility that a
similar sequence of chemical phenomena might be found to
underlie the changes of physical state, which in both cases
result in liberating contractile energy. One is that lactic
acid is an obligatory intermediary in the breakdown of carbo-
hydrates in all animal tissues. The other is that the physical
properties of proteins, being probably dependent on conditions

Fig. 8. — Diagram of ciliary motion
(after Gray).

CILIARY ACTIVITY 25

defined by the Doiman membrane equilibrium, are intimately
affected by the acidity of the medium with which they are in
contact.

Gray has outlined an hypothesis according to which the
chemical events of the contraction cycle in muscle and ciliated
cell are closely analogous. It may be considered under three
headings : («) the possible production of an acid substance
during the contractile process ; (b) the relation of oxygen
consumption to the events of activity and recovery ; (c) the
nature of the substances used up in the transformation of
chemical into kinetic energy by the cell.

The production of acid may be taken first. The only
method at present available for obtaining any evidence on this
question is derived from studying the relation of ciliary activity
to the ionic constituents of its surroundings. Ciliary motion
of Mytilus gill-filaments can be preserved for many hours in a
Van 't Hoff solution containing chlorides of Na, K, Ca; and Mg
in the same proportions as sea- water at a pH. about 7*8. On
addition of acid the cilia on the gill of Mytilus cease to move
when the hydrogen ion concentration of the solution reaches
a limiting value on the acid side of neutrality. What is
especially interesting is the way in which this stoppage is
brought about ; the cilia come to rest in an acid solution by
a gradual slowing of the rate, without reduction in the ampHtude
of the beat, till finally movement is arrested at the end of the
effective stroke, i.e. in the relaxed condition. This fact, while
suggesting that the acid does not exert its effect by damaging
the contractile fibrils of the cilium itself — since the amplitude
is not directly affected — at first sight points to the conclusion
that the effect is a surface one, concerned only with the rate at
which the excitation state is generated.

That the action is not a surface one, however, Gray has
proved by several lines of experimentation. The first depends
upon the fact that the weak organic lipoid-soluble acids
penetrate the lipoid membrane of the cell more readily than do
strong acids. Similarly, weak bases like ammonium hydrate
are more penetrative than strong bases like sodium hydroxide.
If the cessation of ciliary motion in acid medium were a surface

26 COMPARATIVE PHYSIOLOGY

effect the pH. limits with both types of acid would be expected
to be the same. A number of experiments were carried out
by Gray to determine the critical concentration of hydrogen
ions in the external medium which would produce arrest in
one minute. The results are given below.

Acid.

Critical pH.

Acid.

Critical pH.

Hydrochloric

•• 3'4

Formic

.. 4-0

Sulphuric . .

•• 3*1

Acetic

.. 4*8

Nitric

•• 3'4

Butyric

. . 5*2

It will be noticed that the mineral acids are of practically
uniform efficiency ; while the fatty acids form a series of which
the higher members are more efficient, all being more efficient
than the mineral acids. This result strongly suggests that the
acid enters the cell, since the effect is related not to the absolute
hydrogen ion concentration of the medium but the penetrative
power of the acid employed. The value of those observations
are, moreover, reinforced by considering the phenomenon of
recovery of pieces of gills exposed to acid for the same length of
time, when transferred to sea- water made alkaline with weak
and strong alkalis. The time (in minutes) for recovery in
such an experiment is given in the following protocol : —

Sea-water,

Sea-water +NaOH,

Sea -water +NH4OH

pH. 7-8.

pH. 8-4.

Movement begins

Full recovery . .

The experiment may be varied by determining the time for
recovery on transference to sea-water raised to a known
alkalinity by addition of NaOH and NH4OH, respectively, as

given belo

NH4OH

pH.

min.

9*5

9*2

8-7

8-5

8-4

NaOH.

8-10

10-12

12-15

To sum up in Gray's own words, *' The weak acids which
enter the cell are more efficient inhibitors of ciliary movement
than the strong acids which do not enter readily, and con-
versely the weak alkalis are much more efficient restoratives
than the strong alkalis.'*

CILIARY ACTIVITY

Circumstantial evidence also pointing to the conclusion that
the acid enters the cells is gained from a study of the effects of
removal of calcium. Lillie (1900) originally showed that the
toxic effect of pure sodium salts on ciliary movement, as on
muscular contraction and on irritability in nerve, is prevented
by the presence of the alkaline earth metals. In the case of
the ciliated cell either magnesium or calcium will serve in
adequate proportions to maintain the normal semipermeability
of the cell membrane, but magnesium must be present to
ensure recovery (according to Gray) after exposure to a solution

60 6 4 6 8 7-2 7-6 6 0 B '^
Hydrogen Ion Concentration P^

Fig. 9. — Relation of hydrogen ion concentration to ciliary movement

(after Gray).

containing no divalent ions. On the other hand, if calcium
is replaced by an equivalent amount of magnesium, cessation
of ciHary movement is brought about unless the pH. is kept
well above absolute neutrality.

In other words, the cell is more susceptible to the action of
hydrogen ions if the calcium ions, which confer upon the
membrane its normal semipermeability, are removed. More
direct evidence that the acid enters the cell is obtained by
intra vitam staining with an indicator, such as neutral red.
Before drawing conclusions from these data one other fact
may be mentioned. At the critical point the cilia are brought

28 COMPARATIVE PHYSIOLOGY

to rest at the end of the forward stroke. If the hydrogen ion
concentration is increased suddenly to a much lower pH.
than that which suffices to bring about cessation of movement,
the cilia come to rest not at the end of the forward stroke but
in the contracted condition, i.e. at the end of the recovery stroke.
We see then that the cilia are brought to rest in the relaxed
condition at a certain degree of acidity depending on the
penetration of the cell by the acid ; but if a greater quantity of
acid is present, they are brought to rest in the contracted
condition. This Gray interprets as due to the fact that the
cell itself is more permeable to acid than the cilium. In hyper-
tonic solutions arrest is brought about by a reduction not of
the rate, as with stoppage in acid medium, but of the ampHtude
of the beat ; and the fact that the amplitude is affected by an
increase in the osmotic pressure of the external medium
suggests that withdrawal of water from the cell interferes not
with the periodic liberation of energy but with some part of
the contractile mechanism.

To bring all these phenomena within the scope of a
single hypothesis Gray has suggested that the cilium flies
forward owing to imbibition resulting from periodic libera-
tion of acid. This is not essentially very different from a
suggestion put forward many years earlier by Schafer. But
direct proof of the production of acid in ciliary movement is
lacking.

Turning now to the second aspect of the problem, that is,
the relation of oxygen to the contractile mechanism, there is
now satisfactory evidence of a close analogy between the ciliary
and muscular tissues. The mechanical activity of cilia can
be treated quantitatively by timing across a standard distance
of gill filament or other ciliated epithelium the movement of a
minute circular plate of platinum. The oxygen consumption
of ciliated epithelium can be conveniently measured by the
Barcroft manometer. When the rate of oxygen consumption
and mechanical activity are plotted for various temperatures
and reduced to the same scale of ordinates, the curves
correspond closely (Fig. lo), showing that the rate at which
oxygen is consumed is normally a function of the mechanical

CILIARY ACTIVITY

activity of the cell. But the same is true of a rhythmically
contractile muscle such as that of the heart.

Engelmann maintained many years ago that the ciliated
epithelium of the frog's oesophagus remains active for as long
as two hours in an atmosphere of hydrogen, but regarded this

TEMPERATURE

Fig. 10. — The relation of oxygen consumption and mechanical activity of
cilia to temperature (after Gray).

as evidence that the cell stored oxygen in some intramolecular
form. By spectroscopic examination of water in which
haemoglobin was dissolved, Gray has proved conclusively that
in the gills of Mytilus ciliary movement continues long after
all oxygen has been removed from the medium. When
ultimately brought to rest by this means recovery is very slow,

30 COMPARATIVE PHYSIOLOGY

though intra vitrant staining with methylene blue shows that
the diffusion of oxygen back into the depleted tissues is rapid.
The conclusion seems justified that while the ciliary mechanism
is ultimately dependent on the presence of oxygen, oxygen is
not a necessary factor in the contractile process, but only for
the maintenance of the requisite conditions for prolonged
activity. In other words, oxygen would seem to be concerned
with the recovery process, as in muscle. When movement is
abolished by deprivation of calcium or increased osmotic
pressure of the medium, the amount of oxygen consumed is
not affected for a considerable period of time.

By means of Barcroft's manometer it is also possible to
obtain light on the nature of the substance on which a supply
of chemical energy available for transformation into mechanical
energy depends. If the ultimate fate of the substance is to
be oxidised, the " respiratory quotient " or ratio of CO2
evolved to O2 consumed must be unity for carbohydrates, and
about 07 in the case of fats and proteins. Gray finds that the
respiratory quotient for ciliary activity is about 0'8, which
implies that the substance used in ciliary activity is not
exclusively or mainly of carbohydrate nature, a conclusion which
agrees with the failure of micro- chemical m^ethods to detect
glycogen in the cells. Since fats are not stored by the ciliated
epithelial cell, it would appear that the substance on whose
energy the contractile mechanism depends is of a protein
nature.

To sum up, the mechanism of ciliary activity may be
analysed on the basis of Gray's experiments into three com-
ponents ; a reaction by which free energy is liberated from some
chemical reserve, sensitive to monovalent ions (especially the
hydrogen ion), a mechanism by which this free energy is
transformed into mechanical energy, presence of calcium and
a. certain osmotic pressure in the external medium being
essential to its efficient working ; and, finally, an oxidative
recovery process which is necessary only for sustained activity.

Amoeboid Movement. — Amoeboid movement is the character-
istic means of progression in certain Protista (Rhizopoda,
Mycetozoa, etc.) and the wandering cells present in many

CILIARY ACTIVITY 31

Metazoa, Reference to the behaviour of leucocytes, phago-
cytes, etc., will be found in works on bacteriology and medical
aspects of physiology. Attention will here be confined to the
phenomena of amoeboid movement, as they can be studied in
free-living forms. For quantitative treatment of amoeboid
activity the only accessible criterion of the energy of movement
is the rate at which the animal progresses. This has been
studied in relation to changes in external conditions in a
recent series of investigations by Pantin (1923-1925). Marine
amoebae were used in these experiments. Two species
(referred to as type A and B) were used, both being of the
'' Umax " form which progresses by protrusion of a single
anterior pseudopodium, thus tending, in the absence of external
interference, to move in a straight line. If the conditions of
the medium were kept constant, Pantin found that the velocity
of an individual amoeba was constant to within 5 per cent, for
periods of as long as twenty- four hours, and if the conditions
are changed without irreversibly damaging the organism, the
original velocity is regained when the initial state of affairs
is restored. This velocity is readily observed by timing with
a stop-watch the period which is required for an amoeba to
traverse a given number of divisions of the micrometer scale
of the microscope ocular.

By this method Pantin has described the relation of amoeboid
activity to osmotic pressure, temperature, and hydrogen ion
concentration of the external medium. The amoebae studied
were, like many other contractile mechanisms, very insensitive
to OH ions. On the other hand, amoeboid activity is reduced
with mineral acids to zero immediately on the acid side of
neutrality, to be precise at a pH. of 6' 8. As long as the
hydrogen ion concentration is kept at a level above pH. 4*0
the stoppage is reversible, movement recommencing when the
organisms are transferred to an alkaline medium. But below
this pH. limit cytolysis occurs. The velocity of progression
increases continuously up to pH. 9*6 ; and the velocity curves
for hydrochloric, sulphuric, butyric, lactic, and acetic acids
closely agree (Fig. 11).

It has already been suggested that production of acid in the

COMPARATIVE PHYSIOLOGY

cilium, as in
mechanism.

muscle, is an essential part of the contractile
The same appears to be the case with amoeboid

Velocity curves ob-
tained from Type B
amoebae in a medium
acidified with citric
or tartaric acids.. Th«
citric curve marked

thus □ -; was

obtained with car-
bonate-free sea water.

HCl ....+

Citric O and Q

Tartaric ... ^

pH&O

Four velocity
curves obtained suc-
cessively from tlie
same Type B amoeba,.
Each curve taken
with a fresh acid, after
recovery. in "outside
sea water,"

Curve 1 with HCl.
„ 2 ,. Citric.
,, 3 „ HiSO«
,, 4 ,, Tartaric.

pH6*0 7-0 60 5-0

Fig. II. — Relation of hydrogen ion concentration to the velocity of
Amceboid movement (after Pantin).

movement. Bethe, Harvey, and others have used dyes which
stain the living cells and are themselves indicators of hydrogen

CILIARY ACTIVITY 33

ion concentration, to determine the penetration of cells by
acids and alkalis. Pantin has applied the same method of
attack to test whether there is an increase of hydrogen ion
concentration within the cell during pseudopodium formation.
The dye successfully employed for this purpose was neutral
red, which has its turning point in the neighbourhood of
absolute neutrality (becoming red on the acid side), is non-
toxic, readily absorbed by protoplasm, and has a negligible
protein error. The colorometric determination of proto-
plasmic pH. is facihtated by focussing an image of the standard
indicators on the observation slide by aid of the achromatic
condenser.

By this method Pantin found that the hydrogen ion concen-
tration of the normal endoplasm and ectoplasm respectively
corresponded to a pH. of about 7*6 and 7*2 (in the resting
amceba the difference between ectoplasm and endoplasm
tended to be rather less than during active progression), but
the formation of a pseudopodium is preceded by a local
intensification of the red tint, indicating increased hydrogen
ion concentration ; while retraction of the pseudopodium is
accompanied by a local reduction of hydrogen ion
concentration.

Such considerations have led the above-named author to
suggest an hypothesis of the mechanism of amoeboid action on
rather different lines from those advocated by previous workers.
These for the most part have sought to interpret pseudopodium
formation as the consequence of a local lowering of surface
tension in a system which is for the purpose regarded as a
fluid drop in an immiscible medium. There are in the surface
phenomena exhibited by simple fluid systems remarkable
analogies to the behaviour of the amoeba. BayHss quotes from
Rhumbler a picturesque phenomenon which occurs when
particles of glass fibre coated with shellac are brought into
contact with a globule of chloroform suspended in water.
The chloroform drop first envelops the particle, then, after
sufficient time has elapsed to permit of the shellac being
dissolved away, rejects the glass remainder, thus resembling
an amoeba swallowing a diatom and defaecating the insoluble

34 COMPARATIVE PHYSIOLOGY

materials of the cell- wall. The disadvantage of surface tension
theories is that the results of microdissection studies reveal the
gelated character of the ectoplasmic membrane. In the
'' Umax " form the body of the organism is to be regarded,
according to Pantin, as a contracting tube of ectoplasmic gel
closed at its hind end. Endoplasm is continuously streaming
from a point of liquefaction on the inner side of the ectoplasm
at the tail end, while the fluid ectoplasm at the anterior extremity
is continuously adding to the contracting tube by becoming
gelated round the sides of the advancing pseudopodium. The
outer layer of the amoeba is apparently a lipoid-protein system
in the gel state, undergoing liquefaction, as Pantin 's observa-
tions indicate, in consequence of a local increase of hydrogen
ion concentration. Pantin argues that it is natural to regard
the swelling and liquefaction of the advancing tip of the
pseudopodium as a process of imbibition. By surface
coagulation the tube is always adding to itself in front,
and always contracting by synerezis (withdrav/al of water)
as the hydrogen ion concentration falls in the region furthest
behind.

The variation of amoeboid activity with temperature is
seen in Fig. 12. The optimum temperature for the amoebae
which Pantin studied is remarkably low — about 20° C. The
effects of changes in temperature below 15® C. are completely
reversible. But when raised from a lower temperature to one
near the optimum or one above it, the previous rate is not
regained when the amoeba is brought back to a lower tempera-
ture. One of the factors upon which amoeboid activity depends
is therefore some substance or structural arrangement of
substances — probably an enzyme — which is progressively
destroyed near or above the optimum temperature. If the
temperature is raised from, say, 10° to T° (above the optimum)
and then rapidly lowered to 10°, the resulting velocity has a
precisely similar relation to the initial velocity at 10°, as has
the observed velocity at T° to the velocity that would have
been observed at T° if no destruction had taken place, since
the amount of destruction is identical in both cases. In this
way the observed temperature curve can be extrapolated

CILIARY ACTIVITY 35

(Fig. 12). Analogous phenomena are seen in the heart of
the crab Maia. The low temperature at which irreversible
changes in the physical state of protoplasm are produced in
many marine organisms is well worth studying in its ecological
bearing on problems of distribution.

The relation of electrolytes to the amoeba has recently been
studied with microdissection technique by Chambers and
Reznikoff (1925). Their results present a remarkably close

30 ©c /

VELOCITVatr !

VELOCITYatIO' / .

30°C.

Fig. 12. — Effect of temperature on velocity of Amoeboid movement

(Pantin).

analogy to the phenomena described by Clowes in oil- water
systems. When solutions of NaCl and KCl are injected into
the living amoeba a liquefaction of protoplasm takes place in
the neighbourhood of the injected area. When the chlorides of
calcium or magnesium are injected there is, however, an
immediate solidification of the adjacent protoplasm, the
affected portion being pinched off when calcium chloride is
used. In suitable proportions mixtures of sodium and calcium
chlorides have neither of these effects (iCa : 52Na). There

36 COMPARATIVE PHYSIOLOGY

is complete antagonism between the sodium and calcium ions.
This is also true of potassium and calcium, but the critical
ratio is not the same.

When amoebas are torn in water by Chamber's technique
the surface is rapidly repaired after outflow of some of the cell
contents. Slight tearing in NaCl or KCl prevents any repair

M . .

in solutions of — concentration. With weaker solutions

13
repair takes place with increasing ease in NaCl as the dilution

M M

is increased. In solutions of > ^r CaCl2 or > — MgCl2

repair does not occur ; the surface sets into a solid mass.
The antagonism between the monovalent and divalent ions is

M M

seen again in this process : in a solution of NaCl — . CaCl2 —

or of KCl . CaClo — b z repair of a torn surface takes

13 ^ 208 416 ^

place in the normal manner.

Pantin (1925) has also investigated the relation of electrolytes

to the velocity of amoeboid movement, and has obtained

somewhat analogous results. The marine amoeba is destroyed

in isotonic solutions of the chlorides of the four principal

kations present in sea water {i.e. sodium, magnesium, potassium,

and calcium). For actual movement to take place calcium

must be present. The addition of a minute trace of calcium

to a solution of sodium or potassium alone permits movement

to continue for a short time. Movement can be prolonged

indefinitely by increasing to a certain optimum value the

proportion of calcium to sodium in a solution containing these

two kations alone. In this relationship calcium can be

replaced by strontium, but not by any other kation. In a

mixture of sodium and calcium, sodium may be replaced by

any alkali metal ; but it is found that the optimum ratio of

calcium to monovalent ion with respect to the velocity of

amoeboid movement differs for different alkali metals, increasing

with the atomic weight of the latter. In the presence of a trace

of calcium the addition of magnesium to a certain optimum

CILIARY ACTIVITY 37

ratio antagonises the effect of pure solutions of the monovalent
ions on the velocity of amceboid movement, the optimum ratio
differing again with the monovalent ion used. Nevertheless,
amoeboid movement, which, as stated, ceases in pure solutions
of either calcium or magnesium, can be prolonged in mixtures of
the two, this antagonism being evidently of a different nature
from the antagonism of either to monovalent ions.

The Pigmentaxy Effector System.— Colour response in cold-
blooded vertebrates and Crustacea has been regarded by some
investigators as a form of amoeboid activity. The power to
respond to environmental influences in changes in bodily
colour is also met with in molluscs where the pigmentary
organs are muscular structures. In vertebrates and Crustacea
colour response is brought about by the dispersion or aggrega-
tion of pigment granules in special cells or groups of cells,
which may be described as pigmentary effectors to distinguish
them from the chromatophores of molluscs. The pigmentary
effectors of vertebrates (reptiles, amphibia, and fishes) are
unicellular organs charged with pigment granules. Only one
type has been subjected to careful experimental investigation —
the melanophores or black pigment cells. According to some
authors the concentration of pigment granules in the
"contracted" condition is associated with the active with-
drawal of the cell processes, which are supposed to He in
preformed lymph spaces. The majority of recent observers,
with the notable exception of Graham Kerr, however, do not
regard the process as analogous to amoeboid movement, but
consider that the pigment granules stream to and fro within
fixed cell processes.

The co-ordination of vertebrate pigmentary response with
environmental agencies will be considered in a later chapter
(Chapter VII). Here we are concerned chiefly with the
properties of the pigmentary effector as an isolated organ. If
the contracile mechanism is one of dispersion and aggregation
of granules, it is possible that the visible responses of pig-
mentary effectors are more directly related to known properties
of colloid systems than those of any of the mechanisms so far
discussed. Up to the present there has been little quantitative

COMPARATIVE PHYSIOLOGY

work on the responses of pigmentary effectors. To Spaeth
(1914-1918), however, we owe the elaboration of an ingenious
method by which such observations may be made. Spaeth
fixes scales of the Atlantic minnow Fundulus in a glass container,
through which water circulates, on the mechanical stage of a
microscope. By connecting the adjusting screw of the latter
to a lever writing upon a smoked surface, the migration of the
pigment granules is recorded as the observer adjusts the screw
so that the distal extremity of the pigment mass is in alignment
with a particular scale mark on the micrometer (Fig. 13).
Melanophores of Fundulus do not react to visible Hght, but
they respond to electrical and chemical stimulation. Single

make shocks do
not induce complete
contraction, but a
tetanising current of
moderate intensity
produces a complete
contraction, in which
the contraction phase
occupies 25-30
seconds and relaxa-
tion about 65-90
seconds. There is
a latent period of
about 5 sees, before contraction begins. Single make shocks
successively applied produce a summated effect if the inter-
vening period is not more than 2-3 sees.

According to Spaeth, neutral electrolytes exert a specific
physiological action upon fish melanophores. Sodium chloride
(normal saline) promotes melanophore expansion ; while
potassium salts and salts of the alkaline earths bring about
immediate contraction. Distilled water always produces this
latter response. By combining these salts in such proportions
that their characteristic effects are balanced, Spaeth produces
a medium in which, after a preliminary contraction, the melano-
phores display partial expansion, so that, when a pharmaco-
logical reagent is dissolved in such a balanced solution, the

Fig. 13. — Melanophoies of Fundulus (Spaeth).
(a) expanded ; (b) contracted.

CILIARY ACTIVITY 39

specific action of the added substance in either direction can
be conveniently observed. He recommends the following
formula for Fundulus : 6 vols. N/io NaCl, i vol. N/ 10 KCl,
0*35 vol. N/io CaClg.

The most remarkable result which emerges in this con-
nection from Spaeth's experiments is the effect of barium salts,
which, as is well known, exercise a very characteristic excitatory
action upon true contractile tissues. After treatment of scales
for ten minutes in N/io BaCl2 no sign of recovery may appear
for as much as half an hour, during which the melanophores
remain completely contracted. After this period of quiescence,
the melanophores in the periphery of the scale abruptly expand

Fig. 14. — Melanophores of Fundulus (Spaeth).

Effect of adrenaline {a) on the partially contracted pigmentary effectors (b) ;
effect of adrenaline after previous immersion in ergotoxine (c).

and contract almost at once. After a brief interval a second
expansion wave appears at the periphery, and melanophores
lying nearer the centre begin to behave in a similar manner.
This time, however, the peripheral melanophores expand more
completely. Eventually after about an hour the wave of
rhythmic expansion and contraction extends to the centre of
the scale so as to include all the melanophores, which pulsate
between the extremes of contraction and complete expansion.
This pulsation may continue for several hours.

The Crustacean chromatophore consists essentially of a
centre, branches radiating therefrom in a luxuriantly ramifying
system. In the aduh forms of most Crustacea which have been

40 COMPARATIVE PHYSIOLOGY

thoroughly investigated, the chromatophores are highly complex
structures of a multicellular or coenocytic character. However,
the structure of the Crustacean chromatophore differs very
widely in the different groups and in different regions of the
same individual. A necessary prerequisite to quantitative
treatment in this case is a clear recognition of the normal
behaviour of the pigmentary effector organs. The phenomena
of colour response in Crustacea may be illustrated by reference
to the Schizopod Macromysis and the Decapod Hippolyte, as
described by Gamble and Keeble (i 900-1 906).

In Macromysis the chromatophores are localised in definite
regions lying for the most part in a deep situation where their
action is rendered visible by the translucent nature of the
integument. The majority of the chromatophores contain a
large amount of dark brown pigment and a smaller quantity
of a substance which by reflected light has a white or yellow
hue, but appears greyish by transmitted Hght. On a sandy
shore the animal appears transparent and colourless or greyish
in tint ; amid dark surroundings and in deeper water, its
colour deepens and a pattern becomes manifest in the form of
paired arborisations of yellowish-white upon a brown back-
ground corresponding to the chromatophore centres. In the
transparent, colourless, or greyish form the brown and yellow
pigment is withdrawn into the central body of the chromato-
phores ; in the dark condition the brown and yellow pigments
diffuse along separate paths throughout the interlacing tracery
of branches ; the yellow pigment does not mix with the brown,
and it responds to stimuli at a different rate.

In Hippolyte we have to draw a distinction between two
modes of colour response occurring normally in nature. By
day the adult prawns are of a reddish- brown or bright green
hue according to the tint of the seaweeds to which they are
attached. Whatever the diurnal colour of Hippolyte may be,
however, it changes at or soon after nightfall to a beautifully
transparent blue or greenish-blue. The depth of the nocturnal
blue corresponds to the intensity of the diurnal colour ; that
is to say, dark brown prawns become deep blue and light ones
become pale blue. The nocturnal tint ordinarily persists till

CILIARY ACTIVITY 41

daybreak, when the green and brown varieties represent inter-
changeable states. Green Hippolytes placed on brown weeds
may conserve the green colour for about a week before assuming
the brown condition. The extent and rapidity of the change
varies very much from one individual to another. There are
three chromatophore pigments involved : red, yellow, and
blue. In addition to the more deep-seated elements in the
nerve cord, viscera, and principal muscles, the epidermis is
itself in Decapod Crustacea richly charged with chromato-
phores. These consist of groups of as many as eight pear-
shaped cells with tubular branches, each containing pigment of
one colour, yellow or red. Blue pigment is present in the
tracts along which the red pigment flows, but is also found in
branched cells lying among the pear-shaped cells which contain
the yellow or red pigments. The reddish-brown tones are asso-
ciated with the extension of the red pigment into the tubular
processes of the chromatophores. In the green condition the
yellow pigment extends throughout the branches of the cells in
which it occurs, but the red pigment is retracted, its place being
occupied by the blue substance. Finally, the nocturnal blue
colour is produced, when both yellow and red pigments are
withdrawn from the branches, but the blue substance diffuses
along them.

The bulk of recorded observations on Crustacean chromato-
phores are concerned with the more inmiediate problem of
analysing what forces incident to the animal's environment
promote colour response, and the channels through which the
different stimuli gain access. Both Crustacean and Piscine
chromatophores, however, offer admirable material for studying
the physical chemistry of the cell.

Whereas in Fishes and Amphibia, though possibly not in
Reptiles, light and dark backgrounds appear to act merely by
modifying the intensity of illumination, in the case of Crustacea
it would seem from Gamble and Keeble's researches that
something besides intensity is involved. To detail the chief
colour phases induced by various light conditions in Macromysis
and Hippolyte, it will be necessary to call attention to a
difference between the two genera. Freshly- caught Mysids

42 COMPARATIVE PHYSIOLOGY

and some shrimps, e.g. Palaemon, generally have their pigments
moderately expanded ; and this represents the condition of
the chromatophores of the same animals when kept in the light
in glass containers. In Hippolyte, on the other hand, the
yellow and red pigments are always fully expanded in the day
and fully contracted at night. In darkness the yellow, red, and
brown pigments of both Hippolyte and Macromysis become
completely retracted into their chromatophore centres, which
then appear as minute dots so widely separated that they do not
interfere with the translucency of the body as a whole ; the
time required to produce the contraction ranges from a few
minutes to an hour or two, varying considerably with the time
of day and the condition of the animal.

Paradoxically enough, the same result is brought about by
exposing Crustaceans to a pure white reflecting background,
such as a porcelain surface. When subjected to such treat-
ment there occurs a condition essentially the same as the
nocturnal state accompanied in Hippolyte by diffusion of the
blue pigment already referred to, and conditioned in both
Macromysis and Hippolyte by the complete retraction of the
other pigments into the chromatophore centres. This response
is exceedingly rapid, being accomplished in about a minute or
even less. The effect of a dull black, Hght-absorbing back-
ground is no less surprising. In Hippolyte itself it cannot be
demonstrated because the red and yellow pigments are, as
stated, always expanded fully in daylight ; but in Macromysis
and Palaemon, where expansion on a neutral background is
incomplete, the normal accompaniment of transferring indi-
viduals to dark-bottomed vessels is to induce a condition of more
extreme outward migration of the chromatophore pigments
from the chromatophore centres into the branching processes,
so that the animal assumes in the case of Macromysis nigra
a black aspect and in the case of Palaemon serratus a speckled
brown coloration. These colour phases persist so long as
the light conditions remain unchanged. At nightfall the
pigments which have expanded in the " dark background
phase," are withdrawn ; but the expanded phase is again
resumed on the morrow. The white background contracted

CILIARY ACTIVITY 43

phase and the black background expanded phase are produced
no matter whether the Hght intensity is high or low, variable
or constant.

The chromatophores of isolated strips of the integument
of the shrimp react to bright light by expansion. This suggests
that there is a direct response of the chromatophores in
Crustacea. After amputation of the eyes in Macromysis or
Palaemon the chromatophore pigments, instead of assuming
the contracted phase, expand when the anim^al is placed on a
light-reflecting bottom ; in fact, in the case of Palaemon, a
deep chocolate colour even darker than the normal expanded
condition associated in the intact animal with a black back-
ground is produced. But the contracted phase characteristic
of the nocturnal state occurs at night in Hippolyte even after
the eyes have been removed. These data permit the inference
that the nocturnal- diurnal rhythm depends on the direct
reactivity of the chromatophores to the intensity of illumination,
and the white and black background responses are determined
by events conditioned by photic stimulation of the organs of
vision.

The supposition that the eye exerts its influence upon
colour response in virtue of the intensity of illumination alone
is adequate to explain the peculiar reversal of response in seeing
and blinded salamander larvse as described by Babak and
Laurens. But it offers no explanation of the fact that a bright
white background which scatters light produces in the intact
shrimp a response similar to darkness, while a dull black back-
ground which reduces the intensity of illumination acts in the
direction of intensifying the diurnal state. Some other factor
besides intensity of illumination is involved, and the
transcending importance of this factor is clearly seen in experi-
ments which are described as follows in Keeble and Gamble's
own words : —

" To show how, compared with this influence of background,
that of light intensity is of small importance, we mention the
following experiment which we have often performed with
Hippolyte and Macromysis. We take with us on a collecting
expedition four jars, two of glass wrapped round with black

44 COMPARATIVE PHYSIOLOGY

cloth and two of porcelain. For one of the former and two of
the latter we provide covers of white or black paper pierced
with several pin-holes. As the animals are caught they are
distributed between the four jars. When brought into the
laboratory and examined, the animals in " open " porcelain
and in " pin-hole " porcelain are found to be in the fully
contracted phase. Now the light intensity in the open jars
is far higher than that to which the animals are subjected before
capture, and the intensity in the pin-hole jars is probably far
lower. Any effect which a raised or lowered light intensity
might produce is swamped by the background effect. The
latter is of such a nature that an absorbing (black) background
induces expansion, and a scattering (white) background produces
complete contraction, and these effects are produced in the
faintest (pin-hole) Hght."

Apart from the intensity of illumination, an essential
difference between the condition of a shrimp exposed to black
(absorbing) or white (scattering) background is to be sought in
the direction of the incident rays impinging upon the eyes, which
as we have seen are the receptor organs involved in the back-
ground response. When placed against a mirror background
which reflects incident light mainly in one direction, the animals
display a state of partial expansion and not the complete
contraction characteristic of exposure to a white surface which
scatters light rays in all directions. It would appear, then, to
quote the same authors once again, that *' it is, in some way or
other, the ratio direct/scattered light which determines " the
background reaction. This interpretation impHes a dorsi-
ventral differentiation of the photoreceptive elements of the
eye. But whatever be the nature of this dorsiventrality, it
does not reside in any permanent structural arrangement of
the retinal elements, for illumination of the animal from below
against a white surface calls forth the characteristic transparency
of the contracted phase.

Keeble and Gamble have, in addition to the foregoing
analysis of the more rapid responses to light, darkness, and
light or dull backgrounds, attempted to analyse the factors
involved in the production of the different colour forms of

CILIARY ACTIVITY 45

Hippolyte. Hippolyte, as we have seen, displays a remarkably
close resemblance of bodily colour to the hue of the seaweeds
on which it resides. While this similarity is striking, there are
in reaHty two sharply defined categories of colour forms, greens
on the one hand, and reds or browns on the other. The green
state corresponds to an " expansion " of the yellow pigment,
the red being withdrawn into the cell bodies whose branches
are suffused with the blue " nocturnal " substance ; and the
brown or red colour forms display " extension " of the red
pigment into the cell branches ; we should be tempted to
presume that sympathetic coloration depends on the wave-
length of the light reflected respectively by green or brown
weeds ; this interpretation is not, however, compatible with the
results of experiments in which the effect of light of different
wave-lengths has been put to the test.

In carrying out experiments on the effect of monochromatic
light of different qualities, it is evident from what has already
been stated that the direction of the incident rays must be taken
into account by controlling each experiment by comparison of
the response evoked both by absorbing and dispersing back-
grounds. Monochromatic lights, whether red, yellow, green,
or blue, acting on a black background, produce pigment
expansion, the yellow pigment reacting more readily than the
red. Monochromatic lights, whether red, yellow, green, or
blue, acting on a white background, induce, like white light,
the transparent condition. Thus the quality of the light does
not seem to be the effective factor which evokes '' sympathetic "
coloration in Hippolyte. It would, therefore, seem that the
slowly effected transition from one colour to another, when
individuals are transferred from green to red weeds or vice
versa, is in all probability a reaction to the different intensities
and not to the predominant wave-lengths of the light reflected
from the differently coloured surfaces of the green, red, or
brown weeds. It should be added, however, that the diurnal-
nocturnal rhythm of colour change is in part an automatic
process whose periodicity is independent of external stimuli
once there is an appreciable tendency to contract by night or
expand by day, when the animals are kept in continuous light

46 COMPARATIVE PHYSIOLOGY

or darkness. It is to be hoped that in the near future the
physicochemical properties of chromatophores will be the
subject of investigations analogous to those on ciliary and
amoeboid activity described at the beginning of this chapter.

Further Reading

Gray (1922-3). The Mechanism of Ciliary Movement, I-IV. Proc.
Roy. Soc. B. 93-95.

Pantin (1924). On the Physiology of Amoeboid Movement, II. Brit.
Journ. Exp. Biol. i.

Chambers and Reznikoff. Proc. Soc. Exp. Biol. Med. 22, 1925.

Spaeth (19 16). Responses of Single Melanophores to Electrical Stimula-
tion. Am. Journ. Physiol. 41.

Keeble and Gamble (1904-6). The Colour Physiology of the Higher
Crustacea. Phil. Trans. Roy. Soc.

CHAPTER III

SECRETION

The last form of response which we shall consider is secretion.
The phenomena of secretion have been studied far more from
the standpoint of their relation to other functions than with a
view to throwing light on the mechanism by which the gland
cell elaborates and discharges its specific products of its activity.
In vertebrates secretion has been studied especially in relation
to the digestive processes (which will be discussed in a later
chapter), to the elimination of waste products (renal secretion),
heat regulation (sweat glands), and reproduction (the mammary
apparatus, oviducal secretions, etc.). Slime-secreting skin
glands have also received some attention. But the study of
other animal groups opens up an immense variety of secretory
mechanisms, some of which are doubtless of considerable
bionomic interest as a means of defence or attack. Our know-
ledge of a few of these will be briefly indicated by reference to
such phenomena as bioluminescence, gas secretion, silk forma-
tion, etc. Others such as the secretion of ink in cephalopods,
of calcareous matter forming the tubes of sedentary animals or
the exoskeletons of mobile forms, etc., can only be mentioned
in passing. The importance of mucous glands in relation to
digestion will be touched on under that heading. There is a
vast field for biochemical research awaiting inquiry in connexion
with the comparative physiology of secretion.

Renal Secretion.— Concerning the mechanism of secretory
activity, perhaps the study of the vertebrate kidney has yielded
more valuable information than any other secretory organ.
The characteristic differentiation of each glandular element of
the vertebrate kidney into a capsular portion and " renal

48 COMPARATIVE PHYSIOLOGY

tubule " has focussed an immense amount of research on the
attempt to define the role of these two structures in the process
by which non-volatile waste products are eliminated from the
body. Since the secretion includes practically all the crystal-
loidal constituents of the blood plasma, though these are
neither individually nor collectively present in the same con-
centration as that in which they occur in the blood, it is
attractive to explore the possibility that the whole process is
composed of two distinct phases, one consisting of the separa-
tion of a fluid identical in crystalloidal constitution with the
blood — a deproteinised plasma filtrate formed by exudation
from the capillary walls of the glomerulus, and a subsequent
process involving the specific activity of the glandular epi-
thelium of the tubule and resulting in the differential con-
centration of each of the crystalloidal constituents, either by
secretion or by reabsorption or both.

Some colour is lent to this interpretation by a line of
experimentation which is possible owing to the peculiar
vascular arrangements which exist in the amphibian kidney,
where the glomerular blood supply is derived from the aorta
(via the renal arteries), while the renal portal veins only supply
the tubules. It is thus possible in the frog, to cut off the
glomerular blood supply while leaving intact that of the tubules,
or to perfuse separately the vessels of the capsules and renal
tubules.

When the glomerular blood-supply is cut off the secretion
of urine stops. It is still possible, by injection of a solution of
urea, to induce a flow of acid urine containing chlorides,
sulphates, and urea, but the quantity is small compared with
the diuresis produced by the same method in the intact kidney.
Perfusion of the arterial (glomerular) supply of the kidney
with Ringers' solution at normal aortic pressure (20-24 cm.
in the frog) induces a copious flow of urine (of somewhat
more dilute concentration than the perfusion fluid). Perfusion
of the renal portal system at normal venous pressure does not
result in production of urine.

Further support for the belief that the initial stage in the
process of renal secretion is the separation of a deproteinised

SECRETION

plasma filtrate is derived by studying the physical conditions
under which urine is produced. If the capsular membrane
is impermeable to the colloidal constituents of the blood, work
must be done (i) against the osmotic pressure of the proteins,
etc., in removing a filtrate of identical crystalloidal composition ;
(2) in driving the filtrate along the narrow lumen of the tubules
at the observed rate of flow. From the first consideration it
follows that no secretion of urine can take place when the blood-
pressure falls to a value below the osmotic pressure of the
serum proteins. As a matter of fact, secretion of urine has long
been known to cease in the mammal when the arterial blood-
pressure falls below about 35 mm. of mercury. Starling
(1899) was the first to recognise the theoretical significance of
this fact, and on comparing the osmotic pressure of a
crystalloidal filtrate separated from blood by a gelatine filter
with the osmotic pressure of the original serum, showed that
the osmotic pressure of the blood colloids is actually about
30 mm. of mercury. Conversely, urinary secretion can be
prevented by increasing the pressure in the ureter so that the
diflFerence between the pressure of the arterial blood and that
of the fluid in the capsule is of the same order as the osmotic
pressure of the serum proteins. Later it was shown by
Barcroft and Straub (191 1) that a great increase of urinary
secretion follows replacement of normal blood by a suspension
of red corpuscles in Ringers' solution without any rise in
arterial pressure or increase in oxygen consumption by the
kidney.

However, the possibility that the first stage of renal activity
is a process of simple filtration deriving its energy from the
heart beat, does not throw any light on the essentially secretory
function of the kidney, namely, that of modifying the composi-
tion of the filtrate in such a way that the concentration of each
of the crystalloidal constituents is finally diff"erent from its
concentration in the plasma. To effect this, work must be
done by the epithelial portion of the tubule. Barcroft and
Brodie found in experiments on the gaseous metabolism of the
active kidney that the respiratory quotient (ratio of CO2 given
off to oxygen absorbed) is practically unity ; i.e. that the activity

50 COMPARATIVE PHYSIOLOGY

of the kidneys involves oxidation of carbohydrate material,
as in the case of striped muscle.

It has been seen in a previous chapter that activity of muscle
and cilia depends on an anaerobic reaction, oxidative processes
being associated only with the recovery process. That this
is true of secretory response is not easily demonstrated in an
organ which secretes more or less continuously like the kidney.
But in the case of salivary secretion, which can be controlled
for the exigencies of experiment by means of its nervous
connexions, Barcroft and Piper (19 12) were able to demonstrate
that the maximal rate of oxygen consumption occurred
appreciably later than maximal rate of flow from the gland ;
and that the increased oxygen consumption accompanying
secretory activity in the salivary gland of the mammal outlasts
by a considerable interval the cessation of active secretion.
This clearly points to the probability that in the mechanism
of secretion as in that of muscular and ciliary activity oxida-
tive reactions are especially characteristic of the recovery
phase.

For a detailed discussion of the evidence concerning the
way in which the adjustment of concentration of each of the
constituents of the urine is carried out in the renal tubules, the
reader should consult Cushny's monograph on the secretion
of urine. Two rival hypotheses have been advocated.
According to one the renal tubules selectively absorb from
the filtrate ; according to the other they differentially remove
from the blood appropriate quantities of the urinary consti-
tuents. It is certain that dyes are excreted by the cells of the
renal tubules, and innumerable researches of this kind have been
prosecuted in connexion with the segmental excretory organs
and malpighian tubes of Arthropods and the nephridia of
worms. But it is difficult to see what light they throw on the
normal elimination of salts and nitrogenous waste from the
body fluids. The fluid of the excretory glands of molluscs
is particularly rich in uric acid and other purine bodies. A
filtration mechanism in these animals is, however, excluded
by the very low blood-pressures which are found in all Inverte-
brates except Cephalopods. It is possible that reabsorption and

SECRETION

secretion in the ordinary sense each play a part in the activity
of the renal tubules of the Vertebrate, and evidence relating to
one constituent of the urine is not necessarily valid as regards
another.

Nitrogenous Excretion (after v. der Heyde, J. Biol. Chem. 46).

Animal.

Mammal (Homo)
Amphibian (Rana)
Teleost (Lophius)
Elasmobranch (Mustelus)

Total (non-protein) |

N2 (mg.

%).

Blood,

Urine,

181

Blood,

Urine,

Blood,

Urine,

830

Blood,

1000

Urine,

420

Urea (mg. %).

Uric

acid (mg. %).

iS'5

225

213

258

215

1720

729

Looking at the alternative of secretion and reabsorption
from the standpoint of comparative physiology, it seems likely
that a thorough comparison of the state of affairs in fishes with
that which exists in land vertebrates would well repay investiga-
tion. For in fishes the blood from the heart encounters the
resistance of the gill capillaries before reaching the kidneys by
way of the dorsal aorta ; the pressure of blood in the dorsal
aorta is therefore extremely low even as compared with that of
the frog. Exact data are unfortunately lacking, but it seems
unlikely that there is ever a blood-pressure in the renal vessels
of the fish sufficient to overcome the osmotic pressure of the
blood colloids. If this is really the case, filtration clearly plays
no part in the renal function of fishes ; and the hypothesis of
reabsorption cannot be applied to them. The amount of
urine secreted is, according to Denis, about i c.c. per kilo
body weight per hour in the elasmobranch ; and destruction
of the cord, which would presumably lower the blood-pressure,
does not reduce the output. At the same time, the distribution
of nitrogenous (non-protein) materials in blood and urine is
not very different in Teleosts from the condition found in land
vertebrates (see table). In Elasmobranchs an anomalous
feature is the large amount of urea present in the blood and

52 COMPARATIVE PHYSIOLOGY

tissues. According to Baglioni (1903) and Mines (1912),
addition of urea to the saline medium is essential to the mainte-
nance of activity in the artificially-perfused heart of Elasmo-
branchs.

Bioluminescence. — Of all forms of secretion which are of
interest from a bionomic standpoint none have received so
much attention as the phenomena of light-production.
Bioluminescence is widely distributed throughout the animal
kingdom, and is also found among bacteria and fungi. It is
met with in many Protozoa {e.g. Noctiluca) ; in representatives
of all groups of Coelenterates, several Polyzoa {e.g.
Membranipora) ; Polychaetes {e.g. Chsetopterus) ; Ophiuroids,
Urochorda, many Crustacea {e.g. Cypridina), Myriapods and
Insects {e.g. Lampyris), many Cephalopods, a few other
molluscs {e.g. Pholas), and, among Vertebrates, in some genera
of fishes.

The actual intensity of illumination produced by animals
is hardly ever such as to bear the scrutiny of the light-adapted
eye. In some forms the oxidative process involved proceeds
more or less continuously, independently of stimulation, as in
bacteria and a few fish. More generally it is an intermittent
form of activity, luminescence occurring only in response to
stimulation. In some cases, e.g. among Ctenophores, light
has an inhibiting influence. Previous exposure to illumination
is not essential to photogenic response in animals. The most
interesting physical aspect of the phenomenon is that the light
emitted is cold. Within the limits of experimental error a
hundred per cent, of the radiant energy emitted is light. Thus,
though feeble in quantity, light-production in animals is
prodigiously more efficient than any ordinary artificial source
of illumination.

The organs concerned with light-production may be divided
into two categories, according as the production of luminous
material is extra-cellular or intracellular. In the former
category, exemplified by Pholas, Chaetopterus, Cypridina, and
Myriapods, are included those cases in which a slimy secretion
containing the phosphorescent substance is produced by
unicellular glands. These may be either scattered diflPusely

SECRETION 53

over the whole surface of the body (as in Chaetopterus) ;
restricted to various regions — narrow bands on the siphon
and mantle and a pair of triangular regions near the retractor
muscle in Pholas ; or more definitely localised, as in Cypridina,
which possesses on the upper lip above the mouth, a luminous
gland, made up of spindle-shaped cells each opening by a
separate pore with a kind of valve.

Intracellular light-production is characteristic of the more
specialised types. In this category are the complex photogenic
organs of insects, some Crustacea, cephalopods and fishes.
In the fireflies the photogenic organ develops from the fat-
body and consists of a mass of granular luminescent cells
abundantly supplied with nerve fibres and tracheae, and
enveloped above by a layer of supposedly reflecting cells which
probably scatter incident light on account of the crystals of
xanthin, urates, or other purine derivatives contained in them.
In some of the shrimps (Sergestes), Cephalopods (Sepia),
and fishes (Stomias) the photogenic organs are still more
elaborate and diffusely scattered over the body. In these
forms the photogenic organ possesses not only a reflector
behind the photogenic cells, but in addition a cuticular lens, and
in one genus of fishes it is even endowed with an iris diaphragm
by which flash effects are produced. In general, photogenic
organs are richly innervated, and there is no special phyletic
or bionomic significance to be attached to the particular
structural arrangement. Both intracellular and extracellular
types occur within the Crustacea.

For the production of light by the organism it has long been
known that oxygen and moisture are essential. Robert Boyle
(1665) showed that rotten wood infested with luminous fungi
ceases to glow if placed in a vacuum ; and Spallanzani lifted
the subject out of the domain of vitalism by showing that
the luminous material of medusae, if dried, would emit light
when water was added. Pyrosoma, Pholas, Phyllirhoe, fire-
flies, Pyrophorus, copepods, ostracods, pennatulids can all
be desiccated without destruction of the luminescent material.
Desiccated ostracods and copepods will luminesce in the
presence of moisture after being kept for several years. Though

54 COMPARATIVE PHYSIOLOGY

total absence of oxygen brings about cessation of luminescence
in the moistened material, a very low tension of oxygen (3 mm.
in the case of Cypridina) suffices to maintain the emission of
light. Measurements of the electrical conductivity of lumi-
nescent solutions prepared in this way by E. N. Harvey show
no progressive increase in H-ion concentration indicative of
CO2 production.

Bioluminescence, like the secretion of digestive fluids,
affords an instance of the production of the substances known
as enzymes for specific ends. And a few words may here be
inserted with reference to the conception of an enzyme.
According to their velocities, molecular reactions may be
divided into two categories, those involved in the familiar
methods of volumetric analysis, where the combination of
reacting substances is practically instantaneous — ^.^.precipita-
tion of barium salts in the presence of SO4 ions or the union
of hydrochloric acid gas with ammonia — and a large class of
organic reactions, such as the saponification of esters or inversion
of cane sugar, for which an appreciable interval must be allowed
to elapse if a condition of equilibrium is to be attained.
Reactions in the living body are for the m^ost part of the latter
type. It is a familiar fact that many reactions which ordinarily
proceed at an immeasurably slow rate take place with great
rapidity in the presence of substances known as catalysts,
which though influencing the velocity of the reaction do not
enter into the composition of the end-products or in general
shift the point of equilibrium for a given set of conditions.
The facility with which the organism is able to disintegrate
highly stable compounds is due to the agency of a special type
of catalysts known as enzymes, characterised especially by their
extrem^e instability (which is illustrated by the fact that
practically all enzymes are rapidly destroyed at temperatures
well below the boiling-point of water), and by the extremely
minute quantities in which they act. Thus the preparation
known as rennet, of which the active constituent represents
a very small fraction of the total bulk, is able to clot 400,000
times its own weight of the milk protein caseinogen.

Of the general catalytic properties of enzymes, the most

SECRETION 55

important for an understanding of chemical equilibrium in the
organism is the complete reversibility of their action, first
shown by Croft Hill (1898). The same is true of inorganic
catalysts ; thus in the preparation of ethyl acetate, to take an
illustration from the historic researches of Berthelot and Pean
de Saint Gilles, if molar equivalents of acid and alcohol are
mixed, equilibrium is reached when two- thirds of the acid and
alcohol are converted into ester, and precisely the same
equilibrium point is reached when a molar solution of ethyl
acetate is subjected to hydrolysis. It is usual to regard both
reactions as contemporaneous and equiUbrium as the equalisa-
tion of their respective velocities. The catalyst, in this case
usually a mineral acid, acts by accelerating both reactions ;
the actual point of equilibrium is unchanged and the total
energy of the system unaffected. This is also the case with
enzyme catalysis. Thus maltase accelerates both reactions
symbolised by maltose $ ^/-glucose. As with inorganic
catalysts {e.g. finely divided platinum) an enzyme is often
reclaimable at the end of a reaction ; this is well illustrated in
the case to be discussed immediately. Another aspect of
enzyme reaction illustrated by the phenomenon of biolumi-
nescence is specificity. Any enzyme is able to catalyse only
a restricted range of reactions, usually all of a well-defined type.
The recognition of separate enzymes in a tissue-extract depends
on the possibility of differential destruction of one or other of
the catalytic activities of the extract. When different tissue-
extracts have different optima of temperature or hydrogen-
ion concentration, etc., for catalysing precisely the same reaction,
there is good reason to believe that their enzymes are distinct
chemical entities.

The term " enzyme " was introduced by Kiihne (1878) to
obviate the confusion resulting from the use of the older term
" ferment," a term originally used to include the activities of
micro-organisms like yeast. Pasteur's researches led to a
distinction being drawn between the fermenting action of
gastric juice ('* unorganised ferment ") and that of micro-
organisms (" organised ferments "). This confusion was
finally dissipated when Buchner (1903) extracted from crushed

56 COMPARATIVE PHYSIOLOGY

yeast cells an " unorganised " ferment, zymase, capable of
effecting the same transformation previously identified with the
intact organisms.

The foundations of modern knowledge of bioluminescence
were laid by Dubois in the eighties. Dubois' studies elicited
the following facts ; (i) when a preparation of the photogenic
organ of Pyrophorus is dipped in hot water, light-production
is irreversibly stopped ; (ii) when the fresh organ is ground
up the mass only glows for a short time ; (iii) if a hot-water
extract of the gland is added to a cold-water extract in which
all luminescence has ceased the production of light begins
again. This was also found to be the case in Pholas. Dubois
therefore advanced the hypothesis that the hot-water extract
contains a heat-stable substance, luceferm, which is oxidised
in the presence of an enzyme, heifer ase, present in the cold-
water extract. The latter can only go on glowing as long as
any luciferin remains unoxidised. Boiling the mixed extract
brings about cessation of light-production. Luciferin can be
obtained from Pholas by heating the viscous luminous secre-
tion to 70° C. or extracting for some hours in 90 per cent,
alchohol.

Similar phenomena have been studied by Newton Harvey
in luminous bacteria, fireflies and other forms, especially in
the ostracod, Cypridina. The process involved has its parallel
in other biochemical reactions which can be carried out in
vitro. The glucoside, escuHn, in horse-chestnut bark is
oxidised in presence of haemoglobin and hydrogen peroxide
with production of light, and luminescence is also characteristic
of the oxidation of pyrogallol by the peroxidase of potato or
turnip juice. The oxidation product of luciferin can be re-
converted into its precursor by inorganic reducing agents.
We must recognise, therefore, three entities : {a) luciferin, an
oxidisable substance, heat-stable and dialysable ; {h) luci-
ferase, non- dialysable, destroyed by temperatures above 60°
and by tryptic digestion, an enzyme, probably of protein-
like constitution ; {e) oxyluciferin formed from {a) in pre-
sence of {h) with emission of light. As stated, oxyluciferin
can be reconverted by reducing agents into luciferin.

SECRETION 57

Conversely the luciferin of Pholas can be oxidised with emission
of light by various inorganic oxidising agents, e.g, hydrogen
peroxide, potassium permanganate. This is not true of the
luciferin of Cypridina, which differs in other respects from
that of Pholas, being in particular more heat-stable. Whether
the luciferases of different animals are identical is not wholly
certain ; in Cypridina the presence of luciferase is confined
to the photogenic organ. Apparently luciferase — or luci-
ferases, if there are several of such substances — belongs to
the category of enzymes known as oxidases which catalyse
other oxidative processes in the body.

To sum up, photogenic response in at least three groups
of luminous animals, beetles, molluscs, and Crustacea, involves
the interaction of two readily separable components which
have entirely different chemical properties. Probably, how-
ever the " luciferins " and '' luciferases " of different animals
are not identical. Thus the luciferin of Cypridina differs
from that of Pholas in that it cannot be oxidised with light-
production by H2O2 and KMnOi- The luciferase of Cypri-
dina differs from that of Pholas in being less readily destroyed
by lipoid solvents. When the luciferin of Cypridina is
oxidised, no dissolution of the molecule takes place, since the
product can be readily reconverted into its precursor by such
reducing agents as H2S or nascent hydrogen. The luci-
ferases are destroyed by proteoclastic enzymes, and are to be
regarded as oxidases either themselves of protein constitution
or adsorbed to proteins in solution.

Electric Organs.— Under the heading of secretion reference
may conveniently be made to the electrical organs which are
present in several genera of fishes, since in one genus at least
(Malapterurus) the electric organ is a modified gland, though
in other cases the cellular elements have been derived from
muscle fibres. These phenomena serve to draw attention to
a property which gland cells share in common with other
excitable tissues, namely an electrical response accompanying
excitation. The existence of an electrical charge accompany-
ing excitation is well illustrated by the experiments of Anrep
and Harris on pancreatic secretion induced by secretin, and

58 COMPARATIVE PHYSIOLOGY

therefore independent of any concomitant nervous disturb-
ance. In the electrical organ, the P.D. v/hich accompanies
muscular or glandular activity has been elaborated into a
weapon of defence or aggression. The E.M.F. of each cellular
element in a gland or muscle is of very small dimensions. In
a frog's sartorious, where the parallel arrangement of the
elemicnts eliminates any summation of potentials, the cur-
rents of action or injury are rarely greater than 0*05 volt. But
it has been shown that by arranging several frog's muscles
in series with the cut surface of one opposed to the uninjured
surface of the other, a summation of the potential difference
due to injury can be obtained up to a volt or more.

Electric organs have been studied in three genera of fishes
— a Mediterranean ray, Torpedo, Malapterurus, a catfish
of North African rivers, and Gymnotus, the electric eel of
the tropical zone of S. America and Africa. They consist
essentially of disc-like cellular elements richly supplied with
nerve-endings on one surface and arranged in columns in a
manner reminiscent of Volta's pile. In Torpedo there is an
electrical organ on either side of the head, consisting of hori-
zontal rov/s of discs which represent functionally modified
muscle fibres. In Malapterurus the electrical organ is
developed from unicellular glands situated beneath the skin
in the middle region of the body. In Gymnotus it is located
in the tail. The columns are arranged transversely in
Torpedo and longitudinally in the other two genera. The
shock delivered by the electric organ of Torpedo is equivalent
to that obtained from about thirty Daniell cells. Bernstein
and Tschermack, who studied the variation of the P.D. with
cooling and heating, found it to be (within physiological
limits) directly proportional to the absolute temperature, as
would be expected from Nernst's formula for the E.M.F. of a
concentration cell.

Secretion of Poisons.— Secretion of poisonous substances
which are undoubtedly means of attack or defence is met with
throughout the animal kingdom ; and their study raises a
number of points of general biological interest, notably the
questions of anaphylaxis and immunity. Many poisonous

SECRETION 59

substances of colloidal nature produced by animal tissues
when introduced into the body in sublethal doses call forth
a condition of reduced susceptibility as compared with their
effect after an initial dose. This condition is known as immu-
nity. While this mode of reaction to toxic substances is in
some cases of undoubted utility to the organism, it must be borne
in mind that the phenomena of immunity are by no means
exclusively of an " adaptive '* significance. Thus the blood
of the crayfish when injected into the mouse renders the latter
immune against the venom of scorpions, though the crayfish
itself is more susceptible to scorpion venom than the mouse.
Frog serum injected into the body cavity of the crab specifi-
cally protects the latter against the poisonous secretion of the
pedicellari^e of certain Echinoderms. The mechanism of
immunity is exceedingly complex ; in addition to immunity
to poisons produced by secreting glands of larger animals or
by micro-organisms the blood of some animals produces
specific lysins which directly destroy micro-organisms, and
substances, opsonins, which favour phagoc3rtic activity.

The poisons produced by the nematocysts of Coelenterates
illustrate a phenomenon which may be described as the reverse
of immunity and is referred to as anaphylaxis. If extract of
the tentacles of a sea anemone are injected into a dog intense
vascular congestion in the viscera resulting in death follows
after a few hours. When a sublethal dose is given, it is
found that the administration of a very much smaller quantity
of the poison after a certain minimum period of about ten days
has much more severe consequences. This supersensitive-
ness, according to Richet, involves the co-operation of two
factors. From extracts of sea anemone tentacles two toxic
substances can be obtained, one congestin when injected first
increases the sensitivity of the dog to a subsequent dose of
the other known as thalassin. If the order is reversed thalassin
acts as an antitoxin, diminishing the sensitivity to the poisonous
action of congestin. Neither the phenomena of increased
sensitivity or anaphylaxis nor of decreased sensitivity or
immunity to specific poisons are, as the important work of Dale,
Gunn, and others has shown, due simply to the production of

6o COMPARATIVE PHYSIOLOGY

substances analogous to or antagonistic to the effects of the
poison in the blood itself. The supersensitivity in the one case
and increased resistance in the other in part reside in the cells
of the organ affected. Thus Dale (19 12) has shown that the
isolated virgin interns of the guinea pig from an animal treated
with horse serum as the anaphylactic reagent shows specific
supersensitiveness to the reagent. Again, Gunn and Heath-
cote (1921) have shown the greater resistance of the cat as
compared with the rabbit to cobra venom is shown by isolated
organs of the two species. The minimal lethal dose per
kilo of cobra venom for the cat is twenty times that for the
rabbit. Both the isolated heart and gut muscle of the cat
can withstand much higher doses of cobra venom than corre-
sponding preparations from the rabbit. Of the venoms of
Arthropods the most important are those of the spiders, scor-
pions, and hymenoptera. Nearly all spiders possess poison
glands connected with the mouth parts, the poison being
instantly fatal to the small animals on which they prey. The
toxin is destroyed by heat, and like the venom of viperine snakes
displays both coagulant and haemolytic properties with refer-
ence to vertebrate blood. Scorpion venom more closely re-
sembles the venom of the Colubrine snakes {vide infra). The
poison glands of bees contain at least three toxic substances,
one of which possesses haemol3^ic properties and acts on the
nerve centres, but like viperine venom produces marked local
effects. A particularly interesting case from the pharmaco-
logical standpoint is the presence of a substance allied to
tyramine (parahydroxyphenylethylamine) in the salivary secre-
tion of the cephalopod (Henze). By means of it the cuttle-
fish paralyses its decapod prey. Tyramine is closely related
to tyrosine, as is the latter to the melanic secretion of the ink
sac in cephalopods. It is one of a class of compounds allied
to " adrenaline," the hormone of the mammalian suprarenal
glands, prepared synthetically by Barger and Dale. Adrena-
line itself has been isolated by Abel and Macht in association
with an alkaloid bufagin having an action akin to digitalis in
the poisonous parotid and skin glands of the toad, Bufo agua.
The venoms of snakes may be divided into two groups.

SECRETION 6i

That of the colubrine forms (including the cobras) has, Hke
the saHvary secretion of the leech (hirudin) and mosquito,
anticoagulant action on the blood, and produces death chiefly
by asphyxiation through paralysis of the respiratory centre.
Cobra venom is also hsemolytic, i.e. it disintegrates the red
blood corpuscles. The venom of the viperine snakes (including
rattlesnakes) is also haemolytic, and differs from cobra venom
in the more marked local inflammatory reaction and the pre-
sence of a substance which promotes coagulation of the blood ;
its effect on the nerve centres is less marked. Alkaloids are
present in snake venoms, but their toxicity is slight. The
poisonous action is due to constituents of a simple protein-
like structure. Mammals may be artificially immunised by
injection of sublethal doses in increasing quantities. The
serum of animals so treated may be used as anti- toxin. It
is interesting to note that the blood of snakes (which are
immune to snake venoms), and that of animals like the hedge-
hog, which prey on snakes and also possess a high degree of
natural immunity, is toxic to animals which are relatively
susceptible to the poisonous effects of snake-bite.

The phenomena of poison secretion are illustrated in all
large groups of the animal kingdom, and a more extensive
discussion would take us beyond the scope of the present
volume.

Secretion o! Acids. — ^An aspect of the comparative physiology
of secretion which opens up a fascinating field for experimental
investigation that may throw light on the bionomics of some
boring animals is the production of strong acids by the diges-
tive glands of certain animals. More than half a century ago
Troschel noticed that the gasteropod, Dolium galea, squirts
from its mouth a liquid of strongly acidic reaction capable of
producing effervescence on coming into contact with the lime-
stone of the soil. This fluid is the secretion of the salivary
glands and contains as much as 4 per cent, free sulphuric acid
and about 0*5 per cent, hydrochloric acid. Schulz (1905)
has also studied the phenomenon of acid secretion in an opistho-
branch, Pleurobranchia meckelii. This animal not only
ejects an acidic fluid from its pharynx but produces with its

62 COMPARATIVE PHYSIOLOGY

skin glands a very acid slime. Little is known of the
mechanism by which such a very high concentration of acid
can be attained in the cells which secrete it. We are still
in the dark as to the mechanism of acid secretion in the stomach
of the mammal. It may be presumed that the source of
sulphuric acid must be the sulphur of either proteins or sul-
phates of the food. The bionomic significance of acid secre-
tion in the lower organisms has been interpreted as a means of
softening the calcareous skeletons of animal prey or as a pro-
tective device. Neither interpretation is proved ; but the
subject would well repay investigation.

The Secretion of Gas. — Among those fishes (Ganoids
and Teleosts) in which a gas bladder is present a peculiar
form of secretion is often met with. The bladder (homo-
logous with the lungs of air-breathing Vertebrates) is in some
cases a true respiratory organ. In those teleosts which dwell
in deep water and habitually move over a considerable range
of depths, it subserves the function of facilitating movements
from one level to another by altering the specific gravity of
the fish. It was shown over a century ago that in those teleosts
which have a closed duct (physoclistous condition) the bladder
contains oxygen only. The oxygen content of the bladder
changes during inflation and deflation. Though the tension
of oxygen in sea-water is about a fifth of one atmosphere and
in the capillaries of the bladder considerably less, the tension
in the gas bladder may rise to about a hundred atmospheres.
Secretion and absorption of oxygen (in the physoclistous
forms) provide an auxiliary mechanism to promote sinking
and rising in the water. Deflation in physoclistous fishes is
apparently effected by means of the oval, a thin- walled area
on the dorsal wall of the bladder overlying the cardinal
sinuses and enclosed by a sphincter. That oxygen is re-
absorbed into the blood by diffusion through the oval is
indicated by the fact that during active gas-secretion the
sphincter is completely closed so that the thin-walled area is
invisible, while it opens widely when the bladder is com-
pletely inflated.

It is fairly certain that the inflation is brought about by the

SECRETION 63

activity of the gas gland. The form of the gland varies in
different fishes. Typically it is a local proliferation of the
lining epithelium of the bladder, elsewhere composed of
squamous cells. The cells of the glandular region are
columnar where in contact with the endothelial lining of a
peculiar arrangement of blood-vessels called the rete mirabile.
This consists of a closely packed bunch of fine capillaries ;
its essential feature is the juxtaposition and intermingling of
capillaries carrying blood in both directions. It has been
shown by Bohr that the gas gland, like the salivary gland
of the miammal, is under nervous control ; and Dreser states
that oxygen secretion can be induced by the action of pilo-
carpine which provokes activity of salivary and skin glands.
In appearance the gas gland is a bright red diffuse mass typi-
cally situated on the ventral wall of the bladder. Little is
known of its intimate mechanism. Artificial activation of
the gland can be achieved by attaching a weight to the fish.
A load of about five per cent, of the body weight is convenient
for the purpose. In experiments upon the Pollack, Wood-
land found that when the weight is first attached, the fish sinks
immediately to the bottom of the tank. Soon it begins to swim
upwards by active movements. After about twelve hours
or rather less it regains, owing presumably to the activity of
the gas gland, its normal quiescence completely. It is able
to float easily near the surface without the aid of caudal move-
ments. Exposure of the viscera at this stage reveals the
bladder in a distended condition. If instead the weight is
removed and the fish allowed to resume its former state, it
floats at first to the surface ; and has to swim downwards
vigorously in order to keep away from the surface.

Further Reading

Swale Vincent. An Introduction to the Study of Secretion. Arnold.

Bernstein. Elektro biologic. Vieweg.

CusHNY. The Secretion of Urine. Longmans, Green.

Newton Har\^y. The Nature of Animal I>ight. Lippincott.

Calmette. Venoms. Bale & Danielsson.

CHAPTER IV

RESPIRATION

From the knowledge which we have gained concerning the
nature of response in animals we have learned that the utilisa-
tion of oxygen is an essential feature of the processes by
which effector mechanisms are restored to their original
condition after a state of induced activity. The intake of
oxygen by the organism will be our first consideration in
dealing with the sources of vital energy, since the necessity
of oxygen for the maintenance of animal life is a universal
phenomenon.

Under the heading of respiration it is customary to include
not only the intake of oxygen, but the removal of carbon dioxide
which is associated with it. Except where we have to deal
with tissues like ciliated epithelium in immediate contact with
the external world, the intake of oxygen involves : (i) the
absorption of oxygen by the tissues from the body fluids ;
(ii) the absorption of oxygen by the body fluids from the
external medium. It will be convenient (though less logical)
to treat the latter before passing on to the special arrange-
ments for the transport of oxygen to the tissues by the
body fluids, and CO2 from the tissues to the external
medium.

(a) Localised Respiration. — In many animals the absorp-
tion of oxygen takes place to some extent over the entire surface
of the body. Though this is not true of mammals, birds, and
probably reptiles, it is certainly the case with most cold-
blooded vertebrates. Thus Paul Bert showed that the axolotl
larva of the Mexican salamander survives after removal of both

RESPIRATION 65

the lungs and gills with but little diminution of total respira-
tory activity. The following data from Krogh's (1904)
experiments show that the skin is a very important factor
in the respiration of the frog, especially when it is pointed
out that the total surface of the skin is only about one and
a half times the internal surface of the lungs.

Cutaneous respiration.

Pulmonary respiration.

CO2 0.,

CO2 O2

Rana esculenta . .

.. 119 : 62

19 ' 59

Rana fusca

.. 129 : 52

45 : 105

Where a respiratory pigment is present it is sometimes
possible to recognise its respiratory function by the colour
of the blood which enters and leaves an organ. This is true
of course of the lungs and gills of vertebrates, from which the
blood issuing is a brighter red owing to the formation of
oxyhaemoglobin to be discussed below. Analogous evidence
points to the conclusion that the gills of the cephalopod —
whose arterial blood is of a more bluish complexion than that
in the veins — are to be regarded as structures specialised for
the intake of oxygen from the surrounding medium. This
is confirmed by direct measurements of Winterstein (1908)
on the oxygen and carbon dioxide content of arterial and
venous blood in Octopus vulgaris.

Arterial and Venous Blood in Living Octopus

Cephalic

aorta

(per cent.).

Abdominal

vein

(per cent.)

No. of animal.

CO2

CO3

8 ..

4*7

6-31

—

9 ..

4-66

7-09

— •

—

II . .

—

o"o9

5-62

12 . .

—

0-31

9*13

16 ..

—

0'26

7-83

17 ..

. . 4-64

3'94

—

It is highly probable, but not proven, that the gills of
Crustacea are to be regarded in the same light. But it is not
wise to assume that all structures labelled gills by systematists
are special arrangements to facilitate respiration. This is
well illustrated by experiments of Fox (1920) on Chironomus.
This author employed an ingenious method to investigate the
localisation of respiratory exchange in minute organisms, by

66 COMPARATIVE PHYSIOLOGY

making use of the fact that the flagellate Bodo migrates to a
region having a certain optimum oxygen tension. Fox found
that when pupae of Simulium, which respire by means of
filamentous appendages at the junction of head and thorax,
are placed in a suspension of Bodo, the micro-organisms collect
at first round the filaments in a dense congregation, and then
migrate outwards in a crescentic configuration, as the oxygen
concentration falls through absorption to a lower level than the
optimum. When Chironomus larv83 are similarly placed in
a suspension of Bodo, the flagellates show no special concen-
tration with reference to the so-called anal gills ; furthermore,
as this species possesses haemoglobin, it was possible to obtain
independent confirmation by spectroscopic observation for the
conclusion that these structures have no special respiratory
function. It will be seen later that the so-called gills of
lamellibranchs are to be regarded primarily as apparatus for
entrapping food-particles. Bounhiol (1902) brought forward
evidence that respiratory exchange in some Annelids falls
from 25 to 75 per cent, after removal of the gills ; and
Winterstein's (1909) observations on the effect of occluding
the anus in Holuthurians seem to indicate that either the
" respiratory " tree or the alimentary canal is responsible for
about fifty per cent, of the respiratory exchange which occurs
in these forms.

Of localised respiratory organs among Invertebrates the
most fascinating arrangement is the tracheal system of Arthro-
pods. Though the tracheal system of insects was fully
described by Malpighi in 1669, it is only comparatively re-
cently that the respiratory significance of the tracheal apparatus
has been put to conclusive experimental test, initially by the
work of Krogh (19 15). Krogh demonstrated the respiratory
function of the tracheae by two methods of attack. The first
consisted in analysing the gaseous contents of the tracheae of
the limbs in grasshoppers which had remained for some time
in a quiescent condition, as compared with the carbon dioxide
and oxygen in the tracheae of individuals which had been
chased to exhaustion to increase their respiration. For this
purpose the hindmost legs were squeezed out under glycerine

RESPIRATION 67

for gas analysis by Krogh's micro-method. The following
table indicates the results obtained in percentages : —

Quiescent.

Exhausted.

C02

CO2

2*0

I3'5

84-5

7'7

3-0

«9-3

1-4

i6*4

82-2

3*7

7'3

89-0

i"5

i6-7

8r8

1*5

5'8

92-7

3'o

lO'O

8ro

2*3

4' 5

93*2

—

6-0

6-4

87-6

Mean 1*9 15-7 82*4 4*2 5*4 90*4

In a second series of experiments the animals were placed
in a mixture of pure O2 and CO2 till no nitrogen remained in
the tracheae. The requisite time previously determined by
trial was found to be about 10 minutes. Analysis showed that
about one minute after being allowed to breathe ordinary air
the normal nitrogen content of the tracheae is restored. Even
after one second there is 2*5 per cent, of nitrogen in the tracheae,
while half a minute suffices to bring the percentage of nitrogen
up to 62'5. From the first set of data it is clear that oxygen
disappears and carbon dioxide increases in the tracheal tubes
during enforced respiratory activity. From the second it is
clear that gaseous diffusion takes place within the tracheal
system with a surprising rapidity, when one considers the
internal friction encountered by the gases in passing along the
lumen of tubes of such minute dimensions. If the rhythmical
movements of the abdomen which have been supposed to
facilitate the renewal of air in the tracheae are truly acts of
inspiration and expiration, it seems, from the rate at which these
succeed one another, that the air within the tracheal system is
renewed to the extent of about twenty per cent, at each move-
ment.

An admirably thorough investigation of tracheal respira-
tion in insects has recently been carried out by Buddenbrock
and Rohr (1923). The species employed in their researches
was Dixipus morosus, the familiar stick insect, whose barrel-
like configuration renders it specially suitable for some forms
of manipulation. Using Krogh's microrespiration methods,
they first demonstrated the fact that closure of the orifices
(stigmata) of the tracheal system reduces respiratory exchange

COMPARATIVE PHYSIOLOGY

to about a quarter of its normal dimensions. The remaining
twenty-five per cent, might be effected through the mouth or
anus, or by the skin. Buddenbrock and Rohr investigated this
point and found that occlusion of the mouth and anus does not
reduce the total exchange of gas. The alimentary tract does
not therefore contribute materially to the respiratory process.
The oxygen content of the tissues was found by Buddenbrock
to be in equilibrium with a partial pressure of about 76 mm. of

0;:GOj.

GOjv.

0 5^.. ••■ 10 15 20 25 30

Fig. 15. — (After Buddenbrock and Rohr.)

mercury. It might be surmised, therefore, that no effects
would accompany reduction of the oxygen content of the air
so long as there remained at least ten per cent. This proved
to be the case. Below this point physical diffusion does not
compensate for the rate at which oxygen can be used up, and
the respiratory exchange falls.

Very remarkable results accompany variation of the CO2
tension of the inspired air (Fig. 15). The intake of oxygen

RESPIRATION 69

diminishes as the carbon dioxide is increased, down to a
miminum (at about 25 per cent. CO2). With further increase
of CO2, it then suddenly increases till it actually exceeds the
oxgyen consumption in C02-free air. Finally, after attaining
a maximum, the intake of oxygen rapidly falls to zero under
carbon dioxide narcosis. This surprising phenomenon sheds
a new light on the significance of the abdominal contractions
extensively studied by Babak (19 12) and regarded by him as
respiratory movements. Babak had shown that phenomena
analogous to Cheyne- Stokes breathing and asphyxia could be
induced by varying the contents of the inspired medium ; but
crucial evidence that such disturbances of the normal rhythm of
abdominal movements had any compensatory value was lack-
ing. In Dixippus, however, when the normally spasmodic
contractions of the rump and abdomen became rhythmical
in lack of oxygen or excess of carbon dioxide, the increased
rapidity of the movements corresponds with an increase in
oxygen consumption, and it is difficult to escape the conclusion
that they actively facilitate the renewal of air in the tracheal
system. There is, moreover, a correlation between the opening
and closure of the stigmata and the respiratory movements.
In Dixipus, according to the observations of Buddenbrock
and Rohr, the thoracic stigmata open with each expiratory
movement and close with relaxation of the abdominal muscle,
thus, seemingly, promoting the passage of a current of air
from behind forwards.

Lee (1924), working on several other genera of Orthoptera,
has recently observed in their respiratory movements a definite
sequence of valvular motions of the thoracic and abdominal
spiracles (stigmata). According to Lee's account the external
valves or lips of the thoracic and first two pairs of abdominal
orifices of grasshoppers open when the abdomen enlarges and
close when it contracts. The orifices of the last six pairs of
abdominal spiracles are open during expiration and closed when
the abdomen enlarges. Hence if the abdomen of a normal
grasshopper is submerged while the head and thorax are
kept above the level of the water, minute bubbles escape
from the posterior abdominal spiracles ; when the whole

70 COMPARATIVE PHYSIOLOGY

animal is put under water no bubbles escape and asph}^ia
results.

Thus in insects there has been evolved — and in passing
we may note that analogous structures have appeared indepen-
dently in at least two other groups of Arthropods (isopoda
and arachnida) — a system by which oxygen is brought direct
to the tissues by a ramifying system of minute tubules which
penetrate even to the individual cells of the lining epithelium
of the gut. The evolution of this remarkable arrangement
is correlated with a very degenerate condition of the vascular
system. The efficiency of the tracheal system, which as we
have seen accounts for the greater part of the respiratory
exchange of these animals, must be extremely high when it is
remembered that insects in muscular activity surpass all other
invertebrates and many vertebrates also. It is interesting, how-
ever, to note that a limit is set to the efficiency of this device
by the size of the organism, since the internal surface of the
tracheal system cannot increase proportionately to the body-
weight.

Respiratory Pigments.— The body fluids of many animals
are known to contain substances whose affinity for oxygen
enables them to take up far more of this gas than is contained
by serum or sea- water in physical solution. The most familiar
example of a respiratory pigment is provided by the substance,
haemoglobin, present in the erythrocytes of all craniata and in
the serum of some invertebrates, especially annelida.

As is well known, haemoglobin on taking up oxygen assumes
a diiferent colour ; reduced haemoglobin is of a purple tint,
whereas oxy- haemoglobin is bright scarlet. The difference
is correlated with characteristic absorption spectra. Oxy-
haemoglobin has two absorption bands in the green ; reduced
haemoglobin has one which overlaps the space included by the
outer edges of the oxy-haemoglobin bands. Haemoglobin also
combines very readily with carbon monoxide to form carboxy-
haemoglobin, a much more stable compound than the oxygen
derivative. This also has two bands in the green ; it is not
of such a bright red colour. Oxy-haemoglobin can be reduced
to haemoglobin by exposure to a vacuum or neutral gas and

RESPIRATION 71

by various reducing agents. With potassium ferricyanide it
yields up all its oxygen, but the haemoglobin is speedily re-
oxidised by the reagent to form a brown isomer metha^moglobin
with a conspicuous absorption band in the red. Haemoglobin
is a compound of a protein and a nitrogenous pigment called
haemochromogen. The nature of the protein differs in
different animals. Haematin, which is the oxidised form of
haemochromogen, is known to contain four pyrol rings and
one atom of iron in its molecule. The constant relation
between the iron-content and the oxygen- capacity of a solution
of haemoglobin, established by Peters, shows that the formation
of oxy-haemoglobin is an essentially chemical union. Haematin
is separated from its conjugate globulin by dilute alkalis and
acids. The brown solutions formed in the two cases have
slightly different absorption spectra. On reduction of the
alkaline derivative with ammonium sulphide the red pig-
ment haemochromogen is found. Haematoporphyrin is the
purple substance formed by splitting off the iron from the
molecule with strong acids, and is isomeric with the bile pig-
ment bilirubin.

Haemoglobin and oxy-haemoglobin respectively take up or
give up oxygen according to the partial pressure of the gas
in the medium with which they are in contact. In the case
of man 100 c.c. of blood take up about 18*5 c.c. of oxygen
when fully saturated. Human blood is fully saturated at
a partial pressure of 100 mm., which is less than the
partial pressure of oxygen in the atmosphere. The curve
relating oxygen tension to oxygen content (or percentage
saturation) in a haemoglobin solution is a rectangular hyperbola.
Complete saturation of the blood occurs at the same partial
pressure, but the initial part of the curve is steeper in a pure
haemoglobin solution, so that at low tensions its oxygen-content
is higher than in blood at the same partial pressure of oxygen.
Among factors which influence the form of the oxy-haemoglobin
dissociation curve are those which affect the physical chemistry
of proteins generally — neutral salts, hydrogen-ion concentra-
tion, and temperature. Increased hydrogen-ion concentration
facilitates the dissociation of oxygen at low tensions — /.^.flattens

COMPARATIVE PHYSIOLOGY

the initial part of the curv^e ; and differences in salt-content
and pH. play a part in determining differences in the form of
the dissociation curve from the blood of different species.
The effect of acid is of physiological importance, since it implies
(see Fig. i6) that the readiness of the blood to give up oxygen
is greater under the conditions — presence of CO2 — normally
associated with oxygen want. Rise in temperature also
increases the dissociation of oxy-haemoglobin at low tensions.
Thus the warm blood of the mammal or bird is better suited
as a carrier of oxygen to the tissues than a pure solution of

^
t

/ /^^^^"""^

// />

•0 //

/I //

/It /.

il U

■ /Ji A

■^

/-/

i?-.

1 1 i. i_.-j 1 1_

— 1

20»-*OSObOV0^0 90**>0

Fig. 16. — Dissociation curves of mammalian blood (after Parsons).

haemoglobin at atmospheric temperature in the absence of
salts. Seeing that the haemoglobins of different species are
not identical, it is of interest to inquire into the extent to which
the properties of different haemoglobins can be correlated
with the conditions in which a given species pursues its exist-
ence. Investigations with this end in view have been carried
out by Krogh and Leitch (1919) on fishes, and by Barcroft
and Barcroft (1924) on the polychaete, Arenicola.

Krogh and Leitch compared the oxygen dissociation curves
of the blood of several species of fishes between io°-20° C. in
the presence and absence of CO2 to make the data as complete

RESPIRATION 73

as necessary for the purpose. In the absence of carbon dioxide
the blood of such fresh-water genera as the eel, pike, and carp
is half saturated at a partial oxygen pressure of 2-3 mm. (15° C).
In the marine species, represented by the cod and plaice, and
also in the trout, 50 per cent, saturation at 15° C. requires
an oxygen- tension of considerably greater dimensions, viz.
18 mm. with the cod, and about 11 mm. with the plaice or
trout. In both cases the presence of CO2 greatly diminished
the oxygen affinity of the blood at low tensions. In fresh water
the oxygen content is very variable and may sink to extremely
low values. The low loading tension of the blood in the fresh-
water fishes is thus appropriate to their medium, and accounts
for the low oxygen pressure to which Leuciscus (the minnow)
can, according to Winterstein's data, be subjected without
harm. In sea- water the dissolved oxygen is practically always
present in abundance at all depths. The water of the sea is
practically saturated with oxygen ; it therefore has an oxygen
tension above the 70 nmi. which represents the tension below
which (at 15° C.) a cod suffers from oxygen- want. In associa-
tion with this is the fact that sea- water fishes are very sensitive
to oxygen- want. The oxygen dissociation curve explains the
sensitiveness which makes sea-water aquaria more difficult
to maintain than fresh-water. The trout, which, as we have
seen, has a higher oxygen loading tension than such typical
fresh- water fish as the pike and carp, will only live in well-
aerated water, and is easily killed when the water is not
renewed or is insufficiently aerated.

Barcroft and Barcroft investigated not only the entire
range of the dissociation curve for haemoglobin in Arenicola,
but compared its chemical and physical properties with that
of human haemoglobin. As regards the first, which are
graphically set out in Fig. 17, it is to be noted that the blood
of Arenicola has at low tensions a very much higher affinity for
oxygen than mammalian blood. The oxygen loading tension
is of the same order as that described by Krogh and Leitch in
fresh- water fish. Complete saturation is obtained at an oxygen
tension of 10 mm. There are, apart from this, two charac-
teristic differences between the haemoglobins of Arenicola and

COMPARATIVE PHYSIOLOGY

man. The former has a much lower affinity for carbon
monoxide than the latter. At the point of 50 per cent, satura-
tion under exposure to a
mixture of oxygen and
carbon monoxide, the par-
tial pressures of CO and
O2 were not 250 : i as in
the case of man, or 140 : i
as in the mouse, but more
nearly 40:1. Again, as
regards the blood spectra,
it was found that the a-
oxy-hsemoglobin band is
18 Angstrom units nearer
the violet end, and the a-
carboxy-hsemoglobin band
is 1 1 Angstrom units nearer
the violet end than are
those of human blood.

100

1^80

S 60
g 50

-g 30
0^20

^*^

.«

/"^

,^"

,^'

^0^

^' '

L i

j^ c

i -3

b f

) (

) J

I t

i I

5 10

Pressure of O2 , rams.

~A=pU. 7*3 ; B=^^H. 6*9 ; Temp. 20° C.
C and D human blood, C=20°C. ; D =
37°C.;^H.=7-45.

Fig. 17. — Dissociation curves of areni-
colan blood (Barcroft and Barcroft).

Further, Vies has shown that the methaemoglobin of Areni-
cola has not the typical bands in the spectrum of mammalian
haemoglobin. There is a quantitative relation (Barcroft)
between the affinities of the different haemoglobins for carbon
monoxide and oxygen, on the one hand, and the position of
the bands in their absorption spectra on the other. The
logarithm of the reciprocal of the pressure at which the pigment
is half saturated with each is a linear function of the situation
of the a-band. The further the band is situated towards the
violet end, the greater in each case is the affinity for the
gas. From measurement of the blood- volume and oxygen-
consumption of Arenicola, it appears that Arenicola blood
can store just about enough oxygen to last when, sealed up
in its burrow at low tide, it has temporarily no access to
water.

We may now turn to the consideration of the respiratory
pigment of molluscs and Crustacea. Haemocyanin is a term
given to a family of substances which in the presence of
oxygen display a bluish hue, are like the haemoglobins of pro-

RESPIRATION

tein nature, and contain in organic combination a metal
which, however, is not iron but copper.

There is no doubt that the haemocyanin of molluscs and
Crustacea is a reversibly oxidisable pigment. Octopus blood
saturated with air was found by Winterstein (1908) to take
up 4-5 per cent, of oxygen. The oxygen capacity of the blood
of Palinurus was decidedly less — about 1*5 per cent., a differ-
ence possibly correlated with the lower haemocyanin content
of the blood in Crustacea. The blood of the Arachnid,
Limulus, like Crustacean blood contains a bluish pigment
which is a colloidal compound of copper, to which the term
haemocyanin has also been extended. Alsberg and Clark
(19 10-19 14) have, however, stated that the oxygen capacity of
the blood of Limulus or of a lo-per-cent. solution of Limulus
'' haemocyanin " is not significantly greater than that of sea-
water. Their observations have further shown that the com-
position of Limulus haemocyanin is not the same as that
of the haemocyanin of Octopus as determined by Henze
(1904).

Octopus (Henze).

Limulus (Alsberg and Clark)

Carbon

53" 66 per cent.

48*94 per cent.

Hydrogen

7*33

7'io „

Nitrogen

i6"o9

i6-i8

Sulphur

0-86

ri6

Copper

0-38 „

0-28

Oxygen

21-68

25'94

Notable additions have recently been made to our Imow-
ledge of haemocyanin as a respiratory pigment through the
researches of Dhere (1916-1921), and of Quagliariello (19 10-
1923), who have obtained haemocyanin from a number of species
in crystalline form. Haemocyanin of cephalopods can be
prepared by precipitation with concentrated (NH4)2S04,
that of the snail and rock-lobster by crystallisation of the
supernatant serum dialysate in ice. Oxy-haemocyanin so pre-
pared in crystalline form is a protein, completely precipitated
by dialysis, coagulated by heat and alcohol, behaving as an
amphoteric electrolyte with a minimal solubility at its
isoelectric point (pH. 47 in the case of Octopus). The oxy-
haemocyanin of cephalopods crystallises in needles of a greenish-

76 COMPARATIVE PHYSIOLOGY

blue tinge. Reduced haemocyanin does not show any absorp-
tion bands. The spectrum of oxyhaemocyanin from molluscs
and arthropods shows one band in the yellow and the beginning
of another in the blue in virtue of the copper-pynol complex
in the haemocyanin molecule. The position of the yellow
band in molluscs is about A579/X/X, and in Crustacea about
A563/X/X.

As regards the properties of haemocyanin as a respiratory
pigment, Dhere finds that oxy-haemocyanin of both Crustacea

D/ssccfstion Curves oF Haemocydnin In CrustdCesn Blood.

Cancer

/- "

Pahhurus

, ,. -rra

X^ CD

Maia

^. D

Homanis

Oxygen Tension

10 40 60 80 /CO 120 ItO 160 ISO 200 220 2iO 260 280 3C0 3:0 3'tO 366

Fig. 1 8. — (After Stedman and Stedman.)

and molluscs undergoes dissociation by lowering of the oxygen
tension, exposure to an inert gas or heating. There is no
combination between carbon monoxide and haemocyanin,
which however (snail, lobster) form a green compound with
N2O2. The physiological role of haemocyanin in the respira-
tory processes of Crustacea and molluscs is not completely
established ; but is strongly suggested by comparison of the
haemocyanin content (as measured by the amount of copper
present) with the oxygen capacity of the blood in different

RESPIRATION 77

species. The following table from Dhere's data shows how
close the correspondence is :

Oxygen c.c. Copper mg.

Animal. Temperature, per loo c.c. blood, per loo c.c. blood.

Octopus

18''

4' 05

25-6

Helix

i7°-i9°

1-82

Homarus

iS's"

3' I

10-5

Cancer

18°

1-6

5*5

Astacus

22°

2*4

8-0

From the recent work of Fox (1924) it would appear that
the green pigment, chlorocruorin, of certain polychaetes is
like haemoglobin and haemocyanin a substance which may
facilitate the carriage of oxygen to the tissues or its temporary
storage in the body. Chlorocruorin occurs in the Chlorhae-
midae and the Sabelliformia. The blood of these worms is
green by transmitted and red by reflected light. It was first
observed by Milne Edwards (1838) ; and its spectrum was
later studied by Ray Lankester (i 867-1 870). Oxy-chloro-
cruorin has two absorption bands, the limits of which were
given as 618-593/x/Lt and 576-5 54/>t/x. By reduction with
ammonium sulphide a derivative was obtained which had
only one band situated between 625 and 596/x/x. As this
reduced form reassumed the oxy-chlorocruorin spectrum on
shaking with air, Lankester concluded that it was respiratory
in function.

The fact that chlorocruorin is reduced by reducing agents
and reoxidised in air is, however, insufficient reason for be-
lieving that reduction can take place in the body of the worm.
It has now been shown by Fox that oxy-chlorocruorin is reduced
in a vacuum, that is to say, it both takes up and gives up oxygen
according to the oxygen tension of the surroundings. Further-
more, its reduction by living tissues can be demonstrated by
spectroscopic observation of the blood of Spirographs in
contact with a piece of living muscle under a sealed cover-
slip. On removing the cover-slip the oxy-chlorocruorin
bands reappear. Assuming that the gas hberated by potas-
sium ferricyanide is oxygen, the oxygen capacity of the blood
of Spirographis was found by Fox to be about 6*17 per cent,
or 1 0*3 times the quantity dissolved in sea- water. This is
distinctly higher than the figure for the blood of Cephalopods

78 COMPARATIVE PHYSIOLOGY

and about twice or three times the oxygen capacity of the blood

5iO

5(8

57B

576

SifO

Fig. 19. — Absorption spectra. I: oxychlorocruorin in three concentrations ;
II and III give the spectrophotometric axes of the bands of the haemoglobin
of the horse (II) and Arenicola (III) (after Fox).

(,10

—\ —

5S0

Fig. 20. — Absorption spectra. I : chlorocruorin reduced by Na2CO +
Na2S204 ; 1 1 chlorocruorin reduced by AniaS ; III arenicolan haemoglobin
reduced by NagCOs + NaoSaOi ; IV arenicolan haemoglobin reduced by
Am.S : V mammalian haemoglobin (after Fox).

in Crustacea and other molluscs. It is almost equal to the

RESPIRATION 79

oxygen capacity of Arenicolan blood. Bounhiol (1902) found
that the polychaetes with haemoglobin have a more active gas
exchange than those without haemoglobin. It cannot be
stated as yet with confidence that chlorocruorin performs an
essential role in the life of the worm ; for the blood circula-
tion is poorly developed and there appears no danger of oxygen
deficiency in the water in the life of Spirographis under normal
conditions. It is interesting to note that the earthworm will
survive perfectly well after all its haemoglobin has been con-
verted into carboxy -haemoglobin so that the blood pigment can
no longer function as a vehicle for the transport or a means for
storing oxygen. Chlorocruorin in many respects resembles the
haemoglobins. It forms a compound with carbon monoxide.
The oxidised form is reduced by potassium ferricyanide. It
acts as a peroxidase. It is possible to prepare a parallel series
of derivatives of chlorocruorin analogous to met-haemoglobin,
haematoporphyrin, haemochromogen, etc.

Finally, mention must be made of a pigment allied to
haemoglobin in the liver of the crayfish and the gut of all
pulmonates except Planorbis, a genus in which haemoglobin
itself occurs in the blood. Helicoruhin, as this pigment is
called, has been investigated by Anson and Mursky (1925),
who find that it combines loosely with oxygen, its affinity for
the latter being increased, not as in the case of haemoglobin
decreased, by the acidity of the medium. These authors have
put forward a rather different view of the biochemical and
phyletic relationship of the haemoglobins from that hitherto
accepted and indicated earlier in this chapter. They regard
haemochromogen and its oxidised form haematin as themselves
conjugated proteins of which haemoglobin and oxyhaemoglobin
are respectively polymers. To the iron-pyrrol part of the
molecule Anson and Mirsky apply the term hcern. Haemo-
chromogen is a compound of haem and a nitrogenous sub-
stance, protein or otherwise. The haems of different haemo-
globins and of helicorubin are identical ; the specificity of the
haemoglobins depends on the nature of the conjugate globin
or the degree of polymerisation of the haemochromogen. The
separation of haem from its conjugate protein (or other

COMPARATIVE PHYSIOLOGY

nitrogenous compound) takes place in acid medium. At
present the physiological significance of helicorubin is obscure.
The Transport of Carbon Dioxide.— We have seen that in
the recovery phase of muscle the intake of oxygen is accom-
panied by an evolution of carbon dioxide. The removal of
carbon dioxide is generally considered under the heading of
respiration, though it cannot be assumed without direct evi-
dence that the intake of oxygen and the excretion of COo
always takes place in the same organs.

T70

1 1 ' '

" "" _l _^A^6if^-

G^^j^^- __

- . ,_J-''T 1 -.-•-"'^^p^ ..

---"^-''^.^'^ -^

.^^ --^f O-j^*^

^^., <"

^^i^

/^/•-^

^V-

L J

t m

30 40

50 60 70 60
oj C Oj t-vi *«/m. 36a

Fig. 21. — Curves relating the amount of CO2 taken up by a given
volume of blood to CO2 pressure, showing that oxygenated blood takes up
less CO.>, and that therefore oxygenation has made the blood more acid
(after J. S. Haldane).

If the carriage of carbon dioxide in the blood of the mammal
is studied by analogous methods to those employed for plotting
the dissociation curve for haemoglobin — i.e. exposure of the
blood to different tensions of carbon dioxide and estimation
of the carbon dioxide absorbed, — it is found that the amount
of carbon dioxide which can be taken up by the blood is con-
siderably greater than that which would be dissolved by
physical solution even at comparatively high tensions. At
a partial pressure of 40 mm., which corresponds to the lowest
tension of CO2 with which the arterial blood of the mammal

RESPIRATION 8i

is in equilibrium, lOO c.c. of blood contain over 50 c.c. of carbon
dioxide. This combined carbon dioxide exists in the blood
in the form of sodium bicarbonate. It has been shown by
Parsons that the alkali comes from the dissociation of haemo-
globin, which exists in the blood on the alkaline side of its
isoelectric point as a sodium salt. Haemoglobin is a weaker
acid than its oxygen derivative, and oxygenation therefore
favours the displacement of COg in the competition between
the weak carbonic acid and the protein anion for a fixed amount'
of base (Fig. 21).

The conditions of carbon dioxide in the blood of marine
invertebrates have been recently studied by Collip (1920) and
by Parsons and Parsons (1923). The state of affairs existing
in these animals is different in some respects from that which
is found in the mammal. Collip 's investigations, on repre-
sentatives of molluscs, arthropods, annelids, and coelenterates,
indicate that in general the amount of carbon dioxide taken up
at pressures greater than the low tension of CO2 in ordinary
atmospheric air are only such as would be accounted for by
physical solution. Practically all the available alkali is com-
bined to form bicarbonate at a carbon dioxide tension far
below that found even in the arterial blood of the mammal,
where the alkali reserve is not used up until comparatively high
tensions are attained. This peculiarity is of interest, firstly,
in relation to the part played by haemoglobin as an alkaline
salt in the blood of the mammal ; secondly, in relation to the
rate of metabolism which such an arrangement permits ; and
thirdly, as affecting the reaction of the blood which in the
normal life of the mammal is kept constant within fairly narrow
limits by the buffer action of the dissociated protein.

These points have been investigated in several genera
by Parsons and Parsons (1923), from whose observations
emerges a very significant difference between the conditions
of carbon dioxide transport in the blood of comparatively
active free-living forms such as the crustacean genera Maia
and Palinurus or the cephalopod Octopus, and sluggish or
sedentary forms such as the mollusc Aplysia and the tunicate
Phallusia. In Aplysia and Phallusia the uptake of CO2 is

COMPARATIVE PHYSIOLOGY

practically a linear function of the partial pressure for all
values of the latter, and the carbon dioxide capacity of the
blood is not greater than that of sea-water. In the case of
the tunicate that of the whole blood is rather less ; but the
capacity of the plasma is somewhat higher than that of the
whole blood, and the difference is probably correlated with
the distinctly acid reaction of the corpuscles which, according
to Henze, may contain as much as 3 per cent, free sulphuric
acid. In marked contrast with the carbon dioxide capacity
of these comparatively inactive forms are the curves derived

50 60 70

Fig. 22.

from the blood of the crustacean and cephalopod (Fig. 22).
Here the uptake of CO2 by the blood increases steeply up to a
tension of about 15 mm. Hg ; and within this Hmit the reaction
remains well on the alkaline side of neutrality. The steep
portion of the curve is much less protracted than in the
mammal ; but according to Parsons and Parsons there is never
more than about 3-10 c.c. of COo per 100 c.c. of blood in these
animals as against 50 c.c. in 100 c.c. of mammalian arterial
blood. The more stable reaction and greater carbon dioxide
capacity of the blood in the crustacean and cephalopod as
contrasted with the condition in Aplysia is that the blood of
the former is rich in haemocyanin and other proteins. The

RESPIRATION 83

body fluid of Aplysia (Bottazzi) contains less than o"oi per cent,
of protein nitrogen. The isoelectric point of haemocyanin is
well below absolute neutrality, being according to Quagli-
ariello about pH. 47, so that in the slightly alkaline blood of the
invertebrates which possess it, it should be dissociated as an
anion ; and Quagliariello finds, as might be expected on the
assumption that proteins are amphoteric electrolytes, that the
acid-neutralising power of invertebrate blood is roughly pro-
portional to its protein content. Thus it would appear that
the proteins of the blood of invertebrates, as in the mammal,
exist normally as sodium salts capable of giving up their
kation for the carriage of carbon dioxide away from the tissues
as sodium bicarbonate, and that the steep initial portion of
the carbon dioxide dissociation curve exhibited by those
forms which possess haemocyanin is due to the competition
of CO2 and protein anions for the alkali kations of the blood.

Further Reading

Books,
Barcroft's Respiratory Function of the Blood.
Krogh's Respiratory Exchange of Animals and Man.
Haldane's Organism and Environment.
Haldane's Respiration.

On Respiration in Cephalopods.

Polimanti (19 1 2). Beitrage zur physiologic von Sepia II. Arch. f. anat.

u. Physiol, p. 53 (1909)-
WiNTERSTEiN. Zur kenntnis der Blutgase wirbellose Seetiere. Biochem.

Zeitschr. 19.

Annelids.

BoiTNHiGL (1902). Recherches sur la respiration des annelides. Ann. de
sci. nat. i6.

Insects.

BuDDENBROCK AND RoHR (1922). Die Atmung von Dixipus morosus.

Zeit. Allg. Physiol. 20.
Krogh (1913-20). On the Composition of the Air in the Trachea System.

Skand. Arch. Physiol. 29.
Studien ueber Tracheen Respiration II-III. Pflugers Arch. 179.

Lee (1924). On the Mechanism of Respiration in Certain Orthoptera.
Journ. Exp. Zool. 41.

84 COMPAIUTIVE PHYSIOLOGY

Fishes.

Baglioni (1909). Der Atmungsmcchanismus der Fische. Zeit. Allg.

Physiol. 7.
WiNTERSTEiN (1908). Beitrage zur Kenntnis der Fischatmung Pflugers

Archiv. 125.

Pigments.

Alsberg and Clark (19 10-14). Hasmocyanin of Limulus. J. Biol.

Chem. 8. Solubility of Oxygen in the Serum of Limulus. Ibid. 19.
Barcroft and Barcroft (1924). The Blood Pigment of Arenicola. Proc.

Roy. Soc. B. 96.
Dhere (1916-21). Recherche surrhaemocyaninel-VI II. Journ.de Physio.

et Pathol. Gen. 16-20.
Fox (1924). On Chlorocruorin I. Proc. Camb. Phil. Soc.
Krogh and Leitch (1919). The Respiratory Function of the Blood in

Fishes. Journ. Physiol. 52.
QuAGLiARiELLO (1920). Ricerchi etc. sulla emocianina. I-III. Arch. Sci.

Biol. (1922) Pubbl. St. Zool. Napoli.
Stedman AND Stedman (1925). Biochem. Journ. 19.

Carbon Dioxide.

Parsons and Parsons (1923). Transport of Carbon Dioxide in the Blood
of Some Marine Invertebrates. Journ. Gen. Physiol. 6.

CHAPTER V

NUTRITION

When a muscle contracts glycogen disappears. Only part
of this glycogen is reinstated in the recovery phase of muscular
contraction, and probably an analogous phenomenon occurs,
as already indicated, in the case of ciliary and glandular activity.
It is necessary, therefore, that the supply of materials in the
effector organ should be replenished. In the case of growing
organisms, it is necessary also that a supply of the materials
concerned with the manufacture and growth of cells shall
be maintained. In this chapter we shall deal with the means
by which a supply of necessary material is ensured.

It is more than a century since Lavoisier and Laplace
showed that the bodily heat of warm-blooded animals is a
form of slow combustion, and that the amount of oxygen
used up and of carbon dioxide liberated has a definite and
ascertainable relationship to the heat that is generated. By
applying the balance and the thermometer to the phenomena
of life, Lavoisier founded the modern science of nutrition.
By the middle of the following century Liebig had shown
that the three categories of organic compounds known as
proteins, carbohydrates and fats are the substances whose
decomposition and oxidation form the basis of those chemical
changes which occur under the influence of living cells and are
collectively referred to under the term *' metabolism." A signal
advance was made by the researches of Voit and others in the
^sixties, when it was shown that muscular activity does not
increase protein metabolism (estimated by the nitrogenous
content of the urine) ; that the complete combustion of a given
weight of each class of compounds is associated with the

86 COMPARATIVE PHYSIOLOGY

intake of a definite amount of oxygen ; and that the ratio of
carbon dioxide evolved to oxygen used up is different for
carbohydrates, fats, and proteins respectively. This ratio is
known as the respiratory quotient. It is, as would be expected,
unity in the case of carbohydrates, and less than unity for
fats (071) and proteins (cyS). The investigations of Liebig
coincided with the formal statement of the conservation of
energy by Mayer on the basis of Joule's determination of the
mechanical equivalent of heat. The general applicabilit}'- of
the first law of thermodynamics to living organisms was
universally accepted as the basis of physiological research by
the end of the nineteenth century.

For the greater part of this period, however, the impossi-
bility of oxidising animal foodstuffs at such temperatures as
are consistent with organic existence as we know it, or of
stimulating the digestive reactions in vitro without recourse
to reagents which would be fatal to the organism, presented
an inflexible barrier to the probability that the mechanism of
living organisms conforms to the known laws of energetics.
To-day the position has changed in two ways. The study of
those more complex chemico-physical systems which are for
convenience described as *' colloids," and the role of surface
tension, osmotic pressure, and electrolytic dissociation in
modifying their properties, opens up a new horizon of possi-
bilities, while the extension of the principle of catalysis to
enzymes, and its clarification by Ostwald and others, has
thrown a flood of light on the chemical equilibrium of the
organism.

Sources of Animal Food. — Fats, proteins, and carbohydrates
are the principal constituents of a healthy diet. There is a
certain amount of evidence that fats and carbohydrates
are convertible into one another, and that carbohydrates can
be manufactured from the deaminised products of protein
metabolism. Protein as such is not necessary ; but it seems
that it can only be replaced by its hydrolysis products, the
amino-acids. In this connection some interesting bionomic
problems arise.

It was found by Loeb (19 15) that the banana fly (Droso-

NUTRITION 87

phila) can complete its larval stage in a solution of cane sugar
and salts adsorbed in filter paper. The lar/ae grows quickly,
and on addition of ammonium tartrate, glucose, and citric
acid, successive generations can be reared. Here at first
sight appears to be an organism that can flourish in a medium
deficient in nitrogen compounds of the degree of complexity
hitherto thought to be required by animals invariably. This,
however, is not the case.

The nitrogen supply of Drosophila has been made the
subject of recent investigation by Baumberger (1918). The
alimentary tract in Drosophila larvae teems with yeasts. In
order to explore a possible relation between the yeast organisms
and the nutritional processes of the fly, Baumberger steri-
lised eggs and pupae by immersion for a short period in 85 per
cent, alcohol. Sterile individuals having been so obtained,
both sterile and normal individuals were placed on (a) sterile
banana— agar culture media, and (b) a sterile synthetic medium
containing mineral salts, sugar, and ammonium tartarate as
the sole source of nitrogen. The consequences of this treat-
ment on the two classes of individuals w^re striking. Normal
{i.e. unsterilised) individuals deposited eggs which grew into
larvae that pupated normally on both banana-agar and synthetic
media. The larvae which developed from sterilised eggs and
as the offspring of sterilised pupae failed either to grow or
pupate on a sterile medium of either type ; they only survived
a few days. When, on the other hand, similar sterilised
individuals were placed on media of the same nature which
had been previously infected with yeasts, they at once began
to thrive, pupating as usual. Thus in the presence of yeasts
Drosophila can grow on an artificial medium with ammonium
tartarate as its only source of nitrogen.

It may now be asked whether the food requirements are
met by any by-products of fermentation. This Baumberger
tested by boiling the yeast before adding it to the sterile cul-
tures ; fermentation was in this way prevented. The larvae,
however, grew steadily, and the possibility that the fly larva
actually ingests the yeasts alone remained. On cultures of
compressed yeast-agar with yeast nucleo-proteins as the sole

88 COMPARATIVE PHYSIOLOGY

nitrogen supply, the sterile larvae were able to grow and pupate
normally. It thus appears that yeasts are the nitrogenous
food of Drosophila ; the simplest nutrient solution suitable
for the yeasts (and certain other micro-organisms) will replace
fermenting fruit in the ecology of Drosophila larvae. These
experiments of Baumberger are extremely suggestive in relation
to the diet of wood-boring animals, the significance of fungus
gardens, the curious habitat of such organisms as the vinegar
worm, and a host of other bionomic problems.

Researches have also been carried out on Drosophila
larvae in relation to the accessory food-factors or vitamins.
The term " vitamin " is one which at present can hardly be said
to convey more than a recognition of our failure to induce
mammals to grow healthily on a diet of purified carbohydrate,
fat, protein, etc., and our almost complete ignorance of those
constituents of natural foods which must be added to such a
diet to preserve health and normal development. The necessity
for recognising accessory food-factors was first clearly recognised
by Hopkins (1906), and the conception became more concrete
when Funk (191 1) extracted from 100 kilograms of yeast
2*5 grms. of a material of which a dose of 2 mg. sufficed to cure
the polyneuritis induced in pigeons by an exclusive diet of
polished rice. There are at least three chemical entities
included under the term '' vitamin." A is present especially
in animal fats, B in yeast, and C in fruit juices ; but in fresh
animal or vegetable food all three are represented to some
extent. The separate identity of these substances is inferred
from the different clinical results of eliminating one or the
other.

Bacot and Harden (1922) have investigated the extent
to which vitamins are essential to the diet of Drosophila.
Successful growth of larvae can apparently proceed in a nutrient
medium composed of pure caseinogen, starch, sugar, and salts
only if small quantities of yeast extract (as a source of " B ")
and traces of butter fat ('* A ") are included. '' C " was not
found to be essential, though amply available in the normal
diet (fermenting fruit-juice).

The behaviour of Drosophila in regard to yeasts more-

NUTRITION 89

over recalls Keeble's researches on the supposed symbiosis
between the turbellarian Convoluta and the Chlamydomonad
which infests its subintegumentary tissues and is ultimately
destined to be absorbed by intracellular ingestion in the body
of its host after degeneration of the alimentary tract in the
latter. Reference may be made here to an hypothesis put
forward some years ago by Putter (1907) who maintained that
many aquatic organisms absorb dissolved organic matter from
the water as a source of food. This view is provocative,
because as Dakin rightly points out, though structures resem-
bling the alimentary tract of land animals exist throughout the
animal phyla, it is largely on the basis of analogy that these
have been regarded as the only avenue through which food
passes into the organism. Putter's hypothesis was based on
three lines of reasoning : (i) that there exists in sea- water
a comparatively large available quantity of dissolved organic
matter ; (ii) that the quantity of solid food present in sea-
water is insufficient to account for the rate of respiration of
marine organisms ; (iii) that certain animals — e.g. goldfish —
do not lose weight if amino-acids, glycerine, etc., are dissolved
in the water, but do so if kept without food in water containing
no dissolved organic matter. As regards the first, later in-
vestigation has not as yet fully confirmed Putter's analyses,
but recent observations of Harvey (1925) and of Atkins (1925)
point to the conclusion that appreciable quantities of dissolved
organic matter exist in sea- water. The data on which the
second conclusion is based are questionable. Experiments
of Putter on absorption of nitrogenous solutes by goldfish
and axolotls have recently been repeated by Dakin and
Dakin (1925) with negative results. There seems, therefore,
insufficient reason for abandoning the view accepted by most
students of the plankton, that marine organisms prey on one
another, the smaller organisms providing food for larger ones,
as in the following series (Johnstone's " Life in the Sea ") :
Peridinians — Copepoda — Sprats — ^Whiting — Cod — Man.

Feeding Mechanisms. — Appropriate devices (jaws, beaks,
etc.) for the trituration of food in animals which actively
select their diet are described in text-books of zoology. A

COMPARATIVE PHYSIOLOGY

few words may be inserted here with reference to the methods
adopted by animals of sluggish and sedentary habits for
maintaining a supply of food. Most widespread of such feed-
ing mechanisms are those which involve the entanglement
of food particles, such as organic debris and micro-organisms
of the plankton, in mucous slime through the production
of ciliary currents to maintain a constant flow of water over
the slime-glands and to propel the entrapped food-particles
towards the mouth. In Amphioxus, for example (Orton),
water flows from the pharynx into the atrium by the lashing

of cilia which line the sides of
the pores in two lateral rows ;
these cilia do not play a direct
part in the collection of food-
particles, v/hich are caught in a
fine sheet of mucus secreted by
the endostylar gland cells and
thrown on to the sides of the
pharynx by the cilia of the
ventral groove. This sheet of
slime with its entrapped food-
particles is worked up into
cylindrical masses driven to-
wards the dorsal groove by
cilia which line the inner wall
of the pharynx. The cilia of the
hyperpharyngeal groove maintain
a current of this slimy suspension in the direction of the in-
testine where digestion and absorption take place. Essentially
similar arrangements exist in Tunicates, Amphioxus, and in
at least one Vertebrate, the Ammocoete larva of Petromyzon.
In the bivalve molluscs it is again the structures which
descriptive anatomists have labelled gills which constitute
the ciliary net. Water laden with organic debris and micro-
organisms filters between the filaments of the gills through
the action of currents produced by the lateral cilia. A
ventrally directed current due to the frontal cilia washes
the food-particles entangled in slime downwards towards a

Fig. 23. — Ciliary currents on the
Lamellibranch gill.

NUTRITION

ciliated groove formed by the distal ends of the filaments.
In this food - groove a strong anteriorly directed ciliary
current washes the mucous stream on to the labial palps,
whence they are propelled — still by ciliary action — into
the mouth (Orton, Kellogg, Yonge). There are often special
arrangements for excluding coarse particles, sand, etc. In
the primitive gastropod Crepidula — and probably other
marine prosobranchs — we find analogous phenomena.
According to Orton (1913) in Crepidula an ingoing and
outgoing current is established along a definite pathway
and the single gill acts as a strainer between them. The
filaments lie parallel in a horizontal line extending along the
left side of the mantle cavity, dividing it into a left ventro-
lateral inhalent chamber and a right dorsventral exhalent
chamber. In feeding, the front end of the shell is raised
slightly, water is drawn in along the anterior half of the shell
on the left, passed through spaces between the gill-filaments,
and expelled along the front half of the right edge of the shell.
Upon reaching the tips of the filaments, the food-particles,
driven along in a mucous stream by the frontal cilia, are
deposited in a food- groove, like that already seen in lamelli-
branchs, running along the right side of the body. Eventually
the food-masses are seized on by the radula. Ciliary feeding
occurs in Brachiopods, Polyzoa,and some Polychaeta. Entangle-
ment of food-particles in slime is also seen in small Crustacea
such as Daphnids, v/here the labial glands exude a stream
of mucilaginous secretion which entraps suspended matter in
the ventral current produced by the thoracic appendages, to
be seized on by the mouth parts.

The History 0! the Foodstuffs.— We may now turn to con-
sider the changes which the three principal classes of organic
food-constituents undergo in the digestive tract, and their
subsequent fate in the body. For a detailed treatment of the
latter, standard monographs on biochemistry must be con-
sulted ; such knowledge as we possess is derived very largely
from clinical sources and from the study of mammalian
physiology.

As proteins exist in colloidal form, they are incapable of

92 COMPARATIVE PHYSIOLOGY

passing through the membranes of the digestive tract until
broken down into diffusible products by hydrolysis. In the
mammal there appear to be three stages in the process. The
first takes place through the agency of the gastric enzyme
pepsin which exerts its optimum efficiency in an acid medium
which is provided for by the presence of free HCl in the secre-
tion of the gastric mucosa. Prolonged digestion in vitro of
proteins in the presence of pepsin does not carry the process
to the liberation of amino-acids, which are the end-products
of protein hydrolysis in presence of inorganic catalysts. In
the body peptic digestion probably promotes only the initial
stages of splitting into simpler proteins such as proteoses and
peptones.

The enzyme trypsin which is supplied by the pancreatic
juice can bring about the complete hydrolysis of proteins in
vitro. In the body it seems probable that the reaction is not
carried beyond the production of the relatively simple and
diffusible condensation-products of amino acids known as
polypeptides. The final resolution of these into simple amino-
acids is apparently effected with the co-operation of a pro-
teoclastic ferment in the secretion of the intestinal mucosa
(erepsin). Proteoclastic enzymes have been detected in
extracts of the digestive glands of all groups in the animal
kingdom. How far they are identical with those which
occur in the mammalian gut is not certain. Using the facility
with which a gelatine mixture solidifies when cooled for a
fixed period in the ice-bath as a measure of the progress of
protein hydrolysis, Bodansky and Rose (1922) extracted from
the mesenteric filaments of the jelly-fish Stomolophus and the
siphons of Physalia (Siphonophora) a digestive fluid with two
pH. optima at 3*0 and 7*3 respectively, roughly corresponding
to the pH. optima for mammalian pepsin and trypsin. A
rennet-like ferment capable of coagulating the milk protein
caseinogen was also found to be present. Yonge (1924) was
unable to find a pepsin- like enzyme in the digestive gland
of the lobster ; but free amino-acids were obtained from an
alkaline digest with the extract.

In the tissues, especially in the liver of the Vertebrates,

NUTRITION 93

amino-acids are partially decomposed with liberation of
ammonia. The latter combines with CO2 in the blood to be
transformed into urea, which is ultimately excreted ; the
residual portion of the amino-acid molecule is a keto-acid
which forms a common link in the intermediate metabolism
of carbohydrates and fats. This process is known as deamina-
tion. The importance of this lies in the fact that in some
carnivorous animals there is very little carbohydrate present
in the food, though, as we have seen, the chemical energy
of carbohydrates is the ultimate source of the mechanical
energy of molecule contraction. Thus while most animals
feed predominantly on nitrogenous food, the nitrogenous part
of the protein molecule is of little constructive importance
except in growing animals, where new protoplasm is being
formed. We have to distinguish between endogenous meta-
bolism which is concerned with growth and tissue waste on the
one hand, and exogenous metabolism which is concerned with
effector activities and the maintenance of body heat on the other.

Observations have been made concerning the deamination
of amino-acids in the blow-fly (Calliphora) by Weinland
(1908), who showed that both the larvae and a pulp made by
crushing them had the power, in the absence of oxygen, to
split peptones into amino-acids, deaminise them with evolu-
tion of ammonia, and produce higher fatty acids with
evolution of CO 2 — presumably by synthesis from the nitrogen-
free remainder of the amino-acids. The particular enzyme
reactions which occur on a large scale in the digestive
processes are not to be regarded as special properties of the
alimentary secretions, but rather as characteristic of what is
involved, to a greater or less extent, in the metabolism of all
cells in the body. Among the hydrolysis-products of one
important class of proteins, the nucleo-proteins, purine bases
are found in addition to amino-acids. Purine bases derived
from food or tissue waste are excreted in many animals as
uric acid. But enzymes are known to exist which oxidise
purines with formation of urea ; and uric acid is not an
invariable excretory product.

The extent to which fats contribute to the diet varies

94 COMPARATIVE PHYSIOLOGY

greatly with the feeding habits of the organism. In the
mammal the bulk of the fat is unchanged in the stomach.
In the duodenum, aided by the churning movements of the
intestinal wall, bile-salts exert their characteristic effect in
lowering surface tension to effect a fine degree of emulsifica-
tion. The emulsion is acted upon by a lipolytic enzyme of the
pancreatic juice, and broken down completely into its hydro-
lysis products by neutralisation of the fatty acid so formed
with production of soaps. The soaps of the higher fatty acids
form colloidal solutions, and are not diffusible like amino-
acids or sugars. They are absorbed by the cells of the mucous
membrane, which are richly gorged with fat-globules after a
meal of fat- containing food. The colloidal nature of the
higher soaps suggests that the same way of dealing with hydro-
lysis products of fat should hold in other groups, as the work
of Sanford (191 8) on digestion in the cockroaches clearly
demonstrates. Sanford fed cockroaches on a mixture of sugar
and olive oil and showed, by following microscopically the
course of digestion, that after a fatty meal the cells lining the
wall of the crop teem with fat globules. The contents of the
crop exhibited, like the pancreatic juice of the mammal, a
powerful Hpolytic action. This provides good material for
class experiment. The organ is removed from about a dozen
cockroaches, ground in a mortar with sand and about 10 c.c.
of water, a few c.c. of the filtered extract is added to about
the same quantity of olive oil and kept for a few days at room
temperature, when the amount of free acid liberated is deter-
mined by titration and compared with a control tube. For
microscopic examination sections of the wall fixed at varying
intervals after the meal are treated with osmic acid or the
dye known as Sudan III., both of which are specific stains
for fat. Sanford also fed cockroaches on a paste of oil and
sugar mixed with Nile Blue sulphate, which is absorbed by the
fat-globules and gives a red coloration in presence of free
fatty acid. On cutting frozen sections a few hours after such
a meal, a red mass is seen cHnging to the wall of the crop which
is itself blue owing to the dye adsorbed by the fat globules
in the cells.

NUTRITION 95

The fats which occur in the animal body are all derived
from fatty acids with an even number of carbon atoms. The
diabetic animal can form glucose from glycerol ; and a stage
in the intermediary metabolism of the fatty acids is the forma-
tion of keto-acids which further link up the metabolism of
fats with that of carbohydrates. Animals fed on carbohydrate
or protein diet deposit fat in their tissues, and the study of
hibernation indicates that the transformation of fat into
carbohydrates also occurs. In hibernating mammals (Pem-
brey) the respiratory quotient may be as low as 0*3, showing
that there is a conversion of substances with a small quantity
of oxygen (fat) into others with a larger amount (carbohydrate).

Of carbohydrates, the polysaccharides starch and cellulose
are the principal representatives in animal diet. The role
of the former alone is understood in relation to the metabolism
of the mammal, where the breakdown of starch (and glycogen)
occurs in three stages during digestion. An amylase is present
in the saliva which is capable of carr}ang the hydrolysis of
starch in vitro through dextrins to the disaccharide malt
sugar. Actually the acidity of the gastric juice limits con-
siderably the extent of starch- digestion in its initial phase.
The same process is continued in the duodenum by the action
of an enzyme present in the pancreatic juice. There are in
addition present in the secretion of the duodenal glands
enzymes which complete the hydrolysis of maltose and the
other disaccharides (lactose and sucrose) into monosaccharide,
in which form they diffuse into the body. In the vertebrate
the blood, enriched with sugar, after digestion, has to flow
through the capillaries of the liver, where conditions occur
that permit synthesis of the storage carbohydrate glycogen
under the influence of an enzyme (glycogenase). During
starvation the reverse reaction predominates — glycogen is
transformed into sugar, which can be transported by the blood
to the muscle where it is also stored in the form of glycogen.
Glycogen is the universal storage form of carbohydrates in
animals. And the storage of glycogen is also a function
of the so-called liver of Crustacea. The amount of glycogen
in the crab's liver increases before each moult, and is used

COMPARATIVE PHYSIOLOGY

up during the period when the new shell is being formed and
the animal is temporarily deprived of the power to feed.

Amylolytic enzymes have been found in the alimentary
tract of all animals investigated with this end in view. In
the coelenterates mentioned, Bodansky and Rose found that
an amylase and a maltase were present but lactase was absent.
On the other hand, both sucrose and lactose are digested by
extracts of the digestive gland of the lobster.

A remarkable phenomenon connected with starch digestion
in the invertebrate phyla is presented by the structure
known as the crystalline style present in most lamelli-

Digestive gland

stomach

style

heart
Fig. 24. — The crystalline style of the bivalve mollusc.

branchs and in a few prosobranch gasteropods. The crystal-
line style is an elongated hyahne rod of proteinous nature to
which is absorbed an amylolytic enzyme. It lies freely either in
a groove of the intestine or in a separate diverticulum, revolving
about its axis in the ciliary current produced by the epithelium
of its sac. As it revolves its anterior end, which in many forms
projects into the stomach, is worn away against a horny plate
(the gastric shield) in the dorsal wall of the latter, entangling
in its motion a mucous mass laden with diatoms and inorganic
debris. In some forms it is broken down and reformed
periodically ; in the Eastern oyster (Nelson) it disappears
an hour after the fall of tide, and may be reformed in fifteen

NUTRITION

minutes. In other cases, as in Mya (Edmondson), it is per-
manent and may take months to regenerate if excised. As the
hepato-pancreatic secretion which is poured into the stomach
contains a proteolytic enzyme which rapidly dissolves the
style in vitro, the permanence or otherwise of the style
possibly depends (Yonge) simply on whether it is protected by
enclosure in a separate diverticulum or lies exposed in an open
groove of the intestine. Mitra (1901) showed that extracts
of the style have a strong amylolytic reaction. This is abun-
dantly confirmed in a large number of genera. Style extracts
do not digest fats, proteins, inulin, cellulose, or cane sugar.
They break down starch and glycogen completely with the
production of glucose. The temperature-optimum (32° C.) is,
as would be expected, lower than that of the ptyalin in the
saliva ; on the other hand, the amylase found in the digestive
gland of the lobster has a very high optimum — about 56° C,
according to Yonge (1924). Temperature optima in enzyme
reactions deserve further inquiry as a limiting factor in geogra-
phical distribution.

In connection with the digestion of carbohydrates there
is one point which will merit further investigation. Cellulose
is an important ingredient in the diet of all animals living on
plants, yet in vertebrates no cellulose- splitting ferment has
been identified with certainty. Biedermann and Moritz (1898)
found a cellulose-splitting ferment in the digestive gland of
the snail ; and the same authors also detected a cytase in the
hepatopancreas of Astacus ; but Yonge (1924) was unable to
detect any cellulose-splitting action in extracts of the diges-
tive gland of the Norwegian lobster. In herbivorous mammals
cellulose splitting appears to be effected by micro-organisms
living symbiotically in the alimentary tract.

For a detailed account of intermediate carbohydrate meta-
bolism other sources must be consulted. However, mention
must be made of the part played by the pancreas in verte-
brates. The mammalian pancreas contains, in addition to
the exocrine acini, groups of cells known as " islets of
Langerhans." In Teleosts the islet tissue is wholly or partly
separate (Rennie) from the acinar elements. In 1889 Mering

98 COMPARATIVE PHYSIOLOGY

and Minkowski produced glycosuria (increase of sugar in the
urine) by the removal of the pancreas in the dog. Later it
was shown that ligation of the pancreatic duct produces
degeneration of the acini ; extraction of the islet tissue then
yields a product which, when injected into the diabetic dog was
found by Banting and Best (1922) to lower the blood sugar, and
raise the respiratory quotient. Macleod and his co-workers
have shown that extracts of the islet tissues in fishes relieve
the diabetes produced by extirpation of the pancreas in rabbits.
The exact stage at which insulin, the internal secretion of the
pancreas, influences carbohydrate metabolism is still not fully
understood. The lowering of the blood sugar produced by in-
jection of insulin in mammals is accompanied by convulsions,
which can also be reproduced according to Huxley and Fulton
(1923) in frogs. Macleod finds that injection of insulin re-
duces the blood sugar content in fishes, and that removal of
the islet tissue causes hyperglycoemia.

Absorption and passage of Foodstuffs along the Gut.— The
motion of food in the mammalian gut depends upon more
than one mode of response on the part of the circular
and longitudinal musculature of its wall. There are rhyth-
mical movements which tend to churn the food without
moving it predominantly in one direction ; these are an
intrinsic property of the muscle itself, though subject to
inhibitory and excitatory nervous control by the splanchnic
and vagus nerves respectively. Further, when mechanically
stimulated, intestinal muscle shows a relaxation of tone and
inhibition of movement on the aboral side of the point stimxU-
lated, accompanied by increased force of movement on the
oral side (law of the intestine). This is generally believed
to depend on a local nervous mechanism, the myenteric plexus ;
its function is to keep the food moving on the whole towards
the anal end of the gut.

The food is propelled along the gut by the contraction of
its muscular walls in annelids, molluscs, arthropods, and
echinoderms as well as in vertebrates. The intrinsic rhythm
of the muscular system of the gut is beautifully seen in the
excised alimentary tract of many worms, sea-urchins, and

NUTRITION 99

holothurians. Allen observed that the oesophagus of a species
of Syllid pulsates at a rate of 250 per minute. In most cases
rhythmical movement can be induced by applying gentle
stretcliing, e.g. by attaching a strip of gut to a light lever.
This can be shown in the crop of Helix (Ten Gate), Aplysia
(Brucke), the rectum of Astacus (Ten Gate), or in ring prepara-
tions of the pharyngeal musculature of Aphrodite (Hogben
and Hobson). The property very commonly displayed by
plain muscle in responding to gentle stretching by contraction
is probably very important in the production of churning
movement in the gut. In Lamellibranchs there is very little
muscular tissue associated with the alimentary tract, which
is ciliated throughout ; and ciliary movement is the main
factor in propelHng the food from the mouth to the anus in
these animals. In many LameUibranchs, however, the rectum
pierces the ventricle, and by inserting a cannula in the former.
Ten Gate (1924) has shown that the pressure in the rectum
of Anodon undergoes rhythmical variation in unison with the
heart beat. It seems that the heart may here function as a
means of promoting evacuation of rectal contents. The
nervous control of defaecation has been studied in the lobster
(Homarus) by Miller (191 2). The radial musculature that
controls the closure of the anus (there is no true sphincter)
is suppHed by fibres from the last abdominal ganglion, and
stimulation of these nerves produces rhythmical defaecation
movements (incomplete tetanus). It is doubtful whether the
mechanism of defaecation is simply a segmental reflex, since
such movements can be induced after section of the nerves.

There has been little important work relating to the
physico-chemical aspects of absorption based on invertebrate
material. However, two points are worth mentioning. The
phenomenon of intracellular digestion, familiar enough in the
case of coelenterates, is much more widely prevalent in the
invertebrate phyla than is generally recognised. It occurs
in animals with very elaborate digestive systems such as
Gasteropods and Lamellibranchs. Yonge has photographed
ingested diatoms in the cells lining the stomach of the bivalve,
Mya.

100 COMPARATIVE PHYSIOLOGY

The structure of the alimentary canal in Arthropods has
prompted several investigations into the localisation of absorp-
tion in these animals. The greater part of the tract is lined
with chitin. In the case of the cockroach Sanford's observa-
tions conclusively prove that absorption can take place in the
fore- gut of some insects. On the other hand, the work of
MurHn, Cuenot, Jordan, Yonge, and others clearly shows that
in Crustacea absorption is confined to the mid-gut and the
tubules of the digestive gland. Direct experiment shows that
the fore and hind gut behave as semipermeable membranes to
glucose and salts, which, however, penetrate the wall of the
mid-gut.

References

Bacot and Harden (1922). The Vitamin Requirements of Drosophila.

Biochem. Journ. 16.
BoDANSKY AND RosE (1922). Comparative Studies on Digestion I-II.

Am. Journ. Physiol.
Coward and Drummond (1922). On the Significance of Vitamin A in the

Nutrition of Fish. Biochem. Journ, 16.

Dakin and Dakin (1925). The Oxygen Requirements of Certain Aquatic
Animals and its bearing on the Food Supply. Brit. Journ. Exp. Biol. 2.

Hunt (1925). The Food of the Bottom Fauna, etc. Journ. Marine Biol.

Ass. 13.
Miller ( i 9 i o) . On the Rhythmical Contractibility of the Anal Musculature

of the Crayfish and Lobster. Journ. Physiol. 40.

Orton(i9I2). The Mode of Feeding in Crepidula. Journ. Marine. Biol.

Ass. 9.
Sanford (191 8). Experiments on the Physiology of Digestion in the

Blattidas. Journ. Exp. Zool. 25.

Ten Cate (1923-24). Contributions a la physiologie du coeur de Tanodonte.
Arch. Neerland. Physiol. 8.

Contributions a la physiologie comparee du canal stomaco-intestinal.
I-ni. Ibid. 9.

Yonge (1923-24). Studies on the Comparative Physiology of Digestion.
I-II. Brit. Journ. Exp. Biol. i.

CHAPTER VI

THE CIRCULATION OF BODY FLUIDS

Inasmuch as most activities of an organism are intermittent,
and the intake of material sources of energy localised, there
is usually in Metazoa some arrangement for keeping in motion
the body fluids and regulating this motion so as to meet with
the constantly changing requirements of the tissues. Thus
a consideration of the circulatory system may be conveniently
inserted in connexion with the sources of vital energy before
turning to the more specialised aspects of co-ordination in
the chapters which follow.

The circulatory system subserves two functions : it dis-
tributes food to the tissues, and it supplies oxygen to them
removing carbon dioxide at the same time. The fact that
such intensely active organisms as dragon flies, being provided
with a respiratory apparatus which supplies oxygen directly
to the tissues, are able to exist with a circulatory system that
is practically vestigial, indicates that the primary importance
of the latter lies in meeting what Barcroft terms the call of the
tissues for oxygen rather than in distributing foodstuffs and
products of intermediate metaboHsm. It is also interesting
to note that the smaller representatives of groups the majority
of which possess a vascular system are those which are more
often found to be without a well -developed circulation. That
is to say, the necessity for a circulatory system seems to be
greater where the surface for intake of oxygen is relatively less
compared with the mass of the organism. The regulation
of the oxygen supply to the tissues by the blood is a subject
which has been investigated very little except in the higher
vertebrates. Most of the work on the circulatory system of

lOI

102 COMPARATIVE PHYSIOLOGY

invertebrates deals with the action of drugs and nervous
stimulation on the heart. Molluscs, Arthropods, and Annelids
are probably the only invertebrate groups in v^hich the cir-
culatory system plays an important role.

Other things being equal, the rate at which a tissue can take
up oxygen depends upon the amount of blood which flows
through it in unit time. The flow of liquid through a tube
depends upon the force propelling it, the viscosity of the fluid,
the sectional area of the tube, and the length traversed. The
last can be regarded as a constant for present purposes. The
second is probably an important but little known factor in the
circulation. Attention has mainly been focussed on the propel-
ling force, supplied in vertebrates, arthropods, and mulluscs,
by the heart beat, and on the sectional area (constriction or dila-
tation) of the vessels. As the activity of the heart is an inter-
mittent force, the average force exerted by its action depends
upon two variables, the amplitude or force of the individual
beat, and the frequency with which the beats occur. The
output of the heart ir %he vertebrate depends partly upon the
resistance against which it works, being in so far a self-
regulatory mechanism, and is partly determined by a double
nervous control — that of the vagus, which is present in all
craniates including cyclostomes and is inhibitory, and that
of the sympathetic, which is an augmentor factor specially
well developed in the warm-blooded vertebrates. Variation
in the diameter of the vessels determines the resistance against
which the heart works as well as contributing directly to the
rate of flow through, and therefore the oxygen supply of any
given organ. It is provided for by the fact that the arteries,
veins, and capillaries possess contractile elements, plain
muscle in the case of arteries and veins and a special type
of contractile tissue, the cells of Rouget, in the case of the
capillaries. The relaxation and contraction of the contractile
elements investing the walls of the vessels is brought about
by extrinsic agencies (nerves and hormones) and by the local
action of metabolites produced during the activity of an
organ, as illustrated by the work of Bar croft upon salivary
secretion (see Chapter III). The influence of hormones is

THE CIRCULATION OF BODY FLUIDS 103

exemplified by the recent work of Krogh (1922), who has
shown in Amphibia that the pituitary is necessary for the main-
tenance of capillary tone, and that injection of pituitary extract
leads to capillary constriction.

Nervous control of the peripheral circulation in vertebrates
is complex. The constriction of arterioles is brought about
by stimulation of sympathetic fibres which have their cell-
stations in the chain ganglia. Dilatation is probably produced
in the main by a peculiar mechanism which is nervous but
not reflex in the strict sense of the term. It is believed that
certain sensory fibres have branches with motor terminations
in the peripheral vessels, and may thus propagate local dis-
turbances such as arise through mechanical irritation (anti-
dromic action).

The state of constriction of arteries and capillaries has
been studied by the three principal methods : manometer
measurement of the blood-pressure, volume-changes of in-
dividual organs, and rate of flow from vessels artificially per-
fused. The recognition through the work of Krogh that the
capillaries are active agents in determining the resistance to
the flow of blood in the peripheral circulation shows that none
of these methods is wholly satisfactory ; and our views on the
regulation of blood-flow may require considerable revision in
the near future. Stimulation of the central ends of most
sensory nerves produces reflex rise of blood-pressure in
Vertebrates, by exciting the nerve-endings of vaso-constrictor
paths which emerge from the vasomotor centre in the bulb.
In mammals there is a special afferent nerve from the heart
exercising an inhibitory influence on the vasomotor centre
(the depressor branch of the vagus).

Our knowledge of the circulatory system of invertebrates
is very slight. Practically nothing is known of the peripheral
circulation, so that the following account must be confined
to the properties of the circulating fluid and the control of
cardiac rhythm.

Blood of Invertebrates. — The occurrence of respiratory
pigments in the blood of invertebrates has already been treated
under that heading. Two further points are of special interest

104 COMPARATIVE PHYSIOLOGY

from the standpoint of comparative physiology, namely, the
osmotic pressure and coagulative power of blood. The osmotic
pressure of the blood is important as part of the mechanism
of co-ordination. Thus the low blood-pressure of Crustacea
makes it inconceivable that simple filtration could play any
role in their excretory processes (as in the glomerular function
of vertebrates), because of the high content of proteins whose
osmotic force must be overcome in some way in order to
effect any separation of water and diffusible salts from the
blood. In land vertebrates the osmotic pressure of the blood
is a constant quantity for any species. The following figures
(cf. Bayliss' " General Physiology ") for the freezing-point
depression (Bottazzi) bring out a fact of great bionomic interest :

Mammal (whale)

. . o-65°-o-7°

Reptile (turtle)

.. o-6i'

Teleost

. . o-76°-i-04°

Elasmobranch

. . 2-26°

Sea water

.. 2-3°

The osmotic pressure of the blood of elasmobranchs is the
same as that of the sea water in which they live ; and the same
is true of all marine invertebrates ; and, as was first shown
by Fredericq (1885), this is not a mere coincidence, for the
osmotic pressure of the blood adjusts itself to that of the
medium over wide changes of concentration and dilution. We
have already mentioned temperature optima of body proteins
and loading tension of respiratory pigments as possible physio-
logical factors in geographical distribution. The extent to
which an animal can rapidly adjust itself to change in osmotic
pressure is doubtless an important aspect of the ecology of
estuarine forms. Some modification in response to changed
conditions was shown by Dakin in Teleosts, in which we see
the beginnings of a fixed osmotic condition of the body fluids.
It may be noted in passing that the saUne constituents of
elasmobranch blood are not much more concentrated than
those of the blood of Teleostei ; the difference depends upon
the high concentration of urea in selachian blood. Mines
found that the presence of urea was necessary to ensure
successful perfusion of the elasmobranch heart. The hearts
of marine invertebrates in all cases that have been tried will

THE CIRCULATION OF BODY FLUIDS 105

beat in sea water. Adjustments of osmotic pressure of body
fluids to that of the medium probably takes place through
the gills. There is no known advantage in this arrangement ;
on the contrary, the extreme sensitivity of colloidal systems to
electrolytes implies the advantage of a fixed osmotic pressure
for the delicate adjustments of colloidal equilibrium which
underlie the physical processes of life in any medium. In
this connexion reference may be made to the fact that,
while there is a paucity of reliable analyses of the electrolyte
content of the blood in invertebrates, such figures as are
available indicate a rather higher percentage of magnesium
than is present in the blood of the higher vertebrates. It will
be remembered that sea water is richer in magnesium than in
any other kation with the exception of sodium.

A few words may now be said about coagulation, which also
has some bionomic interest ; for while the greatest care should
be exercised in interpreting biological phenomena as protective
mechanisms, it is difficult to deny any utilitarian significance
to the fact that the phenomenon of blood coagulation is
exemplified in no more striking manner throughout the animal
kingdom than in arthropods, whose segmental structure
renders the loss of limbs a common occurrence. The blood
of molluscs has little coagulative power. That of Crustacea
clots with remarkable rapidity ; and the process is often a com-
plex one, taking place in some species in two stages. The first
stage in the clotting of crustacean blood corresponds more or
less to the coagulation process in that of the Arachnid, Limulus,
whose blood also clots rapidly. In Limulus the protein con-
tent of the blood is almost exclusively made up by the haemo-
cyanin and white blood corpuscles. Coagulation is essentially
a phenomenon of cytolysis (Alsberg and Clark), and can be
prevented by reagents which hinder cell agglutination. For
the first stage of coagulation in crustacean blood, which is
brought about by the cytolysis of special '' explosive " cells,
first recognised by Hardy (1892), immediately they come into
contact with foreign substance, it has been shown by L. Loeb
that calcium ions are not necessary, though they are necessary
for the second, which takes place in lobster blood about a

io6 COMPARATIVE PHYSIOLOGY

quarter of an hour later. This consists of a jellying of the
plasma, but according to Tait (191 8) it is accompanied by
a further cytolysis of corpuscles, which he designates thigmo-
cytes. Loeb states that lobster tissues yield specific coaguHns
for the blood of the same species. According to Tait all the
corpuscles of crustacean blood are actively phagocytic. The
subject has attracted a large number of investigators since
attention was focussed upon it by the pioneer labours of
Fredericq ; but much still remains to be done, especially
from the standpoint of a more modern appreciation of colloidal
behaviour.

The Ostiate Heart o£ Limulus and Crustacea. — Investi-
gation into the circulatory system of Arthropods has been
chiefly focussed upon the task of elucidating the origin and
conduction of the cardiac rhythm. The essential characteristic
of the arthropod circulation is the ostiate heart. Generally
speaking the heart is composed of striated muscle fibres, and
lies in a capacious sinus into which, in those forms which lead
an aquatic life, there discharge the lacunar spaces of the gills ;
in the tracheates the circulatory system is poorly developed,
though an ostiate heart is present.

There is no division of the heart into auricular and ventri-
cular portions. It is filled by way of valve-like apertures,
the ostia, which communicate directly with the pericardial
sinus during diastole, but remained closed during the systolic
phase. From the heart proceed the larger arteries. In
Crustacea, at least in Decapods (Baumann), there are valves
at the cardiac end of the larger arteries, so that the pressure
in the arterial system does not fall to zero in diastole. A well-
developed arterial system is found in Limulus, scorpions and
decapod Crustacea. In those forms in which the arterial
system is well developed the heart is a powerful organ, and
beats with a frequency comparable to that of the cephalopod
heart.

Nevertheless, from such data as we have at our disposal,
it would seem that the peripheral resistance is very low, doubt-
less on account of the fact that the arteries discharge into
lacunar spaces and not into a network of capillaries. Brucke

THE CIRCULATION OF BODY FLUIDS 107

and Satacke (19 12) measured the pressure of the blood in the
abdominal aorta of the lobster by means of a water mano-
meter, using hirudin as anticoagulant, and found that with
the heart beating at a rate of 51 per minute the average
pressure-reading was only about 8 mm. of mercury.

The most extensive studies on the circulatory system of
any invertebrate are the investigations of Carlson on the heart
of the king-crab, Limulus. In this arachnid' the heart retains
the original segmental character. It attains to a length of about
six inches in a full-sized animal, and thus provides excellent
experimental material. There are eight pairs of ostia. The
major arteries are located towards the anterior end of the
heart. Histologically it is a syncytium of striated fibres. On

Fig. 25. — Innervation of heart of Palinurus (Carlson).

its dorsal aspect there is an elongated median nerve-ganglion,
which can be easily detached. There are also a pair of laterally-
disposed nerves connected with the abdominal gangHa and
with the cerebrothoracic ganglion of the central nervous
system. Electrical stimulation of the extrinsic cardiac nerves
from the abdominal ganglia leads to acceleration of the normal
rhythm ; the same result may be produced by stimulating
the ventral nerve-cord or abdominal ganglia themselves after
transection of the cord behind the brain. Stimulation of the
brain or its nerve-connections with the heart leads to diastolic
arrest. There is thus a double augmentor-inhibitor mechanism
by which the activity of the heart is subjected to control by
the central nervous system, as in vertebrates. The same has
been shown to be the case in Maia (Bottazzi, Polimanti) and
Palinurus (Carlson) among decapod Crustacea.

io8 COMPARATIVE PHYSIOLOGY

However, the origin and conduction of the heart-rhythm,
in Limulus at all events, is quite different from what is generally
accepted to be the case in vertebrates. In Limulus each
contraction starts at the posterior end of the heart, and travels
forward to the region from which the main arteries have
their origin. When the heart is removed from the body,

10 9 S"7

Fig. 26. — Heart and nerves of Limulus (Carlson).
7-8 Cardiac nerve from brain ; 9-1 1 cardiac nerve from abdominal

ganglia

cardiac ganglion ; p.n.c. lateral nerves.

so that all its connexions with the central nervous system are
severed, it still continues to beat with its normal rhythm of
about twenty to thirty a minute. From this it might be thought
that the heart-rhythm is an inherent property of the cardiac
muscle, as in vertebrates. From two lines of experimental
evidence, however, Carlson has shown conclusively that this
is not the case. He first investigated the effect of stripping
off the ganglion referred to above. When the cardiac ganglion

iliiii__MAiJLLi^

Fig. 27.-

-Inhibition of heart beat of Limulus by electrical stimulation
of brain with weak current (Carlson).

is removed the heart ceases to beat. The normal isolated
heart will only beat in plasma or sea water if the ganglion
is left intact ; after removal of the ganglion the heart may be
made to contract rhythmically in isotonic sodium chloride
after immersion for half an hour ; in this respect it agrees,
however, with vertebrate striped muscle, which acquires a
regular contractile rhythm in the absence of calcium salts.
If, in the isolated heart, the ganglion is divided into four

THE CIRCULATION OF BODY FLUIDS 109

portions, alternate segments being then stripped off, the heart-
beat continues only in those portions to which the remaining
ganglionic tissues adhere, the rhythm of the pulsating regions
being now unco-ordinated.

The last-mentioned experiment leads to a consideration of
another peculiarity of the cardiac rhythm in Limulus. In
vertebrates, not only is it true that the beat arises spontaneously
in the muscle-cells of the sinus venosus, or cardinal sinuses
(Fishes) ; but the conduction of the excitation from one part of
the heart to another takes place through the muscular tissue.
The muscular continuity of the auricles and ventricles in the
higher vertebrates is effected through a bundle of modified
muscle fibres, the bundle of His. Carlson has shown that
section of the heart of Limulus without damage to the median
ganglion does not interrupt the synchronism of the two halves ;
but section of the median ganglion alone abolishes the co-
ordination of rhythm, each half now beating with a rhythm of
its own. It appears fairly certain, therefore, that the origin
and co-ordination of the heart beat in Limulus depends upon
the rhythmical discharge of nervous stimuli from the cardiac
ganglion ; there is no precise parallel to such a mechanism
in vertebrates outside the central nervous system itself. It
is to be noted that stimulation of the heart-muscle or median
ganglion itself with an interrupted current produces a tetanus ;
the period during which the heart remains refractory to a
second stimulus is not protracted to the extent so charac-
teristic of cardiac muscle in the vertebrate.

Some investigators have sought to apply the same interpre-
tation of the origin and conduction of the heart-beat to decapod
Crustacea. It is to be noted, however, that the median gang-
lion of Limulus is an organ sui generis. There may, of course,
be ganglion cells in the heart-muscle of Crustacea ; but there
is no structure comparable to the cardiac ganglion of Limulus
in their gross anatomy, Limulus is not very closely related
to Crustacea in the phyletic scale. Moreover, the embryonic
heart of Limulus, which is at first composed of smooth muscle,
beats before any nerve-fibres reach it : ontogenetically it is
a myogenic heart. And it is quite likely that one would find

no COMPARATIVE PHYSIOLOGY

the common ancestor of Limulus and the Crustacea in a form
which, like Peripatus, possessed only smooth muscle, and
therefore, presumably, a heart whose rhythm was myogenic.
In support of the neurogenic interpretation of the crustacean
heart beat Hoffmann (19 12) has emphasised certain peculiari-
ties which are shared by the electrical response of the crustacean
and arachnid heart. In both cases each major variation
corresponding to a single mechanical contraction displays
small superimposed oscillatory deflections. It is, however,
to be noted that the exposed heart of the crustacean beating
in situ often shows twitching movements superimposed upon
the regular beats ; and the same is true of the isolated heart
beating in sea water or other suitable media. In connexion
with the supposed neurogenic origin of the heart-rhythm in
Crustacea, it is worthy of note that the frequency of the beats
has a temperature coefficient which is fairly high, as may be
seen from the following data taken from Brailsford Robertson *s
paper on Ceriodaphnia :

Ql0ll72I° 1*92

Qioi3723° 1*90

Q1015725'' 2' 20

Qioi9729° 2- 18

The Circulatory System in Molluscs. — Among the Cepha-
lopod molluscs the circulatory system reaches the highest
degree of specialisation met with among the invertebrate
phyla. The anatomical relations are briefly as follows.
Venous blood is collected by the caval vein which bifurcates
to form two branches supplying the gills. These afferent
branchial vessels dilate at the base of the gills into rhythmically
contractile branchial hearts, which drive the colourless venous
blood through the gill-capillaries. Oxygenated blue blood is
collected by the efferent vessels directly to the auricles of the
systemic heart, the ventricle (which may be divided) pumping
the arterial blood into the aortae. This would seem to be a
much more efficient device for supplying oxygenated blood to
the tissues at high pressure than the arrangement which exists
in fishes, where the force of the heart's heat has to overcome
the resistance of the gill-capillaries as well as that of the body

THE CIRCULATION OF BODY FLUIDS in

capillaries. And it is an experimental fact that the blood-
pressure in the arterial system of the Cephalopod is much
higher than in that of the fish. The aortic pressure
of the octopus has been determined by several observers,
first by Fuchs, who found that it varied between 25-80 mm.
Hg ; probably the latter figure is more representative. But
the efficiency of the mechanism becomes even more striking
when the relation between the rhythm of the branchial hearts,
of the systemic hearts, and of the respiratory movements is
studied. The pulsation of the branchial hearts precedes
that of the systemic hearts, but at the same rate (about 30 beats

L. Aurfcle
Ventrfcle
Fig. 28. — Heart of Cephalopod.

per second), and the frequency of the heart-beats is almost
identical with that of the respiratory movements. The means
by which this synchronism is maintained is a fascinating field ;
and several investigations have been made into the innerva-
tion of the cephalopod heart, notably, with modern methods,
by Carlson and by Fry. The later work of H. Fredericq
provides the clearest account of the co-ordination of cardiac
rhythm in this group. Fredericq 's experiments were carried
out on the octopus, and deal with the relation of the branchial
and systemic hearts. When the ventricle is cut off from the
auricles by a ligature, the ventricular beat ceases immediately.
This apparently is not due to a severing of the functional
continuity of the conducting tissue, if any, as in the case of

112 COMPARATIVE PHYSIOLOGY

the Stannius experiment on the heart of the frog, where the
rhythm originates in the sinus. Several considerations which
are borne out by the behaviour of the isolated heart point to
the contrary. When the isolated ventricle is perfused with
sea water or a suitable saline medium, it ceases to pulsate at
once when the pressure falls below about 20 mm. of water,
but starts again when the pressure is raised above that level.
The increase of pressure beyond this point does not lead
to any augmentation of the amplitude ; a certain minimum
driving force is necessary to initiate the rhythm. The cessa-
tion of ventricular movement on ligation of the auriculo-
ventricular junction is due to the fact that the auricles can no
longer maintain the requisite critical pressure necessary to set
the ventricles in action. If, however, the auricles are separated
by a ligature from the efferent branchial vessels they continue
to beat synchronously with one another and in unison with
the branchial hearts. This co-ordination might reside either
in the nature of the contractile tissues, which is unlikely ; or
in some reflex mechanism. That a reflex mechanism may
co-ordinate the pulsation of one auricle with its fellow is
indicated by the fact (Fredericq) that while stimulation of
either intact visceral nerve brings the heart to diastolic stand-
still, stimulation of the peripheral end of the cut nerve pro-
duces inhibition of the auricle on the same side only, whereas
stimulation of the central end produces inhibition on the
opposite side. Here, then, is a mechanism by which one
auricle may reflexly inhibit its fellow, by afferent nerve- con-
nexions which recall in one respect the famihar depressor
nerve of the mammalian heart. That a reflex mechanism
underlies the co-ordination of activity of auricles and branchial
hearts is suggested by the effects of direct electrical stimula-
tion, when the beating of the heart has been stopped by the
faradisation of the inhibitory nerves. If in this condition
one branchial heart is stimulated, it contracts, and its contrac-
tion is followed first by that of the branchial heart of the
opposite gill, and secondly by that of the two auricles, which
is the normal sequence of cardiac rhythm. It is also found
that if under similar conditions the renal vein is stimulated,

THE CIRCULATION OF BODY FLUIDS 113

contraction of the branchial hearts followed by that of the
systemic auricles results. There is thus reason to believe
that there exists in cephalopods, as in vertebrates, a complex
reflex mechanism co-ordinating the constituent parts of the
circulatory system.

The dependence of ventricular rhythm on the stretching
force of the auricular beat in Octopus is a phenomenon which
is characteristic of the ventricular muscle of other mulluscs,
e.g. Anodon, Pecten, Helix, Aplysia, etc. The empty heart
of the mollusc does not beat. This fact depends upon a wide-
spread characteristic of plain muscle, studied by Straub,
Buddington, and others in denervated preparations of the
integumentary muscle of the earthworm. Stretching suffices
to induce rhythmical contraction in isolated rings of the seg-
mental muscles of Lumbricus. In the mammalian bladder,
distension is the normal stimulus for contraction ; and Carey
(1921) has recently found that by introducing fluid into a dog's
bladder under considerable pressure, the walls of the latter
gradually become thicker, come to simulate microscopically
those of the heart and pulsate rhythmically at a rate of two
hundred per minute. Carlson (1906) has utilised the fact that
the empty heart does not beat in seeking for evidence of
cardiac augmentor nerves in his investigations on numerous
molluscan genera. Inhibitory cardiac nerves appear to exist
in all groups (lamellibranchs, cephalopods, and pulmonates)
with the possible exception of marine gasteropods {e.g. Aplysia).
There is a well-developed augmentor nerve supply in Aplysia.

Straub (1904) has made a careful study of the eflFect of
pressure on ventricular rhythm in Aplysia. Between the
pressures 4-20 mm. of water the pulse volume is almost
directly proportional to the pressure applied, and as the auricle
supplies a pressure of about 30 mm. (water), the heart normally
works under approximately the optimum conditions for ven-
tricular contraction. In gasteropods, however, the peripheral
circulation is poorly developed ; a few larger vessels empty
into ill- defined lacunae which finally converge upon the gill-
sinuses. The total resistance of the peripheral circulation
is so small that the blood-pressure is not of the same order

114 COMPARATIVE PHYSIOLOGY

of magnitude as the maximal force which the heart can exert
(Straub).

The same is true of Lamellibranchs. Willen and Minne
found that the ventricular pressure rises from lo mm. (water)
in diastole to 35 mm. in systole in Anodon. The heart-beat
of Anodon is comparatively slow, as may be seen from the
following table, for which temperature data are not in all
cases available : —

Laynellibranchsj per min.

Mytilus . . . . . . 10-15 • • • • Carlson.

Mya . . . . . . 5-10 . . . . ,,

Cardium . . . . . . 15-17 . . ■ . ,,

Anodon 2-4 (at 15° C.) . . Koch.

Gasteropods :

Helix 53 Lang

Cephalopods :

Octopus . . . . . . 55 . . . . . . Bauer.

Sepia . . . . • • 35 • • • • • • Fry.

Koch (19 1 6) has made a very careful study of the heart-
beat in Anodon, which has also been recently investigated
by Ten Gate (1923). The former investigator describes a
phenomenon which is of great interest from the bionomic
standpoint, namely the variation of the pulse-rate with open-
ing and closing of the shell. In the open condition the heart
of an animal may beat five times as quickly as when its shell
is closed. Saturation of the water with oxygen also increases
the frequency of the heart-beat in Anodon.

Periodic Reversal of Rhythm in the Tunicate Heart. — ^An
interesting field for investigation is met with in Tunicates,
whose circulatory system displays a peculiarity which has been
familiar to zoologists for nearly a century. The heart in the
larger ascidians is an elongated cylindrical tube looped round
the base of the pharyngeal sac, so that one end lies dorsal to
the gut and may extend to the nerve gangUon, the other end
being ventral on the opposite side of the pharnyx. The heart-
rhythm is made up of alternating series of beats progressing
respectively from the dorsal (advisceral) and ventral (abvis-
ceral) ends of the heart. In Ascidia mentula, as described by
Day (1921), each pulsation-series lasts from two to four

THE CIRCULATION OF BODY FLUIDS 115

minutes and consists of from twenty to forty beats according
to the size and condition of the animal. In any individual
the number of beats in each advisceral or abvisceral series
tends to be approximately constant. Each pulsation series is
followed by a pause of from ten to twenty seconds before
reversal of the direction of
the contraction wave occurs.
Acceleration is caused by sever-
ing the nerves. Extirpation of
the ganglion or incisions in the
tunic produce an increase in
the number of beats per series,
a slight acceleration of the pul-
sation rate and a diminution of
the pause before reversal. After
a time varying from an hour
to a day the original rhythm
returns. In Ascidia atra, which
has been investigated by Hecht
(19 1 8), the heart has a well-
marked nodal constriction about
halfway along its length. Most
individuals show a marked pre-
ponderance of advisceral over
abvisceral beats. Thus in one

series of observations the beats succeeded one another as
follows :

Fig. 29. — Anatomy of Ascidian to
show circulatory system (after
Hecht).

Abvisceral
Advisceral

36 36

44 51 65

Various hypotheses have been suggested to account for this
phenomenon of reversal. In particular one may mention the
suggestion that the significant factor is back-pressure from the
peripheral circulation. This is dismissed by Hecht on three
grounds. First, because it occurs in the isolated heart.
Secondly, because it can be abolished by raising the tempera-
ture to a certain height (about 35° C. in Hecht 's experiments),
when the direction of the contraction wave remains constant.
Thirdly, because the peripheral resistance cannot be very

ii6 COMPARATIVE PHYSIOLOGY

high on account of the structure of the circulatory system.
The blood-spaces are continuous ; the cavities in which the
blood flows are derived from the primitive blastocoel of the
embryo, where the latter is not filled with strands of con-
nective tissue and gelatinous material. Only in the larger
ascidians do connective tissue cells lining these lacunae limit
well-defined tubes, such as those formed by evagination into
the substance of the test.

In conclusion, some mention is due to the peculiar character
of the blood in some Tunicates. In the colourless plasma,
there are in addition to amoeboid and spherical unpigmented
corpuscles, others packed with large granules of green colour
and markedly acid reaction. It is curious to note that the
green pigment is a compound of Vanadium and a protein.
The low oxygen-capacity of the blood makes it practically
certain that this is not a respiratory pigment. Clotting as
in crustacean blood is brought about by agglutination.

References

Alsberg (19 1 4). The Proteins of the Blood of Limulus. J. Biol. Chem.

19.
Brucke and Satacke (1912). Der Arterielle Blutdruck des Hummers.

Zeitschr. Allg. Physiol. 7.
Dakin (1908). The Osmotic Concentration of the Blood of Fishes, etc.

Biochem. J. 3.
Carlson (1904-5). Nervous Origin of the Heart Beat in Limulus. Am.

Journ. Physiol. 12.
(1906). Chemical Conditions of Heart Action in Limulus. Ibid. 16.

(1905). The Comparative Physiology of the Invertebrate Heart.

Ibid. 13-14.

(1922). A Note on the Action of Drugs on the Invertebrate Heart.

Journ. Gen. Physiol. 4.

Carlson and Meek (1908). On the Mechanism of the Embryonic Heart

Rhythm. Am. Journ. Physiol. 21.
Fredericq (19 1 3). Recherches experimentales sur la physiologic cardiaque

d'Octopus. Arch. Int. Physiol. 14.
Fry (1909). The Influence of the Visceral Nerves on the Heart of Cepha-

lopods. Journ. Physiol. 39.
FucHS (1908). Beitrage zur Physiologic des Kreislaufes bei den Cepha-

lopoden Pflugers Archiv. 60.
Hart (1924). L'action des ions sur les mouvements rhythmiques du sac

muscule cutane du lombric. Arch. Neerland. Physiol. 9.
Hecht (191 8). The Physiology of Ascidia atra. Am. Journ. Physiol. 45.

THE CIRCULATION OF BODY FLUIDS 117

Hoffmann (191 i). Ueber Elektrokardiogram von Evertebraten. Arch
Anat. u. Physiol. 191 1.

Koch (1916). Der Herzschlag von Anodonta. Pflugers Archiv. 156.
LoEB (1903). Ueber die Bedeutung der Blutkorperschen fiir die Blut-
gerinnung. Virchow's Archiv. 173.

PoLiMANTi (1913). Beitrage zur Physiologic von Maia. Arch. Anat.
u. Physiol. 1 91 3.

Straub (1904). Fortgesetzte Studien am Aplysienherz. Pflugers Archiv
103.

Tait (1918). The Blood of Astacus fluviatilis. Q. J. E. P. 12.

CHAPTER VII

ENDOCRINE CO-ORDINATION

Life as understood by the biologist is a term used to denote
a certain combination of reactions exhibited by organic systems
under the influence of external stimuli. In the opening
chapters attention was focussed upon the mechanisms which
underlie the characteristic manifestations of vital activity in
animals. From these we turned to consider the material
exchanges from which the energy of these activities is derived.
The activities of an organism are related in a definite manner
to external conditions ; and it is this co-ordination or integra-
tion of response which gives rise to the conception of the
individual as a physiological unit.

The integration of response presents two aspects. In
all multicellular animals other than sponges, visible manifesta-
tions of activity involve first a receptive surface upon which
the stimulus operates ; secondly, a structure, the effector organ,
specialised for the performance of the appropriate response ;
and thirdly, intervening between these, a mechanism of
co-ordination by which the disturbance is propagated from the
seat of stimulation to the region at which the response is
carried into effect. Co-ordination of this kind is twofold :
all multicellular animals except sponges and a few aberrant
organisms (Mesozoa) of uncertain phyletic relationship possess
specialised conducting tissue in continuity with both receptor
and effector units, which tissue constitutes the nervous system ;
in addition stimuli may in some animals give rise to the pro-
duction of specific chemical entities which make their way
through the body fluids to the organs which they are capable
of activating. Apart from providing a means for co-ordinating

ii8

ENDOCRINE CO-ORDINATION 119

metabolic processes the circulatory system acts as a channel
through which can diffuse the substances known as hormones.
Up to the present no clear evidence of the operation of internal
secretion has been demonstrated outside the vertebrate series.
Accordingly, to illustrate the nature of endocrine regulation,
one cannot select a better instance than the discovery of
BayHss and Starling (1902), who first gave conclusive proof of
the functional role of hormones in the animal body.

Secretin. — By the end of the nineteenth century it had
been established that pancreatic secretion was not entirely
prevented by severance of the nervous connections of the
gut, though it was well recognised that the secretion of the
pancreatic juice followed the signal provided by the entry of
the gastric contents into the small intestine. Having found
that the introduction of acid alone into the denervated gut was
adequate to elicit activity of the secretory cells of the pancreas,
Bayliss and Starling injected into the circulation an aqueous
acid extract of the mucous lining of the duodenum, thereby
activating the pancreas. Subsequent experiments showed
that the liberation from the enteric mucosa of a soluble product,
called by these authors secretin^ provides the immediate
stimulus to pancreatic secretion. Carried by the blood-stream
to the resting gland, " secretin " possesses the property of
producing secretory activity in the pancreatic cells ; the pro-
duction and translocation of the hormone is a mechanism by
which pancreatic activity is co-ordinated with the entry of food
into the small intestine.

Nature 0! Chemical Co-ordination.— With this example before
us .we may proceed to define what is meant by endocrine or
chemical co-ordination and the kind of evidence on which one
can rely for proof of its existence. The essential character-
istics of a hormone are illustrated by secretin, in that a hormone
may be defined as a substance set free in the body by the
activity of a localised organ and capable of evoking a specific
response in tissues remotely situated from its seat of origin.
In one minor respect, however, the production of secretin
differs from that of some other well-established cases of
internal secretion in that the hormone is not produced from a

120 COMPARATIVE PHYSIOLOGY

glandular structure (endocrine organ or ductless gland)
specifically concerned with its manufacture. In defining the
criteria for ascribing to any organ an endocrine function,
attention need only be directed for our present purpose to
the regulation of specific responses in effector organs. If it
is known that an organ contains a substance which evokes a
specific local response in some effector unit {e.g, action of
adrenaline on the pupil or of pituitary extract on frog melano-
phores) its endocrine function may be estabHshed by one (or
both) of two methods. In the first place, it may be shown
that when responses which can be specifically evoked in isolated
effectors by the presence of its active material occur
spontaneously in the intact animal, they are associated with
the Hberation into the blood-stream of a substance having the
same properties as its extract, and in amount significantly
greater from that which is normally present in the blood.
One may formulate the alternative as follows. Given the
fact that an organ contains a specific constituent which evokes
response in an isolated effector unit, it is legitimate to conclude
that such an organ is of endocrine function, when the con-
sequences of its removal upon the given effector system may
be compensated by introducing the active material of its
extract into the circulation.

The study of endocrine mechanisms has been prompted
to a very large extent by clinical interests which lie outside
the scope of the present discussion. Examples of the action
of hormones will be found in other chapters dealing with
developmental and metabolic processes. The pages which
follow will be concerned with the part played by internal
secretions in regulating response of a type which is not met
with in the mammalia, and to consideration of such evidence
as suggests the presence of hormones in the lower animals.
Both examples selected for this purpose deal with the regula-
tion of colour-response in cold-blooded vertebrates.

In fishes the controlling mechanism has been shown by
Pouchet , V . Frisch , and others to be nervous . The melanophores
of fishes are directly supplied with nerve-fibres, and the effects
of local section and stimulation of nerve-trunks conclusively

ENDOCRINE CO-ORDINATION 121

indicate that the melanophores are subject to direct nervous con-
trol. In Amphibia and Reptiles there is as yet no histological
proof of innervation of the pigmentary effector system ; and,
as v^^ill be seen, there exists an alternative method of interpreting
the regulation of pigmentary changes. We shall first consider
the significance of the adrenal glands to the colour responses of
reptiles. A word or two may be inserted with reference to the
comparative physiology of adrenaline, concerning which there
are a few observations which suggest further lines of inquiry.
Adrenaline in the Animal Kingdom.— Oliver and Schafer

Fig. 30. — Action of adrenaline on the heart of Pecten.

(1895) fi^st discovered that the adrenal medulla of the
mammal yields an extract which has a powerful pressor action
on the circulation of the mammal. Later the researches of Lew-
andowsky, Langley, and Elliott showed that in general adrenal
medullary extracts, or adrenaline, the active substance isolated
by Takamine, produces the same effects upon plain muscle in
vertebrate animals as the stimulation of the sympathetic
nerves, e.g. inhibition of intestinal tone and rhythm, dilatation
of the pupil and constriction of the arterioles in all verte-
brates and acceleration of the heart in mammals and birds.
It is, however, to be noted that the action of adrenaline is

122 COMPARATIVE PHYSIOLOGY

not confined to muscles which have a sympathetic innervation ;
and the validity of arguing that an effector organ such as a
melanophore possesses a sympathetic nerve supply from the
fact that it is acted upon by adrenaline, may be questioned.
Adrenaline has a powerful action on both the ganglion and
heart-muscle in Limulus (Carlson), the heart of the crab
Maia (Hogben and Hobson), and the intestine of the crayfish
(Ten Cate). It also acts in very great dilution as an excitant
to the plain muscle of molluscs and annelids, as illustrated

Pjg, 31.— Action of adrenaline on the crop of Aplysia.

by the oesophagus of Aphrodite and Aplysia (Hogben and
Hobson) and of Helix (Ten Cate), also the heart of Pecten
(Hogben and Hobson) . The poison with which the cephalopod
kills its crustacean prey is a natural base, tyramine (CgHpH.-
CH2CH2NH2) of closely allied structure to adrenaline
(CeH3(OH)2CH.OH.CH2NH.CH3) and to certain other
phenolic amines for which Barger and Dale^ have described
a similar physiological action on mammalian tissues. In
this connexion it is interesting to note the presence of adrenaline
in the salivary glands of a toad {cf. Chapter III).

ENDOCRINE CO-ORDINATION 123

The adrenaline-secreting cells of the medulla have a close
ontogenetic relation to the post-ganglionic nerve cells of the
chain ganglia of the sympathetic nervous system ; unlike
other glands the adrenals are innervated by pre-ganglionic
fibres. Owing to the characteristic chrome-staining reaction
discovered by Henle their distribution has been studied in a
number of groups. In the lower vertebrates chrome-staining
cells often occur in the chain ganglia of the sympathetic
nervous system, and in elasmobranchs there are paired chromo-
phil bodies associated with each pair of ganglia, the cortex
of the mammalian gland being represented by a separate
structure, the interrenal body. Gaskell (19 14) has described
chrome-staining cells in the ganglia of Leeches and of those
Polychaetes which have a well-developed musculature in
connexion with the blood-vessels. These chromophil cells
are, according to Gaskell, the cell-bodies of those efferent
neurones which supply the vascomotor system. Gaskell
regards them as the common ancestral representative of both
the adrenalin-secreting cell and post-ganglionic neurone.
He obtained from extracts of the sympathetic ganglia of the
leech an action like that of adrenaHne on the mammaHan uterus.
Adrenalin-secreting cells have also been described in the
mollusc. Purpura^ by Roaf. In view of the widespread action
of adrenaline in the animal kingdom its distribution and possible
function in invertebrates are problems which would well
repay investigation.

The Role o£ Adrenaline in Reptilian Colour-response.— Apart
from the indications provided by the work of Cannon in
favour of the view that excitement phenomena in the mammal
are associated with the Hberation of adrenaline into the blood-
stream, the only evidence for the functional activity of the
adrenals so far available is to be drawn from the study of colour-
response in the reptile. The early experimental work on
reptilian colour-response centred round the bionomic aspect
of the problem, more especially in correcting erroneous
teleological descriptions of protective colour- change in the
chameleon prevalent at the time and still widely credited. We
shall here consider the phenomena of colour- response in the

124 COMPARATIVE PHYSIOLOGY

so-called homed toad, Phrynosoma, the only reptile which
has been made the subject of recent experimental treatment
in this field. In the horned toad bodily colour varies
between a fuscous shade (associated with the outward migration
of pigment granules into the cell processes of the melanophores,
which ramify among the yellow interference cells, which lie
mmediately below the epidermis) and a pale cinnamon bluff
tint (resulting from concentration of the pigment granules in
the cell-body of the melanophore). When it is kept upon a
neutral background, there is seen to be a daily rhythm of
colour-change in Phrynosoma. At night the melanophores
are contracted, giving the skin the appearance of pallor. In
the early morning the skin becomes uniformly dark through
** expansion " of the melanophores. But during the heat of
the day — in its warm natural surroundings — the melanophore
pigment contracts, and as evening approaches a second
expansion supervenes, until night descends. The condition
of pallor in natural surroundings is, therefore, seen at night
and at midday. In the cooler parts of the day the skin is
fuscous.

The exact part which light and temperature respectively
play in promoting this sequence of reactions has been made the
subject of investigation by Parker (1906) and Redfield (1918).
From their work it appears that the melanophores of this
Uzard respond to light and darkness, warmth and cold, in the
manner generally characteristic of reptiles, i.e. bright illumina-
tion and low temperature promote darkening of the skin,
while warmth and darkness bring about pallor. Light and
heat interact so that the effect of the latter predominates at
extremes of temperature, and it is thus that, in natural condi-
tions, living as these creatures do in a warm climate, pallor
intervenes during that part of the day when the temperature
rises to a maximum.

But in addition to this response to direct illumination,
the horned toad reacts in bodily coloration to the character
of the substratum and to mechanical irritation or disturbance.
Any nocuous stimulus, such as electrical excitation of the roof
of the mouth or the cloaca, evokes pallor in fuscous individuals

ENDOCRINE CO-ORDINATION 125

within a few minutes. After being kept for several days on
a white background the animals remain fuscous in bright
light.

Redfield has recorded the results of careful experiments
carried out with a view to locating the receptors involved in
these modes of response. He finds that the local exclusion
of light from and application of heat to restricted areas of the
skin produce a local contraction of the melanophores in the
region to which the stimuli are applied, without affecting the
colour of the skin in other parts of the body. Furthermore,
local illumination produces a local expansion of the
melanophores without affecting the pigmentary effectors of
other regions, while local reduction of temperature maintains
locally a state of melanophore expansion already established,
though it apparently cannot initiate. These experiments
admit the possibility that melanophore response to heat and
to light, in the case of animals kept on a neutral background,
is propriogenic in character and results from the direct reactivity
of the pigmentary effector organs to incident stimuli. It is
not conceivable that these results could be brought about by
hormonic regulation through the circulatory system. Red-
field states that such local responses can be evoked after the
entire nerve- supply of the affected region has been severed.
If this is so, there would seem to be no alternative to accepting
his conclusion that light and temperature can act directly
upon the melanophores, without the intervention of either a
freely-circulating hormone or a nicely-adjusted system of
reflex arcs.

Nevertheless, Redfield is driven to the conclusion that
there is, superimposed on this primary reactivity of the
melanophores of the horned toad to fight and heat, a co-
ordinating mechanism which will account for the generalised
condition of pallor following " excitement," and the peculiar
modification of the normal reaction to light in virtue of the
background upon which the animal is kept. For the latter
response the appropriate receptor is the eye ; since blinded
individuals no longer display the apparently adaptive response
to the brightness or darkness of the substratum. If horned

126 COMPARATIVE PHYSIOLOGY

toads which have been kept upon a background of dark cinders,
are transferred to one of white sand, they become noticeably
paler after one day and reach within five days a condition of
maximum pallor. But when the eyes are bhndfolded, exposure
to the same surroundings for several weeks does not result
in the disappearance of the dark condition ; and the results
of carefully controlled experiments showed that this is not
due to the mechanical influence of the bandage, but can only
be interpreted on the assumption that response to the nature
of the background arises through stimuli received in the first
place through the organs of vision. In considering the
possibilities of pigmentary control through the eyes, it is of
interest to contrast the relatively slow and accumulative nature
of the response to background with the more rapid primary
reaction to incident light. In the case of the horned toad,
the intimate nature of the background response was not
investigated by Redfield, who devoted his investigation of
the co-ordinating mechanism in reptilian colour-response
to the peculiarly characteristic phenomenon of excitement-
pallor, and by an ingenious and painstaking series of experi-
ments has arrived at the conclusion that here too hormonic
regulation plays an important part in the process.

The method adopted to induce pallor by nocuous stimula-
tion was the application of a faradic current to excitable areas,
such as the cloaca or roof of the mouth, a procedure which,
as we have seen, results in general contraction of the melano-
phores throughout the skin of the whole body. That such
treatment results in uniform melanophore contraction in dark
animals even after the denervation of definite areas of the skin,
such as can be achieved by severing all the nervous connections
of a limb, suggests at once that the melanophores are accessible
to stimuli received through the circulatory system. Several
lines of experimental evidence converge to this conclusion ;
in particular, the possibility of evoking pallor by transfusion
of blood from an excited animal. It has been known for some
time that extracts of the suprarenal medulla induce melanophore
contraction in fishes and amphibia. In the horned toad
destruction of the cord between the eighth and thirteenth

ENDOCRINE CO-ORDINATION 127

vertebrae prevents pallor after faradic stimulation of the mouth.
In these circumstances the body-cavity may be opened without
producing melanophore- contraction ; and when the adrenals
of such lizards were stimulated, contraction of the melanophores
occurred throughout the entire body after the lapse of only
a few minutes. It did not occur, however, in the hind limb
after ligature of its arterial supply. When the ligature was
removed, on the other hand, the skin of the leg rapidly assumed
the condition of extreme pallor. Striking collateral evidence
in favour of the possibility that adrenal secretion determines
excitement-pallor in reptiles was provided by analysis of the
blood-sugar content, which is known to rise in Mammals
when adrenalin is Hberated into the circulation, as during
excitement ('' emotional glycosuria "). Redfield found that
the blood-sugar content was significantly higher in lizards
after the production of pallor by nocuous stimulation. To sum
up briefly the evidence from these and other experiments
bearing on adrenalin, it seems clear : (i) that adrenalin causes
the contraction of Reptilian as well as Amphibian melanophores ;
(2) that the adrenal glands of Reptiles contain a substance
which has the same action as adrenal extracts obtained from
Mammalian glands ; (3) that there is indirect evidence that
adrenal activity is associated with *' excitement " in Reptiles
as in Mammals ; (4) that removal of the adrenals in Phrynosoma
in most cases prevented melanophore contraction in response
to nocuous stimulation ; and (5) that in addition Reptilian
melanophores are capable of direct response.

The Function 0! the Pituitary Gland.— The part played by
secretion of the pituitary gland in determining the normal
rhythm of colour-change in amphibia is a further instance
of endocrine phenomena of a type which lies outside
the province of mammalian physiology. There is reason to
believe that adrenalin occurs in the invertebrate phyla. The
pituitary gland, on the other hand, is a specifically vertebrate
structure from which no active substance has yet been isolated
in pure form, so that it is impossible to speculate with profit
upon the possible existence of an analogous mechanism in
other phyla. In fact, strictly speaking, the pituitary belongs

128 COMPARATIVE PHYSIOLOGY

only to the Craniata, and the existence of its homologue in
Amphioxus and Tunicates is doubtful (de Beer).

The physiological activity of extracts of the posterior lobe
of the pituitary was, like that of extracts of suprarenal medulla,
first revealed by the classical researches of Oliver and Schafer
(1895). Extracts of the pituitary of mammals, birds, reptiles,
and fishes (teleostean and elasmobranch) have a powerful
excitatory action on the mammalian uterus and upon mammary
secretion (Herring, 1913 ; Hogben and de Beer, 1925).
Extracts of the pituitary of all classes of Amniota, Amphibia,
and Teleostei exert a pressor- diuretic action on the mammal.
This has not so far been obtained from extracts of the elasmo-
branch pituitary ; such extracts, however, are of a much
lower order of activity than equivalent extracts (by weight of
gland substance) prepared from the teleost pituitary, when
tested on the virgin uterus (Hogben and de Beer). Extracts
of the pituitary of fishes, amphibia, and amniota have also a
specific depressor action on the circulation of the bird. While
Krogh (1922) has given evidence for the conclusion that
pituitary secretion contributes to the maintenance of capillary
tone in the frog, it is an interesting fact that none of the above
responses have any established physiological as opposed to
pharmacological significance. In fact, the posterior lobe of
the mammal, while a storehouse of probably several substances
of prodigious activity and not a little interest to the phar-
macologist, has not yet been proved conclusively to have any
functional significance. There is, on the other hand, the clearest
evidence for regarding pituitary secretion as the main factor
in co-ordinating the pigmentary responses of amphibia to
the changing conditions of its environment.

The anatomical and bionomic aspects of Amphibian colour-
response may now be summarised briefly as follows. The
effector organs concerned with colour-response in Amphibia
are the dermal and epidermal melanophores and the dermal
xantholeucophores. It seems probable that in Amphibia
the activity of the pigmentary effector depends on migration
of pigment-granules rather than the movement of the cell as
a whole. The rhythm of colour-response in Amphibia depends

ENDOCRINE CO-ORDINATION 129

on a balance of natural factors of which humidity, temperature,
oxygen supply, and illumination are the most significant.
Dryness in the terrestrial species promotes pallor (contraction
of the melanophores). Complete melanophore expansion can
only occur in natural surroundings in the presence of moisture.
Warmth tends to produce pallor. Most commonly bright
light has the same effect, at least after continued exposure.

Colour-responses require periods of several hours or even

days to reach their maximum intensity. The following

citation is taken from Laurens' paper on Amblystoma larvae : —

" I. Expansion of the melanophores of seeing larvae in

the light one and a half to two hours.

"2. Expansion of the melanophores of eyeless larvae in
the light two to three hours, and contraction of the melano-
phores of seeing larvae in darkness two to three hours.

"3. Contraction of the melanophores of eyeless larvae in
darkness four to five hours.

** 4. ' Secondary ' contraction of the melanophores of
seeing larvae in the Hght three to five days, and ' secondary '
expansion of the melanophores of seeing larvae in darkness
five days or more."

Laurens does not give the temperature conditions to which
such periods are subject.

" Adaptive " response to background colour depends on
stimuli for which the retina is the receptor. The characteristic
darkening of the skin in the breeding season among Anura
is probably independent of the internal conditions incident
to reproduction, and results from exposure to optimum con-
ditions for melanophore expansion during the period of coupling
and ovulation. The normal pigmentary responses of the frog
are summarised in the table on next page.

The synchronous character of the colour responses which
occur in Amphibia implies that a regulatory mechanism
controls their pigmentary reactions. To earlier workers
nothing seemed more natural than the assumption that this
mechanism must be the nervous system. Every effort was
directed to seek a solution of the problem along this line.
The result has been a bewildering conflict of evidence from

130

COMPARATIVE PHYSIOLOGY

Normal Pigment Responses of the Common Frog

Background.

20° C.

xo»C.

Light background —

(«) Dry

Pallor

Generally pallor

{b) Moist

Pallor (Epidermal

Darkening

melanophores

ex-

panded)

Shade or dark background —

(«) Dry

Pallor

Partial darkening
(Epidermal melano-
phores contract)

(b) Moist

Darkening

Darkness —

(«) Dry

Pallor

Partial darkening

(b) Moist

Darkening

Note. — " Pallor " implies contraction,
both dermal and epidermal melanophores.

darkening " expansion of

equally reliable and competent witnesses on such matters as
the effects of nerve section and stimulation, spinal transection
and extirpation of sympathetic ganglia. After the researches
of Oliver and Schafer (1895) into the physiological effects of
adrenalin a new horizon appeared. Three years later Corona
and Moroni noticed the effect of adrenalin in producing
melanophore contraction in frogs. This observation was
extended by Lieben's researches (1906). But though the
action of adrenalin in promoting melanophore contraction
later suggested to several continental workers, notably Fuchs,
the possible alternative that endocrine factors intervene in
the control of pigmentary responses, it was only when the
action of pituitary extract was tested that it became possible
to envisage a second endocrine system capable of inducing
melanophore expansion.

This line of attack was suggested by the researches of
Adler (19 14), Allen (19 17), and Smith (19 16), who developed
the technique of hypophysectomy in Anuran larvae and called
attention to the condition of extreme pallor which supervenes
in consequence of ablation of the pituitary rudiment. They
did not, however, appreciate clearly the effector character of
the pigmentary change, which was first pointed out by Allen

Fig. 32. — Frog on right injected six hours previously with'extract of
the pituitary of a foetal ox : ! left, control.

Fig. 33. — Two frogs 19 days after operation : on left anterior lobe
only removed ; on right posterior and anterior lobe removed.

ENDOCRINE CO-ORDINATION 131

(19 19). Shortly after it was noticed (Huxley and Hogben,
1 921) that Urodele larvae exhibit darkening of the skin after
pituitary administration, and Swingle (1921) recorded a
similar effect after implantation of the pars intermedia in
tadpoles. At the same time, the writer initiated a series of
experiments on the results of pituitary injection to put to
critical test the hypothesis that these effects were due to a
freely circulating autocoid of the pituitary gland.

The results of these preliminary experiments (Hogben and
Winton, 1922) may be briefly summarised under four
headings : —

1. The pituitary {p. intermedia and nervosa) of Mammals,
Birds, Amphibia, and Fishes, contains a specific stimulant
capable of inducing contracted melanophores of adult and
larval Amphibia (Anura and Urodela) to undergo maximum
expansion.

2. This property is not shared by such drugs {e.g. histamine)
as simulate the physiological action of pituitary extracts in
other respects ; nor is it shared by other tissue-extracts
examined, namely those of spleen, brain, testis, ovary, pancreas,
liver, muscle, adrenal, pineal, and salivary gland.

3. The melanophore response is a very sensitive indicator
of pituitary extracts. The gland of a single frog contains
sufficient to induce darkening in some fifty or more individuals
of the same species.

4. The action of the melanophore stimulant in pituitary
extract is direct and local, independent of concomitant vaso-
motor effects. Taken in conjunction with the phenomena
described in Anuran tadpoles by Smith (1920) and Swingle
(1921), these data present a strong presumptive case for the
view that pituitary secretion forms an important factor in the
regulation of Amphibian colour-response.

In a later series of experiments the effects of removal of
the whole pituitary, controlled by comparison with effects
of exposure of the brain, section of optic nerves, and removal
of the anterior lobe alone were investigated in the common
frog (Hogben and Winton, 1925) and analogous experiments
to those of Smith and Allen on the axolotl larva of the Mexican

132 COMPARATIVE PHYSIOLOGY

salamander were carried out by the writer (Hogben, 1924).
The outstanding results of these researches may be epitomised
as follows : —

1. After removal of the whole pituitary in adult frogs,
as in Axolotls and Anuran tadpoles (Smith, Allen, and Atwell),
the individuals so treated remain permanently pale with the
melanophores in maximum contraction, although subjected
to conditions which inevitably induce darkening of the skin
in normal animals.

2. Melanophore expansion follows pituitary injection in
hypophysectomised individuals ; but hypophysectomised indi-
viduals so treated regain their characteristic pallor, although
subject to conditions which inevitably evoke melanophore
expansion in the normal animal.

3. A comparison of the minimal standardised dose of a
sample of pituitary extract requisite to induce melanophore
expansion in normal and hypophysectomised frogs, under
conditions in which the intensity of external factors tending
to promote pallor were varied, favours the view that melano-
phore contraction and expansion in the intact animal is
correlated with the amount of pituitary secretion in the
circulation.

The regulation of colour-response by fluctuating pituitary
secretion is thus adequate to interpret all the accredited
phenomena in adult Amphibia, without invoking a direct
innervation of melanophores. We may justifiably conclude
that in Urodeles as in Anura pituitary secretion is controlled
by various (e.g. thermic) receptors in the skin, and is reflexly
inhibited by light acting on the retina. This fully explains
why in the salamander Diemyctilus (Rogers), within the
range of external conditions for which light is the significant
factor, section of the optic nerve was found to result in per-
manent melanophore expansion, although transection of the
cord was without eflPect on the rhythm of colour-response.

It may here be noted that there is no conclusive evidence in
favour of the existence of a direct nervous control of the
melanophores in amphibia. Contraction of melanophores
is brought about in frogs by injection of adrenalin ; but

ENDOCRINE CO-ORDINATION 133

for reasons stated above this is not a sufficient reason for
believing that the melanophores are under nervous control.
In tadpoles injection of extract or feeding with substance
of pineal gland produces an extreme condition of pallor
(McCord and F. Allen). But this reaction does not appear
to be shown by urodele larvae, in which the effects of removal
of the pineal (Laurens) on colour response were found to be
negligible. It is of interest to note that pituitary extract does
not produce expansion in the chromatophoresof the chameleon,
and in the melanophores of the Atlantic minnow, Fundulus,
Spaeth found that it produced contraction. The substance
in pituitary extract which produces expansion of amphibian
melanophores has recently been shown by Dreyer and Clark
to be different from the substances which are responsible
for the mammalian pressor and uterine responses.

A group of phenomena of general biological interest that
appear to be subject to endocrine control are illustrated by
the cyclical activity of the genital ducts (secretory and muscular)
in the female of mammalia and their behaviour during the
several of gestation. Some advance has recently been made
by the discovery of Allen and his co-workers (1924) that
definite changes follow the injection of liquor folliculi into
ovariotomised rodents. In young rodents removal of the
ovary inhibits the growth of the genital ducts, and in older
animals produces degenerative changes in the latter with
cessation of cycHcal oestrous changes. When fluid contained
in the ovarian follicle of the normal individual is repeatedly
injected into a spayed female accelerated growth and secretion
characteristic of the oestrous cycle are induced.

Further References

ScHAFER. The Endocrine Organs. Longmans, Green.
HoGBEN. The Pigmentary Effector System. Oliver and Boyd.
Swale Vincent. Internal Secretion of the Ductless Glands. Arnold.

CHAPTER VIII

THE MECHANISM OF NERVOUS CONDUCTION AND EXCITATION

Before attempting to describe the part played by the nervous
system as an arrangement for integrating the response which
an organism exhibits in everyday Hfe, it is desirable to acquaint
ourselves with what is known of the properties of nervous
tissue as a conducting system. It is assumed that the reader
is already familiar with the elements of nervous histology
and the distinction drawn between afferent or sensory and
efferent or motor nerves. The recognition of the excited
state in nerve depends on the response which it calls forth
in an effector organ, either directly in the case of efferent
nerves, or indirectly through the C.N.S. in the case of sensory
fibres. The stimulation of efferent nerves leads either to
initiating or augmenting activity in an effector organ (excitatory
action) or to diminishing or abolishing response (inhibitory
action).

Reference has already been made to the phenomenon of
inhibition, as illustrated by the effects of stimulating the
cardiac branch of the vagus in the vertebrate, or certain nerve-
fibres connected with the heart in Crustacea and molluscs.
Little is known of peripheral inhibition : the phenomena
of excitation, conduction, etc., in nerve have been chiefly
elucidated through the study of excitatory motor nerves
supplying limb-muscles (nerve-muscles preparation) : that
the phenomena are essentially similar in other types of nerve
is inferred by certain similarities such as the electrical con-
ditions which accompany the propagation of the excited state
in all irritable tissues.

The excitation of a muscle through its nerve involves three

134

NERVOUS CONDUCTION AND EXCITATION 135

distinct events : (i) the initiation of some local change at the
point of application of the stimulus constituting nervous
excitation in the restricted sense employed below ; (2) the pro-
pagation of a disturbance of some kind along the nerve-fibre,
or as it is commonly called the conduction of the nervous
impulse ; (3) the production of some change at the junction
between nerve and muscle. We shall consider each aspect
of the process separately.

{a) Excitation. — Nerve, like muscle, may be stimulated
to activity by means of thermal, electrical, mechanical, and
chemical stimuli. Of these only electrical stimuli are appro-
priate for the manipulative requirements of quantitative experi-
ment. In both cases electrical stimulation may be brought
about either by induced or direct currents. The latter on
the whole yield more instructive results. Up to the present
excitation phenomena have been studied pre-eminently in
amphibian motor nerve.

For the purpose of investigating the nature of excitation,
the excitability of nerve is usually defined in terms of the
minimal intensity of stimulus required to evoke an impulse,
other conditions (duration, etc.) being constant. The first
thing to note is that the local change which constitutes excitation
is a reversible one. When two induced electrical stimuli are
sent into a nerve successively, it is found that the effect of
the second depends on the time which elapses between it
and its predecessor. The receipt of the first stimulus is
followed at first by a brief interval during which the nerve is
incapable of being excited by any strength of stimulus ; this
interval, the absolute refractory period , is followed under certain
conditions by a restoration of excitability which increases
beyond its original value, so that for a further brief interval,
the supernormal phasCy the nerve is capable of being excited
by a stimulus appreciably less than that necessary to evoke
response when presented singly. Thus when two stimuli
which exceed the threshold intensity for a single shock are
applied successively with the lapse of an interval less than the
absolute refractory period, the second is completely ineffective ;
when, on the other hand, two stimuli are applied successively,

136

COMPARATIVE PHYSIOLOGY

the first, if adequate to set up a nervous impulse, for a measur-
able interval after the refractory phase leaves the point of
excitation in a more excitable state, so that with an appropriate
period intervening between the two stimuli a second one
of subminimal intensity may become an effective agent of
excitation. The time-relations in the case of the frog's sciatic
gastrocnemius preparation are represented by Adrian and
Keith Lucas, as in Fig. 34. Their possible bearing on the
phenomena of inhibition and summation in the central nervous
system will be dealt with later. We must first consider the

•01 -02

Time since previous stimulus {seconds)

Fig. 34. — Excitability to second stimulus in the sciatic gastrocnemius
preparation of the frog (Adrian and Keith Lucas).

light they throw on the nature of the local change which
precedes the propagated disturbance in nerve. To account
for both the refractory period and the supernormal phase we
may picture this change as a phenomenon of disintegration ;
if excitation involves dissolution of some constituent in the
neighbourhood of the electrode, we should expect no further
stimulus to have any effect so long as the latter state persists ;
and if the disintegrative process is reversible, it is possible
to conceive why this period should be followed by one in which,
restoration being incomplete, a less potent stimulus is required
to reverse the process.

Further light is shed on the problem by taking into account

NERVOUS CONDUCTION AND EXCITATION 137

the direction, duration, and intensity of constant currents in
relation to excitation. The effect of direction was early
recognised through the work of Pfluger and his contemporaries.
The action of the constant current is characteristically polar,
that is definitely orientated with reference to the surface at
which the stimulus is applied. By placing a commutator for
reversing the current in the circuit for exciting a nerve (non-
polarisable electrodes being, of course, used), it is easily
demonstrated that excitation takes place at the cathode, i£,
the point to which positive ions move through the tissue,
while simultaneously excitability is diminished at the anode, i.e.
at the surface from which kations move away, when the current
is made. When conversely the current is broken, stimulation
again results ; but the polar relations are reversed, excitation
occurring at what was the anode, i.e. at the point to which the
kations now tend to revert. The relation of duration and
intensity of stimulus to the excitation process has been
elucidated chiefly through the work of Lapicque and Keith
Lucas. There exists both a time limit and a limit of intensity
for effective stimulation. If the stimulus is of an intensity
less than a certain amount, it cannot excite, however prolonged
its duration may be ; on the other hand, however great the
intensity of the stimulus may be, it cannot excite, if the
duration of the current falls below a certain value. Ahernating
currents of very high frequency may thus be quite ineffective
in provoking physiological responses. This critical duration
during which a minimal stimulus must operate to produce
a manifest effect varies with different nerves in the same
individual and with corresponding nerves in different species
of animals. As a measure of the time-factor Lapicque has
introduced the constant chronaxie, which is defined as the
minimal duration required for excitation with a current whose
intensity is twice the threshold necessary for excitation when
the duration of the stimulus is indefinitely prolonged. The
significant fact emerging from this line of inquiry is that the
length of the refractory period and the interval which must
elapse for production of summation effects in different tissues is
greater or less according as the chronaxie has a high or low value.

138 COMPARATIVE PHYSIOLOGY

These peculiarities of the excitation process and other
considerations based on the study of currents of varying
intensity point in one direction. Stimulation results in a local
change which is associated with the migration of ions to or
from a surface in the neighbourhood of the point of application
of the stimulus. The conditions of duration and intensity
suggest that a certain minimal concentration of ions at this
surface is an essential feature of the process. In order that
such a minimal concentration of ions may be reached there
must naturally be a minimal quantity of electrical energy
imparted, and there must also be a minimal time during which
the directive force of the electrical current may influence the
migration of ions to and from the surface concerned. The
consequences of such a hypothesis are susceptible to mathe-
matical treatment, as was first suggested by Nernst, later
elaborated on the theoretical side by Hill and subsequently
put to experimental test by Keith Lucas. We are thus in
a position to construct a working hypothesis of the excitation
process. If, through a solution enclosed between tw^o
membranes impermeable to ions of a particular kind, a galvanic
current is for a while allowed to pass continuously in one direc-
tion, there will be a local concentration of such ions at one of
the membranes, reaching a Hmit conditioned by their diffusion
constant. A finite time must be allowed to elapse before any
appreciable increase of concentration can take place at the
membrane, the duration depending upon the intensity of the
current. If the current is reversed before the requisite time
has elapsed the flow of ions will be correspondingly reversed.
From this consideration Nernst was led to seek an explanation
of the inefhcacy of alternating currents of very high frequency
as agents of excitation. If the membranes are indefinitely
separated, the relations which must exist between minimal
intensity of current (i), duration (t)y or frequency (n) in order
that an arbitrary critical concentration may be reached are
expressed by the equations :

k=t\/t (for constant current)
i=k'\/n (for alternating currents)

NERVOUS CONDUCTION AND EXCITATION 139

The formulae of Nernst account for the experimental data
relating minimal frequency, duration, and current strength
necessary to set up an excitation within a restricted range.
By taking into consideration the conditions arising when the
membranes at which ions of opposite sign collect are close
together, A. V. Hill deduced a modified expression relating
the duration of an exciting current to its least strength
with greater accuracy than the formulae of Nernst. Hill's
expression is :

_ A

In this equation A, jit, and 6 are constants depending upon the
distance apart of the membranes («), and the distance from
the membrane excited at which concentration changes are
considered (Z>), the number of ions (p) by which a given quantity
of electricity is carried, the diffusion constant of the ions
involved (k) and an arbitrary factor (C). Thus :

A= ix—Aa cos — , v—e a*

whence it is seen that log B is directly proportional to the
diffusion constant of the ions concerned.

Keith Lucas has evaluated the constants of HilFs equation
and shown not only a remarkable correspondence between
observed and calculated values, but that a number of interesting
phenomena are illuminated by the results so obtained. The
evaluation is straightforward, if the equation is written ;

iyiQ^=i — A

Thus in an experiment of Lapicque the corresponding
threshold-values for current strength and duration (in o'ooi
sec.) were :

1/3

2/3

i"S

175

115

140 COMPARATIVE PHYSIOLOGY

To satisfy the above equation the factor /x^* on the left
must vanish for ^=00 , so that A is the smallest current which
will excite at all, i.e. A =60 in this experiment. Substituting
J, t^ and A for any two pairs, yu may be eliminated by dividing
one equation by the other : 0 then becomes 0*375 ^^^ ^» ^Y
substitution, 0*909. The recalculation of i for corresponding
values of t by means of the formula gives the following : —

? (observed) 175 115 91 76 68 64 61 60

I (calculated) 178 115 91*2 75'9 68-9 65-1 62*4 60

For instances of specific phenomena on which further
light is shed by Hill's analysis, the paper by Keith Lucas may
be consulted. It is of considerable interest to note that the
equation holds equally for direct excitation of muscle, so that
we may infer that the event which initiates the changes
described in an earlier chapter is of the same type in both
tissues. This is a fact which it is most important to bear in
mind when discussing the origin of the electrical variation
in muscle ; and in any attempt to unravel the anomaUes which
beset the study of the relation of electrolytes to muscular
activity {cf. pp. 21, 22).

Minimal Duration of Stimulus for Excitation in Muscle (Lapicque).

Gastrocnemius of Rana temporaria . .
Foot of snail, Helix pomatia . .
Adductor of claw of crab, Carcinus msenas
Mantle of sea-slug, Aplysia punctata
Ventricle of tortoise, Testudo graeca

(b) Conduction. — ^When a nerve is stimulated the dis-
turbance set up at the seat of stimulation is propagated along
each neurone at a measurable rate. The rate of conduction
can be determined directly by observing the difference in the
latent period of muscular contraction, when a nerve-muscle
preparation is stimulated at points along the nerve separated
by a measured distance apart : the difference in the latent
period of contraction then represents the interval taken for
the nervous impulse to traverse this distance. This now
familiar class experiment was first performed by Helmholtz
(1852), before whose time it has been supposed that the nervous
impulse travelled at a rate comparable with the velocity of

. . 0-003

sec

. . 0*048

. . 0*300

. . o-8o

. . 0-82

NERVOUS CONDUCTION AND EXCITATION 141

light ; and the train of thought which prompted Helmholtz
to experiment is instructive. *' As long as physiologists thought
it necessary," he argued, " to refer nerve-actions to the move-
ments of an imponderable or psychical principle, it must
have appeared incredible that the velocity of the movement
could be measured within the short distances of the body.
At present we know from the researches of du Bois Raymond
upon the electromotive properties of nerve that those activities
by means of which the conduction of an excitation is
accomplished are in reality actually conditioned by or at least
closely connected with an altered arrangement of their material
particles. Therefore conduction in nerves must belong to
the series of self-propagating reactions of ponderable bodies.
. . ." Only six years before Helmholtz' experiment Johannes
Miiller had denied the possibility of ever attaining the means
of measuring the velocity of nervous transmission ; the history
of mechanistic thought is paved with such denials.

The rate of propagation of the excited disturbance varies
greatly in different animals. In the following tables some
examples are given : —

Animal.

Rate in metres per
second.

Observer.

Method.

"Frog" ..

Graphical and
electrical.

" Snake "
Fishes—

Carlson, 1904

Graphical.

Esox (olfactory) . .

C 1 2-9*2

Nicolai, 1903

Electrical.

Torpedo . .

14-30

Schoenlein, 1895

Malapterurus . .

33 1

Gotch and Burch,
1896

Cyclostomes —

Bdellostoma

4'5 mandibular,
2' 5 vagus

Carlson, 1904

Graphical.

Molluscs —

Loligo . .

Jenkins and Carl-

Graphical.

Octopus . .

son, 1903

Ariolimax

0-44

Limax

1*24 i

Anodon . .
Arthropods —

O'OI

Pick, 1862

Graphical.

Homarus

6 i

Fredericq and
Vandevelde

Graphical.

142 COMPARATIVE PHYSIOLOGY

In addition to observations on those animals in which it is
possible to isolate suitable lengths of nerve-axons, the observa-
tions of Jenkins and Carlson (1903) on conduction through
the nerve- cord of various worms are interesting, though the
values given are not necessarily comparable with the above,
since they involve conduction across synapses in the C.N.S.
All these data are derived from direct (graphical) observation.

Nemertinea i

Cm. per sec.

Cerebratulus

S'4

Hirudinea :

Aulastomum

Polychaeta :

Cirratulus .

Arenicola .

126

Bispira

606

Aphrodite .

Polynoe

230

Sthenelais .

205

Eunice

475

Nereis

..133

Lumbriconer

eis 262

Glycera

441

From these data it would appear that the rate of conduction
in annelids may be higher than in the lower vertebrates as
exemplified by the hagfish, Bdellostoma. Parker found that
the rate of transmission in the nerve-net of the sea-anemone
was about 13 cm. per second.

Propagation takes place on either side of the point of
stimulation in opposite directions. This is shown by the
effects of stimulating the cut ends of fibres which bifurcate,
and by study of the electrical response which is an invariable
accompaniment of the nervous impulse. In seeking for
further light on the nature of the propagated disturbance,
certain facts clearly emerge from experimental investigation
upon the nerve-muscle preparation. These are : first, that
the disturbance depends upon a supply of energy distributed
along the whole course of the neurone, so that if an impulse
is once set up, its intensity is independent of that of the
stimulus which initiates it so long as other factors remain
constant ; second, that the transmission of the excited state
is associated with an electrical variation which travels along

NERVOUS CONDUCTION AND EXCITATION 143

the nerve at the same rate as the propagated disturbance
itself.

The first proposition implies that a means of treating the
intensity of the nervous impulse by quantitative methods is
available. Adrian (19 12) has sought to estabHsh this on the
assumption that the nervous impulse suffers decrement in
passing through a region of narcosis. If a nerve is narcotised
in a gas chamber, a stimulus applied within the region subject
to narcosis (inside the gas chamber) will evoke a response in
its attached muscle after response to stimulation on the far
side of the gas chamber has been abolished ; the depth of
narcosis, as measured by the time of exposure, required to
abolish response becomes continuously less as the length of
nerve exposed is increased. Thus the ability of the nervous
impulse to traverse a region of narcosis can be employed as
a measure of the strength of the impulse. By arranging
electrodes at intervals along the course of a nerve, enclosed
at intervals along its course by gas chambers in which
measurable lengths are subjected to narcosis, it can be shown
that, when a length of nerve is narcotised until its power of
conduction is almost abolished, it retains its ability to transmit
an impulse through a second region of narcosis of length just
sufficient to abolish response to a stimulus applied immediately
outside the latter. If the impulse is not completely
extinguished in passing through a region of decrement, it
recovers its full capacity to face exposure to the same degree
of narcosis ; on re-entering a normal region it regains its
original intensity. Nervous conduction is thus an all-or-
none phenomenon, i.e. its energy of propagation is independent
of the strength of the stimulus which initiates it, being dis-
tributed along the whole course of the neurone.

To avoid undue abstraction an experiment of Adrian (19 12)
may be described in detail. Four gas chambers as in Fig. 35
are arranged. A and B have a diameter of 4*5 mm. The
diameter of C is 9*0 mm. Nerves are arranged for the experi-
ment as indicated. In all chambers the depth of narcosis
for the length of nerve traversed is identical. The narcotic
used by Adrian was alcohol vapour. In each experiment the

144

COMPARATIVE PHYSIOLOGY

depth of narcosis (measured by the time of exposure) which
was just sufficient to prevent the transmission of a stimulus

Alcohol VapOi.

(C:^

Fig. 35. — (After Adrian — see text.)

through A or B alone, through C alone, or through A and B
successively.

Time to Narcosis.

Expt. No.

Through B only,
4'5 mm.

Through A and B
successively, 9 mm.

Through C alone,
9 mm.

10 min.

6 min.

lo's „

10 ,,

6-5 „

13 „

7 „

I2'5 „

16 „

12 ,,
16 „

8 „
8-5 „

i8'5 „

10 ,,

24 „

16 „

Inspection of these figures shows (i) that the degree of
narcosis which extinguishes a nervous impulse transmitted
through a region of narcosis 9 mm. long is quite inadequate
to extinguish it if the section is divided into two halves separated
by unaffected nerve, i.e. recovery of intensity occurs on
emergence from the region of narcosis ; (ii) that the depth of
narcosis which suffices to extinguish an impulse is practically

NERVOUS CONDUCTION AND EXCITATION 145

identical whether the impulse traverses a given distance of
narcotised nerve (A), or double the distance (A+B) when
there is an intervening region of unaffected nerve between
A and B. This indicates that the impulse emerges from the
region of decrement with fully recovered intensity.

The second point can be demonstrated by observing the
interval between the galvanometer deflections obtained by
the use of non-polarisable electrodes, separated by a measured
distance of nerve. The electrical variation recalls what has
already been observed in muscle ; the excited part at a given
instant is electronegative to a non- excited region. The
electrical variation is an invariable accompaniment of the
transmission of the nervous impulse ; and its existence may
be taken as indicative that reversible movements of ions occur
along the track of the nervous impulse. This suspicion is
strongly reinforced by the fact that the propagation of a nervous
impulse is completely blocked by a region through which an
ascending (but not a descending) constant current is passed
(Bernstein's experiment).

Taking into consideration the fact that the energy of the
impulse is distributed along the whole length of the fibre, we
might picture the propagation of the nervous impulse in one
of two ways : either by comparing it to the ignition of a train
of gunpowder, regarding it as a process which is in the thermo-
dynamic sense irreversible, or interpreting it as a succession
of local reversible changes of colloidal equilibrium along the
course of the neurone. In the first case, it would be predicted
that the nerve would be fatiguable, since it could not have
inexhaustible supplies of the material sources of its energy
of transmission. Conduction would be accompanied by in-
creased metabolism . In the second case , the bioelectric currents
propagated along the course of the nervous impulse would
be the only physically measurable accompaniments of its
passage. The facts that nerves are not fatiguable ; and that
nervous conduction is not, according to the extremely delicate
determinations of Hill, associated with any measurable increase
of heat-production, favour the second alternative. At the
same time certain observers, Tashiro among others, have

146 COMPARATIVE PHYSIOLOGY

claimed to demonstrate increased respiratory activity of nerve
during the propagation of the impulse. It remains to be seen
w^hether these claims will be sustained by further research.

In its most recent form the theory of excitation propounded
by Nernst and extended by Hill and Keith Lucas is amplified
by Lillie as follows. The surface-membrane of a resting
neurone is supposed to be impermeable to certain anions
which are present in greater concentration inside the fibre
than outside it. The kations are free to pass through the
membrane, but the electro attraction of the anions forces
the kations to remain close to the membrane, forming an
electrical double layer. The change in ionic concentration at
the membrane during excitation implies the removal of anions
_j- 4- ^^^ therefore the re-

4-"^ JT "^"^ _L moval of the double

layer (Fig. 36). This
depolarisation is accom-
panied by an increase
of permeability at the
_j__l_-l_-|__|_-^-f--l--j--}- kathode, and escape of

anions at the permeable

area initiates an action
Yic^ 36. current. The local bio-

electric current leads to
a depolarisation of adjacent parts of the membrane which
in their turn become permeable and the seat of a bio-
electric current. Thus the change is propagated along the
nerve-fibre, the action current in one section becoming the
stimulus for the setting up of a similar condition in the next.
According to this view all stimulation is electrical. It is
of interest therefore to note that incident light produces an
electrical variation (Piper) in the denervated eye both of
Vertebrates and Cephalopods.

The balance of evidence is at present in favour of the view
that in the phenomenon of nervous conduction we have to
deal with a process which is essentially of the same nature
as the excitation process itself. It might therefore be expected
that the capacity of the nerve to conduct a second impulse

NERVOUS CONDUCTION AND EXCITATION 147

following immediately after a preceding one would be character-
ised by phenomena analogous to the refractory and supernormal
phases described above. Such has been shown to be the case
by Keith Lucas and Adrian. The method for measuring
the conductivity of a nerve after the passage of a previous
impulse was the same as that employed for demonstrating the
all-or-nothing law. The passage of a nervous impulse is
followed first by a refractory period in which the impulse is
more easily extinguished by narcosis, and then by a supernormal
phase in which it is able to traverse a longer distance through
a region of decrement than in the resting state. These facts
are of great interest in connexion with the phenomena of
summation in the central nervous system and in peripheral
tissues.

Summation in Peripheral Tissues.— So far we have dealt
exclusively with excitation and conduction in nerve-trunks.
The phenomena of propagation of nervous impulses in reflex
arcs show two outstanding peculiarities differing — one in kind,
the other only in degree — from the phenomena of nervous
transmission hitherto described. When a synapse intervenes
in the path of the nervous impulse, conduction can take place
only in one direction — from the afferent to the efferent side.
Furthermore, a single stimulus of any strength whatever usually
proves inadequate to induce a response, although sufficient
to set up a disturbance (measured by the electrical variation)
in the afferent nerve. Thus Sherrington, to whose labours
we owe so much of our knowledge of reflex mechanisms,
records cases in which the scratch reflex (alternate responses
of the flexors and extensors of the hind-limb on stimulating
an area of the back in the spinal dog) did not appear till the
fortieth shock had been administered. The significance of
this phenomenon receives some light from a consideration of
the phenomena of summation in peripheral tissues. We have
already considered a condition in the sciatic gastrocnemius
preparation of the frog in which the minimal intensity for a
second stimulus is lower than for the initial one. This depends
upon the mechanism of excitation at the seat of application
of the stimulus : a probable explanation is that summation

148 COMPARATIVE PHYSIOLOGY

of this type occurs when the first stimulus sets up a concentra-
tion of ions at the point of excitation which does not subside
till the second comes into operation. Summation of this kind
can obviously take place only when the nerve is stimulated in
both cases at precisely the same point. But the phenomena
of the refractory and supernormal states in muscle and nerve
allow for at least two other types of summation. These
summations can occur when the electrodes for the successive
stimuli are not at the same point : that is, w^hen the first stimulus
initiates a disturbance in the nerve itself but does not lead to
response in the peripheral organ. This may occur, either
because the second stimulus fell in the supernormal phase
of excitability of the peripheral tissue ; or because the second
impulse falls in the period in which the nerve itself conducts
with supernormal intensity, so that if the intensity of the
first impulse were just insufficient to penetrate the junction
of nerve and muscle the second may succeed.

An example of the latter is provided by the work of Richet,
Lapicque, and Keith Lucas upon the crayfish claw. The
phenomena in this case are complicated, it may be remarked,
by the existence of a double neuromuscular mechanism ; one
concerned with sustained contraction, and the other with
rapid closing of the pincers. This is shown by the fact that
a diflFerent type of response is given by strong currents of short
duration (twitch-like movement) from that elicited by weak
currents of long duration (protracted closure). A similar
state of aflFairs was shown by Keith Lucas to exist in the
lobster, Homarus, where the curve relating duration of con-
denser discharge to the minimum potential requisite to pro-
duce contraction of the adductor showed a discontinuity on
one side, of which the response was a rapid twitch and on the
other a sustained movement. Lapicque showed a somewhat
similar phenomenon in the claw of Carcinus. Richet showed
that when the abductor nerve of the claw in Astacus is initially
stimulated a small twitch results. If this is followed by two
shocks delivered in rapid succession a more powerful con-
traction results. If the preparation is stimulated every half-
minute alternately by a single shock and by two successive

NERVOUS CONDUCTION AND EXCITATION 149

shocks at an interval of 0-0036 second, the response to the single
shock soon vanishes, while the response to the successive shocks
persists (Fig. 37). At this stage we have therefore a summation
of stimuli individually inadequate to elicit a response, and that
this summation does not depend upon polarisation at the seat
of stimulation is shown by several facts ; in particular, first,
that it is obtained by alternating make and break shocks, in
which case the effect of the second stimulus must be to prevent
the persistence and extent of
the polarisation due to the
first ; and second, that it is
equally well seen when the
successive stimuli are applied
at different points along the
course of the nerve. The

interval which must elapse yig. 37.— (After Keith Lucas.)

between successive stimuH if

summation is to occur was found by Keith Lucas to have
about the same range as the supernormal phase for con-
duction in the nerve itself.

Integrative Action o£ the Central Nervous System.— Con-
siderations of the kind which have been advanced already
open to us, in the words of Keith Lucas, " a whole range of
possibilities in the regulation of nervous activity. According
as we time impulses in the nervous system to follow one another
at a shorter or a longer interval, we can make them less or more
capable of being conducted through any regions of decrement
which the system may contain . If there is a region of decrement
such that normal impulse just cannot pass, then impulses of
moderate frequency may pass it successfully, while impulses
of high frequency may not only fail to pass it, but may by their
frequency prevent other impulses finding their way through."
The last sentence offers a possible interpretation of a very
important phenomenon on which Sherrington lays emphasis
in discussing the integrative action of the central nervous
system. The normal organism is subject to an infinite
variety of stimuli : at any moment it is under the influ-
ence of not one but many stimuli, each adequate under

150 COMPARATIVE PHYSIOLOGY

appropriate conditions to elicit a characteristic response. Yet
in fact its responses are at any minute definite and restricted.

Further Reading

Keith Lucas. The Conduction of the Nervous Impulse. Longmans,

Green.
LiLLlE. Protoplasmic Action and Nervous Action. Univ. Chicago Press.
Hill (1921). The Energy Involved in the Electrical Discharge in Muscle

and Nerve. Proc. Roy. Soc, B. 92.

CHAPTER IX

THE ANALYSIS OF BEHAVIOUR IN ANIMALS

In the last chapter the attempt has been made to indicate
some of the evidence available for an understanding of the
nature of the processes involved in the transmission of the
excited state. It will now be necessary to inquire into what
is known regarding the way in which stimuli normally present
in the surroundings operate to produce the characteristic
and more or less appropriate sequence of responses which
constitutes an animal's behaviour. It will coincide best
with our present treatment to consider the question from a
phyletic standpoint.

The simplest form of response is the direct reaction of
an effector organ to external stimuli. This has been already
met with in the pigmentary effector organs of the shrimp
and lizard. The reaction of the iris to light in vertebrates
and cephalopods is another instance. But in both cases we
find a co-ordinating mechanism superimposed upon the local
form of reaction. In the osculum of the sponge with its collar
of primitive muscular elements we have an apparently purely
local mechanism. No co-ordinating arrangement is shown
by the behaviour of other oscula when a neighbouring one
is stimulated. Their normal function is to react to running
water by relaxation and to still water by closure, response
occurring after an interval of several minutes.

Neuroid Transmission.— Undifferentiated protoplasm pos-
sesses the property of propagating the excited state. This
is best seen in the co-ordination of ciliary movement, so
important an aspect of behaviour in large numbers of marine
animals. Ciliary motion is metachronial ; the cilia do not

151

152 COMPARATIVE PHYSIOLOGY

beat simultaneously but in regular succession one after the
other with reference to the axis along which the ciliary current
is maintained. It is this that gives the active ciliary epithelium
an appearance like a field of corn blown by the wind. The
metachronial rhythm of cilia does not depend upon mechanical
stimulation of one cell by the ciliary activity of its neighbour.
There is abundant evidence for this statement, but perhaps
the most convincing is provided by experiments of Kraft,
who showed that by cooling a zone of ciliated epithelium till
mechanical activity subsided, effects of mechanical and
thermal stimulation on one side of the quiescent zone were
transmitted to the other. There is no clear evidence of
nervous agencies at work in connection with ciliary
mechanisms ; and it seems fairly certain that in general the
disturbance which underlies the metachronial rhythm is
propagated through the undifferentiated protoplasm of the
cells. Chambers has shown experimentally that in cells
with intercellular protoplasmic bridges the effects of an
injurious stimulus are transmitted from one cell to another.
Where there are no demonstrable protoplasmic connexions
this is not the case. Such transmission is usually called
neuroid, although there is as yet no evidence of its depending
upon the same type of mechanism as true nervous conduction.
The Elementary Nervous System.— The simplest possible
form of neuromuscular system is met with in the tentacles
of some sea- anemones {e.g. Cerianthus). Here the entire
complex consists of a sensory cell ending in a process
which arborises round the underlying muscle fibres. The
single cell combines all the functions of receptor, afferent, and
efferent neurone. The internal processes tend to run towards
the base of the tentacle, and with this, according to Parker,
is correlated a polarity in the response to stimulating the
tentacle at its distal and basal extremities respectively. To
quote from Parker, " when the tip of a tentacle is vigorously
stimulated the whole tentacle is likely to respond, but when a
part lower down in the side of the tentacle is stimulated, the
reaction is chiefly from this point proximally."

However, in the trunk region of Actinozoa and in general

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 153

among Coelenterates, a more specialised arrangement exists.
Nucleated cell-bodies intercalated in a reticulum of fibrils
connecting the sensory cells and muscular elements are con-
tinuous with one another ; there is between the sensory and
motor apparatus an uninterrupted network.

Ingenious experiments on the physiological properties of
the nerve-net were performed by Romanes more than half
a century ago. Romanes worked on Aurelia, the common
jelly-fish of our coast. From the sensory tentaculocysts
the nerve-net extends inwards over the circular sheet of
muscular tissue round the mouth by whose contraction the
rhythmical pulsation of the swimming bell is brought about.
Excision of all the tentaculocysts brings about a cessation of
the pulsations. If, however, one marginal sense-organ is
left behind it induces a double wave of contraction — one to the
right and one to the left — and the rhythmical movements are
preserved. To ascertain whether special paths of conduction
exist in the nerve-net, Romanes made a series of incisions
in the bell. Spiral, circular, and interdigitating incisions in
the under surface of the body did not prevent pulsation so
long as at any point the nerve-net of the parts separated was
left in continuity, although the muscular coat might be com-
pletely severed. Thus one part of the nerve net is as good
as any other for the transmission of nervous impulses. The
same, according to Parker, is true of transmission in the trunk
region of the sea-anemone. But here definite paths of con-
duction seem to exist in the nerve-net in virtue of the fact
that the fibrils run pre-eminently in an oral-aboral direction.
Hence if the tip of a tongue of tissue cut from the wall of the
body in a longitudinal direction is stimulated, generalised
muscular contraction ensues, while if the tip of a tongue
of tissue cut in the equatorial plane is stimulated, no general
response is evoked. Such polarisation suggests how the
separation of a C.N. S. may in the first place have been brought
about. But the central nervous system of ccelomate animals
is fundamentally different in that the nervous elements are
not structurally continuous. The experiments of Romanes,
which were independent of and contemporaneous with others

154 COMPARATIVE PHYSIOLOGY

by Eimer, have been amplified and extended by those of Mayer
(1905), Loeb (1906), Bethe (1909), Harvey (1912), and others.
These workers have conclusively shown in various ways that
conduction takes place through the nerve-net and not through
the muscle. This is quite easy to show in the jelly-fish
Rhizostoma (Bethe), where the muscle of the bell and nerve-
net are not coextensive, inasmuch as the " sphincter " is
composed not of a continuous band of muscle as in Aurelia,
the form with which Romanes worked, but of sixteen separate
areas with intervening non-muscular tissue across which the
nerve-net extends.

The Synaptic Nervous System. — In contrast with the nerve-
net of the ccelenterate and the generaUsed modes of re-
sponse associated therewith, we now turn to the synaptic
nervous system of echinoderms, annelids, molluscs, arthro-
pods, and vertebrates. The unit of response in these animals
is essentially locaUsed ; the stimulation of any group of re-
ceptors calls forth response in a strictly limited number of
effectors. To illustrate more concretely the conception of
the reflex the following observations of Bethe (1897) on the
behaviour of the shore crab (Carcinus) will serve.

1. When the eye is blackened to exclude photic stimuli,
gentle mechanical stimulation causes the eye which is touched
to be drawn under the carapace ; both of the antennules are
simultaneously withdrawn. When the same eye is subjected
to a more powerful mechanical stimulus, in addition to the
withdrawal of the antennules and the eye itself the second
antenna of the same side is withdrawn. The opposite eye
and corresponding second antenna are not affected.

2. When the covered eye is subjected to electrical stimula-
tion, the walking legs of that side are brought into such a
position that the body tends to be tilted forwards from the
ground at an angle of 45°. If both eyes are stimulated the
body tilts upward symmetrically, both chelae being extended.
Only one chela is involved as a result of unilateral stimulation
of the eye.

3. Weak mechanical stimulation of the second antenna
provokes first the withdrawal of the antennule of the same

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 155

side and then the stimulated organ itself. Stronger stimuli
lead first to the withdrawal of both antennules,then the antenna,
and finally the eyes of both sides.

The unit of structure in the synaptic system is the reflex
arc. But it should be remembered that the barrier between
one reflex arc and another is physiological rather than structural ;
the injection of strychnine results in a condition in which
stimulation of any receptive area will elicit convulsive move-
ments of all the muscles of the body. The simplest reflex

lanql.'oD

Fig. 38.— Diagram of simple reflex arc {stellate ganglion of Cephalopod)

path is one in which the receptor is represented by the
peripheral arborisations of an afferent neurone whose dendrites
terminate distally in connexion with the cell-body of a motor
neurone. Such an ideally simple reflex arc is probably
realised in the reflex paths which have their synapses in the
stellate ganglion of the Cephalopod (Fig. 38). The stellate
ganglion is connected by the pallial commissures with the
brain and by the stellar nerves with the musculature of the
mantle. Frohlich (19 10) showed that local stimulation of its
surface evokes generalised contraction of the musculature of
the mantle so long as the stellar nerves are intact. If the

156 COMPARATIVE PHYSIOLOGY

latter are cut a purely local response is evoked. The generalised
response which occurs when the stellar nerves are intact is
obtained equally well after section of the pallial commissures
so that the stellate ganglion is completely isolated from the
rest of the C.N.S. It follows that the stellate ganglion is
a centre of reflex activity. Generally, however, reflexes make
use of paths in which at least one intermediate (internuncial
or proprio-spinal in the vertebrate) neurone between the
afferent and efferent elements is involved. This condition is
also seen in Fig. 38.

The more specialised forms of perception concerned with
phototropic, geotropic, and chemotropic reflexes discussed
below involve, in practically all cases, separate terminal
organs, receptors, in connexion with the peripheral ends
of the afferent neurones. A great deal has been written
about the anatomy of chemoreceptors, and photoreceptors.
So little is known of the mechanism of these structures
in vertebrates that the scope of the present work does
not permit of more than a reference to Winterstein's " Ver-
gleichende Physiologic " for further information. A brief
reference is, however, due to the receptors for space orien-
tation, which are essentially alike throughout multi-cellular
animals (and one might almost add plants where, however,
they are unicellular in structure). The statocyst is essen-
tially in all cases a sac lined with cells in connexion with
nervous elements, enclosing a solid body with sufficient
space to move freely under the influence of gravity, which
brings it to rest in such a position as to stimulate one or
another group of nerve-endings according to the position of
the body in space. Kreidl's (1893) ingenious experiments
in replacing the statoliths of Crustacea by iron or nickel filings
during ecdysis, showed that when the normal mechanical
effects of gravity are replaced by those of the magnet the indi-
vidual behaves with reference to the field of magnetic attraction
precisely as it should on the hypothesis that equilibration
depends on the group of receptors on which the statoHth
impinges.

The central nervous system usually consists of more or

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 157

less distinct commissural portions composed of internuncial
axons and ganglionic regions in which the cell-bodies and cell-
connections are located. In Vertebrates the commissural
part (white matter) encloses the ganglionic (gray) matter in
the greater part of the C.N.S. ; while among invertebrates
the C.N.S. is built up of discrete ganglionic and commissural
parts. A peculiarity of the Vertebrate C.N.S. Hes in the fact
that all secreting cells and smooth muscle fibres are innervated
by motor neurones whose cell-bodies are located in subsidiary
(autonomic) ganglia receiving efferent impulses from the
cord but not themselves the centre of reflex activity. They

Cerebral qanqljon

Preganglionic

neurone

f?ccepfor

Postgangllonrc neurone

Fig. 39. — Diagram of pedal ganglion ot Razor-shell.

are thus distributive centres for multiplying impulses to be
relayed to a large number of similar effectors of which simul-
taneous and identical action is required. According to experi-
ments by Drew (1908) on the nervous system of the Razor-
shell clam an analogous state of affairs is seen in the pedal
ganglion of the Lamellibranch (Fig. 39). Drew's experiments
indicate that the pedal ganglion of the Lamellibranch only serves
as a distributive centre for impulses from the cerebro-pleural
ganglia. But in the gasteropod Aplysia, the pedal ganglion
is, according to Frohlich (19 10), a reflex as well as a distributive
centre. The cardiac ganghon of Limulus is a structure which
has probably no close analogy in Vertebrates, unless to the

158 COMPARATIVE PHYSIOLOGY

myenteric plexus of the gut ; which is probably in essentials
similar to a nerve-net.

It is possible that many segmental reflexes in Crustacea
may employ a simple afferent- efferent reflex arc. In Verte-
brates this is rarely the case ; the same internuncial neurones
are involved in a multipHcity of reflex arcs. In the adjust-
ment of response an important result of Sherrington's
analysis of the properties of reflex action is the recognition
of the principle of the final common path. If two receptive
areas when stimulated evoke response of one kind or another
in the same set of muscles, the effect of simultaneous stimula-
tion is in general either one of reinforcement, or the complete
exclusion of one reflex response in favour of the other. When
it is remembered that in normal life the organism is subject
to a large variety of stimuli simultaneously impinging upon
reflex systems the majority of which may make use of a common
path in some part of the C.N.S., the general importance of
this fact in defining the behaviour of the animal at any moment
is obvious.

For the scientific analysis of behaviour in animals possessing
a nervous system the term *' tropism " is a convenient label
for grouping reflexes concerned with bodily orientation in
response to a particular type of stimulus. According to
established usage the term is extended also to modes of
behaviour in metazoa which do not possess a central nervous
system, as well as to protista and plants. Thus we speak
of photo tropisms, geotropisms, thigmotropisms, chemotropisms
to classify reflexes concerned with orientation with reference
to light, gravity, contact, or chemical stimuH. The analysis
of behaviour has progressed chiefly by studying : (i) the isolation
of tropisms or other reflex systems by experimental procedure
involving the exclusion of particular receptors or parts of the
C.N.S. itself ; (ii) tropistic reactions or other reflexes which
normally predominate over other modes of behaviour in the
intact animal ; (iii) the modification of normal modes of
response by physico-chemical means.

The study of tropisms in animals has been advanced
especially through the labours of Loeb, whose most valuable

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 159

contributions concern the behaviour of animals under the
influence of light. That in many animals response to light
predominates over other modes of response is proverbial.
The behaviour of moths, and many other nocturnal insects is
very widely known. When the organism moves towards the
light it is said to be positively phototropic. The opposite
type of reaction, negative phototropism, is well shown by
blowfly maggots. Loeb was first attracted by the bending
of sedentary organisms like Tubularia or tubicolous worms
towards the source of illumination, a phenomenon which
superficially resembles the effect of light on the growth of
plants. He observed that in doing so the animal tends to
take up such a position that its photo-sensitive surfaces are
symmetrically illuminated ; and advanced the hypothesis
that orientation depends upon reflex muscular tone maintained
through the photo-receptors. When the animal bends towards
the light it does so because the tone of the muscles on the side
exposed to light is increased by stimulation of the photosensitive
surface on which the incident rays fall. The consequent
flexion of the body eventually brings it into a position when the
photo-sensitive surfaces are equally illuminated, so that the
muscular tension on either side of the body is balanced.
Anthropomorphic bias ascribes the movement of the moth
towards the candle to the preference of the animal for the
light. This view does not permit us to make verifiable
predictions which can be inferred from a more objective
attitude to the problem. It follows that if the insect's move-
ments are mainly concerned with the direction of the rays,
it will move from a strongly to a weakly illuminated situation,
when conditions are so arranged that by doing so it continues
to move along the path of the incident beam with both eyes
equally illuminated. These conditions are easy to arrange
by projecting on to a tube containing some positively photo-
tropic organism such as caterpillars of Porthesia a slanting
beam of parallel rays, the intensity of one half of the beam being
artificially reduced, and in this way the animal is induced to
move from the light into the shade, which is contrary to what
the anthropomorphic view would lead us to anticipate. If

i6o

COMPARATIVE PHYSIOLOGY

unequal muscle tension reflexly excited by unequal illumination
of the eyes leads an animal to turn into the position in which
(moving towards or away from the source of light) the two eyes
are equally illuminated, it follows also that blackening of the
eye should lead to circus movements. Such circus movements
have been shown by Loeb and many other workers — Parker,
Holmes, Lyon, among others — in various insects, and are
well illustrated by Carrey's (19 19) experiments on the Robber
fly (Protacanthus). These experiments support Loeb's
hypothesis and researches on the unilateral removal of the
cerebral ganglia in insects (cf. Matula, 191 1),
and others confirm the view that the ner-
vous mechanism of muscle tonus is pre-
dominantly unilateral. However, crossed
^^/i reflexes are also involved in the forced

v^^<. movements of insects, since Mast (1924)

has shown that when the posterolateral
border of one eye is illuminated the limbs
of one side move forward and those of the
opposite side backwards, the front feet
towards the light and the hinder ones away
from it, thus showing that the location as
well as the intensity of the stimulus in the
photoreceptors of either side is involved in
the reflexes which underlie orientation.
This does not invalidate the fundamental
conception underlying Loeb's contribution
to the problem, though it shows that a
complete analysis of the phenomenon is a rather more
intricate task than he himself supposed.

FHes with normal eyes ascend either a plane or cylindrical
surface vertically. When one eye is blackened they ascend
a plane surface obliquely, veering towards the unblackened
eye (Fig. 40). If made to ascend a cylindrical surface equally
illuminated on all sides, the insect with one eye blackened
ascends with a spiral motion towards the seeing eye, the
number of spirals depending upon the intensity of illumina-
tion. If the cylinder is obliquely placed so that one side is

Fig. 40. — Effect of
unequal illumi-
nation on climb-
ing insect.

THE ANALYSIS OF BEHAVIOUR IN ANIMALS i6i

in the shade, insects with one eye blackened move in very
different paths on the two sides : *' On the shaded side the
spirals are parallel and the pitch is acute, but in the Hght of
the other side the fly's path is more nearly horizontal, as
would be expected from the different conditions of muscle
tonus resulting from light of different intensities " (Garrey).
Loeb (1890) first showed that if rotated on a turn-table flies
describe circus movements in the opposite direction to the
movement of the table. These, as later shown by Lyon (1900),
do not occur when both eyes are blackened. In Carrey's
experiments flies were rotated on a revolving cylinder
illuminated from above. The normal fly circles towards the
opposite direction, in its ascent thus describing a spiral path.
If the speed is sufficiently increased a horizontal path may be
induced. When one eye is blackened, the forced motion is
intensified if the cylinder is rotated towards the same side and
the vertical component is nullified with a much slower motion.
When rotated with the blackened eye in the opposite direction
to the motion the circus movements are diminished and at
a certain speed the fly ascends vertically.

The conclusion that different regions of the eye in
insects are, as implied in Mast's observations, related to
different reflexes involved in orientation is of special interest
in connexion with a phenomenon studied by Parker (1922)
in young turtles. Newly hatched loggerhead turtles find
their way from their nests to the sea in consequence of at least
three factors, one depending on gravity as shown in their
tendency to move down slopes, one which is a response to
localised retinal images in that they move towards regions
of the horizon which are open and clear rather than interrupted,
and finally a response to colour, since they move towards blue
areas. The first of these may be described as geotropism,
but the last two are types of reaction rather more complex
than those to which the term " photo tropism " is customarily
applied.

In most animals it is but rarely that one set of stimuli
predominates over all others to the extent that light does in
many insects. More frequently the normal orientation and

1 62 COMPARATIVE PHYSIOLOGY

movement of organisms is dependent on such a finely balanced
interplay of phototropic, geotropic, chemotropic, and thigmo-
tropic reflexes that only careful analysis can evaluate their
respective influences. Thus the normal swimming move-
ments of Mysids are in ordinary circumstances unaffected
by removal of either the eyes alone or the statocysts alone.
If, however, both statocysts are extirpated the mysid swims
on its back when illuminated from below. Deprived of
both eyes and statocysts, the animals can effect efficient
orientation only when in contact with a surface (through touch-
receptors in the appendages). Other factors which underlie
seemingly inconsequent modes of behaviour are seen in the
effect of physico-chemical conditions on the sign of a tropism.
It was shown many years ago by Loeb that the larvae of Poly-
gordius, and various small Crustacea, which are in normal
life negatively phototropic, can be made either indifferent to
Hght or definitely positive in their phototropic response when
certain salts and acids are added to the medium. Similar
phenomena have been described by Minkiewicz, Moore,
Ewald, and others, who have also succeeded in producing a
reversal of sign in the normal tropism of various animals by
chemical means. An interesting example is recorded by Kanda
(19 14), who studied the effect of electrolytes on the behaviour
of Arenicola larvae with reference to gravity and hght. In
normal circumstances the larvae of Arenicola are positively
phototropic, swimming towards the Hght by ciHary and
muscular movements. When placed in darkness they swim
upwards, i.e. they are negatively geotropic. When excess
of K or Na ions is present {i.e. on addition of a certain amount
of isotonic potassium or sodium chloride solution to the sea
water) the larvae cease to swim towards the source of illumina-
tion when placed in the light ; they tend to swim in the opposite
direction, becoming negatively phototropic, though feebly
so. The geotropic response of the larvae in darkness is not
affected by such treatment. When, however, there is an
excess of Ca, or Mg ions the larvae swim do\vn wards in dark-
ness ; their behaviour towards gravitational attraction is
reversed, they have become positively geotropic. The reaction

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 163

to light is not influenced to the same extent by excess of Ca
and Mg. As a bionomic illustration of these phenomena
the following suggestion by Loeb is based on the fact that larvae
of Porthesia, which as already observed are strongly photo-
tropic, become indifferent to Hght after a meal. Porthesia
lays its eggs on a shrub. The larvae hatch out in autumn and
hibernate on the ground. Provided that the temperature is
raised sufficiently they can be induced to leave the nest at any
time. When they emerge of their own accord they crawl
directly up the shrubs on whose leaves they feed, always
moving upwards, i.e. in the direction of the rays reflected
from the sky. At the top of the shoot they encounter young
buds, where they feed, and becoming phototropically
indifferent in consequence, are free to move downwards and
thus ultimately find another source of food.

Conditioned or Associative Behaviour.— The phenomena of
reflex action as studied in a mammal from which the cortex
has been removed are predictable, and there is reason to hope
that they will fall into line with the phenomena of peripheral
conduction, summation, and inhibition as suggested in the
last chapter. We have seen that considerable progress towards
a knowledge which will enable us to predict the behaviour
of intact animals has been made by (i) analysing the interaction
of reflex systems brought into play by different classes of
stimuli normally present in the surroundings, (2) determining
the way in which reflex responses may be modified by physico-
chemical factors in the external medium. There is, however,
a further aspect of response which has to be taken into account
in discussing the behaviour of animals. It can be illustrated
well enough by the feeding of minnows. If food in the form
of pieces of meat is presented to it the animal behaves in a
predictable way ; it snaps at the food. If paper coloured to
resemble pieces of meat is presented, for the first few times
the fish behaves in a predictable v/ay by snapping. After a
number of trials which can be predicted within limits by
experiment, the fish no longer snaps at the paper. To the
mechanism which conserves the effects of previous stimulation
psychologists still employ the subjective term '' memory."

1 64 COMPARATIVE PHYSIOLOGY

It is the essence of scientific method that it deals only with
relations that are the result of external observation, that it aims
at expressing these relations in quantitative terms, and that
it employs no assumptions that could be eliminated without
affecting the verifiability of its conclusions. We have then
to ask whether it is possible by objective analysis to obtain any
further light on the mechanism by which the simultaneous
application of two classes of stimuli may enter into the result
of the simple operation of one of them on a subsequent
occasion. Associative phenomena probably play comparatively
little part in the lives of any animals outside the vertebrate
series. This special development is characteristic of mammals
in general ; and is the chief glory of man. Though we are
here concerned primarily with the lower organisms, some
account must be given of those properties which pre-eminently
distinguish mammals from those animals which we elect to
regard as " lower " than them, creatures which in any case
appear to have less complex and less flexible possibilities of
behaviour.

In the objective analysis of associative phenomena an
immense advance has been made during the past two decades
through the work of Pavlov and his associates. Pavlov studied
salivary secretion in dogs. In the absence of the cortex, the
entry of food in the mouth is an efficient stimulus for reflex
salivary secretion. With the cortex intact, sight or smell of
food also evokes secretion. Further analysis led to a funda-
mental distinction being drawn between a type of reflex which
is only known in the animal with its higher cortex intact and
the reflexes which exist in both normal and decerebrate pre-
parations. In the intact animal a previously indifferent stimulus
applied at suitable intervals simultaneously with the applica-
tion of a stimulus which unconditionally evokes a particular
response, eventually acquires the capacity to evoke the response
unaccompanied by the " unconditioned " stimulus. A new
non-inherited reflex has been brought into being, known
as a conditioned reflex ; its previously ineffective agent
is known as the " conditioned " stimulus. This appears to
hold good for a large number of reflexes ; the salivary reflex

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 165

is usually chosen because quantitative measurement of the
amount of secretion can be made by using a cannula.

As Pavlov's work is still somewhat inaccessible to the
English reader, the main points will be outlined. We will begin
with the formation of conditioned reflexes, as follows : —

1. Any event in the external world which affects a sense
organ may in the intact mammal become a conditioned stimulus,
provided that its occurrence coincides with the unconditioned
stimulus a sufficient number of times. Even nocuous stimuli
such as intense electrical stimulation or burning of the skin
may, if systematically accompanied by feeding, cease to evoke
their normal consequences and become a signal for salivary
secretion. Nocuous skin stimulation may thus be formed into
a conditioned stimulus for the unconditioned feeding- reflex,
but not for the unconditioned reflex salivary secretion produced
by application of acid to the tongue. Nocuous stimulation
of the skin over the bones, however, cannot be made a con-
ditioned stimulus for either. The response of a conditioned
reflex is essentially similar to the unconditioned reflex from
which it is derived. If a lighted lamp be made a conditioned
signal for food, the dog not only secretes saliva when the
stimulus is presented, but makes groping movements appro-
priate to food itself.

2. It is necessary that the indifferent stimulus with which
it is desired to form a conditioned reflex should be rigidly
isolated ; an unnoticed accompaniment such as an extraneous
smell, sound, sight or movement of the experimentalist may
otherwise become a new conditioned stimulus and vitiate the
interpretation of the phenomena observed.

3. The indifferent stimulus should operate while the animal
is in a quiescent condition with reference to the unconditioned
system into which it is to be incorporated, i.e. it should precede
by a short interval the unconditioned stimulus.

From what has been said, it follows that, since the animal
is normally subject to an immense variety of stimuli, formation
of new conditioned reflexes could only have chaotic con-
sequences unless there exist definable factors which tend to
inhibit the formation or check the operation of conditioned

1 66 COMPARATIVE PHYSIOLOGY

reflexes. The possibility of isolating a conditioned reflex
for study implies that some inhibitory agency preserves the
normal surroundings of the laboratory from exerting very
much influence. Inhibition in conditioned phenomena
presents four distinct aspects depending upon inherent pro-
perties of the central nervous S3^stem ; these are : —

1. Inhibition by extinction. When an indifferent stimulus
has become a conditioned stimulus for salivary secretion, and
is allowed to act alone on several occasions without the con-
ditioned stimulus, it gradually loses its potency, but recovers
it after a period.

2. Conditional- inhibition. If the conditioned stimulus in
a conditioned reflex is accompanied by another indifferent
stimulus the extinction referred to under (i) takes place more
rapidly than it would if allowed to operate alone.

3. Differential inhibition. Stimuli which resemble a
conditioned stimulus fairly closely may at first evoke response
when applied alone, but lose this efiicacy more readily than the
original conditioned stimulus.

4. Retardation. If in the formation of a conditioned
reflex, the new stimulus precedes by a definite interval (from
a half to three minutes) the unconditioned stimulus, the re-
sponse to the conditioned stimulus when the new reflex is
established is delayed by a corresponding interval of time.

In addition to the above may be mentioned the generalised
form of inhibition of the activity of the cortex known as sleep.
This can be regularly evoked in dogs by application of warmth
or cold to an area of the skin. A further complication is
introduced by the fact that external agencies not only give
rise to inhibition but to release from inhibition. The phenome-
non of " inhihition of inhibition " may be illustrated thus.
By repeated synchronous action of the sound and presentation
of food, an organ note of 1000 vibrations per second becomes
a conditioned stimulus evoking salivary secretion in absence
of the food itself. If repeated too often alone it suffers
inhibition by extinction, but recovers its efficacy with a sufficient
period of rest. If during the indifferent period, there is
superimposed on the now ineffective sound stimulus a second

THE ANALYSIS OF BEHAVIOUR IN ANIMALS 167

indifferent stimulus such as lighting a lamp before the dog's
eyes, the sound immediately regains its efficacy. The sound
and the light were each indifferent stimuli ; their combined
effect depends on the fact that the former had previously
been a conditioned stimulus, i.e. that the latter breaks down
the inhibition to which the former was temporarily subject.
The possible bearing of this phenomenon on the phenomenon
of " attention " is evident. How unexpected fields may be
illuminated by study of the conditioned reflex is well seen in
the phenomenon of experimental neurasthenia. When a new
internal inhibition is in process of formation a preformed
inhibition is weakened. Suppose a spheroidal object is
established as a conditioned stimulus for salivary secretion.
When an ellipsoid differing only in the length of one axis
is presented, it is at first an effective stimulus and ceases to
be so by differential inhibition as already described. Now
suppose that we successively present ellipsoids approaching
more nearly the spheroidal form, pushing the process to the
limit of discrimination, marked changes in the dog's behaviour
occur, firmly established inhibitions disappear, its excitability
is greatly increased. After two months' rest the previous
state is regained and old conditioned reflexes reappear.

Finally, the study of conditioned reflexes as implied in
the last type of experiment opens up a new horizon for the
objective and quantitative treatment of sensation, aside from
the consideration of the sense-organs as physical apparatus.
This may be illustrated by employing the conditioned reflex
to define the limits of discrimination. The sound of a tuning-
fork of 256 vibrations (middle C) is accompanied by electrical
stimulation of the paw until it is established as a conditioned
stimulus. A tuning-fork of 264 vibrations presented as a
signal for withdrawal of the paw evokes response which
subsides on successive presentation before the effect of the
original conditioned stimulus is extinguished by internal
inhibition, as may be tested by applying it. A series of pairs
of forks with diminishing differences in tone are now tried
out till no differential inhibition can be established for a
given pair. The limit of discrimination for sound in dogs is

1 68 COMPARATIVE PHYSIOLOGY

represented by a fraction of a tone. Similarly a fine degree of
discrimination of time, doubtless connected with the mechanism
of inhibition by retardation mentioned above, is shown by
the fact that differential response to a metronome beating 104
and 100 per minute can be established and maintained for
periods of over twenty-four hours. The application of this
method of analysis shows, on the other hand, that dogs and cats
are completely colour-bUnd, their world being defined usually
by differences of light- intensity like an ordinary photograph.

Further Reading

Herter. Mechanische Sinnesorgane U. S. W. Leipzig.
LoEB. Forced Movements, Tropisms and Animal Conduct. Lippincott.
Parker. The Elementary Nervous System. Lippincott.
Sherrington. The Integrative Action of the Nervous System. Yale
University Press.

CHAPTER X

THE FERTILISATION OF THE EGG

In the foregoing summary we have taken the existence of an
animate unit or individual for granted, considering its character-
istic properties, their sources of energy and the way in which
they are brought into working relationship with one another
and with the external world. It is one of the characteristic
properties of animate systems that they are self-propagating.
The quantitative analysis of this property is therefore an
important branch of physiology. In spite of the immense
volume of careful quantitative work in this field, the important
fact that living organisms reproduce their kind — and that the
power to do so is one of the most remarkable features which
characterise living beings— is customarily neglected in physio-
logical text-books or summarily treated from a teleological
standpoint which betrays little sympathy with the advances
which have been made in the last two decades through the
work of Loeb on fertihsation and the rediscovery of Mendel's
method in the opening years of the twentieth century. The
explanation of this omission is to be sought chiefly in the fact
that, while exact knowledge of metaboUsm, muscle, nerve,
and respiration has been advanced chiefly by studies on the
higher animals, practically every important discovery in the
field we are now about to consider is based on material of
too humble origin to interest the medical man. Nevertheless
conclusions derived from these studies are, as will be seen, of
wide applicability. For this reason a very brief outline of
existing knowledge of the mechanics of reproduction will now
be given. The subject is full of interest on the bionomic side
in connection with the possibility that living organisms are

169

lyo COMPARATIVE PHYSIOLOGY

related by common ancestry, and with the attempt to trace
out the significant factors which have directed their past
history on the earth in the light of the evolutionary^ hypo-
thesis. As the treatment of the evolutionary problem forms the
subject matter of a separate volume in this series, we must here
confine ourselves solely to those questions which are amenable
to quantitative analysis, leaving out of account issues which
bear specifically on evolutionary biology. One may in passing
legitimately comment upon the importance still attached by
many physiologists to Darwinian concepts, a fact which is
surprising when it is remembered that the exact study of these
problems does not begin till the dawn of the present century ;
that it was not till more than ten years after the issue of the
*' Origin of Species " that the fertilisation of the egg by a single
sperm was clearly established ; and that the material available
for the study of inheritance by Darwin's contemporaries was
largely derived from popular tradition current among stock-
breeders.

The natural starting-point for a study of the physiology
of reproduction is the fertilisation of the egg. The important
fact that the normal process of fertilisation involves the union
of only one gamete of either sex was first clearly established
by Hertwig and Fol (1875). The recognition of this fact
raises two problems. The entry of the sperm into the egg
normally implies (i) the initiation in the egg of active cell-
division culminating in the formation of a new individual ;
(2) the transference to the zygote of something in virtue of
which the nev/ individual so formed resembles the male as
much as it does the female parent. Kupelwieser (19 12)
found that with sufficiently long exposure of the egg and high
concentration of the sperm, it was possible to bring about the
development of a sea-urchin egg Vvdth the sperm of the common
mussel. The offspring reared resembled the former parent
only. Though the sperm was able to penetrate the egg, its
nucleus was eliminated during the subsequent cell- divisions,
and it therefore made no contribution to the hereditary con-
stitution of the fertilisation product. Hence, though it is not
the rule in nature that a sperm can supply the stimulus that

THE FERTILISATION OF THE EGG 171

initiates development without materially contributing to the
structural characteristics of the individual so formed, it is
legitimate to treat these two issues quite independently. The
nature of the hereditary process will be considered later.

The immediate problem of fertilisation has another aspect
besides the elucidation of the mechanism by which the cleavage
process is brought into operation, namely what factors operate
to bring about contact between the sperm and the egg.
Logically perhaps it would be better to consider, first of all,
the attraction (if any) of the egg for the sperm ; but since our
knowledge of the fertilisation process is largely derived from
elimination of the sperm by the use of physico-chemical
reagents, it is just as convenient to begin with the mechanism
which initiates cleavage.

Parthenogenesis exists as a normal occurrence in nature
in many groups of the animal kingdom, though authentic
cases in Vertebrates (observed with experimental safeguards)
are not known. In some species of stick-insects and gall-flies
the male has been eliminated. The existence of natural
parthenogenesis has prompted many biologists to imitate the
operations by which the agency of the sperm can be dispensed
with in nature. The first fruitful work in this field was done
by Loeb (1899), whose labours have enriched so many and
diverse branches of general physiology. Loeb, who con-
temporaneously with Ringer was a pioneer in studying the
relation of contractile tissues to electrolytes, was impressed
with the fact that stale eggs of marine animals sometimes
show signs of cleavage in process of dissolution, and began
his researches in the endeavour to explore the possibility
of producing artificial fertilisation by an increase in the
hydrogen-ion concentration of the sea water. This was not
in the first place successful. The action of other ions was then
investigated, and successful rearing of swimming pluteus
larvae (a stage which is taken as indicating completely successful
development, since the pluteus is self-supporting) from
uncontaminated eggs of the sea-urchin Arbacia was obtained
by exposing the eggs for a certain period to a mixture formed
by adding a hypertonic solution of magnesium chloride to

172 COMPARATIVE PHYSIOLOGY

sea water. Further experiment showed that this was not a
specific effect of the magnesium ion at all, but could be re-
produced by increasing the osmotic pressure of the solution
with a number of different reagents. Exposure for two
hours to any one of the following mixtures suffices to induce
development up to the pluteus stage in eggs of Arbacia
when transferred back into normal sea water : —

50 c.c. sea water, 50 c.c. 1*25 M, MgCIa ;
90 c.c. sea water, 10 c.c. 2*5 M, NaCl or KCl ;
100 c.c. sea water, 25 c.c. 2'o M, cane sugar ;
80 c.c. sea water, 17*5 c.c. 2*5 M, urea.

The freezing point of sea water is about of the order —

to ^ NaCl. From inspection of the above it is clear that
8

one salient feature is common to all these mixtures — they have
an osmotic pressure higher than that of sea water ; and since
the cell is in osmotic equilibrium with its environment, they
must tend to withdraw water from the egg. The following
table, taken from experiments of Loeb on another sea-urchin,
Strongylocentrotus purpuratus, indicates the optimum con-
centration and osmotic pressure of sea-water mixtures for
different reagents

Optimum concen-

Percentage

Osmotic

Substance.

tration in mols.

dissociation.

pressure in

Cane sugar . .

o'96

—

21*53

Grape sugar

1-04

— •

23*33

CaCl,

0*50

25*57

MgCla

0-49

26-47

LiCI

0-74

27*59

NaCl

o'79

30-28

KCl

0-78

30'95

Less reliable results were obtained in subsequent experi-
ments on Strongylocentrotus than with Arbacia. And osmotic
activation alone did not produce a hundred per cent, yield
in either case. The plutei were in some respects abnormal
in that they did not swim near the surface ; there was a fairly
high mortality ; and — most significant of all — the eggs did
not form the characteristic investment, known as the fertilisa-
tion membrane which is an invariable consequence of normal
fertilisation by the agency of the sperm. This last fact

THE FERTILISATION OF THE EGG 173

suggested the possibility of a more perfect imitation of the

natural process. Inquiry was next directed by Loeb to artificial

membrane formation. In earlier experiments on the action

of the hydrogen-ion, mineral acids were used. It was now

found that exposure to ethyl acetate induced the production

of a typical fertilisation membrane ; and further study showed

that this action was due to the acid hydrolysis product. This

suggested that the fatty acids might be successful agents of

membrane formation. By leaving the unfertilised eggs of

Strongylocentrotus in a mixture of 50 c.c. sea water and 2'8

N
c.c. — butyric acid at 15° C. for about two minutes, all the

eggs are induced to form membranes, when replaced in normal
sea water. Eggs of Strongylocentrotus subjected to this
treatment passed through the early developmental stages.

The combination of both methods was next employed.
Exposure to the action of the fatty acid after treatment with
hypertonic sea water gave better results. When, however,
the reverse procedure was adopted, the eggs being treated
with hypertonic sea water (for a shorter period) after artificial
membrane-formation, success was complete. A hundred
per cent, yield of swimming larvae was obtained ; the larvae
were normal in their behaviour ; and the cleavage process
precisely resembled that of the normally-fertilised egg. Loeb
(1904-5) thus made what must be regarded as one of the most
audacious contributions to mechanistic thought in replacing
that mysterious complex the living sperm by familiar physico-
chemical agencies in its role of activating the developmental
process.

Before pursuing the problem further, it will be as well
to form a more concrete picture of the ground so far traversed.
MacLendon (19 12) has shown that fertilised eggs readily
shrink in isotonic sugar solutions ; but that the unfertilised
eggs do not do so with equal readiness. From this and other
experiments by MacLendon and by Gray on the conductivity
of the egg before and after fertilisation there seems good reason
to believe that an essential feature of normal fertilisation is
increased permeability of the cell-membrane. A variety

174 COMPAPvATIVE PHYSIOLOGY

of considerations converge to reinforce this conclusion, Lyon
and Schackell have shown that eggs become more permeable
to dyes as the result of fertihsation. Harvey (1910) has not
only confirmed this, but shown by intravitam staining with
neutral red a temporary increase at fertilisation of the per-
meability of the egg to alkalies. Again, Lyon (1909) found
that fertilised eggs of sea-urchins, three minutes after insemina-
tion, liberate about double as much oxygen from hydrogen
peroxide as do unfertilised eggs, a fact most readily expHcable
on the assumption that the intracellular catalases are more
accessible to the peroxide in the former case. Thus normal
fertilisation may be regarded as a phenomenon of which one
result is that water tends to be withdrawn from the cell ; we
can imitate this process either by withdrawing water from the
cell (osmotic activation), or by changing the surface properties
of the cell-surface so as to increase its permeability, as appears
to be the effect of butyric acid and of cytolytic reagents.
According to Carter (1924) the formation of a fertiUsation-
membrane is not an essential feature of this change.

The surface change which accompanies fertilisation can
be induced in quite a number of ways. The eggs of the poly-
chaete Nereis (which provides more accessible material for
workers in this country than sea-urchins, as it spawns all
through the summer) can be made to segment (i) by osmotic
activation (Fischer), (2) by exposure for a suitable period to
a temperature of 35° -36° C ; (3) by standing them for ten
minutes in the sea-water exudate of Echinoderm eggs or
Echinarachnus lipolysin. Potassium cyanide, radium emana-
tions, fat-solvents, alcohol, distilled water, saponins, bile-
salts, sera, mechanical injury — the method which can be used
for fertilisation of frog's eggs — have all been employed
successfully as substitutes for the fertilising action of the sperm.
It is not profitable in the limited space at our disposal to select
further instances from an extensive literature deahng with
artificial parthenogenesis in representatives of Echinoderms,
Polychastes, Molluscs, Arthropods, Fishes, and Amphibia.
In general we may say that all these agencies have in common
the property of producing cytolysis at the surface of the egg.

THE FERTILISATION OF THE EGG 175

Favourable material is naturally provided by animals which
spawn into the water eggs which contain relatively little yolk ;
and the eggs of marine animals are best for this purpose, the
physico-chemical equilibrium being in such cases of a more
mobile character. From the rapid advances made of late
years in the technique of tissue-culture it would not seem
unlikely that the initiation of developmental stages without
contact with sperm will be accomplished in our own time in
mammalian ova.

In the Echinoid egg, which up till now has yielded the most
satisfactory material for experimental manipulation, an im-
portant aspect of the union of the sperm and egg is the
immediate increase in oxygen consumption which occurs
after entry of the sperm. At an early stage in the study of
this problem, Loeb suggested that the immediate effect of
the penetration of the sperm might be to promote a series of
oxidative processes. Warburg's (1908) determinations of the
oxygen-consumption of fertilised and unfertilised eggs of
Arbacia confirmed Loeb's prediction. Warburg found that
a quantity of eggs (about four million) in sea water, equivalent
to 28 mg. total nitrogen by the Kjeldahl estimation, took up
4-5 c.c. of oxygen during the first hour after insemination,
while only about o'5-o*7 c.mm. were consumed by the un-
fertilised egg in the same time. Warburg's original experi-
ments were carried out by a titration method (Winkler) ; in
later ones the manometer was used for the gas analyses ; and
readings of the rise in oxygen consumption were not taken
till ten minutes after fertiUsation occurred. From the recent
observations of Cresswell Shearer (1922), using the Barcroft
differential manometer, there emerges the remarkable con-
clusion that the mere contact of the spermatozoon with the
external surface of the egg is capable of increasing the oxidation
rate of the latter by rather more than 8000 per cent, in the space
of one minute. The eggs were fertilised in the chamber of
the manometer, so that there was no interruption of the readings
before and after fertiUsation. Within a minute of the libera-
tion of the sperm the increase in oxygen-consumption starts,
but it takes more than two minutes for the sperm to penetrate

176 COMPARATIVE PHYSIOLOGY

the egg-membrane. That is to say, the increase begins when
the sperm is still only in contact with the outside of the egg ;
and the curve for rate of oxygen consumption (and CO2
production) is steepest during the phase of surface contact.

That the sperm brings about profound changes w^hile still
in surface contact with the egg is shown by LiUie's experiments
on eggs of Nereis, in which the sperm does not penetrate
the cytoplasm till thirty minutes after the initial phase of
fertilisation. Meiosis is initiated by the surface contact
of the sperm ; but if the jelly surrounding the egg is separated
from the latter, taking with it the sperm itself, meiosis is not
followed as in the ordinary course of events by cleavage.
The chromosomes break down without the formation of the
first cleavage-spindle derived from the nuclear apparatus of
the egg.

Turning now to another side of the problem of fertilisation,
there is no need to emphasise the fact that the spermatozoa of
practically all animals (and many plants) are flagellate units.
We have, therefore, to inquire how the motility of the sperm
is so regulated that it is brought into contact with the egg of
the same species. In introducing this question it is necessary
to refer to the normal behaviour of spermatozoa. Spermatozoa
are almost without exception immobile while they remain in
the gonad or generative duct of the male. They usually
become active in the medium in which fertiHsation occurs.
Generally in marine animals this is the sea ; but in some
starfishes the sperms do not become very active in sea water,
unless its hydroxyl-ion concentration is raised, or egg secretions
are added. In mammals the sperm becomes motile in the
secretion of the accessory glands (prostate, etc.) ; but sperm
taken from the epididymis becomes active in Ringer's solution.
The sperm appears to possess no means of taking in nourish-
ment— at least in marine forms, though it may be able to do so
in animals such as bees and bats, in which insemination may
take place months or even years before fertilisation ; and it
therefore has a strictly Hmited term of Hfe. Cohn (19 18)
has shown that the total carbon dioxide output of the sperm
is the same whether its life is artificially prolonged or curtailed

THE FERTILISATION OF THE EGG 177

by influencing its motility. Spermatozoa swim with a spiral
motion, adhering to surfaces with which they come in contact,
a fact which maybe of some significance to the present question.
There are really two problems that arise in this connection,
for we have not only to account for the fact that a sperm may
eventually make contact with an egg, but also to explain how
it is that in general eggs are only fertilisable by sperm of the
same species. It is not necessary to suppose that the same
agencies are responsible for both phenomena.

Considering first the influence of the egg upon sperm
motility, one has to face the possibility that the contact of
sperm and egg is a matter of pure chance, or more strictly,
that the only provision made to ensure fertilisation is the
synchronous ripening of the gonads in the two sexes, and the
prodigious fecundity of the species in animals where coitus
does not occur. Where there is congress of the sexes there
is nothing unlikely in this. What evidence is available has
been chiefly derived from studying the effect of egg
" secretions " on the sperm. In practice this amounts to
observing the effect exercised upon the latter when brought into
contact with sea water decanted from an egg-suspension, and
for brevity called egg-water.

In the case of starfishes the influence of egg-secretion is
conclusive. Since it is highly improbable that immotile
spermatozoa can bring about fertilisation, and since in any
case activity must increase the chance that a sperm will make
contact with an egg enormously, the fact that immobile sperms
of Asterias are raised to intense activity by addition of egg-
water points strongly to the belief that, in these creatures
at any rate, the egg exercises some directive influence on the
sperm. In the absence of quantitative methods for studying
the rapidity of motion in spermatozoa, it is impossible to be
certain that egg-secretions have any action upon the
spermatozoa of forms like Arbacia and Nereis in which the
sperms are normally active in sea water. Some support is
given to the affirmative belief by observations of Loeb who
found that the spermatozoa of sea-urchins, which are immobile
but live for days in isotonic sodium chloride, may be made

178

COMPARATIVE PHYSIOLOGY

intensely active by addition of egg-water. There is some
indication — though the evidence is inconclusive — of specificity
in this reaction, as can be seen from the following table, which
summarises the effect of egg-water on sperm of different
genera of starfishes and sea-urchins : —

Egg-water.

?rm.

Asterias

Asterina

Arbacia

Strong>'lo-
centrotus

Asterias

. Very motile

No effect

Moderate
activity

Very slight
effect

Asterina

. No effect

Very motile

Very slight

effect
Activity

Arbacia

. Slight effect

No effect

Activity

Strongylo-

Slight

Activity

centrotus

effect

However, it would not be justifiable to conclude from this
line of argument that the sperm is directed to the egg by the
excretion of substances from the latter whose diffusion sets
up a gradient in favour of greater motility in propinquity to
the egg itself. The question has been further attacked by
two forms of procedure known respectively as the drop and
tube methods. The latter v/as introduced by Pfeffer, and con-
sists of filling capillary tubes with egg-water or other fluid
and observing the reaction of the sperm when the tubes are
placed in a sperm-suspension. The data so obtained are
difficult to interpret. It is true, for example, that capillary
tubes containing sea water which has been in contact with
ripe eggs of Echinus esculentus soon become plugged with
sperm when introduced into a sperm-suspension of the same
species. And Dakin and Fordham (1924) have endeavoured
to establish the chemotactic orientation of the sperm towards
the egg by comparing the accumulation of sperm in the egg-
water tubes with accumulation of sperm in tubes containing
other immobilising agents {e.g. acid) which would act as a
trap collecting the quiescent sperm. With this particular
species the control tubes were always found to contain less
sperm than the egg-water tubes, and it is pointed out by the
authors mentioned that the sperm travelled a greater length
in the egg- water tubes. However, using the same method
for the study of other material, both BuUer and Loeb obtained

THE FERTILISATION OF THE EGG 179

results which do not support the conclusion that in general
the eggs of marine animals give off substances which actually
direct the movement of the sperm. Dakin and Fordham
themselves were unable to demonstrate chemotaxis in the sperm
of the mollusc Teredo. The drop method employed by
Lillie and his pupils yields results which are interesting but
somewhat difficult to bring into relation with those obtained
by Dakin and Fordham, though Lillie himself advocates
the existence of chemotaxis. When a drop of the egg- water
of Arbacia is introduced under a cover slip into a sperm-
suspension of the same species three effects are manifest
on microscopic examination. There is momentarily an
intensification of the normal motility of the sperm. This is
followed by an effect which has the appearance of a precipita-
tion and takes place in two stages : {a) aggregation, the loose
association of spermatozoa in groups which can be imitated
by passing CO2 into a sperm- suspension, and also occurs
spontaneously in dense sperm-suspensions presumably through
the accumulation of their own respiratory products ;
{h) agglutination, in which masses of sperm firmly adhere
together. This latter phenomenon is reversible, when the
egg-water is prepared from the same species as that from
which the open suspension is derived. That is to say, after
a few seconds or minutes, the sperm-masses separate, but
individually the sperms remain immobile. Similar phenomena
have been described in Nereis, Asterias, and Echinarachnius.
It is possible to study the agglutinating reaction
quantitatively by determining the greatest dilution at which
an indisputable reaction occurs for given samples of egg- water.
The properties of the agglutinating substance have been worked
out by LilUe and others, and it has been shown that the sub-
stance is not excreted by the fertilised but only by the
unfertilised egg. It appears to be of colloidal nature. The
egg- waters prepared from Arbacia and from Nereis both con-
tain substances capable of agglutinating the sperm of the same
species. The egg- water of Nereis produces no effect on the
sperm of Arbacia. On the other hand, sperm- suspensions of
Nereis undergo agglutination in presence of egg-water of

i8o COMPARATIVE PHYSIOLOGY

Arbacia, and this reaction is an irreversible and toxic effect,
unlike the reversible reaction of the sperm to egg-secretion of
the same species. According to Lillie's experiments the
** iso " — and *' hetero " — agglutinating reagents are different
substances. He infers this from two lines of evidence : (i) that
egg-water of Arbacia, which originally acted on the sperm of
both genera, on keeping lost its action upon the eggs of Nereis
while retaining its activity with reference to sperm of the same
species ; and (2) after removal of all the agglutinating substance
which affects Nereis sperm by addition of the Arbacia egg- water
to a sperm-suspension of Nereis, the agglutinating action of
the egg-secretion on Arbacia or Arbacia sperm was unimpaired.
It was also found that the sperm of a Teleost would neutraUse
the hetero-active substance. While these phenomena provide
new materials for the serologist, it is perhaps premature to
emphasise very strongly the conclusion stated by Lillie that
" egg substances that thus activate and direct specific
spermatozoa and render them adhesive are well suited to
favour the fertilisation reaction."

Such information as is available with reference to the
specificity of the fertilisation act does not lead to very definite
conclusions. The sperm of one species will not in general
fertilise the eggs of another species. But this specificity
is not by any means absolute, and as illustrated by the rather
extreme example of Kupelwieser's experiment, it can be
overcome to some extent by experimental manipulation.
When this can be done it is possible to search for some factor
which specially distinguishes the normal process from the
experimental procedure. The problem still remains to be
solved. Baltzer, Tennent, Shearer, de Morgan and Fox,
Fischel, and others have successfully hybridised different
species and genera of Echinoderms ; similar experiments have
been made on Teleosts by Newman and Moenkhaus, and on
Amphibia by Bataillon.

Careful investigations into this phenomenon by Fox (19 16)
on Ciona failed to throw very much light on the question.
Ciona exhibits an interesting form of specificity, one that may
be common among hermaphrodite organisms and does not

THE FERTILISATION OF THE EGG i8i

reinforce the teleological view that hemaphroditism, so widely
spread among parasitic and sedentary organisms, is an adapta-
tion to overcome the impediments to sexual intercourse in
these forms. In the Tunicate the eggs are much more readily
fertilised by sperm of another individual than by sperm
derived from the same individual. When a certain number
of eggs of an individual A of Ciona in a given volume of sea-
water are fertilised by the addition of a certain quantity B of
a sperm- suspension of another individual, the number of eggs
which segment is smaller than when an approximately
equivalent suspension A is fertilised by an equivalent amount
of B in the presence of an extract made from the ovary on the
one hand, or from grinding up the eggs either of the individual
from which the eggs were obtained, the individual from which
the sperm was obtained, or a third individual. In the same way
the eggs of Arbacia and Strongylocentrotus contain substances
which increase the fertilising power of the sperm of the same
species. One may say in conclusion that there are a large
number of data available which suggest that eggs secrete sub-
stances which influence the sperm ; that there are indications
that these substances are of the same general character as
*' antibodies " ; and that possibly the action of some such
substances may facilitate fertilisation by a sperm of the same
species, while other substances tend to prevent union with
sperm of another species. But the last proposition remains to
be proved.

Further Reading

LiLLiE. Problems of Fertilisation. Chicago University Press.

LoEB. Artificial Parthenogenesis and Fertilisation. Ide7n.

For later work consult :

Carter (1924). On the Early Development of the Echinoderm Egg.

Proc. Camb. Phil. Soc. (Biol.) i.
Shearer (1922). On the Oxidation Processes of the Echinoderm Egg

during Fertilisation. Proc. Roy. Soc. B. 93.

CHAPTER XI

INHERITANCE

Before proceeding to a consideration of the subsequent
history of the fertiUsed egg it is necessary to take into account
an aspect of fertilisation which is significant to the final result
of the process. The entry of the sperm into the egg not
only provides the stimulus for further development, but
influences the development so that the new individual bears
resemblance to the male as well as to the female parent. The
study of reflex phenomena has provided an instance of a field
of physiological inquiry which has been made susceptible to
quantitative treatment by the work of Sherrington, Pavlov,
and others, though the physicochemical basis of the phenomena
themselves is but little understood. The ultimate mechanics
of hereditary transmission is perhaps even more obscure ;
but there is hardly any branch of biological research which
has attained a higher degree of precision in the quantitative
treatment of those relations with which it is concerned. The
excellent presentation of the existing state of knowledge in
such recent works as that of Crew is sufficient excuse for
omitting a large mass of experimental detail. To such
the reader may turn for confirmation of statements which,
owing to lack of space for detailed treatment, may appear to
be dogmatic. One can, however, hardly omit all reference
in an account of this nature to properties common to all
organisms, and properties concerning which, moreover, the
bulk of our knowledge is derived from the study of the lower
organisms.

The Factorial Kypothesis.— The exact study of inheritance
begins in the opening years of the present century with the
rediscovery by Tschermak, Correns and de Vries of certain
principles originally formulated by a contemporary of Darwin,

INHERITANCE 183

the Abbot Mendel. In their original form Mendel's laws
were based on the study of plant types, but they were at an
early stage extended to animals by Bateson (1902), whose
brilhant critique of the speculations of nineteenth- century
naturahsts in his " Materials for the Study of Variation "
(1895) had done so much to prepare the way for the
development of genetic physiology.

In Mendel's original experiments inheritance was studied
in the common pea, which possesses a number of true-breeding
strains distinguished by well-defined characteristics such as
colour (yellow or green) of the seed coat, or size of shoot
(tall or dwarf). Mendel's method, which is the basis of all
truly quantitative treatment of inheritance on experimental
lines, differed from that of his predecessors in three particulars :
he confined his attention exclusively to the transmission of
single well-defined characteristics ; he recorded separately
the progeny of the individuals employed ; and he used in
his crosses only individuals from stocks proved to breed true
for such characteristics. Experiments not safeguarded by
this precaution have no value for purposes of scientific reason-
ing * in relation to our present problem, namely, the extent to
w^hich the sperm and egg respectively contribute in maintain-
ing the continuity of resemblance between parent and offspring.

As an introduction to the problem let us consider the
results of mating an individual from a pure v/ild stock of the
fruit-fly Drosophila, in which the wings extend beyond the
tip of the abdomen, with an individual from a pure stock of
the well-established mutant (sport) in vvhich the v/ings are
vestigial. We shall attempt to build up the argument at each
stage in the experiment. The offspring of the first generation,
commonly referred to as the F.i (first fiHal), are all of the normal
(long- winged) type. Bodily they are indistinguishable from
the long-v/inged parent of the cross ; but when mated among
themselves they behave in a different manner, in that a definite
proportion of their progeny have vestigial wings. The long-

* The pure line experiments of Johannsen are not so much to be re-
garded as having constructive significance, but rather as a means of clearing
up confused habits of thought resulting from the neglect of this precaution
in earlier work. — Author.

1 84 COMPARATIVE PHYSIOLOGY

winged (referred to henceforth as " longs " for the sake of
brevity) flies of the F.i thus differ from the parental longs
in producing gametes some of which are characterised by the
possession of a material something — let us call it 2igene, with-
out discussing its nature — which leads to the production of the
vestigial condition of the wings. If we denote the gene which
determines the long-winged condition by the symbol V and
the gene which determines the vestigial condition by the
symbol v, we may refer to the F.i longs as Yv, to denote that
they form gametes bearing both V and v. By analogy the
parental long and vestigial types, which since they breed true
may be regarded as forming one type of gamete only, may be
denoted by the symbols VV and vv. Let us proceed to examine
the progeny (F.2) of the F.i longs mated inter se. One quarter
are vestigial indistinguishable from the original vestigial,
breeding true to type w^hen mated with their like or with
the original vestigial type. They may therefore be denoted
by the symbol vv as before. The remaining three-quarters
are longs. They do not all behave in the same way on crossing.
If they are individually crossed back to the original vestigial
stock, one-third of the F.2 longs produce, like the original
longs, only long- winged offspring ; and individuals which
behave in the back-cross in this way, when mated inter se
breed true to type ; they may therefore be denoted by the
symbol VV. The remainder when back-crossed to vestigial
give offspring half of which are longs and half vestigials. These
impure F.2 longs when mated inter se behave like the F.i longs,
giving a 3 : I ratio of long to vestigial. Thus they may again
be denoted by the symbol V^^. The constitution of the F.2
is therefore i VV : 2 Vz; : i z;?^. Now if we make a very simple
assumption about the distribution of the genes V and v in
the formation of the gametes the quantitative relations of all
these crosses fall into line. Let us suppose that on the average
the gametes produced by an individual consist of equal numbers
containing the gene derived from one or the other parent. The
F.I long receives from its vestigial parent the gene v, and from
its long parent the gene V : one-half of the gametes it produces
carry V and the other half v. V may fertilise V or v. Similarly

INHERITANCE

185

V may fertilise V or v. Since the probability of two events
happening together is the product of their separate probabilities,
the resulting probabilities of all possible combinations are
(1)2 VV : (i)2 \v : (1)2 v\ : (i)2 vv. This gives the pro-
portions I : 2 : I for pure longs, impure longs, and vestigials,
or a 3 : I ratio of longs and vestigials.

In general characters distinguishing different hereditary

Vestigial,

Vv
Vv

Long-winged.

n
V

vv P,
Vv Fi

vv F\

1
Vestigial.

Fig. 41. — Genetic segregation.

Strains are distributed in hereditary transmission according
to the assumption that they depend upon genes derived from
both parents which segregate in the formation of the gametes,
so that a gamete either contains the paternal or the
maternal gene. The individual bearing dissimilar paternal
and maternal genes (heterozygous condition) is not always
predominantly like one or the other parent ; it may be quite
intermediate, or unlike either. When the character of one

i86 COMPARATIVE PHYSIOLOGY

parent predominates in the heterozygous condition, it is called
the dominant character (long in this case) in contradistinction
to the recessive (vestigial in this example).

An immense variety of characters both in plants and animals
have been found to follow^ the rule of segregation. To mention
but a few, colour of the hair in mam^mals, duration of life in
Drosophila, fecundity and absence of feathers on the neck in
fowls, brachydactyly in man, absence of eyes and wings in
flies. These suffice to show what diverse types of hereditable
characteristics, anatomical and physiological, depend on
segregating hereditary factors or genes.

However, factorial analysis, as this method of investigation
is sometimes called, is not often as simple as in the case cited.
And those who have criticised the universal applicability of
the gene hypothesis usually do so in the expectation of a text-
book simplicity in eveiy instance. When we cross two strains,
it may, and often does, happen that the difference which
distinguishes them depends on more than one gene. The
applicabihty of the factorial hypothesis can here be substantiated
by the possibility of recovering types identical with both
parents in the F.2 generation. Of course, as the number of
genes involved increases, the number of possible combinations
in the F.2 increases, and the likelihood of reclaiming the
parental types diminishes.

There is another criterion of segregation which has been
successfully applied to the analysis of a phenomenon which
has been held up as a stumblingblock to the general validity
of the gene hypothesis, namely the inheritance of size. In a
good many cases clear-cut size differences depending on single
genes have been found out. Very often, however, the F.2
form a continuous unimodal series. If segregation took place
in a cross involving a large number of factors, it follows from
quite elementary statistical principles that the coefficient of
variation in the F.i should not be greater than that of either
parent ; but that the coefficient of variation of the F.2 should
be greater than that of the F.i ; and that the coefficient of
variation of every subsequent generation would be on the whole
less and never greater than that of the F.2. Furthermore,

INHERITANCE 187

the range of variability in F.2 should extend to or beyond the
limits of the two parental ranges. This has been shown to
be true in cases w^orked out by East and Jones.

We have next to inquire how the transmission of one gene
reacts upon that of another ; and what results occur when two
or more pairs of genes are involved in a cross.

Independent Assortment.— In the wild form of the banana fly,
Drosophila, which has been the material of a considerable volume
of research by Morgan and his school, the body is gray and
the wings extend beyond the tip of the abdomen. Two true-
breeding mutants have appeared in Morgan's cultures respec-
tively distinguished by the shade of body colour known as ehony
and by a vestigial condition of the wings. Both are recessives
to the wild condition. On crossing an ebony fly with long-
wings with a gray fly with vestigial wings, all the F.i are of the
gray-long type ; and the F.2 the four combinations : gray-
long, ebony-long, gray- vestigial, and ebony-vestigial in the
proportions 9:3:3:1. On the assumption that the pair
of genes responsible for the ebony and gray characters, on the
one hand, and the long and vestigial characters on the other
are transmitted quite independently, there is a 3 : i chance
of any individual having either dominant character in the F.2.
The probability of an individual having both dominant
characters is (|)2, that of it having one dominant but not the
other and vice versa is | x J, and that of having neither dominant
character (J)^. This gives the 9:3:3:1 ratio and proves that
the assumption is correct. This is further borne out by the fact
that identical results follow the mating of an individual of ebony
colour and vestigial wings (double recessive) with the wild type.

Linkage. — This independent assortment of separate pairs
of genes is very com^mon in all organisms investigated. If,
how^ever, separate pairs of genes always segregated in this
way, we should be compelled to postulate an indefinite num.ber
of structural units to provide for the material basis of inheritance.
As a matter of fact, independent assortment is not a universal
rule. Association of genes belonging to different allelo-
morpliic {i.e. segregating) pairs in the process of transmission
in contradistinction to the independent assortment illustrated

1 88 COMPARATIVE PHYSIOLOGY

by the experiment just described is known as linkage. Linkage
may be partial or complete.

Both types of linkage are illustrated by the cross between
the recessive mutants of the fruit fly known respectively as
black (already mentioned) and vestigial. When a black fly
with long wings is crossed with a gray fly with vestigial wings
all the off'spring as in the foregoing experiment are gray with
long wings ; and the same is true if a black fly with vestigial
wings is crossed with a fly that is homozygous for the gray body
colour and long- winged condition. But whereas, when the
F.I male from the cross between gray- vestigial and black-
long is mated to the double recessive (black- vestigial) female,
one-half of the off'spring are gray-vestigial and the other half
black-long ; when the F.i male of the cross between black-
vestigial and gray-long are mated to the double recessive female
one-half of the progeny are black- vestigial and the other half
gray-long. The genes re-emerge in the same combinations
as those in which they were present in the original parent.
Here linkage is complete.

Partial Hnkage is seen when the F.i females are crossed to
the double recessive males. The off'spring of the mating
between the double recessive male and F.i female from the
cross between black-vestigial and gray-long are not fifty per
cent, black- vestigial and fifty per cent, gray-long but 41*5 per
cent, black- vestigial, 41-5 per cent, gray-long, 8*5 per cent,
black-long and 8*5 per cent, gray- vestigial. Similarly the
offspring of the mating between the double recessive male and
the F.I females from the cross between black-long and gray-
vestigial are 41*5 per cent, black long, 41*5 per cent, gray-
vestigial, 8'5 per cent, black-vestigial and 8*5 per cent, gray-
long. In seventeen per cent, of the offspring there has been
'' crossing over " ; the genes for black-gray and long- vestigial
have become detached, though not to such an extent as to
segregate with complete independence.

Linkage has been studied in several hundred mutants of
Drosophila, and two important general results emerge from
these researches of Morgan's school : (i) if a gene a is linked
with a gene b which is also linked with a gene c, then a and c

INHERITANCE 189

are also linked, and the percentage of crossing-over between
a and c is in linear relation to the percentage crossing-
over between a and h on the one hand, and h and c on the
other ; similar phenomena appear to hold in the sweet pea,
where linkage was first discovered by Bateson and Punnett
(1906) ; (2) if the gene a segregates independently of d, then
b and c which are linked with a also segregate independently
of d. Thus in the fruit fly all the mutant genes can be classified
in four groups such that members of a given group show linkage
inter se and independent segregation with respect to members
of other groups. The genes of Drosophila thus appear to
be associated in four pairs of material units.

Sex-linked Inheritance.— One group of linked characters
in Drosophila is of special importance to a consideration of
the general applicability of the gene hypothesis, and is equally
important because of the Hght which it sheds on the problem
of sex- determination. A single instance will suffice to make
clear the characteristic feature of this group. In the wild
fruit-fly the eye is red ; there is a mutant form with white
eyes. A red-eyed female crossed with a white-eyed male
yields an F.i composed exclusively of red-eyed individuals ;
but in the F.2, which consists of three reds to one white, all the
females are red- eyed, and all the whites are male. Now when
a pure red-eyed male is crossed with a white-eyed female
the result is quite different ; all the females in the F.i as
before have the dominant red eye ; but the males are white-
eyed. When the F.i are mated inter se^ equal numbers of
white-eyed and red-eyed females and males are produced.
The inability of the male to transmit red to his offspring of
the same sex is readily explained on the asumption that the
red gene is linked to something which, if present in the zygote
in duplicate, leads to the production of a female, and if present
in the zygote unpaired (diagram) leads to the production of
a male ; the red-eyed male produces sperm of two kinds, one
bearing the " red " gene destined to fertilise an egg which
must become a female, and one which cannot bear the red
gene and which is destined to lead to the production of another
male (Fig, 42). This implies that sex itself is predetermined

1 90

COMPARATIVE PHYSIOLOGY

by genetical factors for which one sex is heterozygous, so that
a I : I sex ratio is maintained by the normal consequences of a
homozygous-heterozygous mating. Since in this case maleness
is the state associated with the single condition and femaleness
with the duplex state as regards the sex-linked genes, the
male may be represented symbolically as F/ and the female as
FF, using the symbol F for that which determines femaleness.
This type of sex-linked inheritance occurs in most insects
and in mammals ; and for reasons
given later may be anticipated to
occur in practically all higher bi-
sexual animals except birds and
lepidoptera (moths and butterflies).
The phenomenon of sex-linked in-
heritance was first discovered in the
latter group by Doncaster (1906).
A variety of the currant moth
Abraxas grossulariata is distin-
guished by the pale colour of the
wings as lacticolor. If a lacticolor
female (wings of a pale cream tint)
is crossed with the normal dark-
winged (grossulariata) male, all the
offspring of both sexes are of the
grossulariata type : in the F.2 there
is a 3:1 ratio of grossulariata to
lacticolor, but all the males are of
the grossulariata type. In the reciprocal mating the grossu-
lariata female can only transmit the grossulariata pattern to
her sons ; all the female offspring are of the lacticolor type.
When the F.i are mated inter se^ equal numbers of lacticolor
and grossulariata males and females are produced (diagram).
Here the female moth is constitutionally simplex with
respect to the sex-Hnked genes. This is the exact reverse of
the state of affairs in sex-linked characters in Drosophila.
Femaleness is associated with the simxplex condition of genes
which in the duplex condition give rise to maleness.

The predetermination of sex by genetic factors does not

Fig. 42. — Sex-linked inherit-
ance in Drosophila.

INHERITANCE 191

mean that sex is irrevocably fixed at fertilisation ; like all
other genes those which ordinarily determine sex require
appropriate external and internal conditions in which to
operate; and we shall return to this question in considering
sex- differentiation as part of the physiology of development.

It is interesting to note how nicely balanced the genetic
factors influencing sex- differentiation may be. This is well
seen in experiments of Goldschmidt on the gypsy moth,
Lymantria. Individuals from the same local races of this
widely distributed form when bred among themselves produce
a normal sex ratio ; when individuals of different local races are
crossed the relations of the sexes among the offspring may be
abnormal. If females from a European race are crossed with
males from a Japanese race, the offspring are normal males
together with females showing a num.ber of modifications in
the direction of maleness ; the would-be females are intersexual .
Reciprocally, when a Japanese female is mated to a European
male the F.i generation is normal but a definite proportion of
the males in the F.2 are intersexual, i.e. show modifications in
the direction of femaleness. Varying grades of intersexuality
characterise the results of crossing individuals from different
local races ; the grade from the cross between any two given
races is always the same. In the extreme case all the individuals
of such a cross may be of one sex, but half of these on being
bred back to a parent stock can be shown to have the genetic
constitution of the alternate sex. By making two assumptions
Goldschmidt has brought into line the results of a very large
number of such racial crosses : (i) in addition to the genes
for maleness for which the female has the constitution Mw
and the male MM (as in Abraxas where the same type of
sex-linked inheritance occurs) there is something dependent
on the constitution of the tggy transmitted therefore through
the female parent only, that mxodifies the degree of maleness,
and is denoted by the symbol F. Thus a female MmF forms
gametes MF and rriF while the male MMF forms gametes M
only ; (2) it|is^/urther assumed that in different local races the
efficiency of M and F respectively to influence differentiation
in the direction of maleness or femaleness are quantitatively

192

COMPARATIVE PHYSIOLOGY

different. By denoting M and F in terms of an arbitrary
system of numerical symbols Goldschmidt has elaborated
a system by which the appearance and degree of intersexuality
of his crosses between local races can be faithfully predicted.

Race A

T^ac€ B.

Female

F^ femaUs

^walc
CM>F)

Fig. 43. — Simplified representation of the results of Goldschmidt's
investigation of crossing in the gypsy moth Lymantria. The suffixes to
M and F denote the ** strengths " of the male and female factors. When
the sum of M is > F, the animal is male ; when < F, a female ; when =
F, intersexual.

Many thousands of individuals have been employed in these
crosses. The accompanying scheme is a simplified representa-
tion by the author of Goldschmidt's essential ideas (Fig. 43).

INHERITANCE 193

Structural Basis of Inheritance.— Just as it is possible to
identify in the reflex arc the structural basis of neuro-muscular
co-ordination in the higher animals, so it is possible for the
genetic physiologist to identify in the chromosomes the
structural basis of hereditary transmission in animals and
plants. It is probable that prevailing ignorance of the cellular
morphology of inheritance accounts in no small measure for
the neglect of Mendel's work by his contemporaries. By
the middle of the latter half of the nineteenth century the work
of Hertwig, Fleming, Strasburger, Boveri, Van Beneden, and
others had led to the recognition of the union of the nuclei
of the male and female gametes as the essential fact of sexual
reproduction ; of the constancy in number for every species
of the chromosomes or nuclear components in cell- division ;
and the maintenance of this constancy by the reduction of
the chromosomes to half the species-number in the production
of the gametes. By the beginning of the twentieth century
the studies of Strasburger and Sutton on the sizes and shapes
of chromosomes, the detailed study of the reduction division
and its antecedents by von Winiwarter, and the study of sexual
differences in the chromosome complex by McClung, whose
work was extended and elaborated by Stevens, Wilson, and
others, had accumulated sufficient evidence to locate in the
chromosomes the anatomical basis of MendeHan segregation,
and encourage the belief that the principles revealed by factorial
analysis were of widespread applicability.

Let us now consider separately the conclusions derived
from experimental study in the light of microscopic knowledge
available to-day.

I. Factorial analysis leads to the conception of material
units present in the fertilised egg in dupHcate, and segregating
before the formation of the gametes into maternal and paternal
components, one member of each pair and one only being
present in each gamete. As is well known, the chromosomes
in all animals and plants are present in the fertilised egg in
twice the number found to be present in the gametes . Further-
more, in many animals (and plants) from the most diverse
phyla, the chromosome complex of a species is characterised

194 COMPARATIVE PHYSIOLOGY

not only by a definite number but a definite configuration.
It is possible to distinguish among the chromosomes pairs
of different sizes and shapes (this is true of man, and many
mammals) ; and the maintenance of this constant configuration
implies that when reduction takes place one member of each
pair passes into each gamete. In other words, the chromosomes
are present in the fertilised egg in pairs, segregating in the
formation of the gametes into maternal and paternal com-
ponents, one member of each pair and one only being repre-
sented in each gamete.

2. The material units on v/hich hereditary transmission
depends are associated in groups, the members of which are
independently segregated with reference to members of other
groups. There are four such groups in Drosophila ; and in
Drosophila there are exactly four pairs of chromosomes. No
organism is known in which the number of linkage groups
is numerically greater than the number of pairs of chromosomes.

3. Lastly, we have seen that with respect to one group of
linked characters the sexes are differently constituted. Sex-
linked inheritance has been described in several groups of
the animal kingdom, including mammals ; there are several
well-established cases in man, v/here, as in Drosophila and the
cat, it is the male that produces two types of gam.etes. In
several hundreds of animal species from the most widely
divergent groups it is now established that one pair of chromo-
somes which is equally paired in one sex is represented in
the other sex by a single member, or a pair of unequal elem.ents.
For instance, in the cockroach the male has thirty-three
chromosomes (Morse) and the female has thirty-four chromo-
somes (Hogben). The female produces eggs which contain
seventeen chromosomes, while the male produces sperms
half of which possess sixteen and half seventeen chromosomes.
Clearly, if a sperm of the former type fertilises an egg the
resulting zygote v/ill have the chromosome number (33)
characteristic of the male, while if the latter type fertilise
an egg the resulting zj^gote will have the female constitution
(34). Drosophila, on the other hand, is a case where one pair
of chromosomes (XX) similar in the female is represented by

INHERITANCE 195

two unequal elements (XY) in the male. The same is true
of man and numerous genera of mammals. In the vast
majority of cases the male is the heterogametic sex. We have
seen, however, that in moths and birds the female is the hetero-
gametic sex. In birds the chromosomes are too small and
numerous to provide satisfactory material for investigation,
though recent work of Hance shows that the female is hetero-
gametic. In some moths (Seiler) there is a pair of dissimilar
elements in the female which are equally paired in the male.

The coincidence between the genetic and microscopic
data has been illustrated still further by the phenomenon of
'' non- disjunction " described by Bridges in connection with
several sex-linked nutant characters of which our original
instance of white eye-colour will serve as an example. There
appeared among the white-eyed m.utant stock of Drosophila
certain strains of which the females when crossed to normal
red-eyed males gave a certain proportion of red- eyed males
and w^hite-eyed females in addition to the usual red-eyed
females and white-eyed males alone. When the v/hite-eyed
female offspring of such abnormal crossings were mated back
to red-eyed males, they in their turn gave all four classes —
red-eyed males and females, white-eyed males and females.
The white-eyed females behaved like their mothers, giving
abnormal results in all cases. Half of the red-eyed females
gave normal and half abnormal results in crossing. Of the
male progeny the red-eyed individuals were normal, whereas
only half the white-eyed individuals were normal, the remainder
begetting daughters whose progeny was exceptional. Bridges
found that in the F.i abnormal white females the chromosome
complex of the dividing cells show^ed a Y element in addition
to the XX pair. This is explicable on the understanding
that at reduction of the egg in a certain proportion of cases
the X elements failed to disjoin, so that the ripe egg contained
either two X elements or none at all . If we represent the sperms
of a red male as X' or Y, two additional types of individuals
will result from fertilisation by a Y or X' sperm respectively :
an XXY or white female, and X'O or red male. This accounts
for the exceptional individuals in the F.i, and accords with

196 COMPARATIVE PHYSIOLOGY

the facts elicited. Next consider the resuhs of back-crossing
these XXY abnormal F.i white females to a normal X'Y male.
According to whether the X elements segregate with respect
to one another or the Y chromosome, the F.i white females
will lay four types of eggs : XX, Y, XY, X. If these are
fertilised by a Y sperm (which cannot bring in the red factor),
we get four types : {a) XXY white females which will obviously
behave in the same way, thus agreeing with breeding
experience ; (b) YY — individuals with such constitution
cannot exist ; (c) XYY — white males which should produce
XY sperms so that in crossing with normal white females
daughters of the XXY type producing exceptional progeny
would result ; (d) XY — normal white males. When, on the
other hand, the same four classes of eggs are fertihsed by an
X' sperm carrying the red factor, four red types of offspring
would result, as follows : (a) X'XX — a triploid female which
usually dies ; (b) X'Y — normal red males ; (c) X'YX red
females with abnormal offspring ; (d) X'X normal red females.
Thus the non- disjunction of the X chromosome in the forma-
ion of the eggs of some of the females of the parental white-
eyed stock accounts for the entire series of exceptional genetic
phenomena which occur in these strains.

Recently Bridges has shed further light on the genetical
aspect of sex- determination by the discovery of non- disjunction
in chromosomes other than the sex-chromosomes, sometimes
referred to in contrast to the latter as autosomes. In an
experiment in which a brown mutant of Drosophila was
crossed back to a parental stock, a culture was obtained in
which the individuals were almost exclusively females or sex
intermediates. These " intersexes " displayed intermediate
sex-characters throughout, notably in the abdomen and in
the sex-combs of the tarsal joint of the forelegs, and also
genitalia. On the whole they fell into two groups, one tending
more to the female, the other to the male condition. Genetical
evidence led Bridges to conclude that for one group of genes
at least the female individuals of these cultures were triploid,
i.e. inherited a double instead of a single set of genes from
their fathers. Microscopic examination of the germ cells

INHERITANCE 197

revealed the fact that the second and third chromosomes
were present in triplicate, while an additional fourth chromo-
some was present in some but lacking in others, there being
thus two degrees of the triploid condition, that with three
fourth chromosomes being more female (Morgan, Bridges and
Sturtevant, 1925). The X chromosome was present in dupli-
cate in the intersexes but the females possessed three X
elements. Thus using the symbol A for autosome and X
for the sex-chromosomes, the genetical constitution of these
intersexes and abnormal females were respectively 3A-I-2X
and 3A4-3X, as contrasted with the normal female con-
stitution 2A -h 2X. Abnormal males were also found with the
constitution 3A-hX, as contrasted with the normal male con-
stitution 2A + X. Therefore if X : A= i or> i the individual
is a female, if X : A | or<-| it is a male, but when X : A lies
between the i and | the intersexual condition is manifested.

Ratio of sex chromosomes
to autosomes.

Sex.

I or>i

ior<i

>ibut <i

Intersexual

General Validity of the Factorial Hypothesis.— The fac-
torial hypothesis has aroused a good deal of hostility, not
unnaturally, for it conflicts with many accepted speculations
as to the evolution of living organisms and throws doubt on
not a few beliefs still current in the medical profession. The
remarkable diversity of inherited characteristics, anatomical
and physiological, with which it deals ; the truly amazing
correspondence between the conclusions derived from experi-
mental and microscopic studies ; and finally, the established
fact that the nucleus is the only recognisable cell-element
which is universally contributed by the sperm to the develop-
ment of a new individual can leave little room for doubt in
the minds of impartial students of the subject that, in broad
general outline, it will be found to apply to all the essential
phenomena of biparental inheritance. If this is so its
importance for the study of animal function does not lie merely
in the account it gives of the contribution which the male and

198 COMPARATIVE PHYSIOLOGY

female respectively make to the constitution of a new animate
unit.

Bearing on Other Branches of Physiology.— For example,
many kinds of physiological experiment involve comparisons
for which controlled observations can only be carried out on
different individuals, e.g. effect of diet or endocrine substances
on growth. It is clearly established that growth-phenomena
and size-differences in a number of cases depend upon factorial
inheritance. Experiments of this kind, therefore, unless
based on large numbers of animals (which is often impossible)
and subjected to statistical analysis, are of doubtful value,
when the material is not known to be genetically homogeneous.

To achieve this end it is not necessary to set about breeding
pure strains with reference to every characteristic it is desired
to study. With the aid of the conventional symbols the reader
can easily satisfy himself that in a cross involving one pair of
factors, the proportion of heterozygotes diminishes generation
by generation in a continuously convergent series, if any
system of close inbreeding is employed. After about twenty
generations of brother and sister mating, for instance, or ten of
self-fertilization, the proportion of heterozygotes is indefinitely
small. If a stock that is not undergoing mutation is bred from
generation to generation by close inbreeding, it must eventually
become for all practical purposes homozygous for any charac-
teristic. Such stocks of white rats have been reared by the
Wistar Institute, and are employed in physiological experi-
mentation increasingly by American workers (see East and
Jones).

Again, the following citation, from an important physio-
logical memoir, illustrates how easily genetical bias enters into
physiological reasoning : —

" Now the history of the surfaces in the hearts of rays
on the one hand and of the dogfish and angel-fish on the other,
differs in this significant respect. They have been laved for
years, or one might properly say for generations before the
experiment, with solutions of different hydrogen ion concentra-
tion."

It is here implied that the action of a stimulus upon the

INHERITANCE 199

body of the parent can affect the reaction of the offspring to
the same stimulus. This beHef, at one time universally held,
still awaits confirmation by properly controlled experiment. If
we attempt to analyse in exact terms the belief that '* acquired
characters " are inherited, it appears to involve two separable
issues : (a) whether the inhibition or destruction of a characteE
is accompanied wholly or partially by the destruction of its
material antecedent in the germ cells ; (b) whether, if a stimulus
of a given magnitude is required to call forth a given response,
the application of that stimulus to the parent carries with it
the possibility of evoking the corresponding response in the
offspring with a stimulus of smaller intensity. Without
committing oneself to a dogmatic negative, it can be stated
as a matter of fact that in a good many cases the answer is
certainly in the negative (within the limitations of experiment
on these lines), and that in no single instance where a positive
answer has been given has independent and rigorous
reinvestigation confirmed the observations recorded. The
principle of economy of hypothesis is therefore best preserved
if the Lamar ckian principle is eliminated from consideration,
when the bearing of hereditary transmission on other branches
of experimental biology is under discussion.

Further Reading

GoLDSCHMiDT. Mechanism and Physiology of Sex Deteniiination.
Methuen.

Morgan. The Physical Basis of Heredity. Lippincott.

Crew. Introduction to the Science of Animal Breeding. Oliver and
Boyd.

East and Jones. Inbreeding and Outbreeding. Lippincott.

Morgan, Bridges and Sturtevant (1925). The Genetics of Drosophila.
Bibliographica Genetica (the Hague), vol. 2,

CHAPTER XII

THE PHYSIOLOGY OF DEVELOPMENT

Inheritance, which was discussed in the last chapter, is a
rhythmical repetition in each species of a definite and (on the
whole) similarly repeated series of events in which the pro-
duction of one individual leads up to the formation of gametes
whose union initiates a new being. The fertiHsed egg bears
within it the power to develop into an individual resembling
the parents from which the sperm and egg were derived.
Fertilisation starts in the egg a period of active cell-division.
In the initial stages of cleavage all the cells may be, and often
are, for a considerable period very much alike. As they go
on dividing they differentiate individually and regionally to
build up the structural architecture of the new individual.
In the early stages there is no increase in size ; at some point,
however, the developing organism begins to augment in weight
and volume. This process usually goes on long after the final
morphological order characteristic of the individual is
completely established. Developmental phenomena may thus
be considered under two headings : diiferentiation, individua-
tion and growth.

Individuation, or the differentiation of structural pattern
in cellular animals, raises perhaps the most recondite issues
in the whole field of biological inquiry. It is convenient to
consider it separately in its spatial and chronological aspects,
that is to say (i) the agencies which determine whether a
particular region is to differentiate into one type of structure
rather than another ; and (2) the agencies which determine
the orderly sequence in which the differentiation of one
structure follows another. In this chapter no attempt will

THE PHYSIOLOGY OF DEVELOPMENT 201

be made to deal with that large body of inquiry in the field
of " experimental embryology " which is not as yet susceptible
to quantitative treatment * ; we shall merely attempt to indicate
directions in which what are ordinarily called physiological
methods have been brought to bear upon developmental
phenomena. As the subject is a difficult one a certain amount
of latitude in defining concepts which may assist to clarify
the issues may be permitted.

The normal end-product of development — the individual
as we know it — is only one of a large number of ways in which
the hereditary constitution can be reahsed spatially. By the
methods of regeneration and implantation of organs, and by
varying the physicochemical constituents of the external
medium other structural patterns can be induced. Modifica-
tion by physicochemical agencies alone falls within the scope
of this treatment. As in deaHng with the problems of fertilisa-
tion, if the experimentalist can modify the course of events by
physicochemical means, some progress will have been made
towards an understanding of the mechanical basis of the
natural process. A measure of success has already attended
the efforts of experimentalists in modifying the course of
individuation by physicochemical agencies. One may recall
the well-known experiments of Stockard (1906), who found
that by placing eggs of the Atlantic minnow in a mixture of
sea water and magnesium chloride (19/60 M) about half
the individuals developed into one-eyed forms. These
Cyclopean monsters were of two varieties ; in one kind the
two optic rudiments approximated at an early stage in the mid-
dorsal line and coalesced ; in the other only one eye developed,
shifting into a median dorsal situation. Many of these
embryos hatched out, and were able to swim like the normal
individual. Again, there is the well-known method, due to
Herlitzka, of producing Siamese twins in newts by mechanical
means. If the two cells of the first cleavage in the newt's
egg are separated by a fine noose of hair in the plane of the
first furrow each half may segment as a whole, developing

* On this, readers should consult Jenkinson (1909). Diirken (i9i9)>
and Wilson (1925).

202 COJMPARATIVE PHYSIOLOGY

into a complete larva ; but if the constriction is incomplete
double-headed forms result. Double-headed monsters can
be produced from frog's eggs by inverting them in the two
cell stage, or (Bellamy) by the action of cyanides in appropriate
concentration at a later stage in development. Lastly may
be mentioned the production from sea urchins by Herbst
(1893) of plutei vi^ithout arms or spicules by exposing the egg
to the action of potassium salts.

Pioneer work of this kind provided a wealth of spectacular
instances of ways in which differentiation and individuation
can be partially controlled in a predictable manner by experi-
mental procedure. But in many cases the characteristic
abnormalities were on subsequent examination found to be
procurable by such a variety of methods as to defy analysis.
Thus Stockard's cyclopean embryos, thought at first to be
due to the specific action of the magnesium ion, can be obtained
with alcohol and other very different reagents. McLendon
(19 1 2) produced cyclopia in fish embryos with isotonic
solutions of lithium chloride, sodium hydrate, and a number
of other equally dissimilar substances.

During the past few years a new impulse has been given
to experiment on these lines by a hypothesis which has been
elaborated by Child. The evidence brought forward by Child
and his co-workers in favour of his hypothesis can hardly
as yet be said to be crucial. But its effect has been to introduce
new concepts v/hich, whether the main body of this work
stands or falls, are bound to simplify the nature of the problems
of individuation and prove the starting-point of new lines of
investigation. Of these concepts, it is not least im.portant
that Child, by emphasising the idea of polarity and describing
the architecture of the organism with reference to axial sym-
metries, has provided us with the very useful term, axiate
pattern. The arrangement of structural parts in the higher
organisms is so immensely complex that one must limit the
field in order to make the search thorough. One way of doing
this is to confine attention to the arrangement of parts with
reference to som^e axis of symmetry, e.g. the oral-aboral axis
of the body.

THE PHYSIOLOGY OF DEVELOPMENT 203

The first postulate of Child's hypothesis which need
here concern us may be stated thus : — the morphological
differentiation of parts (axiate pattern) with reference to a given
axis is preceded by the appearance of a gradient of physio-
logical activity (axial gradient) along this axis. This proposi-
tion may be shown to be true in a number of ways ; and is
quite independent of the particular interpretation of the axial
gradient which may be stated later. The existence of a ph3^sio-
logical gradient is here taken to imply that there exist between
the properties of the cells quantitative differences following a
definite orientation with reference to the future axiate pattern.
A very clear instance of this is provided by experiments on
asexual reproduction in Planaria dorotocephala. If a large
number of individuals of this species are cut up into strips,
it is found that the frequency with which corresponding strips
taken from different regions regenerate a head and develop into
complete new individuals varies in a perfectly definite way. If
we plot the statistical results of such an experiment with fre-
quency of head-formation as ordinates and regional position of
the strip along the abscissa (taking the head extremity as zero),
the ordinates gradually diminish as we pass along the x-axis
up to a certain point, then increase abruptly to a new maximum
and then diminish (Child, 19 15). Thus before there exists
any outward structural appearance of the formation of a new
head, there exists a physiological difference in the tissue at
the point where the new head is to be formed.

The second proposition brings us on to more debateable
ground. To do justice to the author, it may be stated, in his
own words : " Axial gradients have often been called metabolic
gradients, because differences in metabolism, or more
specifically of oxidative metabolism as indicated by various
experimental methods, appear to be characteristic and con-
spicuous features of them." In this sentence for the first
time an attempt is made to put the problem of individuation
on a basis which is accessible to quantitative methods of
physiological inquiry.

Various methods have been employed by Child and his
co-workers in the attempt to demonstrate regional differences

204 COMPARATIVE PHYSIOLOGY

in metabolic activity in small organisms and embryos. Two
of these, neither of which are wholly satisfactory, may be
mentioned. One is the use of potassium permanganate as
a colorimetric indicator of oxidative processes. Potassium
permanganate is readily reduced by protoplasm with the pro-
duction of a brown coloration, the intensity of which may be
taken as a measure of oxidation in a particular region, but
depends on other things besides. Child and Hyman (19 19)
studied the effect of placing small organisms and embryos
in very dilute solutions (M/ 10,000), and described in all cases
a gradient along the oral-aboral axis with maximum activity
at the anterior end.

The other method is the so-called susceptibility method.
The results obtained with this, though more striking still,
provide evidence of a somewhat indirect nature. In this
method Child has concentrated on defining the effects of
reagents like the cyanides which are known to reduce oxidative
activity. The method of interpretation is elaborate and
requires further investigation before it can be applied indis-
criminately, and one would feel more assured if Child had
confined his observations to the action of the cyanides alone.
Child and his co-workers have carried out experiments on
tissues at different temperatures and in different states of
activity which point to a quantitative relation between suscepti-
bility to the toxic action of cyanides on the one hand and to
metabolic rate, or, at least, to some form of physiological
activity, on the other. This relation is according to these
observations a complex one ; in lethal doses which are not
sufficiently concentrated to produce death within a short
period of exposure, the regions of higher activity are always
affected first, so that above a critical concentration suscepti-
bility varies directly as the physiological activity, w^hile below
this concentration the reverse relation is seen, in that regions
of higher activity recover and adjust themselves to the reagent
more successfully than regions of lower activity. In applying
the susceptibility method to embryonic development, lethal
concentrations may be used but not allowed to act long
enough to produce death in the embryo, and in such cases

THE PHYSIOLOGY OF DEVELOPMENT 205

they will, according to Child's interpretation, inhibit regions
of higher activity to a more marked degree than regions of
lower activity ; while, on the other hand, in very low con-
centrations of the reagent such as to permit acclimation
and recovery, the region of higher activity will be inhibited,
according to Child's interpretation, less than regions of lower
activity.

Two instances of the use of the susceptibility method
must suffice to indicate some positive results of the application
of Child's hypothesis. Eggs of Polychaetes were placed in
lethal concentrations of KCN. Initially dissolution begins at
the anterior end. As development proceeds the region of
maximum susceptibility shifts to the posterior region (where
growth is most active) so that when the larva is ready to undergo
metamorphosis the posterior extremity is the region which
succumbs most readily to lethal and recovers most easily
from sublethal concentrations. Child (19 17) finds that
embryos submitted to short exposure of lethal concentration
in the earliest developmental stages develop into individuals
with abnormally small heads. Embryos which are similarly
exposed at the later stage develop into individuals with
abnormally large heads.

In a similar study by Child of development in sea-urchins,
long exposure to sublethal concentration as well as the short
exposure to lethal concentration was investigated. Two
resultant types of plutei are figured, that produced by " dif-
ferential acclimation " (long exposure) with abnormally
enlarged oral lobe and widely divergent arms and that produced
by " differential inhibition " (short exposure) with diminished
oral lobe and angle of divergence between the arms. It is
impossible in the short space at our disposal to do justice to
Child's voluminous pubHcations, which must be consulted
for further information. The susceptibility method may well
prove a useful instrument of research, when its theoretical
assumptions are independently substantiated by accurate
gas analysis. Shearer (1924) in a recent publication records
results of an investigation based on direct measurement of
oxygen consumption with a technique that for the purpose is

206

COMPARATIVE PHYSIOLOGY

above criticism ; the new data are decidedly confirmatory of
the main contentions of Child's work *

Shearer first studied (a) portions of the living chick embryo
from the anterior and posterior ends of the blastodisc during
the first ten days of incubation ; (b) pigments from the anterior
and posterior extremities of the earthworm. In the first case
the oxygen consumption was measured by Barcroft's differential
m.anometer. In the second case Haldane's apparatus was

-Head.

en
rr>

E
E

Tail )t--

--^.^

^^t— = —

^--^

Days 4 5 6 7 8 9

Fig. 44. — Oxygen consumption of chick embryos (Shearer).

employed. In both sets of experiments the tissue used was
incinerated and estimated for protein content by Kjeldahl's
method. Thus all values of oxygen were expressed in absolute
units by reference to an equivalent amount of NH3 liberated
in the Kjeldahl determination at the end of the experiment.
The results of the experiments on chick embryos are repre-
sented graphically on Fig. 44. Here the point to notice is
that the gradient of the oxygen consumption curve for the head
and tail portions become identical at the seventh day, when

* [Note by the Editor, Sept., 1925.] Later, as yet unpublished, work
has somewhat modified these conclusions, Dr. Shearer informs me.

THE PHYSIOLOGY OF DEVELOPMENT 207

the axial pattern is established and further development is
mainly concerned with increase in size. The existence of an
antero-posterior metabolic gradient in the earthworm is seen
from the following data : —

Earthworm Experiments (Haldane's Methxod).

Temperature 12-5° C. 760 mm. Hg. All values reduced to 5 c.c.
NH3 Kjeldahl.

(i) In 3 hours worm consumed head 1*3 cub. mm. oxygen.

tail 0-3

(2) „ „ head 0-85
„ „ tail 0-27

(3) „ ,, head o*75
„ ,, tail o'25

In a second series of experiments Shearer investigated
the action of acetone powders instead of living tissues, the
former being prepared after the manner of acetone yeast
preparations by dehydrating the fresh tissue in acetone and
subsequently desiccating it. Such powders on being made
into a thin emulsion in distilled water take up oxygen. If the
powder prepared from the head region has a greater oxygen
capacity than powder prepared from the tail region, it is
impossible to escape the conclusion that the head region has
an intrinsic power to consume oxygen more rapidly,
independently of its structural organisation. This is indeed
the case. The results were again reduced to a fixed amount
of protein, i.e. in terms of 100 c.c. NH3 (Kjeldahl). The dura-
tion of each experiment was one hour twenty minutes at a
temperature of 40 C. and standard pressure. The oxygen
consumption in three experiments with acetone powders
from 6-7 embryos estimated by Barcroft's method is given
in cubic millimetres as follows : —

(i) Head o'62
Tail o"23

(2) Head 0*52
Tail 0*29

(3) Head 0*47
Tail o"27

The axial gradient hypothesis has naturally aroused con-
siderable hostility ; it is subversive of the underlying assump-
tions of the germ layer theor^^ For this reason the independent

2o8 COMPARATIVE PHYSIOLOGY

experiments of Shearer have been cited in some detail. It
opens up the possibility of substituting for architectural
mnemonics quantitative experiment along two lines : (i) how
external agencies acting on the egg or embryo set up spatially
orientated differences in cellular oxidative or other processes ;
in this connection important w^ork on electrical gradients in
organisms has been done by Lund (1921-22) and Hyman and
Bellamy (1922). (2) How the structural features of isolated
tissues are affected by artificially induced differences in oxygen
consumption. In relation to the last issue papers by Huxley on
de- differentiation maybe mentioned for suggestive indications.
Let us now turn from the spatial aspect of individuation to
the mechanism which determines the orderly succession of
developmental stages. The nature of the issue is clearly
presented by reference to experiments of Uhlenhuth (1912-17)
who has studied the effects of grafting eyes and skin of larval
salamanders into individuals of different ages. His observa-
tions show conclusively that when such organs are transplanted
they assume the adult characteristics not at the time when their
original possessor attains maturity, but always when the animals
into which they have been grafted attain metamorphosis.
Hence for the development of, say, the adult skin characteristics
(^e.g. yellow pigment areas) there must be present something
which is normally produced at the time of metamorphosis
and is produced by the body as a whole or by some special
organ or organs. The nature of this factor is now clearly
established as regards the case selected.

Animals which like Amphibia undergo a metamorphosis
are pecuUarly suitable for the study of the time factor in
development. Up to a certain point individuation proceeds
actively. It is then checked ; growth continues for a period
without much structural rearrangement. Then a second
phase of active structural differentiation is intercalated. In
the Anura this involves (i) closure of the gill clefts ; (2) resorp-
tion of the tail ; (3) full development of the limb rudiments.
InUrodeles the events are (i) resorption of the external gills ;

(2) resorption of the dorsal fin and shedding of the larval skin ;

(3) closure of the gill clefts. The Urodele larva has fully-

THE PHYSIOLOGY OF DEVELOPMENT 209

developed limbs and the tail persists into the adult stage.
The nature of the physiological change which initiates this
series of events is the same in either case.

The first experiment which threw any light on this was
the discovery of Babak (191 1) that the axolotl larva of the
Mexican salamander {Amhlystoma tigrinu7n ) which is nor-
mally neotenous, can be induced to undergo transformation
into the adult form by thyroid administration. Gudernatsch
(19 1 2- 14) showed that this was true of frog tadpoles. If
tadpoles are fed on diets of various tissues, ovary, liver, thymus,
brain, pancreas, spleen, pituitary, and thyroid — those fed on
thyroid gland develop limbs and lose their tails long before
the others. Thus in Rana catesbana (Swingle), a species
which normally requires three seasons to attain to the stage
at which metamorphosis occurs in nature, the six- weeks-old
tadpole will transform into a pigmy frog if fed on ox thyroid.
These observations received abundant confirmation both as
regards urodele larvae and anuran tadpoles (Morse, Barthelemez,
Jensen, Huxley and Hogben, Uhlenhuth).

Bennet Allen (19 16- 18) succeeded in overcoming the
manipulative difficulties of extirpating the thyroid gland in
tadpoles of the toad. The thyroidectomised tadpoles behave
in a perfectly normal manner until the limb-buds develop,
when transformation should occur. Instead of undergoing
metamorphosis at this stage they remain permanently in the
larval state, attaining as age advances dimensions far exceeding
those of a normal tadpole. They can, however, be induced, as
Swingle (19 18) showed, to develop into normal frogs if fed
on thyroid tissue. Later E. R. and M. M. Hoskins confirmed
the work of Allen by similar experiments on frogs and others
on urodele larvae.

Thus both inAnura andUrodeles it is certain (i) that the
removal of the thryoid normally prevents metamorphosis ;
(2) the administration of thyroid substance (or implantation
of thyroid tissue) accelerates normal metamorphosis and
initiates metamorphosis in thyroidless individuals. It has
long been known that the thyroid gland is essential to normal
growth in mammals, that it contains a high percentage of

210 COMPARATIVE PHYSIOLOGY

iodine, and that administration of tlie gland substance com-
pensates for the clinical disturbances resulting from its removal
or disorder — notably a reduction in basal metabolism.
Recently Kendall has isolated a substance having the properties
of thyroid extracts and the constitution of a tri-iodo derivative
of tryptophane. Helff (1923) and Huxley (1925) have shown
that thyroid extract increases respiratory exchange in the
tadpole ; but this does not appear to be the case in the adult ;
and Champy (19 19) claims that the thyroid hormone acts
selectively on larval tissues. The significance of the thyroid
has been further explored by Swingle (19 19) and by Uhlenhuth
(1921 and 1922). Swingle's observations concern the relation
of the organism to its iodine supply : the iodine content of
water and food is the limiting factor in thyroid develop-
ment and consequently in metamorphosis. This observation
throws a flood of light upon a phenomenon of no little bionomic
interest — neoteny in Anura. Neoteny in urodeles, however,
is a different matter. Two different grades are illustrated in
Amblystoma tigrinum by the European strain of the Mexican
species and the species from Colorado. The former never
undergoes metamorphosis in aquaria ; the latter is easily
induced to transform into the terrestrial salamander form by
external disturbances of one kind or another. Inorganic
iodine administered to urodele larvae does not induce metamor-
phosis as was shown by Swingle to be the case in frog tadpoles.
The experiments of Uhlenhuth and of Swingle (1922) show that
the axolotl thyroid, while containing the thyroid hormone,
requires some special stimulus to bring about its discharge
into the blood stream. Neoteny may be due then to lack of
sufficient iodine in the environment or inadequacy of the
mechanism which controls the discharge of the thyroid
hormone. It may finally be asked whether either of these
explanations extend to the true perennibranchiate genera
(Proteus, Necturus, Typhlomolge, and Siren). The first
three (Jensen, Huxley and Hogben, Uhlenhuth, Swingle) have
not been found to respond to the action of the thyroid hormone
at all. But since it is almost certain (i) that thyroid does not
increase basal metabolism in adult frogs ; (2) that the action

THE PHYSIOLOGY OF DEVELOPMENT an

of thyroxin on the frog tadpole is correlated with an increased
respiratory exchange ; and since all the perenibranchiates used
for these experiments were of fairly advanced age, it cannot be
stated with certainty that these forms are not representatives
of the persistent larval stage of species whose '' adult " form
has been eliminated through decreased sensitivity to the
action of the thyroid hormone.

These considerations lead us to go back a step and inquire
what it is that controls the development and activity of the
thyroid in Anura. Light has been shed on this by the work
of several investigators of the American school. In Anuran
tadpoles the hypophysial rudiment lies above the mouth in
a very accessible situation ; by pricking the surface of the head
in the embryo at a certain stage the ablation of the pituitary
anlage can be accomplished. This was first done by Adler
(1914). Smith (19 1 7) and Bennet Allen (19 17) simultaneously
and independently discovered that hypophysectomized tadpoles
in addition to showing the pigmentary disturbances already
mentioned, fail to undergo metamorphosis, this failure being
associated with arrested development of the thyroid gland.
Later it was shown by Bennet Allen (19 19) that such individuals
can be made to complete their development by thyroid
administration, and by Swingle (1922) that the same result
can be brought about by implantation of the pars anterior.
Metamorphosis of the Axolotl by injection of fresh extracts
of ox anterior lobe was recorded by Hogben (1922), and meta-
morphosis of hypophysectomised frog tadpoles by Smith
(1922) (see Spaul, 1925). Smith found that thyroidless tad-
poles will not respond to this treatment. On the whole the
evidence points to the following sequence : development of
the pituitary ; development of the thyroid under the influence
of a substance secreted by the pars anterior ; closure of gill-
clefts accompanied in Anura by development of limbs and
resorption of tail and in urodeles by shedding of the larval
skin and resorption of external gills under the influence of
the thyroid hormone, with the discharge of which in Urodeles
special — at present unknown — agencies are involved.

Another set of problems connected with the chronological

212 COMPARATIVE PHYSIOLOGY

aspect of individuation is presented by the appearance of
sexual differences. In some animals, e.g. mammals and birds,
there is a sex metamorphosis (puberty). The effect of
castration in the male of mammals is too well known to call
for elaborate comment, and the work of Steinach, Lipschiitz,
Sand, Moore, and others definitely establishes between sexual
metamorphosis and the glandular constituents (the " interstitial
cells ") of the gonads, a relation analogous to that seen in the
phenomena just described. But there is no complete proof
that the interstitial tissue exerts its influence on metabolism
by discharging a hormone into the blood. The same remark
applies to birds, where spaying of the female leads to assump-
tion of male characteristics of comb, plumage and spurs ; and to
Amphibia. All the established phenomena are equally well
explained by the hypothesis suggested by Geoffrey Smith :
that is, the gonads quantitatively affect the metabolism of
one or other blood constituents by their own activity in situ.
Some reference is due to a conception introduced by
Goldschmidt because of its suggestive bearing on the general
consideration of time relations in development, and because it
at once disposes of any difficulty we might find in harmonising
the established role of genetic factors in sex- determination
with the undoubted facts of sex-reversal in the animal kingdom.
The production of intersexes in the gypsy moth, Lymantria,
by crossing local races has already been mentioned (p. 191).
In moths the sexual and somatic metamorphoses are synchro-
nous. The sex differences are very marked in the copulatory
devices, colour, wing pattern, feathering of the antennae, shape
of abdomen, etc. And intersexuality in Lymantria is not
an intermediate condition of sex differentiation affecting all
parts alike. The intersexual individual is a sex mosaic.
Females with a low grade of intersexuality may display modifica-
tion in the antennae alone, these being of the feathered, i.e.
completely male type. A higher grade of intersexuality is
seen when the wing-colour as well as the antennae are character-
istically male, all other organs being of the female type. The
most advanced stage of recognisable intersexuality is that in
which the individual is externally a perfect male but internally

THE PHYSIOLOGY OF DEVELOPMENT 213

possesses ovaries instead of testes. ''If we now," states
Goldschmidt, ** try to formulate a rule which governs this
strange seriation ... we find the most important fact that
this series is the inverse order of differentiation of the organs
in development. The last organs to differentiate in the
pupa and the first to be intersexual are the branching of the
antennae and coloration of the wings. The first imaginal organ
differentiated is the sex gland, and if we apply this law even to
the parts of a single organ like the copulatory organ we find
it also holds good.

From these facts Goldschmidt elaborates a hypothesis
which may be stated in the following three propositions :
(i) that the relative potencies of one or other type of sex-
determining reaction-system are not the same throughout the
whole course of development ; (2) that genetic factors normally
ensure that one or other system predominates at the time
when sex- differentiation normally occurs ; (3) that if sex-
differentiation can be induced at an earlier or later stage, it
may be made to synchronise with the predominance of the
system alternative to that which controls differentiation in
the normal course of events. To put it in another way, if
we represent the potency of the male- and female- determining
reactions by ordinates and the time of development along the
abscissa, there is usually some stage in the life cycle of either
sex where the two curves intersect ; this point generally lies
either well before or well after the stage at which sex- differentia-
tion actually occurs. Such being the case we should anticipate
the possibility of sex-reversal by influencing the growth-rate
in forms where the two systems are, as in Lymantria, fairly
delicately balanced ; and Goldschmidt has produced female
intersexes in pure strains of Lymantria through rearing the
embryos at a very low temperature.

In concluding the foregoing sketch of developmental
physiology, one may frankly admit that we are only at the
beginning of a scientific treatment of the problem, and no
useful purpose is served by understating the difficulties
inherent in the subject and the distance which must still be
traversed before we can begin to envisage a purely physico-

214 COMPARATIVE PHYSIOLOGY

chemical treatment of individuation. However, when in
the Hght of such recent inquiries as have been touched on in
this chapter, one considers the fact that Httle more than a
decade has passed since entelechy was the centre of discussion
in developmental physiology, there is no justification for a
pessimistic attitude to the possibility of arriving at predictable
conclusions in this field of knowledge.

Further Reading

Child, C. M. (1915). Individuality in Organisms. Chicago.

(1924). Physiological Foundations of Behaviour. New York.

DCrken, B. (1919). Einfiihrung in die Experimentalzoologie. Berlin.

GoLDSCHMiDT, R. (1923). The Mechanism and Physiology of Sex Deter-
mination. London.

Jenkinson, J. W. (1909). Experimental Embryology. Oxford.

Wilson, E. B. (1925). The Cell in Development and Heredity (3rd ed.).
New York.

Recent Papers
Champy (1919). Arch. Morphol. Exp. et Gen., 1919.
Helff, O. M. (1924). The Oxygen Consumption of Thyroid and Diiodo-
tyrosine-fed Tadpoles. Proc. Soc. Exp. Biol. Med., 21, 34.

Huxley, J. S. and de Beer, G. R. (1923). Studies in Dedifferentiation,

IV. Quart. Journ. Micr. Sci., 67, 473.
Huxley, J. S. (1925). Studies in Amphibian Metamorphosis, II. Proc.

Roy. Soc, (B) 98, 113.

Hyman, L. H. and Bellamy, A. W. (1922). Studies on the Correlation
between Metabolic Gradients, Electrical Gradients and Galvanotoxis.
Biol. Bull. 43.

Lund, E. J. (1921-22). Experimental Control of Polarity by the Electric
Current, I and II. Journ. Exp. Zool., 34 and 36.

Shearer, C. (1924). On the Oxygen Consumption Rate of Parts of the
Chick Embryo, etc. Proc. Roy. Soc, (B) 96, 146.

Spaul, E. A. (1925). Experiments on the Localization of the Substances
in Pituitary Extracts, etc. Brit. Journ. Exp. Biol., 2, 427.

Stockard (1907). Artificial Production of a Single Median Eye, etc.

Arch. Entro. Mech., 23, 249.
Swingle, W. W. (1919). Studies on the Relation of lodin to the Thyroid,

I and II. Journ. Exp. Zool., 27, 397.

(1922). Experiments on the Metamorphosis of Neotenous Amphi-
bians. Journ. Exp. Zool., 36, 397.

Uhlenhuth, E. (1921). The Internal Secretions in Growth and Develop-
ment of Amphibia. Amer. Nat., 55, 193.

(1922). The Effect of Iodine and lodothyrin on the Larvae of

Salamanders, III. Biol. Bull., 42, 143.

INDEX

Abraxas, 190

Absorption spectra, 70, 73, 76, 78

Acids, secretion of, 62

Adler, 130

Adrenaline, 39, 122

Adrian, 14, 136, 143

Agglutination, 179

Aggregation, 179

Allen, 130, 209 et seq.

All or nothing law, 143

Alsberg and Clark, 75, 105

Amino acids, 92, 93

Amphibian metamorphoses, 208 et

seq.
Amphioxus, 90
Amoeboid movement, 30
Amylolytic enzymes, 96
Anaphylaxis, 59
Anodon, 15, 114
Anson and Mir sky, 79
Antidromic action, 103
Aphrodite, 99, 122
Aplysia, 81, 99, 113, 122, 157
Arbacia, 71
Arenicola, 73, 162
Ascidia, 114

Associative behaviour, 163
Astacus, 16, 77, 97
Atlantic minnow, 20, 38, 201
Atzvell, 132
Aurelia, 153
Autonomic ganglia, 157
Axial gradients, 203
Axolotl, 209 et seq.

Babak, 69

Bacot and Harden, 88

Baglioni, 52

Banting and Best, 98

Barcroft, 28, 49, 50, 73, loi, 102,

175
Barium, 39
Bateson, 183

Baumberger, 87

Bayliss, 33, 119

Bernstein, 13, 58, 63, 145

Bert, 64

Bethe, 32, 154

Biedermann and Moritz, 97

Bioluminescence, 52 et seq.

Blood pressure, 103, 107, 11 1

Bridges, 195

Briicke, 99, 106

Bodansky and Rose, 92, 96

Bodo, 66

Botazzi, 82, 104

Bounhiol, 66

Robert Boyle, 53

Buchner, 55

Buddenbrock and Rohr, 67 et seq.

Calcium ions, 16-22, 27, 35, 163

Calliphora, 93

Cancer, 77

Capillaries, 102

Carbon dioxide, tension of, 68, 80

Carbon monoxide, 74

Carcinus, 16, 154

Carlson, 107 et seq., iii

Carter, 174

Catch muscle, 15

Cellulose, 97

Cephalopods, eye of, 146

nervous system, 155
Cerianthus, 152
Ceriodaphnia, no
Chaetopterus, 52
Chambers, 35, 152
Chemotaxis of sperm, 178-9
Child, 201 et seq.
Chironomus, 65
Chlamydomonad, 89
Chlorhaemidse, 77
Chlorocruorin, 77 et seq.
Chromatophores of Crustacea, 39
Chromosomes, 193 et seq.

215

2l6

INDEX

Chronaxie, 137
Ciliary feeding, 89-91

motion, 23 et scq., 151
Ciona, 180
Clark, 19
Clowes, 121

Coagulation of crustacean blood, 105
Coelenterates, 52
Cockroach, 94
Colin, 176
Collip, 81

Conditioned reflexes, 164 et seq.
Conduction of nervous impulse, 140
Convoluta, 89
Crew, 182
Crepidula, 91
Crystalline style, 96
Ctenophores, 52
Ciishny, 50

Cutaneous respiration, 65
Cyanides, 20
Cyclopea, 201
Cypridina, 52 et seq.

Dakin, 89, 104, 178

Dale, 60

Darwin, 170

Day, 114

Demarcation current, 12

Denis, 51

Depressor nerve, 103

Dhere, 75-7

Diemyctilus, 132

Dioxinia, 15

Dissociation of oxyhaernocyanin, 76

of oxyhaemoglobin, 71-4
Dixipus, 67
Dolium, 62
Dominant, 186
Doticaster, 190
Drezv, 157

Drosophila, 7, 183 et seq.
Dubois, 56

Edmondson, 97

Edwards, Milne, 77

Eimer, 154

Elasmobranch, blood of, 104

Electric organs, 57-8

Electrocardiogram, 12, no

Electrolytes, 16 et seq., 35-6, 39

Elementary nervous system, 152

Elliott, 121

Endogenous metabolism, 93

Enzymes, 54 et seq.

Ergotoxine, 39

Esculin, 56

Lovatt Evans, 16

Excitation in nerve, 135 et scq.

Exogenous metabolism, 93

Explosive cells, 105

Fat digestion, 94
Fienga, 17

Final common path, 158
Fletcher, 7
Fox, ds^l^J, 180
Fredericq, 104, in
Frog's heart, 19
Frohlich, 155, 157
Fry, III
Fuchs, III
Fulton, 13, 98
Fundtilus, 20, 38

Gamble, 40 et seq,

Garrey, 160-1
j Gaskell, 123
i Gas, secretion of, 62
i Glomerulus, 48
i Glycogen, 95
i Goldschmidt , 191, 212

Grasshopper, 66, 69

Gray, 23 e/ 5-?^., 173

Giinn, 60

Gymnotus, 58

Haem, 79

Haematin, 71, 79

Haematoporphyrin, 71

Hasmochromogen, 71, 79

Heemocyanin, 75-7

Haemoglobin, 71 et seq., y^

Hardy, 105

Hartree, 10

Hartridge, 12

Harvey, 32, 54, 154. 1/4

Heat production in muscle, 9-10

Hecht, 115

Helicocrubin, 179

Helix, 17, 77, 97, 122

Helntholtz, 141

Henle, 123

Henry, i

Henze, 60

Herring, 128

Her twig and Fol, 170
I Heterozygote, 185
i Hibernation, 95

INDEX

217

Hilly 4 et seq., 139, 145
Croft Hill, SS
Hippolyte, 40 et scq.
Hirudin, 61
Hisy bundle of, 109
Hogben, 17, 122, 128
Holothurian, 66
Homarus, 14, 77
Hopkins, 7, 88
Hormones, 119

Hydrogen concentration, 8,14, 25-
33

Imbibition 28
Immunity, 59
Inhibitory nerves, 107, 113
Insulin, 98

Intersexuality, 191, 196, 212
Intestinal movements, 98
Intracellular digestion, 99
Islets of Langerhaus, 97
Isometric contraction, 6
Isotonic contraction, 2

Johnstone, 89
Joule, 86

Keeble, 40 et seq., 89

Koch, 114

Kreidl, 156

Krogh, 65, 66, 72, 103, 128

Kuhne, 55

Kupelweiser, 170

Lactic acid, 8-1 1

Lampyris, 52

Langley, 121

Lankester, Ray, 77

Lapicque, 16, 137, 148

Laplace, 85

Laurens, 129, 133

Lavoisier, 85

Law of Intestine, 98

Lee, 69

Leeches, 123

Lezvandozvsky, 121

Lieben, 130

Lillie, 20, 176 et seq.

Limulus hsemocyanin, 75

blood, 105

heart, 107
Linkage, 187

Loeb, 20, 105, 154, 158 et seq., 171
et seq.

I Loggerhead turtles, 161
j Lophius, 51

Lucas, Keith, 11, 136-9, 147-9

Luciferase, 56-7

Luciferm, 56-7

Lumbricus, 113

Lymantria, 191, 212

Lyon, 161, 174

Lysins, 59

Macromyses , 40 et seq.

Macleod, 98

Magnesium ions, 17-20, 27, 35, 163

Maia, 81, 107, 122

Malapterurus, 57-8

Maltase, 55

Mast, 160

Matula, 160

Mayer, 86, 154

Melanophores, 38, 123-32

Membranipora, 52

Memory, 163

Mendel's Law, 183 et seq.

Mering and Minkozvski, 97

Methaemoglobin, 71

Mexican salamander, 64, 209 et seq.

Microdissection, 35

Mines, 14, 16 et seq., 52

Mitra, 97

Monochromatic light, 45

Mustelus, 51

My a, 99

Myenteric plexus, 98

Mysids, 162

Mytilus, 23 et seq.

Negative variation, 2

Nelson, 96

Nereis, 174-6

Nernst, 138

Neurogenic heart beat of Limulus,

108-10
Neuroid transmission, 151
Neutral red, 33
Nitrogenous excretion, 51
Non-disjunction, 195
Nucleoproteins, 93

Octopus, 65, 75, 77, 81, III
Osmotic pressure, of blood, 104

and fertilisation, 172
Ostiate heart, 106
Oval, 62
Overton, 16

2l8

INDEX

Oxygen consumption of cilia, 28-9

of eggs, 17s

embryos, 206

muscle, 8-1 1
Oxy haemoglobin, 70

Palaemon, 42

Palinurus, 81, 107

Pancreas, 97, 119

Pantin, 32 et seq.

Parker, 124, 152, 161

Parnas, 15

Parsons, 81

Pasteur, 55

Pathenogenesis, i']i et seq,

Pazvlozv, 164 et seq.

Pecten, 15

Pedicellariae, 59

Pepsin, 92

Peters, 12, 71

Pfluger's Laio, 137

Phallusia, 81

Pholas, 53

Phototropism, 159 e^ ^^3-

Phrynosoma, 124

Phyllirhoe, 53

Physalia, 92

Physoclisti, 62

Pigmentary effect on system, 37 e/

seq.
Pineal, 133
Piper, 146

Pituitary gland, 103, 127, 211
Pollack, 63
Porthesia, 159, 163
Pouchet, 120
Protacanthus, 160

Purpura, 123
Putter's hypothesis, 89
Pyrosoma, 53
Pyrophorus, 53

Quagliariello, 75 > ^3

Raia, 17
Recessive, 186
Redfield, 124
Refractory period, 135
Renal secretion, 47 et seq.

tubules, 48
Rennet, 54
Rennie, 91
Respiratory movements, 67, 69

quotient, 30, 49, 95

Reversal of rhythm in tunicate

heart, 115
Rhumhler, 33
Ritchie, 8
Roaf, 123
Robber fly, 160
Robertson, Brailsford, no
Rogers, 132
Romanes, 153
Rouget, cells of, 102

Sabelliformia, 77
Sanford, 94
Schafer, 121
Schulz, 61
Scy Ilium, 17
Secretion, 119
Self-sterility, 180
Sepia, 53
Sergester, 53

Sex determination, 189-93
differentiation, 211
Shearer, 175, 206
Sherrington, 147, I49> 158
Simulium, 66
Smith, P. E., 120, 130
Snail, 17, 77, 97
Spftth, 38 ei seq.
Spallanzani, 53

Sperographis, 77

Sponges, 151

Stannius experiment, 112

Starling, 49

Statoliths, 156

Stedman, 76

Stick insect, 67

Stigmata, 67

Stockard, 201

Strauh, 113

Stretching, effect of, on muscle, 113

Strongylocentrotus, 172

Strong acids and bases, 25-6

Suprarenals, 127

Susceptibility method, 204

Syllid, 99

Synoptic nervous system, 154

Swingle, 131, 209

Takamine, 121
Teleosts, 97,

blood of, 104
Temperature, 28-9, no
Ten Gate, 99, 114, 122
Tentaculocysts, 153
Tetanus, 3

INDEX

219

Thyroid gland, 209 et seq.
Tonus, 15
Torpedo, 58

Tracheal respiration, 66 et seq,
Trivalentious, 18
Tropisms, 158
Tunicate heart, 114
Tyramine, 60, 122

Uexkull, 15
Uhlenhuth, 208 et seq.
Urea in blood and urine, 51

in selachian blood, 104
Uric acid, 93

in blood and urine, 5 1
Urine, 51

Vanadium, no

Van 't Hoff solution, 25

V. der Heyde, 5 1

Venoms, 60

Vitamins, 88

Vies, 74

Voit, 85

V, Frisch, 120

Warburg, 175

Weak bases and weak acids, 25-6
Weinland, 93
Winterstein, 65-6
Wohler, i
I V/oodland, 13

I Yeasts, 87-9
I Yonge, 92, 97

PRINTED IN GREAT BRITAIN BY WILLIAM CLOWES AND SONS, LTD., LONDON AND BECCLES.

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"

See other formats