THE LIBRARY

OF

THE UNIVERSITY
OF CALIFORNIA

PRESENTED BY

PROF. CHARLES A. KOFOID AND
MRS. PRUDENCE W. KOFOID

BIOLOGICAL LECTURES

AND ADDRESSES

BY THE SAME A UTHOR.

THE FROG.
AN INTRODUCTION TO ANATOMY, HISTOLOGY AND EMBRYOLOGY.

Fifth Edition. 1894. 45.
MANCHESTER : J. E. CORNISH.

A JUNIOR COURSE OF PRACTICAL ZOOLOGY.

In conjunction with C. HERBERT HURST, Ph.D.,

Demonstrator and Assistant Lecturer in

Zoology in Owens College.

Third Edition. 1892. ios. 6d.
VERTEBRATE EMBRYOLOGY.

A TEXT-BOOK FOR STUDENTS AND PRACTITIONERS.
First Edition. iSgj. 2is.

LONDON :
SMITH ELDER & CO., 15 WATERLOO PLACE.

IN PREPARA TION.

LECTURES ON THE DARWINIAN THEORY.

With Numerous Illustrations.
Edited by C. F. MARSHALL, M.D., B.Sc., F.R.C.S.

LONDON :
DAVID NUTT, 270-271, STRAND.

BIOLOGICAL LECTURES

AND ADDRESSES

DELIVERED BY THE LATE

ARTHUR MILNES MARSHALL

M.A., M.D., D.Sc., F.R.S.

PROFESSOR OF ZOOLOGY IN OWENS COLLEGE; LATE FELLOW OF
ST. JOHN'S COLLEGE, CAMBRIDGE

EDITED BY

C. F. MARSHALL

M.D., B.Sc., F.R.C.S.

LONDON

DAVID NUTT, 270-271, STRAND
1894

PREFACE

THE majority of the lectures and addresses col-
lected together in this volume have already been
printed in the Transactions of several Societies —
viz., the lecture on Animal Pedigrees, published
in the Midland Naturalist; the Presidential address
to the Biological Section of the British Association ;
and several reprinted from the Transactions of the
Manchester Microscopical Society. Of these printed
addresses I have reproduced as many as possible
without involving too much repetition. In the
case of the British Association address however it
will be found that many of the points discussed
are dealt with in other addresses, especially in the
lecture on Animal Pedigrees, which is indeed based
on that address. It appeared however desirable
to include the British Association address, even
at the risk of repetition, on account of the im-
portance of its scientific value, and for this reason
I have placed it at the end of the series, the others
being arranged chronologically.

With regard to the lectures in manuscript,

vi PREFACE

hitherto unpublished, I have selected a few of
those which appeared to be of most interest, and
in this I have of necessity been obliged to confine
myself to those which were most fully written
out. Where amplification was required I have
endeavoured, as far as possible, to do this in
words which, from my own personal knowledge, I
believe would have been used.

The lectures on the Darwinian Theory, which
form a distinct course by themselves, will be
published as soon as possible in a separate volume,
together with other series of lectures, if there
appears to be a sufficient demand for them.

I must express my thanks to the Committees
of the Manchester Microscopical Society, the
Birmingham Natural History Society, and the
British Association for permission to reproduce the
addresses printed in their Transactions.

I am under great obligations to Professor G. B.
Howes for his kindness in reading the proofs, and
for supervising the technical points. My thanks
are also due to Professor Ray Lankester for valuable
suggestions, to my brother Mr. P. E. Marshall for
correcting the proofs, and to Dr. C. H. Hurst for
assistance on several points.

C. F. MARSHALL.

LONDON, April 1894.

CONTENTS

I. THE MODERN STUDY OF ZOOLOGY.

PAGE

An address delivered at Owens College, introductory to
the Session 1879-80 I

II. THE INFLUENCE OF ENVIRONMENT ON
THE STRUCTURE AND HABITS OF ANIMALS.

A lecture delivered to the Owens College Debating Society,
January 1881 27

III. EMBRYOLOGY AS AN AID TO ANATOMY.

A lecture delivered to the Owens College Medical Students'
Debating Society, January 1881 41

IV. THE THEORY OF CHANGE OF FUNCTION.

A lecture delivered to the Owens College Biological Society,
1881 ....... . 53

V. BUTTERFLIES.
A popular lecture (date unknown) . . . . -63

VI. FRESH- WATER ANIMALS.

An address delivered at the Manchester Athenaeum on the
occasion of the Annual Soiree of the Manchester Micro-
scopical Society, January 1887. Reprinted from the
Society's Transactions 78

VII. INHERITANCE.

The President's address delivered at the Manchester Micro-
scopical Society, 1888. Reprinted from the Society's
Transactions . 87

viii CONTENTS

VIII. THE SHAPES AND SIZES OF ANIMALS.

PAGE

The President's address delivered at the Manchester Micro-
scopical Society, 1889. Reprinted from the Society's
Transactions • .116

IX. SOME RECENT DEVELOPMENTS OF THE

CELL THEORY

The President's address delivered at the Manchester Micro-
scopical Society, February 6th, 1890. Reprinted from
the Society's Transactions . .159

X. ANIMAL PEDIGREES.

An address delivered before the Birmingham Natural
History Society, October i4th, 1890; and based upon
the Presidential address to the Biological Section of the
British Association at Leeds in September 1890. Re-
printed from the Midland Naturalist . . . .. .192

XI. SOME RECENT EMBRYOLOGICAL
INVESTIGATIONS.

An address delivered on the occasion of the Annual Conver-
sazione of the Manchester Microscopical Society, January
2ist, 1893. Reprinted from the Society's Transactions . 249

XII. DEATH.

The President's address delivered at the Manchester Micro-
scopical Society, February 2nd, 1893. Reprinted from
the Society's Transactions 271

XIII. THE RECAPITULATION THEORY.

The President's address to the Biological Section of the
British Association delivered at Leeds, September 1890.
Reprinted from the Transactions of the Association . 289

I

THE MODERN STUDY OF
ZOOLOGY

THE man of business knows full well — at times too
well — the importance of periodical stock-taking ;
of comparing his actual position with his estimated
one, of ascertaining exactly how he stands, of
assuring himself that his affairs are in a sound and
healthy condition, and that the gain on the year's
transactions is a real one. The man who neglects
such precautions is apt, sooner or later, to find
himself in difficulties : his latest transaction proves
a failure, and on attempting to fall back on his
former position and start afresh, he finds the ground
cut away from beneath him, his reserve fund myste-
riously vanished, and his affairs in hopeless con-
fusion.

As in business, so in science, it is well to have
periodical stock-takings. Scientific facts accumulate
rapidly, and give rise to theories with almost equal
rapidity. These theories are often wonderfully
enticing, and one is apt to pass from one to another,

A

2 THE MODERN STUDY OF ZOOLOGY

from theory to theory, without taking care to estab-
lish each before passing on to the next, without
assuring oneself that the foundation on which one
is building is secure. Then comes the crash ; the
last theory breaks down utterly, and on attempting
to retrace our steps to firm ground and start anew,
we may find too late that one of the cards, possibly
at the very foundation of the pagoda, is either
faultily placed or in itself defective, and that this
blemish — easily remedied if detected in time — has,
neglected, caused the collapse of the whole structure
on whose erection so much skill and perseverance
have been spent.

Thus men of science find it well occasionally
to take stock, to look back for the moment instead
of forward, to assure themselves that their operations
since the last stock-taking have really resulted in
a gain, and to define accurately the nature and
extent of that gain.

Science has been aptly compared to a globe,
similar to our own earth — a globe with a solid hard
crust bounded by an irregular surface. The solid
crust represents ascertained facts, facts that have
been confirmed and stowed away in their proper
places ; the irregularity of its surface indicates the
unequal accumulation of facts in the various branches
of knowledge. The atmosphere by which the whole
globe is invested represents the world of speculation,
of theories — an atmosphere heavily laden with germs
and particles of truth, but germs as yet immature,
particles whose position relative to the solid crust
is not yet a fixed and determined one.

THE MODERN STUDY OF ZOOLOGY 3

Our process of stock-taking consists in defining
the boundary line between the crust and the atmo-
sphere, between earth and air ; such a process
becomes periodically necessary because the contour
of the surface is constantly changing ; particles are
continually being added to the crust, while those
whose places are already determined are liable by
reason of these additions to have their relative
positions and importance altered. Thus what was
at one time a lofty peak, a startling though estab-
lished generalisation, may become overshadowed
by the formation of a far loftier one by its side, of
which the original peak becomes but an insignifi-
cant shoulder whose original importance is soon
forgotten.

I propose, then, in the present paper to take
stock of our zoological knowledge, to attempt to
define the actual position and aims of zoological
thought, the steps by which this position has been
attained, and the methods by which it is hoped to
achieve these ends.

Such a process is of special and peculiar interest
as applied to zoology, firstly, by reason of the great
and rapid accumulation of facts that has occurred
of late years ; secondly, because of the far-reaching
and fiercely contested theories to which these facts
have given birth ; and, thirdly, because the study
of zoology includes the study of man, so that
generalisations concerning the rest of the animal
kingdom must apply also to man himself. For
these reasons, and more especially for the third one,
the theories and generalisations of zoology are

4 THE MODERN STUDY OF ZOOLOGY

always subjected to rigid and jealous scrutiny, not
only by those who make zoology a special study,
but by the world at large.

In order to know clearly with what we are deal-
ing we may, with Professor Huxley, define zoology
as " the whole doctrine of animal life," as being in
fact, if such marked alliteration may be excused, all
about animals. Now, from very remote times in-
deed there have existed not only names for different
animals, but also collective names for groups of
animals agreeing with one another in certain re-
spects but differing widely amongst themselves in
others ; collective names such as fish, under which
head a great number of animals are commonly
included, some of which, such as the whale, are
not fish at all ; or birds, including forms as diverse
as a starling and a stork, a humming bird and an
ostrich. The introduction of such collective names
marks the earliest attempts at zoological classifi-
cation.

Of such classifications we meet with examples in
the Old Testament. Thus we read of Solomon
that " he spake of trees, from the cedar tree that is
in Lebanon even unto the hyssop that springeth
out of the wall : he spake also of beasts, and of
fowl, and of creeping things, and of fishes."* The
object of the writer in the above passage is mani-
festly to bring into prominence the extent of
Solomon's knowledge, and we are certainly led to
believe that Solomon had made a personal study of
the several groups of animals mentioned — i.e., that
* i Kings iv. 33.

THE MODERN STUDY OF ZOOLOGY 5

he was a zoologist. The passage quoted bears
evidence in itself that the four groups named were
intended to include the whole of the animal king-
dom ; but any doubt on this point is removed by
the fact that in other parts of the Old Testament
the animal kingdom is distinctly divided into these
same four groups.* We are therefore justified in
speaking of this as a zoological classification.

If we examine this classification more closely we
see that the habits of the different animals, and
more especially the media in which they live, are
made the basis on which the several divisions are
founded. Thus beasts include terrestrial animals,
animals living on dry land ; fowl are those animals
that possess the power of flight and so are enabled
to live as denizens of the air ; fishes are animals
adapted for living in water ; while creeping things
probably included what we now call insects, and
any other small forms that could not be referred
readily to either of the other groups. Such a
system may be spoken of as a classification by dis-
tribution. It is one of easy application, and so far
a convenient one, but inasmuch as it takes no
account whatever of structural and physiological
resemblances and differences between the several
animals with which it deals it must be regarded as
an exceedingly primitive one.

The next classification of any great importance
that we meet with is that given by Aristotle, per-
haps the greatest and most truly scientific man in
the highest sense of the word that the world has
* e.g. Deuteronomy iv. 17, 18.

THE MODERN STUDY OF ZOOLOGY

ever known. Aristotle, like Solomon, is better
known in connection with other branches of know-
ledge than zoology ; still he devoted much attention
to the study of animals, and placed zoology on a
far more scientific basis than his predecessors had
done. It would appear that Aristotle never drew
up a formal scheme of classification ; the system
commonly ascribed to him, which is in reality
compiled from his various writings and was never
given by him in its modern form, is as follows,*
the modern equivalents of the several groups being
indicated in the right hand column.

A. Animals with red blood and a

backbone

1. Provided with four legs.

(a) Viviparous .

(b) Oviparous .

2. Provided with two legs

and two wings .

3. Devoid of legs, but pro-

vided with fins

B. Animals without red blood and

with no backbone .

1. Soft externally

2. Soft internally, hard ex-

ternally .

Vertebrata.

Mammalia.
Reptilia.

Aves.
(Pisces.
\ Cetacea.

Invertcbrata.

Mollusca.

Crustacea.

Testacea.

Insecta.

Such a classification is manifestly based on a

* Vide Claus : " Grundzuge der Zoologie " : French Translation
by Moquin-Tandon. Note B., p. 1099.

THE MODERN STUDY OF ZOOLOGY 7

totally different system to that of Solomon ; the
several groups are now characterised not by their
habits or the media in which they live, but by
resemblances and differences in anatomical struc-
ture. The branch of zoology that treats of the
structure of animals is called Morphology; hence
Aristotle's classification may be contrasted with
that of Solomon as being not a classification by
distribution, but a morphological classification.
Inasmuch as the latter springs from a closer
and more accurate acquaintance with animals than
the former, it is a better and more scientific one
and may be taken as marking a distinct and very
important advance in the study of zoology.

The next writer on zoology of any great
importance is the elder Pliny, who lost his life
A.D. 79, at the celebrated eruption of Mount
Vesuvius by which Pompeii and Herculaneum were
destroyed. Pliny was to a far greater extent than
Aristotle a professed zoologist, and left a volumin-
ous work on natural history in thirty-seven books.
He divided the animal kingdom into four main
groups, which he named as follows :

1. Animalia terrestria ;

2. Animalia aquatilia;

3. Volucres ;

4. Animalia insecta ;

i.e.t he classified animals according as they lived on
the ground, in the water, or in the air; dividing
them into terrestrial, aquatic, and volatile, with a
distinct class for those animals, such as insects,
which do not belong to any element exclusively.

8 THE MODERN STUDY OF ZOOLOGY

This is clearly a classification by distribution, and
therefore differs totally from that of Aristotle, while
it agrees in principle with that of Solomon. This
agreement, however, is not only in principle ; if the
two schemes of classification be compared it will be
seen that they are really identical, a point of some
interest. Thus, Pliny's Animalia terrestria are the
same as the beasts of Solomon ; the Animalia
aquatilia as the fishes ; while Volucres are obviously
equivalent to fowl, and Animalia insecta to creeping
things. The sole difference between the two
systems is in the order in which the several groups
are arranged. It would, therefore, appear, that
while Aristotle was a long way in advance of any
of his predecessors, Pliny, who lived more than
400 years after Aristotle, not only made no
advance, but even fell back on the very empirical
classification that was in use in the days of
Solomon, iioo years previously, and that had
probably been in use for a still longer time.

As Pliny is a writer who owed a considerable
part of his reputation to his work on natural
history, it may not be inappropriate here to quote
the criticism passed on him many centuries after
by Cuvier, in order to support my statement that
Pliny, instead of placing zoology on a more
scientific basis, in reality did it incalculable damage,
and threw it back as a science to the condition in
which it had been before Aristotle's time. Cuvier's
words are as follows :* — " In general, he is only a
compiler, and, indeed, for the most part, a compiler
* " Biographic Universelle," xxxv.

THE MODERN STUDY OF ZOOLOGY g

who has not himself any idea of the subjects on
which he collects the testimony of others, and
therefore cannot appreciate the truth of their
testimonies, nor even always understand what they
mean. In short, he is an author devoid of
criticism, who, after having spent a great deal of
time in making extracts, has ranged them under
certain chapters, to which he has added reflections
that have no reference to science properly so called,
but display alternately either the most superstitious
credulity or the declamations of a discontented
philosophy, which finds fault continually with man-
kind, with nature, and with the gods themselves."

Pliny's influence on zoological thought, though
most pernicious, was sufficiently great to com-
pletely outweigh his illustrious Greek predecessor ;
and to this must, I think, be ascribed in great part
the almost complete gap in zoological literature of
any value that extends from the time of the Roman
zoologist to about the sixteenth century. It was
not, indeed, until nearly the middle of the eighteenth
century that a system of zoological classification of
any permanent value was proposed. For this we
are indebted to the great Swedish naturalist Lin-
naeus, the founder of modern natural history as he
has been well called.

The system of classification proposed by Linnaeus
was, like that of Aristotle, a morphological one,
based on resemblances and differences of structure
in the several animals and groups of animals. He
divided the whole animal kingdom into six classes,
defined as follows : —

io THE MODERN STUDY OF ZOOLOGY

A. Cor biloculare biauritum, sanguine

calido rubro.

1. Viviparis. Mammalia.

2. Oviparis. Aves.

B. Cor uniloculare uniauritum, san-

guine frigido rubro.

1. Pulmone arbitrario. Amphibia.

2. Branchiis externis. Pisces.

C. Cor uniloculare inauritum, sanie

frigida alba.

1. Antennatis. Insecta.

2. Tentaculatis. Vermes.

Of morphological classifications there are two
principal varieties, classification by definition and
classification by type. The Linnaean classification
is a typical example of the former of these. In it
the whole animal kingdom is divided up into groups
of convenient size, each characterised by the pre-
sence or absence of some one, two or more easily
recognisable features ; stress being laid on the
differences between the several groups, rather than
on the resemblances between the several animals
included in each individual group. An illustration
will perhaps serve to give a clearer idea of what is
meant. Take a piece of paper and make a number
of dots on it in a perfectly irregular manner. We
want to classify these dots, to arrange them in
groups : if we were to classify them by definition,
we should divide the paper by means of lines pass-
ing between the dots into a number of compartments
of convenient size, to which we should give distinc-

THE MODERN STUDY OF ZOOLOGY 11

tive names ; we should then define the position of
any one dot by simply saying in which division it
was. As our whole paper is divided up, every dot
must fall into some one or other of these divisions,
so that our classification is at any rate a simple
and a convenient one.

Or, again, imagine a map of England in which
the county boundaries are laid down, but all the
towns and villages are left out ; such a map would
give us a classification of the inhabitants of Eng-
land, and a classification by definition. Stress is
laid simply on the boundary lines between the
several divisions, and the sole interest attaching to
any particular individual consists in the question on
which side of a given arbitrary line he happens to
reside. It follows also that in such a scheme those
individuals who reside in the centres of the several
counties are subordinate in interest to those near
the margins of the counties, since about these latter
there may be doubt as to which division they should
be referred to, while such doubt can hardly exist in
the case of the former. Such a map might be very
useful, and for purposes of minor importance, such
as a parliamentary election or a cricket match,
might contain all the information necessary, the
sole interest consisting in which side of an arti-
ficially drawn line a given individual happened to
live.

As the knowledge of anatomy advanced ; as
zoologists became gradually acquainted with the
structure of a larger and continually increasing
number of animals ; as the microscope in the hands

12 THE MODERN STUDY OF ZOOLOGY

of Malpighi, Swammerdam, and their successors
gradually revealed the details of minute structure
and rendered possible a correct appreciation of the
anatomy of animals previously too small to be
investigated, it was gradually realised that the
Linnaean system, with its hard and fast lines of
division, no longer represented the actual state of
our knowledge, and classification by definition
gradually gave way to the second form of morpho-
logical classification — classification by type.

We may explain the difference between the two
by means of our former illustrations : thus, to take
our first case, we no longer divide our paper by
artificial lines, we now look to the dots themselves ;
we find that the dots are not always the same
distance apart, that many of them fall naturally
into groups of various sizes ; each well-marked
group we give a name to, and the central member
of the group round which the others seem to be
arranged we call the type of the group : of the
remaining dots, some are so close to our big groups
that we include them with these, others form
distinct smaller groups of their own, whilst some
solitary ones stand quite apart and isolated from all
the rest. Or we may, to take our second instance,
illustrate classification by type by a map of
England, in which the county boundaries are left
out, but all the towns and villages marked. Here
we have large centres such as London or
Manchester, containing large numbers of inhabi-
tants, and representing distinct types ; smaller
centres lying immediately round them, not definitely

THE MODERN STUDY OF ZOOLOGY 13

connected with them as yet, but destined ultimately
to be so, other small centres lying at a distance
from the large ones, and constituting distinct
types, and, finally, isolated houses representing
species of animals widely separated from their
fellows, and forming for the time at any rate small
but distinct types of their own.

The distinguishing characteristics of classification
by type, and especially the points in which it con-
trasts most strongly with classification by definition,
have been admirably stated by the late Master of
Trinity College in the following words : — "The
class is steadily fixed, though not precisely limited ;
it is given though not circumscribed ; it is deter-
mined, not by a boundary line without, but by a
central point within ; not by what it strictly
excludes, but by what it eminently includes ; by an
example, not by a precept ; in short, instead of a
definition we have a type for our director. A type
is an example of any class, for instance, a species
of a genus, which is considered as eminently possess-
ing the characters of the class. All the species
which have a greater affinity with the type-species
than with any others form the genus, and are
ranged about it, deviating from it in various direc-
tions and different degrees."*

Such a classification represents the real affinities
of animals much more truthfully than classification
by definition. The sharp boundary lines, of which
nature knows nothing, and which formed the main

* Whewell, " The Philosophy of the Inductive Sciences,"
vol. i. pp. 476-7.

14 THE MODERN STUDY OF ZOOLOGY

feature of the older system, are here swept
away ; the resemblances of animals are made
of more weight than their differences ; and no
attempt is made to define the limits of the several
groups.

This doctrine of animal types was first brought
forward prominently by Cuvier and Von Baer at
the commencement of the present century. Cuvier,
in his latest system of classification, distinguished
four leading types or plans of structure in the animal
kingdom, to one or other of which all animals could,
according to him, be referred. Mainly owing to
the weight of Cuvier's authority this doctrine of
types made considerable progress during the first
half of the present century ; it never, however,
wholly replaced classification by definition, and
probably never would have done so ; for, in the
first place, the essence of classification is conveni-
ence, and classification by definition is far more
convenient for the ordinary purpose of a zoologist
than classification by type ; and, secondly, although
the idea of types expressed a great and important
truth; yet it was but the partial expression of a
still greater one, which, when fully developed,
was destined to completely overthrow all former
attempts, and to reveal the only true and un-
assailable basis of classification. The gradual
rise of this new doctrine we have now to notice
briefly.

About the commencement of the present century
two new influences began to make themselves felt
in zoology, two new branches that were afterwards

THE MODERN STUDY OF ZOOLOGY 15

to exert great influence on zoological thought began
for the first time to receive serious attention.
These were Palaeontology and Embryology.

Palaeontology, the investigation of extinct animal
forms, of those animals and portions of animals
known to us only through their fossil remains, was
first studied systematically and raised to the rank
of a science by Cuvier. Previous to his time
fossils had not received serious attention ; even
their animal origin was far from being commonly
recognised, and the most absurd ideas were in
vogue as to their nature and origin ; some sup-
posing them to be mere freaks of nature, others
that they were models used by the Creator
when he was preparing to stock the earth with
animals.

Cuvier did not confine himself to demonstrating
that these fossil remains must have proceeded from
animals that once lived on the surface of the earth;
he studied the distribution of fossils in the different
geological strata with great care, and was led to
form generalisations of extreme value and interest.
The most important of these conclusions are con-
tained in his " Theory of the Earth,"* and are to the
following effect : — In the oldest strata of all there
are no fossil remains at all ; organised beings were
not all created at the same time, but at different
times, probably very remote from one another ; the
fossil remains of the recent strata approach far

* First published in 1798 as the preliminary discourse to the
" Recherches sur les Ossemens Fossiles ; " republished separately,
with many additions, in 1825.

16 THE MODERN STUDY OF ZOOLOGY

nearer to the existing forms of animals than do
those of the older strata; finally, of the highest
forms of animal life — man and the quadrumana —
there are no fossil remains whatever.

From these conclusions, the importance of which
it is impossible to overrate, Cuvier was led to found
his doctrine of Catastrophtsm, according to which
there have been periodical annihilations at long
intervals of time of all the animals living on the
earth at the time ; each cataclysm being followed
by the creation of a totally new set of animals,
which though agreeing in many points with their
predecessors, yet presented many marked differences
from them.

Cuvier's doctrines, however, did not meet with
general acceptance among geologists, and the pub-
lication of the first edition of " The Principles of
Geology," by Sir Charles Lyell, in 1830, two years
before Cuvier's death, may be said to mark the
complete overthrow of the doctrine of catastrophism
so far as the changes that have taken place in the
earth's crust are concerned. A closer study of
what is at present occurring on the earth's surface
showed that there are now in action forces amply
sufficient, given time enough, to produce changes
as great as any of which we have geological record ;
that the elevation of great mountain chains is not
due to the sudden action of immeasurably great
forces but to the long continued action of appar-
ently insignificant ones ; and that there is not only
no evidence whatever of the occurrence of the
supposed catastrophic periods, but that all the

THE MODERN STUDY OF ZOOLOGY 17

evidence on the point tends to prove that such
periods never have occurred.

Though catastrophism thus received its death-
blow so far as the crust of the earth was concerned,
men still hesitated to apply the same reasoning to
the fossil remains of animals, and in spite of the
geological evidence the doctrine of catastrophism,
i.e., of periodical annihilations and re-creations,
continued to meet with acceptance so far as these
fossil remains were concerned.

All this time there was steadily developing and
gradually acquiring definite shape a doctrine des-
tined ultimately to overthrow Cuvier's theories
concerning fossils as completely as the geologists
had done those dealing with the earth's crust.
This was the doctrine of the Mutability of Species.

Cuvier, as we have seen, maintained that species
were all due to separate acts of creation ; the new
doctrine maintained that species were not immutable,
but that one species might give rise to two or more
new ones. The actual birth of this doctrine is
involved in some obscurity ; it is not quite clear
when it first arose, or to whom the credit of its
origination is due. It was clearly recognised and
advocated by the illustrious Goethe in 1796, but
whether this is the date of its birth is not clear.

Its greatest advocates were Lamarck and St.
Hilaire, its greatest opponent Cuvier, and long and
bitter was the struggle. Though the two former,
and more especially Lamarck, worked out the
doctrine in the most elaborate manner, yet they
were unable to point out the causes at work in the

B

i8 THE MODERN STUDY OF ZOOLOGY

supposed transformation of species ; they were
unable to show why species should become modified
into other species, and so, the onus probandi lying
with them, victory in the eyes of the world rested
with Cuvier. So complete was this victory con-
sidered at the time that for nearly thirty years
after his death Cuvier's authority was sufficient to
keep this new doctrine in abeyance.

At length came the most eventful epoch in the
history of zoology, the simultaneous announcement
by two independent investigators, Charles Darwin
and Alfred Russel Wallace, of the doctrine of
Natural Selection, at the meeting of the Linnaean
Society on July 1st, 1858. This doctrine effected
for the animal world exactly what the geologist had
already done for the earth's crust ; it showed that
there are now in operation causes sufficient, given
time enough, to produce all the changes requisite
to convert the extinct fossil species into those now
living on the earth, causes that must have been in
operation since life first dawned on the earth, causes
that must inevitably have led to the passage of
species into species.

In this way a complete and consistent theory of
the history of life on the earth was at length
obtained — not only what had actually occurred, but
how and why it had occurred. And now at length
the true meaning of the laws of Cuvier regarding
the distribution of fossil remains was seen, those
laws which had led him to form his erroneous
theory of catastrophism. For instance, it was seen
now that the reason why the fossils of recent beds

THE MODERN STUDY OF ZOOLOGY 19

resemble existing forms more closely than do those
of the older beds is simply that there has been a
continuous process of evolution ; that the fossils of
the recent beds are the ancestors of the now living
forms, the descendants of the fossils of the older
beds, and thus occupy an intermediate position
genealogically ; in which case their intermediate
position structurally becomes at once intelligible.

Recognising these facts, attempts were soon
made to reconstruct the path of descent, to trace
out the pedigree of some given species. This was
first accomplished with any degree of success in
the case of the horse. The horse, the zebra, and
the ass stand alone among mammalia in pos-
sessing but one complete functional toe on each
limb ; they must either have been specially created
as such, or must have been derived from some
more typical mammalian form. Owing to their
large size, the fact that bones are fairly easily pre-
served as fossils, and the great time that must have
elapsed during the gradual transformation of some
typical mammal to the highly specialised horse, it
is only reasonable to expect that some direct evidence
of this transformation should be forthcoming.
Consequently this furnishes a very good test case.
Without entering into details, which would be un-
necessary in what is now so familiar an instance,
it will suffice to state that a series of fossil forms is
now known furnishing a complete gradation from
older tertiary forms with four or five toes on each
foot, through newer tertiary forms to the horse with
its one functional toe on each limb ; that, in other

20 THE MODERN STUDY OF ZOOLOGY

words, the pedigree of the horse has been completely
and satisfactorily worked out.

Another familiar but striking example is afforded
by birds. These are very highly specialised forms,
and stand apart from other vertebrates in a number
of anatomical points. We are now acquainted
with a large number of fossil forms serving to
connect birds with reptiles, and showing the several
gradations by which reptiles gradually lost their
teeth, acquired wings and feathers, and became
birds.

If then it is possible by the aid of fossil remains
to reconstruct the pedigree of some particular form
or group of forms, why should it not be possible
for all animals ? As the possibility of this recon-
struction gradually dawned on man, so it was slowly
realised that such a reconstruction, such a pedigree,
would in itself be a perfect system of classification,
and the only really natural one. Such a classifica-
tion may be spoken of as a genetic or genealogical
classification.

The idea of such a classification is familiar to
all of us through the medium of our own special
genealogies : these generally take the form of a tree
in which the stem represents our earliest ancestor,
who, in this country at least, usually takes the form
of some impoverished adventurer, whom we should
probably be intensely ashamed of could we see him
in the flesh, whose sole virtue lies in the fact that
he " came over with the Conqueror," and whose
sole possessions of any importance appear to have
been a crest, a motto, and a coat of arms ; the

THE MODERN STUDY OF ZOOLOGY 21

primary branches of the stem represent the off-
spring of this all-important ancestor ; the secondary
branches their offspring, and so on, each branch
denoting a generation. Some of the branches die
off, stop, and become extinct ; others persist and
thrive : the ultimate branchlets of these last bear
the leaves, which are the actually living representa-
tives of the family, on the topmost of which we
inscribe our own name.

A genetic classification of the animal kingdom
takes a similar form ; the stem represents the
earliest and most primitive form of animal life, the
branches the several forms derived successively
from this and from one another, while the terminal
leaves represent the actually existing forms. If we
draw horizontal lines across our tree, these may be
taken to represent the different geological ages ;
then the branches cut by any one line or plane will
represent the life on the earth at that period. Some
of the branches never reach the top of the tree ;
these represent forms of life that attained their full
development in some of the earlier epochs and then
became extinct.

Such a genetic classification is at once felt to be
the only really natural one, to be in fact an ideal
classification. It is simply an embodiment of fact,
not an expression of opinion : it is therefore ab-
solutely unassailable, and what is more, true for all
time. The recognition of the possibility of such a
classification is the distinguishing feature of modern
zoology, and the determination of such genetic
histories or phylogenies of the several groups of

22 THE MODERN STUDY OF ZOOLOGY

animals is the goal towards the attainment of which
the efforts of zoologists are directed.

Having traced the way in which this idea arose,
how increase of knowledge and perfection of
methods caused classification by definition to re-
place the crude ideas embodied in the classifications
adopted by Solomon and Pliny, and then in its turn
to give way to classification by type, and having
seen how finally the doctrine of natural selection
rendered possible the conception of a genetic
classification, it may not be out of place to devote
a few words to considering the methods by which
it is proposed to attain this end. These are as
follows :

1. The accumulation of anatomical facts con-
cerning as large a number of animals as possible ;
this requires no further explanation.

2. The systematic study of the geographical
distribution of the different animals and groups of
animals, both recent and extinct. This, which is
a comparatively new branch of zoology, has already
yielded results of immense importance, and pro-
mises to yield others of even greater value in the
future.

3. Palaeontology, or the study of the anatomy
of extinct animals and their relations to existing
forms. We have seen above that it is to palaeon-
tology that we owe the first successful attempt to
reconstruct the pedigree of some one given form ;
and inasmuch as the connecting links between the
several groups of animals are almost necessarily
extinct, it is clear that the evidence yielded by

THE MODERN STUDY OF ZOOLOGY 23

those extinct forms must be invaluable. Still it
must be borne in mind that it is only certain
animals, and of these animals only certain parts, that
are capable of being preserved as fossils, so that
palaeontology can only help us as regards certain
groups of the animal kingdom, and even concerning
these its evidence must necessarily be very im-
perfect and fragmentary. Indeed, had we only
these three methods to aid us, the attempt to re-
construct the pedigree of the animal kingdom could
only be successful to a very partial and limited
extent. Fortunately there is another method which
supplies us with evidence of the very kind we
most want, and which bids fair to far outweigh
in importance the other three methods. This is :

4. Embryology, or the study of the actual de-
velopment of existing forms, the several changes
which they undergo during their gradual evolution
from the ovum. This, which is the second of the
two new influences referred to in a preceding page,
is, like palaeontology, a young science, but one that
has thriven mightily of late years. It was not till
towards the middle of the present century that it
began, in the hands of von Baer, to assume its
modern form, and it is only of recent years that it
has revealed its enormous powers as an instrument
for the solution of the hard problems of phylogeny.
Von Baer propounded the doctrine of animal types
quite independently of Cuvier ; and he went further
than his illustrious contemporary, for he showed
that not only as far as structure was concerned
might animals be arranged under certain definite

24 THE MODERN STUDY OF ZOOLOGY

types, but that each type had its own mode of
development, and that all the animals included in
any one type agreed with one another in the funda-
mental features of their development. Thus the
several members of the type of animals known as
Vertebrata all develop in a fundamentally similar
manner ; in their earlier stages they resemble one
another so closely that it is often no easy matter
to distinguish one from another, the characteristic
differences not appearing till the later stages of
development. A good example of this is given by
von Baer himself, as follows : " In my possession
are two little embryos in spirit, whose names I
have omitted to attach, and at present I am quite
unable to say to what class they belong. They
may be lizards, or small birds, or very young mam-
mals, so complete is the similarity in mode of
formation of the head and trunk in these animals.
The extremities, however, are still absent in these
embryos ; but even if they had existed in the
earliest stages of their development we should learn
nothing, for the feet of lizards and mammals, the
wings and feet of birds, no less than the hands and
feet of men, all arise from the same fundamental
form."

As soon as the idea of the mutability of species
was fairly grasped, the reason of these embryo-
logical resemblances became apparent, and it was
seen^that if two species or groups of animals develop
in the same way this is due to their being geneti-
cally allied to one another.

Further consideration of this question led ulti-

THE MODERN STUDY OF ZOOLOGY 25

mately by a series of steps, which space forbids me
to notice, to the promulgation of what is known as
the recapitulation hypothesis. This which is found
lurking in the works of von Baer, but first received
definite expression from Fritz Muller, is commonly
stated thus : The development of the individual is
an epitome of the development of the species — i.e., the
actual changes through which an animal passes in
its development from the egg to the adult repre-
sent in a condensed form its genealogical history or
pedigree.

Thus embryology provides us with a completely
new and very accessible clue to the determination
of animal genealogies — a clue of such extreme
value that we cease to wonder at the immense
importance attached to it by the modern zoologist,
or at the time and perseverance that are expended
in attempts to penetrate its mysteries. For the
clue thus afforded is the one on which we have
chiefly to rely in our efforts to unravel the knotty
problems of genealogy, and it is a clue in the
following up of which extreme caution is necessary.
If the recapitulation hypothesis were strictly true,
if the ontogenetic history of an individual were an
accurate and complete reflection of its specific
history or phylogeny, then our task would be a
comparatively light one. But it is not so. As
Fritz Miiller has pointed out with admirable clear-
ness* the ontogeny of an individual is not a simple
abbreviated history of its phylogeny, but a history
obscured and falsified by a variety of causes, the

* Fur Darwin (English translation), chap. xi.f pp. no seq.

26 THE MODERN STUDY OF ZOOLOGY

most important of which are the tendency to shorten
the processes of development, thereby causing the
omission of stages which may be of extreme his-
torical significance ; and, secondly, modifications
introduced by natural selection into the ontogenetic
history in consequence of the struggle for existence
which free-living larvae have, in common with adult
forms, to undergo.

Thus, in spite of the powerful aid afforded us by
embryology, our problem is still one of extreme
difficulty, requiring for its solution great patience,
great manipulative skill, and, above all, the faculty
of distinguishing accurately between essentials and
accidentals. A good beginning has already been
made, mainly through the energy of zoologists in
other countries than our own. England has not
yet contributed her fair share towards the solution
of a problem the conception of which was first ren-
dered possible by the magnificent labours of one of
the greatest of her children, Charles Darwin.

II

THE INFLUENCE OF ENVIRON-
MENT ON THE STRUCTURE
AND HABITS OF ANIMALS

I PROPOSE to ask your attention for a short time
to the consideration of the dependence of any
body, living or dead, organised or unorganised,
organic or inorganic, on its environment, i.e., on
the external conditions surrounding and acting on
it at a given time, and I propose to consider this
more especially with regard to animals.

Of environment, as affecting inanimate objects,
many admirable examples are afforded us by Physics.
For instance, we speak of water as a definite sub-
stance having a definite existence, but we know well
enough that for the existence of water as such
certain definite relations of temperature and pres-
sure are essential. At the normal pressure water
will not exist as water at a temperature below o° C.
Similarly, if we raise the temperature beyond a
certain limit, 100° C., the existence of water as
water again becomes impossible. If we alter the
pressure the relations change again, for when the

28 THE INFLUENCE OF ENVIRONMENT ON

pressure is increased beyond the normal amount we
find that the water will remain liquid at a tempera-
ture at which it would previously have assumed the
solid form of ice.

Let us take another example from Chemistry.
Chemical formulae have a very definite appearance,
yet every chemist knows well that the reactions
they indicate will only occur provided certain
conditions of environment be fulfilled, especially
those of temperature and pressure ; and that a
change of no great extent in environment may cause
either a completely different reaction or even an
actual reversal of the previous action. For in-
stance, Mercury at 300° C. combines with oxygen to
form oxide of mercury ; now if the temperature is
raised to 400° C. this action is reversed.

I have purposely chosen simple and familiar
examples in order to illustrate with as little loss of
time as possible what we mean by environment. The
points to which I wish to direct attention are these:
Firstly, that the actual condition of any body may
be regarded as the resultant of the several forces
acting on it at the time ; secondly, that therefore,
if we know all the conditions of environment, we can
calculate with certainty the actual condition of the
body ; thirdly, that a change of known amount in
one of the elements of the environment will produce
a definite and calculable change in the conditions
of the body acted on. In other words, that not
only has environment a marked and definite action,
but that this action is capable of exact measurement.

Now let me turn to the more immediate subject

THE STRUCTURE AND HABITS OF ANIMALS 29

of my paper, viz., the influence of environment on
the structure and habits of animals. It will, of
course, be at once conceded that changes in environ-
ment not only can but do produce changes of very
great magnitude in both the structure and habits of
animals — for example, domestic breeds of cattle,
&c.; but in most of these cases the conditions are
very complicated, and it is impossible to assign to
each factor its true value. What I wish to prove
to you, if possible, is that change of environment
not only produces changes of structure and habit,
but also produces definite and calculable changes,
so that a definite change in environment is always
followed by a definite change of structure, and the
result can be predicted with as much certainty as a
chemical reaction.

For this purpose all ordinary examples fail us
because the environment is too complicated, e.g.,
pigeon fanciers can produce at pleasure a bird with
any number of feathers in the tail they wish, and
with almost any variety of plumage, but as to which
of the elements of a complicated and artificial en-
vironment the result is due we are completely in the
dark. For this purpose it is necessary to choose
our examples carefully, and the number of available
ones that are not of too technical a character for
general discussion is very limited.

We have seen that it is very easy to establish a
general relation between changes of environment
and changes of structure, and that the one series is
followed at once by the other. Now this general
relation is not sufficient for our proposition, much

30 THE INFLUENCE OF ENVIRONMENT ON

more rigid proof being necessary before we can
accept it. I wish first to ask your attention to one
or two cases in which this proof appears to be
present, and in which we are able to establish not
only a general but a direct causal relationship
between environment and structure ; as unquestion-
able indeed as that which holds in the case of oxide
of mercury.

Now we know of at least one well authenticated
case in which the structure of an animal responds
in as precise a manner to simple and definite
changes in one of the elements of the environment
as certain chemical reactions do to alterations of
temperature.

Artemia salina is a small aquatic crustacean
about half an inch in length, and somewhat like a
small shrimp in appearance ; its distribution is
peculiar, inasmuch as it is only found in water
containing from 4 to 8 per cent, of salt, a much
higher proportion than is found in the sea. It
consequently can only occur in salt lakes and in
such places as the salt pans of Lymington and
other similar places where the sea-water is allowed
to gradually concentrate by evaporation. Another
form, Artemia Milhausenii, is known, whose habits
are still more peculiar, for it is only found in water
containing at least 25 per cent, of salt ; its dis-
tribution is consequently a very limited one, since
it can only occur in certain inland lakes in which,
owing to special circumstances, the water attains
this very high percentage of saltness.

Artemia Milhausenii differs from Artemia salina

THE STRUCTURE AND HABITS OF ANIMALS 31

in so many points of structure that the two forms
have always been unhesitatingly ranked as separate
species. Of these differences one will suffice for
the present purpose. In Artemia salina the tail
is split at the end into two long pointed lobes,
each of which bears a large number of bristles and
hairs of a peculiar structure ; the tail in the other
species, Artemia Milhausenii} is abruptly terminated,
and presents only a very slight separation into two
blunt rounded lobes, which lobes are completely
devoid of hairs or bristles. There are besides this
many other points of difference between the two
forms which need not here be noticed in detail.

Schmankewitsch has recently made the very in-
teresting discovery that the anatomical differences
between the two species depend directly and solely
on the different percentages of salt in the water
they inhabit. He finds if he takes Artemia salina,
living in water containing 4 per cent, of salt, and
gradually increases the percentage of salt, that the
structure of the Artemia becomes gradually modi-
fied ; the lobes of the tail become shorter and
shorter, and the hairs diminish in size and number.
Ultimately, by increasing the strength of the salt
up to 25 per cent., he succeeded in completely
transforming Artemia salina into the distinct species
Artemia Milhausenii. Not only can the trans-
formation of one species into another be effected
in the laboratory, but it can be demonstrated
to actually occur in Nature. Of this fact
Schmankewitsch has recorded the following very
striking instance : Two lakes in the neighbour-

32 THE INFLUENCE OF ENVIRONMENT ON

hood of the Black Sea were separated from one
another by a dam ; of these the upper and larger
lake contained water with about 4 per cent, of salt
in which Artemia salina occurred in great numbers.
The water of the lower lake contained about 25
per cent, of salt and had no Artemia living in it at
all. From some cause or other the dam burst, and
the water from the upper lake rushed down into
the lower and smaller one, large numbers of Artemia
salina being carried down by the flood. The dam
was repaired, and the water of the lower lake, the
strength of which had been reduced by the flood
to about 8 per cent, of salt, began slowly and
gradually to regain its former strength. Specimens
of the Artemia were taken out and examined from
time to time, and it was found that as the per-
centage of salt rose the tail-lobes of the Artemia
gradually got shorter and blunter, the hairs got
smaller and fewer, and the whole animal gradually
became transformed from Artemia salina to Artemia
Milhausenii. In about three years' time the water
had regained its former strength of 25 per cent.,
and the Artemice had become completely trans-
formed into Artemia Milhausenii.

Not satisfied with this, Schmankewitsch next
tried the reverse experiment. Starting with
Artemia Milhausenii he gradually decreased the
strength of the water in which they lived by adding
fresh water, and had the satisfaction of finding that
as the percentage of salt diminished so the tail-
lobes of his Artemice became longer and their
hairs more abundant, and when the strength of

THE STRUCTURE AND HABITS OF ANIMALS 33

the water was reduced to about 6 per cent, all the
Artemice were converted to typical specimens of
Artemia salina. Having thus satisfied himself of
the causal relations between the two species of
Artemia he determined to push his experiments
still further. He therefore went on adding fresh
water after he had obtained Artemia salma, and by
this means converted Artemia salma into a well-
known fresh-water form that had previously been
considered not only a separate species but even a
distinct genus — Branchipus.

From these very striking experiments it would
appear that the relation between the structure of
the animal and the degree of saltness of the water
it inhabits is a perfectly definite one, so much so,
that if the percentage of salt be told us, we are
able to foretell with perfect confidence and certainty,
not only what will be the general character of the
animal inhabiting it, but such minutiae as the size
and shape of its tail-lobes, and even the number
of hairs borne by these lobes.

Here then is a very striking proof of the direct
action of environment on structure ; a phenomenon
as definite as a chemical reaction. Nor must the
case be made light of or explained away by saying
that it simply shows that zoologists were at fault
in describing these forms as distinct genera and
species, and that the experiment showed them to
be mere varieties ; for it must be noticed that the
characters by which the three forms differ are
characters of the very kind that are invariably
employed by zoologists to distinguish the genera

34 THE INFLUENCE OF ENVIRONMENT ON

and species of Crustacea ; while the ultimate cri-
terion of distinct species — fertile breeding — bears
the matter out also, for no one of the three forms
is capable of living in the medium dwelt in by
either of the others.

We also know of several cases among butterflies
in which, out of a very complex environment, we
are able to put our finger on the one element which
is the actual determining cause of the structural
changes that we observe. One of the most im-
portant of these examples is furnished by two
butterflies of widely different appearance, Vanessa
levana and Vanessa prorsa.

Vanessa levana has a red ground colour, with
black spots and dashes and a row of blue spots
round the margin of the hind wings ; Vanessa prorsa
is deep black, with a broad yellowish-white band
across both wings and with no blue spots. These
were formerly called distinct species, but have
recently been shown to be varieties of one and the
same species. The relation between them is a
very definite one, and they are what is called
seasonally dimorphic : that is, they are double
brooded, and have two generations in the course of
the year — a winter brood and a summer brood. In
the summer or autumn Prorsa is alone met with,
and the pupae developed from the autumn brood of
eggs of Prorsa hybernate during the winter and
hatch in April or May of the following spring, not
as Prorsa, but as Levana. This in turn produces
a brood of eggs which develop into Prorsa again.
This phenomenon was investigated by Weismann,

THE STRUCTURE AND HABITS OF ANIMALS 35

who found it was to be attributed to the direct
action of cold and heat. He found that if the
pupae of Levana, which would ordinarily have
produced Prorsa, were put in a refrigerator
for one or two months the butterflies emerged
partly as Levana and partly as another form, inter-
mediate in many respects between Levana and
Prorsa. Thus the direct action of cold causes
pupae, which in the normal state of things would
have produced Prorsa, to produce Levana.

The reverse process, however, is not successful,
for heat does not change Levana into Prorsa. The
explanation of this is, perhaps, not out of place,
although it has no direct bearing on the subject in
hand. Levana is the parent form, and in the
glacial epoch there was only one single brooded
variety and one generation in the year. The
lengthening of the summer gave rise to a second
generation, and the gradual action of climate caused
the Prorsa form to arise. Now, Levana being the
parent form, the effect of cold in causing pupae
that would normally have become Prorsa to become
Levana is simply a case of reversion. The reverse
action could clearly not occur. It is a long time since
the glacial epoch was over, and therefore probably a
long time since the two forms Levana and Prorsa
were established ; hence the reconversion by cold
is only partial.

A similar example is afforded by another butter-
fly, Pieris napi} the green-veined white. This is
also double brooded, and occurs as two varieties,
each producing the other. The winter form is

36 THE INFLUENCE OF ENVIRONMENT ON

rather smaller, and has more black at the bases
and less at the tips of the wings than the other
variety. The green veining on the under surface
is also more marked. If the pupae of the summer
brood are placed in a refrigerator for three months,
and then in a hothouse, they hatch in about three
weeks as the winter form.

These illustrations I have cited show the direct
action of cold producing a result which can be
calculated on beforehand with certainty. By the
long-continued application of heat the summer form
has been produced, and if the environment is altered
and cold substituted for heat, we reverse the action
and bring back the summer form to the original
winter one.

Now let me refer to an instance in which the
evidence is as yet incomplete — an instance some-
what more technical than the previous ones. I
choose it purposely because it is of a somewhat
abstruse kind, and will therefore serve to show
that some of the more intricate problems con-
cerning the exact nature of the relation between
environment and structure and habits may be well
within our grasp.

Many animals possess the power of changing
colours — e.g., the chameleon, frog, many fish, and
the cuttle-fish. This power depends on the
presence of chromatophores or pigmented bodies
found in the skin, which have the power of
changing shape. These are of different colours
and are arranged in layers, some near the surface,
some deeper; the light yellow cells being most

THE STRUCTURE AND HABITS OF ANIMALS 37

superficial, then the red and brown, and the black
deepest of all. If the superficial ones contract,
the deeper ones will become more apparent, and
vice versa. These changes of colour always bear a
relation to the surface on which the animal is placed
at the time, and are therefore supposed with good
reason to be protective. Many attempts have been
made to explain this.

Lister believed that the irritation which excites the
action of the chromatophores does not act on them
directly, but through the intermediation of the optic
nerve. It is essential for this action that the eye
and optic nerve should be present and healthy, the
action ceasing if the eye is destroyed or the nerve
cut. In support of this view he pointed out that
blind fish are always dark. Pouchet suggested
two possible paths of impulse — the spinal cord
and the sympathetic system. By dividing the
spinal cord the path of impulse was shown to pass
along the sympathetic nerves. Again, by dividing
some branches of these, he produced zebra-like
markings of the skin. Dewar showed that
different colours of the spectrum influence the
eye differently and cause electric currents of
different intensity according to the light employed.
These currents are strongest with yellow light,
weakest with purple, and nil with black. Semper
pointed out that, though electric currents and nerve
currents are far from being the same thing, we may
not unfairly assume that the nerve current, like the
electric current, is greatest with yellow light and nil
with black. Thus a black ground, which absorbs

38 THE INFLUENCE OF ENVIRONMENT ON

nearly all the light, and therefore reflects little or
none, will stimulate the eye to a very faint extent,
and hence cause a very feeble current, if any at all,
and so the chromatophores will remain unstimulated,
and as in the case of the blind fish, the fish will
remain dark. On the other hand, light from a
red or blue object will cause a current of rather
greater intensity, and will cause the brown or
red chromatophores to expand, and so more or
less conceal the black ones. Finally, yellow
light will cause the strongest stimulation and all
the chromatophores will give out processes ; the
yellow ones, being most superficial, will be the
most conspicuous, and will hide the deeper and
darker ones to a greater or less extent.

In the foregoing remarks I have endeavoured to
show, first in a general way, and afterwards by the
consideration of a few selected examples, that the
relation between an animal and its environment
is a very definite and direct one, and in many cases
a calculable one.

We have seen from certain special cases that an
organism responds promptly and with certainty
to certain changes in its environment. Can we
not extend this and speak of the actual structure
of an animal as the resultant of the various forces
acting on it, both external and internal ? Small
changes in environment seem to give rise to results
altogether disproportionate with themselves ; but
then we must remember that the bodies of animals
are of very complex chemical composition and in
very unstable equilibrium, and in such cases a

THE STRUCTURE AND HABITS OF ANIMALS 39

small force may very easily cause a very con-
siderable rearrangement. Turning to psychology,
we see that the tendency of modern research is
to show that whatever we do or think is merely
the resultant of what we have done or thought
or suffered previously.

It is not difficult to show that, even in our own
worldly transactions, changes of environment often
produce not only direct and immediate changes
and readjustments, but also definite and calculable
ones.

Thus, as an example, let a be a merchant, and
b his purse : the combination ab will at once strike
you as a natural and stable one. Indeed there
is something about the word merchant that seems
to suggest inevitably a well-lined purse. Now
'et c be a highwayman, and d his pistol : the
combination cd is again recognised as a natural
and stable one. Now bring the compound ab
into the presence of the compound cd, and mark
how the stability of the former is shaken : ab, a
previously stable compound, becomes at once
curiously unstable ; the merchant and the purse
that we were wont to view as inseparable are
split asunder at once, and the several elements
become rearranged in a manner that finds perfect
expression in the formula :

ab + cd = a + bed.

Again to take another instance which will remind
us of the operation known in chemistry as double
decomposition : Let a be a small boy, b a penny
in his pocket, c an old woman with an apple stall,

40 THE INFLUENCE OF ENVIRONMENT, ETC.

d the apples. The combination ab is a remarkably
unstable one in the presence of cd, and the affinity
between a and d is very great ; an interchange
takes place as follows :

ab -f cd = ad + be.

These examples are no doubt somewhat crude, yet
they will serve to indicate that if we could define
all the conditions we should be able to draw up an
equation for a man's life and show how, given the
man and the conditions acting on him, there was
but one path open to him.

In conclusion let me refer to a statement by
Herbert Spencer in which he discusses the influence
of environment on animals ; starting, however,
from a standpoint that I cannot accept, viz., the
Lamarckian hypothesis of animals responding by
efforts of their own to changes in environment.

The more complete the correspondence between
external and internal changes, the more complete
the power of response to changes of environment,
the higher is the grade of the animal and, as a rule,
the longer is its life and the less its fertility. In
this respect man stands the highest, for instance, in
the power of counterbalancing changes of tempera-
ture, and in the supply of food consequent on
seasons.

Perfect correspondence to change in environ-
ment would be perfect life, eternal existence.
Death, whether from natural decay, disease or
accident, is simply due to failure of the organism to
respond effectively to changes of environment.

Ill

ON EMBRYOLOGY AS AN AID
TO ANATOMY

THERE are perhaps few sciences, or branches of
science, in_wJiich progress has been so rapid and
whose importance has been so suddenly and un-
expectedly revealed as that of embryology. When
we consider the vast number of papers, memoirs,
and treatis£S__that are continually being poured
forth it is^difficult to realise that embryology, as
we now understand it, practically commenced in
1837 with Ihe publication of von Baer's treatise on
the Development of Animals. The Study of
Invertebrate Embryology is still more recent, and in
its modern form is only about a dozen years old.
No one who compares our text-books of biological
science of the present day with those of twelve or
even six years ago can fail to be struck with
the great change effected by embryology in the
methods of teaching this science.

At the outset men hardly knew what to do with
this new branch of science ; for its study the

42 ON EMBRYOLOGY AS AN AID TO ANATOMY

delicate methods used in histological research were
necessary, and hence for a time embryology was
included as a branch of physiology and taught by
physiologists. But anatomists soon saw that this
science belonged by right to them, and that it was
absolutely essential for them to master it. A
zoologist of the present day relies quite as much
on his razor as his scalpel in his endeavour to
comprehend the structure of animals.

It is not difficult to see how it is that em-
bryology has so rapidly attained such importance,
the reason being that it has afforded an altogether
new and very tangible clue to many of the most
interesting problems of biology which have been
thereby first brought within the comprehension and
grasp of man. Thus the zoologist has in em-
bryology obtained a very important clue to the
determination of the affinities of animals ; in fact we
should hesitate to definitely assign a place to some
animals whose development was unknown. To the
philosophical biologist a clue is thus given to the
determination of what, to him, is the problem of
problems — viz., phylogeny, or the genetic connec-
tion between one animal and another and between
their various classes. While to the anatomist em-
bryology offers an explanation of many otherwise
completely unintelligible anatomical facts. To the
members of a medical society anatomy means, and
perhaps I may be permitted to say rather too often
means, simply human anatomy. I therefore
choose for closer consideration a problem, the
main conditions of which are familiar to every one

ON EMBRYOLOGY AS AN AID TO ANATOMY 43

who has dissected a head and neck. The problem
to which I refer is that of the nerve supply of the
muscles by which the various movements of the
eyeball are effected.

Whoever has dissected the orbit cannot fail to
have been struck with the large number of nerves
which not only traverse it but supply parts con-
tained therein ; and also with the additional fact
that these are either distinct nerves or branches of
distinct nerves, and not merely separate branches
of one or two cranial nerves. Thus, of the
twelve cranial nerves no less than six are dis-
tributed wholly or in part to the contents of the
orbit. When we consider that of the remaining
six, two — the olfactory and the auditory — are
nerves of very special function and limited dis-
tribution, and that another one — the spinal
accessory — has but slight claim to be considered
a cranial nerve at all, the fact becomes still more
striking. A further and far more significant fact
than the above is that of these six nerves four
are exclusively distributed to the orbit, and send
branches nowhere else. Of these four nerves
one is the optic nerve, while the other three supply
the muscles by which the movements of the eye
are effected. I must ask your indulgence for a
moment if I refer to such an exceedingly familiar
fact as the arrangement and distribution of these
muscles and their nerves. I only do this because
it is absolutely necessary to the argument I wish
to put before you that this arrangement and
distribution should be fresh in your minds.

44 ON EMBRYOLOGY AS AN AID TO ANATOMY

The movements of the eyeball are effected by
six muscles — four recti and two obliqui. Of these,
speaking roughly, the superior rectus turns the eye
upwards ; the inferior rectus turns it downwards ;
the internal and external turn it inwards and
outwards respectively ; the superior oblique rotates
the eye outwards and downwards, the inferior
oblique outwards and upwards. Of these six
muscles four are supplied by the third cranial nerve,
while each of the remaining two has a separate
and distinct cranial nerve to itself, the fourth
cranial nerve supplying the superior oblique muscle,
and the sixth cranial nerve the external rectus.

Now it is a very remarkable fact that this
small group of muscles should have three cranial
nerves to supply them, and still more remarkable
that these nerves should do little or nothing else.
The third nerve, besides supplying the muscles
mentioned, also supplies the elevator muscle of the
upper eyelid and sends branches which penetrate
the eye and supply the ciliary muscle and iris.
The fourth and sixth nerves, however, do nothing
else, and form unique instances of nerves with
separate origins from nerve centres distributed ex-
clusively to single muscles, and it is very curious
that these two muscles should both be connected
with the eye. In order to complete the conditions
of the problem it is only necessary to add that the
distribution of the nerves and muscles I have
just noticed is not peculiar to man but occurs
throughout the whole vertebrate series almost
without exception. Wherever we meet with a

ON EMBRYOLOGY AS AN AID TO ANATOMY 45

vertebrate having well-developed eyes, there we find
this same arrangement of muscles and nerves.
The exceptions to this rule that have been quoted
are probably only apparent. The few exceptions
that are quoted are worth noticing. In Amphioxus
the eye is a mere spot of pigment and has no
muscles at all; in Myxinoids the eyes are small
and the eye-muscles imperfect, but all three nerves,
though very small, are found to be present and with
their usual distribution ; in Lepidosiren there are
no oblique muscles, and the third, fourth, and sixth
nerves are said to be absent ; the observations on
this point are, however, imperfect. In some of the
lower Amphibia, the newt for instance, all three
nerves are present, at any rate proximally, although
distally they may become closely bound up with
branches of the fifth nerve. With these exceptions,
in all fish, amphibia, reptiles, birds, and mammals
these six muscles are all present and all have their
typical nerve supply. Let me again remind you
that these nerves supply nothing else ; the third
nerve supplies no other parts than those I have
mentioned ; the fourth nerve never supplies any-
thing besides the superior oblique muscle ; the
sixth nerve invariably supplies the external rectus,
but may also supply two other muscles when these
are present — viz., the retractor muscle of the eye,
which is found in some newts, many reptiles and
most mammals, except the monkeys and ourselves,
and the special muscle of the nictitating membrane,
which is met with in birds and reptiles.

Such a state of things so widely spread must

46 ON EMBRYOLOGY AS AN AID TO ANATOMY

have a reason for its existence, but hitherto no
satisfactory or even intelligible answer has been
given to the problem, and anatomy still leaves us
in total ignorance as to why the external rectus
muscle has a special nerve all to itself. Explan-
ations have of course been attempted, for the
problem cannot fail to strike any one who is
acquainted with the anatomy of the parts in
question. A common attempt is to say that the
movements of the eye are so complex and have
to be so extremely nicely adjusted that a complex
nerve supply is necessary. This, I would submit,
is no explanation at all ; it does not tell us in the
least why a small group of six muscles should
have three cranial nerves to supply them. For we
find the same nerve supply in the lowest vertebrates,
in which we cannot suppose these movements to be
very complex. Nor does it suffice to say that the
nerve supply is due to the two eyes having to work
together, and the internal rectus of one eye having
to work habitually with the external rectus of the
other, for in these lowest vertebrates the eyes are
situated at the sides of the head, so that they have
totally distinct fields of vision and hence do not
work together at all. Sir Charles Bell, the illus-
trious anatomist, who was keenly alive to the
interest of the problem before us, attempted an
explanation as follows : Observing that during sleep
and also during certain involuntary acts, such as
sneezing, the eyeball is turned upwards beneath
the upper eyelid, and finding by experiment that
the recti muscles are voluntary, he was led to

ON EMBRYOLOGY AS AN AID TO ANATOMY 47

regard the oblique muscles as involuntary, especially
the superior oblique, whose separate nerve supply
was supposed to be thus accounted for. However,
as Bell himself points out, this theory involves an
assumption which more recent research has failed
to verify. He showed that division of the superior
oblique muscle causes the upward rolling of the
eyeball to be increased, and that therefore if this
movement is due to an impulse transmitted along
the fourth nerve this impulse must be of such a
nature as to cause not contraction but relaxation of
the muscle — i.e.t that stimulation of the fourth nerve
causes relaxation of the superior oblique muscle and
not contraction.

I have referred to the views of Sir Charles
Bell for two very sufficient reasons : firstly, because
this is, so far as I am aware, the only rational
attempt that has been made to grapple with the
problem before us ; secondly, because any account
of such a question without reference to the opinion
of the great anatomist who did so much to render
it possible for us to understand problems connected
with the distribution of the nerves would be a gross
injustice.

So far we have failed utterly to master our prob-
lem ; anatomy and physiology have alike failed to
give us the clue we require, and for my part I see
no reason why, if we had nothing else to help us,
this should not always be so, and the problem be
ranked as insoluble. It is under circumstances
such as these that the anatomist turns to embry-
ology— that sheet-anchor of philosophical anatomy —

48 ON EMBRYOLOGY AS AN AID TO ANATOMY

to which, when in difficulty, he has of late years
turned so often and so rarely in vain ; and I wish
now to ask your attention for a short time while I
endeavour to make clear to you the evidence which
embryology offers on this point. 1 must again ask
your indulgence if I have to refer to matters
many of which are perhaps painfully familiar to
you, while others may appear at first sight to have
nothing whatever to do with the matter in hand.
This evidence applies only at present to Elasmo-
branch fishes, but the conditions of the problem are
absolutely the same as in man.

You all know that there is in the trunk a space
between the body walls and the alimentary canal
called the pleuro-peritoneal cavity or body cavity,
or, still better, the ccelom, this space or cavity
being absent in the head and neck. That is, if
you were to push a sharp instrument through the
walls of the abdomen into the intestine the instru-
ment would not pass through solid tissue along its
whole course but would traverse a cavity — the
ccelom — before reaching the intestine. While if
you performed a similar operation in the neck the
instrument would pass through no cavity but would
penetrate solid tissue all the way until the ceso-
phagus was reached. Now the whole body is formed
from three cellular layers or strata — the epiblast,
mesoblast, and hypoblast — and if we make a
diagrammatic section through the body the epiblast
will form the outermost layer or skin ; the hypoblast
will form the lining of the alimentary canal ; all the
rest will be mesoblast. The body cavity or ccelom

ON EMBRYOLOGY AS AN AID TO ANATOMY 49

is entirely within the substance of the mesoblast.
In the earliest stage of development the mesoblast
is solid and there is no coelom. In the trunk the
mesoblast very early splits into two layers which
become separated from each other and so give rise
to the coelom. It was formerly assumed to be a
sharp distinction between the trunk on the one
hand and the head and neck on the other, that in
the former the mesoblast splits in this manner and
that in the latter it does not do so. This distinction
is now known not to hold good, for the splitting
of the mesoblast has been shown to extend to the
head, but the cavities on each side do not meet in
the median ventral line. Again, in the trunk the
coelomic cavity becomes divided on either side into
an upper or dorsal part, and a lower or ventral
part. Each of the dorsal portions becomes cut up
transversely into a number of segments arranged
one behind the other in series, one such segment
occurring in each primary body segment or
protovertebra. In the head the coelomic cavity is
first cut up transversely by the visceral clefts into
a series of cavities, one in each visceral arch, and
then each cavity divides again into dorsal and
ventral portions. The ultimate result is the same as
in the trunk, but the order of division is different.

Of these divisions of the coelom in the head, or
head cavities as they are called, we are only
concerned with the three most anterior. The
first head cavity — the premandibular — is in front
of the mandible and immediately behind the eye ;
the second — the mandibular — is situated in the

D

50 ON EMBRYOLOGY AS AN AID TO ANATOMY

mandibular arch ; the third — the hyoidean — is in
the hyoid arch. Now in the trunk the walls of
both the dorsal and ventral divisions of the coelom
become converted into muscles, and the cavities of
the dorsal division become obliterated owing to the
great increase in the thickness of their walls. In
the head cavities the walls of both dorsal and
ventral divisions become converted into muscles in
the same way, and the only difference is that both
dorsal and ventral cavities become obliterated
instead of the dorsal only. If we bear in mind
that the transverse divisions of the dorsal
portion of the coelom into segments is obviously
part of the general segmentation of the body,
and that the similar division of the head cavity
is manifestly of the same nature, we see that we
may speak of these first three head cavities as
indicating three distinct and successive segments
of the head comparable to three distinct and
successive body segments.

We have now accomplished by far the most
difficult and tedious portion of our task, but let me
direct your attention to another set of organs. The
central nervous system consists at an early period
in all Vertebrates of a tube closed at both ends
stretching along the back of the animal from head
to tail. This tube is not of uniform calibre, but its
anterior part — the future brain — is from the first
wider than the hinder part — the future spinal cord.
The anterior part is bent in the shape of a hook,
and presents along its whole length a series of
alternate dilatations and constrictions which are

ON EMBRYOLOGY AS AN AID TO ANATOMY 51

much larger and more conspicuous in the brain than
elsewhere, but occur along the whole length of the
cord. The first dilatation is the fore brain, and
from it the cerebral hemispheres project ; the
second is the mid brain ; this is followed by a series
of vesicles, rapidly decreasing in size, and called
collectively the hind brain ; behind this we have the
spinal cord. In the spinal cord the dilatations,
though only feebly marked, manifestly correspond
to the segments, and from each one a pair of spinal
nerves arises. Similarly in the brain from each
well-marked vesicle a pair of nerves arises. Thus,
from the mid brain the third pair of nerves arises
and runs backwards to the interval between the
first and second head cavities. From the first
vesicle of the hind brain arises the fifth nerve, and
this passes down between the second and third
head cavities. From the second vesicle of the hind
brain the seventh or facial nerve springs and runs
down behind the third head cavity. The division
of the head into segments is thus very clearly and
satisfactorily shown, for all the different elements
that could afford evidence — viz., brain vesicles,
nerves, head cavities, and visceral clefts and arches —
all point to the same divisions and agree among
themselves.

We have now got all the links of our evidence
nearly complete ; let us fit them together and give
them the finishing strokes. We have seen that the
walls of the head cavities like their homologues in
the body develop into muscles. Now the first head
cavity lies immediately behind the eye and extends

52 ON EMBRYOLOGY AS AN AID TO ANATOMY

round it so as to invest it like a cup. From
its walls are developed the rectus superior, rectus
inferior, rectus internus and obliquus inferior
muscles — i.e.} all the muscles supplied by the third
nerve. At last we see why the third nerve supplies
these and no other muscles ; it does so because it
is the nerve belonging to the segment in which the
first head cavity lies; and therefore supplies the
muscles that are formed out of the walls of that
cavity. This reason is a complete and sufficient one.

Concerning the obliquus superior we are still in
the dark ; it does not appear to have any connection
with the first head cavity, and this is sufficient
reason why it should not be supplied by the third
nerve. The development of the nerve that does
supply it — the fourth — is at present absolutely
unknown. Concerning the rectus externus muscle
we are in a better position. The sixth nerve which
supplies it bears the same relation to the seventh
nerve that the anterior root of a spinal nerve does to
the posterior root. The rectus externus has no
connection with the first head cavity, but lies
altogether superficially to it, and for some time
behind it ; it appears to be developed partly from
the third head cavity and possibly in part from the
second. We see now clearly why it is not supplied
by the third nerve, the reason being that it does
not belong to the segment which is supplied by that
nerve, but to one further back.

Thus, we have not quite a complete, but quite a
novel and, I think, an intelligible clue to the solution
of our problem.

IV

THE THEORY OF CHANGE OF
FUNCTION

WHEN asked to read the opening paper of the
second session of the Owens College Biological
Society, I felt that the subject most suitable for the
occasion would be one gathered from the works of
that great Englishman whose name it was at one
time proposed that this Society should bear ; that
prince of biologists who has taught us not to be
satisfied with a knowledge of facts but ' to seek
earnestly for the reason of these facts ; who has
taught us both what to seek and how to seek, and
who has thus rendered philosophical biology not
only a possibility but an actuality.

In selecting the particular aspect of Mr. Darwin's
wonderful theory, to which I should draw your
attention, I have been influenced mainly by two cir-
cumstances. In the first place it appeared more
profitable, and therefore more likely to prove accept-
able to the Society, to select some portion to which
exception has been taken rather than one which has

54 THE THEORY OF CHANGE OF FUNCTION

met with general acceptance ; to consider one of the
numerous difficulties in the way of accepting the
theory of Natural Selection, difficulties pointed out
by none so clearly as Mr. Darwin himself.

The second circumstance to which I have
alluded as influencing my choice was the accidental
fact that some work on which I happened to
be engaged some little time ago brought me into
very violent, and for the time very embarrasing
contact with one of these said difficulties, and thus
compelled me to take its bearings and measurements
very accurately in order to discover in what way it
might most conveniently be circumvented or sur-
mounted, or, if possible, removed altogether.

I purpose then asking your attention to one of
the more serious of the many alleged difficulties in
accepting the doctrine of Natural Selection and to
a brief consideration of the means proposed for
removing that difficulty. Let me first attempt to
define clearly the nature of this obstacle. In the
higher animals we meet with a great number of
very complex organs, each with very definite func-
tions, such as the eye, the ear, and the hand. Now,
according to the Darwinian theory, the mode in
which such organs have attained their present com-
plexity and perfection is as follows :

Once upon a time the ancestors of these animals
possessed eyes less perfect and less complex than
those they have at present. Now no two animals
of the same species have eyes absolutely identical
with one another ; slight differences always occur,
and of these slight differences it must happen that

THE THEORY OF CHANGE OF FUNCTION 55

certain ones are improvements, are changes for the
better, or what we call useful modifications. All
the animals possessing these useful modifications
will be slightly better off and will have a slight
advantage over their companions who have not got
them. Now more animals of every species are born
than can possibly live. Hence the greater propor-
tion of animals that are born die, and die young
before they have produced any offspring. It is
clear that those animals which have this slight
advantage over their brethren will by virtue of this
fact have a better chance of living. Now we know
as a fact that such variations tend to be inherited —
i.e., to be handed down from the parents in whom
they first appeared to their offspring, and not only
so but also that they tend to appear in the offspring
in a rather more strongly developed form.

Hence we get this state of affairs ; certain forms
of a particular species, say of deer, tend to survive
because they have some slight accidental modifica-
tion of their eyes which gives them a slight but
distinct advantage over their brethren ; they trans-
mit this modification and with it this advantage to
their offspring ; certain of their offspring present the
modification in a greater degree not only than their
brethren but also than their parents. These will
get a slight additional advantage and will tend to
survive, and so on for successive generations. The
gradual accumulation of these slight modifications
will in time cause a perceptible change in the
structure of the eye, and as each successive modifi-
cation is preserved only because it is useful it is

S6 THE THEORY OF CHANGE OF FUNCTION

clear that the eye will as a whole have improved
and have become more perfect. Similarly with the
hand or with any other of the organs of the body ;
formerly less complex, it has acquired its present
complexity and perfection as the result of accumu-
lations of a long series of small but useful modifica-
tions. Each of these modifications must give some
distinct advantage to its possessor or else it would
not be preserved or perpetuated.

It has been objected that this explanation is not
sufficient ; that although it will account for the
gradual perfection and increasing complexity of an
•organ after it has attained a certain size and after
it has assumed its definite function, yet that it fails
completely to explain the first origin of such
organs ; for in their earliest origin they must have
been very minute and absolutely incapable of
fulfilling or even aiding in the function which they
are afterwards destined to perform.

This objection, which is the one I wish to
consider to-night, must be confessed at once to be
of very great weight, and is indeed freely acknow-
ledged to be so by Darwin himself. By some
writers, such as Mivart (almost the only English
naturalist of any repute who at the present day
rejects the doctrine of Natural Selection), it is
indeed considered as absolutely fatal to the whole
theory. We shall perhaps get a clearer idea of
the nature and force of the objection by consider-
ing one or two examples which bring it forward
prominently.

The limb of a Vertebrate in its earliest stage is a

THE THEORY OF CHANGE OF FUNCTION 57

small bud arising from the surface of the body and
•is absolutely useless for any of the purposes to
which we put our arms. If this represents the
primitive condition, why should such an organ
have been preserved at all ? Again, a more striking
example and one yet not thoroughly explained is
found in the wing of the bat. This is a great fold
of skin connecting together the much elongated
fingers, and also connecting these with the sides of
the body extending downwards so as to involve
the legs as well. This constitutes as we know
a very efficient flying apparatus. The theory
of evolution undoubtedly requires that bats
should be descended from, should have had as
ancestors, mammals possessing fingers of normal
length and devoid of the lateral expansion of the
skin. The theory of Natural Selection demands
further that the various steps in the transformations
of this ancestral form to the existing bat should
have been slow and gradual. Now it is obvious
that a very slight lengthening of the fingers, coupled
with a very slight increase in the extent to which
they were webbed, would in no way enable the
ancestral mammal to fly ; and hence that such
accidental variations could never have been pre-
served and perpetuated because of their utility as
flying organs.

Cases such as these have long been felt to offer
very serious difficulty and to throw serious doubt,
not on the reality of Natural Selection, but on its
sufficiency. Darwin himself suggested a possible
solution of the difficulty, at any rate in certain cases,

58 THE THEORY OF CHANGE OF FUNCTION

by pointing out that an organ which had already
attained considerable size and complexity might,
owing to some change in the habits or surroundings
of the animal possessing it, have not its structure
but its function changed — i.e., that an organ already
in use for one purpose might become employed for
some other purpose, might gradually lose its original
function and undergo further modifications fitting it
better for its adopted function.

This idea, which clearly affords us a possible
mode of getting over the difficulty, is only briefly
alluded to by Darwin, but was taken up and
developed by Dr. Dohrn who, in a pamphlet
published in 1875, discussed it at considerable
length.

A very good example of the principle in question
is afforded by the swimming bladder of fishes.
This, in most fish, is a closed sac lying just
underneath the vertebral column. In many fish
it acquires a connection by a duct with some part
of the alimentary canal. It then becomes an
accessory breathing organ, especially in those fish
which are capable of living out of water for a time —
e.g.} the Protopterus of Africa. An interesting
series of modifications exists connecting the air
bladder with the lung of the higher Vertebrates,
which is undoubtedly the same organ. The air
bladder of fishes is in fact a very good example of
change of function, inasmuch as it is used originally
for purposes of flotation and afterwards it is pre-
served as a lung. This example is instructive also
because the evidence of embryology, on which we

THE THEORY OF CHANGE OF FUNCTION 59

are accustomed to rely, fails utterly here. The lung
develops as a pit-like depression in the floor of the
oesophagus. Now this could not have been the
earliest origin of the lung, for it would be utterly
useless as such for the simple reason that food
would always be falling into it.

In developing the theory of Change of Function
Dr. Dohrn points out that it is a very common
thing, if not the general rule, for an organ to have
not one function alone but a number of different
functions to perform, of which functions one at any
given time is predominant and the rest subordinate ;
but that it is quite conceivable that some slight
change of circumstance might cause the relative
importance to be changed, and one of the subor-
dinate functions to become the primary one, or to
put it in his own words : " Each function is a
resultant of several components of which one is
the principal or primary function, the other
secondary. Diminution of the primary function
and increase of a secondary function alters the
total function ; the secondary gradually becomes
the primary, the total function is changed and the
issue of the whole process is the transformation of
the organ."

For instance, the primary function of the stomach
is undoubtedly the secretion of gastric juice ; a
secondary function is the movement of its muscular
walls aiding the action of the gastric juice by
bringing the contents into closer contact with
it. Now in no animal are the glands abso-
lutely uniformly dispersed over the stomach walls,

60 THE THEORY OF CHANGE OF FUNCTION

and it is readily conceivable that both glands
and muscular wall might be better developed in
one half of the stomach than the other, and indeed
this condition actually occurs in the stomach of the
rat. A continuance of this modification brings us
to the condition met with in the ruminant stomach.
Here the first compartment is a mere receptacle
for the storing of food, and in this no digestion
takes place at all. Again, from the same starting-
point, we have another series. Suppose the
glands to aggregate at one end of the stomach, and
the muscular coat to be thickened at the other ; if we
carry this far enough we get to the condition of the
stomach of the bird where peptic glands are con-
fined to one part, muscles to the other. Here the
second portion has completely lost its original
primary function, this having become replaced by
a secondary function which has now become
primary.

The modifications of the limbs of Arthropods
afford numerous and admirable illustrations of the
case before us. Here we have organs whose
primary function is undoubtedly locomotion, but
of which a certain number, greater or less, in
different groups, have become modified so as to
aid in the mastication of food. Naturally, those
nearest the mouth are the ones so modified, and
of these, those at the actual sides of the mouth
are most likely to undergo the greatest modifica-
tions. Such modifications, if extreme, would unfit
them for their primary locomotor function, hence
we find the first post oral pair of appendages,

THE THEORY OF CHANGE OF FUNCTION 61

the mandibles, in most Arthropods completely
and permanently altered, both structurally and
functionally, the hinder ones being altered to less
extent.

Of the two divisions of the limb the inner or
endopodite is the nearer to the mouth, and conse-
quently most likely to be useful for purposes of
feeding. On the other hand, for swimming, the
exopodite is of equal, if not greater, importance ; for
walking the endopodite is of most use as being more
directly beneath the body. Another good example
of change of function is found in the hyomandibular
cleft. This, like certain other clefts, has become
saved from destruction by becoming modified into
an accessory organ of hearing.

The olfactory organ furnishes another illustration
of the theory, and with regard to this let us first
consider what appear to be the difficulties of the
case. All Vertebrates possess olfactory organs with
the solitary exception of Amphioxus, that curiously
exceptional Vertebrate whose anatomy seems to be
made up of contradictions. These olfactory organs
have in all cases the same essential structure.
Certain differences occur, but these are so slight as
to leave no doubt whatever that the vertebrate
olfactory organs are, wherever they occur, the
same organs. As soon, however, as we get beyond
Vertebrates we meet with a difficulty. There is no
doubt that Vertebrates came into existence later
than Invertebrates, therefore Vertebrates must either
have inherited their olfactory organs from their
invertebrate ancestors, or must have acquired them

62 THE THEORY OF CHANGE OF FUNCTION

for themselves. An olfactory organ is found in
Invertebrates ; some insects almost certainly have
them, but these could not have been the ancestors
of the vertebrate olfactory organ. Now if the
Vertebrates did not inherit them as such they must
have got them some other way, either acquired de
novo or by change of function. Now in its time of
appearance, its mode of development, the histology
of its epithelium, there is a very close resem-
blance between the olfactory organ and a gill.
Hence there are strong reasons for regarding the
olfactory organ of Vertebrates as a modified gill.
It has been stated that the mode of development of
the olfactory nerve as an outgrowth from the
cerebral hemisphere constitutes a serious objection to
this view as differing from the other cranial nerves.
This objection does not now hold good, since it has
been shown that the olfactory nerve develops in the
same way as other cranial nerves.

In conclusion, we seem to have in this principle
of change of function a real and practicable solution
to the chief difficulty in the way of accepting the
doctrine of Natural Selection, a solution that has
long been overlooked, and whose real importance
has, I believe, yet to be appreciated.

BUTTERFLIES

THE beauty of butterflies is partly due to their
shape and proportions and to their very graceful
outline, but it is in their colouring that their chief
beauty lies ; and concerning this colouring I wish
to say a few words, even at the risk of repeating
what is already familiar.

The colour of a butterfly is due to the scales
which cover both surfaces of the wings and which
are easily rubbed off by our fingers. These scales
are of various shapes and overlap each other like
the tiles on a roof, each having a short stalk for
insertion into a small depression in the wing,
Their variety of colour is extraordinarily great and
is due partly to actual pigments ; the brightest tints
however, and more especially the metallic lines
seen in some foreign butterflies, and perhaps best
of all in humming-birds, are due to fine lines on the
surface of the wings producing what are known to
physicists as interference colours.

Let us consider the question. Why should

64 BUTTERFLIES

butterflies be thus gaily clothed ? The older
naturalists thought it sufficient to say that butter-
flies were beautiful because their beauty gave us
pleasure, and that it was for our enjoyment that
these delicate tints and gorgeous hues existed.
But this is clearly incorrect, for the most magnifi-
cent butterflies are found in parts where man but
seldom visits, for instance, tropical America and the
East Indies. The true explanations are varied
and not the same in all cases.

Sexual Colours. — With butterflies as with birds
there is often a striking difference between the
sexes, the male as a rule being the more gaily
coloured. Among birds, for instance, the cock bird
is often most gaily coloured ; the hen bird being more
soberly coloured, and indeed often plain. The
peacock, drake, and bird of paradise will at once
occur to our minds as illustrations. Mr. Wallace
pointed out that the sober colouring of the female
was explained as protective so as to escape detection
when sitting on eggs on an open nest. The male
usually taking no share in incubation had no
special need for protective colouring. Mr. Darwin
explains the specially brilliant colours of males as
due to what he calls sexual selection, that is, to the
fact of the female preferring a smart male to an
untidy one. So with butterflies and moths ; and if
we watch butterflies or birds coquetting we soon
convince ourselves that the males know they are
beautiful, and mean to show off their beauty to its
best advantage in the hopes of gainii|g the affection
of the female.

BUTTERFLIES 65

It has been objected that this is going too far ;
that the proper appreciation of the colours and tints
of butterflies, birds, or flowers requires considerable
training and a distinct intellectual effort on our
part, and that it is absurd to suppose that what we
ourselves only do imperfectly and with difficulty a
butterfly can do just as well and without effort.
But I do not think we ought to reject an explanation
just because it happens not to be particularly
flattering to ourselves. In the case of flowers it
has been conclusively shown that their colours, mark-
ings and odours are developed to attract insects —
bees, beetles, and butterflies — to visit them and so
fertilise them ; in other words, to advertise them-
selves. The colours of flowers are as beautiful and
as varied as those of butterflies. And if, as it is
acknowledged, butterflies and birds understand and
appreciate the differences between the colours of
gay flowers, and if those colours are developed
simply to attract them, why deny them the power
of appreciating similar colours in themselves and
in one another ? Again, the appreciation of colour
in ourselves has greatly developed ; for if we study
the history of painting and compare the oldest
pictures, whether oil or water colour, with more
recent ones we note a distinct development in
the power of appreciating the full effect of colour,
or what is known as the colour sense.

Protective Colours. — So far we have considered
the colours of the upper surface only. With regard
to the colours of the under surface the case is
different; here the colours are protective, and butter-

E

66 BUTTERFLIES

flies have the habit of turning up their wings when
resting, so as to expose the under surface only.
The under surface is coloured like the leaves, twigs,
and especially the flowers on which they most
love to perch, for it is then that they are most
exposed to the attacks of enemies and have the
most need for protection. To appreciate this it is
necessary to see them in their haunts, and to watch
them at home. The best marked instance is Kal-
lima, which is met with in India, the Malay Archi-
pelago and Sumatra. It is very common in these
places, and is a large showy butterfly with orange
and purple colouring on the upper surface, and is
a rapid flier frequenting dry forests. It always
settles where there is some dead and decaying
foliage, for the colouring on the under surface of
the wings bears a remarkable resemblance to that
of a dead leaf, and when the wings are turned up
and the head and body hidden between them, it is
often very difficult to distinguish them from dead
leaves, the resemblance being rendered even more
close by the short tail which looks like the stalk
of a leaf, and by the markings on the under surface
which closely imitate the mid-rib and veins of a leaf.
Speaking of this insect Mr. Wallace says : " The
colour is very remarkable for its extreme amount
of variability, from deep reddish-brown to olive or
pale yellow, hardly two specimens being exactly
alike, but all coming within the range of leaves in
various stages of decay. Still more curious is the
fact that the paler wings, which imitate leaves most
decayed, are usually covered with small black dots,

BUTTERFLIES 67

often gathered into circular groups, and so exactly
resembling the minute fungi on decaying leaves
that it is hard at first to believe that the insects
themselves are not attacked by some such fungus.
The concealment produced by this wonderful imita-
tion is most complete, and in Sumatra I have often
seen one enter a bush and then disappear like magic.
Once I was so fortunate as to see the exact spot on
which the insect settled, but even then I lost sight
of it for some time, and only after a persistent
search discovered that it was close before my
eyes."

This example serves as an extreme case of what
is really a general law among butterflies.

The mode in which this protective colouring is
acquired is explained by Natural Selection. At first
there is a more or less accidental resemblance.
Now large numbers of butterflies are killed by birds,
lizards, and other animals, and any whose mark-
ings and habits of perching rendered them less
easy to see would have a better chance of escaping
their enemies ; these, therefore, will survive and
transmit their peculiarities to their offspring ; the
survivors of these in turn transmitting the peculi-
arities in a more strongly marked form, and so the
protective colouring becomes more marked from
generation to generation and the unprotected ones
perish.

Warning Colours. — These are found in specially
protected individuals which are usually uneatable,
or at any rate unpalatable, and whose nastiness it
is desirable to advertise in order that they may

68 BUTTERFLIES

not be eaten, or at any rate killed, by mistake.
Their object, therefore, is not to escape notice, but
to be readily seen and recognised. The best
examples of these are found in three great
families of butterflies — the Heliconidae, found in
South America, the Danaidae, found in Asia and
tropical regions generally, and the Acraeidae of
Africa. These have large but rather weak wings
and fly slowly. They are always very abundant,
and all have conspicuous colours or markings and
often a peculiar form of flight ; characters by
which they can be recognised at a glance. The
colours are nearly always the same on both upper
and under surfaces of the wings, and they never try
to conceal themselves, but rest on the upper
surfaces of leaves and flowers. Moreover, they all
have juices which exhale a powerful scent ; so that
if they are killed by pinching the body a liquid
exudes which stains the fingers yellow, and leaves
an odour which can only be removed by repeated
washing. This odour is not very offensive to man,
but has been shown by experiment to be so to
birds and other insect-eating animals.

Another example is furnished by the skunk,
which, although not included in the immediate
subject of this lecture, may be mentioned as an
extreme case illustrating the point we are con-
sidering. Concerning this animal I cannot do
better than quote Mr. Wallace again: "This
animal possesses, as is well known, a most offensive
secretion which it has the power of ejecting over
its enemies, and which effectually protects it from

BUTTERFLIES 69

attack, The odour of this substance is so penetra-
ting that it taints and renders useless everything it
touches or in its vicinity. Provisions near it
become uneatable, and clothes saturated with it will
retain the smell for several weeks even though
repeatedly washed and dried. A drop of the liquid
in the eyes will cause blindness, and Indians
are said sometimes to lose their sight from this
cause. Owing to this remarkable power of offence
the skunk is rarely attacked by other animals, and
its black and white fur and the bushy white tail
carried erect when disturbed, form the danger
signals by which it is easily distinguished in the
twilight or moonlight from unprotected animals.
Its consciousness that it needs only to be seen to
be avoided gives it that slowness of motion and
fearlessness of aspect which are, as we shall see,
characteristic of most creatures so protected."

Recognition Colours. — A good instance of this
class of colouring is seen in the upturned white tail
of the rabbit which, although making it conspicuous
to its enemies as well as friends, is probably a
signal of danger to other rabbits ; and when feed-
ing together, in accordance with their social habits,
soon after sunset or on moonlight nights, the
upturned tails of those in front serve as guides to
those behind to run home on the appearance of an
enemy. Many birds, antelopes, and other animals
have markings believed to serve a similar purpose,
and probably the principle of distinctive colouring
for recognition has something to do with the great
diversity of colour met with in butterflies.

70 BUTTERFLIES

Mimicry. — Many butterflies escape destruction
through mimicking the appearance of uneatable or
venomous forms ; for instance, Leptalis, a form allied
to the common garden white, mimics the Heliconidce,
a widely distinct family, in the shape of its body
and wings, in its colour, and even in its habits and
mode of flight, so much so that they are difficult to
distinguish. They are much rarer than the forms
they mimic, and may aptly be compared to the Ass
in the Lion's Skin. It is often only the female
which mimics, for this has greater need for protec-
tion. The deception is indeed often so good that
it may deceive not merely an expert naturalist but
even the insects themselves, and Fritz Muller says :
" I have repeatedly seen the male pursuing the
mimicked species till after closely approaching and
becoming aware of his error he suddenly returned."
Other examples of mimicry are found in the
beehawk moth, which mimics the humble bee ; in
the clearwing moths, which receive names such
as apiforme, vespiforme, from the insects they
mimic ; and, among other animals, the harmless
snakes mimic the venomous ones. Again, the
" devil's coach horse," the beetle with the habit of
turning its tail over its back and pretending to have
a sting, certainly deceives children and perhaps
grown-up people as well. Many other instances of
mimicry could be cited ; a certain spider simulates
the droppings of birds for instance, and a blue
butterfly was actually seen by Wallace resting on
what was apparently dung but in reality a spider.
Nor is the advantage of the power of mimicry

BUTTERFLIES 71

confined to animals, for among ourselves the power
of pretending to be other than what we really art
is often of the greatest possible service to us. For
example, the success of a detective largely
depends on his being able to pass himself off as
something else and to entirely conceal his real
calling.

There is an old view which holds that colours
are due to the direct action of the sun, but this
will not explain the constancy of colouring in most
species, or the bright colours of butterflies in
climates such as our own. There are some
cases, however, in which differences in temperature,
at any rate in the season of the year, appear
to have had a distinct influence on colour, the
best known examples being the common continental
butterflies Vanessa levana and Vanessa prorsa and
the English butterfly Pieris napi*

Colour is perhaps in part a matter of indifference,
and a part may be red because there is no reason
why it should be otherwise. The red colour of
blood, the red colour of many deep sea animals, the
blue colour of the sky, and perhaps the green colour
of the grass and vegetation generally, may, for all we
know, be indifferent.

The Senses of Butterflies. — Our knowledge of this
very interesting subject is at present very imperfect.
We know that the sense of smell must, at any rate
in some forms which have a liking for dead and
decaying animal matter, be very acute. The eyes

* For a full account of these see paper on "Environment,"
p. 34.— ED.

72 BUTTERFLIES

are compound and their facets extraordinarily
numerous, as many as 12,000 being present in the
eye of the death's-head moth, and 17,000 in the
swallow-tailed butterfly. The range of vision in
butterflies is unknown. Of the sense of hearing
in butterflies we know very little, but in moths and
other insects the antennae are often fringed ; and
some most interesting experiments have been made
with the antennae of the male mosquito, special hairs
of which were found to vibrate to particular notes,
those hairs being most affected which were at right
angles to the direction from which the sound came.
The sound is thus most intense if directly in front
of the head ; if one antenna is affected more than
the other the head is turned till both are equally
affected, and so the mosquito is enabled to direct its
flight directly towards the point from which the
sound originates ; this power being specially used
to aid it in finding the female.

The Life-History of Butterflies. — A butterfly is
not always a butterfly, but goes through several
phases — viz., the egg, the larva or caterpillar, the
pupa or chrysalis, and the adult form or imago. The
caterpillar phase is the nutritive one ; the adult
phase is reproductive and usually short-lived, often
not more than a few days, and sometimes limited
to a few hours. The eggs are usually protec-
tively coloured, and often show very beautiful
markings under the microscope. It has been shown
that while at least nine-tenths of the eggs and
larvae of North American butterflies are destroyed
by parasites or disease, such parasites never attack

BUTTERFLIES 73

the eggs of the Danaidae, and it is possible that
these are distasteful to their foes at all stages of
their existence. The eggs of butterflies are always
laid in safe places and as near as possible to the
proper food for the young ; the butterfly never sees
her young and can have no clear idea as to the
respective merits of cabbages, carrots, and oak-
leaves, yet she makes no mistake.

The larva, or caterpillar, is soft and fleshy and is
very easily injured. It is furnished with a biting
mouth, soft fleshy fore-legs, and a terminal sucker.
It has three pairs of harder jointed legs correspond-
ing to the legs of a butterfly and which are
sometimes very long, as in the lobster moth. The
great and only work of caterpillars is to eat, and
this they do all night, many of them all day as well.
They feed on the leaves of plants by means of their
powerful jaws, and have no need even to stop to
take breath, for their breathing is carried on by
means of spiracles or pores along the sides of the
body which lead to the tracheal tubes by which the
respiration of insects is effected. Their rapid
growth soon renders their skin too tight and the
outer layer, or cuticle, is thrown off like that of a
crab or lobster ; after this they eat more ravenously
than ever to make up for lost time, often commenc-
ing with their cast-off clothes. As a rule there are
several of these castings before the caterpillar
attains its full size. Their gain in weight is
prodigious, the caterpillar of one of the hawk moths
for instance, Acherontia, weighing about one-eighth
of a grain on leaving the egg, in thirty days

74 BUTTERFLIES

increases its weight 10,000 times. Even this is not
the limit, for some caterpillars live three years ; for
instance, the goat moth, which grows to 72,000
times its weight on hatching.

The shapes and colours of caterpillars follow
the same laws as those concerning the adult
insect, the colour being generally protective be-
cause the caterpillar is soft and fleshy and good
eating for birds. The colour for this reason is
usually green, and those feeding on grass are
striped longitudinally, those on larger leaves
obliquely, this forming a very effective protection.
Others are brown, and so like twigs in shape that
even the most experienced may be deceived. Mr.
Jenner Weir writes: " After being thirty years
an entomologist I was deceived myself. I took
out my pruning knife to cut from a plum tree a
spur which I thought I had overlooked. This
turned out to be a larva of a geometer two inches
long. I showed it to several members of my
family and defined a space of four inches in which
it was to be seen, but none of them could per-
ceive that it was a caterpillar."

Warning colours are well seen in caterpillars.
All green and brown ones are readily and greedily
eaten by birds, lizards, and frogs, but many are
conspicuously coloured and do not attempt to
conceal themselves, and these are usually nauseous
to the taste. For instance, the gooseberry moth
caterpillar was given to frogs to eat, whereupon
they " sprang forward and licked them eagerly into
their mouths ; no sooner had they done so than

BUTTERFLIES 75

they seemed to become aware of the mistake they
had made, and sat with gaping mouths rolling
their tongues about until they had got quit of the
nauseous caterpillars, which seemed perfectly un-
injured and walked off as briskly as ever."

Terrifying colours are also met with ; in the
caterpillars of the puss moth and hawk moth the
eye spots are of this nature, and the caterpillar of
Bombyx regia, the " hickory horned devil " of the
Southern States of North America, is ornamented
with an immense crown of orange-red tubercles
which if disturbed it erects and shakes from side
to side in a very alarming manner, the negroes
believing it to be as deadly as the rattlesnake,
whereas it is really perfectly harmless.

The pupa, or chrysalis, is the last stage of
existence of the caterpillar. The change from
caterpillar to butterfly is an enormous one, not
merely as regards the possession of wings, but
as regards the whole organisation ; the mouth,
the digestive system, the nerves, the eye and all
parts being most profoundly modified. The chry-
salis period is the stage of rest, and constitutes
a protected condition in which these changes can
be effected. This stage sometimes lasts a few
weeks or even days only, sometimes months or
during a whole winter. In the case of the small
eggar moth, insects of the same brood appear a
few at a time each year up to fourteen or fifteen
years. Now the time of appearance of the moth
being February, it is obvious that in a severe
winter the whole brood might be killed off if they

76 BUTTERFLIES

all appeared at the same time, hence the advantage
of appearing a few at a time in successive years.

Pupae are of various forms : (i) Underground
pupae, which are found in crevices in walls and
the roots of trees, the holes in which they lie
being often carefully lined. These pupae have
some power of motion, and work their way to
the surface before the imago emerges from the
pupa. (2) Suspended pupce. A good example of
these is found in the pupa of the common tor-
toiseshell butterfly. The larva having chosen a
suitable place, by means of the glands which open
on the under lip, spins a little button of silk strong
enough to support its weight ; it then thrusts its
tail into the button and swings head downwards.
It has now to get out of its skin without letting go
its hold, and this it does by gradually working out
of its skin, which has previously been split along
the back, and pushing it towards its tail. Before
completely escaping, it gets the tip of its tail free
while the part in front of the tail is still fixed, the
lining of the trachea and intestine helping to
suspend it ; it then stretches its tail up to the
button of silk, fixes itself and spins round several
times to secure a firm hold, finally casting away
its larval skin. Other larvae sling themselves up
by a girth of silk, and some are flexible enough
to attach the thread on one side, carry it over and
fix it on the other side, repeating this process
several times. Others again spin the girth of
silk first and then slip their heads under it ; some,
such as the swallow-tail, hold the silk in their

BUTTERFLIES 77

claws till it is strong enough, and then slip it over
their heads.

Many other interesting points could be men-
tioned in connection with butterflies, but those I
have briefly touched upon will serve to indicate
what interesting questions arise from the study
of butterflies and other insects, and what good
examples they afford us of many of the problems
of Natural History.

VI

FRESH-WATER ANIMALS

I PROPOSE to speak this evening about the fauna
of fresh-water ponds and streams, a subject of very
special interest to the members of the Society,
many of whom have a far more intimate and accu-
rate acquaintance with special groups than I can
either claim or reasonably hope to possess. Still,
there are some points of general interest which
may have escaped the observation of those who
have chiefly concerned themselves with one or two
particular groups ; and it is with regard to these
more general conclusions that I propose to speak
to-night.

As regards geographical distribution the entire
animal kingdom may conveniently be divided into
terrestrial and aquatic forms, and the latter again
subdivided into those that live in the sea, and
those inhabiting fresh water. The marine fauna
is infinitely more abundant, and includes a far
greater number of species than that found in fresh
water ; and we can hardly be surprised at this. A

FRESH-WATER ANIMALS 79

glance at a map of the world will show that the
area covered by the ocean is much greater than
that occupied by land — the proportion being about
three to one. Moreover, only a very small por-
tion of the land is taken up by rivers and ponds ;
so that the area over which fresh-water forms can
exist is necessarily much more restricted than that
inhabited by marine animals.

Again, geologists show us that the land is sub-
ject to constant change in level and in extent,
owing to upheaval or depression ; and these oscil-
lations, especially those of depression, must cause
great disturbance, or even local annihilation, of the
fresh-water fauna. By upheaval on the other
hand, shallow water marine areas may be cut off
from the sea, and converted into brackish marshes,
and ultimately into fresh-water ponds ; the animals
inhabiting them being either killed off, or else
adapting themselves to the altered environment, and
becoming converted into fresh-water forms.

From these and other similar considerations it
may be concluded that fresh-water animals must,
with few exceptions, have been derived from marine
forms, and my chief object this evening is to con-
sider the various modifications, either in structure or
life-texture, which marine animals undergo in
consequence of becoming adapted to fresh-water
life.

Why should marine animals strive to work their
way up rivers ? Why should they not stop in the
sea ? The reason is not far to seek, and is to be
found in that struggle for existence we hear so

8o FRESH-WATER ANIMALS

much about, but the full extent of which we too
often fail to realise. Very large numbers of every
species of animal perish before reaching maturity.
If we reflect for a moment on the really enormous
number of eggs which fish and other marine animals
lay, it is clear that if all these eggs, in the case
of any single species, were to come to maturity,
there would very soon be little room in the sea for
anything else. A mackerel of a pound weight, for
instance, will lay eighty thousand or ninety thousand
eggs, a cod may produce five millions, and a conger-
eel no less than fifteen millions. However, the vast
majority in every species never come to maturity,
but are devoured while young by other animals, as
food ; while a very small minority, favoured by the
possession of some slight accidental advantage,
escape destruction, survive, and perpetuate the
species. So keen is the struggle for existence in
the sea, especially in the shallow waters round the
coasts, where life is far more abundant and
competition more severe than at greater depths,
that many forms, belonging to different groups,
have, to escape it, worked their way up the rivers,
and adapted themselves to fresh-water life.

Let us now see what these fresh-water animals
are like, to what groups they belong, and what are
their most important characteristics. The first of the
eight large groups into which the animal kingdom
is usually divided is that of the Protozoa. These
include the simplest unicellular forms of animal
life, and occur very abundantly in fresh water,
though certain important subdivisions — e.g., the

FRESH-WATER ANIMALS 81

Radiolaria, are almost exclusively marine. Of the
next group, or Sponges, some thirty-nine or forty
families are recognised, of which one only occurs
in fresh water, the rest being exclusively marine.
Among the Coelenterata, to which the polypes, sea
anemones and corals belong, much the same state of
things occurs, for of about seventy families only three
have fresh-water representatives ; whilst of the
next group, the Echinodermata or starfish, sea-
urchins, etc., not a single species inhabits fresh
water. In the remaining groups we find the fresh-
water forms rather more abundant. The Vermes,
or " worms," using the term in its widest sense,
have a large number of fresh-water representatives,
such as the river worms, Rotifera ; but here also
the greater number of families of worms have no
fresh-water members. Among the Mollusca we
find that several of the best marked groups never
get into fresh water, while other groups, as the
snails and bivalves, have a large number of fresh-
water representatives. The seventh group, or
Arthropoda, tells much the same tale ; for though
the lower forms, such as the Entomostraca, occur
abundantly in fresh water, the higher families are
very poorly represented. Turning to the last
group, that of the Vertebrates, we find that there
are a very large number of fresh-water fish ; yet
even here, out of 137 families of fish recognised by
Dr. Giinther and other leading authorities, there are
only thirty-five that get into fresh water, and many
of 'these are only represented by single species.
On the other hand the Amphibians, such as newts

F

82 FRESH-WATER ANIMALS

and frogs, form perhaps the most purely fresh-
water group of animals known. They are, with
very few exceptions, aquatic, at any rate in their
earlier or tadpole stages ; and yet no single
Amphibian is found in salt water.

The general conclusion we arrive at is that only
a very small proportion of the families of marine
animals have made their way into fresh water. Of
the eight large groups into which the animal
kingdom is divided, one is exclusively marine, two
others have an exceedingly small number of fresh-
water representatives, whilst in the remainder,
fresh-water forms, though more abundant, are con-
fined to certain families.

It becomes now of interest to enquire why it is
that the bulk of marine animals do not make their
way up the rivers, and thereby escape from the
enemies that devour them in such enormous
numbers. For a long time it was supposed that
the real and sufficient reason was that marine
animals are unable to live in fresh water ; but it
is now known that though this may apply to some
cases, it certainly will not to all. A. series of
remarkable experiments were made some years
ago, at Marseilles, by Beudant. He took a large
number of specimens of different species of marine
snails and bivalves, and by gradually adding fresh
water to the salt water in which they were origin-
ally living he succeeded in gradually converting
them into fresh-water animals. During the experi-
ments only 37 per cent, of the animals died, 63
per cent, surviving the change from sea water to

FRESH-WATER ANIMALS 83

fresh water. This result is rendered still more
remarkable by a check experiment performed at the
same time, in which an equal number of individuals
of the same species as before were taken, and kept
in salt-water tanks, when it was found that 34 per
cent, died ; so that the difference in mortality
between those living in their natural medium, and
those constrained to change from marine to fresh-
water habitat, was but 3 per cent.

A still more remarkable case is that of the two
Entomostracan genera, Artemia and Branchipus,
of which the former is marine, the latter fresh-
water. Of Artemia two species are known, the
differences between which are very well marked.
Artemia sahna lives in water containing from 4 to
6 per cent, of salt, and is found in the brine pans
of salt works, and in other similar places. The
other species, Artemia Milhausenii, requires water
containing not less than 25 per cent, of salt. By
direct experiment it has been shown that the
differences between the two species depend simply
on the percentage of salt in the water in which
they live ; and that by gradually adding salt, or
adding water, Artemia salina may be converted into
Artemia Milhausenii^ or vice versa. This experi-
ment has been performed not merely in the labora-
tory, but also by Nature herself.*

The reverse experiment also succeeded, and by

gradually adding fresh water, and so reducing the

strength of the solution, Artemia Milhausenii has

been converted into Artemia salina. The ex-

* Vide Lecture on " Environment," p. 31. — ED.

84 FRESH-WATER ANIMALS

periments have been pushed still further : and by
addition of fresh water Artemia salina has been
converted into the fresh-water genus Branchipus,
which is of larger size, and in many respects very
different.

A number of other cases could readily be quoted
illustrating the same point, namely, that many
marine animals which never do make their way
into rivers, are quite capable of living in fresh
water, if the change is made sufficiently gradually.
If therefore it is not the difference between sea
water and fresh water that prevents marine animals
from entering rivers, there must be some other
cause or causes at work. One of these is, un-
doubtedly, the severity of the climatic conditions
to which fresh- water animals are exposed, as
compared with the practically uniform temperature
of the sea. Further, the variation in the amount
of water in rivers at times of drought and flood
respectively, the changes in the strength of the
current and modifications in the character of the
water from sewage and other contamination, all
conspire to render the environment a very shifting
one and at times a very harmful one, as compared
with the much greater uniformity of external con-
ditions under which the marine fauna exists.

But even these causes, potent as they un-
doubtedly are, will not suffice to account for so
few marine families getting into fresh water.
Some further explanation is required, and this is,
I believe, to be found in a suggestion made
originally by Professor Sollas, of Dublin, who has

FRESH-WATER ANIMALS 85

pointed out that though the adult animal might be

thoroughly well adapted to fresh-water life, yet

it would by no means follow that the early

stages of development would be equally well

suited to the change. Professor Sollas also shows

that the early larval stages of most of the marine

animals are forms peculiarly ill-adapted to living in

fresh- water streams, forms indeed which could not

hold their own under such conditions. Thus, in

the Echinodermata, which we have already seen

are exclusively marine, the young hatch as very

minute larvae swimming freely in the water by

means of cilia ; and larvae similarly occur in

almost all the other groups of marine invertebrates.

Now such small ciliated larvae are altogether

unsuited to fresh-water streams, and could never

hold their own in them. They are quite incapable

of swimming against even weak currents, and the

inevitable consequence is that each succeeding

generation would be carried further and further

down stream, and ultimately the whole species

carried out to sea. This is a very important point,

and as it is one that has not yet received general

attention I think it will be well that we should

enquire into it more fully, taking the several groups

one by one and seeing how far it will serve to

explain the special characters in the distribution of

fresh-water animals.

Before doing so, there is one further point of a
preliminary nature that will require attention. It
is a very familiar fact that the eggs of different
animals vary greatly in size. Thus, the egg of a

86 FRESH-WATER ANIMALS

herring is about the size of a pin's head, that of a
salmon is as big as a pea, while that of a dog-fish,
or of a hen is very much larger still. Every egg
contains two chief kinds of matter, germ-yolk and
food-yolk, the former of which develops directly
into the embryo, while the latter is simply a store
of food material generally in the form of minute
granules dispersed through the germ-yolk, which
can be drawn upon as required and at the expense
of which the embryo is able to develop. The
difference in size between one egg and another
concerns almost exclusively the amount of food-
yolk, which is very abundant in large eggs, but
comparatively scanty in small ones. Until the
food-yolk is used up there is no need for the
embryo to hatch : consequently embryos developed
from large eggs will be of much larger size and
greater strength at the time of hatching than those
developed from smaller eggs. This is well illus-
trated in the development of frogs. In the
ordinary frog the egg has but a comparatively
small amount of food-yolk and the embryo hatches
as a tadpole, an animal of small size and of much
simpler organisation than the frog. In the little
West Indian frog Hylodes however, the eggs are
of larger size — i.e., they contain more food-yolk —
and the embryo consequently hatches, not as a
tadpole, but as a fully-formed frog.

As small free-swimming larvae are as a rule unable
to hold their own in fresh water it follows that
any increase in the size of egg — i.e., in the amount
of food-yolk — will be a direct advantage to a fresh-

FRESH-WATER ANIMALS 87

water animal, enabling it to hatch of greater size
and strength and consequently better able to resist
the currents of the river. Hence we find that in
fresh-water animals the eggs are usually larger
than in their marine allies. A crayfish, for ex-
ample, though only a third the length of a lobster
lays actually bigger eggs. Having thus seen
what are the special conditions under which fresh-
water animals exist, we may now proceed with our
inquiry and consider in what way the fresh-water
representatives of the several groups of animals
meet these conditions.

Concerning the Protozoa I am not aware of any
points, either in structure or life-history, that
would distinguish the fresh-water from the marine
forms. The mode of life is apparently the same in
the two cases ; though it is very possible that
closer examination would show that differences do
exist, at any rate in certain cases. From this point
of view a careful study of the fresh-water Protozoa
might very possibly yield results of interest and
importance. Among Sponges Spongilla is the most
familiar one occurring in fresh water. Spongilla
reproduces sexually, like the marine sponges ; but,
unlike the latter, it can also reproduce by means of
special buds or " gemmules," Each gemmule
consists of a little spherical group of cells, formed
in the deeper part of the sponge, of which the
superficially placed cells become specially modified
so as to form a thick projecting capsule strengthened
by peculiarly formed silicious spicules. Such gem-
mules are formed usually in the autumn. Owing

88 FRESH-WATER ANIMALS

to their protective capsules they are enabled to
survive the cold, and other adverse conditions of
the winter months, which are often fatal to the
parent sponge; and in the spring the capsule
ruptures, and the contained cells crawling out
commence to develop into little sponges. Such
gemmules, or clusters of specially protected cells,
are unknown among marine sponges, and there can
be no doubt that their occurrence in Spongilla is to
be regarded as a special adaptation to the fresh-
water habits of this genus.

Among Ccelenterates the production of small free
swimming ciliated embryos is almost universal,
and this is almost certainly the reason why so
very few fresh -water members of the group are
found. Of those that do occur, the best known is
the fresh-water Hydra, whose marvellous powers
of recovery from injury have been so admirably
investigated and described by Mr. Dunkerley.* As
regards its life-history Hydra exhibits several
peculiarities, which I believe, are to be associated
with its fresh-water habits. In the first place, un-
like its marine allies, it not only does not give
rise to free-swimming reproductive zooids or jelly-
fish, but does not even show the slightest tendency
to form such zooids. Secondly, only a single
ovum, and this of very large size, is found in the
ovary, a point in which Hydra is unique among
Ccelenterates, and a point the special advantage of
which, in fresh-water forms, has already been fully

* " Hydra : its Anatomy and Development," by J. W. Dun-
kerley, F.R.M.S.

FRESH-WATER ANIMALS 89

discussed. Furthermore, this single egg has a
special protective capsule formed around it, and the
young at the time of hatching has the form of the
parent and adopts at once its mode of life.

Although jelly-fish, which are weak swimmers,
are apparently altogether unsuited to fresh-water
life, yet it is of interest to note that one truly fresh-
water form, Limnocodium, is known, while several
others occur in brackish water.* Limnocodium has
however only been found as yet in ponds, not in
streams, and the mode of its development is not
known. It was discovered a few years ago in
large numbers in an artificially heated pond of the
Regent's Park Botanical Gardens, and may possibly
have been introduced with the Victoria regia or
other plants living in the pond.

Concerning " worms," few worms are more
familiar to microscopists than the Rotifers ; active
little animals, sometimes swimming freely, some-
times attached and living in tubes, and sometimes
massed together into colonies. Nothing is more
remarkable concerning them than their extraordinary
power of resisting desiccation. Mr. Sykes has
recently sent me some dust he collected more than
a year ago from a gutter on a house-top, and con-
taining some of these dried-up Rotifers. On
placing a little of the dust in warm water for about
half an hour, a number of the Rotifers came out
and swam about actively. Rotifers have been
known to exist in this dried-up condition for many

* A second genus of fresh-water Medusae (Limnocnida) has
since been found in Lake Tanganyika — ED.

90 FRESH-WATER ANIMALS

years, and there is no doubt that their geographical
distribution, and their persistence as fresh-water
forms is largely dependent on this power, for while
in the desiccated condition they can withstand the
action of cold or drought, and can also be blown
by the wind for very considerable distances from
pond to pond, or stream to stream.

Polyzoa again, of which some of the most beauti-
ful and interesting forms are fresh -water, exhibit
special modifications in accordance with their
habitat. Almost all the fresh-water Polyzoa give
rise to specially protected buds or " statoblasts,"
which are collections of cells very similar to the
gemmules of Spongilla,* and like these, enclosed in
hard protective capsules. These statoblasts survive
the winter, which is usually fatal to the parent,
and in the spring a young fully formed Polyzoon
emerges from each, and at once adopts the mode
of life of the parent.

Among the Entomostraca, such as Daphnia
and Cyclops, we meet with what seem at first
sight to be striking exceptions to the rules we have
laid down concerning fresh- water animals ; for
while some, such as Daphnia, produce large and
specially protected eggs which hatch in the full
form of the parent, yet a very large number of
others, such as Cyclops and Cypris, produce very
small eggs, from which small free-swimming
embryos arise. We may note however that in

* These have since been shown (Ephydatia Mullen) to retain
the power of giving rise to free embryos after two years desicca-
tion.— ED.

FRESH-WATER ANIMALS 9'

these cases the small free-swimming larva, or
nauplius, though very unlike the parent, is provided
with very powerful swimming appendages, and is
therefore able to hold its own against streams
which would be fatal to those small larvae which
depend for locomotion on the action of cilia. In
fact in these Entomostraca the mode of life of
larva and adult is the same, and hence if the adult
can hold its own, there is little reason why the
larva should not do so also. Still it is a point that
requires further investigation, how it comes about
that Daphnia should lay large eggs, while the
allied genera, Cyclops and Cypris, living under the
same conditions, and in the same ponds and streams,
produce very small eggs.

Concerning these fresh- water Entomostraca there
is another point of much interest, that would well
repay further examination. In the sea there are
marked differences between the shore and shallow-
water animals living near to the land, and on the
other hand the oceanic or pelagic forms that are
met with in the open sea hundreds of miles from
land. The same distribution applies to the large
fresh- water lakes, in which shallow- water and
pelagic fauna may also be recognised.

The pelagic forms are very generally character-
ised by the possession of eyes of unusual and often
of gigantic size, and by their habit of remaining
down at some depth during the daytime and only
coming to the surface at night. Professor Weis-
mann has suggested that these two peculiarities
may be associated together, in this way. Ordinary

92 FRESH-WATER ANIMALS

daylight only penetrates to a limited depth in water,
and it has been shown that below 25 fathoms
photographic paper is not acted on by daylight.
It is also found that 25 fathoms is about the depth
to which the pelagic Entomostraca descend during
bright sunshine. The object of so descending is
apparently to utilise the light, so as to range during
the four and twenty hours over their whole hunting
ground for food. Were they to remain at the
surface in the daytime they would lose their sole
chance of obtaining food from the greater depths.
It is for this reason again that the eyes are of such
great size, for it is clear that forms with larger and
more perfect eyes would have a better chance in
the pursuit of food than those less well equipped ;
and hence by the action of natural selection, the
large-eyed forms would survive, and any further
improvement of the eyes would be preserved and
transmitted to their descendants.

Turning now to the Mollusca, one of the best
known and interesting fresh-water forms is the
common river mussel, Anodonia cygnea. Like
bivalves in general, Anodonta produces eggs of
small size ; however, these are not immediately
passed from the body of the parent, but are
transferred to the outer gills where they remain for
some time, and where they pass through all the
early stages of their development.

The gill consists of a couple of laminae fused
together along their ventral edges ; each lamina is
not a continuous membrane, but a trellis-work
composed of very numerous vertical bars, crossed
and connected together by a smaller number of

FRESH-WATER ANIMALS 93

horizontal bars. The two laminae are further
bolted together at intervals by cross bars. The
eggs, which are exceedingly numerous, form a
bulky mass lying between the two laminae. Owing
to the cilia which clothe the outer surface of the
bars, streams of water are constantly passing
through the meshes of the trellis-work ; and in this
way the embryos not merely obtain perfect pro-
tection during the early stages of development, but
are also abundantly supplied both with oxygen for
respiration, and with food in the shape of minute
particles of vegetable or animal matter suspended
in the water. The outgoing stream will also serve
to carry away any excretory or faecal matter passed
out by the embryos.

At the time of leaving the parent the young
Anodonta is a fully formed bivalve, but is still very
unlike the parent, and of too small size, and too
weak to hold its own against the currents of the
streams in which the adult lives. The young at
this period have bivalved shells, the two halves of
which are triangular, with the apices incurved so as
to form sharp serrated projections, which, when the
valves are closed, form a very efficient pair of
pincers ; they have a very rudimentary foot and gills,
and in other respects differ markedly from the
adult.

On passing out from the gills of the parent into
the water, they swim by snapping movements of
the valves, like a Pecten, but very speedily attach
themselves by the pincer-like processes of their
shells to fish, such as the stickleback, or else to the
legs of water birds. The skin either of the fish or

94 FRESH-WATER ANIMALS

bird, irritated by the pincers, swells up and forms
a capsule within which the young Anodonta
completes its development. When it has attained
the adult form it leaves its host, drops down to the
bottom of the stream, and henceforth leads the life
of the adult, half buried in the mud at the bottom
of the stream, along which it slowly ploughs its
way by means of its powerful muscular foot.

No better illustration could possibly be given of
the special modifications acquired by fresh-water
animals as regards their life-history. Marine
bivalves, such as the oyster or cockle, all give rise
to small ciliated embryos, adapted to a free
swimming existence. So also does Anodonta, but
the embryos are not set free in the streams, where
as we have seen they would be entirely unable to
hold their own, but are retained within the gills of
the parent until they have attained sufficient size
and strength to look after themselves. The second
phase in their life-history, during which they are
attached to fish or birds, has probably been acquired
rather as a means of ensuring wider distribution
than as a precaution against the strength of the
currents. The adult Anodonta leads a very
sedentary life, while through the stickleback or
bird, new colonies may be started miles further up
stream, or even in other streams some distance off.

Among fish very numerous examples are met
with of marine animals that have taken partly or
completely to fresh-water life. In many cases, as
in the salmon, the fish live normally in the sea, and
run up rivers merely for the sake of laying eggs in
places where they are exposed to less danger than

FRESH- WATER ANIMALS 95

in the sea. It is the species and not the individual
that here benefits, for while the salmon on running
up the rivers in the winter months is fat and in
good condition, on its return journey, after laying
eggs, it is in wretched condition, and very many
die on their way back to the sea. Many other
fish, such as lampreys and sturgeons, have similar
habits, ascending rivers during the spawning
season. On the other hand many characteristically
fresh-water fish, as the stickleback, frequently
descend the rivers to the sea, in which they are
perfectly at home.

The large size of the eggs of fresh-water fish, as
the trout, or such fish as breed in fresh water, as
the salmon, is worthy of notice. In consequence
of the larger supply of food contained in the eggs,
as compared with those of marine fish, the young
hatch of larger size and greater strength, and are
therefore better able to cope from the time of their
birth with the downward currents of the streams
and rivers.

The Amphibians may be briefly alluded to in
conclusion. They are a characteristically fresh-
water group of animals, the chief interest about
which is that during their development, at least in
the higher forms, such as the frogs, they pass from a
gill-breathing aquatic fish-like stage, to a lung-
breathing terrestrial condition, and show us very
clearly steps by which air breathing vertebrates have
been evolved from the more primitive water-breath-
ing forms.

Summarising the results arrived at, we may say
that fresh-water animals have almost certainly

96 FRESH-WATER ANIMALS

descended from marine animals which have become
habituated to fresh-water life: that very little
modification, if any, is required in the adult
structure to fit an animal for the change in habitat;
and that many purely marine animals will live readily
in fresh-water, if the change from one to another is
sufficiently gradual. We have further seen that the
reasons why so few groups of marine animals have
given rise to fresh-water representatives, concern
not so much the adult condition as the earlier larval
stages ; and that it is the inability of small free-
swimming ciliated larvae to hold their own against
the currents in the streams and rivers that is pro-
bably the main cause of this paucity of fresh-water
species.

An examination of the life-history of those forms
that have established themselves in fresh-water,
has shown that in many cases there are very
special devices of a curiously interesting kind to
enable the animals to get over these difficulties.
We have only had time to notice a few of the
more striking instances, and very much yet remains
to be done before we can explain fully all the
peculiarities of the fresh-water fauna.

The aspect of the question which I have
touched on this evening, has only very recently
attracted notice ; the subject is one of great im-
portance and interest, and I would venture to
commend it very earnestly to the attention of
members of the Society, as one the further and
more systematic investigation of which would,
beyond doubt, yield results of high scientific value,

VII

INHERITANCE

THE members of a Microscopical Society may find
very legitimate cause of congratulation in the pro-
gress that is being daily made in the use and
application of their favourite instrument. As
regards natural history — the history of nature —
it may rightly be said that the microscope has
effected a veritable revolution ; and this not in one
branch only, but in all. In Zoology, it has
rendered possible the detailed examination of forms
barely perceptible, or even invisible, to the naked
eye : witness the five huge volumes lately published
in the reports of the Challenger Expedition,
in which are recorded Professor Haeckel's re-
searches on the Radiolaria, and those of Professor
Brady on the Foraminifera. It has also revealed
to us numberless facts of the utmost importance
concerning the minute structure of animals whose
general anatomy was previously well known : facts
which in many cases have shown that our earlier
ideas as to the affinities of these forms were

G

98 INHERITANCE

defective or erroneous. But far more than all
this, it has opened up to us the science of Embry-
ology— the most fascinating study a man can
pursue — which not merely teaches us the several
stages through which the complexities of adult
structure are reached, but also enables us to
reconstruct the pedigrees, the genealogical histories
of the various groups of animals, and proves to us
that morphological resemblances are no mere
accidents, but are indications of true blood relation-
ship between one form and another.

In Botany, results of similar nature have been
attained ; and the amount of space accorded in our
more recent text-books to the lower orders of plants
is cogent evidence of the importance attached to
the results of microscopical examination by those
best qualified to pass judgment upon them. Nor
has the benefit been confined to the student of
biological science. To the geologist, the micro-
scope is rapidly becoming as important, and in-
dispensable as it has long been to the zoologist or
botanist. Like his biological brethren, the modern
geologist is no longer content with a knowledge,
however accurate and minute, of the structure and
present condition of the rock he is examining, but
recognises that in the microscope he has a means
which, used aright, will enable him to unravel the
pedigree of the rocks, to reconstruct their past
history, and to determine through what series of
changes, in what order, and by what agencies,
their present condition has been brought about.

The limits of time and space forbid that I should

INHERITANCE 99

refer, save in the briefest manner, to the ever-
widening applications of the microscope in other
branches of knowledge, and the benefits which it
is conferring on mankind. In trade and commerce
the microscope is employed more frequently and
relied on more fully than is generally appreciated ;
and in such matters as the adulteration of food,
and in criminal enquiries, it often yields evidence
of the most material and convincing character.
There is perhaps no direction in which microscop-
ical enquiry has advanced more rapidly of late
years than its employment as a means of detecting
disease, or even of determining its true nature and
causation. Pathology is one of the most actively
growing of modern sciences ; and though its fulness
of time has not yet come we may feel well assured,
from the results already attained, that the scientific,
and especially the microscopical, investigation of
disease, will in the immediate future afford us
most powerful and welcome assistance in the
alleviation of human suffering. It is in con-
siderations of this kind — presented here, I am but
too well aware, in the crudest possible form — that
we find the justification for the high position to
which a Microscopical Society may rightly aspire.
In the microscope we have perhaps the most potent
instrument of research that mankind has ever
possessed ; and in the ever-widening circle of its
influences, in its far-reaching applications, we may
see opportunity for enrolling amongst our numbers
men of the most varied interests and pursuits, and
so gaining that free interchange of independent

loo INHERITANCE

opinion which is one of the highest privileges and
delights of civilised humanity.

One sometimes hears it said that a microscopist,
being occupied with small things, is usually,
perhaps of necessity, a man of small ideas. This
may possibly be true in individual cases, but as a
general statement I believe it to be utterly false.
No better justification of this belief could be found
than is afforded by the present state of our know-
ledge with regard to the subject I have chosen for
my address.

The real nature and modus operandi of Inheri-
tance are problems of the widest possible interest
and importance, problems which have baffled many
in the past, and which are at this moment being
attacked by different observers, working from
different sides, and along different lines of attack,
but all relying for their evidence on microscopical
observation. In dealing with the subject of
Inheritance it is well to bear in mind that the
problem is as yet unsolved, the question still an
open one. Neither can I myself make any material
contribution towards its solution : all I propose to
do here is to indicate the main conditions of the
problem, and to point out what appear, in the
light of recent investigation, to be the most
promising lines of attack. The problem itself is
familiar enough, and may be expressed in its
simplest form by the question — Why is a child like
its father ? Why is it ; how does it come about
that a young animal resembles and grows into the
likeness of its parent ? Stated thus, the problem

INHERITANCE 101

seems definite enough. However, it is really much
more complicated than appears at first sight, and
it will be well to consider briefly certain preliminary
matters before dealing with the more serious
attempts that have been made to grapple with its
difficulties.

In the first place it should be remembered that
reproduction, whether of animals or plants, is
effected in two principal ways ; asexual and
sexual ; and that the phenomena of inheritance are
seen in both cases. Thus, to take a familiar case,
the common fresh-water Hydra reproduces either
by budding, or by the formation of eggs. The
former is an asexual process, the bud appearing as
a hollow outgrowth from the body wall of the
parent, which acquires mouth and tentacles at its
distal end, and after a longer or shorter time
detaches itself and becomes an independent Hydra.
An egg, on the other hand, is a specialised cell of
the ectoderm, or outer layer of cells of the parent,
which is incapable of development of any sort until
it has been fertilised by a spermatozoon, from the
same or another animal. After fertilisation the
egg segments, i.e., divides repeatedly so as to give
rise to a number of cells from which, by further
growth and differentiation, an embryo and ulti-
mately an adult Hydra is produced. The two
modes of reproduction, sexual and asexual, are
absolutely unlike, and yet the final results are the
same ; for so far as we are aware, there are no
points of difference that will distinguish with
certainty a Hydra produced by budding from one

102 INHERITANCE

produced from a fertilised egg. Inasmuch as these
two forms are not only similar to each other, but
similar also to their parents, it follows that a
Theory of Inheritance, to be of any real value,
must apply equally to sexual and asexual pro-
cesses of reproduction. I lay stress on this point
as it is one which appears to me to have
been very frequently overlooked, especially of late
years.

The power of repairing mutilation, that is
possessed in so marked a degree by many animals,
is another phenomenon of which account must be
taken in any Theory as to the real nature and
mode of action of Inheritance. A Hydra may be
cut into many pieces, and each piece will regenerate
the missing portions and give rise to a perfect
animal. One or more of the arms of a starfish
may be removed, the whole of the viscera of an
Antedon may be turned out, the leg of a crab or
the limb or tail of a newt may be cut off, and the
loss will in each case be made good. Spallanzani
cut off the tail of a salamander six times in suc-
cession, and Bonnet eight times ; while the eye-
bearing tentacle of a snail has been removed
twenty times ; and yet after each mutilation the
missing organ has been reproduced. In the more
highly organised animals, such as birds and
mammals, this power of repairing mutilation is
much restricted : removal of parts of the epidermis
is, however, readily made good ; while such
operations as skin-grafting, and transfusion of
blood from one animal to another, show that

INHERITANCE 103

isolated parts of even the highest animals may
retain their vitality and special properties when
placed under favourable conditions, More striking
examples are afforded by such cases as those
quoted by Mr. Darwin : in one instance, the spur
of a cock inserted into the ear of an ox lived for
eight years, and grew to a length of nine inches ;
while in another, the tail of a pig removed from its
natural position and grafted into the middle of the
animal's back lived for a time and recovered
sensibility.

The phenomena of Reversion are again of great
importance in reference to the problem of Inherit-
ance. It is well known that animals may trans-
mit to their offspring characters which are not
manifested in themselves : the tendency of gout
and some other diseases to appear in alternate
generations is perhaps the most familiar instance.
In such cases we must regard the disease, or other
peculiarity, as present in a latent form in the
generations which it apparently skips, for how
otherwise could we understand its reappearance in
a later generation ? The tendency of domestic
animals, and more especially of cultivated flowers
and fruits, to revert — either in form, colour, or
other characteristics — to the ancestral wild con-
dition, is another good illustration of what may
well be termed Latent Inheritance. Readers of
Mr. Darwin's works will call to mind his famous
experiments on pigeons ; more especially that
crucial one in which he first paired a black Barb
with a red Spot ; then another black Barb with a

104 INHERITANCE

white Fantail; and then paired the mongrel Barb-
Spot with the mongrel Barb-Fantail, the result being
that he obtained a family of birds which in colour
and markings were almost identical with the blue
Rock pigeon, the common ancestor of all domestic
pigeons.

The tendency of cultivated and domesticated
plants and animals to revert to a former ancestral
condition may perhaps be illustrated mechanically
in this way. Take a pack of cards, and lay it on
the table ; the cards will all lie on their sides, and
be in a condition of stable equilibrium, so that the
table may be struck or shaken without materially
affecting their position : this represents the normal,
i.e., the wild or ancestral condition of the race.
Now arrange the cards on their edges, resting them
against one another, and so building them up into
a pagoda: the resulting structure is a far more
imposing one than the pack of cards when laid flat
on the table, but it is also an eminently unstable
one, its instability being directly proportional to
the extent to which it departs from the initial con-
dition : a very slight shake or push of the table
will cause the whole structure to collapse, and
revert to its condition of initial stability, the cards
all falling flat on their sides as at first. So 4:he
Pouter or the Fantail are much more impressive
and remarkable birds than the blue Rock, but the
former are artificial productions, in a condition of
great instability, and very readHy revert to the
ancestral condition. These are but a few of
the considerations which must be kept in

INHERITANCE 105

mind when dealing with Inheritance. Let us now
consider in what way the problem may best be
attacked.

Of Theories of Inheritance there are two which
have attracted special attention, and which demand
careful consideration. These are, first, Mr. Darwin's
" Provisional Hypothesis of Pangenesis ; " and
secondly, the view more recently advanced by
Professor Weismann, of Freiburg.

Mr. Darwin's Theory is stated by himself as
follows : " It is universally admitted that the cells or
units of the body increase by self-division or pro-
liferation, retaining the same nature, and that they
ultimately become converted into the various tis-
sues and substances of the body. But besides
this means of increase, I assume that the units
throw off minute granules, which are dispersed
throughout the whole system ; that these when
supplied with proper nutriment, multiply by self-
division, and are ultimately developed into units
like those from which they were originally derived.
These granules may be called gemmules. They
are collected from all parts of the system to consti-
tute the sexual elements, and their development in
the next generation forms a new being ; but they
are likewise capable of transmission in a dormant
state to future generations, and may then be
developed. Their development depends on their
union with other partially developed or nascent
cells, which precede them in the regular course
of growth. Why I use the term union, will be
seen when we discuss the direct action of pollen

io6 INHERITANCE

on the tissues of the mother-plant. Gemmules are
supposed to be thrown off by every unit, not only
during the adult state, but during each stage of
development of every organism ; but not necessarily
during the continued existence of the same unit.
Lastly, I assume that the gemmules in their
dormant state have a mutual affinity for each other,
leading to their aggregation into buds or into the
sexual elements. Hence, it is not the reproductive
organs or buds which generate new organisms,
but the units of which each individual is composed.
These assumptions constitute the provisional hypo-
thesis which I have called Pangenesis. Views in
many respects similar have been propounded by
various authors."

It will be seen that Pangenesis is a mechanical
theory of Inheritance ; and that it recognises and
faces fully the difficulties of Reversion and of
Repair of Mutilations, and explains how organs
may become abnormally multiplied and transposed
through the gemmules developing accidentally in
wrong places, as in the case of supernumerary
fingers or toes, or the development of hairs or teeth
in unusual situations.

Pangenesis, in spite of the ready explanation
it gives of many difficulties, has never met with
anything like general acceptance. Its illustrious
author, however, while careful to speak of it always
as a Provisional Hypothesis, regarded it with much
affection ; and alludes to it almost pathetically as
his neglected child, for which he predicts confi-
dently a future career of greatness. Such an
opinion claims the highest respect, and compels

INHERITANCE 107

the utmost caution in criticising the theory :
yet it cannot be denied that it involves certain
difficulties which seem of great weight, and which
have not yet been satisfactorily met. In the first
place, there is the mechanical difficulty of the
extraordinary numbers of these gemmules which
must be present. Gemmulee must be derived from
every component cell of the body : for Mr. Darwin
lays much stress on the independence of these cells,
quoting Virchow to the effect that " Every single
epithelial and muscular fibre cell leads a sort of
parasitical existence in relation to the rest of the
body .... every single bone corpuscle really
possesses conditions of nutrition peculiar to itself."
Again, Sir James Paget speaks of each cell as
living its appointed time, and then dying and being
cast off or absorbed ; while further on Mr. Darwin
continues, "I presume that no physiologist doubts
that, for instance, each bone-corpuscle of the finger
differs from the corresponding corpuscle in the
corresponding joint of the toe ; and there can
hardly be a doubt that even those on the two
sides of the body differ, though almost identical
in nature."

Again, these gemmules must not only be
formed from every cell, but must be present in
enormous numbers from every cell, and at every
period of life : for it is well known that a portion
of a leaf of Begonia or Asplenium can reproduce
the whole plant ; and to do this there must, on the
theory of Pangenesis, be present in this portion of
the leaf gemmules from every part of the plant. So
again, the repair of mutilations can only be possible

io8 INHERITANCE

through the presence of a great reserve stock of
gemmules of every kind, from all parts, Moreover,
it must be borne in mind that the component cells
of the body are not simple, homogeneous structures,
but that, as is daily becoming more and more
evident, some at any rate of them have an exceed-
ingly complex structure, consisting of parts of very
different composition, and discharging very different
functions. Hence it must follow that many kinds
of gemmules, each in enormous numbers, must be
required from a single cell in many cases. Inas-
much as the body structure of the young and of
the adult animal are different, there must also be
different sets of gemmules for the several stages
of existence : and it becomes a matter of the ut-
most difficulty to form the slightest conception of
how these different sets take up the running in
successive stages of development. If we remember
that gemmules from all parts, from each component
cell of the body, have not only to be formed in
great numbers, but have also to find their way
about the body to be collected together from the
most remote parts, and planted in due proportion
in each of the ova or spermatozoa of the parent,
we become fairly staggered at the magnitude of the
operations we are asked to believe that each animal
performs with such apparent ease.

It must also be remembered that Pangenesis
requires that besides the active gemmules there
must be enormous numbers of latent gemmules,
corresponding to ancestral characters, present in
each egg and spermatozoon : for Pangenesis ex-

INHERITANCE 109

plains such instances of Reversion as the production
of a pigeon practically identical with the blue Rock
from a Barb-Fantail and a Barb-Spot as due to
the development of ancestral blue Rock gemmules,
which must be supposed to be present in all
pigeons in sufficient numbers to produce fully
formed offspring, though they usually remain in
a latent condition. Considering the enormous
number of generations that must have intervened
between the original ancestral blue Rock and the
present Barb or Fantail, and that each member of
each of these generations must be supposed to
have possessed these ancestral germs in sufficient
numbers to cause Reversion if an opportunity
occurred, the magnitude of the operation and the
numbers of such germs originally present become
simply inconceivable.

A further difficulty is found in the consideration
that Pangenesis, involving as we have seen the
presence of gemmules corresponding to the different
periods of life of the parent, fails altogether to
explain the inheritance of the characters of old age,
or of any period beyond that at which the ova or
spermatozoa were discharged from the parent. It
is hardly sufficient to say in answer to this that " in
all the changes of structure which regularly super-
vene in old age, we probably see the effects of
deteriorated growth, and not of true development,"
for the objection may apply not merely to the
period of old age, but to three-fourths or more of the
entire life of the animal.

One further objection may be alluded to : On

no INHERITANCE

the theory of Pangenesis we should certainly
expect that the removal of a part, such as the tail
of a sheep or horse, especially when effected in
early life before the breeding period has been
reached, would lead at any rate to diminution of
size of the part in the offspring : for surely the
removal at an early age of the source from which
the gemmules arise ought to have at least some
effect on the transmission, through the gemmules,
of this part to the offspring ; yet it is well known
that such mutilations do not tend to be inherited.

Considerations such as these show clearly that
whatever may be the ultimate fate of the Theory of
Pangenesis, it is not yet in a position to command
acceptance : indeed, some of the objections seem of
so important and fundamental a nature as to
compel us to regard them as fatal to the Theory, at
any rate in its present form. Quite recently, Mr.
Francis Galton has published the results of a series
of most laborious statistical enquiries, undertaken
with the view of ascertaining whether inheritance
takes place according to definite laws, and if so to
determine as accurately as possible what these
laws are. His researches, which are of the greatest
possible interest and importance, and must exercise
great influence on future speculations, lead him to
a view which he refers to as Paniculate Inheritance,
and which may be described as an aggravated
form of the Theory of Pangenesis, propounded by
his illustrious kinsman. Mr. Galton states his view
thus: "All living beings are individuals in one
aspect, and composite in another. They are

INHERITANCE ill

stable fabrics of an inconceivably large number of
cells, each of which has in some sense a separate
life of its own, and which have combined under
influences that are the subjects of much speculation,
but are as yet little understood, We seem to in-
herit, bit by bit, this element from one progenitor,
that from another, under conditions that will be
more clearly expressed as we proceed, while the
several bits are themselves liable to some little
change during the process of transmission.
Inheritance may therefore be described as largely,
if not wholly, particulate." Farther on, he compares
the process of inheritance to the construction of a
modern building out of the corresponding parts of
the ruined edifices of former days. " This simile,"
he says, " gives a rude though true idea of the
exact meaning of Particulate Inheritance, namely,
that each piece of the new structure is derived from
a corresponding piece of some older one, as a lintel
was derived from a lintel, a column from a column,
a piece of wall from a piece of wall. We appear
then to be severally built up out of a host of
minute particles of whose nature we know nothing,
any one of which may be derived from any one
progenitor, but which are usually transmitted in
aggregates, considerable groups being derived from
the same progenitor. It would seem that while the
embryo is developing itself, the particles more
or less qualified for each new post wait as it were
in competition to obtain it. Also, that the
particle that succeeds must owe its success
partly to accident of position, and partly to

U2 INHERITANCE

being better qualified than any equally well
placed competitor to gain a lodgment. Thus the
step by step development of the embryo cannot fail
to be influenced by an incalculable number of small
and most unknown circumstances."

These views are boldly expressed but they are
also distinctly crude, and the metaphor, which Mr.
Gallon is dangerously fond of using, rather confuses
than aids the explanation. Of more real value are
his attempts to determine the numerical ratio in
which characters, such as height, colour of eyes,
etc., tend to be transmitted to successive gener-
ations. On this point Mr. Galton comes to the
following very definite and important conclusions :
"The average contributions of each separate
ancestor to the heritage of the child were deter-
mined apparently within narrow limits, for a
couple of generations at least. The results proved
to be very simple ; they assign an average of one
quarter from each parent, and one-sixteenth from
each grandparent. According to this geometrical scale
continued indefinitely backwards, the total heritage
of the child would be accounted for." Results of
this kind are of the greatest possible value, and open
up a most promising field for further enquiry.

I turn now to a consideration of the important
series of researches by Professor Weismann, which
have of late years attracted so much attention.
These are of especial interest to microscopists,
because the data on which Prof. Weismann bases
his arguments are obtained from a careful study,
with the most refined histological methods, of the

INHERITANCE 113

minute structure of the egg, and of the changes
which it undergoes during development. This is a
perfectly philosophical standpoint to adopt ; for the
egg, say of a hen, has the power of developing
into a chick without any further assistance from
the parent, provided only that certain conditions of
temperature and moisture are fulfilled : and it
follows that the problem of heredity is centred
in the egg, and that it is therefore reasonable to
hope that patient investigation of the egg-structure
may throw some light on the question.

Weismann in the first instance lays special
stress on the fundamental difference between
the unicellular animals, or Protozoa, on the one
hand, and the Metazoa, or multicellular animals,
on the other. In Protozoa there is no distinction
between body cell and germ cell : the entire animal
is but one single cell. Reproduction is effected by
simple division of this cell, and every individual
peculiarity in the parent must therefore be trans-
mitted directly to the offspring. Heredity therefore
in Protozoa is no problem at all, but a simple and
direct consequence of the mode of reproduction.
In the Metazoa however the case is different :
here the adult animal consists not of a single cell,
but of many cells arranged variously so as to form
the epithelial, muscular, nervous, and other compo-
nent tissues of the animal. Of these cells certain
ones are early distinguished as genital cells, and to
these the power of reproduction, at any rate sexual
reproduction, is confined, and in them are centred
the hereditary tendencies of the whole organism.

H

II4 INHERITANCE

The problem is to explain how it is that in
the Metazoa, one particular cell, the ovum, should
have acquired and retained this special power of
transmitting the characters of the entire animal.
This problem Weismann proceeds to attack. He
calls special attention to the general agreement
among competent observers, that the part of the
cell directly concerned in the transmission of
hereditary features is the nucleus ; and he brings
forward observations of his own in support of this
view. He assumes the presence in the nucleus of
a special substance to which he gives the name
germ-plasma, and to which he supposes the power
of hereditary transmission to be confined. He
maintains that this germ-plasma is of exceedingly
complex structure, and that it has the power of
indefinite growth without loss of its essential
characters. He further supposes that the germ-
plasma of an egg is not wholly employed in build-
ing up the body of the embryo, or young animal,
but that a certain portion of it remains unchanged,
and produces the germ cells of the succeeding
generation. In this way the germ-plasma is
supposed to pass unchanged from one generation to
another, and this continuity of the germ-plasma is
regarded by Weismann as the fundamental cause of
heredity. It cannot be said that this explanation is
a satisfactory one. In the first place it is not
really an explanation of inheritance at all ; for
unlike Darwin's theory of Pangenesis, it does not
attempt to explain the actual modus operandi of
inheritance, but merely localises the power of

INHERITANCE 115

transmitting hereditary characters to the germ-
plasma ; and asserts that the power is due to the
special and utterly unknown molecular constitution
of this germ-plasma, the very existence of which is
altogether hypothetical.

A more valuable part of Weismann's work con-
cerns the share which sexual reproduction takes, in
his opinion/as the direct cause of specific variation.
Inasmuch as the germ-plasma is supposed to be
of constant composition, and to be transmitted
unchanged from generation to generation, it follows
that characters acquired by the parent cannot be
transmitted to the offspring. This statement, that
acquired characters are not inherited, is one that
has attracted much attention, and that at the
present time is being most keenly debated.
Should it prove to be true, it becomes necessary to
look elsewhere for the cause of specific variation in
animals : and this, Weismann maintains, is to be
found in the mingling of germ-plasma from two
separate animals in the act of fertilisation of the
ovum by a spermatozoon.

Limits, both of time and space, forbid that I
should follow these arguments further. We are at
present very ignorant concerning the nature,
properties, and reactions of living things, while life
itself remains as great a mystery as ever. Perhaps
it will not be until we have gained some clue to the
mystery that we shall be able to understand what
is the real nature of inheritance : in the meantime
we may heartily welcome all earnest efforts towards
the solution of the problem.

VIII

THE SHAPES AND SIZES OF
ANIMALS

IT may, I believe, be laid down as a general rule,
that when a man deliberately selects a particular
subject for conversation or for more serious
consideration, his choice is made, consciously or
unconsciously, for one of the following reasons.
Either it is because he has himself paid attention
to the subject, and has something to say about it
which he thinks will be novel or at any rate
interesting to his hearers ; or it is because he is
addressing others better instructed in the subject
than himself, whose opinions he is desirous to
obtain; or finally, it may be that he introduces
the subject with the express though probably
unavowed purpose of finding out what his own
opinions are about it.

I confess at once that it is this last motive that
has determined my choice of a theme for the
Presidential Address, and I make no apology
for this. It must have happened many times

THE SHAPES AND SIZES OF ANIMALS 117

to each of us, that ideas have occurred unex-
pectedly on subjects to which we had pre-
viously paid but little attention ; ideas, which
though recognised at once as crude and disjointed,
are yet felt instinctively to contain germs of interest,
worthy of future development. Such ideas should
not be let slip : it is well to docket them, and
without attempting too soon to frame a consistent
notion of their real bearing, or of the conclusions
to which they may lead, it is well also to keep a
look-out for any additional facts or ideas bearing
on the subject, to take due note of them, and
after a time to turn out one's box, go over one's
notes, and take stock of one's material. At this
stage conversation and discussion with others will
probably afford material assistance : and in bring-
ing before you this evening a subject concerning
which my state of mind is exactly expressed by
saying that I want to find out what I think about
it, I believe I am utilising in a very proper and
legitimate manner the advantages which such a
Society as ours confers upon its members.

The problem that I desire to deal with, that of
the forms and dimensions, or in more popular
language, the shapes and sizes of animals, may be
stated thus : Is it a mere matter of chance that
animals, say butterflies and birds, have certain
characteristic shapes ? Clearly it is not altogether
so : it is plain to every one that the shapes of
animals are correlated with their habits, and that
if it were a purely haphazard business there would
be no reason why butterflies should so much

n8 THE SHAPES AND SIZES OF ANIMALS

resemble one another, or why birds should be con-
structed on one common plan. If then it is not mere
matter of chance, can we determine in any way
what are the causes which govern the shapes of
animals, and what are the laws in accordance with
with which their effects are produced ?

So with size : for each animal we have a certain
standard of size, which is rarely very greatly
departed from. Such names as cat, pigeon, cock-
roach, convey to our minds not merely impressions
of animals of certain shape, structure, and
habits, but also of tolerably fixed dimensions.
A cockroach as big as a cat would at once arrest
our attention as unusual. The problem we have
to deal with is to find out, if possible, the causes
which regulate the dimensions of animals, and
determine that there shall be for each kind of
animal a certain average size. These are most
elementary considerations, but they will serve to
show us what our subject is, and that it is one well
worthy of attention.

With regard to the shapes of animals, we find
that among the simplest animals, or Protozoa, it is
characteristic of the more primitive genera that
there should be no definite or consistent shape.
This is well seen in Amceba, which consists of a
minute speck of protoplasm equally contractile
in all directions ; protrusions of the protoplasm,
known as pseudopodia, being put out from any
part of the surface, and in any direction. In
Amoeba and in all forms which exhibit similar
" amoeboid " movements, there is no distinction of

THE SHAPES AND SIZES OF ANIMALS 119

ends, sides, and surfaces, such as we are familiar
with in the higher animals. Anterior and posterior
ends, right and left sides, dorsal and ventral sur-
faces, are terms which have no meaning in
reference to an Amoeba, for any part of the animal
may go first in locomotion, and when crawling the
animal moves along on whatever part of its surface
happens to be in contact with foreign bodies. The
only distinction in such an animal is between
inside and outside, and even this is not permanent
in all cases. In the higher Protozoa the body
presents a clear distinction between its outer or
superficial layer, which is clearer, firmer, and more
contractile ; and the inner or central part, which is
more fluid, less contractile, and usually less trans-
parent ; the former being named ectosarc, the latter
endosarc. In the simpler forms however as may
be seen in many Amoebae, this distinction between
ectosarc and endosarc is not a constant or
permanent one ; the protoplasm of the whole body
exhibits constant flowing movements, by which
parts that at one time are at the surface, at another
are carried into the interior : in such cases the
ectosarc is merely the layer of protoplasm that at
any one moment is at the surface ; and if this
differs in appearance from the more deeply placed
protoplasm, such difference is perhaps due to the
effects of contact with the water in which the
animal dwells, rather than to any fundamental
distinction in structure or composition of the pro-
toplasm itself.

Even amongst the Amoeboid Protozoa it is how-

120 THE SHAPES AND SIZES OF ANIMALS

ever possible to speak of a distinct shape, for
when perfectly at rest they all tend to withdraw
their processes, and to assume more or less
definitely a spherical form. This spherical shape
is very characteristic either of the normal condition
or of the resting state of a large number of
Protozoa, and deserves further notice. When
widely departed from, such departure is commonly
associated with induration of part of the surface
either as a mere thickening of the ectosarc, or in
the form of a definite shell ; a condition which is
clearly a more modified one, and usually involves
and is associated with further differentiation, such
as the presence of a definite mouth, and usually of
distinct oral and aboral ends to the body.

Concerning the spherical forms, it may be noted
that they are all aquatic and free swimming :
aquatic, because in the case of animals with soft
body-substance, it is only in water that the body of
the animal can be sufficiently supported on all sides
to enable the spherical shape to he retained : free
swimming, because a habit of crawling leads, as we
shall see directly, to definite modification of external
shape, involving a distinction between dorsal and
ventral surfaces, and almost invariably a further
distinction between anterior and posterior ends. In
the typically spherical animal, all parts of the
surface are equidistant from the centre, and are
alike in all respects ; and such an animal will float
suspended in the water with any part of the surface
uppermost, or any part undermost.

There is much reason for thinking that the

THE SHAPES AND SIZES OF ANIMALS 121

spherical shape is not merely the simplest which an
animal can offer, but is also the most primitive.
In evidence of this we may quote the spherical
shape of the ovum, which is characteristic of all the
higher animals as regards the earliest stage of its
formation, and of most as regards its fully matured
condition. If we are right in regarding the
embryological development of an animal as a re-
capitulation of its ancestral history, then the earliest
developmental condition — i.e., the ovum, or egg —
must represent the most primitive ancestral phase,
and the significance of the spherical shape so
characteristic of the earliest condition becomes at
once apparent.

A further argument in favour of the primitive
nature of the spherical form may be drawn from
the development of the more modified forms of
cells, even in adult animals. Thus, the shapes of
the epithelial cells vary greatly, according to the
part from which they are taken. Cells from the
surface-layer of the epidermis, such as those lining
the mouth or covering the hand, are thin scales
fitted together edge to edge, and with their flat
surfaces parallel to the surface which they cover.
Cells from the epithelial lining of the stomach
or intestine on the other hand, are columnar or
rod-like in form, being placed side by side with
their long axes vertical to the surface they clothe.
Yet a section through the epidermis shows that
its deepest layer consists of spherical cells, which
gradually approach the surface as those lying over
them get rubbed away, and which, as they move

122 THE SHAPES AND SIZES OF ANIMALS

towards the surface, get flattened more and more
until ultimately they become converted into the
scale-like cells. Thus, each scale-like or pavement
epithelial cell is in the first instance, in its earliest
stage of existence, a spherical cell. So also with
the columnar cells of the stomach or intestine.
Each such cell is formed in the deeper layers of
the epithelium as a spherical cell, and gradually
becomes elongated into a columnar cell as it
approaches the surface. The spherical cell is
therefore the link connecting the scale-like and the
columnar cells; an indifferent or primitve form,
from which either of the more modified forms may
be derived, and which is really the earliest stage in
the developmental history of both. It would be
easy to multiply instances of this kind, but I have
said enough to show that there is really strong
ground for .holding that the spherical form is to be
regarded as a primitive one ; perhaps as the most
primitive form met with amongst animals.

The truly spherical shape, in which all parts of
the surface are alike and of equal value, is only
seen in the unicellular animals, or Protozoa, and in
the individual cells of higher animals. The early
embryonic phases of many of the higher animals,
known as Morula and Blastula, are also spherical,
the former being a solid heap of spherical and
polygonal cells resulting from the repeated division
of the fertilised ovum ; while the blastula is a later
stage, having the form of a hollow ball with a wall
composed of a single layer of cells surrounding
a cavity filled with fluid. Neither morula nor

THE SHAPES AND SIZES OF ANIMALS 123

blastula however is absolutely spherical in the
sense in which I have used the word above,
for the boundary lines between the individual cells
must be of different value to the cells themselves,
so that all parts of the surface cannot be identical.
Moreover, the component cells very usually, perhaps
always, present differences of size or structure by
which upper and lower hemispheres may be
marked off from each other, and by which the true
spherical symmetry becomes still further disturbed.
Volvox and Pandorina may be quoted as examples
of permanent blastulse in which the component
cells present no such differences, but they are
forms the animal nature of which is still extremely
doubtful.

Leaving the spherical forms, the next character-
istic shape we meet with among animals is that
known as radially symmetrical, of which the most
typical instances are met with in the group of
Ccelenterates ; an ordinary jelly-fish affording as
excellent an example as one could wish to find.
A sphere is said to have an infinite number of
axes, all equal to one another ; but a jelly-fish may
be described, if the mathematicians will pardon the
phrase, as having of axes a number that can only
be expressed as being one more than infinity, for
its bell-shaped body, besides having an infinite
number of transverse equal axes, has one definite
longitudinal axis round which all the parts are
symmetrically arranged. Watch a jelly-fish swim-
ming in still water, and you will note that while
locomotion is always effected in the direction of the

I24 THE SHAPES AND SIZES OF ANIMALS

main or longitudinal axis of the animal, the rounded
end of the bell going first, the open mouth of the
bell last, yet that it is a matter of indifference
which part of the rim of the bell is uppermost.
The animal, in fact, may be said to have anterior
and posterior ends, but no distinction between
dorsal and ventral surfaces, or between right and
left sides ; and this is the characteristic arrange-
ment in a radially symmetrical animal.

If the spherical form is primitive, then the
radially symmetrical form must be derived from it,
and of this we have direct evidence in the fact that
every radially symmetrical animal is developed
directly or indirectly from a spherical egg. Con-
cerning the actual historical mode of derivation of
the radial from the spherical form however there
has been much discussion, and the question cannot
yet be regarded as settled. The chief difficulty arises
from the fact that in actual development there are
at least two quite different ways in which the
spherical ovum may give rise to a radial larva, and
it has not yet been determined which of these
modes is the more primitive, and in what way one
of them could have been derived from the other.

The more usual mode of development is as
follows : The ovum, after fertilisation, divides into
two cells ; each of these again divides, giving four
in all ; the process is repeated until a solid heap of
cells, the morula, is produced ; then this becomes
converted into a blastula by the cells moving away
from the centre and becoming arranged so as to
form a spherical ball, consisting of a single layer of

THE SHAPES AND SIZES OF ANIMALS 125

cells enclosing a central space filled with fluid.
The blastula now becomes flattened on one side
and the flattened side becomes doubled up within
the rounded part, so that the larva now assumes
the form of a hemispherical cup, the walls of which
consist of two layers of cells, outer and inner,
between which is a narrow chink-like space con-
taining fluid, which is really the last disappearing
remnant of the blastula cavity of the earlier stage.
The cup-shaped larva is spoken of as a gastrula.
A gastrula developing in this fashion is said to
be formed by invagination. Such an invaginate
gastrula is of very wide occurrence, occurring as an
early larval stage in members of all the large
groups of the animal kingdom above the Protozoa —
i.e., in Sponges, Ccelenterates, Echinoderms, Worms,
Molluscs, Arthropods, and Vertebrates.

The second mode, referred to above, in which a
gastrula is formed is by what is called delamination.
The starting-point, the egg, is the same as before,
and so also is the gastrula itself, for the delaminate
and invaginate gastrulae, though formed in entirely
different ways, cannot always be distinguished
from each other. In the development of the
delaminate gastrula the egg segments, giving rise
to a solid heap of cells, the morula ; and this
becomes a blastula as before. Each cell of the
blastula now divides into inner and outer parts, so
that the blastula wall becomes double, consisting
of outer and inner layers of cells, surrounding a
central cavity filled with fluid. By perforation of
one pole the cavity is placed in communication

126 THE SHAPES AND SIZES OF ANIMALS

with the exterior, and the embryo becomes a
gastrula.

It is by no means easy to determine which of
these two forms is the more primitive. The in-
vaginate gastrula is much more widely distributed
in the animal kingdom, occurring, as we have seen,
in all the large groups, while the delaminate gas-
trula is much more restricted. On the other hand
it is not easy to see how the invaginate gastrula
first came into existence, for it is by no means
clear what advantage a spherical blastula-like
animal gets by becoming flattened on one side,
or what further advantage is conferred by a very
slight depression or cupping of this flattened sur-
face. Such a depression is useful enough after it
has reached such proportions as to give rise to a
sac-like cavity suitable for the reception and diges-
tion of food, but the early stages of its formation
are useless, and could not have been preserved for
such purpose. In the case of the delaminate'
gastrula however there is no such difficulty, and it
is possible to construct a hypothetical series of
forms which may well represent the ancestral
series in the pedigree of the gastrula, each step
marking a distinct advance in organisation, and
being a sufficiently definite improvement to justify
its perpetuation ; and the whole series correspond-
ing to the successive stages of development of the
delaminate gastrula of the present day.

Starting with the blastula stage, a hollow ball
whose wall is but one cell thick, we note that the
inner and outer ends of each of the cells are

THE SHAPES AND SIZES OF ANIMALS 127

exposed to very different environment. The outer
ends being on the surface of the blastula can come
in contact with and gain cognisance of the outer
world, while the inner ends facing towards the
blastula cavity can have no direct contact or con-
cern, except with bodies that have passed inwards
through the outer parts of the cells. Hence
merely as a consequence of this arrangement,
physiological differentiation will be set up between
the outer and inner ends of each cell ; the outer
parts of the cells will become the seats of sensation
and of locomotive activity, while the inner ends,
freed from these functions, apply themselves to
other purposes and become specially nutritive or
digestive in function. The next stage is a simple
one. The differences between the outer and inner
ends of each cell once established will tend to
increase as each part of the cell learns to discharge
its special functions more efficiently : a mechanical
separation of the two parts of the cell is but a slight
further differentiation ; each cell dividing into two —
an outer cell, sensitive and locomotive and probably
respiratory in function, and an inner cell specially
digestive in purpose.

So far we have supposed the food to consist of
small particles captured by the outer parts of the
cell, or the surface cells where two layers are
established, and passed inwards to the inner parts,
or the inner cells, to be digested. The act of di-
gestion is still intra-cellular — i.e., is effected entirely
within the substance of the cells just as in an
Amoeba. If now we suppose particles of larger

128 THE SHAPES AND SIZES OF ANIMALS

size to be taken in as food, we can well imagine
how these might be passed on by the inner cells
into the central cavity, which will then become a
digestive cavity; the inner cells pouring out into
this cavity the secretions which dissolve the food,
and which it is their special purpose to manufacture.
The formation of a mouth by thinning away of the
wall at one point, will be a manifest advantage, as
it will avoid the necessity of the food particles
having to pass through both layers of cells in order
to reach the digestive cavity ; and on its appear-
ance the gastrula is completed.

The Theory of Natural Selection requires that
each stage in the gradual evolution of a complex
organ or system should be a distinct, if slight,
advance on the stage immediately preceding it ; an
advance so distinct as to confer on its possessor an
appreciable advantage in the struggle for existence.
This condition is often overlooked ; we are apt to
assume, though most erroneously, that if it can be
shown that the ultimate stage is more advantageous
than the initial or earlier condition, then the
whole problem of the evolution of the organ in
question is solved. It is indeed seldom that we
are able to refer to so complete a series of inter-
mediate stages as that given above in the case of
the delaminate gastrula, each step being but a very
slight advance beyond the previous condition, and
yet each step conferring on its possessor a distinct
and tangible advantage. The fact that such a series
of forms can be pointed out, every one of which is
repeated in the life-history of certain jelly-fish, is a

THE SHAPES AND SIZES OF ANIMALS 129

strong argument in favour of the primitive nature
of the delaminate gastrula ; while, as already
noticed, it tells strongly against the claims of its
rival, the invaginate gastrula, that it is at present
not possible to point out the progressive advantage
gained by the successive stages of gradual flattening
and gradual invagination through which the gastrula
stage is acquired. Indeed, we seem here driven to
suppose a much more rapid change to have occurred
than is commonly recognised as possible. Perhaps
the real explanation may be that the delaminate
gastrula is the older form historically, and that the
formation of a gastrula by invagination is merely an
embryological device to save time and facilitate the
course of individual development.

The formation of the mouth, which in the de-
laminate gastrula is the final stage of development,
is an event of first-rate importance, both from the
morphological and physiological standpoints. Now
in the invaginate gastrula a mouth is, by the very
mode of development, present from the first com-
mencement of the process of invagination, and it
may be that the advantage gained by the early
formation of this important organ, which at once
obviates the necessity of the food having to traverse
the ectoderm cells in order to reach the digestive
layer or endoderm, it may be that this advantage
has led to the substitution in actual or individual
development of the invaginate for the historically
delaminate older type of gastrula formation.

Turning from this somewhat lengthy digression
to our more immediate subject, we find that radial

i

130 THE SHAPES AND SIZES OF ANIMALS

symmetry, seen in its most typical form in the
gastrula, is confined to aquatic forms. The reason
for this is the same as in the case of the still more
primitive spherical form — i.e., that it is only in the
case of animals, whether young or adult, which
live immersed in fluid, that the relations between
the animal and the surrounding medium are such
as to allow of the animal having identical relations
to the environment, whichever part of its circum-
ference happens to be uppermost or undermost. It
is also a matter of common observation that radi-
ally symmetrical animals are not merely all aquatic,
but are almost entirely marine. The reason for
the paucity of radially symmetrical forms in the
fresh-water fauna appears to be that they are weak
swimmers, depending for locomotion on the action
of cilia when the animals are of small size, and
having no special locomotive organs when of larger
dimensions. Fresh-water animals, at any rate
such as dwell in rivers and streams, have to be
able to hold their own against the currents in which
they live ; nay more, they must not merely be
able to hold their own but also to make their way
up stream as well as down, or else in the long
run they will be carried slowly but steadily lower
and lower down the river, until ultimately they
become swept out to sea. For this reason weakly
swimming animals cannot, unless under excep-
tional circumstances, establish themselves in fresh
water.

While the typically radial animal is a free swim-
ming form, there are a large number that in the

THE SHAPES AND SIZES OF ANIMALS 131

adult condition are attached either temporarily or
permanently. Of these the common fresh-water
Hydra, and the whole of the great group of Hydro-
zoa, known popularly as zoophytes, are familiar
examples. The majority of these attached radial
animals reproduce by budding, the buds usually
remaining attached to their parent, and so giving
rise to plant-like colonies. In many of these, and
especially in the higher group of Coelenterates, the
Actinozoa, of which sea anemones and corals are
instances, a curious modification of the typical
radial symmetry is manifested, to which the term
biradiate symmetry is commonly applied. In a
biradiate animal, while the radial symmetry is well
preserved, there is superadded to it a further
change in the shape and arrangement of certain of
the internal organs, whereby a definite plane of
symmetry is established, on either side of which
the organs are perfectly similarly and symmetrically
arranged, but which is the only plane by which the
animal can be so divided.

The simplest case of biradiate symmetry would
be of this kind : Imagine a Hydra with the body,
as usual, cylindrical, i.e., circular in transverse
section, and with the mouth also circular in outline;
and, for the sake of simplicity, imagine the tentacles
to be absent. Such an animal has any number of
planes of symmetry, for any plane of division
passing along the whole length of the animal and
along its axis will divide the Hydra into two
perfectly symmetrical halves. Now, imagine the
mouth of our Hydra to become oval or elliptical in

132 THE SHAPES AND SIZES OF ANIMALS

outline instead of circular ; there will now be only
two planes of symmetry, which will divide the
animal into exactly similar and corresponding
halves; one of these planes passing along the
longer diameter of the elliptical mouth, the other
along its shorter diameter. Next imagine the
mouth, instead of being elliptical, to be ovoid or
egg-shaped in outline, with a larger and a smaller
end. There is now only one single plane of
symmetry possible, namely, that passing along the
longer diameter of the mouth opening, for any
other plane will divide the mouth, and therefore the
animal, into two unlike halves.

The origin of this biradiate symmetry is a little
obscure. There are reasons for thinking that it
first arose in colonial forms, such as Alcyonium
or Pennatula, inasmuch as in these colonial forms
it is very well marked, and furthermore the plane
of symmetry of the individual polypes always has
a definite relation to the axis of the entire colony,
and the differences between the two sides of the
animal on which the biradiate symmetry depends
seems to be associated with a special provision for
securing a rapid and efficient circulation of water,
not merely through the individual polypes them-
selves, but throughout the whole colony. This
explanation seems fairly satisfactory in most cases.
It must be noted however that it involves the
descent of solitary forms, such as Cerianthus and
many other Anemones in which biradiate symmetry
is well marked, from colonial ancestors, a line of
ancestry for which there is but little independent

THE SHAPES AND SIZES OF ANIMALS 133

evidence. Much greater difficulty is offered by
the Ctenophora, in which biradiate symmetry is
usually well established, and by some Medusae, in
which the number of tentacles arising from the
margin of the bell may be reduced to two, or even
to a single one ; in these latter cases however
the biradiate symmetry is probably independently
acquired, as theoretically it might readily be by
any radiate animal.

We have next to consider the type of animal
shape spoken of as bilaterally symmetrical, which
must be carefully distinguished from the biradiate
symmetry we have just been describing. In
biradiate symmetry, as in a sea anemone, there is
one divisional plane or plane of symmetry, by
which the animal can be divided into identical
halves ; this plane however concerns the internal
organs only, and has no constant relation to the
movements of the animal. In cases of bilateral
symmetry on the other hand, the animal, as in a
worm, a lobster, or a frog, is divided by a median
vertical plane into symmetrical right and left halves,
while furthermore a distinction may be readily
made between dorsal and ventral surfaces and
between anterior and posterior ends.

Just as the radiate and biradiate shapes are
associated with free-swimming habits, or else with
an attached condition, so is the presence of bilateral
symmetry similarly connected in its earlier phases
with the habit of crawling along the sea-bottom. A
most instructive series of gradations is shown by
the simpler Turbellarian worms, commencing with

I34 THE SHAPES AND SIZES OF ANIMALS

such forms as Anonymus, in which the body is
greatly flattened and almost circular in outline, and
in which the mouth is almost in the centre of the
ventral or oral surface, while a row of eye spots
occurs all round the edge of the animal, though
more thickly set at the two extremities. Starting
from such a form as Anonymus, which may be
compared to a very flat jelly-fish, like Aurelia,
which instead of swimming freely in the water, has
taken to crawling about on the sea-bottom mouth
downwards, we find two diverging series in both
of which the body gradually becomes more and
more elongated and vermiform in shape, while the
sense organs tend to become concentrated at the
anterior end. In one series, that of the Turbel-
laria Acotylea, the mouth gradually moves backwards
as the shape of the body becomes more markedly
oval and elongated ; in the other series, the Tur-
bellaria Cotylea, which receive their name from the
presence of a muscular sucker on the ventral sur-
face, the mouth, starting as in the Acotylea from a
central position, gradually shifts further and further
forwards until it ultimately, in the genus Prosthio-
stomum, becomes placed quite at the anterior end of
the body.

There is little room for doubt that, just as among
epithelial cells the spherical form is the primitive
one from which both the columnar and the squamous
forms have been derived, so also in the series of
Turbellarian worms, those with the body approxi-
mately circular in outline and with a centrally
placed mouth are really the primitive ones from

THE SHAPES AND SIZES OF ANIMALS 135

which the more modified Cotylea and Acotylea
have alike sprung. It is very significant that these
more primitive Turbellarians should present many
points of affinity with the radiate animals, such as
Ccelenterates, not merely in general shape but in
the position of the mouth and central gastral
chamber, the radial arrangement of the diverticula
of the gastral chamber by which nutriment is dis-
tributed to all parts of the animal, the disposition
of the sense organs all round the margin of the
animal, and the position and relations of the ner-
vous system and reproductive organs. These
resemblances are too close and too fundamental to
be accidental, and they lend much support to the
view hinted at above, that bilateral animals are
descended from radiate ancestors, that bilateral
symmetry is something additional to and imposed
on the radiate symmetry, and that this further
modification is a direct consequence of the animals
having exchanged their pelagic free-swimming
habits for crawling ones ; a change that would at
once lead to the establishment of a difference, both
structural and physiological, between the ventral
or oral surface along which locomotion is effected,
and the opposite or dorsal surface ; while the
further differentiation between anterior and posterior
ends would very soon follow as a necessary con-
sequence of this same crawling habit. It is very
possible also that the Turbellarians are themselves
the simplest group in which the crawling habit has
been acquired, and are directly descended from
Ccelenterate ancestors, a view that finds favour

136 THE SHAPES AND SIZES OF ANIMALS

with many zoologists, but which can at present
hardly be regarded as more than an hypothesis.

With regard to the mechanical origin of bilateral
symmetry suggested above, Herbert Spencer,
whose writings on the fundamental laws govern-
ing animal form and structure are of the greatest
possible interest, speaks as follows : " Where the
movements subject the body to different forces at
its two ends, different forces on its under and upper
surfaces, and like forces along its two sides, there
arises a corresponding form, unlike at its ex-
tremities, unlike above and below, but having its
two sides alike."

We have seen above that there are reasons of
very great weight for regarding the radiate type as
more primitive than the bilaterally symmetrical one,
and further than this, for regarding the latter as
directly descended from the former. The group of
Echinodermata, including the starfish, brittle stars,
sea urchins, and their allies, warn us however that
we must not generalise too hastily. About the
radiate symmetry of an Echinoderm there can be no
possible doubt ; a starfish is as markedly, as con-
spicuously, radiate as any animal in existence,
indeed it has been by older writers taken as the
type of radiate animals. Take one of the ordinary
five-fingered starfish for instance : note its shape ;
the central disc-like body produced into five equal
and symmetrically arranged arms, the mouth
placed centrally on the lower, or oral surface, the
anus subcentrally on the dorsal surface. As re-
gards internal structure, the radiate symmetry is

THE SHAPES AND SIZES OF ANIMALS 137

equally well marked ; the muscular, skeletal, diges-
tive, circulatory, nervous, and reproductive systems
all extend radially and symmetrically along the
arms. There are no anterior or posterior ends,
right or left sides to the animal, for in its dilatory
ramblings a starfish moves indifferently in any
direction, any one of its five arms leading.

Bearing in mind what has been said above as to
the primitive nature of radiate symmetry, and of
the relation between it and bilateral symmetry,
bearing in mind also that Echinodermata are not
merely all aquatic, but are exclusively marine, and
that there are the most cogent reasons for regard-
ing the marine fauna as the primitive one, from
which both the fresh-water and terrestrial have
sprung, we should I think naturally conclude that
the radiate symmetry of Echinodermata is primitive,
and that a starfish is a radiate animal, which has
adopted crawling rather than pelagic habits, pre-
sumably for convenience in obtaining food, and in
which consequently a distinction between ventral
and dorsal surfaces has been established ; but that
the further structural modification by which anterior
and posterior ends, right and left sides, become
differentiated, has not yet appeared in it. If we
want further evidence in support of this primitive
character of Echinodermata, we may obtain it from
the past history of the group, for Echinoderms are,
geologically considered, a group of extreme antiquity,
and a group in which the characteristic radiate sym-
metry is as marked in the older as in the more
recent members. If however we consider the

138 THE SHAPES AND SIZES OF ANIMALS

actual embryological development of a starfish or
other Echinoderm, we find difficulties in the way of
the view sketched out above, difficulties which have
not yet been overcome and which may not improb-
ably prove fatal.

A starfish lays small eggs, from which a radially
symmetrical larva, a gastrula, is developed. This
larva however soon acquires a very marked and
unmistakable bilateral symmetry ; ventral and
dorsal surfaces, anterior and posterior ends, right
and left sides may readily be distinguished in it,
and the internal organisation shows the bilateral
symmetry as clearly as does the external shape.
This bilateral symmetry is not confined to starfish,
but is present in the larval stages of all other
Echinoderms as well. The importance of the point
is at once apparent : it shows us that the radiate
symmetry of the adult Echinoderm is not directly
continuous with, and may indeed not be the same
thing as, the radiate symmetry of the early larva,
for between the two radiately symmetrical stages a
bilaterally symmetrical stage is intercalated. If
the developmental history of an Echinoderm is a
true recapitulation of the pedigree of the race, then
the history can only be interpreted as meaning that
Echinoderms are descended from bilaterally sym-
metrical ancestors, and that the radiate symmetry
of the adult Echinoderm is secondary, and of later
origin. The matter is one of great interest, espe-
cially when we bear in mind that the relations of
Echinoderms with other groups of animals are at
present entirely unknown to us, and that conse-

THE SHAPES AND SIZES OF ANIMALS 139

quently any light that can be obtained from a study
of embryology would be peculiarly welcome.

It is impossible to discuss the question at all
adequately here, but it is perhaps worth while
pointing out that though the adult starfish is an
animal showing radiate symmetry in a marked
manner, yet that the radiate symmetry of a starfish,
or indeed of any other Echinoderm, differs in some
important respects from the radiate symmetry of
the Coelenterate. Thus, in the first place, the
symmetry of an Echinoderm is pentamerous, the
typical number of arms in a starfish being five, and
five being the typical number of corresponding
parts met with in the other groups of Echinoderms
also. This may seem an unimportant point, but it
becomes significant when we note that though the
actual number of corresponding parts in a Coelen-
terate varies very greatly in different forms, yet
that it is almost invariably either four or six, or
some multiple of these numbers, while five or any
multiple of five is unknown. Then again, in the
actual development of an Echinoderm the radial
symmetry of the adult is first shown by a set of
organs, the ambulacral system, which is absolutely
and entirely unrepresented in Coelenterates ; and it
is apparently on this radially arranged ambulacral
system that the radial symmetry of other parts and
systems is based. When further we bear in mind
that the whole structure of an Echinoderm is alto-
gether different to that of a Coelenterate, and in
many respects very much more complex, any real
comparison between the two groups becomes very

140 THE SHAPES AND SIZES OF ANIMALS

nearly impossible, and we find it less anomalous than
it at first appeared to regard the adult symmetry
of an Echinoderm as something quite distinct from,
and acquired perfectly independently of, that of a
Coelenterate.

Limits both of space and time forbid that I
should pursue further the discussion of the shapes
of animals. Where once bilateral symmetry is
established however the further modifications seen
in the higher forms become comparatively easy to
follow. The development of a head, with accom-
panying concentration of the nervous system and
sense organs, are merely further developments of
processes and tendencies which we have seen
already established in the Turbellarian worms,
while the formation of limbs, perhaps shadowed
forth in the parapodia of Chaetopods, is the most
important step in the upward progress to the
highest groups of animals.

Bilateral symmetry is characteristic of all the
higher groups, though it may be masked or modi-
fied by further development, as the twisting of the
body of a snail or other gastropod, the asymmetrical
form of the tail of a hermit crab, or the shifting of
the eye in a sole. Speaking generally we may say
that the forms of the higher animals are derived from
those of the lower bilaterally symmetrical worms,
by exaggerating the differences between one part of
the body and another already present in these latter.
Thus the differences between the dorsal and ventral
surfaces or rather halves of the body, and between
the anterior and posterior ends of the body, gradually

THE SHAPES AND SIZES OF ANIMALS 141

become intensified, attaining their maximum in birds
and mammals, the two highest groups of animals.

I can hardly conclude this part of my address
more fittingly than by the following quotation from
Herbert Spencer who, in his " Principles of Biology,"
has discussed in most philosophical fashion, and
with far greater thoroughness than I can pretend to
here, the laws regulating the shapes of animals. "The
one ultimate principle," says Spencer, " that in any
organism equal amounts of growth take place in
those directions in which the incident forces are
equal, serves as a key to the phenomena of mor-
phological differentiation. By it we are furnished
with interpretations of those likenesses and un-
likenesses of parts which are exhibited in the
several kinds of symmetry ; and when we take into
account inherited effects wrought under ancestral
conditions, contrasted in various ways with present
conditions, we are enabled to comprehend, in a
general way, the actions by which animals have
been moulded into the shapes they possess."

Passing from the consideration of the shapes to
that of the sizes of animals, is very like turning
from a well-made road into a ploughed field, across
which progression becomes not only slow but
difficult and irregular. Hitherto the problems
concerned with the sizes or magnitudes of animals
have received but very scant attention, and we are
not only ignorant of the principles and laws that
govern them but of the directions in which to
seek for these principles. Indeed we have at
present but a very limited number or range of

142 THE SHAPES AND SIZES OF ANIMALS

facts on which to base our arguments. Still the
questions are of much interest, and it is certainly
worth while enquiring into them, even though we
may not be able to make much progress to-night.

In the natural or wild state the size of each
kind of animal in the adult condition is fairly
well defined, and often very sharply so. The
words cat, rabbit, sparrow, convey to our minds
the impression of animals, not merely of certain
appearances, habits, and structure, but also of a
certain well understood and fairly constant size.
The limits of variability are much wider in some
cases than in others. Speaking generally they are
much wider in the case of aquatic, and especially
of marine, than of terrestrial animals. If we say
of an animal that it is as large as a fox, we know
fairly exactly what is meant ; but to speak of
anything being as large as a salmon, would convey
a very vague notion of magnitude. As a standard
of size, the salmon is indeed but little better than
the traditional " lump of chalk." The same is
true of most other fish, and indeed is characteristic
of aquatic animals in general. The explanation
seems to be that in terrestrial animals the period of
growth is practically limited to the earlier stages of
existence ; while aquatic animals continue to grow
for a much longer period, or indeed throughout
their entire lives. Why an animal should stop
growing on reaching a given size is a very difficult
question to answer, but one that, if time permits,
I will return to later on.

As regards the actual dimensions attained, here

THE SHAPES AND SIZES OF ANIMALS 143

again the aquatic animals lead the way. Of all
animals now existing, whales are incomparably the
largest ; next to the aquatic come the terrestrial
forms, with the elephant in the forefront ; while
last and smallest of all, come the aerial, or flying
animals. The actual size seems here to be
associated with the density of the medium in which
the animal lives. In water an animal has to
support but a very small part of its weight by its
own muscular effort, for more than half the weight
of a fish or whale is water, and of the solid com-
ponents the fats are lighter than water, so that
the specific gravity of such an animal is not much
in excess of the water in which it dwells. Flying
animals live under very different circumstances, for
here every part of the body is considerably
heavier than the air and great muscular efforts
are necessary to sustain the animal during flight.
Large size or great weight of the body becomes
therefore impossible.

Again, we may lay it down as a general rule
that the largest animals belong to the higher, or
even the highest groups, i.e., that great size is
associated with great complexity of organisation.
Exceptions are readily met with, but on the whole
the statement is correct. Vertebrates are clearly
on the whole of much larger size than any group
of Invertebrates. Amongst Vertebrates, mammals
rank as the highest group and to them belong the
largest animals now living, both aquatic and
terrestrial. So also amongst Invertebrates ; in
the important group of Mollusca, the highest

144 THE SHAPES AND SIZES OF ANIMALS

forms are undoubtedly the Cephalopoda, and it is
amongst these that the largest members of the
group occur.

It has been suggested recently that the rule is a
stricter one than has been hitherto recognised ;
and evidence has been quoted in support of the
further statement that the ancestral forms or
progenitors of the higher groups, such as mammals
or birds, were of distinctly small size, i.e., much
below the average stature attained by their
descendants — the present members of the groups
in question. The direct influence of size on
structure has as yet been but very imperfectly
investigated, but there are many cases known in
which among animals of the same zoological group
the larger forms are distinctly of more complicated
structure than their smaller allies, and in which
there are very valid grounds for holding that the
greater complication of structure is connected
casually with the increased dimensions.

We shall perhaps best deal with the several
points just mentioned by taking the large groups
into which the animal kingdom is divided one by
one, and noticing with regard to each, the principal
facts concerning the sizes of the several members of
the group.

Protozoa, the simplest of all animals, are defined
as those forms which, not merely in the earliest
phases of their existence, but throughout their
entire lives, remain in the condition of single
cells. This unicellular nature is associated with
great simplicity of organisation and with extremely

THE SHAPES AND SIZES OF ANIMALS 145

small size. The differentiation of parts or organs
within a single cell can only proceed up to certain
limits, and if of great bulk parts of the cell would
be too far removed from the surface to obtain
nourishment or to get rid of excretory matter.
What the actual limits of size are among unicellular
animals we perhaps do not know accurately.
Stentor, which may attain a length of •£$ inch, is
usually regarded as one of the largest. Individual
cells in higher animals may however attain much
larger dimensions : thus, if histologists are right in
regarding a single muscle fibre as formed by
differentiation within a single cell, then we must
grant that a cell may be some inches in length ;
while, if the statement be true that a nerve fibre
is merely a process of a single cell, it will follow
that a single cell may in a man extend from the
spinal cord, perhaps from the brain itself, to the
extremity of the fingers or toes — i.e., may attain a
length of some feet, or in an animal the size of a
whale as much as 60 feet or more. Such nerve
fibres are however exceedingly slender.

Among the largest cells known to us are the
eggs of reptiles, of Elasmobranch fishes, and,
largest of all, those of birds. Embryology shows
us that the yolk of a bird's egg, which is the only
part from which the embryo is directly developed,
is morphologically a single cell, and is indeed in
the earlier stages of its formation indistinguishable
from the ordinary epithelial cells covering the
surface of the ovary. The largest eggs are those
laid by the Struthious birds, and the yolk of the

K

I46 THE SHAPES AND SIZES OF ANIMALS

egg of an ostrich is perhaps the largest individual
cell known to us, though it was very greatly
exceeded in recent geological times by that of the
gigantic ^Epyornis. It must however be remem-
bered that the large size of the yolk of an egg is
due mainly to its distension by the granules of
food-yolk imbedded in it, which cannot properly be
regarded as part of the living substance of the
cell.

In the next group — that of the Sponges — it is
difficult to speak of the size of the individual
animals, owing to the very general habit of forming
colonies by continuous gemmation, in which
colonies the outlines of the component individuals
are impossible to determine. The solitary sponges
are usually of small size, not exceeding an inch or
two in height, but some forms, as the beautiful
Euplectella, attain a height of a foot or more.

Among Ccelenterates the same habit of colony
formation prevails, but the boundaries and hence
the size of the component members of the colony
can almost always be determined. The majority of
Ccelenterates are of small size : the individual
zooids of a hydroid colony are commonly but a
fraction of an inch, rarely much over an inch in
length, though occasionally attaining a larger size.
The entire colonies often reach great dimensions ;
the branching zooids may form brush-like masses
some feet across, while the reef-building corals
reach a far larger size.

Speaking generally the Anthozoa, which are the
higher group of the two, are of larger size than the

THE SHAPES AND SIZES OF ANIMALS 147

Hydrozoa ; and it is also among the more highly
organised groups that the largest individuals, the
giant Coelenterates, are met with. Thus, of the
two groups of jelly-fish, the Hydromedusae, which
belong to the Hydrozoa, are small, while the
Scyphomedusse, which in many points of structure
and also in some peculiarities of development are
more nearly allied to the Anthozoa, are of much
greater average size and occasionally reach ex-
traordinary dimensions, individuals measuring as
much as seven feet in diameter with tentacles
nearly fifty feet in length having been met with.
Sea anemones again, the typical Anthozoa, are
much larger than the hydroids ; a height of three
or four inches with a diameter of one or two
inches is not at all uncommon, while from tropical
seas anemones have been described over three feet
in diameter.

The Echinodermata are as a rule of small size
and are measured by inches. The vermiform
Holothurians are sometimes greatly elongated, some
species of Synapta measuring five feet or more in
length, while the extinct Crinoids had stems up to
seventy feet long.

Among the heterogeneous and in many respects
unnatural assemblage of animals grouped together
by zoologists under the name Vermes, the average
dimensions are distinctly small. Turbellarian
worms, perhaps the most primitive members of the
group, average less than an inch in length, though
some forms may measure four to six inches.
Trematodes are similarly small ; and the great

148 THE SHAPES AND SIZES OF ANIMALS

length attained by some Tapeworms, thirty feet or
more, is due rather to their peculiar mode of a
sexual reproduction than to increase in individual
dimensions.

Nemertines may attain an enormous length,
fifteen or twenty feet being not uncommonly
exceeded. Amongst Annelids the average length
is certainly below six inches ; but individual genera,
as Halla, often exceed three feet, and the giant
earthworm of Australia measures six feet or more.

Rotifers are a very interesting group ; invariably
of small size, and often actually microscopic, they
yet exhibit very great complexity of organisation.
A Rotifer may be actually smaller than a Stentor,
and yet while the latter is but a single cell exhibit-
ing but very slight differentiation of parts, the
former is an animal with well developed cutaneous
and muscular systems, a large body cavity, an ali-
mentary canal in which the various regions are
modified in special manner for special purposes, a
definite system of excretory organs, a very perfect
nervous system, with well developed and sometimes
highly specialised sense organs ; and all these
various parts formed of specially differentiated cells.
Comparison of a Rotifer with a Protozoon shows
us very forcibly that although it may be true that
larger animals are, on the whole, more highly
organised than smaller ones, yet it is not magnitude
alone that determines structure.

Arthropoda, though an extraordinarily numerous
group, far exceeding in number of species all the
other groups put together, yet do not present any

THE SHAPES AND SIZES OF ANIMALS 149

unusually great diversity of size. The majority of
arthropods are small, and a length of four inches is
a long way above the average, while many large
and very numerous groups such as Entomostraca
are minute, indeed almost microscopic, in their
dimensions. Crabs may reach a large size, but
this is almost entirely due to great elongation of
their legs.

Amongst Mollusca a much larger size may be
attained. Of Lamellibranchs the genus Tridacna
affords the giants, shells of these huge clams having
been found measuring two feet across and weighing
upwards of five hundred pounds, the animal itself
exceeding twenty pounds in weight. Gastropods
also reach a large size, but are far exceeded by the
Cephalopods, or cuttle-fish, which group yields the
largest invertebrates known to exist.

From time to time the bodies of enormous cuttle-
fish are cast up on the shores in various parts of
the world. Two or three very large specimens
have occurred on the West Coast of Ireland, others
are described from New Zealand, but by far the
largest and also the most numerous cases are from
the East Coast of North America, more especially
the Coast of Newfoundland. Some of the speci-
mens are of really gigantic proportions, a length
of body of as much as twenty feet having been
measured, while the long arms, or tentacles as
they are often named, were thirty or even forty
feet in length.

The Gastropods yield some very instructive series
of forms, illustrating in a remarkable way the influ-

150 THE SHAPES AND SIZES OF ANIMALS

ence of size on structure. Of these one of the
most interesting is that afforded by the genera
Aplysia, Doris, Eolis, and Pontolimax, all four of
which belong to the group of Opisthobranchiate
Gastropods, or sea slugs. Aplysia, the sea hare,
is an animal of some size, four or five inches in
length, and has a very well-developed gill covered
over and protected by a thin uncalcified shield-like
shell. Doris is of smaller size, has no shell, and
has its gill processes less developed than in Aplysia,
but still grouped together in a tuft on the hinder
part of its back. In Eolis, there are a number of
elongated papillae covering the back of the animal,
some of which, at any rate, act as respiratory
organs. Pontolimax, finally, is a very small slug-
like animal only a twelfth of an inch in length, and
having a perfectly smooth dorsal surface devoid of
respiratory processes of any kind.

Herbert Spencer has pointed out very clearly
how in two animals of the same zoological type,
but of markedly different size, the smaller one may
be able to do without respiratory organs, while to
the larger one, merely in consequence of its larger
size, such organs are absolutely essential. Suppose
you have two animals of identical shape, but of
different size : for the sake of simplicity let us
suppose the animals to be spherical, and let the
diameter of the smaller animal be one inch, that of
the larger two inches. The extent of surface in
the two will be proportionate to the squares of their
linear dimensions — i.e.t in this case, the larger animal
will have a surface four times as extensive as the

THE SHAPES AND SIZES OF ANIMALS 151

smaller one. But the mass of bulk of the animals
will be proportionate to the cubes of their linear
dimensions — i.e., the larger animal will be eight
times the bulk of the smaller one. Hence the
larger animal, with a bulk or mass eight times that
of the smaller one, will have a surface only four
times as extensive. This fact, that as animals
increase in size their bulk or mass increases at a
much faster ratio than their surface, explains how it
is that a small animal, such as Pontolimax, may
find its body surface sufficient for the interchange
of gases that constitutes respiration, while hi a
larger animal Doris, of similar shape and constitu-
tion, the body surface may be altogether insufficient
and special respiratory organs may become neces-
sary. Again, in a small animal no part of the
body is very far removed from either the surface or
from the digestive organs, hence each part is able
to effect its respiratory interchange of gases and to
obtain its due supply of nutriment directly, without
the intermediation of a system of circulatory
organs or blood-vessels. In a larger animal how-
ever such blood-vessels are absolutely necessary,
for without them the more deeply lying parts would
be unable to obtain the oxygen which is necessary
for their vital activity, or to get rid of the carbonic
acid and other poisonous products of that activity ;
while the nutrition of the surface and more remote
parts of the body could not be properly kept up
unless there were direct communication between
the digestive organs and these parts. Such con-
siderations must suffice to illustrate in what way

152 THE SHAPES AND SIZES OF ANIMALS

mere increase of size may involve and necessitate
greater complexity of structure.

Amongst Vertebrates we have already noticed
the extreme individual variability of size seen
amongst fish, in which apparently growth continues
throughout the whole period of life. The actual
limits of size of fish are probably only imperfectly
known to us. The blue shark, Carcharias, attains
a length of twenty-five feet ; specimens of Car-
charadon have been measured over forty feet in
length, while of the genus Rhinodon examples of
fifty, sixty, or even seventy feet in length have
been described. This is very probably the limit
of size reached by fish at the present day, but
judging from the fossil teeth of Carcharodon,
Cestracion, and other forms, sharks of these genera
must have existed in tertiary times more than
twice the dimensions of any now living.

It is difficult to speak with any certainty about
the size of Amphibians, but the existing genera
are all small. Some of the Salamanders attain a
length of four feet or more, but the majority of
recent Amphibians are of much smaller size.
Fossil forms occur of much larger dimensions,
some, such as the Labyrinthodonts, being veritable
giants. It is however impossible in most cases
to speak with certainty as to the Amphibian char-
acter of these fossils, but we are probably right
in regarding recent Amphibians as diminutive and
pigmy representatives of a group formerly of much
larger size.

Reptiles are in much the same position, many

THE SHAPES AND SIZES OF ANIMALS 153

of the largest groups being now extinct. Of these
the Plesiosauri and Ichthyosauri attained lengths up
to twenty feet, while the flying Pterodactyls were
also of large size, some measuring as much as
twenty feet in expanse of wing ; but these
dimensions were greatly exceeded by other forms.
It has been estimated that Mosasaurus was as much
as seventy feet in length, while in the genus Bron-
tosaurus, Professor Marsh believes we have a truly
terrestrial reptile eighty feet in length, thirty feet
high, and estimated to have weighed not less than
twenty tons. Of recent reptiles the Crocodiles and
some of the Chelonians attain large dimensions,
while snakes are described up to thirty feet in
length and as thick as a man's body.

It is however amongst Mammals that we find
as already noticed the real giants, and it is in-
teresting to note, that in different geological periods
different groups of Mammals have worked up to a
maximum of size and then disappeared. Thus the
Edentates at one time gave rise to giant forms
culminating in the huge Glyptodon and still larger
Megatherium, but the existing Edentates like ex-
isting Amphibians are a very puny race compared
to their forefathers. Ungulates seem in past times
to have had a greater number of large forms than
at present, though even now they are among
terrestrial animals second only to the Elephants.
It is interesting also to note that there seems to
be an actual limit to the size of a terrestrial
mammal, which has been approached more than
once by separate groups. It can hardly be an

154 THE SHAPES AND SIZES OF ANIMALS

accident that the Megatherium attained dimensions
very closely comparable to those of an elephant,
while the biggest fossil Ungulates were not far
short of this size.

It has been suggested that the limits of size are
due to the nature of the materials of which animals
are constructed, and that the difficulty in increasing
that size is a mechanical one ; that just as it would
be impossible to construct a Forth Bridge of stone,
for in order to get sufficient strength it would be
necessary to employ so great a quantity of material
that the structure would be crushed by its own
weight, so the bones, ligaments, and muscles of
which the animal frame is constructed will permit
of size up to that of an elephant but not beyond.
It is not hard to find points in the anatomy of the
elephant that support this view. The several
joints of each limb, which in smaller quadrupeds are
in the natural standing position bent on one
another, often at considerable angles, are in the
elephant placed vertically one above another, an
arrangement that enables them to support with
less muscular effort the enormous weight of the
body and head. The massive pillar-like form of
the legs, which take up the greater part of the
space below the body, are also indications that the
limits of size are being approached, and when we
find that though approached closely time after time,
the size of the elephant has not been exceeded in
past times by terrestrial animals, save by a few
members of its own group, and with the possible
but as yet doubtful exception of Brontosaurus and

THE SHAPES AND SIZES OF ANIMALS 155

some other reptiles, we are probably not far wrong
in assuming that this size is somewhere near the
mechanical limit imposed by the strength of the
materials of which the animal frame is composed.

With aquatic animals in which the body is
supported on all sides, and the specific gravity of
the entire animal is not greatly in excess of that
of the water in which it lives, the case is very
different ; and among the whales we meet with
genera which attain the length of ninety feet, and
whose weight is estimated at upwards of one
hundred tons, figures which according to some
authorities may in exceptional cases be consider-
ably exceeded. Whales are not merely the biggest
animals that live, they are probably also the biggest
that ever have lived, for we have no satisfactory
evidence of larger forms at any period in the
world's history.

It is worth while also to note that the giants,
whether aquatic or terrestrial, are doomed to
extinction. Elephants have been tolerated because
of the readiness with which they can be tamed and
made to place their great strength at the service of
man ; but it is many centuries since the African
elephant was so employed, and no one of the
African tribes at the present day makes the
slightest attempt to so use him ; while with regard
to the Indian elephant, though large numbers are
still employed, experts are becoming more and
more doubtful as to the profitable nature of this
mode of labour. Elephants require much attention,
they consume enormous quantities of food, and

156 THE SHAPES AND SIZES OF ANIMALS

they are subject to a variety of ailments, which
more or less incapacitate them for their work, so
that the elephant seems in real danger of being
shouldered out of existence by the steam engine
and hydraulic jack, which are more reliable, easier
to apply, and on the whole less costly. Enormous
numbers of elephants, both African and Indian, are
killed each year for the sake of the ivory of which
their tusks consist ; it is difficult to form anything
like a correct estimate of the actual number, but
competent authorities state that it cannot be less
than 100,000 annually. This terrible destruction is
admittedly thinning their numbers, and elephants are
on all hands recognised as doomed to destruction ;
and with the whales, which are yearly becoming
scarcer, to form the last and greatest victims of a
ruthlessly advancing civilization.

A final problem, that time only permits me to
hint at but which has been keenly discussed by
many writers, is this. Why is it, how does it
come about, that some animals are so much bigger
than others ? To take a concrete case : why is a
cow bigger than a sheep ? The two animals
belong to closely allied groups ; we see them side
by side in our fields, eating the same food and
digesting it in the same characteristic fashion ;
why then should there be so great and constant a
difference in size ? Herbert Spencer, who discusses
the problem at some length, is inclined to explain
it by the size of the animal at birth, and attempts
to establish the position that those animals which
are larger at the time of birth or of hatching are

THE SHAPES AND SIZES OF ANIMALS 157

those which are larger also when adult. It is
true that an ostrich lays a bigger egg than a hen,
and a hen than a sparrow; but it is very easy
to show that the relation is not a general one,
and that size at birth has no necessary relation
to size in the adult condition. For example,
a crayfish and a lobster are two closely allied
animals, and yet the crayfish lays the larger egg
of the two, though the adult is not more than
a quarter the length of a lobster. Or again, were
Spencer's contention right, then an eight months'
child, being born of smaller size than the average,
should not attain to average stature. It is more
probable that the real explanation is a much more
complicated one, and that there is for each animal
a certain average size which is most advantageous
to the animal when living wild in its natural
condition. Natural Selection will then tend to
preserve this average size by placing at a dis-
advantage those individuals which depart from it
conspicuously, whether in the way of excess or
diminution.

In connection with this point, reference must be
made to Mr. Galton's law of Regression, enunciated
in his recently published and most important book
on Natural Inheritance. It is impossible to discuss
the subject further on the present occasion, and I
will conclude by giving the Law in Mr. Galton's
own words, referring my readers for further details
to his most fascinating and suggestive work. " If
the word ' peculiarity ' be used to signify the
difference between the amount of any faculty

158 THE SHAPES AND SIZES OF ANIMALS

possessed by a man, and the average of that
possessed by the population at large, then the Law
of Regression may be described as follows : Each
peculiarity in a man is shared by his kinsmen, but
on the average in a less degree. It is reduced to
a definite fraction of its amount, quite independ-
ently of what its amount might be. The fraction
differs in different orders of kinship, becoming
smaller as they are more remote." In other
words, according to Mr. Galton's Law of Regres-
sion, any tendency to individual deviation — say,
from the normal stature — whether in the way of
excess or of diminution, is counteracted by an
inexorable law, which acting to a definite degree
in each succeeding generation, constrains return to
the normal height.

IX

SOME RECENT DEVELOPMENTS
OF THE CELL THEORY

MY address last year was concerned with big rather
than with little animals, and might perhaps with
greater propriety have been delivered to a macro-
scopical rather than to a microscopical society. For
this however I propose to make amends to-night ;
and to the subject-matter of my address the most
ardent microscopist could hardly take exception,
however legitimate may be his dissatisfaction at the
mode of its presentation. For these more recent de-
velopments of the cell theory not merely depend
absolutely on the use of the microscope, but require
for their elucidation the very highest powers
obtainable, and the most refined methods of modern
histological research. In what I have to say to-
night I can make no claim to originality : my aim is
rather to give a summary of recent progress along
certain important lines of investigation, and a state-
ment of the present position of problems recognised
by all as of the highest interest and importance.

160 SOME RECENT DEVELOPMENTS

A few words concerning the origin and the
earlier phases of the cell theory may fitly preface
my remarks. Like all great theories it is im-
possible to date precisely its first enunciation.
Commonly stated to have been founded in 1839,
first for plants by Schleiden, and almost directly
afterwards extended to animals by Schwann, the
cell theory in its main outlines is to be found
lurking in an incomplete unformed condition,
rather hinted at than clearly expressed, in the
earlier writings of Robert Brown, Dutrochet, Von
Mohl, and others. These earlier and less precise
attempts were however confined to vegetable
tissues. As regards animals, to which I propose
to limit myself this evening — from no desire to
minimise the importance of the sister science of
Botany, but merely because I am unfortunately less
familiar with it and unable to speak from my own
knowledge and observation — as regards animal
tissues, Schwann is rightly regarded as the founder
of the cell theory.

By the cell theory, Schwann meant that the
bodies of all animals, as also of all plants, are
formed of cells : that it is by evolution of, and
changes in these cells, that the tissues of which
the various parts of the body consist are formed,
and further, " that the differences in the properties
of the different tissues and organs of animals and
plants depend on differences in the chemical and
physical activities of the constituent cells." Con-
cerning the mode of formation of cells, Schwann
held that they might arise independently in a

OF THE CELL THEORY u6i

surrounding matrix or blastema. In 1845 how-
ever Goodsir first promulgated the doctrine that
cells never originate without pre-existing cells, from
which they arise by fission ; a view strongly
supported by Remak, Kolliker, and Virchow, and
soon universally accepted. The next great step
was von Baer's discovery of the mammalian egg or
ovum, and his recognition of the fact that it was a
single nucleated cell ; a discovery that threw an
entirely new light on Harvey's dictum, omne vivum ex
ovo.

The cell theory was now firmly established.
Histology, or the microscopical study of the
animal body, showed that all its parts and tissues
were really built up of cells, as a wall is of bricks ;
while embryology furnished even more cogent proof,
showing that the ovum or egg of all animals is one
single cell, and that from this single cell by
repeated division all the component cells of the
adult animal are derived. Thus expressed the
theory is the grandest generalisation and the
most firmly established fact in all morphology ;
and the division of the animal kingdom into Pro-
tozoa and Metazoa — i.e., into animals which on the
one hand remain single cells all their lives, and
on the other hand commence as single cells or
ova but speedily become multicellular, takes rank
as the most fundamental and most natural of all
zoological distinctions.

The cell theory early gained acceptance as
regards its main conclusions, but concerning the
detailed structure of the unit or cell discussions

162 SOME RECENT DEVELOPMENTS

quickly arose, which the progress of knowledge has
served rather to widen than to restrict. Schwann's
idea of a cell was " a vesicle closed by a solid mem-
brane, containing a liquid in which floats a nucleus
enclosing a nucleolus, and in which also one may
find small granular bodies." This definition was
soon challenged. The cell wall was first attacked,
and shown by Nageli in 1845 to be not essential
to a cell. Others speedily confirmed Nageli on this
point, and in 1857 we find Leydig defining a cell
as " a soft substance containing a nucleus." Next
the nucleus received attention. Max Schultze, one
of the most careful of observers, denied its existence
in Amceba porrecta ; other investigators quickly
followed on the same line, and by Haeckel a dis-
tinct sub-kingdom, the Protista, was proposed for
the reception of forms such as Protamceba, Proto-
myxa, and Protomonas, which were supposed to be
devoid of nucleus and were regarded as possibly
representing the parent forms from which animals
and plants have alike descended. Further research
has however tended to check rather than to con-
firm these statements; and the use of special re-
agents and more refined histological methods has
shown that nuclei are present in many forms, such
as the Foraminifera, in which their existence was
at first denied.

The more recent investigations have also shown
that the nucleus is not as was at first supposed a
structure always presenting the same characters,
but that it may vary greatly in form, structure, and
relations, even amongst the simplest animals, or

OF THE CELL THEORY 163

Protozoa. In the young of many forms no nucleus
can be detected, while the adult possesses a con-
spicuous one. Some forms, as Amoeba, have a
single nucleus ; others, as Opalina, or Arcella, have
many nuclei. In Paramcecium there are two
nuclear structures, the nucleus proper and the
paranucleus, which have different properties and
very different functions. Then as to shape, the
nucleus is in most cases spherical or approximately
so ; in Vorticella it is greatly elongated ; in Sten-
tor it is elongated, and furthermore constricted in a
moniliform manner ; and in such forms as Dendro-
soma it presents an extraordinarily branched and
extremely complex condition. A still more curious
state is seen in Opalinopsis, an Infusorian living
parasitically in the liver of the squid. Here the
nucleus appears at times as a much branched reti-
cular structure extending the whole length of the
animal ; at other times the reticular nucleus breaks
up or becomes " pulverised " into an immense
number of extremely minute particles, which be-
come diffused through the protoplasm of the animal,
and would most certainly escape detection but for a
knowledge of their previous condition. These few
illustrations will serve to show how extraordinarily
variable the nucleus may be even amongst the
Protozoa. I shall return later on to the condition
of the nucleus in higher animals, but propose first
to deal with the cell-body itself.

The cell- body or cell contents — i.e., the whole
cell except the cell membrane and the nucleus — were
spoken of by Dujardin as consisting of " sarcode,"

164 SOME RECENT DEVELOPMENTS

which he defined as " a kind of mucus endowed
with spontaneous movement and contractility." His
idea of the physiological activity of the cell-body
was very early recognised as important, and
Schwann himself expressed the conviction that
the study of the cell-body or cell-substance was
essential to a true knowledge of the processes of life.
The term " protoplasm " that has since come into
general use for the cell-substance was first applied
to the cell-contents of plants, and was adopted by
Max Schultze for animals as well, when in 1861
he maintained the fundamental identity of the living
matter of animals and plants, however high or how-
ever low in grade they might be.

In 1868 Huxley, in his celebrated lecture at
Edinburgh on " The Physical Basis of Life," laid
great stress on the fundamental unity of the living
matter of animals and plants, maintaining that it is
right to speak of life as a property of protoplasm ;
and that just as the properties of water — e.g., con-
vertibility under given conditions of temperature
and pressure into steam — result from the nature and
disposition of its component molecules, so do the
properties of protoplasm result from the nature and
disposition of its molecules. This idea, really
originating with Schwann himself, that the riddle,
the mystery of life, was to be solved by study of
the molecular constitution of living matter or proto^
plasm was a great conception and in all probability
a correct one. However for the moment it
checked rather than aided the progress of biological
science ; it was too big a jump. Our knowledge

OF THE CELL THEORY 165

of molecular physics is far too incomplete to enable
us to tackle the problem of life from this side, with
even the remotest prospect of success ; while the
habit which this conception almost necessarily en-
gendered, of speaking of protoplasm as a substance
of a certain chemical constitution, containing so
much carbon, hydrogen, nitrogen, oxygen, phos-
phorus, and so on, as a substance allied to the
albumens, perhaps most akin to white of egg, by
diverting attention from other and more promising
lines of enquiry, threatened to hamper rather than
to further real progress.

The reaction soon came. Physiologists reminded
themselves that it is the essence of protoplasm that
it should be alive, and that their wisest course was
to study directly the phenomena of life as manifested
by it. The more conspicuous of these phenomena,
contractility and irritability, the powers of movement
and of sensation, first attracted attention. More
recently attention has been specially directed to a
further problem, resulting from the consideration
that protoplasm is never idle, never at rest, but is
always wearing itself away, incessantly wasting.
Every living animal or plant wastes — i.e., loses
weight ; and this in all its parts and at all times,
though at unequal rates. The loss of weight must
mean loss of actual matter, and this loss is due
to the breaking down of the body substance into
simpler chemical bodies or excretory matters, of
which carbonic acid and urea are the most im-
portant and the most characteristic.

But this is not all. A dead body also wastes

166 SOME RECENT DEVELOPMENTS

away, and the products of its wasting are much the
same as the excretory matters formed by a living
animal. The difference is that the living animal
or plant has the power of repair, renewal, or
regeneration of its living substance, while the dead
body has no such power. This renewal is effected
at the expense of the food. Food is essential, or
we starve and die; and it has been well said that
hunger is the essential and diagnostic character of
living things. This power of building up living
from non-living matter is peculiar to protoplasm, of
which we have seen that the cell-bodies, whether
of animals or plants, consist. The process is
partly a chemical one, a building up of simpler into
more complex bodies ; but in part it is something
more, something peculiar, the true nature of which
we do not understand. The final touch, the con-
version of the dead food into living brain, muscle,
etc., is effected by a process which we name
assimilation, but of which the modus operandi is at
present absolutely unknown.

This conception of protoplasm or the living
matter of animals and plants, as undergoing in-
cessant change or metabolism as it is termed, is
one of much importance. Living protoplasm has
been compared to a fountain in which the form
remains constant though each component particle
of water is in constant movement. In protoplasm
as in the fountain, we distinguish two main pro-
cesses, an uphill or anabolic process, as the water
rises to the crest of the wave, or the food is being
built up into the living tissues ; and a downhill or

OF THE CELL THEORY 167

katabolic process, as the water falls from the crest
back into the basin, or as the living brain, muscle,
etc., become broken down into the various excretory
products. The up-hill or anabolic processes are
synthetic and require or absorb energy. The
water of the fountain will not rise of itself, but
must be forced above the level of the water of the
basin ; and similarly the digestion and assimilation
of food, the building-up of protoplasm, demand
an expenditure of energy on the part of the
animal.

On the other hand, the downhill or katabolic
processes are analytic and are sources of energy ;
and just as the falling water of the fountain may be
used to drive a wheel or other machine, so in the
living body the katabolic changes, like a series of
explosions, give out energy which as muscular
movement or mental activity or in other ways may
be employed to do the work of the body. And
just as in a fountain, the energy required to raise
the water a given height is equal to the energy
liberated by the water in falling from the same
height ; arid if work is to be done by the fountain,
the water must be allowed to fall to a lower level
than that from which it was raised ; so also in the
living protoplasm of an animal, if work, muscular,
mental, or of other kind is to be performed, the
katabolic fall must be greater than the synthetic
rise ; or in other words the excretory matters
must be of simpler constitution than the food
taken in.

Considerations of this kind which form the basis

168 SOME RECENT DEVELOPMENTS

of modern physiology, lead us to regard the proto-
plasm of which the cell-bodies consist, in rather a
different manner, and from a new standpoint.
Protoplasm may, to adopt the simile given above,
be regarded as the topmost point, the crest of the
physiological wave, consisting of matter in a con-
dition of extremely unstable equilibrium, and varying
greatly in structure and in composition in different
cells, or even in the same cell at different times.
The essential character however of protoplasm is
that it is alive ; and although we are unable to
formulate precisely what we mean by life, there is
at any rate one very definite idea which we asso-
ciate with living matter and with it alone, and this
is the power which living matter possesses of
building itself up, renewing itself from the dead
matter which is taken in as food. To define
protoplasm is difficult, perhaps impossible, but the
one essential thing to remember about it is that
it is alive. To speak of dead protoplasm is a
misnomer.

Starting from this new position attempts were
soon made to determine whether the protoplasm of
which cell-bodies consist possesses any structure,
and more especially whether there are any structural
changes which occur normally in cell-bodies, in
association with vital acts. It was soon found that
the cell-body or cell-substance is not necessarily
homogeneous. In many, perhaps in most cases, a
more or less pronounced reticular structure is
present ; the protoplasm consisting of a network of
firmer strands, the meshes of which are filled with

OF THE CELL THEORY 169

a fluid or gelatinous substance in which are often
present minute particles or granules.

Furthermore the action of reagents of various
kinds shows that the network or reticulum is of
different character, presumably of different composi-
tion, to the substance filling the meshes, and that in
some cases the network itself may vary in different
parts. Examples of such reticular protoplasm are
frequent among Protozoa — as for example in the
genera Noctiluca, Trachelius, and many other Infu-
soria ; while extreme cases of reticulation or vacuo-
lation are seen among the Heliozoa, Radiolaria, and
many of the Foraminifera — e.g., Globigerina, in which
the protoplasm has a characteristic bubbly or frothy
appearance from the great number and size of the
vacuoles. The contents of the meshes are usually
fluid, or semi-fluid, and a study of their behaviour
under various conditions, with a comparison of
the mode of formation of fat cells, suggests strongly
that the reticulum or network is the active part of
the protoplasm, and that the substance filling the
meshes is of secondary importance to it. Of this
view, a most suggestive application has been made
with regard to the structure of striated muscular
fibre, which appears to consist of a reticulum, the
strands of which are arranged in regular geometric
pattern, and the meshes filled with a more fluid
sarcous substance. It has been suggested that the
reticulum is the active agent by contraction of
which shortening of the muscle is produced ; and
that the reticulum of a striated muscle fibre cell is
equivalent to the reticulum of a Protozoan or of any

iyo SOME RECENT DEVELOPMENTS ,

ordinary cell, differing mainly in its more regular
arrangement.

The changes that actually go on within the

protoplasm of the cell-body in different phases of

its activity have been studied most closely, and

with the most definite results in secreting cells,

either from special glands, such as the pancreas or

salivary gland, or with still greater success in the

epithelial cells lining the digestive stomach of

certain insects. These secreting cells have their

outer ends bathed in lymph, which diffuses from

the neighbouring blood-vessels, while their inner

ends form the bounding surface either of the gland

or of the alimentary canal itself. The cells separate

from the lymph certain products which they

modify in various ways, and finally discharge at

their free surface as the special secretion of the

gland. Careful histological examination has shown

that during rest the gland-cell enlarges, the special

secretions accumulating in the meshes of the

reticulum as a fluid or semi-fluid substance, often

granular in appearance. During digestion, this

secreted matter is discharged from the cells at

their free surface, the meshes of the network

becoming more or less completely emptied. As to

the part played by the network itself the evidence

is incomplete, but there is some reason to think

that the secreted substance is formed in part at the

expense of the reticulum. The discharge of the

secretion does not necessarily involve the death of

the cell, which may fill up again, time after time,

with the secreted matter.

OF THE CELL THEORY 171

This very brief and imperfect account must
suffice to indicate the lines along which investiga-
tion is at present advancing, in the attempt to
determine the nature of the processes which go on
within active living protoplasm.

Turning now to the nucleus, we find that an
extraordinary number of minute researches have
been made of late years, more especially with a
view to determine as far as possible the part
played by the nucleus in the acts of fertilisation
and reproduction, in the hope of obtaining some
clue to the attractive but bewildering problem of
heredity. The important part played by the
nucleus in initiating, indeed determining, the act
of cell-division, was first definitely established by
F. E. Schulze, in Amceba polypodia. He describes
the successive changes as follows : The nucleus
first elongates, then becomes dumb-bell shaped ;
then the bridge between the two knobs becomes
thinner and thinner, and finally breaks, so that
there are now two separate nuclei, formed by
division of the single original one. Next, the
body of the Amceba begins to elongate in the
same direction as the nucleus did previously ; then
follow constriction, and finally the division of
the Amceba into two parts, each containing one of
the two nuclei already formed. The entire pro-
cess occupied ten minutes ; division of the nucleus
taking about a minute and a half, and the remain-
ing eight and a half minutes being occupied by
division of the body of the Amceba. A similar
process of direct nuclear division, as it is called,

172 SOME RECENT DEVELOPMENTS

has been described by Ranvier in the leucocytes of
the Axolotl, in which case the process occupied an
hour and a half ; by Waldeyer in Infusoria ; and
since then by many other writers.

Of recent years attention has been more
specially directed to a far more complicated series
of changes, which have been seen by many
observers to occur in the nucleus during division,
and which are spoken of as karyokinesisor mitosis,
or indirect nuclear division, in contrast with the
former or direct mode. To understand these
more complicated processes, we must first describe
the structure of the nucleus as determined by more
recent and more detailed investigations. It now
appears that in the cell-nucleus there may normally
be distinguished four elements : the network or
reticulum, the nucleoli, the nuclear membrane, and
the nucleoplasm.

The network or reticulum consists of finer or
coarser threads, which differ in their arrangement
in different cells. In epithelial cells from a
Chironomus larva, Balbiani found the nuclear
reticulum to be one single complexly coiled thread.
In a large number of other cases it has been
found by Rabl and others that the typical arrange-
ment is as follows : Two kinds of threads may be
distinguished ; thicker primary threads and thinner
secondary ones. The primary threads may be
twenty or more in number : each is folded on
itself so as to form a loop, and the several looped
threads are placed in a symmetrical manner in the
nucleus ; the loops being arranged around one pole

OF THE CELL THEORY 173

of the nucleus, and the free ends of the threads
interlacing at the other pole. The secondary
threads form a fine network, connecting together
the primary threads, and may be so numerous as
to conceal more or less completely the definite
arrangement of these latter. The nucleoli are
spherical bodies of various size, which stain deeply
with the ordinary colouring reagents. It is not
certain whether they are in all cases corresponding
structures. Sometimes they are merely nodes or
local thickenings of the network of threads, while
at other times they appear to be quite independent
of the network. The general tendency at present
is to regard them as non-essential, or at any rate
secondary structures ; and it is held by many that
their purpose is to serve as a store of reserve
matter for the nutrition of the other constituents of
the nucleus.

The nuclear membrane has recently attracted
renewed attention. By many it is regarded merely
as a denser and superficial part of the nuclear
reticulum ; while by others it is held that it is,
at any rate in certain cases, a continuous mem-
brane, though authorities are not agreed as to
whether it belongs really to the nucleus itself, or
is rather to be regarded as a limiting layer of
the cell protoplasm, immediately surrounding the
nucleus.

The nucleoplasm or nuclear sap, which fills up
the whole of the rest of the nucleus, is an
albuminous coagulable liquid. A fine reticulum has
been described in some cases as traversing the

174 SOME RECENT DEVELOPMENTS

nucleoplasm, and apparently independent of the
main nuclear reticulum.

As regards the chemical constitution of the
various parts of the nucleus, very little is as yet
known. The network and the nucleoli take up
stains more readily than the rest of the nucleus :
the substance of which they consist is hence con-
trasted as chromatin with the non-staining nucleo-
plasm. In some cases the threads of the reticulum
appear to consist of rows of minute chromatin
granules arranged in an achromatin basis.

We are now in a position to consider the changes
which occur in the nucleus during the process of
indirect nuclear division or mitosis. The first
change is that the secondary threads of the
reticulum disappear, apparently through being
absorbed into the primary threads. The primary
threads thus become much more conspicuous, and
their arrangement in loops is clearly seen. The
threads forming the loops next become somewhat
shorter and thicker ; they may also by transverse
division become more numerous. There is some
reason for thinking that in cells of the same kind
from the same animal, the number of threads
is constant, but this has not yet been proved to be
a general rule. In the epidermal cells of the
Salamander twenty-four loops are said to be
constantly present. The next stage is an extremely
important one, for the discovery of which we are
indebted to Flemming. Each of the looped threads
splits longitudinally along its whole length into
two parallel threads. The pair of threads formed

OF THE CELL THEORY 175

by the splitting of a single primary thread are

spoken of as " sister-threads." As they are of

exactly equal size, and as each of the primary

threads becomes divided into a pair in this way,

the upshot of the process is that the whole of the

chromatin becomes divided into two precisely equal

portions, a process the full significance of which will

become apparent immediately. About or shortly

after the time of the splitting of the primary threads

into sister-threads, a structure known as the

nuclear spindle appears. This is a fusiform figure,

bounded by a number of exceedingly fine and

feebly staining threads. It is said to be formed

not from the nucleus, but from the cell protoplasm

immediately surrounding the nucleus. At the time

of its first appearance it has no clear relation to the

looped threads of the nucleus, but very soon the

two structures take up definite positions with

regard to each other, the looped threads being

grouped in a ring round the equator of the spindle,

with the loops directed inwards towards the

centre and the free ends outwards. At the same

time the spindle itself becomes more clearly

marked; at its poles are two rounded bodies, the

" pole-bodies " of van Beneden, from which fine

threads radiate outwards in all directions through

the protoplasm of the cell-body. The nuclear

membrane disappears about this time, though what

exactly happens to it is unknown. It was formerly

supposed to give rise, at any rate in part, to the

nuclear spindle, but it is now generally agreed that

the spindle is extra- nuclear and a product of the

176 SOME RECENT DEVELOPMENTS

cell protoplasm alone. The chromatin loops now
begin to move from the equator along the threads
of the spindle, towards its poles. This takes place
in a perfectly regular manner, the sister-threads of
each pair moving towards opposite poles. As the
two sister-threads of each pair are precisely equal,
the two groups of threads around the two poles of
the spindle will contain precisely equal quantities of
chromatin. The two groups separate completely
from each other, and each group becomes arranged
in a manner corresponding to that of the primary
threads in the original nucleus. In this way two
daughter-nuclei are formed by division of the
mother-nucleus : each daughter-nucleus containing
half the chromatin of the mother-nucleus. After
completion of the daughter-nuclei, the protoplasm
of the cell-body divides, precisely as in the process
of direct division, and the division of the mother-
cell into two daughter-cells is complete.

The above description of the phenomena of
mitosis or indirect nuclear division, represents the
generally accepted view as to the sequence of
events, but several points in it are still very
imperfectly understood. One of the most important
of these is the origin of the nuclear spindle and
its pole-bodies. Van Beneden, whose long-
continued and extraordinarily careful observations
entitle him to speak with great authority, maintains
that the pole-bodies are of primary importance.
He says that the pole-bodies appear before the
spindle ; that they arise distinctly in the cell
protoplasm, quite independently of the nucleus ;

OF THE CELL THEORY 177

and that they appear to be centres from which the
spindle threads diverge : the spindle threads them-
selves he describes as " apparently muscular," and
as governing the movements of the chromatin
threads. Van Beneden would therefore regard the
pole-bodies as the most important structures, and
as really determining cell division rather than the
nuclei. Other observers however are not yet
prepared to accept this view, but regard the nuclei
as the determining agents in all cases of cell divi-
sion.

A still more important problem is as to whether
the direct and indirect processes of nuclear division
are entirely independent, or whether they are not
really related to each other, and if so in what
manner. On this point very keen discussion has
been and is still being carried on. As regards
the relative frequency of the two processes, direct
nuclear division has only been seen in comparatively
few cases : in leucocytes by Rabl and Flemming ;
in the intestinal epithelium of Crustacea by
Frenzel ; in regenerating tissues by Fraisse ; in
the Infusorian Euplotes by Mobius ; in the early
stage of spermatozoon formation in Amphibia, and
in a few other cases. On the other hand indirect
nuclear division, or mitosis, has been seen in both
animal and vegetable cells of almost every kind,
and in processes both normal and pathological, and
would certainly appear to be by far the more usual
method.

With reference to a possible relation between
the two methods, Waldeyer has strenuously main-

M

i;8 SOME RECENT DEVELOPMENTS

tained that there is one fundamental form of nuclear
division, presenting a series of gradations from the
simple direct method described by Remak up to
the extreme complication seen in typical cases of
mitosis ; and in favour of this view there is a con-
siderable amount of evidence already accumulated.
Thus Waldeyer himself has shown that if a frog's
epidermis, in which mitosis usually occurs, be
treated with silver nitrate, direct nuclear division
can alone be demonstrated, from which he argues
with great justice that the distinction between the
two methods may in other cases also be only
apparent, and due to the particular mode of histo-
logical treatment adopted. Pfitzner has succeeded
in staining both the nucleoplasm and chromatin
simultaneously, and maintains that mitosis concerns
not merely the chromatin but the entire nucleus ;
and further that the cell protoplasm takes no part
in the process. Btitschli, Hertwig, Schewiakoff,
and others have shown that in Protozoa, in which
all the phenomena of mitosis occur, the nuclear
membrane may remain complete the whole time
until the final division of the nucleus into the two
daughter-nuclei ; while Boveri and others have
found that in segmenting eggs of various animals
the various stages of mitosis may sometimes be
very clear, and at others very obscure. Finally,
Carnoy asserts that mitosis, so far from being a
uniform process in all cases, presents so many
varieties that no one series can be viewed as essen-
tial ; he even denies that the longitudinal splitting
of the chromatin threads into sister-threads is uni-

OF THE CELL THEORY 179

versal : a point on which Waldeyer differs from
him emphatically.

On the whole it would appear that the balance
of evidence is against the existence of a sharp
distinction between the direct and indirect methods
of nuclear division ; and in favour of there being
one common form, subject to considerable variation
in detail. Much however yet remains to be done
before the question can be regarded as settled.
We are at present absolutely in the dark as to the
real meaning and significance of the chromatin
threads ; and it is especially important that the true
relation of van Beneden's pole-bodies and of the
nuclear spindle should be clearly determined, for
until this is done, no theory of nuclear division can
be held to be definitely established. Perhaps the
most striking points brought to light by the recent
researches on nuclear division are the extreme
complication of a process formerly regarded as a
perfectly simple one, and the mathematical precision
with which the division of the chromatin into equal
halves is effected in typical cases.

One of the most interesting problems arising
from the advance of our knowledge concerning the
minute changes that occur during nuclear division,
is the relation of the phenomena of mitosis to the
act of fertilisation of the egg. By the older writers
it was supposed that either prior to, or as a con-
sequence of the act of fertilisation, the nucleus of
the egg, or germinal vesicle as it is commonly
termed, disappears completely. It is now however
known that part of the egg-nucleus persists and

i8o SOME RECENT DEVELOPMENTS

fuses with part of the spermatozoon, this fusion
constituting the act of fertilisation. Auerbach was
the first to show, in 1874, that in the egg of
Ascaris two nuclei may be seen shortly after the
spermatozoa reach the egg, and that these two
nuclei fuse together. In 1875 Oscar Hertwig
stated, as the result of a careful series of observa-
tions on the eggs of Echini, that one of the two
nuclei seen by Auerbach is the head of the sper-
matozoon, and that the other is part of the nucleus
or germinal vesicle of the egg, probably the nucleolus
or germinal spot. In 1875 and 1876 van Beneden
saw the fusion of two nuclei, or pronuclei as he
named them, in the ovum of the rabbit, as a result
of fertilisation ; one of these pronuclei he found to
arise from the egg nucleus, though not as Hertwig
had supposed from the nucleolus: the other, or
male pronucleus, van Beneden recognised as in
some way connected with the spermatozoon, though
he failed to trace it directly to the head of the
spermatozoon as Hertwig had done in the sea
urchins. These discoveries naturally attracted
great attention, as they for the first time rendered
possible a theory of fertilisation based on the
changes actually known to occur during the pro-
cess. A number of other investigators, prominent
among them being Fol, Greeff, Selenka, Flemming,
Hensen, and Boveri, studied the phenomena in
various groups of animals and definitely established
the fact that the essence of the act of fertilisation
consists in fusion of part of the egg nucleus with
the head or nucleus of the spermatozoon — i.e.t

OF THE CELL THEORY 181

fusion of the male and female pronuclei. Con-
cerning the details of this fusion, and especially
its relations to the processes of mitosis or karyo-
kinesis described above, many elaborate and careful
investigations have been made of late years which
have brought to light points of extraordinary
interest.

Before describing these in detail it is necessary
to say a few words concerning a phenomenon
closely connected with fertilisation, though no part
of the actual process itself; I mean the extrusion
from the egg, prior to fertilisation and independent
of it, of the so-called polar bodies. Carus first
noticed, in 1828, minute globules on the surface of
an egg which took no part in the formation of the
embryo. These were described carefully by
Friedrich Muller in 1848 as seen by him in the
snail's egg, and were called by him directive cor-
puscles because of the constant relation they
appeared to have to the first plane of segmentation ;
he showed clearly that they were derived from the
egg itself. Hensen was the first to show that the
extrusion of the directive corpuscles or polar bodies
occurred independently of fertilisation, for he found
that in the rabbit and guinea-pig the polar bodies
were formed while the eggs were still in the
Graafian follicles, before their discharge from the
ovary. This has been confirmed by many other
investigators, who have shown that as a rule two
polar bodies are extruded in succession ; that both
are usually extruded previous to fertilisation ; but
that in some cases as in the lamprey and the

182 SOME RECENT DEVELOPMENTS

frog, the first polar body may be extruded prior to
fertilisation, and the second after or during the
process ; while in some cases, as in Ascaris,
according to van Beneden, both polar bodies are
extruded after the entrance of the spermatozoon.
In 1875 Btitschli showed that in Nematodes the
extrusion of polar bodies is accompanied by the
formation of a nuclear spindle, and by other
changes similar to those seen in indirect nuclear
division. He in consequence suggested the view
that the formation of a polar body is really a
division of the egg cell into two very unequal
portions, and is accompanied by the ordinary
nuclear changes seen in typical cases of mitosis.
This view has since been widely adopted, though
authorities are not yet agreed as to whether any
part of the protoplasm of the egg is extruded with
the daughter nucleus in the polar body. The point
is of considerable interest, for on one view the
whole process would be one of cell division, on
the other merely nuclear division with extrusion
of one of the daughter nuclei. The balance of
opinion inclines strongly towards the former of
these views, but the point cannot be considered
definitely settled as yet.

The changes that occur in the egg nucleus
during the formation of polar bodies have been
recently studied with great care by many observers.
In the frog's egg Oscar Schultze has shown that
at the time of ripening of the egg the nucleus
undergoes important changes. Previous to this
time it has been of very large size, as much as a

OF THE CELL THEORY 183

third the diameter of the egg itself, and consists of
a well-marked nuclear membrane enclosing fluid
nucleoplasm in which float a number of chromatin
globules of varying size and apparently unconnected
with one another. At the time of ripening of the
egg, shortly before it leaves the ovary, the nuclear
membrane becomes crumpled and indistinct ; the
whole nucleus shrivels very greatly, its fluid con-
tents being diffused through the egg; the majority
of the chromatin globules disappear, but a group of
very small ones near the centre persist ; these unite
together to form a convoluted varicose chromatin
thread which soon breaks up into loops ; then a
nuclear spindle appears with which the loops soon
become connected as in ordinary mitosis. The
spindle with the associated loops, which together
are of exceedingly small size, move to the surface
of the ovum ; the loops and the spindle now divide
in the ordinary way into two equal halves which
become the daughter nuclei. One of the daughter
nuclei is extruded as the first polar body, while the
other retreats into the interior of the egg. The
formation of the second polar body is a repetition
of the process by which the first was formed. As
the amount of chromatin is halved at each nuclear
division, the female pronucleus, which is the
portion of the egg nucleus left after the formation
of the second polar body, will contain exactly one-
fourth of the amount of chromatin present in the
egg nucleus at the formation of the nuclear spindle.
Van Beneden's account of the process in Ascaris
agrees exactly with this. At each division of the

184 SOME RECENT DEVELOPMENTS

nucleus to form a polar body the chromatin, which
is present rather in the form of spherules than of
threads, is exactly halved in amount.

Concerning the fusion of the male and female pro-
nuclei in the act of fertilisation some points of great
interest have recently been brought to light. In some
cases, as shown by Nussbaum in Ascaris, a direct
fusion of the two pronuclei is described as occurring,
while in other forms and by other observers the
process is said to be of a more complicated char-
acter. In the Nematodes, genus Leptodera, Nussbaum
says that the two pronuclei, male and female, take
up a position parallel to the long axis of the egg,
which is ovoid in shape, and then fuse together
lengthways. The first segmentation plane is a
longitudinal one and passes along the axis of the
fused pronuclei, so that each of the two cells
formed by the first cleft contains one-half of the
male pronucleus and one-half of the female pro-
nucleus. Inasmuch as all the cells of the body of
the adult animal are derived by division from the
two primary ones, it follows, as Nussbaum points
out, that if this equal division of male and female
nuclear elements obtains in the later stages of cell
division, each cell of the adult animal will possess
a nucleus, one half of which is derived from the
father and the other half from the mother.

This suggestion, the bearing of which on theories
of heredity is of the greatest importance, has
received most striking confirmation from the
extraordinarily minute observations of van Beneden
on the eggs of Ascaris. These investigations,

OF THE CELL THEORY 185

made in the years 1883 and 1884, rank deservedly
among the most remarkable of microscopical
triumphs. Van Beneden finds that after extrusion
of the polar bodies, and entrance of the sperma-
tozoon into the egg, the two pronuclei, male and
female, which are precisely equal in all respects,
come very close together but do not fuse directly.
Each pronucleus bears at first a single much convo-
luted and varicose thread of chromatin, which soon
divides transversely into two, each of which
becomes bent into a U-shaped loop. There are
then four loops in all, two male and two female.
Each loop now splits longitudinally into two sister
threads. A spindle figure with pole-bodies and
polar rays at its apices now appears, and the out-
lines of the pronuclei, previously distinct, disappear.
The chromatin loops, of which owing to the longi-
tudinal splitting there are now eight, four male
and four female, take up a position at the equator
of the spindle. The two sister threads of each
pair now separate, one moving towards one pole
of the spindle the other towards the opposite pole;
so that at each pole of the spindle there is a group
consisting of two male threads and two female
threads. Each group now forms a daughter
nucleus, and then the entire egg divides into the
first two segmentation cells. This account agrees
with Nussbaum's as regards the equal division of
male and female threads between the first two
segmentation cells ; but differs inasmuch as
according to Nussbaum, the two pronuclei fuse
directly, while, according to van Beneden, there is

i86 SOME RECENT DEVELOPMENTS

no fusion of the pronuclei, but merely a re-arrange-
ment of the chromatin threads. Zacharias and
Hertwig have suggested that fusion really does
occur but may have been overlooked, but for the
present the point cannot be considered settled. It
is possible indeed that the details are not the
same in all cases.

The equal division of male and female elements
in the first segmentation, a point in which Nuss-
baum and van Beneden agree and which has been
confirmed by others, is of the highest possible
interest and importance. If it should prove to
occur in the later as well as in the earliest stages
of development, then, as pointed out above, it will
follow that the nucleus of each cell in the body of
the adult animal will contain male and female
elements derived from the male and female pro-
nuclei — *'.*., from the father and mother — in precisely
equal amounts. In other words, each cell of the
adult body may be spoken of as hermaphrodite. If
this is true, then the egg, which is merely an epi-
thelial cell indistinguishable in its early stages from
the surrounding cells, must itself be hermaphrodite.
The further suggestion at once presents itself: Is
not the extrusion of the polar bodies a casting out
of the male elements of the egg ?

This is a view which in one form or another
has commended itself to many embryologists.
Balfour first suggested that the extrusion of the
polar bodies was a device for ridding the egg of
its male elements, and so ensuring that cross ferti-
lisation must occur, the advantage of which as

OF THE CELL THEORY 187

regards vigour of offspring is well known. Minot
has independently suggested and strongly supported
the same view. A serious difficulty however has
been raised to this view by Weismann, who points
out that if the male element in the egg is got rid
of completely by extrusion of the polar bodies, then
it becomes very difficult to understand the trans-
mission of male ancestral characters through the
mother. It is well known that children may
inherit peculiarities of their grandfather on the
mother's side, and this, according to Weismann,
should not be possible if the extrusion of the polar
bodies removes all male elements from the egg.
We must suppose that a small remnant is or may
be left behind ; but then, asks Weismann, why
should any be eliminated at all if it is not necesary
that all should be got rid of ?

The matter is very far from a simple one.
Weismann himself has shown that in various
parthenogenetic Crustacea, Polyphemus, Moina,
Daphnia, and others, only one polar body is
extruded ; and Blochmann in 1887 announced
that in Aphis the parthenogenetic eggs extrude one
polar body only, while those that require fertili-
sation extrude two. This seems to indicate that
the two polar bodies may have very different value
in spite of the close similarity in their modes of
formation ; for the extrusion of one polar body
seems to leave the egg still capable of developing
without fertilisation ; while after two polar bodies
are extruded, fertilisation is necessary for develop-
ment to occur.

i88 SOME RECENT DEVELOPMENTS

Weismann's explanation of the process is a
complicated one. Regarding the nucleus as
specially concerned with heredity, he distinguishes
in the nuclear substance two kinds of matter,
which he names histogenic plasma and ancestral
plasma respectively. The former or histogenic
plasma he regards as specially concerned with the
growth, nutrition, and shaping of cells ; while the
second or ancestral plasma is supposed to be
specially concerned with reproduction and heredity.
He then assumes that the histogenic plasma
having completed its work when the egg has
reached maturity, it is convenient to get rid of it,
which is done by extruding it bodily from the egg
as the first polar body. The remaining half of
the nuclear substance is the ancestral plasma ; and
with regard to the second polar body, Weismann
argues that if the whole of the ancestral plasma of
the egg were retained, and were to fuse with the
whole of the ancestral plasma of the spermatozoon,
then the total amount of ancestral plasma would be
doubled at each generation. This he argues
would cause inconvenience, and so the device is
adopted of getting rid of half of this ancestral plasma
in the form of the second polar body.

Apart from its extreme complexity, this theory
of Weismann's is open to grave objection. In the
first place the supposed difference between histo-
genic plasma and ancestral plasma is a pure
assumption, in support of which no direct evidence
has been advanced. Secondly, it is very difficult
to understand why, on Weismann's view, the

OF THE CELL THEORY 189

histogenic plasma should be exactly half the
nucleus. Thirdly, it cannot be overlooked that
the modes of formation of the first and of the
second polar bodies are precisely similar to each
other, and also that they agree precisely with
the changes that occur in the nucleus of an
ordinary epithelial cell during indirect division,
similarities which become very difficult to under-
stand if, as Weismann supposes, the two polar
bodies are of totally different nature, and do not
correspond to the halves of the nucleus of a
dividing epithelial cell. Lastly, it may be noted
that cases have been described in which more than
two polar bodies have been extruded.

On the whole therefore, while acknowledging
the extreme ingenuity of Weismann's theory and
the service which its publication has rendered
by stimulating investigation and discussion, it
cannot be said that the theory is in accordance
with all the facts it seeks to explain, or that it
helps us very materially towards an understanding
of these facts. So far direct observation has
failed to show any difference in structure or in
mode of formation of the two polar bodies, or any
important respects in which their development
differs from the phenomena of ordinary mitosis.
These processes are so complicated and require
such extraordinary care and patience for their
successful investigation, that it is but natural to
assume that they have some deep and far-reaching
significance ; and it is well therefore to remember
that they are in no way specially concerned with

190 SOME RECENT DEVELOPMENTS

the processes of sexual reproduction, but occur as
characteristically in the act of division of an ordinary
epithelial cell as in the formation of a polar body
by a ripe ovum.

It is well also to bear in mind that although
the arrangements and divisions of the chromatin
threads naturally attract special attention, it is by
no means certain that these threads are the main
factors in the act of nuclear division. Indeed the
comparison of the direct method of division with
the indirect method which we have discussed
above, rather suggests that the nucleoplasm or
nuclear substance itself is the really essential
element, and that the chromatin threads need not
even be present, at any rate in a form capable of
being brought into view by the reagents at present
in use.

And so it would appear from these more recent
researches, of which time has only permitted me to
give a brief and most imperfect summary, that the
cell theory, great and important as it is most
undoubtedly, is rather the commencement of a
great movement, a fresh starting-point from which
to begin investigations anew, than a complete
scheme or final explanation ; and the one great
lesson for us to learn is that processes of
apparently the simplest kind are really of an
extremely complicated nature, and will well repay
the most minute and attentive study : for a right
understanding of the changes that occur during the
act of division of an ordinary epithelial cell, and
of the causes determining those changes, would

OF THE CELL THEORY 191

throw most welcome light on the more complicated
processes accompanying the ripening and fertili-
sation of the egg, which microscopists of all
nationalities are at present studying with such
intense earnestness.

ANIMAL PEDIGREES

THERE are few things in which men take greater
pride or from which they derive more solid enjoy-
ment, than in tracing out and publishing to the
world their pedigrees ; and we must acknowledge
that the proceeding itself, and the satisfaction
obtained from it, are entirely legitimate. For we
all have pedigrees ; for two or three generations
each one of us could give his descent, trace his
pedigree, offhand ; and if we fail in attempting to
go further back we know that this is from lack of
knowledge, not of facts.

Would we learn these facts, we know that there
are those whose profession it is to supplement our
deficiencies of memory or of information on these
points, and who are prepared for a sum of half-
a-crown, to provide the enquirer with a duly
attested pedigree dating from the time of the
Conquest. For a guinea a Roman emperor can.be
obtained ; while the avaricious in such matters,
who are prepared to spend a five-pound note, may

ANIMAL PEDIGREES 193

satisfy themselves, and for all we know, truthfully, of
their descent from a Pharaoh of the ipth dynasty.

The mode of construction of such pedigrees, or
genealogical trees, is familiar to us all from our
school days. We begin by ruling a series of
horizontal lines, which we agree shall represent
successive generations. Then, assuming that we

J.S.

v

r\

\i v

\ v

V

DIAGRAM A.

are of those whose aspirations are satisfied by a two-
and-sixpenny ancestry, we commence with the dawn
of respectability in the year 1066, and gradually
trace upwards from this date the line representing
our descent from the selected progenitor.

We indicate in capitals, or in italics, any kings,
or statesmen, or poets, or other eminent people
whose memories we like to think derive renewed
lustre from association with ourselves ; and to

N

194 ANIMAL PEDIGREES

include a sufficient number of these we trace the
side branches of our tree for some distance.
Finally, on the topmost twig, and in largest letters,
we write — John Smith.

We all have such pedigrees, whether we can
declare them or not ; and for the prince and the
pauper they are of equal length, dating back not
merely to the Norman Invasion of England, but
to the first appearance of man on the earth.

A special point with regard to these genealogical
trees, and one to which attention may well be
directed, is their absolute truthfulness. Suppose
for example, you have three young friends, two of
whom, Tom and Dick, are brothers, while Harry,
the third, is their first cousin ; then the diagram,
or genealogical tree, representing their mutual
relationships will be as follows, the horizontal lines
marking successive generations : —

T. D. H.

DIAGRAM B.

Go back one generation, and Tom and Dick's lines
meet, for they are sons of one father. Before
Harry's line joins in it is necessary to go back one
generation further — i.e., to the grandfather of our

ANIMAL PEDIGREES 195

young friends, who was one and the same person
for all three, and who forms the true link or bond
of union between them. If once these relationships
are determined correctly, and the diagram con-
structed aright, then it expresses facts, and facts
only, and can never be disturbed. It matters not
what the occupations or residences of the three
may be ; they may never see one another, never
even suspect one another's existence. It makes
no difference how many people may be living con-
temporaneously, how many may have preceded,
how many may be born in after ages ; this little
bit of history remains untouched and absolutely
true for all time. It is for this reason that gene-
alogy or blood relationship affords the only satis-
factory basis for a classification of men ; and we
shall find that the same considerations apply to the
lower animals as well.

One further matter of a preliminary nature
requires mention. It is customary in preparing
genealogical tables to construct them as is done
above, starting from some one more or less remote
ancestor, and following upwards the branches
representing his descendants, so that the whole
diagram takes the form of an upright tree. It
should be noted however that in a certain sense
the diagram would be more correct if it were
inverted. In tracing back human pedigrees there
is a marked tendency to follow out one special line
of descent, and to concentrate attention on one
particular ancestor from whom we desire to make
our line arise, ignoring the fact that there were

196 ANIMAL PEDIGREES

many other contemporary ancestors from any of
whom our path might with equal truth have com-
menced.

A man has two parents, four grandparents, eight
great-grandparents — i.e., in tracing back his pedi-
gree from the present time the number of his
ancestors in each generation is double those of
the generation that succeeded it in time. This
is graphically expressed in the following diagram :

M.

l\/\ l(\ l(\ /A 7s

Parents.
Grandparents.

Great-grand-
ents.

great-
DlAGRAM C. grandparents.

which takes, as noted above, the form of an inverted,
not an erect, tree. If we allow three generations
to a century there will have been twenty-five
generations between the Norman Invasion and the
present time : so that a man now living may
be descended not merely from one ancestor who
came over to England in 1066, but directly and
equally from over sixteen million ancestors who
lived at or about that date. I say advisedly " may
be descended;" for unless we assume that many of
these ancestors were identical individuals, we shall

ANIMAL PEDIGREES 197

find that the existence of a single man to-day
involves the existence, a thousand years ago, of
over a thousand millions of ancestors; and at the
commencement of the Christian era of nearly
seventy thousand millions of millions of ancestors :
a state of things which would involve serious re-
consideration of the dimensions of the earth.

Genealogical trees, such as I have described, we
are all familiar with. Furthermore we know that
the principles employed in constructing them are
not confined to Smith and Jones and the kings and
queens of England, but apply to the lower animals
as well. The pedigrees of racehorses, and of other
artificially bred animals, such as cattle, sheep, pigs,
dogs, pigeons, poultry, etc., have for many years
past been kept with the most scrupulous care; and
there are men who could tell you in detail the
pedigree of the winner of last year's Derby or Leger,
who would be sorely perplexed if asked for their
own, and would perhaps prefer that the results of
researches on this point should not be made too
public.

We all recognise that the cats and rabbits and
dogs of to-day did not come into existence spon-
taneously, but are descended from the cats, rabbits,
and dogs of preceding generations, decades, or
centuries ; and that the same applies to birds, to
butterflies, to sea anemones, or to any other animals
we like to think of. We have now merely to
enlarge our sphere of action, to widen our boundaries
with regard to such genealogies, and we find
ourselves face to face with the great problem with

198 ANIMAL PEDIGREES

which naturalists are confronted, and which they
are attacking on every side and by all means in
their power.

We recognise that Diagram A represents cor-
rectly the relation between man and man: and we
admit that it is equally true when applied to horses,
to cows, to dogs, or to canaries. In other words,
we acknowledge that the principle on which the
diagram is constructed is true in all cases in
which historical or documentary evidence is forth-
coming.

Can we not go further than this ? Is this
written testimony essential ? Would the facts be
in any way altered if no documentary evidence
were forthcoming ? Do we not agree that all
animals have had pedigrees of this kind ; and is it
not worth enquiring whether we cannot reconstruct,
unravel these pedigrees, even in cases where of
necessity documentary evidence cannot be ob-
tained ?

Again, to return for a moment to the human
argument. So far our horizontal lines have been
used to indicate successive generations, and the
relation, admitted by all, has been that each gene-
ration has sprung from the preceding generation,
and has in its turn given birth to the next succeed-
ing one.

Supposing now that we widen our boundaries,
and agree that the intervals between successive
horizontal lines shall indicate, not generations but
longer intervals, say centuries, the relations will
remain unaltered.

ANIMAL PEDIGREES 199

XIX Century.

XVIII Century.

XVII Century.

XVI Century.

XV Century.

XIV Century.

If for example we fix our attention on the six-
teenth century, we find that the men of that century
did not arise spontaneously, but were the direct
descendants of those of the preceding or fifteenth
century ; furthermore we all admit that from the
men of the sixteenth century those of the seven-
teenth, eighteenth, and nineteenth centuries have
all directly sprung. And what holds good with
regard to the men of the sixteenth century applies
equally well to the horses, the cats, the dogs, the
birds, the butterflies, the starfishes of that time.
Now widen the intervals still further : let them
represent not merely centuries but thousands, tens
of thousands, of years ; let them finally indicate
the great geologic periods, and the argument will
still hold :

Quaternary and
Recent.
Pliocene.

Miocene.

200 ANIMAL PEDIGREES

Eocene.

Cretaceous.

The animals, and for that matter the plants too,
of the Miocene age, did not come into existence
irrespective of pre-existing animals and plants, but
were the direct lineal descendants of the Eocene
animals and plants ; and the forefathers of those of
Pliocene and Recent times.

Kainozoic.

Mesozoic.

Palaeozoic.

So also were the animals of Mesozoic times the
children of those of the Palaeozoic age, and the
parents of the Kainozoic fauna. And so this idea
of continuity of life from its earliest dawn on the
earth, through age after age, down to the present
time, forces itself upon us : an idea involving the
further conception of the evolution of existing
animals from unlike ancestors of former times : an
idea constraining us to admit that animals, like men,
have pedigrees, and that between the animals of all
ages a kinship, a blood relationship, exists.

The recognition of this kinship is the determining
feature of the Natural History of to-day. The
reconstruction of these pedigrees is the great work
of the future : the rewriting of the past histories,
not merely of one or two groups for a limited

ANIMAL PEDIGREES 201

number of generations, but of all animals and for
all time : — a formidable, but an entrancing problem;
and whatever misgivings I may have as to my
power of presenting it aright, no apology is needed
for asking your attention to a consideration of the
means at our disposal for attacking it, of the evi-
dence on which we rely in our attempts to recon-
struct the past histories, to determine the pedigrees
of animals.

On the present occasion, it is not with the
whole evidence, but with one special side of it, that
we shall be concerned, that namely which is
derived from a study of the development of existing
animals. Every one knows that animals in the
earlier stages of their existence differ greatly in
form, in structure, and in habits from the adult
condition ; a lung-breathing frog for example
commences its life as a gill-breathing tadpole ; and
a butterfly passes its infancy and youth as a
caterpillar. It is clear that these developmental
stages, and the order of their occurrence, can be no
mere accidents ; for all the individuals of any
particular species of frog, or of butterfly, pass
through the same series of changes. It is not
however until recent years that naturalists have
realised that each animal is constrained to develop
along definitely determined lines ; and that the suc-
cessive stages in its life-history are forced on an
animal in accordance with a law, the determination
of which ranks as one of the greatest achievements
of biological science.

The doctrine of Descent, or of Evolution, teaches

202 ANIMAL PEDIGREES

us that as individual animals arise, not spon-
taneously, but by direct descent from pre-existing
animals, so also is it with species, with families,
and with larger groups of animals, and so also has
it been for all time ; that as the animals of
succeeding generations are related together, so also
are those of successive geologic periods ; that all
animals living or that have lived are united
together by blood relationship of varying nearness
or remoteness; and that every animal now in
existence has a pedigree stretching back, not
merely for ten or a hundred generations, but
through all geologic time since the dawn of life on
this globe.

The study of Development, in its turn, has
revealed to us that each animal bears the mark of
its ancestry, and is compelled to discover its
parentage in its own development ; that the phases
through which an animal passes in its progress
from the egg to the adult are no accidental freaks,
no mere matters of developmental convenience, but
represent more or less closely, in more or less
modified manner, the successive ancestral stages
through which the present condition has been
acquired. Evolution tells us that each animal has
had a pedigree in the past. Embryology reveals to
us this ancestry, because every animal in its own
development repeats its history, climbs up its own
genealogical tree. Such is the Recapitulation
Theory, hinted at by Agassiz, and suggested more
directly in the writings of Von Baer, but first
clearly enunciated by Fritz Mttller, and since

ANIMAL PEDIGREES 203

elaborated by many, notably by Balfour and Ernst
Haeckel.

A few illustrations from different groups of
animals will best explain the practical bearings of
the theory, and the aid which it affords to the
zoologist in his attempts to reconstruct the pedigrees
of animals ; while these will also serve to illustrate
certain of the difficulties which have arisen in the
attempt to interpret individual development by the
light of past history ; difficulties which I propose to
consider at greater length.

A very simple example of recapitulation is
afforded by the eyes of the sole, flounder, plaice,
turbot, and their allies. These " flat fish " have
their bodies greatly compressed laterally, and the
two surfaces, really the right and left sides of the
animal, unlike, one being white or nearly so, and
the other coloured. The flat fish has two eyes,
but these, in place of being situated as in other
fish one on each side of the head, are both on the
coloured side. The advantage to the fish is clear,
for a flat fish when at rest lies on the sea bottom,
with its white surface downwards and the coloured
one upwards. In such a position an eye situated
on the white surface could be of no use to the fish,
and might even become a source of danger, owing
to its liability to injury from stones or other hard
bodies on the sea bottom.

No one would maintain that flat fish were
specially created as such. The totality of their
organisation shows clearly enough that they are
true fish, akin to others in which the eyes are

204 ANIMAL PEDIGREES

symmetrically placed one on each side of the head,
in the position they normally hold among verte-
brates. We must therefore suppose that flat fish
are descended from other fish in which the eyes are
normally situated.

The Recapitulation Theory supplies a ready test ;
and on employing it — i.e., on studying the develop-
ment of the flat fish — we obtain a conclusive answer.
A young flounder or other fish, on leaving the egg,
is shaped just as any ordinary fish, and has the
two eyes placed symmetrically on the right and left
sides of the head. As the young fish increases in
size, the shape gradually approaches that of the
adult ; the body increases in height and becomes
flattened laterally, the median, dorsal, and ventral
fins becoming greatly developed at the same time ;
and the fish now begins to adopt the habit of the
adult of lying on one side on the sea bottom.
Another change occurs : the eye of the side on
which the fish lies, usually the left side in a flounder,
becomes shifted slightly forwards, then rotated on
to the top of the head, and finally twisted completely
over to the opposite or right side.

Crabs differ markedly from their allies, the lob-
sters, in the small size and rudimentary condition
of their abdomen or "tail." Development how-
ever affords abundant evidence of the descent of
crabs from macrurous ancestors. A crab leaves
the egg in what is termed the zoea condition,
possessing a long and clearly jointed abdomen;
and throughout all the earlier stages of existence
the abdomen remains at least as long as the body.

ANIMAL PEDIGREES 205

At the megalopa stage the shape and proportions
are very similar to those of a lobster or other
macrurous decapod. It is only in the last stages
of development when the shape, though not the
size, of the adult crab is attained, that the abdomen
becomes relatively smaller, and is turned forwards
out of sight beneath the hinder part of the thorax.

Molluscs afford excellent illustrations of recapitu-
lation. The typical gasteropod has a large spirally
coiled shell ; the limpet however has a large
conical shell, which in the adult gives no sign of
spiral twisting, although the structure of the animal
shows clearly its affinity to forms with spiral shells.
Development solves the riddle at once, telling us
that in its early stages the limpet embryo has a
spiral shell, which is lost on the formation subse-
quently of the conical shell of the adult.

Recapitulation is not confined to the higher
groups of animals, and the Protozoa themselves
yield most instructive examples. A very striking
case is that of Orbitolites, one of the most complex
of the porcellanous Foraminifera, in which each
individual during its own growth and development
passes through the series of stages by which the
cyclical or discoidal type of shell was derived from
the simpler spiral form.

The fully formed Orbitolite shell is a thin cal-
careous disc. It is hollow, and the central cavity
is divided into chambers by concentric partitions or
septa. These chambers are further subdivided by
incomplete radial partitions. The concentric parti-
tions are perforated by numerous holes, which place

206 ANIMAL PEDIGREES

the chambers in communication with one another ;
and the outermost or marginal chamber communi-
cates with the outer world through a series of holes
round the edge or rim of the disc. During life all
the cavities are filled with a slimy protoplasm, very
similar to that of which an Amoeba consists ;
through the perforations in the septa the protoplasm
of one chamber communicates freely with that of
the neighbouring chambers ; and through the mar-
ginal apertures at the rim of the shell pseudopodia
can be protruded and food captured and digested.

An Orbitolite grows by addition of new chambers
round the margin of the shell. The protoplasm,
becoming too abundant to be contained within the
cavities of the shell, protrudes as a rim all round
the margin of the shell, and by deposition of
calcareous matter gives rise to successive new
chambers to the shell.

The discoidal shape of shell, so characteristic of
Orbitolites, is very unusual amongst Foraminifera,
and it is a matter of great interest to determine in
what way it has been acquired. Bearing in mind
what has just been said as to the mode of growth
of the shell, it is clear that the oldest part — i.e., that
which alone was present in the young animal — is
the central portion, and that the successive concen-
tric rings are younger and younger as we pass
outwards towards the circumference, the marginal
chamber being the youngest and latest formed of
the whole series. Now if we look at the central
or oldest part of an Orbitolite we find that it has
not the concentric arrangement of the peripheral

ANIMAL PEDIGREES 207

part, but is coiled spirally like the majority of
Foraminifera. The central or oldest turns of this
spiral are not chambered ; the outer turns are
divided by partitions into chambers, and these
chambers, as we follow the spiral round, become
wider and wider, so as to overlap and wrap round
the older part of the shell, at first partially but
ultimately completely ; the first chamber that com-
pletely surrounds the shell marking the transition
from the spiral to the discoidal type.

We thus find that the discoidal Orbitolite shell
commences its development as a spiral shell, and
acquires the discoidal character merely through an
exaggerated mode of growth on the part of the
spiral. The Recapitulation Theory tells us that
this is to be interpreted as meaning that the dis-
coidal shells are descended from spiral ancestors,
and the close agreement between a young Orbito-
lite and an adult Peneroplis suggests that either
Peneroplis itself, or forms closely allied to it, were
the actual ancestors.

The Orbitolite is peculiarly instructive, owing to
the fact that the addition of new chambers during
growth takes place in such a way as to leave the
older parts of the shell unaltered and fully exposed
to view, so that simple inspection of an adult shell
reveals the whole course of development, and shows
us not merely the anatomy but the embryology as
well. It is as though a kitten were to develop
into a cat, not by interstitial growth in all its parts,
but by the addition of successive lengths to its
nose, its ears, its legs, and its tail; the additions

208 ANIMAL PEDIGREES

being cleverly effected so as to leave the original
kitten unaltered in the middle, and fully exposed
to view the whole time.

The above examples, selected almost haphazard,
will suffice to illustrate the Theory of Recapitulation.
The proof of the theory depends chiefly on its
universal applicability to all animals, whether high
or low in the zoological scale, and to all their parts
and organs. It derives also strong support from
the ready explanation which it gives of many
otherwise unintelligible points.

Of these latter, familiar and most instructive
instances are afforded by rudimentary organs — i.e.,
structures which, like the outer digits of the horse's
leg, or the intrinsic muscles of the ear of a man,
are present in the adult in an incompletely developed
form, and in a condition in which they can be of
no use to their possessors ; or else structures which
are present in the embryo, but disappear completely
before the adult condition is attained; for example,
the teeth of whalebone whales or the gill-slits
present in the neck during the embryonic phases of
all higher vertebrates.

Natural selection explains the preservation of
useful variations, but will not account for the for-
mation and perpetuation of useless organs, and
rudiments such as those mentioned above would be
unintelligible but for Recapitulation, which solves
the problem at once, showing that these organs,
though now useless, must have been of functional
value to the ancestors of their present possessors,
and that their appearance in the ontogeny of

ANIMAL PEDIGREES 209

existing forms is due to repetition of ancestral
characters.

Rudimentary organs are extremely common,
especially among the higher groups of animals, and
their presence and significance are now well under-
stood. Man himself affords numerous and excellent
examples, not merely in his bodily structure, but by
his speech, dress, and customs. For the silent
letter b in the word " doubt," the g in " reign," or
the w of " answer," or the buttons on his elastic-
side boots are as true examples of rudiments,
unintelligible but for their past history, as are the
ear muscles he possesses but cannot use ; or the
gill-clefts, which are functional in fishes and tad-
poles, and are present, though useless, in the em-
bryos of all higher vertebrates ; which in their early
stages the hare and the tortoise alike possess, and
which are shared with them by cats and by kings.

The fossil remains of animals and of plants yield
results of the greatest importance when studied in
the light of the Recapitulation Theory. I have
thought it well to ask special attention to these,
even at the risk of repeating what has been said
elsewhere and by others, for it seems to me that
zoologists are too apt nowadays to neglect palaeon-
tology, while palaeontologists have a tendency to
regard embryology as something beyond their own
ken, and concerning them but little. Moreover
there are certain points arising from a study of
fossils which, I venture to think, may possibly
commend themselves to some of our members as
suitable subjects for practical investigation.

210 ANIMAL PEDIGREES

The elder Agassiz was the first to point out, in
1858, the remarkable agreement between the em-
bryonic growth of animals and their palaeontological
history. He called attention to the resemblance
between certain stages in the growth of young fish
and their fossil representatives, and attempted to
establish, with regard to fish, a correspondence
between their palaeontological sequence and the
successive stages of embryonic development. He
then extended his observations to other groups of
animals, and stated his conclusions in these words :
" It may therefore be considered as a general
fact, very likely to be more fully illustrated as
investigations cover a wider ground, that the phases
of development of all living animals correspond to
the order of succession of their extinct representa-
tives in past geological times."

This point of view is of great importance. If
the development of an animal is really a repetition
or recapitulation of its ancestral history, then it is
clear that the agreement or parallelism which
Agassiz insists on between the embryological and
palaeontological records must hold good, and a
most important field of work is thus opened up to
us. It is sometimes urged however that such work
is necessarily unfruitful and inconclusive, because of
the scantiness of our knowledge concerning life in
the earlier geologic periods, or as it is commonly
termed, the imperfection of the geological record.
I have elsewhere protested against this objection,
and would repeat my protest here. The actual
number of fossils already obtained, especially from

ANIMAL PEDIGREES 21 1

the more recent formations, is prodigious; and what
we have to do is to make the most of the material
already accumulated, rather than to fold our hands
and idly lament the absence of forms that perhaps
never existed.

It is true that with all groups the chances are
not equal. But by judicious selection of groups in
which long series of specimens can be obtained,
and in which the hard skeletal parts, which alone
can be suitably preserved as fossils, afford reliable
indications of zoological affinity, it is possible to
test directly this alleged correspondence between
the palaeontological and embryological histories;
while in some instances a single lucky specimen
may afford us, on a particular point, all the evidence
we require. Many serious attempts have already
been made to work out in detail this comparison
between fossils and the developmental stages of
living forms, and the results obtained are most
promising.

Following the lines laid down by his father,
Alexander Agassiz has made a detailed comparison
between the fossil series and the embryonic phases
of recent forms in the case of the Echinoids or Sea
Urchins, a group peculiarly well adapted for such
an investigation, as the fossil representatives are
extremely numerous and well preserved, and the
existing members well known and comparatively
few in number. Agassiz shows that the two
records in this case agree remarkably closely ;
more especially in the independent evidence they
give of the origin of the asymmetrical forms from

212 ANIMAL PEDIGREES

more regular ancestors. The young Clypeastroid
for example has an ovoid test, a small number of
coronal plates, few and large primary tubercles and
spines, simple straight ambulacral areas, and no
petaloid ambulacra ; in fact has none of the
characteristic features of the adult Clypeastroid,
while the characters it does possess are those of
geologically older and preceding forms. So again,
in the group of Echinidae, the members of the com-
paratively recent polyporous group, in which each
ambulacral plate bears more than three pairs of
ambulacral pores, commence their existence in the
older and more primitive oligoporous condition, and
become polyporous through fusion of originally dis-
tinct ambulacral plates.

Agassiz gives many other examples, and from
a careful consideration of the entire group, arrives
at the conclusion that " comparing the embryonic
development with the palaeontological one, we find
a remarkable similarity ; " and again, " the com-
parison of the Echini which have appeared since
the Lias with the young stages of growth of the
principal families of recent Echini, shows a most
striking coincidence, amounting almost to identity,
between the successive fossil genera and the various
stages of growth."

In this connection Agassiz makes a suggestion of
much interest. We are apt he says to assume,
and perhaps rightly, that enormous periods of time
have elapsed during the conversion of genus into
genus, but the fact that these very changes can be
repeated before our eyes in a few days' or even

ANIMAL PEDIGREES 213

hours' time, during the development of the individual
animal, may perhaps afford us a hint that such
enormous periods are not really necessary in
historical development, and that transformation of
one form to a widely different one may, under
favourable circumstances, be effected with con-
siderable rapidity.

The Echinoids, and other groups of Echinoderms
as well, have been worked at from the same stand-
point and with the same results by Neumayr, in
whom we have recently lost one of the most gifted
and painstaking of palaeontologists.

As an example of the extreme value in certain
cases of a single fossil specimen, the singular fossil
bird Archaeopteryx may be referred to. In recent
birds the metacarpal bones of the wing are firmly
fused with one another and with the distal row of
carpal or wrist bones, but in development the
metacarpals are at first and for some time distinct.
The first specimen of Archaeopteryx discovered,
which is now in the British Museum, showed that
in it this distinctness was preserved in the adult —
i.e., that what is now an embryonic character in
recent birds was formerly an adult one.

Another very excellent illustration of the paral-
lelism between the palaeontological and the develop-
mental series is afforded by the antlers of deer,
which as is well known are shed annually, and
grow again of increased size and complexity in each
succeeding year. In the case of the red deer,
Cervus elaphus, the antlers are shed in the spring,
usually between the months of February and April ;

214 ANIMAL PEDIGREES

during the summer the new antlers sprout out, and
growing rapidly attain their full size at the pairing
season in August or September : they persist
throughout the winter, and are shed in the follow-
ing spring. The antlers of the first year are small
and unbranched ; those of the second year are
larger and branched; in the antlers of the third
year three tynes or points are present ; in the
fourth year four points, and so on until the full
size of the antler and the full number of points are
attained.

The geological history of antlers has been worked
out by Professor Gaudry and by Professor Boyd
Dawkins, and is of great interest. In the Lower
Miocene and earlier deposits no antlers have been
found. In the genus Procervulus from the Middle
Miocene, a pair of small, erect, branched, but non-
deciduous antlers were present, intermediate in
many respects between the antlers of deer and the
horns of antelopes. From slightly later deposits
a stag has been found with forked deciduous
antlers, which however do not appear to have
had more than two points. In Upper Miocene
times antlered ruminants were more abundant, and
the antlers themselves larger and more complex :
while from Pliocene deposits very numerous fossils
have been obtained showing a gradual increase in
the size of the antlers and the number of their
branches down to the present time.

Antlers are therefore, geologically considered,
very recent acquisitions : at their first appearance
they were small, and either simple or branched once

ANIMAL PEDIGREES 215

only ; while in succeeding ages they gradually
increased in size and in complexity. The palaeon-
tological series thus agrees with the developmental
series of stages through which the antlers of a stag
pass at the present day before attaining their full
dimensions.

There is another point of view from which
fossils acquire special interest in connection with
the Recapitulation Theory. If the theory is
correct, it must apply not merely to the animals
now living on the earth, but to all animals that
ever have lived ; and it becomes a matter of
considerable interest to enquire whether we have
any evidence whereby we can test this point, and
determine whether or not the fossil animals in
their own individual development repeated the
characters of their ancestors.

At first sight the enquiry does not seem a
promising one, for it may well be asked what
possibility there is of determining the embryology
or mode of development of animals which are
only known to us through the chance preservation
of their bones or shells as fossils. In most cases
it is true that such determination is impossible,
but in some groups as for example the Trilobites,
great numbers of well preserved specimens have
been obtained, not merely of adults, but of young
forms in various stages of growth ; and the study
of these young forms has already yielded results
of considerable interest. According to Barrande,
to whom our knowledge of ' these early stages
is mainly due, four chief types of development

216 ANIMAL PEDIGREES

may be recognised, differing from one another much
as existing Crustaceans do in the relative size and
perfection of the three regions, head, thorax, and
tail, into which the body is divided.

Evidence of a very different kind, and often of
far greater value, is afforded by the study of
shells, whether of Mollusca or of Foraminifera.
Such shells, like those of Orbitolites already
noticed, have no power of interstitial growth, and
increase in size can only be effected by the addition
of new shelly matter to the part already in
existence. In most instances these additions take
place in such manner that the older parts of the
shell are retained unaltered in the adult ; and
examination of the adult or fully formed shell will
then reveal the several stages through which the
shell passed in its development. In such a shell
for instance as Nautilus or Ammonites, the
central chamber is the oldest or first formed one,
to which the remaining chambers are added in
succession. If therefore the development of
the shell is a repetition of ancestral history, the
central chamber should represent the palseonto-
logically oldest form, and the remaining chambers,
in succession, forms of more and more recent
origin.

Ammonite shells present, more especially in
their sutures and in the markings and sculpturing
of their surface, characters that are easily recog-
nised. Upwards of four thousand species are
known to us, of many of which large numbers of
specimens can be obtained, in excellent preserva-

ANIMAL PEDIGREES 217

tion. The group consequently is a very suitable
one to study from our present standpoint ; and the
enquiry gains additional interest from the fact that
Ammonites are an entirely extinct group of
animals, no single species having survived the
cretaceous period, so that our only chance of learn-
ing anything about their embryology is to study the
fossil shells themselves.

Wurtenberger, who has made a special study of
the Jurassic Ammonites, has shown that there is
the same correspondence between historic and
embryonic development that obtains among living
animals. In the middle Jurassic deposits, for
instance, the older Ammonites are flattened and
disc-like, with numerous ribs ; in later forms the
shell bears rows of tubercles near the outer side
of the spiral, and later still a second inner row of
tubercles as well, while the ribs gradually become
less conspicuous and ultimately disappear. In
forms from more recent deposits the outer row of
tubercles disappears, and then the inner row, the
shell becoming smooth, swollen, and almost
spherical. On taking one of these smooth spheri-
cal shells, such as Aspidoceras cyclotum, and
breaking away the outer turns of the spiral so as
to expose the more central and older turns,
Wiirtenberger found first an inner and then an
outer row of tubercles appearing, which nearer the
centre disappeared, and in the oldest part of the
shell were replaced by the ribs characteristic of
the earlier, and presumably ancestral forms.
Results such as these open up to us a new field of

2i8 ANIMAL PEDIGREES

inquiry, which if energetically worked must yield
results of great interest and importance.

In order to understand fossils aright, and to
derive from them the full amount of information
they are capable of yielding us, it is necessary that
we should have a thorough knowledge of the deve-
lopment of their living descendants ; and more es-
pecially that we should be fully acquainted with
the several stages of formation of the shells or
other hard parts of the recent forms, which in
their fossil representatives are, with rare excep-
tions, the only parts sufficiently well preserved to
give trustworthy evidence.

Embryologists have too often confined them-
selves to the earlier stages of development, and
have unduly neglected the later stages, and more
especially the later stages of the skeletal structures.
By so doing they have failed to afford to palaeon-
tologists the aid which they are peculiarly qualified
to give, and which to the palaeontologist would be
of the utmost value. Fortunately the mistake is
now recognised, and serious efforts are being made
to remove the reproach.

We must now turn to another side of the ques-
tion. Although it is undoubtedly true that
development is to be regarded as a recapitulation
of ancestral phases, and that the embryonic
history of an animal presents to us a record of
the race history ; yet it is also an undoubted fact,
recognised by all writers on embryology, that the
record so obtained is neither a complete nor a
straightforward one. It is indeed a history, but a

ANIMAL PEDIGREES 219

history of which entire chapters are lost, while in
those that remain many pages are misplaced and
others are so blurred as to be illegible ; words,
sentences, or entire paragraphs are omitted, and
worse still alterations or spurious additions have
been freely introduced by later hands, and at
times so cunningly as to defy detection.

Very slight consideration will show that develop-
ment cannot in all cases be strictly a recapitulation
of ancestral stages. It is well known that closely
allied animals may differ markedly in their mode of
development. The common frog is at first a tadpole,
breathing by gills, a stage which is entirely omitted
by the West Indian Hylodes. A crayfish, a lobster,
and a prawn are allied animals, yet they leave the
egg in totally different forms. Some developmental
stages, as the pupa condition of insects, or the
stage in the development of a dogfish in which the
oesophagus is imperforate, cannot possibly be
ancestral stages. Or again, a chick embryo of, say
the fourth day, is clearly not an animal capable of
independent existence and therefore cannot cor-
rectly represent any ancestral condition, an objection
which applies to the developmental history of many,
perhaps of most, animals.

Haeckel long ago urged the necessity of distin-
guishing in actual development between those
characters which are really historical and inherited
and those which are acquired or spurious additions
to the record. The former he termed palingenetic
or ancestral characters, the latter cenogenetic or
acquired. The distinction is undoubtedly a true

220 ANIMAL PEDIGREES

one, but an exceedingly difficult one to draw in
practice. The causes which prevent development
from being a strict recapitulation of ancestral
characters, the modes in which these came about,
and the influence which they respectively exert,
are matters which are greatly exercising embryo-
logists ; and the attempt to determine them has as
yet met with only partial success.

The most potent and the most widely spread of
these disturbing causes arise from the necessity of
supplying the embryo with nutriment. This acts
in two ways. If the amount of nutritive matter
with the egg is small, then the young animal must
hatch early, and in a condition in which it is able
to obtain food for itself. In such cases there is of
necessity a long period of larval life, during which
natural selection may act so as to introduce modi-
fications of the ancestral history, spurious additions
to the text.

If on the other hand the egg contain within
itself a considerable quantity of nutrient matter,
then the period of hatching can be postponed
until this nutrient matter has been used up. The
consequence is that the embryo hatches at a much
later stage of its development, and if the amount of
food material is sufficient may even leave the egg
in the form of the parent. In such cases the earlier
developmental phases are often greatly condensed
and abbreviated ; and as the embryo does not
lead a free existence, and has no need to exert it-
self to obtain food, it commonly happens that these
stages are passed through in a very modified form,

ANIMAL PEDIGREES 221

the embryo being, as in a four-day chick, in a con-
dition in which it is clearly incapable of independent
existence.

The nutrition of the embryo prior to hatching is
most usually effected by granules of nutrient matter,
known as food yolk, and embedded in the proto-
plasm of the egg itself; and it is on the relative
abundance of these granules that the size of the
egg chiefly depends.

Large size of eggs implies diminution of their
number, and hence in that of the offspring ;
and it can be well understood, that while some
species derive advantage in the struggle for exist-
ence by producing the maximum number of young,
to others it is of greater importance that the young
on hatching should be of considerable size and
strength, and so better able to begin the world on
their own account. In other words, some animals
may gain by producing a large number of small
eggs, others by producing a smaller number of eggs
of larger size — i.e., provided with more food yolk.

The immediate effect of a large amount of food
yolk is to mechanically retard the processes of
development ; the ultimate result is to greatly
shorten the time occupied by development. This
apparent paradox is readily explained. A small
egg, such as that of Amphioxus, starts its develop-
ment rapidly, and in about eighteen hours gives
rise to a free-swimming larva, capable of indepen-
dent existence, with a digestive cavity and nervous
system already formed ; while a large egg, like that
of the hen, hampered by the great mass of food

222 ANIMAL PEDIGREES

yolk with which it is distended, has in the same
time made but very slight progress. From this
time however other considerations begin to tell.
Amphioxus has been able to make this rapid start
owing to its relative freedom from food yolk. This
freedom now becomes a retarding influence, for the
Jarva, containing within itself but a very scanty
supply of nutriment, must devote much of its
energies to hunting for and to digesting its food,
and hence its further development will proceed more
slowly.

The chick embryo on the other hand has an
abundant supply of food in the egg itself; it has
no occasion to spend time in searching for food, but
can devote its whole energies to the further stages
of its development. Hence, except in the earliest
stages, the chick develops more rapidly than Am-
phioxus, and attains its adult form in a much
shorter time.

The tendency of abundant food yolk to lead to
shortening or abbreviation of the ancestral history,
and even to the entire omission of important stages,
is well known. The embryo of forms well provided
with yolk takes short cuts in its development,
jumps from branch to branch of its genealogical
tree, instead of climbing steadily upwards.

An excellent illustration of the influence of food
yolk on development is afforded by the life histories
of frogs. The common frog, Rana temporaria, lays
as is well known eggs of small size, about ^ inch
in diameter. A small egg can only contain a
limited amount of food yolk, and hence the young

ANIMAL PEDIGREES 223

frog can only accomplish a small part of its de-
velopmental history within the egg, and must then
hatch in order to obtain food from without. Con-
sequently a frog hatches not as a frog but as a
tadpole — i.e., at the fish stage in the ancestral
history of frogs. At the time of hatching there
are no limbs and no lungs ; the heart, the ali-
mentary canal, and the nervous system are in an
extremely imperfect condition ; while other organs
of the adult frog, such as the kidneys, have not yet
commenced to appear. The frog has therefore to
effect the greater part of its development after the
time of hatching.

In the little West Indian frog, Hylodes, the
course of events is very different. This frog
which is of small size — less than a couple of
inches in length — lays its eggs not in water but on
the leaves of plants. The eggs are large, having
a diameter of about i inch — i.e., are about three
times the diameter and twenty-seven times the
bulk of the eggs of the common English frog. The
large size of the egg is caused as we have seen
by great abundance of food yolk ; and the con-
sequence of this large supply of food yolk in the
egg of Hylodes is that the frog is enabled to
complete the whole of its development before
hatching, and emerges from the egg capsule in a
form differing from the adult merely in the posses-
sion of a rudimentary stump of a tail ; and even
this disappears before the close of the first day of
its existence.

A further and direct consequence of this de-

224 ANIMAL PEDIGREES

velopment within the egg is that the successive
stages are hurried over, and are at best but
imperfectly recapitulated. Thus although Hylodes
passes through what may be called a tadpole stage,
it never develops gills ; so that were Hylodes the
only frog known to us, it is very likely that we
should have arrived at other conclusions concern-
ing the pedigree of frogs.

The influence of food yolk on the development
of animals is closely analogous to that of capital in
human undertakings. A new industry, for example
that of pen-making, has often been started by a
man working by hand and alone, making and sell-
ing his own wares; if he succeed in the struggle
for existence, it soon becomes necessary for him to
call in others to assist him, and to subdivide the
work ; hand labour is soon superseded by machines,
involving further differentiation of labour ; the
earlier machines are replaced by more perfect and
more costly ones ; factories are built, agents en-
gaged, and in the end a whole army of workpeople
employed. In later times a man commencing the
same business with very limited means will start
at the same level as the original founder, and will
have to work his way upwards through much the
same stages — i.e., will repeat the pedigree of the
industry. The capitalist on the other hand is
enabled, like Hylodes, to omit these earlier stages,
and after a brief period of incubation, to start busi-
ness with large factories equipped with the most
recent appliances, and with a complete staff of work-
people— i.e., to spring into existence fully fledged.

ANIMAL PEDIGREES 225

There is no doubt that abundance of food yolk is
a direct and very frequent cause of the omission of
ancestral stages from individual development ; but
it must not be viewed as the sole cause. It is quite
impossible that any animal, except perhaps in the
lowest zoological groups, should repeat all the
ancestral stages in the history of the race ; the
limits of time available for individual development
will not permit this. There is a tendency in all
animals towards condensation of the ancestral
history, towards striking a direct path from the egg
to the adult. This tendency is best marked in the
higher, the more complicated members of a group—
i.e., in those which have a longer and more tortuous
pedigree ; and, although greatly strengthened by
the presence of food yolk in the egg, is apparently
not due to this in the first instance.

Thus the simpler forms of Orbitolites, such as
O. tenm'ssima, repeat in their development all the
stages leading from a spiral to a discoidal shell ;
but in the more complicated species, as Dr. Car-
penter has pointed out, there is a tendency towards
precocious development of the adult characters, the
earlier stages being hurried over in a modified
form ; while in the most complex examples, as in
O. complanata,) the earlier spiral stages may be entirely
omitted, the shell acquiring almost from its earliest
commencement the discoidal mode of growth. There
is no question here of relative abundance of food
yolk, but merely of early or precocious appearance
of adult characters.

Of causes other than food yolk, or only in-

226 ANIMAL PEDIGREES

directly connected with it, which tend to falsify the
ancestral history, many are now known, but time
will only permit me to notice the more important.
These are distortion, whether in time or space ;
sudden or violent metamorphosis ; a series of modi-
fications, due chiefly to mechanical causes, and which
may be spoken of as developmental conveniences ;
the important question of variability in development ;
and finally the great problem of degeneration.

Concerning distortion in time, all embryologists
have noticed the tendency to anticipation or pre-
cocious development of characters which really
belong to a later stage in the pedigree. The early
attainment of the discoidal form in the shell of
Orbitolites complanata is a case in point ; and
Wilrtenberger has specially noticed this tendency
in Ammonites. Many early larvae show it markedly,
the explanation in this case being that it is
essential for them to hatch in a condition capable of
independent existence — i.e., capable at any rate of
obtaining and digesting their own food. Ana-
chronisms, or actual reversal of the historical order
of development of organs or parts, occur frequently.
Thus the joint surfaces of bones acquire their
characteristic curvatures before movement of one
part on another is effected, and before even the
joint cavities are formed.

Another good example is afforded by the develop-
ment of the mesenterial filaments in Alcyonarians.
Wilson has shown, in the case of Renilla, that in
the development of an embryo from the egg the six
endodermal filaments appear first, and the two long

ANIMAL PEDIGREES 227

ectodermal filaments at a later period ; but that in
the formation of a bud this order of development is
reversed, the ectodermal filaments being the first
formed. He suggests in explanation, that as the
endodermal filaments are the digestive organs, it is
of primary importance to the free embryo that they
should be formed quickly. The long ectodermal
filaments are chiefly concerned with maintaining
currents of water through the colony ; in bud-
development they appear before the endodermal
filaments, because they enable the bud during its
early stages to draw nutrient matter from the body
fluid of the parent ; while the endodermal filaments
cannot come into use until the bud has acquired its
own mouth and tentacles.

The completion Of the ventricular septum in the
heart of higher vertebrates before the auricular
septum is an often quoted anachronism, and every
embryologist could readily furnish many other
cases. A curious instance is afforded by the
development of the teeth in mammals, if recent
suggestions as to the origin of the milk dentition
are confirmed, and the milk dentition prove to be a
more recent acquisition than the permanent one.*

Distortion of a curious kind is seen in cases of
abrupt metamorphosis where, as in the case of
many Echinoderms, of Phoronis, and of the
metabolic insects, the larva and the adult differ
greatly in form, habits, mode of life, and very
usually in the nature of their food and the mode of
obtaining it ; and the transition from the one stage
* This has since been disproved. — ED.

228 ANIMAL PEDIGREES

to the other is not a gradual but an abrupt one, at
any rate so far as external characters are concerned.
Sudden changes of this kind, as from the free
swimming Pluteus to the creeping Echinus, or from
the sluggish leaf-eating caterpillar to the dainty
butterfly, cannot possibly be recapitulatory, for
even if small jumps are permissible in nature, there
is no room for bounds forward of this magnitude.
Cases of abrupt metamorphosis may always be
viewed as due to secondary modifications, and
rarely if ever have any significance beyond the
particular group of animals concerned. For
example, a Pluteus larva may be recognised as
belonging to the group of Echinoidea before the
adult urchin has commenced to be formed within it,
and the Lepidopteran caterpillar is already an un-
mistakable insect. Hence, for the explanation of
the metamorphoses in these cases it is useless to
look outside the groups of Echinoidea and Insecta
respectively.

Abrupt metamorphosis is always associated with
great change in external form and appearance, in
manner of life, and very usually in mode of
nutrition. A gradual transition in such cases is
inadmissible, because in the intermediate stages the
animal would be adapted to neither the larval nor
the adult condition ; a gradual conversion of the
biting mouth of the caterpillar to the sucking
proboscis of a moth would inevitably lead to
starvation. This difficulty is evaded by retaining
the external form and habits of one particular stage
for an unduly long period, so that the relations of

ANIMAL PEDIGREES 229

the animal to the surrounding environment remain
unchanged, while internally preparations for the
later stages are in progress.

Cinderella and the princess are equally possible
entities, each being well adapted to her environ-
ment. The exigencies of the situation do not
permit however of a gradual change from one to
the other : the transformation, at least as regards
external appearance, must be abrupt.

Embryology supplies us with many unsolved
problems, and it is not to be wondered at that this
should be the case. Some of these may fairly be
spoken of as mere curiosities of development, while
others are clearly of greater moment. I do not
propose to catalogue these, but will merely mention
one which I happen to have recently run my head
against, and remember vividly.

The solid condition of the oesophagus in dogfish
embryos, first noticed by Balfour, is a very curious
point. The oesophagus has at first a well-developed
lumen, like the rest of the alimentary canal ; but at
an early period, stage K of Balfour's nomenclature,
the part of the oesophagus overlying the heart, and
immediately behind the branchial region, becomes
solid and remains solid for a long time, the exact date
of reappearance of the lumen not being yet ascer-
tained. A similar solidification of the oesophagus
occurs in tadpoles of the common frog. In young free
swimming tadpoles the oesophagus is perforate, but
in tadpoles of about £ inch length it becomes solid
and remains so until a length of about J inch has
been attained. The solidification occurs at a stage

230 ANIMAL PEDIGREES

closely corresponding with that in which it first
appears in the dogfish, and a curious point about it
is that in the frog the oesophagus becomes solid just
before the mouth opening is formed, and remains
solid for some little time after this important event.
This closing of the oesophagus clearly cannot be
recapitulation, but the fact that it occurs at corre-
sponding periods in the frog and the dogfish
suggests that it may possibly, as Balfour hinted,
"turn out to have some unsuspected morphological
bearing."

A matter which at present is attracting much
attention is the question of degeneration. Natural
selection, though consistent with and capable of
leading to steady upward progress and improvement,
by no means involves such progress as a necessary
consequence. All it says is that those animals will
in each generation have the best chance of survival
which are most in harmony with their environment,
and such animals will not necessarily be those
which are ideally the best or most perfect.

If you go into a shop to purchase an umbrella
the one you select is by no means necessarily that
which most nearly approaches ideal perfection,
but the one which best hits off the mean between
your idea of what an umbrella should be and the
amount of money you are prepared to give for it ;
the one in fact that is on the whole best suited to
the circumstances of the case, or the environment
for the time being. It might well happen that you
had a violent antipathy to a crooked handle, or else
were determined to have a catch of a particular kind

ANIMAL PEDIGREES 231

to secure the ribs, and this might lead to the
selection — i.e., the survival, of an article that in other
and even in more important respects was manifestly
inferior to the average.

So is it also with animals : the survival of a form
that is ideally inferior is very possible. To animals
living in profound darkness the possession of eyes
is of no advantage, and forms devoid of eyes would
not merely lose nothing thereby, but would actually
gain, inasmuch as they would escape the dangers
that might arise from injury to a delicate and
complicated organ. In extreme cases, as in animals
leading a parasitic existence, the conditions of life
may be such as to render locomotor, digestive,
sensory, and other organs entirely useless ; and in
such cases those forms will be most in harmony
with their surroundings which avoid the waste of
energy resulting from the formation and maintenance
of these organs.

An excellent illustration of this downhill progress
is afforded by the Rhizocephala, a curious group of
parasitic Crustaceans, of which the genus Sacculina
is perhaps the best known member. The adult
Sacculina is found as a soft shapeless bag, an inch
or so in length, attached to the under surface of the
tail of a crab by a fleshy stalk which, passing
through the skin of the crab, spreads out within it
into a complicated system of branching tubular roots
by which the parasite sucks up the juices of the
crab, on which it depends for food. As regards
structure, the Sacculina in its fully developed form
is little more than a bag of eggs enclosed in a

232 ANIMAL PEDIGREES

loosely fitting outer skin, with a single orifice
through which the young escape. A nervous
system is present, but there are no traces of limbs,
digestive system, heart, breathing organs, or sense
organs. Indeed, examination however careful of
an adult Sacculina would fail to afford any clue as
to its real zoological affinities.

Development however in obedience to the potent
law of recapitulation, shows us at once that
Sacculina is a Crustacean, more closely allied to the
Barnacles than to any other of the more familiar
members of the group. From each of the exceed-
ingly numerous eggs which a Sacculina produces,
there emerges a minute, free swimming larva of the
type known as a Nauplius, characterised by possess-
ing a somewhat pyriform body, a single median eye,
and three pairs of swimming appendages or legs.
Nauplius larvae are widely spread amongst Crusta-
ceans. All Entomostraca, except the Cladocera,
hatch in this form, and Nauplius larvae are found
in individual members in nearly all the higher
groups as well. The only special peculiarity about
the Sacculina Nauplius is that it has neither mouth
nor digestive organs of any kind, these being
unnecessary by the presence of a considerable
quantity of food yolk. The Sacculina larva con-
tinues its free existence for a time; it casts its skin,
or rather its cuticle many times, emerging each time
rather more complicated in structure though actually
smaller in size, for it cannot yet take in food.
Ultimately it reaches the condition spoken of as the
pupa stage. The pupa is enclosed in a bivalved

ANIMAL PEDIGREES 233

carapace, very similar to that of a Cypris. It
possesses a pair of well-developed antennules in
front and six pairs of swimming-legs behind. The
Nauplius eye is still present, a little way behind the
basal joints of the antennules.

Now comes the great change. The pupa meet-
ing with a crab fastens itself to the under surface
of the crab's tail by its antennules, and then goes
to the bad with startling rapidity. Within three
hours of the time of fixing itself to the crab, the
six pairs of swimming legs, with the muscles
moving them, and the whole posterior part of
the body disintegrate and are cast off. The
antennules become modified into a tube, piercing
the skin of the crab ; the head of the Sacculina
remains as a bottle-shaped mass in connection with
the modified antennules, but the bivalved carapace,
with all the other organs including the eye, are
cast off and lost. The Sacculina now passes for a
time completely into the interior of the crab : later
on, after increasing in size, it comes once more to
the surface and becomes the bag-like mass which
we have found to be the adult condition.

This is a typical instance of degeneration or
retrograde development, the animal being more
highly organised, and standing at a higher mor-
phological level in its early stages than when adult.
Yet inasmuch as the organs that are lost, such as
the limbs and eye, would be of no use to it in its
changed conditions of life, there is nothing in the
whole history that is in any way inconsistent with
natural selection. This principle of degeneration,

234 ANIMAL PEDIGREES

recognised by Darwin as a possible, and under
certain conditions a necessary consequence of his
theory, has been since advocated strongly by Dohrn,
and later by Lankester, in an evening discourse
delivered before the British Association at the
Sheffield meeting, in 1879. Both Dohrn and
Lankester have suggested that degeneration may
occur much more widely than is commonly sup-
posed.

In animals which are parasitic when adult, but
free swimming in their early stages, as in the case
of Sacculina, degeneration is clear enough ; so also
is it in the case of the solitary Ascidians, in which
the larva is a free swimming animal with a noto-
chord, an elongated tubular nervous system, and
sense organs, while the adult is fixed, devoid of
the swimming tail, with no notochord, and with a
greatly reduced nervous system and aborted sense
organs. In such cases the animal, when adult, is,
as regards the totality of its organisation, at a
distinctly lower morphological level, is less highly
differentiated than it is when young, and during
individual development there is actual retrograde
development of important systems and organs.

About such cases there is no doubt ; but we
are asked to extend the idea of degeneration much
more widely. It is urged that we ought not to
demand direct embryological evidence before accept-
ing a group as degenerate. We are reminded of
the tendency to abbreviation or to complete omis-
sion of ancestral stages of which we have quoted
examples above ; and it is suggested that if such

ANIMAL PEDIGREES 235

larval stages were omitted in all the members of a
group, we should have no direct evidence of de-
generation in a group that might really be in an
extremely degenerate condition. Supposing for
instance the free larval stages of the solitary
Ascidians were suppressed, say through the acqui-
sition of food yolk, then it is urged that the
degenerate condition of the group might easily
escape detection. The supposition is by no means
extravagant. Food yolk varies greatly in amount
in allied animals, and cases like Hylodes, or
amongst Ascidians, Pyrosoma, show how readily a
mere increase in the amount of food yolk in the egg
may lead to the omission of important ancestral
stages.

The question then arises whether it is not
possible, or even probable, that animals which now
show no indication of degeneration in their de-
velopment are in reality highly degenerate, and
whether it is not legitimate to suppose such de-
generation to have occurred in the case of animals
whose affinities are obscure or difficult to determine.
It is more especially with regard to the lower
vertebrates that this argument has been employed ;
and at the present day zoologists of authority,
relying on it, do not hesitate to speak of such
forms as Amphioxus and the Cyclostomes as de-
generate animals, as wolves in sheep's clothing,
animals whose simplicity is acquired and deceptive
rather than real and ancestral.

I cannot but think that cases such as these
should be regarded with some jealousy ; there is

236 ANIMAL PEDIGREES

at present a tendency to invoke degeneration rather
freely as a talisman to extricate us from mor-
phological difficulties ; and an inclination to accept
such suggestions, at any rate provisionally, without
requiring satisfactory evidence in their support.

Degeneration of which there is direct embryo-
logical evidence stands on a very different footing
from suspected degeneration, for which no direct
evidence is forthcoming ; and in the latter case the
burden of proof undoubtedly rests with those who
assume its existence. The alleged instances among
the lower vertebrates must be regarded particularly
closely, because in their case the suggestion of
degeneration is admittedly put forward as a means
of escape from difficulties arising through theoretical
views concerning the relation between vertebrates
and invertebrates.

Amphioxus itself, so far as I can see, shows
in its development no sign of degeneration, except
possibly with regard to the anterior gut diverticula,
whose ultimate fate is not altogether clear. With
regard to the earlier stages of development, con-
cerning which, thanks to the patient investigations
of Kowalevsky and Hatschek, our knowledge is
precise, there is no animal known to us in which
the sequence of events is simpler or more straight-
forward. Its various organs and systems are
formed in what is recognised as a primitive manner ;
and the development of each is a steady upward
progress towards the adult condition. Food yolk,
the great cause of distortion in development, is
almost absent, and there is not the slightest indi-

ANIMAL PEDIGREES 237

cation of the former possession of a larger quantity.
Concerning the later stages our knowledge is
incomplete, but so much as has been ascertained
gives no support to the suggestion of general
degeneration.

Our knowledge of the conditions leading to
degeneration is undoubtedly incomplete, but it
must be noticed that the conditions usually asso-
ciated with degeneration do not occur. Amphioxus
is not parasitic, is not attached when adult, and
shows no evidence of having formerly possessed
food yolk in quantity sufficient to have led to the
omission of important ancestral stages. Its small
size, as compared with other vertebrates, is one
of the very few points that can be referred to as
possibly indicating degeneration, but by no means
proving its occurrence.

A consideration of much less importance, but
deserving of mention, is that in its mode of life
Amphioxus not merely differs, as already noticed,
from those groups of animals which we know to
be degenerate, but agrees with some, at any rate,
of those which there is reason to regard as primitive
or persistent types. Amphioxus, like Balanoglossus,
Lingula, Dentalium, and Limulus, is marine, and
occurs in shallow water, usually with a sandy
bottom, and like the three smaller of these genera
it lives habitually buried almost completely in the
sand, into which it burrows with great rapidity.

I do not wish to speak dogmatically. I merely
wish to protest against a too ready assumption
of degeneration ; and to repeat that, so far as I can

238 ANIMAL PEDIGREES

see, Amphioxus has not yet, either in its develop-
ment, in its structure, or in its habits, been shown
to present characters that suggest, still less that
prove, the occurrence in it of general or extensive
degeneration. In a sense all the higher animals
are degenerate ; that is, they can be shown to
possess certain organs in a less highly developed
condition than their ancestors, or even in a rudi-
mentary state. Thus a crab as compared with a
lobster is degenerate in the matter of its tail,
a horse as compared with Hipparion in regard to
its outer toes ; but it is neither customary nor
advisable to speak of a crab as a degenerate animal
compared to a lobster ; to do so would be mislead-
ing. An animal should only be spoken of as
degenerate when the retrograde development is well
marked, and has affected not one or two organs
only, but the totality of its organisation.

It is impossible to draw a sharp line in such
cases, and to limit precisely the use of the term
degeneration. It must be borne in mind that no
animal is at the top of the tree in all respects.
Man himself is primitive as regards the number of
his toes, and degenerate in respect to his ear
muscles ; and between two animals even of the
same group it may be impossible to decide which
of the two is to be called the higher and which the
lower form. Thus, to compare an oyster with a
mussel. The oyster is more primitive than the
mussel as regards the position of the ventricle of
the heart and its relations to the alimentary canal ;
but is more modified in having but a single

ANIMAL PEDIGREES 239

adductor muscle ; and almost certainly degenerate
in being devoid of a foot.

Care must also be taken to avoid speaking of
an animal as degenerate in regard to a particular
organ merely because that organ is less fully
developed than in allied animals. An organ is not
degenerate unless its present possessor has it in
a less perfect condition than its ancestors had. A
man is not degenerate in the matter of the length
of his neck as compared with a giraffe, . nor as
compared with an elephant in respect of the size
of his front teeth, for neither elephant nor giraffe
enters into the pedigree of man. A man is how-
ever degenerate, whoever his ancestors may have
been, in regard to his ear muscles ; for he possesses
these in a rudimentary and functionless condition,
which can only be explained by descent from some
better equipped progenitor.

We have now considered some of the more
important of the influences which are recognised
as affecting developmental history in such a way as
to render the recapitulation of ancestral stages less
complete than it might otherwise be ; which tend
to prevent ontogeny from correctly repeating the
phylogenetic history. It may at this point reason-
ably be asked whether there is any test by which
we can determine whether a given larval character
is or is not ancestral. Most assuredly there is no
one rule, no single test, that will apply in all cases ;
but there are certain considerations which will help
us, and which should be kept in view. A character
that is of general occurrence among the members

240 ANIMAL PEDIGREES

of a group, both high and low, may reasonably be
regarded as having strong claims to ancestral rank;
claims that are greatly strengthened if it occurs at
corresponding developmental periods in all cases ;
and still more if it occurs equally in forms that
hatch early as free larvae, and in forms with large
eggs, which develop directly into the adult. As
examples of such characters may be cited the mode
of formation and relations of the notochord, and of
the gill-clefts of vertebrates, which satisfy all the
conditions mentioned. Characters that are tran-
sitory in certain groups, but retained throughout
life in allied groups, may with tolerable certainty
be regarded as ancestral for the former; for
instance, the symmetrical position of the eyes in
young flat-fish, the spiral shell of the young limpet,
the superficial position of the madreporite in
Elasipodous Holothurians, or the suckerless con-
dition of the ambulacral feet in many Echinoderms.
A more important consideration is that if the
developmental changes are to be interpreted as a
correct record of ancestral history, then the several
stages must be possible ones, the history must be
one that could actually have occurred — i.e., the
several steps of the history as reconstructed must
form a series, all the stages of which are practicable
ones. Natural selection explains the actual
structure of a complex organ as having been
acquired by the preservation of a series of stages,
each a distinct, if slight, advance on the stage
immediately preceding it, an advance so distinct as
to confer on its possessor an appreciable advantage

ANIMAL PEDIGREES 241

in the struggle for existence. It is not enough
that the ultimate stage should be more advantageous
than the initial or earlier condition, but each inter-
mediate stage must also be a distinct advance. If
then the development of an organ is strictly
recapitulatory, it should present to us a series of
stages, each of which is not merely functional, but
a distinct advance on the stage immediately pre-
ceding it. Intermediate stages — e.g., the solid
oesophagus of the tadpole, which are not and could
not be functional — can form no part of an ancestral
series ; a consideration well expressed by Sedgwick
thus : " Any phylogenetic hypothesis which pre-
sents difficulties from a physiological standpoint
must be regarded as very provisional indeed."

A good example of an embryological series
fulfilling these conditions is afforded by the de-
velopment of the eye in the higher Cephalopoda.
The earliest stage consists in the depression of a
slightly modified patch of skin ; round the edge of
the patch the epidermis becomes raised up as a
rim ; this gradually grows inwards from all sides,
so that the depressed patch now forms a pit,
communicating with the exterior through a small
hole or mouth. By further growth the mouth of
the pit becomes still more narrowed, and ultimately
completely closed, so that the pit becomes converted
into a closed sac or vesicle ; at the point at which
final closure occurs formation of cuticle takes place,
which projects as a small transparent drop into the
cavity of the sac; by the formation of concentric
layers of cuticle this drop becomes enlarged into the

Q

242 ANIMAL PEDIGREES

spherical transparent lens of the eye ; and the
development is completed by histological changes
in the inner wall of the vesicle, which convert it
into the retina, and by the formation of folds of
skin around the eye, which become the iris and the
eyelids respectively. Each stage of this develop-
mental history is a distinct advance, physiologically,
on the preceding stage, and furthermore, each
stage is retained at the present day as the perma-
nent condition of the eye in some member of the
group Mollusca. The earliest stage, in which the
eye is merely a slightly depressed and slightly
modified patch of skin, represents the simplest
condition of the Molluscan eye, and is retained
throughout life in Solen. The stage in which the
eye is a pit with widely open mouth, is retained
in the limpet ; it is a distinct advance on the former,
as through the greater depression the sensory cells
are less exposed to accidental injury. The narrow-
ing of the mouth of the pit in the next stage is a
simple change, but a very important step forward.
Up to this point the eye has served to distinguish
light from darkness, but the formation of an image
has been impossible. Now, owing to the smallness
of the aperture, and the pigmentation of the walls
of the pit which accompanies the change, light from
any one part of an object can only fall on one
particular part of the inner wall of the pit or
retina, and so an image, though a dim one, is
formed. This type of eye is permanently retained
in the Nautilus. The closing of the mouth of the
pit by a transparent membrane will not affect the

ANIMAL PEDIGREES 243

optical properties of the eye, and will be a gain, as
it will prevent the entrance of foreign bodies into
the cavity of the eye. The formation of the lens
by deposit of cuticle is the next step The gain
here is increased distinctness and increased bright-
ness of the image, for the lens will focus the rays
of light more sharply on the retina, and will allow
a greater quantity of light, a larger pencil of rays
from each part of the object, to reach the corre-
sponding part of the retina. The eye is now in the
condition in which it remains throughout life in the
snail and other gastropods. Finally the formation
of the folds of skin known as iris and eyelids
provides for the better protection of the eye, and
is a clear advance on the somewhat clumsy method
of withdrawal seen in the snail,

It is not always possible to point out so clearly
as in the above instance the particular advantage
gained at each step, even when a complete develop-
mental series is known to us ; but in such cases, as
for instance in Orbitolites, our difficulties may be
largely ascribed to ignorance of the particular con-
ditions that confer advantage in the struggle for
existence, in the case of the forms we are dealing
with. That ontogeny really is a repetition of phylo-
geny must I think be admitted, in spite of the
numerous and various ways in which the ances-
tral history may be distorted during the actual
development.

Before leaving the subject, it is worth while
inquiring whether any explanation can be found
of recapitulation. A complete answer can certainly

244 ANIMAL PEDIGREES

not be given at present, but a partial one may
perhaps be obtained. Darwin himself suggested
that the clue might be found in the consideration
that at whatever age a variation first appears in the
parent, it tends to reappear at a corresponding age
in the offspring ; but this must be regarded rather
as a statement of the fundamental fact of embry-
ology than as an explanation of it. It is probably
safe to assume that animals would not recapitulate
unless they were compelled to do so : that there
must be some constraining influence at work,
forcing them to repeat more or less closely the
ancestral stages. It is impossible, for instance, to
conceive what advantage it can be to a reptilian or
mammalian embryo to develop gill-clefts which are
never used, and which disappear at a slightly later
stage ; or how it can benefit a whale, that in its
embryonic condition it should possess teeth which
never cut the gum, and which are lost before birth.
Moreover, the history of development in different
animals or groups of animals offers to us, as we
have seen, a series of ingenious, determined, varied,
but more or less unsuccessful efforts to escape
from the necessity of recapitulating, and to sub-
stitute for the ancestral process a more direct
method.

A further consideration of importance is that
recapitulation is not seen in all forms of develop-
ment, but only in sexual development ; or at least
only in development from the egg. In the several
forms of asexual development, of which budding
is the most frequent and most familiar, there is no

ANIMAL PEDIGREES 245

repetition of ancestral phases ; neither is there in
cases of regeneration of lost parts, such as the
tentacle of a snail, the arm of a starfish, or the tail
of a lizard. In such regeneration it is not a larval
tentacle, or arm, or tail, that is produced, but an
adult one.

The most striking point about the development
of the higher animals is that they all alike com-
mence as eggs. Looking more closely at the egg
and the conditions of its development, two facts
impress us as of special importance. First, the
egg is a single cell, and therefore represents mor-
phologically the Protozoon, or earliest ancestral
phase ; secondly, the egg, before it can develop,
must be fertilised by a spermatozoon, just as the
stimulus of fertilisation by the pollen grain is neces-
sary before the ovum of a plant will commence to
develop into the plant-embryo.

The advantage of cross-fertilisation in increasing
the vigour of the offspring is well known, and in
plants devices of the most varied and even extra-
ordinary kind are adopted to ensure that such
cross-fertilisation occurs. The essence of the act
of cross-fertilisation, which is already established
among Protozoa, consists in combination of the
nuclei of two cells, male and female, derived from
different individuals. The nature of the process is
of such a kind that two individual cells are alone
concerned in it ; and it may I think be reasonably
argued that the reason why animals commence
their existence as eggs — i.e., as single cells — is
because it is in this way only that the advantage

246 ANIMAL PEDIGREES

of cross-fertilisation can be secured, an advantage
admittedly of the greatest importance, and to
secure which natural selection would operate power-
fully.

The occurrence of parthenogenesis, either occa-
sionally or normally, in certain groups is not I
think a serious objection to this view. There are
very strong reasons for holding that partheno-
genetic development is a modified form, derived
from the sexual method. Moreover, the view
advanced above does not require that cross-fertilisa-
tion should be essential to individual development,
but merely that it should be in the highest degree
advantageous to the species ; and hence leaves
room for the occurrence, exceptionally, of partheno-
genetic development.

If it be objected that this is laying too much
stress on sexual reproduction, and on the advantage
of cross- fertilisation, then it may be pointed out in
reply that sexual reproduction is the characteristic
and essential mode of multiplication among Metazoa :
that it occurs in all Metazoa, and that when
asexual reproduction, as by budding, occurs, this
merely alternates with the sexual process which,
sooner or later, becomes essential.

If the fundamental importance of sexual repro-
duction to the welfare of the species be granted,
and if it be further admitted that Metazoa are
descended from Protozoa, then we see that there is
really a constraining force of a most powerful nature
compelling every animal to commence its life-history
in the unicellular condition, the only condition in

ANIMAL PEDIGREES 247

which the advantage of cross-fertilisation can be
obtained — i.e., constraining every animal to begin
its development at its earliest ancestral stage, at the
very bottom of its genealogical tree.

On this view the actual development of any
animal is strictly limited at both ends: it must
commence as an egg, and it must end in the
likeness of the parent. The problem of recapitula-
tion becomes thereby greatly narrowed ; all that
remains being to explain why the intermediate
stages in the actual development should repeat the
intermediate stages of the ancestral history.

Although narrowed in this way, the problem still
remains one of extreme difficulty. It is a conse-
quence of the theory of Natural Selection that
identity of structure involves community of descent ;
a given result can only be arrived at through
a given sequence of events : the same morphological
goal cannot be reached by two independent paths.
A negro and a white man have had common
ancestors in the past ; and it is through the long-
continued action of selection and environment that
the two types have been gradually evolved. You
cannot turn a white man into a negro merely by
sending him to live in Africa: to create a negro the
whole ancestral history would have to be repeated ;
and it may be that it is for the same reason that the
embryo must repeat or recapitulate its ancestral
history in order to reach the adult goal.

I am not sure that we can get much further than
this at present. However, be the explanation what
it may, there can I think be no doubt as to the

248 ANIMAL PEDIGREES

general truth of the Recapitulation Theory, and the
wonderful assistance which it gives us in recon-
structing the pedigrees of animals. Yet it must
not be supposed that all we have to do in order to
determine the past history of a species is to study
the actual development of the existing members of
that species.

Embryology is not to be regarded as a master
key that will open the gates of knowledge, and
remove all obstacles from our path without further
trouble on our part ; it is rather to be viewed and
treated as a delicate and complicated instrument, the
proper handling of which requires the utmost nicety
of balance and adjustment, and which unless
employed with the greatest skill and judgment may
yield false instead of true results. We are indeed
only just beginning to understand the real power of
our weapons and the right way of employing them ;
and in the future embryology, especially when
studied as it should be in conjunction with palaeont-
ology, may be confidently relied on to afford a far
clearer insight than we have yet obtained into the
history of life on the earth.

XI

SOME RECENT EMBRYOLOGICAL
INVESTIGATIONS

THE close resemblance between the embryos of
animals which, when adult, are widely different
in form, in size, and even in structure, greatly
impressed the earlier embryologists, and was often
insisted on by them. A reptile, a bird, and a
mammal are in their early stages of development
so closely similar that v. Baer himself was unable
to decide to which of these groups three unnamed
embryos in his collection were to be referred.

A still more striking illustration is afforded by
the controversy which raged for many years over
Krause's famous embryo. In 1875 Krause de-
scribed an early human embryo which appeared to
differ from all known human embryos in having
a large vesicular allantois like that of a chick or
reptile, instead of the thick allantoid stalk by
which the human embryo is normally connected with
the chorion. This peculiarity with regard to the
allantois was so marked that doubts were at once

250 SOME RECENT

raised as to the embryo being really a human one ;
and Professor His, one of the most expert of
embryologists, asserted roundly that Krause must
have made a mistake, and that his specimen was
a chick embryo and not a human one at all. An
ardent, almost furious, discussion arose, and con-
tinued for many years : it is indeed only within
the last twelve months that the points at issue have
been finally put at rest, and it has been shown
that while Krause was right in describing his
embryo as a human one, he was mistaken in regard
to the supposed peculiarity in the allantois, the
bladder-like vesicle which he took for the allantois
being merely a pathological dilatation of the allantoic
stalk.

Among Invertebrates the resemblances between
the early larval forms of allied groups are equally
striking : the veliger and trochophore larvae, for
example, or the nauplius larva of Crustacea, having
not merely very wide zoological distribution, but
presenting marked constancy in essential, or even
in minor characters in the groups in which they
occur.

That the early larval stages of a prawn, a cyclops,
and a barnacle ; or of a reptile, a chick, and a man,
should be so closely similar that it is possible to
mistake one for the other, is undoubtedly a very
remarkable thing; and it was perhaps inevitable
that there should be at first a tendency to over-
estimate the exactness of this resemblance. Em-
bryologists have indeed too often overrun their
facts; and, misled by the undoubtedly striking

EMBRYOLOGICAL INVESTIGATIONS 251

similarity between embryos of forms zoologically
akin, have not hesitated to fill the gaps in their
knowledge of the developmental history of certain
animals, by reference to the known processes in
allied forms. More especially is this the case with
regard to man himself : material for direct observa-
tion is difficult to obtain, and it is only too common
to find that descriptions purporting to be of human
embryos are really founded on observations derived
from the study of pigs and rabbits, or even of
animals so remote zoologically as chicken, lizards,
or dog-fish. Of late years however there has
been a marked reaction in this respect, and the
pendulum has shown a tendency to swing over to
the opposite side. More exact observations on
many groups of animals have proved that even
in allied forms the course of development may be
markedly different ; in some cases indeed not only
genera and species, but even the eggs of the same
brood, may develop in ways curiously unlike one
another.

This inconstancy in the mode of development,
more especially in the earliest stages, is one of the
most striking results of recent embryological inves-
tigation, and may well claim attention on the
present occasion. If at first sight it appear be-
wildering or even unwelcome to those who, from
the study of development of existing animals,
would seek to unravel the past history of the race,
it is to be remembered that facts must always be
accepted ; and it may even be that in this varia-
bility, by which his labours are so greatly extended,

252 SOME RECENT

the embryologist will find the clue he has so
long waited for, which will enable him to distinguish
the real from the spurious, the ancestral from the
acquired characters.

The earliest stage of development in every Meta-
zoon is the act of cell-division, or segmentation as
it is commonly termed, by which the single cell, or
egg, becomes divided into a number of cells or
blastomeres. The details of the process vary greatly
in the eggs of different animals, and are affected
more particularly by the amount of food yolk, or
deutoplasm, present in the egg; this food yolk
offering mechanical hindrance to the division of the
egg. In each species or genus, or even in larger
groups, the mode of segmentation was formerly
assumed to be constant, but the more recent re-
searches have shown that this is by no means
always the case.

In 1878, Kleinenberg noticed that the eggs of an
earth-worm, Lumbricus trapezoides, which he was
investigating, showed considerable individual differ-
ences in the mode and order in which the successive
cell-divisions were effected ; an observation which
was afterwards confirmed by Wilson, and extended
to other species of earthworms. In 1884, while
studying the development of Renilla, one of the Pen-
natulida or Sea- Pens, Wilson found an extraordinary
range of variation in the mode of segmentation of
eggs, even of the same brood. In some cases the
egg divided into two blastomeres in the normal
manner, each of these in its turn again dividing ; in
other cases however the egg divided at once into

EMBRYOLOGICAL INVESTIGATIONS 253

eight, sixteen, or even thirty-two blastomeres, which
in different specimens were approximately equal or
markedly unequal in size. Sometimes a preliminary
change of form occurred without any further result,
the egg returning to its spherical shape, and pausing
for a time before recommencing the attempt to
segment. Segmentation sometimes commenced at
one pole, as in the telolecithal eggs of birds or
reptiles, with the formation of four or five small
segments, the rest of the egg breaking up later,
either simultaneously or progressively, into segments
about equal in size to those first formed ; while
lastly, in some instances segmentation was very
irregular, following no apparent law. Similar
modifications in the segmentation of the egg have
been described in the oyster by Brooks, in Anodonta
and in other Mollusca, and in Hydra. In the
different species of Peripatus there appear also
to be considerable variations in the details of
segmentation.

It has long been known that the eggs of
Amphibia may vary greatly in the mode in which
the early stages of segmentation are effected ; and
recently Jordan and Eycleshymer have described
these variations in detail. The first cleft usually
corresponds with the median sagittal plane, dividing
the egg into two blastomeres, which give rise
respectively to the right and left halves of the
embryo. .The cleft may however be oblique, or
even transverse to this plane. The two first blasto-
meres are usually equal, but may be very unequal,
one being sometimes twice the size of the other.

254 SOME RECENT

The second cleft is usually at right angles to the
first, and divides the egg into anterior and posterior
halves; but it may cut the egg obliquely. The
third and fourth clefts also present considerable
variability in their position and relations to the
earlier clefts. In all the cases described above the
variability was confined to the earliest phases of
segmentation : the earlier stages of development
were the same, whatever the mode in which seg-
mentation was effected ; and apparently identical,
and certainly normal embryos, resulted in all
cases.

Some recent observations of Loeb give a possible
clue to these phenomena. He found by experi-
menting on the eggs of sea urchins that the process
of segmentation could be retarded, or modified, by
varying the proportion of sodium chloride present
in the sea- water in which the eggs were laid.
A slight increase in the normal proportion of sodium
chloride delayed the occurrence of segmentation ;
but on the return of the eggs to normal sea-water
they very quickly divided, often simultaneously,
into a number of blastomeres. He states his results
thus : " If we bring impregnated eggs into sea-
water of a certain higher concentration, no segment-
ation takes place ; but if we bring them back into
normal sea-water, they divide in about twenty
minutes directly into nearly, but not quite so many,
cleavage spheres as they would contain by that time
if they had remained in normal sea-water all the
time." Further investigation showed that, although
when placed in water containing more than the

EMBRYOLOGICAL INVESTIGATIONS 255

normal proportion of sodium chloride, the eggs did
not segment, yet that division of the nucleus
occurred : unsegmented eggs were in this way
obtained with from four to as many as thirty
nuclei.

It is at least possible that some of the variations
in the normal process of segmentation mentioned
above as observed in other animals may be due to
changes in the composition, or perhaps of the
temperature of the water in which the eggs were
developing. Herbst's interesting experiments on
the modifications in the development of Echinoderm
larvae produced by, the addition of potassium or
lithium salts to the sea-water in which they were
contained, led him to conclude that the influence on
the developing ova was not directly chemical, but
was due to the altered osmotic pressure of the sea-
water. The field of enquiry thus opened up is a
most promising one, and is certain to receive the
attention of embryologists in the immediate future.
We are at present ignorant of the causes which
determine the division of a cell, or the segmentation
of an egg ; but we shall have made an appreciable
step towards a right understanding of the pheno-
mena when we have determined with precision in
what way they can be modified by slight, but known,
alterations in the environment.

Variations occur in the later as well as in the
earlier stages of development ; and the differences
between allied genera or species, or even between
closely related individuals, may be very marked,
Among Ccelenterates, for example, the mode of

256 SOME RECENT

formation of the inner germinal layer, or hypoblast,
presents most perplexing modifications : it may
arise as a true gastrula invagination ; as cells
budded off from one pole of the blastula into its
cavity ; as cells budded off from various parts of
the wall of the blastula ; by delamination, or actual
division of each cell of the blastula wall into outer
or epiblastic, and inner or hypoblastic elements ;
or it may be present from the first as a solid mass
of cells enclosed by the epiblast cells. Another
good illustration is afforded by the extraordinary
modifications in the position, and in every detail of
formation of the middle germinal layer, or mesoblast,
in different and often in closely allied forms of the
higher Metazoa : differences which have given rise
to ardent discussion, and have led to the proposal
of theory after theory, each rejected in its turn as
affording only a partial explanation, and finally
culminating in Kleinenberg's protest against the
use of the term mesoblast at all, at any rate in a
sense implying any possibility of comparison with
the primary cellular layers, epiblast and hypoblast,
of Coelenterata. Amongst Vertebrates the frog and
the newt differ greatly in important developmental
points. The stages immediately following segmen-
tation of the egg are very unlike in the two cases.
The epiblast is but one cell thick in the newt, while
in the frog it consists of two distinct layers from
the first ; the ear of the newt develops as a pit-
like depression of the skin, while in the frog it is
from the first a closed sac ; and many other differ-
ences will readily occur to embryologists. In the

EMBRYOLOGICAL INVESTIGATIONS 257

common English frog, Rana temporaria, and in the
closely allied Rana esculentay the branchial blood-
vessels develop in very different manner, although
ultimately reaching the same condition.

But the most remarkable series of facts with
regard to variation in the later stages of develop-
ment is afforded by the recently published observa-
tions of Professor Brooks and Mr. Herrick on the
metamorphoses of certain small decapod Crustacea
of the genus Alpheus. These are essentially
tropical forms, not unlike small crayfish in appear-
ance, and about an inch or so in length. They are
brilliantly coloured and occur most abundantly
in shallow water, more especially amongst coral
reefs. A few species live freely, but the majority
are found inhabiting the tubes of sponges, or dwell-
ing in holes and crannies in the porous coral lime-
stone. The development of thirteen species was
carefully studied by Messrs. Brooks and Herrick,
and the remarkable conclusion was arrived at that
" individuals of a single species sometimes differ
more from each other as regards their metamor-
phoses than do the individuals of two very distinct
species." It is not merely that different individuals
hatch at different stages of development : one indi-
vidual may appear with characters, e.g., the number
and form of the legs, which occur at no stage in
the development of other individuals of the same
species. In some cases the differences in develop-
mental history were associated with differences in
the locality from which the specimens were obtained.
Thus individuals of a given species observed at

R

258 SOME RECENT

Key West, in Florida, differed constantly from
individuals of the same species from N. Carolina,
and these again from individuals living at the
Bahamas. In other instances the differences in
development appeared to be related to the conditions
of life. At the Bahamas a species of Alpheus was
observed dwelling in the chambers of sponges.
Two kinds of sponge were employed by the prawn,
one sponge being green, the other brown. The
prawns living in the green sponge produced con-
siderable numbers of small eggs ; while the indi-
viduals of the same species which dwelt in the
brown sponge gave rise to a few eggs of large size,
from which the young prawns were hatched at a
stage corresponding to that reached at the third
moult by the young developed from the smaller eggs
in the green sponge.

Another very curious series of facts has been
made known to us by the careful researches of the
Russian embryologist, Salensky, on the early
development of certain Ascidians. The normal
course of development in a Metazoon, as is well
known, is that the fertilised egg segments, i.e.,
divides into a number of nucleated cells from which,
by further division, and accompanying differentia-
tion, all the various parts and tissues of the embryo,
and finally of the adult animal, are produced.
Every individual cell of the adult, whether an
epithelial cell, a nerve cell, a muscle cell, or a bone
cell, owes its origin to direct descent from the
original egg-cell or ovum. The most striking fact
in the whole range of embryology is that all

EMBRYOLOGICAL INVESTIGATIONS 259

animals above Protozoa commence their existence
as eggs — i.e., as single cells — from which all parts of
the adult body are ultimately derived. The pur-
pose of this arrangement appears to be to secure
the advantage to be derived from cross fertilisation,
a process which concerns two cells only, male and
female respectively ; and which necessitates a uni-
cellular stage as the initial one in individual de-
velopment. The eggs of Metazoa, while still in the
ovary, are very commonly enclosed in follicles.
These consist of one or more layers of cells which
are of epithelial origin, and which are of the same
order as the ova themselves, and in their earlier
stages indistinguishable from these. The follicle
cells play an important part in the nourishment of
the egg which they surround ; but they are left
behind in the ovary, or are rubbed off after the egg
has ripened and discharged from the ovary, and
they have nothing to do with the formation of the
embryo.

In the genus Pyrosoma however, a well-known
colonial and pelagic Ascidian, Salensky discovered
a very peculiar condition of things. Each Pyrosoma
individual produces one very large egg, which is
closely invested by a follicle or capsule. The egg
is meroblastic ; and as in the hen's egg, segmenta-
tion is confined to one pole, and results in the
formation of a small cap of cells or blastoderm^
lying on the top of the mass of unsegmented yolk
formed by the rest of the egg. As segmentation
proceeds, certain of the follicle cells investing the
egg grow in between the segmentation cells or

26o SOME RECENT

blastomeres. These follicle cells, or kalymmocytes
as they are called by Salensky, are at first very
distinct in appearance from the blastomeres, and
easily recognised from these ; but as development
proceeds the differences become less and less
marked, and finally cease to be evident. These
ingrowing follicle cells, or kalymmocytes, enter
directly into the formation of the embryo, which is
thus built up partly from cells derived, as in other
Metazoa, from division of the fertilised egg or
ovum ; and partly from cells of independent origin,
the kalymmocytes, which are unfertilised, and have
grown into the egg from the surrounding follicle.
In the genus Salpa a similar condition of things
has been described by Salensky ; the only difference,
as compared with Pyrosoma, being that in Salpa
the kalymmocytes are actually more numerous and
more bulky than the blastomeres formed by seg-
mentation of the ovum, so that the major part of
the embryo is composed of unfertilised cells.

These observations are of the greatest possible
interest, and we wait anxiously to learn whether
this curious mode of development is confined to the
groups of Ascidians, or whether it occurs in other
animals as well. It is perhaps worth while
suggesting that if the process were carried one
stage further than it actually is in Salpa, we should
arrive at a condition of things curiously resembling
the mode of formation of gemmules in Sponges, or
of statoblasts in Polyzoa. In Pyrosoma, the follicle
cells or kalymmocytes form part, but the smaller
part, of the embryo. In Salpa, the kalymmocytes

EMBRYOLOGICAL INVESTIGATIONS 261

are more abundant than the blastomeres, so that the
greater part of the embryo is formed from the
unfertilised kalymmocytes or follicle cells. If now
we imagine this carried one step further, if we sup-
pose the blastomeres, or fertilised elements, to be
completely absent, then the whole embryo would
be formed by an aggregation of unfertilised cells,
and the process would become most suggestively
similar to that by which a Sponge gemmule is formed.
The last series of phenomena to which I wish
to refer are those resulting from the natural
or artificial division of an egg, in the early
stages of its development, into two or more
fragments. In 1869, Haeckel, while studying the
development of Crystallodes, a genus of Siphono-
phora or pelagic Hydroids, was struck with the
fact that the polyhedral cells of which the egg
consisted at the close of segmentation exhibited
active amoeboid changes of shape, and appeared to
possess a certain amount of independence. The
idea occurred to him to test this power of inde-
pendent existence by breaking up the embryo into
fragments, and following their fate. By means of
needles, eggs at the close of segmentation on the
second day of development were broken into two,
three, or four pieces, and it was found that these
not only lived for eight or ten days, but developed
and gave rise, in almost normal manner, to
rudimentary Siphonophoran colonies. These ob-
servations of Haeckel's are the earliest experiments
on lines which have recently led to remarkable
results.

262 SOME RECENT

Kleinenberg, in 1878, published an account of
the early stages of development of an earthworm,
Lumbncus trapezoides, which he obtained abun-
dantly in the gardens of the island of Ischia. The
worms lay their eggs in capsules : each capsule is
from one to eight millimetres in length, and is
filled with an albuminous mass in which are
contained bundles of ripe spermatozoa, and from
three to eight eggs. Of the eggs only one
develops as a rule, the remainder gradually dis-
appearing. At an early stage in development, as
soon as the germinal layers are established, and
before the appearance of any of the organs or
parts of the embryo, the ovum divides into two
parts which are usually equal and similar. Each
oT these parts develops into an embryo, and sub-
sequently into an adult worm ; the two remaining
for a time connected together like Siamese twins,
but ultimately separating.

The upshot of the process is that what Haeckel
effected artificially for Crystallodes — i.e., the division
of the segmented egg into two parts each of which
develops into an embryo — is found to be a natural
or even normal occurrence in Lumbricus trapezoides.
Kleinenberg supposed that the tendency to form
twins, which is so marked a feature in Lumbricus
trapezoides, was due to a process of double fertilis-
ation, two spermatozoa instead of one being con-
cerned in the act. Vejdovsky however suggested
that the twinning was perhaps influenced by warmth,
for it occurred most frequently in warm weather ; a
suggestion which receives much support from some

EMBRYOLOGICAL INVESTIGATIONS 263

experiments made by Driesch on Echinoid eggs,
which when artificially warmed were found to have
a marked tendency to develop twin-embryos.

Another line of research was initiated independ-
ently by Chabry and by Roux, who investigated
the effects of injuring or destroying one of the two
or four blastomeres resulting from the first or
second cleavage of the egg. Chabry, whose results
were published in 1887, use<^ f°r h*5 experiments
the egg of an Ascidian, Ascidia aspersa. Taking
eggs in which segmentation had just commenced,
and division into two cells had been effected, he
destroyed one of the two cells by pricking with a
needle. Under such circumstances he found that
the surviving cell developed into a half-embryo.
By experimenting in similar manner on eggs which
had divided into four cells or blastomeres, he was
able to produce either quarter, half, or three-quarter
embryos. Chabry concluded from his results that
each blastomere, at any rate in the early stages of
development, has a determined destiny and repre-
sents a definite part of the larva. This view was
first propounded by Professor His who, in 1874,
maintained the existence of " special regions in the
germ, which give rise to special organs," and held
that each organ of the embryo is represented by a
definite part of the body of the egg.

This view is closely similar to the old doctrine of
" Evolution" or "Preformation," according to which
it was held that all the organs and parts of
the embryo were already present in the egg when
laid; and that development consisted in an unfolding

264 SOME RECENT

and perfecting of these parts, much as the flower is
formed by the expansion of parts already present in
the bud. This doctrine of Preformation, in its
original form, was overthrown in 1759 by Wolff,
who substituted for it the doctrine of Epigenesis — />.,
that there is no trace of the embryo or of any of its
parts in the egg, and that the formation of the
embryo is an entirely new process. It is not
a little curious to find the older doctrine, which
embryologists thought disposed of for ever, coming
up again in somewhat modified form, as a result of
more recent investigations.

Almost simultaneously with Chabry, Roux in-
vestigated in similar manner, but in more detail, the
results of localised injuries to frog embryos at early
stages of development. His method consisted in
destroying one or more of the cells of a segmenting
egg by puncturing them with a hot needle. He
took a frog's egg at the completion of the first cleft —
i.e., an egg which had just divided into the first two
cells or blastomeres — and destroyed one of the two
cells. The surviving cell developed into a half-
embryo, from which by a process of regeneration
the missing half was gradually formed, a whole
embryo ultimately resulting. A more satisfactory
method of experimenting, inasmuch as it does not
involve the destruction of any of the cells, consists
in shaking apart the blastomeres at an early stage,
and then following their subsequent fate. This was
first done, in 1877, by Chun, and has since been
repeated by other observers. Chun experimented
with the eggs of Ctenophora, and found that if the

EMBRYOLOGICAL INVESTIGATIONS 265

two first segmentation cells, or blastomeres, were
separated by shaking, each developed into a half-
larva, which actually became sexual, and which
ultimately regenerated the missing half by a process
of budding.

Driesch, in 1891, carried out a similar but more
complete series of experiments at Trieste, employing
for the purpose the eggs of sea-urchins, chiefly
Echinus microtuberculatus and Sphcerechinus granu-
laris. Selecting eggs which had just completed the first
division, into two segmentation cells or blastomeres,
he put from fifty to a hundred in a small quantity
of sea-water in a glass tube about an inch and a
half long and a quarter of an inch in diameter. By
vigorously shaking the tube for five minutes or
longer he succeeded in rupturing the egg mem-
branes, and isolating the blastomeres. The contents
of the tube were then poured into a shallow vessel,
and the isolated blastomeres picked out with a fine
pipette, and placed in separate watch-glasses, in
which their further development could be followed.
Each blastomere so treated developed at first into
a half-embryo. By the evening of the first day a
hemi-blastula was formed. During the night a
change took place, and by the following morning
each hemi-blastula had become a complete blastula
or sphere, but of half the normal size. By the end
of the second day invagination commenced. Each
blastula became a gastrula in the normal manner ;
and the formation of the various organs, mouth,
arms, coelom, and ambulacral system was effected in
the typical fashion. In some cases the shaking had

266 SOME RECENT

effected imperfect separation of the two blastomeres,
without rupturing the egg membrane. From these
eggs either twins or double embryos were produced,
a result which suggests that the twinning which
occurs so commonly in Lumbricus trapezotdes may
not after all be a result of double fertilisation as
was supposed by Kleinenberg.

The most recent and the most complete of the
experiments on the results of shaking apart the
blastomeres of segmenting eggs are those performed
by Wilson on Amphioxus eggs in 1892. The
eggs of Amphioxus are very minute, about O'l mm.
in diameter ; and their early stages of development
are extremely simple. The egg divides by a
vertical cleft into two equal blastomeres : by a
second vertical cleft, at right angles to the first,
each of the two blastomeres is bisected, and four
blastomeres of equal size result. The third cleft
is a horizontal one, and is rather nearer the upper
pole than the lower : by it each of the four
blastomeres is divided into an upper and rather
smaller cell, and a lower and rather larger one.
In the later stages the number of cells is rapidly
increased ; a blastula, and then by invagination
a gastrula is formed, and the several organs of the
embryo quickly appear. In the early stages the
blastomeres hang together very loosely, and are
easily separated by shaking. Commencing with
the stage at which two blastomeres are present,
Wilson isolated them by shaking, and found that
each developed as though it were a complete egg,
segmenting in the normal fashion, giving rise to

EMBRYOLOGICAL INVESTIGATIONS 267

a blastula, and then a gastrula, and ultimately
becoming a larva, which differed from a normal one
merely in being half the usual size.

Wilson next took the stage with four blasto-
meres, and found that each of the four, if isolated,
developed into a larva of typical form, but one-
fourth the usual size : if two or three of the
blastomeres held together a larva resulted which
was half or three-quarters the normal size. It
appears therefore that at the stage with four
blastomeres, either one, two, three, or all four of
the blastomeres have the power of developing into
an embryo which will be normal in all respects,
saving only as regards size. Anxious to deter-
mine the limits to which this extraordinary process
could be continued, Wilson tried the next stage,
that in which eight blastomeres are present, of
which four are rather smaller, and four rather
larger in size. Each blastomere when isolated
commenced to develop, but never became a
complete embryo. Flat plates, curved plates, even
blastulas one-eighth the normal size were formed,
which swam about freely by means of cilia, but
which underwent no further development.

These results are of very great interest. They
must I think be regarded as fatal to the doctrine
of " Evolution " in its new form, as maintained by
His and Weismann ; for if any one of the first four
blastomeres is able by itself to develop directly
into an embryo, it seems impossible to hold that
each blastomere has a predestined part to play in
the formation of the embryo.

268 SOME RECENT

Another very interesting result is the precise
point at which the power of developing an embryo
from a single blastomere ceases. Discussion has
taken place as to whether the failure is due to
quantitative or to qualitative considerations ; to an
insufficiency in amount of living matter, or to
incompleteness in its structure or composition.
Wilson is of opinion that the difficulty is mainly
a qualitative one, and has advanced arguments in
support of his view. Perhaps the most important
consideration in its favour is that so long as the
blastomeres are precisely alike — i.e., up to the stage
with four blastomeres — each one possesses the
power when isolated of developing into an
embryo ; while as soon as a distinction between
smaller and larger blastomeres appears — t.e., at the
third cleavage — this power ceases.

This last consideration becomes still more
significant if we regard a Metazoon as comparable
to a colony of Protozoa, in which differentiation
between the component units, and consequent
mutual dependence, and the necessity for holding
together have become established ; and if further
we view the early development of a Metazoon as a
shadowing of the process by which this condition
was originally attained. So long as the cell units
or blastomeres remain identical, so long does each
one retain the power of independent existence and
development ; but as soon as differentiation is
established between one and another this power of
separate life ceases. This view, though extremely
suggestive, must however be regarded as entirely

EMBRYOLOGICAL INVESTIGATIONS 269

provisional, until we gain more complete and exact
knowledge concerning the conditions under which
the power of independent development can be
exerted, not only in Amphioxus, but in other
animals as well.

The present paper is merely a record of some
recent investigations, and has no pretence to
completeness. These researches however all
tend in one direction, namely to emphasise the
fact that the development of animals takes place
in far more varied manner than is generally
supposed, and that marked differences may occur
both in the earlier and the later stages of the
embryological history, not merely in allied genera
and species, but even between individual members
of the same brood. I will conclude with a brief
summary of the principal modes in which the
development of a Metazoon may be effected. It
should be noted that the adult structure as a rule
gives no indication of the particular mode in which
development has occurred : a sponge developed
from a gemmule is, so far as we know, indistin-
guishable in its adult form from one developed from
a fertilised egg.

AN ADULT METAZOON MAY BE DEVELOPED.

i. From a group of (mesoblast ?) cells,
which are originally independent of one
another, and of which none are fertilised.
Examples: The gemmules of Sponges,
or the statoblasts of Polyzoa.

270 EMBRYOLOGICAL INVESTIGATIONS

2. Partly from a fertilised ovum, and partly

from unfertilised follicle cells. Examples:
Salpa and Pyrosoma.

3. From a single fertilised ovum, together
with a greater or less number of yolk
cells, which are at first independent
cells, but which ultimately become
absorbed into the ovum. Examples:
Fasciola, Pisidium.

4. From a single fertilised ovum, aided by

nutrient matter derived from the sur-
rounding follicle cells. Examples : Ce-
phalopoda, and most Vertebrates.

5. From a single fertilised ovum, without

the aid of follicle cells. Example:
Amphioxus.

6. From an unfertilised cell, or ovum. £!r-

amples : The parthenogenetic ova of
Entomostraca or of Rotifera.

7. From one or more of the blastomeres

resulting from the segmentation of a
fertilised ovum. Examples: Echinus,
Amphioxus.

8. From a portion, large or small, of the

body of the parent. Examples : Hydra,
Sponge.

9. By local proliferation of the cellular

elements of the parent, forming an
external or internal bud. Examples :
Sporocyst, Hydra.

XII

DEATH

THE subject of my address is I admit open to
objection from more than one standpoint. It is
commonly regarded as dull and uninviting ; and
although in the long run it must concern us all,
there would seem to be little reason for urging
its consideration prematurely. Then again it
may be objected that having, from the nature of
the case, no personal experience of the pheno-
menon, I am not in a position to speak about it
with authority.

I can only reply that a matter which concerns
vitally all living things, animal or vegetable, cannot
be devoid of interest. The problems of life are the
most fascinating of all scientific enquiries ; and
surely death, the cessation of life, must have some-
thing to teach us, must throw some light on the
nature of life itself. The beginnings of life are at
present hidden from us ; the other end of the series,
the termination of life, we have daily opportunities
of studying. With regard to the second objection,

272 DEATH

it is true that from personal experience I can say
nothing on the subject of death ; but this is a
disqualification which I share with all members of
our Society, and with all living men. If nobody is
to be allowed to talk about death until he has
qualified by personal acquaintance with the " fell
sergeant," it is clear that any knowledge to be
derived, any lessons to be learnt from its study, are
lost to us for ever.

My choice of the subject has been determined
mainly by the consideration that of recent years it
has attracted considerable attention, and has given
rise to interesting speculations, many of which are
based on facts made known to us by those extra-
ordinarily minute investigations into the structure
of the lower animals, which the modern improve-
ments in microscopical methods and appliances have
rendered possible. My purpose this evening is to
give a summary of these more recent contributions
to our knowledge of the nature and causation
of death, and an indication of the paths along
which it seems probable that advances will be made
in the future.

In a scientific enquiry it is above all things
desirable to have clear ideas as to the nature of the
subject we are dealing with. Unfortunately it is
not easy to define with any degree of precision what
it is that we understand by death. Gotte regards
death as something inherent in life itself, a view
which is held more or less explicitly by the majority
of mankind. This is however disputed by Weis-
mann, whose contributions to the subject are of

DEATH 273

importance. Weismann defines death as "an arrest
of life, from which no lengthened revival, either of
the whole or any of its parts, can take place ; or to
put it concisely, as a definite arrest of hfe." " The
real proof of death," according to Weismann, "is
that the organised substance which previously gave
rise to the phenomena of life, for ever ceases to
originate such phenomena." He adds to this the
corollary that death involves the presence of some-
thing dead — i.e., a corpse. Weismann next challenges
the statement that death is a necessity, inseparable
from the idea or existence of life. He calls atten-
tion to the conditions which obtain among animals
such as Protozoa, in which reproduction is normally
effected by fission ; and points out that in the life-
history of such animals natural death does not occur.
An Amoeba for example reproduces by simply
dividing into two. In such an act of fission the
parent generation disappears, but nothing has died.
If the original Amoeba be called Tom, and the
products of fission Dick and Harry, the upshot of
the process may be expressed by saying that Tom
has disappeared without having died, while Dick
and Harry have come into existence without having
been born. Nothing has died, there is no corpse
to bury, and our ordinary ideas with regard to
individuality and identity fail altogether to afford
answer to the question — Where is Tom at the end
of the process ?

Hence arises the idea of the immortality of the
Protozoa. An Amoeba or other Protozoon repro-
ducing by simple fission can indeed be killed, as

s

274 DEATH

by boiling the water in which it is contained ; but
it does not, in the ordinary course of events, die.
The production of one generation involves the dis-
appearance but not the death of the parent genera-
tion. From the first Amceba to the present day
there has been direct continuity of living matter.
Death may occur through violence, but it is not a
necessary accompaniment or consequence of life.
Moreover death, when it does happen in the case
of an Amceba, causes a final interruption, an abso-
lute break in the chain. No Amceba that has died
has left offspring, for such offspring can only arise
by the division of the living body of the parent.
" No Amceba," it has been well expressed, " has
ever lost an ancestor by death."

The above considerations, which clearly apply
not only to Protozoa but to any other animals which
reproduce by fission, form the basis on which Weis-
mann has built up his theory of the origin of death.
This theory may be briefly summarised as follows :
Protozoa, reproducing by fission, are immortal in
the sense that death does not occur as a necessary
or natural termination of the life-cycle. Natural
death must therefore be limited to Metazoa, or
multicellular animals. In Metazoa a distinction is
always found between somatic cells and reproduc-
tive cells ; the former being the component elements
of which the body of the individual is constructed,
while the latter are the units from which the
individuals of the next succeeding generation will
be developed. The somatic cells are concerned
with the existence and welfare of the individual ;

DEATH 275

the reproductive cells with the perpetuation of the
species.

The normal life-cycle of a Metazoon is as follows :
The fertilised egg, or reproductive cell, by repeated
division gives rise to a number of cells, of which
some become the somatic cells — i.e., the body of the
individual animal — while others remain as its repro-
ductive cells ; from these latter, in due course and
in similar manner, the individuals of the next
generation are formed. Of the two kinds of cells,
the somatic cells alone are liable to natural death :
the reproductive cells survive as the individuals of
the succeeding generation. The reproductive cells
of Metazoa are therefore immortal, in exactly the
same sense as are Amoebae. The reproductive
cells, like the Amoebae, can be destroyed, but they
do not die naturally. Each reproductive cell is
derived by fission from a corresponding cell of the
preceding generation ; and in Metazoa, as in
Protozoa, there has been from generation to genera-
tion direct continuity of living matter. We are
apt to think of the somatic cells — i.e., the body of
the individual animal — as the essential part, by
reason of its greater bulk and impressiveness ; and
to regard the reproductive cells as structures whose
purpose it is to give rise to the somatic cells — i.e.,
the individuals of the next generation. In doing
so however we lose sight of the true relation
between the two groups of cells. The reproductive
cells are the really essential elements, and the part
of the somatic cells is a subordinate one ; their
purpose being to nourish and protect the repro-

276 DEATH

ductive cells in such way as to afford them the best
chance of completing their special duty, the per-
petuation of the species.

The above considerations bring us to the final
point in Weismann's argument. If natural death
affects the somatic cells only, and is a character
acquired by them, it remains to inquire why such
natural death should occur, and what determines
the time of its occurrence. Weismann answers
these questions by saying that death occurs because
it is advantageous to the species that it should do
so ; and that the normal time for such death to
occur is the end of the reproductive period of the
individual. Both these points require further con-
sideration. With regard to the former it is of great
importance to distinguish clearly between what is for
the good of the individual on the one hand, and on
the other hand what is advantageous for the species.
A good illustration is afforded by the elaborate
provision which insects make for their offspring,
which they will never see. Certain wasps have
the habit of stinging the larvae of beetles in their
nerve centres in such manner as to paralyse their
victims without killing them. On the body of the
paralysed larva a single egg is laid by the wasp,
and then left to its fate. From the egg a grub is
hatched in due time, which at once begins to suck
the juices of the larva ; the victim supplying it with
food sufficient for the whole period of its develop-
ment. The grub changes to a pupa on the skin of
its victim, and passing through the winter in the
pupa state, emerges in the spring as a wasp with

DEATH 277

the same instincts and habits as its parent.
Difficulty is sometimes felt in accounting for such
instincts. The individual wasp, it is true, derives
no advantage whatever from its ingenuity ; but the
gain to the species is enormous ; and the preserva-
tion of the habit is due to the fact that those indi-
viduals which took the greatest care to make pro-
vision for their young would be most likely to give
rise to offspring which would survive in the struggle
for existence. Natural selection would tend to
preserve the instinct because it is advantageous to
the species, though of no benefit whatever to the
individual.

So with regard to death; Weismann argues
that the origin thereof is to be found in the con-
sideration that it is advantageous to the species
that individuals should die. The argument is
perhaps best stated in his own words. "Let us
imagine," he says, " that one of the higher animals
became immortal ; it then becomes perfectly obvious
that it would cease to be of value to the species to
which it belonged. Suppose that such an immortal
individual could escape all fatal accidents through
infinite time, — a supposition which is of course
hardly conceivable. The individual would never-
theless be unable to avoid from time to time slight
injuries to one or another part of its body. The
injured parts could not regain their former integrity,
and thus the longer the individual lived the more
defective and crippled it would become, and the less
perfectly would it fulfil the purpose of its species.
Individuals are injured by the operation of external

278 DEATH

forces, and for this reason alone it is necessary that
new and perfect individuals should continually arise
and take their place, and this necessity would re-
main even if the individuals possessed the power
of living eternally." " From this follows, on the
one hand the necessity of reproduction, and on the
other the utility of death. Worn-out individuals
are not only valueless to the species, but they are
even harmful, for they take the place of those which
are sound. Hence by the operation of natural
selection the life of our hypothetically immortal
individual would be shortened by the amount which
was useless to the species. It would be reduced to
a length which would afford the most favourable
conditions for the existence of as large a number as
possible of vigorous individuals at the same time."

The passage just quoted is extremely ingenious,
but is hardly convincing, for it does not attempt to
explain the real nature of death, nor how it came
about in the first instance. The distinction between
somatic and reproductive cells is a real one in
Metazoa. The actual steps by which it was estab-
lished have yet to be traced : but it would seem
probable that in the earliest Metazoa all the cells
originally retained their reproductive power, and
that the process by which this power became re-
stricted to particular groups of cells was a gradual
one. The gemmules of sponges, or the statoblasts
of Polyzoa, are perhaps to be interpreted as
examples of the retention of reproductive power
by groups of cells which in allied animals have
become exclusively somatic in character ; and

DEATH 279

similar instances could be quoted from the vegetable
kingdom.

Weismann's contention that the reproductive
cells of Metazoa are immortal in the same sense as
an Amoeba, must also be admitted to be established.
He however leaves altogether undecided the ques-
tion of the way in which death of the somatic cells
arose in the first instance ; and we shall find later
on that there are strong grounds for holding that
natural death appeared first, not as Weismann sup-
posed among Metazoa, but in the Protozoa them-
selves.

Before considering these more recent aspects of
the problem it will be well to refer briefly to Weis-
mann's views, which have already been mentioned,
in regard to the causes determining the duration of
life in different animals. According to Weismann
the duration of life in a given case is that which is
most advantageous for the species. In the simpler
cases death occurs at the close of the reproductive
period. In the silkworm moth, or the May fly, the
adult existence is only of a few hours' duration ;
the insect laying all its eggs simultaneously, and
then dying. In other insects, as in many of the
hawkmoths, and in most butterflies, the eggs are
laid at intervals and in different places ; in such
cases the life of the adult is prolonged until a
sufficient number of eggs have been laid to ensure
the perpetuation of the species, and then the insect
dies. In birds, owing to the small number of the
eggs that are produced at any one time, and the
great destruction to which the eggs are liable from

28o DEATH

the attacks of enemies, several years may elapse
before enough eggs are produced to ensure survival
of the species ; and the life of the adult is conse-
quently prolonged for many years. A further
lengthening of life takes place in mammals and other
animals which tend or rear their young, either by
retaining the eggs or embryos within their bodies
for a longer or shorter portion of their development,
or by protecting and feeding the young after birth.

The whole subject thus opened up is of extreme
interest, but it would be impossible to treat it
adequately on the present occasion. Weismann
himself has accumulated a large number of statistics
with regard to different groups of animals, which
lead him to the conclusion that " the end of the
reproductive period is usually more or less coinci-
dent with death ; " while in cases in which the
duration of life is prolonged, owing to the parent
tending or nursing the young after birth, he con-
cludes that " as a general rule the increase in
length of life is exactly proportional to the time
which is demanded by the care of the young."

With regard to the actual cause of death in
Metazoa, Weismann suggested in his earlier essay
published in 1881, that death may be due to the
somatic cells losing the power of reproduction by
cell-division after a certain number of generations.
In 1883 in a further essay on life and death, he
repeated this suggestion and developed it more
fully. After referring to his former view that the
varying duration of life in the animal kingdom " is
determined in different species by the varying

DEATH 281

number of somatic cell-generations," he frankly
admits that he is quite unable to indicate the
changes in the physical constitution of protoplasm
upon which the variations in the capacity for cell-
division depend, or the causes which determine the
greater or smaller number of cell-generations ; but
he urges that if we must wait until we understand
the molecular structure of cells before advancing
views concerning the nature and limits of their
activities we shall probably never solve the
problem. " Therefore," he continues, " it is in my
opinion an advance if we may assume that length
of life is dependent upon the number of generations
of somatic cells which can succeed one another in the
course of a single life ; and furthermore that this
number, as well as the duration of each single cell-
generation, is predestined in the germ itself. This
view seems to me to derive support from the obvious
fact that the duration of each cell-generation, and
also the number of generations, undergo con-
siderable increase as we pass from the lowest to
the highest Metazoa."

This bold suggestion, based entirely on theo-
retical considerations, received striking confirmation
a couple of years later, from the results of Maupas'
famous researches into the reproduction of Infusoria.
The normal mode of reproduction among the ciliated
Infusoria is by means of fission, essentially similar
to the fission of Amoebae. In addition to this a
process of conjugation has long been known to
occur in Infusoria, though its real nature has been
much disputed. Balbiani described the process in

282 DEATH

great detail as long ago as 1858, and was led to
the conclusion that it was a true sexual act, com-
parable to fertilisation in the higher animals. Bal-
biani's views were for a long time discredited,
mainly through the unwillingness of zoologists to
admit the possibility of sexual reproduction occur-
ring in unicellular animals. More recent researches
however and in particular those of Maupas, have
shown that while Balbiani was not absolutely right
with regard to all the details of the process, yet
that he was correct in his interpretation of the
conjunction of Infusoria as a true sexual act.

The problem which Maupas set himself to solve
was to determine the relation between the two
modes of reproduction in Infusoria, to ascertain the
conditions under which asexual and sexual repro-
duction respectively occur, and to find out what
causes lead to the substitution at particular times
of one process for the other. To do this it was
necessary to isolate an individual Infusorian, to
place it under known conditions as to temperature
and food supply, and then to follow the fate of the
successive generations of offspring to which it gave
rise by fission.

The investigation was an extraordinarily labo-
rious one. Continuous observations for over five
months were necessary ; the production and fate
of from 200 to 300 successive generations had to
be followed accurately, and in some cases the
observations extended over more than 600 genera-
tions. The most suitable temperature at which to
conduct the experiments, and the kind of food best

DEATH 283

suited to the particular Infusorian under observa-
tion, had in each case to be determined. No less
than twenty different species were experimented
upon, and the whole research, both as regards
the extreme patience necessary for its proper
conduct, and the great importance of the results
obtained, justly ranks among the most famous of
its kind.

The general results of Maupas' investigation,
which deserve to be followed in detail by all micro-
scopists, were to this effect : In conjugation the
paranucleus of each of the conjugating individuals
acts as a hermaphrodite sexual element : it under-
goes successive divisions, and parts are extruded
and lost completely. The remaining parts of each
paranucleus become differentiated into male and
female pronuclei ; interchange of the male pronuclei
takes place between the conjugating individuals,
and the male pronuclei then fuse with the female
pronuclei of the individuals to which they have
been transferred. The entire nuclear apparatus of
each of the conjugating individuals is now recon-
stituted, and the two Infusoria separate and
become independent once more. The purpose of
conjugation appears to be to stimulate the asexual
act of fission. Individuals which previous to con-
jugation showed no tendency to divide, begin to
do so actively as soon as conjugation has been
completed. In order however for conjugation to
be effective, Maupas finds that it must take place
between individuals which are not closely related
to each other. Members of the same family show

284 DEATH

no tendency to conjugate with each other, and
appear indeed to be incapable of doing so, for
attempts at such conjugation have been found to
prove abortive.

By his method of isolating Infusoria and follow-
ing the fate of the successive generations produced
by fission, Maupas was enabled to prove that there
are natural and definite limits to the continuance
of asexual reproduction. An Infusorian, Stylonychia
pustulata, was isolated in November 1885, and the
successive generations followed until March 1886.
By that time there had been 215 generations pro-
duced by successive acts of fission. Conjugation
had not occurred, since nearly related individuals
will not conjugate, and unrelated forms were
excluded by the conditions of the experiment.
Towards the close of the experiment the tendency
to multiply by fission became less manifest, and the
individuals themselves were in a condition well
described as senescent ; they were of reduced size,
often distorted in shape, or actually malformed ;
their nuclear apparatus was in a degenerate con-
dition, and they no longer had the power, which
all the younger generations possessed, of conjugat-
ing effectively with unrelated individuals when
transferred to water containing such. Finally the
nutritive powers of these senescent members failed
and death of these exhausted forms brought the
experiment to an end. The same results were
obtained from experiments conducted in a similar
way with other Infusorians. The general conclu-
sions arrived at were : (i.) That in those Protozoa

DEATH 285

which reproduce both by fission and by a sexual
process of conjugation there are definite limits to
the number of generations which can be produced
asexually; (ii.) that conjugation only occurs be-
tween individuals which are not nearly related to
each other ; (iii.) that if conjugation with unrelated
forms be prevented, senescence, and finally death
occur.

These results are of the utmost interest in connec-
tion with Weismann's views regarding the nature
and origin of death. They show that Weismann
was wrong in supposing that death occurred first
amongst Metazoa. Natural death occurs among
Protozoa ; and the tendency to it and inability to
escape from it, are probably inherited by Metazoa
from their Protozoon ancestors. On the other hand,
Maupas' results confirm in the fullest manner Weis-
mann's bold suggestions, (i.) that the original
occurrence of death is intimately connected with
sexual reproduction, if not indeed an actual conse-
quence of it; (ii.) that the number of generations
of somatic cells which can succeed one another in
the course of a single life may be strictly limited.
Maupas' experiments seem to me to afford the very
evidence of which Weismann was in search. They
prove that amongst Infusoria asexual reproduction
by cell division cannot be continued indefinitely, but
that it leads in time to senescence and ultimately to
death.

If we apply these results to Metazoa the conclu-
sions become very striking. In Metazoa, as in
Infusoria, there is alternation of sexual and asexual

286 DEATH

modes of reproduction. The fusion of male and
female pronuclei in the act of fertilisation of the egg
is the sexual process, and is equivalent to the
similar fusion of male and female pronuclei of
unrelated cells, seen in the conjugation of Infusoria.
On the other hand the successive acts of cell
division, by which the fertilised egg gives rise to
the embryo and the embryo becomes converted into
the adult, are asexual processes, equivalent to the
repeated acts of cell division by which the successive
generations of the Infusorian are produced. In the
Infusorian the number of such asexually produced
generations that can succeed one another is limited ;
so also is it in the Metazoon ; and the gradual
failure of the power to divide further leads in both
cases alike first to senescence or old age, and
ultimately to death.

This comparison between Protozoa and Metazoa
in regard to the modes in which reproduction is
effected appears to be a just one. The striking
difference, that in the Protozoon the products of the
asexual process of cell division become independent
and similar unicellular animals, while in the Meta-
zoon they are component and differentiated units in
the body of a multicellular animal, does not affect the
comparison so far as concerns the essential point —
i.*., the mode in which successive cell generations
come into existence in the two cases alike. A
further point of difference is- found in the consider-
ation that in the Infusoria all the asexually produced
cells retain, at any rate for a number of generations,
the power of conjugating with other cells ; while in

DEATH 287

Metazoa the power appears to be lost very early by
the majority of the cells, and retained only by the
reproductive cells. This however does not in any
way invalidate the comparison, and is merely an
example of that structural and physiological differ-
entiatiation which distinguishes Metazoa from
colonial Protozoa, and which affords the key to all
Metazoon structure.

Moreover, it is at present entirely unknown to
us at what period or to what extent somatic cells of
a Metazoon lose their power of conjugating. From
this standpoint it would be of the greatest interest
to know precisely what happens in cases of intro-
duction of cells from without into a living Metazoon;
for example, in vaccination or in other methods of
inoculation ; or in cases of transfusion of blood on
a large scale. Theoretically it seems possible that
rejuvenescence of the somatic, as of the reproductive
cells of a Metazoon might be effected, and a new
lease of life obtained for these cells and their
descendants. This however is a matter of mere
speculation, and not to be lightly entered upon.
The general conclusions to which these recent
investigations have led us may be briefly summarised
as follows : —

i. Death is not an intrinsic necessity, either

of life or of organisation,
ii. .Natural death first appeared, so far as we
know at present, among the higher
Protozoa.

iii. Death is closely associated with the occur-
rence of conjugation, and the consequent

DEATH

alternation of sexual and asexual modes
of reproduction.

iv. The asexual mode of reproduction, by
fission, is the more primitive one. Con-
jugation, or sexual reproduction, gives an
advantage in the struggle for existence;
and at first a luxury, has through the
action of natural selection become a
necessity.

v. The normal duration of life of a given
species is that which is most advantage-
ous to the species — i.e.t an animal dies
when it has produced sufficient young to
ensure the perpetuation of the species
under existing conditions.

vi. In Metazoa, as in Protozoa, there is direct
continuity of living matter by cell division
from generation to generation.

vii. The statement that " no Protozoon has
ever lost an ancestor by death" may now
be extended thus : There is not a single
component cell in the body of a Metazoon
that has ever lost an ancestor by death.
For each component cell in the body of
a Metazoon is descended by direct fission
from the egg. The egg was a body cell
of the parent, and in its turn was derived
by cell division from the egg cell from
which the parent was developed ; and so
on, generation behind generation, there
has been unbroken continuity of living
matter.

XIII
THE RECAPITULATION THEORY

As my theme for this morning's address I have
selected the Development of Animals. I have
made this choice from no desire to extol one par-
ticular branch of biological study at the expense of
others, nor through failure to appreciate or at least
admire the work done and the results achieved in
recent years by those who are attacking the great
problems of life from other sides and with other
weapons. My choice is determined by the neces-
sity that is laid upon me, through the wide range
of sciences whose encouragement and advancement
are the peculiar privilege of this Section, to keep
within reasonable limits the direction and scope of
my remarks ; and is confirmed by the thought that,
in addressing those specially interested in and
conversant with biological study, }^our President
acts wisely in selecting as the subject-matter of his
discourse some branch with which his own studies
and inclinations have brought him into close
relation.

T

290 THE RECAPITULATION THEORY

L_Embryology, referred to by the greatest of
naturalists as " one of the most important subjects
in the whole round of Natural History," is still in
its youth, but has of late years thriven so mightily
that fear has been expressed lest it should absorb
unduly tfie attention of zoologists, or even check
the progress of science by diverting interest from
otheFjmjDsqually important branches. Nor is the
reason of this phenomenal success hard to find.
The actual study of the processes of development ;
the gradual building up of the embryo, and then of
the young animal, within the egg ; the fashioning
of its various parts and organs ; the devices for
supplying it with food, and for ensuring that the
respiratory and other interchanges are duly per-
formed at all stages : all these are matters of
absorbing interest. Add to these the extraordinary
changes which may take place after leaving the
egg, the conversion, for instance, of the aquatic
gill-breathing tadpole — a true fish as regards all
essential points of its anatomy — into a four-legged
frog, devoid of tail, and breathing by lungs ; or
the history of the metamorphosis by which the
sea-urchin is gradually built up within the body of
its pelagic larva, or the butterfly derived from its
grub. Add to these again the far wider interest
aroused by comparing the life-histories of allied
animals, or by tracing the mode of development of
a complicated organ — e.g., the eye or the brain — in
the various animal groups, from its simplest com-
mencement, through gradually increasing grades of
efficiency, up to its most perfect form as seen in

THE RECAPITULATION THEORY 291

the highest animals. Consider this, and it becomes
easy to understand the fascination which embry-
ology exercises over those who study it.

But all this is of trifling moment compared with
the great generalisation which tells us that the
development of animals has a far higher meaning ;
that the several embryological stages and the order
of their occurrence are no mere accidents, but are
forced on an animal in accordance with a law, the
determination of which ranks as one of the greatest
achievements of biological science. The doctrine
of descent, or of Evolution, teaches us that as
individual animals arise, aig_l spontaneously, but by
direct descent from pre-existing' animals, so also is
it with species, with families, and with larger
groups of animals, and sjajalso has it been for all
time ; that as the animals of succeeding generations
are related together, so also are those of successive
geologic periods ; that, all animals, living or that
have lived, are united together by blood relation-
ship of varying nearness or remoteness ; and that
every animal now in existence has a pedigree
stretching back, not merely for ten or a hundred
generations, but through all geologic time since the
dawn of life on this globe.

[_JThe study of Development, in its turn, has
revealed to us that each animal bears the mark of
its ancestry, and is compelled to discover its
parentage in its own development ; that the phases
through which an animal passes in its progress
from the egg to the adult are no accidental freaks,
no mere matters of developmental convenience, but

292 THE RECAPITULATION THEORY

} represent more or less closely, in more or less
^modified manner, the successive ancestral stages
through which the present condition has been
acquired^ Evolution tells us that each animal has
had_a__peidigree in the past. Embryology reveals
to us this ancestry, because every animal in its
own_ji£yelopment repeats this history, climbs up
itsown_genealogical tree. Such is the Recapitula-
3on_TheoryJ hinted at by Agassiz, and suggested
more directly in the writings of von Baer, but first
clearly enunciated by Fritz Muller, and since elabo-
rated by many, notably by Balfour, and by Ernst
HaeckeJ^. It is concerning this theory, which
forms the basis of the science of Embryology, and
which alone justifies the extraordinary attention
this science has received, that I venture to address
you this morning. A few illustrations from differ-
ent groups of animals will best explain the practical
bearings of the theory, and the aid which it affords
to the zoologist of to-day, while these will also
serve to illustrate certain of the difficulties which
have arisen in the attempt to interpret individual
development by the light of past history — diffi-
culties which I propose to consider at greater
length.

A very simple example of recapitulation is
afforded by the eyes of the sole, plaice, turbot, and
their allies. These " flat fish " have their bodies
greatly compressed laterally ; and the two surfaces,
really the right and left sides of the animal are,
unlike, one being white, or nearly so, and the other
coloured. The flat fish has two eyes, but these

THE RECAPITULATION THEORY 293

in place of being situated; as in other fish> one on
each side of the head, are both on the coloured
side. The advantage to the fish is clear, for the
natural position of rest of a flat fish is lying on the
sea bottom; with the white surface downwards and
the coloured one upwards. In such a position an
eye situated on the white surface could be of no
use to the fish, and might even become a source of
danger, owing to its liability to injury from stones
or other hard bodies on the sea bottom. No one
would maintain that flat fish were specially created
as such. The totality of their organisation shows
clearly enough that they are true fish, akin to
others in which the eyes are symmetrically placed
one on each side of the head, in the position they
normally hold among vertebrates. We must there-
fore suppose that flat fish are descended from other
fish in which the eyes are normally situated.

The Recapitulation Theory supplies a ready test.
On employing it — i.e., on studying the development
of the flat fish — we obtain a conclusive answer.
The young sole on leaving the egg is shaped just
as any ordinary fish, and has the two eyes placed
symmetrically on the two sides of the head. It is
only after the young fish has reached some size,
and has begun to approach the adult in shape, and
to adopt its habit of resting on one side on the sea
bottom, that the eye of the side on which it rests
becomes shifted forwards, then rotated on to the
top of the head, and finally twisted completely over
to the opposite side.

The brain of a bird differs from that of other

294 THE RECAPITULATION THEORY

vertebrates in the position of the optic lobes, these
being situated at the sides instead of on the dorsal
surface. Development shows that this lateral
position is a secondarily acquired one, for through-
out all the earlier stages the optic lobes are, as in
other vertebrates, on the dorsal surface, and only
shift down to the sides shortly before the time of
hatching.

Crabs differ markedly from their allies, the
lobsters, in the small size and rudimentary condi-
tion of their abdomen or " tail." Development
however affords abundant evidence of the descent
of crabs from macrurous ancestors, for a young crab
at what is termed the Megalopa stage has the
abdomen as large as a lobster or prawn at the same
stage.

Molluscs afford excellent illustrations of recapitu-
lation. The typical gasteropod has a large spirally
coiled shell ; the limpet however has a large
conical shell, which in the adult gives no sign of
spiral twisting, although the structure of the animal
shows clearly its affinity to forms with spiral shells.
Development solves the riddle at once, telling us
that in its early stages the limpet embryo has
a spiral shell, which is lost on the formation
subsequently of the conical shell of the adult.

Recapitulation is not confined to the higher
groups of animals, and the Protozoa themselves
yield most instructive examples. A very striking
case is that of Orbitolites, one of the most complex
of the procellanous Foraminifera, in which each
individual during its own growth and development

THE RECAPITULATION THEORY 295

passes through the series of stages by which the
cyclical or discoidal type of shell was derived from
the simpler spiral form.

In Orbitolites tenmsstma, as Dr. Carpenter has
shown,* "the whole transition is actually presented
during the successive stages of its growth. For it
begins life as a Cornuspira, .... its shell forming
a continuous spiral tube, with slight interruptions
at the points at which its successive extensions
commence ; while its sarcodic body consists of a
continuous coil with slight constrictions at intervals.
The second stage consists in the opening out of its
spire, and the division of its cavity at regular
intervals by transverse septa, traversed by separate
pores, exactly as in Peneroplis. The third stage is
marked by the subdivision of the ' peneropline '
chambers into chamberlets, as in the early forms of
Orbiculina. And the fourth consists in the exchange
of the spiral for the cyclical plan of growth, which
is characteristic of Orbitolites ; a circular disc of
progressively increasing diameter being formed by
the addition of successive annular zones around the
entire periphery."

The shells both of Foraminifera and of Mollusca
afford peculiarly instructive examples for the study
of recapitulation. As growth of the shell is effected
by the addition of new shelly matter to the part
already existing, the older parts of the shell are
retained, often unaltered, in the adult ; and in
favourable cases, as in Orbitolites tenuissima, all the

* W. B. Carpenter, "On an Abyssal Type of the Genus Or-
bitolites," Phil. Trans. 1883, part ii. p. 553.

296 THE RECAPITULATION THEORY

stages of development can be determined by simple
inspection of the adult shell.

Jt is important to remember that the Recapitula-
tion Theory, if valid, must apply not merely in a
general way to the development of the animal body,
b<ut~nmst hold good with regard to the formation of
each organ or system, and with regard to the later
e_qually with the earlier phases of development. Of
individual organs, the brain of birds has been
already cited. The formation of the vertebrate liver
as a diverticulum from the alimentary canal, which
is at first simple, but by the folding of its walls
becomes greatly complicated, is another good
example ; as is also the development of the vomer
in Amphibians as a series of toothed plates, equiva-
lent morphologically to the placoid scales of fishes,
which are at first separate, but later on fuse
together and lose the greater number of their
teeth.

Concerning recapitulation in the later phases
of development and in the adult animal, the mode
of renewal of the nails or of the epidermis generally
is a good example, each cell commencing its exist-
ence in an indifferent form in the deeper layers
of the epidermis, and gradually acquiring the adult
peculiarities as it approaches the surface, through
removal of the cells lying above it.

.The above examples, selected almost haphazard,
will suffice to illustrate the Theory of Recapitulation,
of the theory depends chiefly on its
unversaap^licability to all animals, whether high
or low in the zoological scale, and to all their parts

THE RECAPITULATION THEORY 297

and organs. It derives also strong support from
the ready explanation which it gives of many
otherwise unintelligible points.

Of these latter a familiar and most instructive
instance is afforded by rudimentary organs — i.e.,
structures which, like the outer digits of the horse's
leg, or the intrinsic muscles of the ear of a man,
are present in the adult in an incompletely devel-
oped form, and in a condition in which they can be
of no use to their possessors — or else structures
which are present in the embryo, but disappear
completely before the adult condition is attained,
for example the teeth of whalebone whales, or the
branchial clefts of all higher vertebrates.

Natural Selection explains the preservation of
useful variations, but will not account for the
formation and perpetuation of useless organs ; and
rudiments such as those mentioned above would be
unintelligible but for Recapitulation, which solves
the proBIem at once, showing that these organs,
though^ now useless, must have been of functional
valire^Tolhe ancestors of their present possessors,
and that their appearance in the ontogeny of existing
forms is due to repetition of ancestral characters.
Such rudimentary organs are, as Darwin pointed
out, of larger relative or even absolute size in the
embryo than in the adult, Because the embryo
represents the stage in the pedigree in whiclTthey
were functionally activer^

Rudimentary^ergans are extremely common,
especially among the higher groups of animals, and
their presence and significance are now well under-

298 THE RECAPITULATION THEORY

stood^ Man himself affords numerous and excellent
examples. _not merely in his bodily structure, but
byjiis speech, dress, and customs. For the silent
letter b in the word doubt, or the w of answer, or

.the buttons on his elastic side boots are as true
examples of rudiments, unintelligible but for their
past history, as are the ear muscles he possesses

JjJULt cannot use, or the gill-clefts, which are func-
tional in fishes and tadpoles, and are present
though useless in the embryos of all higher verte-

~brates, which in their early stages the hare and
the tortoise alike possess, and which are shared

"with them by cats and by kings.

Another consideration of the greatest importance
arises from the study of the fossil remains of the
animals that formerly inhabited the earth. . It was
the elder Agassiz who first directed attention to
the remarkable agreement between THiTembrvomc
growth of animals and their palaeontological history.
He pointed out the resemblance between certain
stages in the growth of young fish and their fossil
representatives, and attempted to establish, with
regard to fish, a correspondence between their
palaeontological sequence and the successive stages
of embryonic development. He then extended his
observations to other groups, and stated his conclu-
sions in these words : *?_" It may therefore be con-
sidered as a general fact, very likely to be more
fully illustrated as_investigations cover a wider
ground, that the phases (^development of all living
animals correspond to the order of succession
* L. Agassiz, " Essay on Classification," 1859, p. 115.

THE RECAPITULATION THEORY 299

of their extinct representatives in past geological
times."

This point of view is of the utmost importance.
If the development of an animal is really a repeti-
tion of its ancestral history, then it is clear that the
agreement or parallelism which Agassiz insists on
between the embryological and palaeontological
records must hold good. Owing to the attitude
which Agassiz subsequently adopted with regard
to the theory of Natural Selection, there is some
fear of his services in this respect failing to receive
full recognition, and it must not be forgotten that
the sentence I have quoted was written prior to
the clear enunciation of the Recapitulation Theory
by Fritz Miiller.

The imperfection of the geological record has
been often referred to and lamented. It is very
true that our museums afford us but fragmentary
pictures of life in past ages ; that the earliest
volumes of the history are lost, and that of others
but a few torn pages remain to us ; but the later
records are in far more satisfactory condition. The
actual number of specimens accumulated from the
more recent formations is prodigious ; facilities for
consulting them are far greater than they were ;
the international brotherhood of science is now
fully established, and the fault will be ours if the
material and opportunities now forthcoming are not
rightly and fully utilised.

By judicious selection of groups in which long
series of specimens can be obtained, and in which
the hard skeletal parts, which alone can be suitably

300 THE RECAPITULATION THEORY

preserved as fossils, afford reliable indications of
zoological affinity, it is possible to test directly
this correspondence between palaeontological and
embryological histories, while in some instances a
single lucky specimen will afford us, on a particular
point, all the evidence we require.

Great progress has already been made in this
direction, and the results obtained are of the most
encouraging description. By Alexander Agassiz
a detailed comparison was made between the fossil
series and the developmental stages of recent forms
in the case of the Echinoids, a group peculiarly
well adapted for such an investigation. The two
records agree remarkably in many respects, more
especially in the independent evidence they give as
to the origin of the asymmetrical forms from more
regular ancestors. The gradually increasing com-
pilation in some of the historic series is found to
be repeated very closely in the development of
their existing representatives; and with regard to
the whole group, Agassiz concludes that,* " com-
paring the embryonic development with the
palaeontological one, we find a remarkable similarity
in both, and in a .general way there seems to be a
parallelism in the_a£pearance of the fossil genera
and the successive stages of the development of
the Echini." Neumayr Tias followed similar lines,
and by him, as by other authorities on the group,
there seems to be general agreement as to the

* A. Agassiz, " Palaeontological and Embryological Develop-
ment," an Address before the American Association for the
Advancement of Science, 1880.

THE RECAPITULATION THEORY 301

parallelism between the embryological and palaeon-
tological records, not merely for Echini, but for
other groups of Echinodermata as well.

The Tetrabranchiate Cephalopoda are an excel-
lent group in which to study the problem, for
though no opportunity has yet occurred for study-
ing the embryology of the only surviving member
of the group, the pearly nautilus, yet owing to the
fact that growth of the shell is effected by addition
of shelly matter to the part already present, and to
the additions being made in such manner that the
older part of the shell persists unaltered, it is
possible, from examination of a single shell — and
in the case of fossils the shells are the only part
of which we have exact knowledge — to determine
all the phases of its growth ; just as in the shell
of Orbitolites all the stages of development are
manifest on inspection of an adult specimen. In
such a shell as Nautilus or Ammonites the central
chamber is the oldest or first formed one, to which
the remaining chambers are added in succession.
If therefore the development of the shell is a
repetition of ancestral history, the central chamber
should represent the palasontologically oldest form,
and the remaining chambers in succession, forms of
more and more recent origin. Ammonite shells
present, more especially in their sutures, and in
the markings and sculpturing of their surface,
characters that are easily recognised, and readily
preserved in fossils ; and the group consequently
is a very suitable one for investigation from this
standpoint. Wtirtenberger's admirable and well-

302 THE RECAPITULATION THEORY

known researches* have shown that in the Ammon-
ites such a correspondence between historic and
embryonic development does really exist ; that,
for example, in Aspidoceras the shape and mark-
ings of the shells in young specimens differ greatly
from those of adults, and that the characters of the
young shells are those of palaeontologically older
forms. Another striking illustration of the cor-
respondence between palaeontological and develop-
mental records is afforded by the antlers of deer,
in which the gradually increasing complication of
the antler in successive years agrees singularly
closely with the progressive increase in size and
complexity shown by the fossil series from the
Miocene age to recent times.

Of cases where a single specimen has sufficed to
prove the palaeontological significance of a develop-
mental character, Archaeopteryx affords a typical
example. In recent birds the metacarpals are
firmly fused with one another, and with the distal
series of carpals ; but in development the meta-
carpals are at first, and for some time, distinct.
In Archaeopteryx this distinctness is retained in
the adult, showing that what is now an embryonic
character in recent birds, was formerly an adult
one.f Other examples might easily be quoted, but
these will suffice to show that the relation between

.Palaeontology and Embryology, first enunciated by
Agassiz, and required by the Recapitulation

* L. WUrtenberger, " Studien iiber die Stammesgeschichte der
Ammoniten. Ein geologischer Beweis fur die Darwin 'sche
Theorie." Leipzig, 1880.

THE RECAPITULATION THEORY 303

Theory, does in reality exist. There is much yet
to be done in this direction. A commencement, a
most promising commencement, has been made,
but as yet only a few groups have been seriously
studied from this standpoint.

It is a great misfortune that palaeontology is not
more generally and more seriously studied by men
versed in embryology, and that those who have so
greatly advanced our knowledge of the early
development of animals should so seldom have
tested their conclusions as to the affinities of the
groups they are concerned with by direct reference
to the ancestors themselves, as known to us
through their fossil remains. I cannot but feel
that for instance the determination of the affinities
of fossil Mammalia, of which such an extraordinary
number and variety of forms are now known to us,
would be greatly facilitated by a thorough and
exact knowledge of the development, and especially
the later development, of the skeleton in their
existing descendants, and I regard it as a reproach
that such exact descriptions of the later stages of
development should not exist even in the case of
our commonest domestic animals.

The pedigree of the horse has attracted great
attention and has been worked at most assiduously,
and we are now, largely owing to the labours of
American palaeontologists, able to refer to a series
of fossil forms commencing in the lowest Eocene
beds, and extending upwards to the most recent
deposits, which show a complete gradation from a
more generalised mammalian type to the highly

304 THE RECAPITULATION THEORY

specialised condition characteristic of the horse and
its allies, and which may reasonably be regarded
as indicating the actual line of descent of the horse.
In this particular case, more frequently cited than
any other, the evidence is entirely palaeontological.
The actual development of the horse has yet to be
studied, and it is greatly to be desired that it
should be undertaken speedily. Klever's* recent
work on the development of the teeth in the horse
may be referred to as showing that important and
unexpected evidence is to be obtained in this way.

A brilliant exception to the statement just made
as to the want of exact knowledge of the later
development of the more highly organised animals
is afforded by the splendid labours of Professor
W. K. Parker, whose recent death has deprived
zoology of one of her most earnest and single-minded
students, and zoologists, young and old alike, of a
true and sincere friend. Professor Parker's extra-
ordinarily minute and painstaking investigations into
the development of the vertebrate skull rank among
the most remarkable of zootomical achievements
and afford a rich mine of carefully recorded facts, the
full value and bearing of which we are hardly yet
able to appreciate.

If further evidence as to the value and import-
ance of the Recapitulation Theory were needed, it
would suffice to refer to the influence which it
has had on the classification of the animal king-
dom. Ascidians and Cirripedes may be quoted as

* Klever, " Zur Kenntniss der Morphogenese des Equidenge-
bisses," Morphologisches Jahrbuch, xv. 1889, p. 308.

THE RECAPITULATION THEORY 305

important groups, the true affinities of which were
first revealed by embryology ; and in the case of
parasitic animals the structural modifications of the
adult are often so great that, but for the evidence
yielded by development, their zoological position
could not be determined. It is now indeed gene-
rally recognised that in doubtful cases embryology
affords the safest of all clues, and that the zoo-
logical position of such forms can hardly be
regarded as definitely established unless their
development, as well as their adult anatomy, is
ascertained.

It is owing to this Recapitulation Theory that
Embryology has exercised so marked an influence
on zoological speculation. Thus the formation in
most, if not in all, animals of the nervous system
and of the sense organs from the epidermal layer
of the skin, acquired a new significance when it
was recognised that this mode of development was
to be regarded as a repetition of the primitive
mode of formation of such organs ; while the
vertebral theory of the skull affords a good ex-
ample of a view, once stoutly maintained, which
received its death-blow through the failure of
embryology to supply the evidence requisite in
its behalf. The necessary limits of time and space
forbid that I should attempt to refer to even the
more important of the numerous recent discoveries
in embryology, but mention may be very pro-
perly made here of Sedgwick's determination of
the mode of development of the body cavity in
Peripatus, a discovery which has thrown most

u

306 THE RECAPITULATION THEORY

welcome light on what was previously a great mor-
phological puzzle.

We must now turn to another side of the ques-
tion. Although it is undoubtedly true that develop-
TneTTr~!s to be regarded as a recapitulation of
ancestral phases, and that the embryonic history
of an animal presents to us a record of the race-
history, yet it is also an undoubted fact, recognised
by all writers on embryology, that the record so
obtained is neither a complete nor a straightforward
one. It is indeed a history, but a history of which
entire chapters are lost, while in those that remain
many pages are misplaced and others are so blurred
as to be illegible ; words, sentences, or entire para-
graphs are omitted, and, worse still, alterations or
spurious additions have been freely introduced by
later hands, and at times so cunningly as to defy
detection.

^ Very slight consideration will show that develop-
ment cannot in all cases be strictly a recapitulation
oT ancestral stages. It is well known that closely
allied animals may differ markedly in their mode
of development. The common frog is at first a
tadpole, breathing by gills, a stage which is entirely
omitted by the West Indian Hylodes. A crayfish,
a lobster, and a prawn are allied animals, yet they
leave the egg in totally different forms. Some
developmental stages, as the pupa condition of
insects, or the stage in the development of a dog-
fish in which the oesophagus is imperforate, cannot
possibly be ancestral stages. Or again, a chick
embryo of say the fourth day is clearly not an

THE RECAPITULATION THEORY 307

animal capable of independent existence, and
therefore cannot correctly represent any ancestral
condition, an objection which applies to the
developmental history of many, perhaps of most
animals.

Haeckel long ago urged the necessity of distin-
guishing in actual development between those
characters which are really historical and inherited,
and those which are acquired or spurious additions
to theTecord. The former he termed palingenetic
or ancestral characters, the latter cenogenetic or
acquired. The distinction is undoubtedly a true
one, but an exceedingly difficult one to draw in
practice. The causes which prevent development
from being a strict recapitulation of ancestral
characters, the mode in which these came about,
and the influence which they respectively exert,
are matters which are greatly exercising embry-
ologists, and the attempt to determine which has as
yet met with only partial success.

The most potent and the most widely spread
of these disturbing causes arise from the necessity
of supplying the embryo with nutriment. This
acts in two ways. If the amount of nutritive
matter within the egg is small, then the young
animal must hatch early, and in a condition in
which it is able to obtain food for itself. In such
cases there is of necessity a long period of larval
life, during which natural selection may act so as
to introduce modifications of the ancestral history,
spurious additions to the text.

If on the other hand, the egg contain within

308 THE RECAPITULATION THEORY

itself a considerable quantity of nutrient matter,
then the period of hatching can be postponed until
this nutrient matter has been used up. The conse-
quence is that the embryo hatches at a much later
stage of its development, and if the amount of food
material is sufficient may even leave the egg in the
form of the parent. In such cases the earlier
developmental phases are often greatly condensed
and abbreviated ; and as the embryo does not lead
a free existence, and has no need to exert itself to
obtain food, it commonly happens that these stages
are passed through in a very modified form, the
embryo being, as in a four-day chick, in a condition
in which it is clearly incapable of independent
existence.

The nutrition of the embryo prior to hatching is
most usually effected by granules of nutrient matter,
known as food yolk, and embedded in the proto-
plasm of the egg itself ; and it is on the relative
abundance of these granules that the size of the
egg chiefly depends. Large size of eggs implies
diminution of number of the eggs, and hence of the
offspring ; and it can be well understood that while
some species derive advantage in the struggle for
existence by producing the maximum number of
young, to others it is of greater importance that
the young on hatching should be of considerable
size and strength, and able to begin the world on
their own account. In other words, some animals
may gain by producing a large number of small
eggs, others by producing a smaller number of eggs
of larger size — i.e., provided with more food yolk.

THE RECAPITULATION THEORY 309

The immediate effect of a large amount of food
yolk is to mechanically retard the processes of
development; the ultimate result is to greatly
shorten the time occupied by development. This
apparent paradox is readily explained. A small
egg, such as that of Amphioxus, starts its develop-
ment rapidly, and in about eighteen hours gives
rise to a free swimming larva, capable of indepen-
dent existence, with a digestive cavity and nervous
system already formed ; while a large egg like that
of the hen, hampered by the great mass of food
yolk by which it is distended, has in the same
time made but very slight progress. From this
time however other considerations begin to tell.
Amphioxus has been able to make this rapid start
owing to its relative freedom from food yolk. This
freedom now becomes a retarding influence, for the
larva, containing within itself but a very scanty
supply of nutriment, must devote much of its
energies to hunting for and to digesting its food,
and hence its further development will proceed more
slowly.

The chick embryo, on the other hand, has an
abundant supply of food in the egg itself; it has
no occasion to spend time searching for food, but
can devote its whole energies to the further stages
of its development. Hence, except in the earliest
stages, the chick develops more rapidly than
Amphioxus, and attains its adult form in a much
shorter time.

The tendency of abundant food yolk to lead to
shortening or abbreviation of the ancestral history,

310 THE RECAPITULATION THEORY

and even to the entire omission of important stages,
is well known. The embryo of forms well provided
with yolk takes short cuts in its development, jumps
irom ~branch to branch of its genealogical tree,
instead of climbing steadily upwards. Thus the
little West Indian frog, Hylodes, produces eggs
which contain a larger amount of food yolk than
those of the common English frog. The young
Hylodes is consequently enabled to pass through
the tadpole stage before hatching, to attain the
form of a frog before leaving the egg ; and the
tadpole stage is only imperfectly recapitulated, the
formation of gills for instance being entirely
omitted.

The influence of food yolk on the development
of animals is closely analogous to that of capital in

human undertakings. A new industry, for example
"jhat of Jen-making, has often been started by a
man working by hand and alone, making and selling
his own wares ; if he succeed in the struggle for
existence, "it soon becomes necessary for him to
"£all in others to assist him, and to subdivide the
work ; hand labour is soon superseded by machines,
Involving further differentiation of labour ; the
earlier machines are replaced by more perfect and
more costly ones ; factories are built, agents
engaged, and in the end a whole army of work-
people employed. In later times a man commencing
Ijvith very limited means will start at the
level as the original founder, and will have
to work his way upwards through much the same
stages—?^., will repeat the pedigree of the industry.

THE RECAPITULATION THEORY 311

Thp_papitalist on the other hand, is enabled like
f to omit these earlier stages, and after a

brief period of incubation, to start business with
large factories equipped with the most recent appli-
ances, and__wjth a complete staff of workpeople
existence fully fledged.

There is no doubt that abundance of food yolk
is a direct and very frequent cause of the omission
of ancestral stages from individual development ;
but it must not be viewed as the sole cause. It is
quite impossible that any animal, except perhaps
in the lowest "zoological groups, should repeat all
the ancestral stages in the history of the race ; the
limits of time avail a¥Ie for individual development
will not permit .Jhis^ There is a tendency in all
animals towards condensation of the ancestral his-
tory, towards striking a direct path from the egg
to the adult. This Telio!ency is best marked in the
higher, the more complicated members of a group —
i.e., in those which have a longer and more tor-
tuous pedigree — and, though greatly strengthened
by the presence of food yolk in the egg, is appa-
rently not due to this in the first instance.

Thus the simpler forms of Orbitolites, as O.
tenm'sstma, repeat in their development all the
stages leading from a spiral to a cyclical shell ; but
in the more complicated species, as Dr. Carpenter
has pointed out, there is a tendency towards pre-
cocious development of the adult characters, the
earlier stages being hurried over in a modified
form ; while in the most complex examples, as in
O. complanata, the earlier spiral stages may be

312 THE RECAPITULATION THEORY

entirely omitted, the shell acquiring almost from its
earliest commencement the cyclical mode of growth.
There is no question here of relative abundance of
food yolk, but merely of early or precocious
appearance of adult characters.

The question of the relations and influence of
food yolk, involving as it does the larger or smaller
size of the egg, is however merely a special side
of the much wider question of the nutrition of the
embryo, one of the most potent of the disturbing
elements affecting development. Speaking gene-
rally, we may say that large eggs are more often
met with in the higher than the lower groups of
animals. Birds and Reptiles are cases in point,
and if Mammals do not now produce large eggs,
it is because a more direct and more efficient mode
of nourishing the young by the placenta has been
acquired by the higher forms, and has replaced the
food yolk that was formerly present, and is now
retained in quantity by Monotremes alone. Molluscs
afford another good example, the eggs of Cephalo-
poda being of larger size than those of the less
highly organised groups. The large size of the
eggs of Elasmobranchs, and perhaps that of Cepha-
lopods also, may possibly be associated with the
carnivorous habits of the animals ; for it is of im-
portance that forms which prey on other animals
should hatch of considerable size and strength.

The influence of habitat must also be considered.
It has long been noticed as a general rule that
marine animals lay small eggs, while their fresh-
water allies have eggs of much larger size. The

THE RECAPITULATION THEORY 313

eggs of the salmon or trout are much larger than
those of the cod or herring ; and the crayfish,
though only a quarter the length of a lobster, lays
eggs of actually larger size. This larger size of the
eggs of fresh-water forms appears to be dependent
on the nature of the environment to which they
are exposed. Considering the geological instability
of the land as compared with the ocean, there can
be no doubt that fresh-water fauna are, speaking
generally, derived from the marine fauna ; and the
great problem with regard to fresh-water life is to
explain why it is that so many groups of animals
which flourish abundantly in the sea should have
failed to establish themselves in fresh water.
Sponges and Coelenterates abound in the sea, but
their fresh-water representatives are extremely few
in number ; Echinoderms are exclusively marine ;
there are no fresh-water Cephalopods, and no
Ascidians ; and of the smaller groups of Worms,
Molluscs, and Crustaceans, there are many that do
not occur in fresh water.

Direct experiment has shown that in many cases
this distribution is not due to inability of the adult
animals to live in fresh water ; and the real ex-
planation appears to be that the early larval stages
are unable to establish themselves under such
conditions. This interesting suggestion, which has
been worked out in detail by Professor Sollas,*
undoubtedly affords an important clue. To

* W. J. Sollas, "On the Origin of Freshwater Fauna,"
Scientific Transactions of the Royal Dublin Society, vol. iii. Ser. n.,
1886.

314 THE RECAPITULATION THEORY

establish itself permanently in fresh water an animal
must either be fixed, or else be strong enough to
withstand and make headway against the currents
of the streams or rivers it inhabits, for otherwise
it will in the long run be swept out to sea, and
this consideration applies to larval forms equally
with adults.

The majority of marine Invertebrates leave the
egg as minute ciliated larvae, and such larvae are
quite incapable of holding their own in currents of
any strength. Hence it is only forms which have
got rid of the free swimming ciliated larval stage,
and which leave the egg of considerable size and
strength, that can establish themselves as fresh-
water animals. This is effected most readily by
the acquisition of food yolk — hence the large size
of the eggs of fresh-water animals — and is often
supplemented, as Sollas has shown, by special
protective devices of a most interesting nature. For
this reason fresh-water forms are not so well
adapted as their marine allies for the study of
ancestral history as revealed in larval or embryonic
development.

Before leaving the question of food yolk, refer-
ence must be made to the proposal of the brothers
Sarasin, to regard the yolk cells as forming a dis-
tinct embryonic layer, the lecithoblast,* distinct
from the blastoderm. I do not desire to speak
dogmatically on a point the full bearings of which
are not yet apparent, but I venture to think that

* P. and F. Sarasin, " Ergebnisse naturwissenschaftlicher
Forschungen auf Ceylon," Bd. ii. Heft iii. 1889.

THE RECAPITULATION THEORY 315

this suggestion will not commend itself to embry-
ologists. The distinction between the yolk granules
and the cells in which they are embedded is a real
and fundamental one ; but I see no reason for
regarding the yolk cells as other than originally
functional endoderm cells in which yolk granules
have accumulated to such an extent that they have
in extreme cases become devoted solely to the
storing of food for the embryo.*

Of all the causes tending to modify development,
tending to obscure or falsify the ancestral record,
food yolk is ther most frequent and the most im-
portant ; its position in the egg determines the
mode of segmentation, and its relative abundance
affects profoundly Jhe entire embryonic history, and
decides at what particular stage, and of what size
and form, the embryo, shall hatch.

The loss of food_yolk is another disturbing
element, the full influence of which is as yet im-
perfectly understopd,J)ut the possibility of which
must be always kept in mind. It is best known
in the case of mammals, where it has led to appa-
rent, though very, ^deceptive, simplification of
development ; jmd it will probably not be until the
embryology of the large-yolked monotremes is at
length described, that we shall fully understand the
formation of the germinal layers in the higher
placental mammals. Amongst invertebrates we
know but little as yet concerning the effects of loss
of food yolk. It has been suggested that the

* Cf. E. B. Wilson, " The Development of Renilla," Phil
Trans. 1883, p. 755.

3i6 THE RECAPITULATION THEORY

extraordinary nature of the segmentation of the
egg of Peripatus capensis, made known to us
through Mr. Sedgwick's admirable researches, may
be due to loss of food yolk : a suggestion which
receives support from the long duration of uterine
development in this case. Our knowledge is very
imperfect as to the ease with which food yolk may
be acquired or lost ; but until our information is
more precise on this point it seems unwise to lay
much stress on suggested pedigrees which involve
great and frequent alternations in the amount of
food yolk present.

f_ Ofjrauses other than food yolk, or only in-
directly connected with it, which tend to falsify the
ancestral history, many are now known, but time
will only permit me to notice the more important.
These are distortion, whether in time or space ;
sudden or violent metamorphosis ; a series of
"modifications, due chiefly to mechanical causes, and
which may be spoken of as developmental con-
Veniences ; the important question of variability
in development ; and finally the great problem of
degeneration.

Concerning distortions in time, all embryologists
have noticed the tendency to anticipation or pre-
cocious development of characters which really
belong~to a later stage in the pedigree. The
early attainment of the cyclical form in the shell
of Orbitolites complanata is a case in point ; and
Wiirtenberger has specially noticed this tendency
in Ammonites. Many early larvae show it mark-
edly, the explanation in this case being that it is

THE RECAPITULATION THEORY 317

essential for them to hatch in a condition capable
of independent existence — i.e., capable at any rate
of obtaining and digesting their own food.

Anachronisms, or actual reversal of the historical
order of development^ of organs, or parts, occur
frequently.^ Thus the joint surfaces of bones
acquire their characteristic curvatures before move-
ment of one part or another is effected, and before
even the joint cavities are formed. Another good
example is afforded by the development of the
mesenterial filaments in Alcyonarians. Wilson has
shown in the case of Renilla that in the develop-
ment of an embryo from the egg the six endodermal
filaments appear first, and the two long ectodermal
filaments at a later period ; but that in the forma-
tion of a bud this order of development is reversed,
the ectodermal filaments being the first formed.
He suggests in explanation, that as the endodermal
filaments are the digestive organs, it is of primary
importance to the free embryo that they should be
formed quickly. The long ectodermal filaments
are chiefly concerned with maintaining currents of
water through the colony ; in bud-development
they appear before the endodermal filaments,
because they enable the bud during its early stages
to draw nutrient matter from the body fluid of the
parent ; while the endodermal filaments cannot
come into use until the bud has acquired both
mouth and tentacles. The completion of the ven-
tricular septum in the heart of higher vertebrates
before the auricular septum is a well-known
anachronism, and every embryologist could readily

318 THE RECAPITULATION THEORY

furnish many other cases. A curious instance is
afforded by the development of the teeth in mam-
mals, if recent suggestions as to the origin of the
milk dentition are confirmed, and the milk dentition
prove to be a more recent acquisition than the
permanent one.*

s But the most important case in reference to dis-
j:ortion in time concerns the reproductive organs.
If development were a strict and correct recapitula-
tion of ancestral history, then each stage would
possess reproductive organs in a mature condition.
This is not the case, and it is clearly of the greatest
importance that it should not be. It is true that
the first commencement of the reproductive organs
may_occur_ at a very early stage, or even that the
very first step in development may be a division
of the egg into somatic and reproductive cells ; and
it is possible that, as maintained by Weismann,
JJiis latter-condition is a primitive one. Still even
in. these jgses the reproductive organs merely com-
mence their development at these early stages, and
^la-aot-become functional until the animal is adult.

<L_Exceptionally in certain animals, and as a
normal occurrence in others, precocious maturation
of the~1Feproductive organs takes place, and a larval
form becomes capable of sexual reproduction. This
may lead to arrest of development, either at a late
larval period as in the Axolotl, or at successively

* C/. Thomas Oldfield, " On the Homologies and Succession
of the Teeth in the Dasyuridae, with an attempt to trace the
history of the evolution of the Mammal and Teeth in general,"
Phil. Trans. 1887.

THE RECAPITULATION THEORY 319

earlier ajjdearlier stages, as in the gonophores of
the Hydromedusae, until finally the extreme condi-
tion seen in HycTra is produced. We do not know
the causes that determine the period, whether late
or early, at whicji the reproductive organs ripen,
but the question is one of great interest and
importance,__and, deserves careful attention. The
suggestion "has been made that entire groups of
animals, such as the Mesozoa, are merely larvae,
arrested through such precocious acquiring of re-
productive power, and it is conceivable that this
may be the case. Mesozoa are a puzzling group
in which the life history, though known with toler-
able completeness, has as yet given us no reliable
clue concerning their affinities to other animals,
a tantalising distinction that is shared with them by
Rotifers and Polyzoa.

Distortion of a curious kind is seen in cases of
abrupt metamorphosis, where, as in the case of
many Echinoderms, of Phoronis, and of the meta-
bolic insects, the larva and the adult differ greatly
in form, habits, mode of life, and very usually in
the nature of their food and the mode of obtaining
it ; and the transition from one stage to the other
is not a gradual but an abrupt one, at any rate so
far as external characters are concerned. Sudden
changes of this kind, as from the free swimming
Pluteus to the creeping Echinus, or from the slug-
gish leaf-eating caterpillar to the dainty butterfly,
cannot possibly be recapitulatory, for even if small
jumps are permissible in nature, there is no room
for bounds forward of this magnitude. Cases of

320 THE RECAPITULATION THEORY

abrupt metamorphosis may always be viewed as due
to secondary modifications, and rarely, if ever, have
any significance beyond the particular group of
animals concerned. For example, a Pluteus larva
may be recognised as belonging to the group of
Echinoidea before the adult urchin has commenced
to be formed within it, and the Lepidopteran cater-
pillar is already an unmistakable insect. Hence,
for the explanation of the metamorphoses in these
cases it is useless to look outside the groups of
Echinoidea and Insecta respectively.

Abrupt metamorphosis is always associated with
great change in external form and appearance, and
in mode of life, and very usually in mode of nutri-
tion. A gradual transition in such cases is
inadmissible, because in the intermediate stages
the animal would be adapted to neither the larval
nor the adult condition ; a gradual conversion of
the biting mouth parts of the caterpillar to the
sucking proboscis of a moth would inevitably lead
to starvation. The difficulty is evaded by retaining
the external form and habits of one particular stage
for an unduly long period, so that the relations
of the animal to the surrounding environment
remain unchanged, while internally preparations
for the later stages are in progress. Cinderella
and the princess are equally possible entities, each
being well adapted to her environment. The exi-
gences of the situation do not permit however of
a gradual change from one to the other : the trans-
formation, at least as regards external appearance,
must be abrupt.

THE RECAPITULATION THEORY 321

Kleinenberg has recently directed attention to
cases in which the larval and adult organs develop
independently — the larval nervous system for
instance, aborting completely and forming no part
of that of the adult. I am not sure that I fully
understand Kleinenberg's argument, but it seems
very possible that such cases, which are probably
far more numerous than is yet admitted, may be
due to what may be termed the telescoping of
ancestral stages oiie within another, which takes
place in actual development, and may accord-
ingly be grouped under the head of develop-
mental convenience. Undue prolongation of an
early ancestral stage, as in cases of abrupt meta-
morphosis, must involve modification, especially
in the muscular and nervous systems; in such
cases a telescoping of ancestral stages takes place
as we have seen, the adult being developed within
the larva. , Such telescoping must distort the re-
capitulatofVjjiistory, and as the shape of the larva
and adult may differ widely, an independent origin
of organs, especially the muscular and nervous
systems, may be_acquired secondarily.

The stage in the development of Squilla, in which
the three posterior maxillipedes disappear com-
pletely, to reappear at a later stage in a totally
different form, is not to be interpreted as meaning
that the adult maxillipedes are entirely new struc-
tures unconnected historically with those of the larva.
Neither is the annual shedding of the antlers
of deer to be regarded as the repetition of an an-
cestral hornless condition intercalated historically

X

322 THE RECAPITULATION THEORY

between successive stages provided with antlers.
In both cases the explanation is afforded by con-
venience, whether of the embryo or adult.

Many embryological modifications or distortions
may be attributed to mechanical causes, and may
fairly be considered under the head of develop-
mental conveniences. The amnion of higher
vertebrates is a case in point, and is probably
rightly explained as due in the first instance to
sinking or depression of the embryo into the yolk,
in order to avoid distortion through pressure
against a hard unyielding eggshell. A similar
device is employed, presumably for the same reason,
in the early development of many insect embryos ;
and the depression of the Taenia head within
the cyst is a phenomenon of very similar nature.

Restriction of the space within which develop-
ment occurs often causes displacement or distortion
of organs whose growth, restricted in its normal
direction, takes place along the lines of least resist-
ance. The telescoping of the limbs and other
organs within the body of an insect larva is a simple
case of such distortion ; and a more complicated
example, closely comparable in many ways to the
invagination of the Taenia head, is afforded by the
remarkable inversion of the germinal layers in
Rodents, first described by Bischoff in the guinea
pig, and long believed to be peculiar to that animal,
but subsequently and simultaneously discovered by
three independent observers, Kupffer, Selenka, and
Eraser, to occur in varying degrees in rats, mice,
and in other rodents.

THE RECAPITULATION THEORY 323

One of the most recent attempts to explain
developmental peculiarities as due to mechanical
causes is Mr. Dendy's suggestion with regard to
the pseudogastrula stage in the development of
calcareous sponges. It is well known that while
the larva is in the amphiblastula stage, and still
imbedded in the tissues of the parent, the granular
cells become invaginated within the ciliated cells,
giving rise to the pseudogastrula stage. At a
slightly later stage, when the larva becomes free,
the invaginated granular cells become again everted,
and the larva spherical in shape ; while still later
invagination occurs once more, the ciliated cells
being this time invaginated within the granular
cells. The significance of the pseudogastrula stage
has hitherto been undetermined, but Mr. Dendy
points out that the larva always occupies a definite
position with reference to the parental tissues ; that
the ciliated half of the larva is covered by a soft
and yielding wall, while the opposite half, composed
of the granular cells, is covered by a layer stiffened
with rigid spicules ; and his observations on the
growth of the larva lead him to think that the
pseudogastrula stage is brought about mechanically
by flattening of the granular cells through pressure
against this rigid wall of spicules.

Embryology supplies us with many unsolved
problemgjmcHt is not to be wondered at that this
should be the case. Some of these may fairly be
spoken of as mere curiosities of development, while
others are clearly of greater moment. I do not
propose to catalogue these, but will merely mention

324 THE RECAPITULATION THEORY

two or three which I happen to have recently run
my head against and remember vividly. The solid
condition of the oesophagus, in Elasmobranch
embryos, first noticed by Balfour, is a very curious
point. The oesophagus has at first a well-developed
lumen, like the rest of the alimentary canal ; but at
an early period, stage K of Balfour's nomenclature,
the part of the oesophagus overlying the heart, and
immediately behind the branchial region, becomes
solid, and remains solid for a long time, the exact
date of reappearance of the lumen not being yet
ascertained. Mr. Bles and myself have recently
noticed that a similar solidification of the oesophagus
occurs in tadpoles of the common frog. In young
free swimming tadpoles the oesophagus is perforate,
but in tadpoles of about 7j mm. length it becomes
solid and remains so until a length of about loj
mm. has been attained. The solidification occurs
at a stage closely corresponding with that in which
it first appears in the dogfish, and a curious point
about it is that in the frog the oesophagus becomes
solid just before the mouth-opening is formed, and
remains solid for some little time after this important
event. This closing of the oesophagus clearly
cannot be recapitulation, but the fact that it occurs
at corresponding periods in the frog and dogfish
suggests that it may possibly, as Balfour hinted,
" turn out to have some unsuspected morphological
bearing."

Another developmental curiosity is the duplication
of the gill slits by growth downwards of tongues
from their dorsal margins ; a duplication which is

THE RECAPITULATION THEORY 325

described as occurring in Amphioxus and in Balano-
glossus, but in no other animal ; and the occurrence
of which, in apparently closely similar fashion, is
one of the strongest arguments in favour of a real
affinity between these two forms. It is hardly
possible that such a modification should have been
acquired independently twice over.

A much more litigious question is the significance
of the neurenteric canal of vertebrates, that curious
tubular communication between the central canal of
the nervous system and the hinder end of the ali-
mentary canal that is conspicuously present in the
embryos of lower vertebrates, and retained in a
more or less disguised condition in the higher groups
as well. The neurenteric canal was discovered by
that famous embryologist Kowalevsky in Ascidians
and in Amphioxus. He drew special attention to
the occurrence of a stage in both Ascidians and in
Amphioxus in which the larva is free swimming
and in which the sole communication between the
alimentary cavity and tire exterior is through the
neurenteric canal and the central canal of the nervous
system ; and suggested * that animals may have
existed or may still exist in which the nerve tube
fulfilled a non-nervous function, and possibly acted
as part of the alimentary canal ; a suggestion that
has recently been revived in a somewhat extra-
vagant form. A passage of food particles into the
alimentary cavity through the neural tube has not

* A. Kowalevsky, " Weitere Studien iiber die Entwickelungs-
geschichte des Amphioxus lanceolatus," Archiv. Jurmihroskopisch
Anatomic , Bd. xiii. 1877, p. 201.

326 THE RECAPITULATION THEORY

yet been seen, and probably does not occur, as the
larva still possesses sufficient food yolk to carry it
on in its development. It is therefore permissible
to hold that the neurenteric canal may be a mere
embryological device, and devoid of any deep mor-
phological significance.

The question of variation in development is one
of very great importance, and has perhaps not yet
received the attention it deserves. We are in some
danger of assuming tacitly that the mode of develop-
ment of allied animals will necessarily agree in all
important respects or even in details, and that if
the development of one member of a group be known,
that of the others may be assumed to be similar.
The more recent progress of embryology is showing
us that such inferences are not safe, and that in
allied genera or species, or even in different indi-
viduals of the same species, variations of develop-
ment may occur affecting important organs and at
almost any stage in their formation.

Great individual variations in the earliest pro-
cesses of development — i.e.t the segmentation of the
egg — have been described by different writers. In
Renilla, Wilson found an extraordinary range of
variation in the segmentation of eggs from which
apparently identical embryos were produced. In
some cases the egg divided into two in the normal
manner ; in other cases it divided at once into eight,
sixteen, or thirty-two segments, which in different
specimens were approximately equal or markedly
unequal in size. Sometimes a preliminary change
of form occurred without any further result, the egg

THE RECAPITULATION THEORY 327

returning to its spherical shape, and pausing for a
time before recommencing the attempt to segment.
Segmentation sometimes commenced at one pole, as
in telolecithal eggs, with the formation of four or
five small segments, the rest of the egg breaking
up later, either simultaneously or progressively,
into segments about equal in size to those first
formed ; while lastly, in some instances segmenta-
tion was very irregular, following no apparent law.
It is noteworthy that the variability in the case of
Renilla is apparently confined to the earliest stages,
for whatever the mode of segmentation, the embryos
in their later stages were indistinguishable from one
another. Similar modifications in the segmentation
of the egg have been described in the oyster by
Brooks, in Anodon and other Mollusca, in Hydra,
and in Lumbricus, in which last Wilson has recently
shown that marked differences occur in the eggs
even of the same individual animal. In the different
species of Peripatus there appear also to be con-
siderable variations in the details of segmentation.

In the early embryonic stages after the comple-
tion of segmentation very considerable variation
may occur in allied species or genera. Among
Coelenterates for instance the mode of formation of
the hypoblast presents most perplexing modifica-
tions : it may arise as a true gastrula invagination ;
as cells budded off from one pole of the blastula
into its cavity ; as cells budded off from various
parts of the wall of the blastula ; by delamination
or actual division of each cell of the blastula wall ;
or it may be present from the first as a solid mass

328 THE RECAPITULATION THEORY

of cells enclosed by the epiblast cells. It is in con-
nection with these variations that controversy has
arisen as to the primitive mode of development of
the gastrula, a point to which I shall return later
on.

Among the higher Metazoa or Ccelomata the
extraordinary modifications in the position and in
every conceivable detail of formation of the meso-
blast in different and often in closely allied forms
have given rise to ardent discussion, and have led
to the proposal of theory after theory, each rejected
in turn as only affording a partial explanation, and
now culminating in Kleinenberg's protest against
the use of the term mesoblast at all, at any rate in
a sense implying any possibility of comparison with
the primary layers, epiblast and hypoblast, of
Coelenterata.

This is not the place to attempt to decide so
difficult and technical a point, even were 1 capable
of so doing f but we may well take warning from
this extraordinary diversity of development, the
full extent ~oT~Wh~ich I believe we as yet realise
most imperfectly, that in our attempts to recon-
struct ancestral history from ontogenetic develop-
ment we have taken ^n hand no light task. To
reconstruct Latin from modern European languages
would in comparison be but child's play.

Of the readiness with which special develop-
mental characters are acquired by allied animals
the brothers Sarasin* have given us evidence in

* P. and F. Sarasin, "Ergebnisse naturwissenschaftlicher
Forschungen auf Ceylon," vol. ii. chap. i. pp. 24-38.

THE RECAPITULATION THEORY 329

the extraordinary modifications presented by the
embryonic and larval respiratory organs of Amphi-
bians. Confining ourselves to those forms which
do not lay their eggs in water, and in which con-
sequently development takes place within the egg,
we find that Ichthyophis and Salamandra have
three pairs of specially modified external gills.
Nototrema has two pairs ; Alytes and Typhlonectes
have only a single pair, which in the latter genus
take the form of enormous leaf-like outgrowths
from the sides of the neck. In Hylodes and Pipa
there are no gills, the tail acting as the larval
respiratory organ ; and in Rana opisthodon, accord-
ing to Boulenger, larval respiration is effected by
nine pairs of folds of the skin of the ventral surface
of the body.

Most of these extraordinarily diversified organs
are clearly secondarily acquired structures ; it is
possible that they all are, and that external gills,
as was suggested by Balfour for Elasmobranchs,
are to be regarded as embryonal respiratory organs
acquired by the larvae and of no ancestral value. The
point however cannot be considered settled, for on
this view the external gills of Elasmobranchs and
Amphibians would be independently acquired and
not homologous structures, a view contradicted by
the close agreement in their relations in the two
groups, as well as by the absence of any real
break between external and internal gills in Am-
phibians.

It is well known that the frog and the newt
differ greatly in important points of their develop-

330 THE RECAPITULATION THEORY

ment. The two-layered condition of the epiblast
in the frog is a marked point of difference, which
involves further changes in the mode of formation
of the nervous system and sense organs. The
kidneys and their ducts differ considerably in their
development in the two forms, as do also the
blood-vessels. Concerning the early development
of the blood-vessels, there are considerable differ-
ences even between allied species of frogs. In
Rana esculenta Maurer finds that there is at first in
each branchial arch a single vessel or aortic arch,
running directly from the heart to the aorta : from
the cardiac end of this aortic arch a vessel grows
out into the gill as the afferent branchial vessel,
the original aortic arch losing its connection with
the heart, and becoming the efferent branchial vessel.
Afferent and efferent branchial vessels become
connected by capillaries in the gill, and the course
of the circulation, so long as gill-breathing is main-
tained, is from the heart through the truncus arte-
riosus to the afferent branchial vessel, then through
the gill capillaries to the efferent branchial vessel,
and then on to the aorta. When the pulmonary
circulation is thoroughly established the branchial
circulation is cut off by the efferent vessel reacquir-
ing its connection with the heart, when the blood
naturally takes the direct passage along it to the
aorta, and so escapes the gill capillaries.

In Rana temporaria the mode of development is
very different : the afferent and efferent vessels
arise in each arch independently and almost simul-
taneously ; the afferent vessel soon acquires

THE RECAPITULATION THEORY 331

connection with the heart, but unlike R. esculenta,
the efferent vessel has no connection with the heart
until the gills are about to atrophy. In other
words the continuous aortic arch, from heart to
aorta, is present in R. esculenta prior to the de-
velopment of the gills ; it becomes interrupted
while the gills are in functional use, but is re-
established when these begin to atrophy. In R.
temporaria, on the other hand, there is no con-
tinuous aortic arch until the gills begin to atrophy.
The difference is an important one, for it is a
matter of considerable morphological interest to
determine whether the continuous aortic arch is
primitive for vertebrates — i.e., whether it existed
prior to the development of gills. This point could
be practically settled if we could decide which of
the two frogs, R. esculenta and R. temporaria, has
most correctly preserved its ancestral history in
this respect. About this there can be little doubt.
The development of the vessels in the newts, a less
modified group than the frogs, agrees with that of
R. esculenta, and interesting confirmation is afforded
by a single aberrant specimen of R. temporaria, in
which Mr. Bles and myself found the vessels de-
veloping after the type of R. esculenta — i.e., in which
a complete aortic arch was present before the gills
were formed. We are therefore justified in con-
cluding that as regards the development of the
branchial blood-vessels, R. esculenta has retained a
primitive ancestral character which is lost in R.
temporaria ; and it is interesting to note that were
our knowledge of the development of amphibians

332 THE RECAPITULATION THEORY

confined to the common frog, the most likely form
to be studied, we should in all probability have
been led to wrong conclusions concerning the
ancestral condition of the blood-vessels in a point
of considerable importance.

t A matter which at present is attracting much

attention is the question of degeneration. Natural
selection, though consistent with and capable of
leading to steady upward progress and improve-
ment, by no means involves such progress as a
necessary consequence. All it says is that those
animals will, in each generation, have the best
chance of survival which are most in harmony
with their environment, and such animals will not
'necessarily be those which are ideally the best or
most perfect.

If you go into a shop to purchase an umbrella,
the one you select is by no means necessarily that
which most nearly approaches ideal perfection, but
the one which best hits off the mean between your
idea of what an umbrella should be and the amount
of money you are prepared to give for it : the one
in fact that is on the whole best suited to the
circumstances of the case or the environment for
the time being. It might well happen that you
had a violent antipathy to a crooked handle, or else
were determined to have a catch of a particular
kind to secure the ribs, and this might lead to
the selection — i.e., the survival — of an article that
in other and even in more important respects was
manifestly inferior to the average.

So is it also with animals : the survival of a

THE RECAPITULATION THEORY 333

form that is ideally inferior is very ^possible. To
animals living in profound darkness_the__possession
of eyes is of no advantage, and forms devoid of
eyes would not merely lose nothing_^thereby, but

would actually gain, inasmuch as they would

escape the dangers that might arise from injury
to a delicate and complicated organ. In extreme
cases, as in animals leading a parasitic existence,
the conditions of life may be juch as to render
locomotor, digestive, sensory, and other organs
entirely useless ; and in such cases those forms
will be best in harmony with their surroundings
which avoid the waste of_energy resulting from the
formation and maintenance of theae organs.

Animals which have in this wav^fallen from the
high estate of their forefathers, which have lost
organs or systems which their progenitors possessed,
are commonly called degenerate. The"principlej3f
degeneration, recognised by Darwin as a possible,
and under certain conditions a necessary conse-
quence of his theory of natural selectioji^has been
since advocated strongly by Dohrn, and later by
Lankf£ler~ia-ari EveningJDiscourse delivered before
the Association at the Sheffield Meeting in 1879.
Both Dohrn and Lankester suggested that degenera-
tion occurred mucITntore "WldelyjKan was generally
recognised f

In animals which are parasitic when adult, but
free swimming in their early stages, as in the case
of the Rhizocephala whose life-history was so
admirably worked out by Fritz Mtiller, degenera-
tion is clear enough. So also is it in the case of

334 THE RECAPITULATION THEORY

the solitary Ascidians, in which the larva is a free
swimming animal with a notochord, an elongated
tubular nervous system, and sense organs, while
the adult is fixed, devoid of the swimming tail,
with no notochord, and with a greatly reduced
nervous system and aborted sense organs. In
such cases the animal when adult is, as regards
the totality of its organisation, at a distinctly
lower morphological level, is less highly differen-
tiated than it is when young, and during individual
development there is actual retrograde development
of important systems and organs.

About such cases there is no doubt ; but we
are asked to extend the idea of degeneration much
more widely. It is urged that we ought not to
demand direct embryological evidence before accept-
ing a group as degenerate. We are reminded of
to abbreviation or to complete omis-

sion of ancestral stages of which we have quoted
examples, above ; and it is suggested that if such
larval stages were omitted in all the members of a
group we should have no direct evidence of de-
generation in a group that might really be in an
extremely degenerate condition. Supposing for
instance the free larval stages of the solitary
Ascidians were suppressed, say through the acqui-
sition of food yolk, then it is urged that the
degenerate condition of the group might easily
escape detection. The supposition is by no means
extravagant ; food yolk varies greatly in amount
in allied animals, and cases like Hylodes, or
amongst Ascidians Pyrosoma, show how readily a

THE RECAPITULATION THEORY 335

mere increase in the amount of food yolk in the
egg may lead to the omission of important ancestral
stages.

The question then arises whether it is not
possible, or even probable, that animals which now
show no indication of degeneration in their develop-
ment are in reality highly degenerate, and whether
it is not legitimate to suppose such degeneration to
have occurred in the case of animals whose affinities
are obscure or difficult to determine. It is more
especially with regard to the lower vertebrates that
this argument has been employed ; and at the pre-
sent day zoologists of authority, relying on it, do
not hesitate to speak of such forms as Amphioxus
and the Cyclostomes as degenerate animals, as
wolves in sheep's clothing, animals whose simpli-
city is acquired and deceptive rather than real and
ancestral. I cannot but think that cases such as
these should be regarded with some jealousy.
There is at present a tendency to invoke degenera-
tion rather freely as a talisman to extricate us from
morphological difficulties ; and an inclination to
accept such suggestions, at any rate provisionally,
without requiring satisfactory evidence in their
support.

Degeneration of which there is direct embryo-
logical evidence stands on a very different footing
from suspected degeneration, for which no direct
evidence is forthcoming ; and in the latter case the
burden of proof undoubtedly rests with those who
assume its existence. The alleged instances among
the lower vertebrates must be regarded particularly

336 THE RECAPITULATION THEORY

closely, because in their case the suggestion of
degeneration is admittedly put forward as a means
of escape from difficulties arising through theoretical
views concerning the relation between vertebrates
and invertebrates. Amphioxus itself, so far as I
can see, shows in its development no sign of
degeneration, except possibly with regard to the
anterior gut diverticula, whose ultimate fate is not
altogether clear. With regard to the earlier stages
of development, concerning which, thanks to the
patient investigations of Kowalevsky and Hatschek,
our knowledge is precise, there is no animal known
to us in which the sequence of events is simpler or
more straightforward. Its various organs and
systems are formed in what is recognised as a
primitive manner ; and the development of each is
a steady upward progress towards the adult con-
dition. Food yolk, the great cause of distortion in
development, is almost absent, and there is not the
slightest indication of the former possession of a
larger quantity. Concerning the later stages our
knowledge is incomplete, but so much as has been
ascertained gives no support to the suggestion of
general degeneration.

t Our knowledge of the conditions leading to
degeneration is undoubtedly incomplete, but it must
J>e noticed that the conditions usually associated
with degeneration do not occur. Amphioxus is not
parasitic, is not attached when adult, and shows no
evidence of having formerly possessed food yolk
in quantity sufficient to have led to the omission
of important ancestral stages. Its small size, as

THE RECAPITULATION THEORY 337

compared with other vertebrates, is one of the very
few points that can be referred to as possibly indi-
cating degeneration, and will be considered more
fully at a later point in my address.

A consideration of much less importance, but
deserving of mention, is that in its mode of life
Amphioxus not merely differs as already noticed
from those groups of animals which we know to
be degenerate, but agrees with some at any rate
of those which there is reason to regard as primi-
tive or persistent types. Amphioxus, like Balano-
glossus, Lingula, Dentalium, and Limulus, is marine,
and occurs in shallow water, usually with a sandy
bottom, and, like the three smaller of these genera,
it lives habitually buried almost completely in the
sand, into which it burrows with great rapidity.

I do not wish to speak dogmatically. I merely
wish to protest against a too ready assumption of
degeneration ; and to repeat that so far as I can
see, Amphioxus has not yet, either in its develop-
ment, in its structure, or in its habits, been shown
to present characters that suggest, still less that
prove, the occurrence in it of general or extensive
degeneration. In a sense, all the higher animals
are degenerate ; that is they can be shown to
possess certain organs in a less highly developed
condition than their ancestors, or even in a rudi-
mentary state. Thus a crab as compared with
a lobster is degenerate in the matter of its tail,
a horse as compared with Hipparion in regard to
its outer toes ; but it is neither customary nor
advisable to speak of a crab as a degenerate animal

Y

338 THE RECAPITULATION THEORY

compared to a lobster ; to do so would be mis-
leading. An animal should only be spoken of as
degenerate when the retrograde development is
well marked, and has affected not one or two
organs only, but the totality of its organisation.
c. It is impossible to draw a sharp line in such
cases, and to limit precisely the use of the term
degeneration. It must be borne in mind that no
animal is at the top of the tree in all respects,
^an Himself is primitive as regards the number
of his^ toes, and degenerate in respect to his ear
muscles ; and between two animals even of the
same group it may be impossible to decide which
of the two is to be called the higher and which the
lower form. Thus to compare an oyster with a
mussel : the oyster is more primitive than the
mussel as regards the position of the ventricle of
the heart and its relations to the alimentary canal ;
but is more modified in having but a single adductor
muscle, and almost certainly degenerate in being
devoid of a foot.

Care must also be taken to avoid speaking of an
animal as degenerate in regard to a particular organ
merely because that organ is less fully developed
than in allied animals. An organ is not degenerate
unless its present possessor has it in a less perfect
condition than its__ancestors had.c — A~man is not
degenerate in the matter of the length of his neck
as compared jvith a giraffe, nor as compared with
an elephant irTrelpect of the size of his front teeth,
for neither elephant nor giraffe enters into the
pedigree of man! STman is however degenerate,

THE RECAPITULATION THEORY 339

whoever his ancestors may have been, in regard to
his ear muscles ; for he possesses these in a rudi-
menTarjTand functionless condition, which can only
be e'xpfeified by descent from some better equipped
progenitor.

Closely connected with the question of degenera-
tion is that of the size of animals, and its bearing
on their structure and development — a problem
noticed by many writers, but which has perhaps
not yet received the attention it merits.

If we are right in interpreting the eggs of
Metazoa as representing the unicellular or protozoon
stage in their ancestry, then the small size of the
egg may be viewed as recapitulatory. But the
gradual increase in size of the embryo, and its
growth up to the adult condition, can only be
regarded as representing in a most general way, if
at all, the actual or even the relative sizes of the
intermediate ancestral stages of the pedigree.

It is quite true that animals belonging to the
lower groups are, as a general rule, of smaller size
than those of higher grade ; and also that the giants
are met with among the highest members of each
division. Cephalopoda are the highest molluscs,
and the largest cephalopods greatly exceed in size
any other members of the group ; decapods are at
once the highest and the largest crustaceans ; and
whales, the hugest animals that exist, or so far
as we know, that ever have existed, belong to the
highest group of all, the mammalia. It would be
easy to quote exceptions, but the general rule
obtains admittedly. However, although there may

340 THE RECAPITULATION THEORY

be, and probably is, a general parallelism between
the increase in size from the egg to the adult, and
the historical increase in size during the passage
from lower to higher forms ; yet no one could
maintain that the sizes of embryos represent at all
correctly those of the ancestors ; that for instance
the earliest birds were animals the size of a chick
embryo at a time when avian characters first declare
themselves, or that the ancestral series in all cases
presented a steady progression in respect of actual
magnitude.

In the lower animals — e.g., in Orbitolites — the
actual size of the several ancestral stages is probably
correctly recapitulated during the growth of the
adult ; and it is very possible that it is so also in
such forms as the solitary sponges. In higher
animals, except in the early stages of those forms
which are practically devoid of food yolk, and which
hatch as pelagic larvae, this certainly does not
obtain. This is clear enough, but is worth point-
ing out, for if as most certainly is the case the
embryos of animals are actually smaller than the
ancestral forms they represent, it is possible that
the smallness of the embryo may have had some
influence on its organisation, and be responsible
for some of the modifications in the ances-
tral history ; and more especially for the disap-
pearance of ancestral organs in free swimming
larvae.

In adult animals the relation between size and
structure has been very clearly pointed out by
Herbert Spencer. Increased size involves by itself

THE RECAPITULATION THEORY 341

increased complexity of structure ; the determining
consideration being that while the surface area of
the body increases as the squares of the linear
dimensions, the mass of the body increases as their
cubes. If for example we imagine two animals of
similar shape and proportions, but of different size ;
for the sake of simplicity, we may suppose them to
be spherical, and that the diameter of one is twice
that of the other ; then the larger one will have
four times the extent of surface of the smaller, but
eight times its mass or bulk : and it is quite possible
that while the extent of surface, or skin, in the
smaller animal might suffice for the necessary
respiratory and excretory interchanges, it would be
altogether insufficient in the larger animal, in which
increased extent of surface must be provided by
foldings of the skin, as in the form of gills. To
take an actual instance ; Limapontia is a minute
nudibranchiate, or sea-slug, about the sixth of an
inch in length ; it has a smooth body, totally devoid
of respiratory processes, while forms allied to it,
but of larger size, have their extent of surface
increased by branching processes, which often take
the form of specialised gills. This is a peculiarly
instructive case, because Limapontia in its early
developmental stages possesses a large spirally
coiled shell, and shows other evidence of descent
from forms with specialised breathing organs. We
are certainly right in associating the absence of
respiratory organs in the adult with the small size
of the animal ; and comparison with allied forms
suggests very strongly that there has been in its

342 THE RECAPITULATION THEORY

pedigree an actual reduction of size, which has led
to the degeneration of the respiratory organs.

This is an important conclusion : it is a well-
known fact that the smaller members of a group
are as a rule more simply organised than the
larger members, especially with regard to their
respiratory and circulatory systems ; but if we are
right in concluding that reduction in size may be
an actual cause of simplification or degeneration in
structure, then we must be on our guard against
assuming hastily that these smaller and simpler
animals are necessarily primitive in regard to the
groups to which they belong. It is possible for
instance that the simplification or even absence of
respiratory organs seen in Pauropus, in the
Thysanura, and in other small Tracheata, may be
a secondary character, acquired through reduction
of size.

An interesting illustration of the law discussed
above is afforded by the brains of mammals ; it has
been noticed by many anatomists that the extent of
convolution, or folding of the surface of the cerebral
hemispheres in mammals, is related not to the degree
of intelligence of the animal, but to its actual size,
a beaver having an almost smooth brain and a cow
a highly complicated one. Jelgersma, and inde-
pendently of him Professor Fitzgerald,* have
explained this as due to the necessity of preserving
the due proportion between the outer layer of grey
matter or cortex, which is approximately uniform in
thickness, and the central mass of white matter.
* Cf. Nature. June 5, 1890, p. 125.

THE RECAPITULATION THEORY 343

But for the foldings of the surface the proportion of
white matter to grey matter would be far higher in
a large than in a small brain.

It must not be forgotten on the other hand, that
many zoologists hold the view, in favour of which
the evidence is steadily increasing, that the primitive
or ancestral members of each group were of small
size. Thus Furbringer remarks with regard to
birds that on the whole small birds show more
primitive and simpler conditions of structure than
the larger members of the same group. He
expresses the opinion that the first birds were
probably smaller than Archaeapteryx, and notes that
reptiles and mammals also show in their earlier and
smaller types more primitive features than do their
larger descendants. Finally, Furbringer concludes
that "it is therefore the study of the smaller
members within given groups of animals which
promises the best results as to their phylogeny."

Again, one of the most striking points with
regard to the pedigree of the horse, as agreed on
by palaeontologists, is the progressive reduction in
size which we meet with as we pass backwards in
time from stage to stage. The Pliocene Hipparion
was smaller than the existing horse, in fact about
the size of a donkey ; the Miocene Mesohippus
about equalled a sheep ; while Eohippus, from the
Lower Eocene deposits, was no larger than a fox.
Not only is there good reason for holding that as
a rule larger animals are descended from ancestors
of smaller size, but there is also much evidence to
show that increase in size beyond certain limits is

344 THE RECAPITULATION THEORY

disadvantageous, and may lead to destruction rather
than to survival. It has happened more than once
in the history of the world, and in more than one
group of animals, that gigantic stature has been
attained immediately before extinction of the group,
a final and tremendous effort to secure survival, but
a despairing and unsuccessful one. The Ichthyo-
sauri, Plesiosauri, and other extinct reptilian groups,
the Moas, and the huge extinct Edentates, are well-
known examples, to which before long will be added
the elephants and the whales, and it may be iron-
clads as well. The whole question of the influence
of size is of the greatest possible interest and
importance, and it is greatly to be hoped that it
will not be permitted to remain in its present un-
certain and unsatisfactory condition.

It may be suggested that Amphioxus is an
animal which has undergone reduction in size, and
that its structural simplicity may, like that of
Limapontia, be due in part at least to this reduc-
tion. Such evidence as we have tells against this
suggestion ; the first system to undergo degenera-
tion in consequence of a reduction in size is the
respiratory, and the respiratory organs of Amphioxus,
though very simple, are also for a vertebrate
unusually extensive.

/\Ve have now considered the more important of
jhe Influences which are recognised as affecting
developmental history in such a way as to render
the recapitulation of ancestral stages less complete
than it might otherwise be, which tend to prevent
ontogeny from correctly repeating the phylogenetic

THE RECAPITULATION THEORY 345

history. It may at this point reasonably be asked
whether there is ~any way of distinguishing the
palingenetic history from the later cenogenetic
modifications grafted on to it ; any test by which
we can determine whether a given larval character
is or is not ancestral. Most assuredly there is no
one rule, no single test that will apply in all cases ;
but there are certain considerations which will help
us, and which should be kept in view.

A character that is of general occurrence among
the members of a group, both high and low, may
reasonably be regarded as having strong claims to
ancestral rank ; claims that are greatly strengthened
if it occurs at corresponding developmental periods
in all cases ; and still more if it occurs equally in
forms that hatch early as free larvae, and in forms
with large eggs, which develop directly into the
adult. As examples of such characters may be cited
the mode of formation and relations of the notochord,
and of the gill clefts of vertebrates, which satisfy
all the conditions mentioned. Characters that are
transitory in certain groups, but retained throughout
life in allied groups, may with tolerable certainty
be regarded as ancestral for the former ; for instance
the symmetrical position of the eyes in young flat
fish, the spiral shell of the young limpet, the
superficial positions of the madreporite in Elasi-
podous Holothurians, or the suckerless condition of
the ambulacral feet in many Echinoderms.

i_&_jmore important consideration is that if the
developmental changes are to be interpreted as a
correct record of ancestral history, then the several

346 THE RECAPITULATION THEORY

stages must be possible ones, the history must be
one thatTOuld actually have occurred — i.e., the
several steps of the history as reconstructed must
form zTseries, all the stages of which are practicable
ones.^_Natural selection explains the actual structure
of a ^complex organ as having been acquired by the
preservation of a series of stages, each a distinct
if ^slight advance on the stage immediately pre-
ceding it, an advance so distinct as to confer on its
possessor an appreciable advantage in the struggle
forjExistence. It is not enough that the ultimate
stage should be more advantageous than the initial
oF earlier condition, but each intermediate stage
must also be a distinct advance. If then the
development of an organ is strictly recapitulatory,
ft shouldjresent to us a series of stages, each of
which is not merely functional, but a distinct
advance on the stage immediately preceding it.
Intermediate stages — e.g., the solid oesophagus of
the tadpole — which are not and could not be
functional, can form no part of an ancestral series ;
a consideration well expressed by Sedgwick* thus :
" Any phylogenetic hypothesis which presents diffi-
culties from a physiological standpoint must be
regarded as very provisional indeed."

A good example of an embryological series
fulfilling these conditions is afforded by the de-
velopment of the eye in the higher Cephalopoda.
The earliest stage consists in the depression of a

* Sedgwick, " On the Early Development of the Anterior Part
of the Wolffian Duct and Body in the Chick," Quarterly Journal
of Microscopical Science, vol. xxi. 1881, p. 456.

THE RECAPITULATION THEORY 347

slightly modified patch of skin ; round the edge of
the patch the epidermis becomes raised up as a rim ;
this gradually grows inwards from all sides, so that
the depressed patch now forms a pit, communi-
cating with the exterior through a small hole or
mouth. By further growth the mouth of the pit
becomes still more narrowed, and ultimately com-
pletely closed, so that the pit becomes converted
into a closed sac or vesicle ; at the point at which
final closure occurs formation of cuticle takes place,
which projects as a small transparent drop into the
cavity of the sac ; by formation of concentric layers
of cuticle this drop becomes enlarged into the
spherical transparent lens of the eye, and the
development is completed by histological changes
in the inner wall of the vesicle, which convert it
into the retina, and by the formation of folds of
skin around the eye, which become the iris and the
eyelids respectively.

Each stage in this developmental history is a
distinct advance, physiologically, on the preceding
stage, and furthermore each stage is retained
at the present day as the permanent condi-
tion of the eye in some member of the group
Mollusca. The earliest stage, in which the eye is
merely a slightly depressed and slightly modified
patch of skin, represents the simplest condition of
the Molluscan eye, and is retained throughout life in
Solen. The stage in which the eye is a pit, with
widely open mouth, is retained in the limpet ; it
is a distinct advance on the former, as through
the greater depression the sensory cells are less

348 THE RECAPITULATION THEORY

exposed to accidental injury. The narrowing of
the mouth of the pit in the next stage is a simple
change, but a very important step forwards. Up
to this point the eye has served to distinguish
light from darkness, but the formation of an image
has been impossible. Now, owing to the smallness
of the aperture, and the pigmentation of the walls
of the pit, which accompanies the change, light
from any one part of an object can only fall on one
particular part of the inner wall of the pit or retina,
and so an image, though a dim one, is formed.
This type of eye is permanently retained in the
Nautilus. The closing of the mouth of the pit by
a transparent membrane will not affect the optical
properties of the eye, and will be a gain, as it will
prevent the entrance of foreign bodies into the
cavity of the eye. The formation of the lens by
deposit of cuticle is the next step. The gain here
is increased distinctness and increased brightness
of the image, for the lens will focus the rays of
light more sharply on the retina, and will allow a
greater quantity of light, a larger pencil of rays
from each part of the object, to reach the corre-
sponding part of the retina. The eye is now in
the condition in which it remains throughout life
in the snail and other gasteropods. Finally the
formation of the folds of skin known as iris and
eyelids provide for the better protection of the eye,
and is a clear advance on the somewhat clumsy
method of withdrawal seen in the snail.

The development of the vertebrate liver is
another good but simpler example. The most

THE RECAPITULATION THEORY 349

primitive form of the liver is that of Amphioxus, in
which it is present as a simple saccular diverti-
culum of the intestinal canal, with its wall consisting
of a single layer of cells, and with bloodvessels on
its outer surface. The earliest stage in the forma-
tion of the liver in higher vertebrates — the frog,
for instance — is practically identical with this. In
the frog the next stage consists in folding of the
wall of the sac, which increases the efficiency of
the organ by increasing the extent of surface in
contact with bloodvessels. The adult condition is
attained simply by a continuance of this process ;
the foldings of the wall becoming more and more
complicated, but the essential structure remaining
the same — a single layer of epithelial cells in con-
tact on one side with bloodvessels, and bounding
on the other directly or indirectly the cavity of the
alimentary canal.

It is not always possible to point out the
particular advantage gained at each step even
when a complete developmental series is known
to us, but in such cases, as for instance in
Orbitolites, our difficulties arise chiefly from ignor-
ance of the particular conditions that confer
advantage in the struggle for existence in the case
of the forms we are dealing with.

The early larval stages in the development of
animals, and more especially those that are marine
and pelagic in habit, have naturally attracted much
attention, since in the absence, probably inevitable,
of satisfactory palaeontological evidence, they afford
us the sole available clue to the determination of the

350 THE RECAPITULATION THEORY

mutual relations of the large groups of animals, or
of the points at which these diverged from one
another.

In attempting to interpret these early ontoge-
n&icjstages as actual ancestral forms, beyond which
development at one time did not proceed, we must
keejTcIearly in view the various disturbing causes
whiclTtend to falsify the ancestral record ; such as
the influence of food yolk, or of habitat, and the
tendency of diminution in size to give rise to
simplification of structure, a point of importance if
it be granted that these free larvae are of smaller
size than the ancestral forms to which they corre-
spond; If on the other hand, in spite of these
powerful modifying causes, we do find a particular
larval form occurring widely and in groups not
very closely akin, then we certainly are justified in
attaching great importance to it, and in regarding it
as^ having strong claims to be accepted as ancestral
for these groups.

Concerning these larval forms, and their possible
ancestral significance, our knowledge has made no
great advance since the publication of Balfour's
memorable chapter on this subject ; and I propose
merely to allude briefly to a few of the more strik-
ing instances. The earliest, the most widely
spread, and the most famous of larval forms is the
gastrula, which occurs in a simple or in a modified
form in some members of each of the large animal
groups. It is generally admitted that its signifi-
cance is the same in all cases, and the evidence
is very strong in favour of regarding it as a stage

THE RECAPITULATION THEORY 351

ancestral for all Metazoa. The difficulty arising
from its varying mode of development in different
forms is however still unsolved, and embryologists
are not yet agreed whether the invaginate or delami-
nate form is the more primitive. In favour of the
former is its much wider occurrence ; in favour of
the latter the fact that it is easy to picture a series
of stages leading gradually from a unicellular proto-
zoon to a blastula, a diblastula, and ultimately a
gastrula, each stage being a distinct advance, both
morphological and physiological, on the preceding
stage ; while in the case of the invaginate gastrula
it is not easy to imagine any advantage resulting
from a flattening or slight pitting in of one part
of the surface, sufficient to lead to its preservation
and further development.

Of larval forms later than the gastrula, the most
important by far is the Pilidium larva, from which
it is possible, as Balfour has shown, that the
slightly later Echinoderm larva, as well as the
widely spread Trochosphere larva, may both be
derived. Balfour concludes that the larval forms
of all Ccelomata, excluding the Crustacea and verte-
brates, may be derived from one common type,
which is most nearly represented now by the Pili-
dium larva and which " was an organism something
like a Medusa, with a radial symmetry." The
tendency of recent phylogenetic speculations is to
accept this in full, and to regard as the ancestor
of Turbellarians and of all higher forms, a jelly-fish
or Ctenophoran, which in place of swimming freely
has taken to crawling on the sea-bottom.

352 THE RECAPITULATION THEORY

Of the two groups excluded above, the Crustacea
and the Vertebrata, the interest of the former
centres in the much discussed problem of the
significance of the Nauplius larva. There is now
a fairly general agreement that the primitive crus-
tacea were types akin to the phyllopods — i.e., forms
with elongated and many-segmented bodies, and
a large number of pairs of similar appendages. If
this is correct, then the explanation of the Nauplius
stage must be afforded by the phyllopods them-
selves, and it is no use looking beyond this group
for it. A Nauplius larva occurs in other Crustacea
merely because they have inherited from their
phyllopod ancestors the tendency to develop such
a stage, and it is quite legitimate to hold that
higher crustaceans are descended from phyllopods,
and that the Nauplius represents in more or less
modified form an earlier ancestor of the phyllopods
themselves.

As to the Nauplius itself the first thing to note
is that though an early larval form, it cannot be a
very primitive form, for it is already an unmis-
takable crustacean ; the absence of cilia, the forma -
tion of a cuticular investment, the presence of
jointed schizopodous limbs, together with other
anatomical characters, proving this point conclu-
sively. It follows therefore either that the earlier
and more primitive stages are entirely omitted in
the development of Crustacea, or else that the
Nauplius represents such an early ancestral stage
with crustacean characters which properly be-
long to a later stage, thrown back upon it and

THE RECAPITULATION THEORY 353

precociously developed. The latter explanation is
the one usually adopted ; but before the question
can be finally decided more accurate observations
than we at present possess are needed concerning
the stages intermediate between the egg and the
Nauplius. The absence of a heart in the Nauplius
may reasonably be associated with the small size
of the larva.

Concerning the larval forms of vertebrates, it is
only in Amphioxus and the Ascidians that the
earliest larval stages are free-living, independent
animals. In both groups the most characteristic
larval stage is that in which a notochord is present,
and a neural tube, open in front, and communicating
behind through a neurenteric canal with the digestive
cavity, which has no other opening to the exterior.
This is a very early stage, both in Amphioxus and
Ascidians ; but so far as we know, it cannot be
compared with any invertebrate larva. It is
customary, in discussions on the affinities of verte-
brates, to absolutely ignore the vertebrate larval
forms, and to assume that their peculiarities are
due to precocious development of vertebrate cha-
racteristics. It may turn out that this view of the
matter is correct ; but it has certainly not yet
been proved to be so, and the development of both
Amphioxus and Ascidians is so direct and straight-
forward that evidence of some kind may reasonably
be required before accepting the doctrine that this
development is entirely deceptive with regard to
the ancestry of vertebrates.

Zoologists have not quite made up their minds

354 THE RECAPITULATION THEORY

what to do with Amphioxus : apparently the most
guileless of creatures, many view it with the utmost
suspicion, and not merely refuse to accept its mute
protestations of innocence, but regard and speak
of it as the most artful of deceivers. Few questions
at the present day are in greater need of authorita-
tive settlement.

t^That ontogeny really is a repetition of phylogeny
rnusTTThink be admitted, in spite of the numerous
and various ways in which the ancestral history
jnav be distorted during actual development.
Before leaving the subject, it is worth while inquir-
ing __wjietrier any explanation can be found of
recapitulation. A complete answer can certainly
not be___giyen at present, but a partial one may
perhaps be obtained. Darwin himself suggested
tliaT the clue might be found in the consideration
that at whatever age a variation first appears in the
parent, it tends to reappear at a corresponding age
in the offspring^_but this must be regarded rather
as a statement _of the fundamental fact of embry-
ology than as an explanation of it. It is probably
safe to assume that animals would not recapitulate
unless they were "compelled to do so : that there
must be some constraining influence at work, forcing
them to repeat more oFless closely the ancestral
stages. It is impossible, for instance, to conceive
what advantage it can T>e to a reptilian or mam-
malian embryo to develojTgill clefts which are never
used, and which disappear~at a slightly later stage ;
or how it can Eeneti t a whale, that in its embryonic
condition it should possess teeth which never cut

THE RECAPITULATION THEORY 355

the gum, and which are lost before birth. More-
over, the history of development in different animals
or groups of animals offers to us, as we have seen,
a series of ingenious, determined, varied, but more
or less unsuccessful efforts to escape from the
necessity of recapitulating, and to substitute for
the ancestrg__proc£ss a more direct method. A
further consideration of importance is that recapitu-
lation is noTseerTin all forms of development, but
only in sexual development ; or at least only in
developmentTfrom' the egg. In the several forms
of asexual development, of which budding is the
most frequent and most familiar, there is no repeti-
tion of ancestral phases ; neither is there in cases
of regeneration of lost parts, such as the tentacle of
a snail, the arm of a starfish, or the tail of a lizard ;
in such regeneration it is not a larval tentacle, or
arm, or tail, that is produced, but an adult one.

The most striking point about the development
of the higher animals is that they all alike com-
mence as eggs. Looking more closely at the egg
and the conditions of its development, two facts
impress us as of special importance : first, the egg
is a single cell, and therefore represents morpho-
logically the Protozoon, or earliest ancestral phase ;
secondly, the egg, before it can develop, must be
fertilised by a spermatozoon, just as the stimulus
of fertilisation by the pollen grain is necessary be-
fore the ovum of a plant will commence to develop
into the plant-embryo.

The advantage of cross-fertilisation in increasing
the vigour of the offspring is well known, and in

356 THE RECAPITULATION THEORY

plants devices of the most varied and even extra-
ordinary kind are adopted to ensure that such
cross-fertilisation occurs. The essence of the act
of cross-fertilisation, which is already established
among Protozoa, consists in combination of the
nuclei of two cells, male and female, derived from
different individuals. The nature of the process is
of such a kind that two individual cells are alone
concerned in it ; and it may I think be reasonably
argued that the reason why animals commence
their existence as eggs — *>., as single cells — is
because it is in this way only that the advantage
of cross-fertilisation can be secured, an advantage
admittedly of the greatest importance, and to secure
which natural selection would operate powerfully.

The occurrence of parthenogenesis, either occa-
sionally or normally, in certain groups is not I
think a serious objection to this view. There are
very strong reasons for holding that partheno-
genetic development is a modified form, derived
from the sexual method. Moreover, the view
advanced above does not require that cross-fertilisa-
tion should be essential to individual development,
but merely that it should be in the highest degree
advantageous to the species, and hence leaves room
for the occurrence, exceptionally, of parthenogenetic
development.

If it be objected that this is laying too much
stress on sexual reproduction, and on the advantage
of cross-fertilisation, then it may be pointed out in
reply that sexual reproduction is the character-
istic and essential mode of multiplication among

THE RECAPITULATION THEORY 357

Metazoa ; that it occurs in all Metazoa, and that
when asexual reproduction, as by budding, &c.,
occurs, this merely alternates with the sexual pro-
cess which sooner or later becomes essential. (!£_
the fundamental importance of sexual reproduction
to the welfare of the"speries be granted, and if it
be further admitted that Metazoa are descended
from Protozoa, then we see that there is really a
constraining force of a most powerful nature com-
pelling every animal to commence its life-history in
the unicellular condition, the only condition in
which the advantage of ^cross^fertilisation can be
obtained — *.*., constraining every animal to begin
its development at its ^earli£st_ancestral stage, at
the very bottom of its genealogical tree.

On this view the actual development of any
animal is strictly limited at both ends ; it must
commence as an egg, and it must enoTirf the like-
ness of the parent. The problem ofrecapitulation
becomes thereby greatly narrowed ; all that remains
being to explain why the intermediate stages in the
actual development should repeat ^he^intermediate
stages of the ancestral history. Although narrowed
in this way, the problem still remains__one of ex-
treme difficulty.\

It is a consequence of the Theory of Natural
Selection that identity of structure involves com-
munity of descent : a given result_can only be
arrived at through a given sequence of events :
the same morphological goal cannot be reached
by two independent paths. A negro and a white
man have had common ancestors in the past ;

358 THE RECAPITULATION THEORY

and it is through the long-continued action of
selection, and environment that the two types
have heen gradually evolved. You cannot turn
a^ white man into a negro merely by sending
him__tp___live in Africa : to create a negro the
whole ancestral history would have to be repeated ;
and ITImay be that it is for the same reason
that the embryo must repeat or recapitulate its
ancestral iiistorv inl>rder to Teach the adult goal.
I am not sure that we can at present get much
further ; but the above considerations give oppor-
tunity for brief notice of what is perhaps the
most noteworthy of recent embryological papers,
Kleinenberg's remarkable monograph on Lopado-
rhynchus.

Kleinenberg directs special attention to what is
known to evolutionists as the difficulty with regard
to the origin of new organs, which is to the effect
that although natural selection is competent to
account for any amount of modification in an
organ after it has attained a certain size, and
become of functional importance, yet that it cannot
account for the earliest stages in the formation of
an organ before it has become large enough or
sufficiently developed to be of real use. The
difficulty is a serious one ; it is carefully con-
sidered by Mr. Darwin, and met completely in
certain cases ; but as Kleinenberg correctly states,
no general explanation has been offered with regard
to such instances.

As such general explanation Kleinenberg pro-
poses his theory of the development of organs

THE RECAPITULATION THEORY 359

by substitution. He points out that any modi-
fication of an organ or tissue must involve
modification, at least in functional activity, of other
organs. He then continues by urging that one
organ may replace or be substituted for another,
the replacing organ being in no way derived
morphologically from the replaced or preceding
organ, but having a genetic relation to it of this
kind : that it can only arise in an organism so
constituted, and is dependent on the prior existence
of the replaced organ, which supplies the necessary
stimulus for its formation. As an example he
takes the axial skeleton of vertebrates. The noto-
chord, formed by change of function from the wall
of the digestive canal, is the sole skeleton of the
lowest vertebrates, and the earliest developmental
phase in all the higher forms. The notochord
gives rise directly to no other organ, but is gradu-
ally replaced by other and unlike structures by
substitution. The notochord is an intermediate
organ, and the cartilaginous skeleton which re-
places it is only intelligible through the previous
existence of the notochord ; while, in its turn, the
cartilaginous skeleton gives way, being replaced,
through substitution, by the bony skeleton.

The successive phases in the evolution of weapons
might be quoted as an illustration of Kleinenberg's
theory. The__bow and arrow are a better weapon
than a stick or stone ; they are used for the same
purpose, and the importance or need for a better
weapon led tojhe replacement of the sling by the
bow. The bow does not arise by further develop-

360 THE RECAPITULATION THEORY

mentor increasing perfection of the sling ; it is
an entirely new weapon, towards the formation
of which the older and more primitive weapons
have acted as a stimulus, and which has replaced
tftese latter by substitution, while the substitution
at a later date of firearms for the bow and arrow
is merely a further instance of the same principle.

It is too early yet to realise the full significance
of Kleinenberg's most suggestive theory ; but if it
be really true that each historic stage in the evolu-
tion of an organ is necessary as a stimulus to the
development of the next succeeding stage, then it
becomes clear why animals are constrained to re-
capitulate. Kleinenberg suggests further that the
extraordinary persistence in embryonic life of
organs which are rudimentary and functionless in
the adult may also be explained by his theory, the
presence of such organs in the embryo being indis-
pensable as a stimulus to the development of the
permanent structures of the adult. It would be
easy to point out difficulties in the way of the
theory. (_JThe omission of historic stages in the
actual ontogenetic development, of which almost
all groups of animals supply striking examples, is
one of the most serious ; for if these stages are
necessary as stimuli for the succeeding stages,
then^ their omission requires explanation ; while if
such stimuli are not necessary, the theory would
appear to need revision. Such objections may
however prove to be less serious than they appear
at first sight ; and in any case Kleinenberg's
theory may be welcomed as an important and

THE RECAPITULATION THEORY 361

original contribution, which deserves — indeed
demands — the fullest and most careful considera-
tion from all morphologists, and which acquires
special interest from the explanation it offers
of recapitulation as a mechanical process, through
which alone it is possible for an embryo to attain
the adult structure.

That recapitulation does actually occur, that the
several stages in the development of an animal are
inseparably linked with and determined by its
ancestral history, must be accepted. _J!To take
any other view is to admit thaL-the— structure of
animals and the history of their development form
a mere snare to entrap oui judgment." Em-
bryology however is not to be regarded as a
master-key that is to open the gates of knowledge
and remove all obstacles from our path without
further trouble on our part;, it is rather to be
viewed and treated as a delicate and complicated
instrument, the proper handling of which requires
the utmost nicety of balance and adjustment, and
which, unless employed with fhe greatest skill and
judgment, may yield false instead_o£_true results.

Embryology is indeed a most powerful and
efficient aid, but it will not, and cannot, provide
us with an immediate or complete answer to the
great _ridolle_- of life. Complications, distortions,
innumerable_and bewildering, confront us at every
step, and the progress of knowledge has so far
served rather to increase the number .and magnitude
of these_pit falls than to teach us how to avoid them.
Still JhereJ^ no cause for despair — far from it.

362 THE RECAPITULATION THEORY

If our difficulties are increasing, so also are our
means of grappling with them ; if the goal appears
harder to reach than we thought, on the other
hand its position is far better defined, and the
means of approach, the lines 'of attack, are more
clearly recognised. One thing above all is
apparent — that embryologists must not work
single-handed, and must not be satisfied with an
acquaintance, however exact, with animals from
the side of development only ; for embryos have
this in common with maps, that too close and too
exclusive a study of them is apt to disturb a man's
reasoning power.

I Embryology is a means, not an end. Our
ambition is to explain in what manner and by
what stages the present structure of animals has
been attained. Towards this embryology affords
most potent aid ; but the eloquent protest of the
great anatomist of Heidelberg must be laid to
heart, and it must not be forgotten that it is
through comparative anatomy that its power to
help is derived. What would it profit us, as
Gegenbaur justly asks, to know that the higher
vertebrates when embryos, have slite in their
throats, unless through comparative anatomy we
were acquainted with forms now existing in which
these slits are structures essential to existence ?
Anatomy defines the goal, tells us of the things
that have to be explained ; embryology offers a
means, otherwise denied to us, of attaining it.
Comparative anatomy and palaeontology must be
studied most earnestly by those who would turn

THE RECAPITULATION THEORY 363

the lessons of embryology to best account, and it
must never be forgotten that it is to men like
Johannes Muller, Stannius, Cuvier, and John Hunter
— the men to whom our exact knowledge of com-
parative anatomy is due — that we owe also the
possibility of a science of embryology.

Printed by BAI.LANTYNE, HANSON & Co.
London & Edinburgh

A SELECTION FROM DAVID NUTT'S LIST
OF PUBLICATIONS

POETRY

WILLIAM ERNEST HENLEY

A BOOK OF VERSES. In Hospital, Rhymes and
Rhythms — Life and Death (Echoes) — Bric-a-Brac: Ballads,
Rondels, Sonnets, Quatorzains, and Rondeaus, xii., 175 pages,
i6mo. 1893. Etched Title-page Vignette of the Old Infirmary,
Edinburgh, by W. HOLE, A. R.S. A. Fourth Edition. Cloth.5s.net.
Mr. R. L. Stevenson says at the close of his Christmas Article

(Scribner, 1888) : ' ' From a recent book of verse, where there is more

than one such beautiful manly poem, I take this memorial piece; it says

better than I can what I love to think."

The Spectator says " the author is a genuine poet."

LONDON VOLUNTARIES. Being the Second Edition

of the ' ' Song of the Sword." 1893. i6mo, 130 pages. Cloth, top
gilt, 53. net.

*„* This Second Edition has been enlarged by the inclusion of
" Arabian Nights' Entertainment" and Epilogue "Something is Dead''

LYRA HERO 1C A. An Anthology selected from the best
English Verse of the Sixteenth, Seventeenth, Eighteenth, and Nine-
teenth Centuries. Library (Third) Edition. Crown 8vo, xviii.,
362 pages. Cloth, uncut edges, stamped cover, 35. 6d. ; or, School
(Second) Edition, ismo, cloth, as.

*** Unanimously recommended by organs of every shade of opinion,
critical and political, as the best Anthology of Heroic Verse in the
language.

BLISS CARMAN

LOW TIDE ON GRAND PRE. Poems. 1893. Small

4to, 116 pages. Cloth, 55. net.

Academy.— "All these Acadian lyrics are dominated by an over-
mastering sense of beauty, and permeated with a love of nature, for
which there is no epithet so apt as the much misused and hackneyed
word intense."

CHARLES SAYLE

MUSA CONSOLATRIX. Poems. 1893. I2mo, 130

pages. Wrapper. 35. 6d.

H. D. RAWNSLEY

IDYLLS AND LYRICS OF THE NILE. 1894. I2mo,
148 pages. Cloth.

A Selection from David Nutts

CRITICISM

WILLIAM ERNEST HENLEY

VIEWS AND REVIEWS. Essays in Appreciation.
Literature. Second Edition. 1893. i6mo, xii., 228 pages.
Cloth, top gilt, 55. net.

Spectator. — " One of the most remarkable volumes of literary criticism
— in more senses than one it is the most striking — that have appeared
for years."
Athenaum. — " Attracts and delights the reader."

JOSEPH JACOBS

GEORGE ELIOT; MATTHEW ARNOLD; BROWNING;

NEWMAN ; Essays and Reviews from the Athenceum. i6mo.
1891. xxiv., 152 pages. Cloth. 25. 6d.
Spectator.— '' Full of suggestion."
Literary World. — "As important as it is interesting."
Pall Mall Gaxette. — "Scholarly and moderate in tone and broad in
view."
TENNYSON: and, IN MEMORIAM. i6mo. 1892.

108 pages. Cloth, 25.

Spectator: —' ' A thoughtful and kindly critic."

Anti-Jacobin. — " Exceptionally well written, and the work of a
clear, acute, and independent mind."

GEORGE MOORE

IMPRESSIONS AND OPINIONS. (Balzac, TurgueniefT,
Verlaine, Ibsen, Mummer Worship, &c). i6mo, 346 pages,
Cloth. 1891. 55.

National Observer. — " Packed full of ideas."

Literary World. — "One of the freshest and most interesting volumes
of criticism that have appeared for some time.' '

G. S. STREET

MINIATURES AND MOODS. i6mo. 1893. vii., 113

pages. Cloth, 33. 6d.

National Observer. — " In its charmingly foppish fashion a decided
success."

W. C. BROWNELL.

FRENCH TRAITS. I2mo, 400 pages. Cloth, uncut, ;s.

*** The most suggestive and penetrating study of the social and
moral ideas of French Society in the English language,

Athenceum. — " We doubt whether so thoughtful, so valuable a page
of international criticism has appeared since M. Taine's ' Notes on
England.' "
FRENCH ART. Crown 8vo. 1892. 248 pages. Cloth, 6s.

Athenceum. — "Rare perspicacity of judgment and clearness of
expression."

List of Publications

GIFT BOOKS FOR THE YOUNG

JOSEPH JACOBS

THE FAIRY TALES OF THE BRITISH EMPIRE.

Collected and Edited from printed sources and oral tradition.
With Notes exhibiting the sources, parallels, and literary history of
each tale. Profusely illustrated with full page-plates, head and
tail pieces, cuts in the text, and initials, by J. D. BATTEN. Square
crown 8vo. Each volume, upwards of 270 pages, sumptuously
printed on special paper, with wide margins, in specially designed
cloth cover, 6s.

Already Issued

ENGLISH FAIRY TALES. 1890. CELTIC FAIRY TALES. 1891.
INDIAN FAIRY TALES. 1892. MORE ENGLISH FAIRY TALES.

1893.

*^* The laudatory press notices are too numerous to quote from. The
Series has unanimously been recognised as one of the best ever issued.

OSCAR WILDE

THE HAPPY PRINCE, AND OTHER TALES. 116

pages, small 410. Old-faced type, on cream-laid paper with wide
margins, Japanese vellum cover printed in red and black. With
Three full-page Plates and Eleven Vignettes by WALTER CRANE
and JACOMB HOOD. Second Edition. Price 33. 6d.
Athenceum. — "The gift of writing fairy tales is rare, and Mr. Oscar
Wilde shows that he possesses it in a rare degree."

DOLLIE RADFORD

SONGS FOR SOMEBODY. With designs by G. M.
BRADLEY. Square 8vo. 32 pages printed in monotint, and 6
pages printed in colours, by EDMUND EVANS. With specially
designed cover, printed in colours. 35. 6d.

*** In addition to the ordinary issue, a few copies have been struck o/
o/i the finest J apanese vellum, -with the cover design stamped in gold.
The price of these is i$s. net.
Sylvia' s Journal. — "A most delightful little book for children."

FOLK-LORE

G. L. COM ME

A DICTIONARY OF BRITISH FOLK-LORE. Part I.

—The Traditional Games of England, Scotland, and Ireland, with
Tunes, Singing-rhymes, and methods of playing according to the
variants extant and recorded in different parts of the kingdom.
Vol. I.— Accroshay— Nuts in May. Demy 8vo, xx., 433 pages.
Numerous Illustrations. Cloth, 123. 6d. net.

(Vol. II. Christmas 1894.)

4 A Selection from David Nut? s List of Publications

MISS L. M. GARNETT

THE WOMEN OF TURKEY AND THEIR FOLK-
LORE. Complete Edition, with Ethnological and Sociologic.il
Appendices, by J. S. STUART-GLENNIE. Two Vols., large 8vo,
(noo pages). 1890-91. £1 6s. 6d. ; or, Text Edition in One
Volume (looo pages), los. 6d.

%* The standard English work on the Folk-lore of the Balkan Penin-
sula and the Levant.

G. L. GRINNELL

GRINNELL'S PAWNEE HERO STORIES AND
-FOLK-TALES— GRINNELL'S BLACKFOOT LODGE TALES
Two Vols. 8vo. 1893. (445 and 310 pages). Each 75. 6d.
*»* Mr. Grinncll lived many years among the Prairie Indians, and
his two volumes are the most compact and faithful account of their
Mythology, Legendary History, and Ceremonial Customs.

ALFRED NUTT

STUDIES OF THE LEGEND OF THE HOLY GRAIL,
with special reference to the Hypothesis of its Celtic Origin. 8vo.
1888. 300 pages. los. 6d.
%* The only English work on this branch of the Arthurian Romance.

DR. DOUGLAS HYDE

BESIDE THE FIRE. Irish Folk-Tales, Collected,
Edited, and Translated by Dr. DOUGLAS HYDE. With Additional
Notes by ALFRED NUTT. Crown 8vo, 260 pages. 1891. 75. 6d.

ALICE MORSE EARLE

CUSTOMS AND FASHIONS IN OLD NEW ENG-
LAND. i2mo, 320 pages. Cloth, uncut, gilt top. 75. 6d.

CONTENTS : — Child Life — Courtship and Marriage — Domestic
Service — Home Interiors — Table Plenishings— Larder Supplies —
Drinks and Drinkers — Travel, Tavern, and Turnpike — Holidays and
Festivals— Sports and Diversions — Books and Bookmakers — Artifices
of Handsomeness — Raiment and Vesture— Doctors and Patients —
Funeral and Burial.

%* A mine of information for the Student of English over-sea manners
and speech in the ijth and i8th centuries. .

KUNO MEYER

THE VISION OF MACCONGLINNE. A I2th Century

Irish Wonder Tale. Edited and Translated by KUNO MEYER.

With Literary Introduction by W. WOLLNER. Crown 8vo. 1892.

liv. and 212 pages. Cloth, los. 6d.

%* A most picturesque and spirited Rabelaisian tale dealing with a

supernatural land of plenty. Equally important to the Irish philologist

and to the student of mediaeval legend.

U.C.BERKELEY LIBRARIES