THE BIOLOGY OF SENESCENCE

v-*| r «-- HL.y

c

THE BIOLOGY OF

SENESCENCE

by

Alex Comfort

RINEHART & COMPANY, INC.

Publishers 1956 New York

Printed in Great Britain

by Butler & Tanner Ltd

Frome and London

To the question propounded . . ., I
can make only one answer: yes, it is
useful to prolong human life.

ILYA METCHNIKOFF (1907)

fc*A*P

PREFACE

This book is a compilation. It was written as an aid to my own
research, in a subject where it is difficult to know where to
begin, but I hope that the references, at least, will be useful to
others.

The denunciation of a subject and its current theoretical
basis as 'unsatisfactory' is a relatively easy exercise — dealing
with it satisfactorily is quite another matter. No biological
treatment of senescence can hope to be satisfactory in the
absence of a great deal of factual information which at present
is not there. I have attempted to collect as much of this informa-
tion as possible: since most of it comes from fields in which I
have no experience, there are bound to be errors both of fact
and of deduction in such a survey, and I hope that they will
be pointed out to me.

I am deeply grateful to Professor Peter Medawar, F.R.S.,
under whom I have worked, to Professor J. B. S. Haldane,
F.R.S., for kindly drawing my attention to a number of refer-
ences I would not otherwise have seen, and to many colleagues
whom I have molested for information or criticism, and whose
help and advice has been invaluable, though they bear no
responsibility for the result. I am also profoundly indebted to
the Nuffield Foundation for several years' financial support, to
Dr. Harrison Matthews, Director of the London Zoo, for access
to its records, and to Miss Rosemary Birbeck for much help in
preparing the manuscript and bibliography.

ALEX COMFORT

December 1954

vn

CONTENTS

PREFACE vii

INTRODUCTORY AND HISTORICAL 1

1 THE NATURE AND CRITERIA OF SENESCENCE

1 • 1 The Measurement of senescence 1 7

1 -2 Forms of senescence 33

1-2-1 Mechanical senescence 34

1-2-2 'Accumulation' and 'depletion' 35

1-2-3 Morphogenetic senescence

36

1 -3 Senescence in evolution 37

2 THE DISTRIBUTION OF SENESCENCE

2-1 Character of the evidence 42

2-2 Maximum longevities in animals 45

2-2-1 Mammals 46

2-2-2 Birds 49

2-2-3 Reptiles 51

2-2-4 Amphibians 52

2-2-5 Fish 53

2-2-6 Invertebrates 54

2-3 Maximum life-span in man 59

2-4 Distribution of senescence in vertebrates 63

2-4-1 Fish 68

2-4-2 Reptiles 77

2-5 Distribution of senescence in invertebrates 79

2-5-1 Porifera 80

2-5-2 Coelenterates 81

2-5-3 Sundry invertebrates S3

2-5-4 Rotifers 84

2-5-5 Arthropods 92

2-5-6 Molluscs 102

ix

71047

Contents

2-6 Senescence in wild populations 108

2-6-1 Vertebrates 108

2-6-2 Invertebrates 112

3 SENESCENCE IN PROTOZOA

3-1 Individual cells 114

3-2 The 'senescence' of clones 116

4 THE INFLUENCE OF GENETIC CONSTITUTION

ON SENESCENCE AND LONGEVITY

4-1 Inheritance of life-span 121

4-1-1 General 121

4-1-2 Parental age 125

4-2 Heterosis or hybrid vigour 127

4-3 Sex differences 130

4-4 Progeria 134

4-5 Choice of material for the experimental study of age

effects 136

5 GROWTH AND SENESCENCE

5-1 'Rate of living' 138

5-2 Experimental alteration of the growth rate 143

5-2-1 Invertebrates 143

5-2-2 Insect metamorphosis and senescence 146

5-2-3 Vertebrates 148

5-3 Growth-cessation and mammalian senescence 153

6 THE MECHANISMS OF SENESCENCE 162
6-1 Senescence in cells 163

6-1-1 'Irreplaceable' enzymes 163

6-1-2 Cell turnover 166

6-1-3 Somatic mutation 1 68

6-1-4 Specificity 169

6-2 Endocrine senescence 171

6-2-1 General 171

6-2-2 Gonad-pituitary system 177

6-2-3 Hormonal regulation of growth in vertebrates. 1 82

CONCLUSION 189

BIBLIOGRAPHY 201

INDICES 245

x

ILLUSTRATIONS

1 Annual rate of mortality per 1,000 by sex: United States,
1939-41 15

2 Number of survivors out of 100,000 born alive, for each
race by sex: United States, 1939-41 15

3 Number of Survivors out of 100,000 male live births, from
recent life- tables for selected countries 16

4 Frequency distribution of ages at death in a cohort starting
with 100,000 live births, based on the mortality of white
males: United States, 1939-41 16

5 (a) Survival curve at a constant rate of mortality 19
(b) Survival curve of a population which exhibits sene-
scence 19

6 Types of survival curve 20

7 Survival curves of a German population 22

8 Survival curves for cafeteria tumblers 23

9 African Ibis (Threskiornis aethiopicus) . Survival of 21 indi-
viduals 27
Night Heron (Nycticorax nycticorax). Survival of 17 indi-
viduals 27

10 Orkney Vole (Microtus orcadensis) . Survival of 24 individuals 27

11 Patagonian Cavy (Dolichotis patagona). Survival of 55
individuals 28

12 Mouflon Sheep (Ovis musimon). Survival of 77 individuals 28

13 Irish Wolfhounds. Survival of 67 individuals from 12
months of age — sexes combined 29

14 The decline in egg production in successive years of laying

— domestic fowls 63

15 (a) Growth in length of male fish of the genera Xiphophorus,
Lebistes and Heterandria during the first year of life 66
(b) Growth in length of female fish of the genera Xipho-
phorus, Lebistes and Heterandria during the first year of life 66

16 Growth of trout in Windermere and the small tarns 71

xi

Illustrations

1 7 (a) Growth-constant for growth in length of the sturgeon,
Acipenser stellatus, at various ages 72

(b) Growth in weight of the sturgeon (Acipenser stellatus)

and the bream (Abramis brama) 73

(c) Growth in weight of the Bream (Abramis brama) 74

1 8 (a) Life-table of Lebistes reticulatus 75
(b) Growth of female Lebistes reticulatus 76

19 Growth of Emys 78

20 Survival curves for Lecane inermis 86

21 Growth in length of Philodina citrina 87

22 Life-span and egglaying of Philodina citrina over 6 genera-
tions in normal culture 88

23 Progressive decline in life-span of a strain of Philodina
citrina (Rotifera) raised in each generation from eggs laid

by old mothers 89

24 Life-span of successive generations of Philodina reared in
each generation from the eggs of 4, 11, and 1 7 day old
mothers 90

25 (a) Growth of Daphnia magna — first type 93

(b) Growth of D. magna — second type 94

(c) Growth of D. magna — third type 95

26 Survival curves of 143 isolated virgin females and 44
isolated, fertilized females of the moth Fumea crassiorella 99

27 Egg production of Eulota fruticum 103

28 Life-span of the pulmonate Limnaea columella 105

29 Growth and longevity of Patella vulgata in various stations,
showing the short life of rapidly-growing populations 106

30 Smoothed survival curve for the vole, Microtus agrestis in
captivity 109

31 Survival curves of mice in laboratory culture — breeding
females 122

32 Drosophila subobscura. Strain K. Survival curves of flies
raised in each generation from eggs laid by adults which

had passed the 30th day of imaginal life 1 24

33 Drosophila subobscura — hybrid vigour and longevity. Sur-
vival curves for the inbred lines B and K, and for the
reciprocal hybrids between them 128

34 Survivorship curves for 82 males and 45 females of the
black widow spider Latrodectes mactans 131

xii

Illustrations

35 Survivorship, death and death-rate curves for the black
widow. 131

36 Survivorship curves for male and female Tribolium madens 131

37 Effect of restricted food upon the longevity of Daphnia
longispina 144

38 Effect of restricted food upon the duration of instars in
Daphnia longispina 145

39 Effect of restricted food upon the rate of senile change in

the heart rate of Daphnia longispina 146

40 Survival curves of normal and retarded male and female
rats, showing the effect of dietary restriction 149

41 Neutral 1 7-Ketosteroids, 24-hour urinary excretion of
human males 1 73

42 The postnatal growth in weight of male children 1 84

43 Annual growth increment in boys, from the data of
Quetelet 184

INTRODUCTORY AND HISTORICAL

Man throughout history, and every individual since his child-
hood, has been aware that he himself, and those animals which
he has kept in domestication, will undergo an adverse change
with the passage of time. Their fertility, strength and activity
decreases, and their liability to die from causes which, earlier
in life, they could have resisted, increases.

This process of change is senescence, and senescence enters
human experience through the fact that man exhibits it him-
self. This close involvement with human fears and aspirations
may account for the very extensive metaphysical literature of
ageing. It certainly accounts for the profound concern with
which humanity has tended to regard the subject. To a great
extent human history and psychology must always have been
determined and moulded by the awareness that the life-span of
any individual is determinate, and that the expectation of life
tends to decrease with increasing age. The Oriental could say
'O King, live for ever!' in the knowledge that every personal
tyranny has its term. Every child since the emergence of
language has probably asked 'Why did that man die?' and has
been told 'He died because he was old.'

Interesting psychological and historical speculation could be
made on the part which this awareness has played in human
affairs. From the biologist's standpoint, its main importance has
been the bias which it has injected into the study of senescence.
The child who asks the question, and receives the answer, is
familiar with 'old' clothes and 'old' toys. He has always known
that he, his pets, his cattle and his neighbours will become
increasingly prone to breakdown and ultimate death the older
they get. He has observed from the nursery that inanimate and

1

The Biology of Senescence

mechanical systems also deteriorate with the passage of time.
He appears at a later age to derive some degree of comfort from
the contemplation of the supposed generality, universality and
fundamental inherence of ageing — or alternatively from draw-
ing a contrast between Divine or cosmic permanence and his
own transience. However inspiring this type of thinking may
have been — and it features largely in the past artistic and
philosophical productions of all cultures — its influence and its
incorporation as second nature into the thought of biologists
throughout history has seriously handicapped the attempt to
understand what exactly takes place in senescence, which
organisms exhibit it, and how far it is really analogous to pro-
cesses of mechanical wear. One result of the involvement of
senescence with philosophy and the 'things that matter' has
been the prevalence of attempts to demonstrate general theories
of senile change, including all metazoa and even inanimate
objects, and having an edifying and a metaphysical cast. Pro-
minent among these have been attempts to equate ageing with
development, with the 'price' of multicellular existence, with
hypothetical mechano-chemical changes in colloid systems,
with the exhaustion induced by reproductive processes, and
with various concepts tending to the philosophical contempla-
tion of decline and death.

It is not unreasonable to point out that these theories have
for the most part deeper psychological and anthropological
than experimental and observational roots. Some of them have
a few facts on their side. 'Reproductive exhaustion' does appear
to induce senescence in fish and in mollusca, and flowering is
a proximate cause of death in monocarpic plants, but the general
concept, especially when it is made a universal, owes a large
debt to the widespread belief in human cultures that sexuality
'has its price'. Extensions of mechanical analogies from the
wearing out of tools to the wearing out of animal bodies are
justifiable in a limited number of cases where structures such
as teeth undergo demonstrable wear with use, and where this
process limits the life of the organism; but they have also shown
a tendency to become generalized in the hands of biologists
who are devoted for philosophical, political or religious reasons,
to mechanism in the interpretation of human behaviour. State-

2

Introduction and Historical

merits that 'senescence is no more than the later stage of
embryology' resemble Benjamin Rush's great discovery,
that all disease is disordered function. They belong to the
category of word-rearrangement games, which have long been
played in those fields of study where there is as yet no 'hard
news'.

Although the religious, poetic, metaphysical and philo-
sophical literatures of senescence will not be examined here, the
detection and examination of analogies based upon them,
which have had a great, and generally adverse, influence on the
growth of our knowledge of age processes, must clearly play a
large part in any critical examination of the subject. The com-
ments of Francis Bacon, who was both a philosophical originator
of the scientific method, and the first systematic English geronto-
logist,1 provide one of the best critiques of the influence of such
analogies and thought-patterns, and they will be quoted without
scruple here.

The practical importance of work upon the biology of
senescence, beyond the fundamental information which such
work might give about the mechanisms of cell differentiation and
renewal, can best be seen from the diagrams at Fig. 1-3 and 7.
The advance of public health has produced a conspicuous shift
in the shape of the survival curve in man so far as the privileged
countries are concerned, from the oblique to the rectangular
form. This has been due almost entirely to a reduction in the
mortality of the younger age groups — the human 'specific age'
and the maximum life-span have not been appreciably altered.
The medical importance of work on the nature of ageing lies at
present less in the immediate prospect of spectacular interfer-
ence with the process of senescence than in the fact that unless
we understand old age we cannot treat its diseases or palliate
its unpleasantness. At present age-linked diseases are coming to
account for well over half the major clinical material in any
Western medical practice. The physician is constantly referring
to the biologist for a scientific basis for geriatrics, and finding

1 I dislike this word, but it is probably too well grown for eradication.
It should mean 'a student of old men' (yegcov) and gerontology the study
of old men. For the study of age itself, the subject of this book, we require
geratology (yfjQag), upon which it would be fruitless to insist.
B 3

The Biology of Senescence

that it is not there. The amount of material on which such a
foundation could be built has increased, though not very
rapidly, during the present century. Its quantity is still inversely
proportional to the importance of the subject.

There are not many adequate reviews of the modern bio-
logical literature. The most recent are those of Lansing ( 1 95 1 , 52) .
A previous review of mine contains little which is not repeated
here (Comfort, 1954). Some of the more celebrated 'general
theories' have received spirited treatment in a review by
Medawar (1945). The literature of animal population statistics
has been reviewed by Deevey (1947) and that of invertebrate
senescence by Szabo (1935) and by Harms (1949). It is a
pleasure to acknowledge my indebtedness to these reviews and
to the bibliography of Shock (1951). A great deal of clinico-
pathological material upon the age-incidence of various human
diseases and the weights of organs throughout life has been col-
lected by Burger (1954). In a depressingly large number of
fields, there has been little new information in the last twenty
years. Other reviews of specific topics will be cited in their
place. The senescence of plants is not discussed here: it has been
reviewed in some detail by Crocker (1939), to whose paper
there seems little to add.

Senescence is probably best regarded as a general title for the
v group of effects which, in various phyla, lead to a decreasing
expectation of life with increasing age. It is not, in this sense, a
'fundamental', 'inherent', or otherwise generalizable process,
and attempts to find one underlying cellular property which
explains all instances of such a change are probably misplaced.
It is important and desirable to recognize the origins of many
such general theories, which owe much to folk-lore on one
hand and to the emotional make-up of their authors on the
other. The demoralizing effect of the subject of senescence, even
upon biologists of the highest competence and critical intel-
ligence, is well illustrated by the following passage from Pearl
(1928), the father of animal actuarial studies:

'(Somatic death in metazoa) is simply the price they pay for
the privilege of enjoying those higher specializations of structure
and function which have been added on as a sideline to the

4

Introductory and Historical

main business of living things, which is to pass on in unbroken
continuity the never-dimmed fire of life itself.'

Warthin (1929), whose insistence upon the fundamental
impossibility of modifying the tempo of human ageing, now or
at any time in the future, has an orgiastic tone quite out of
keeping with the rashness of such a prediction, writes:

'We live but to create a new machine of a little later model
than our own, a new life-machine that in some ineffable way
can help along the great process of evolution of the species
somehow more efficiently than we could do were we immortal.
The Universe, by its very nature, demands mortality for the
individual if the life of the species is to attain immortality
through the ability to cope with the changing environment of
successive ages. ... It is evident that involution is a biologic
entity equally important with evolution in the broad scheme of
the immortal process of life. Its processes are as physiologic as
those of growth. It is therefore inherent in the cell itself, an
intrinsic, inherited quality of the germ plasm and no slur or
stigma of pathologic should be cast upon this process. What its
exact chemicophysical mechanism is will be known only when
we know the nature of the energy-charge and the energy-release of
the cell. We may say, therefore, that age, the major involution,
is due primarily to the gradually weakening energy-charge set in
action by the moment of fertilization, and is dependent upon
the potential fulfillment of function by the organism. The
immortality of the germ plasm rests upon the renewal of this
energy charge from generation to generation.'

This passage is highly typical of the literature of old age to the
present day. There can be few branches of biology in which
uplifting generalization of this kind has so long been treated as
a respectable currency for scientific thought.

In general, the more elaborate the attempts to depict
senescence in overall mathematical terms, the more intellec-
tually disastrous they have proved. One of the most celebrated
incursions of metaphysics into biology, that which postulates a
separate 'biological time', is best expounded in the words of
its sponsor, Lecomte du Noiiy (1936):

5

The Biology of Senescence

'When we refer to sidereal time as being the canvas on which
the pattern of our existence is spread, we notice that the time
needed to effectuate a certain unit of physiological work of
repair is about four times greater at fifty than at ten years of
age. Everything, therefore, occurs as if sidereal time flowed four
times faster for a man of fifty than for a child often. It is evi-
dent, on the other hand, that from a psychological point of
view many more things happen to a child in a year than to an
old man. The year therefore seems much longer to the child. . . .
Thus we find that when we take physiological time as a unit of
comparison, physical time no longer flows uniformly. This
affirmation revolts one if the words are taken in a literal sense.
But . . . the expression "flow of time" ... is entirely false and
does not correspond to a reality. When ... we say that physical
time measured by means of a unit borrowed from our physio-
logical time no longer flows uniformly, it simply means that it
does not seem to flow uniformly . . . Must one consider this fact
as the indication of a difference of magnitude between our
short individual period and the immense periods of the uni-
verse? Must we see a proof of the existence of such periods?
Who knows? All that we can say at present is that our crude
language, lacking appropriate words, translate this knowledge
into improper, inadequate expressions such as "There are two
species of time" or "Physiological time does not flow uniformly
like physical time" . . . We must not let ourselves be duped
by these words, etc. . . .'

It is startling how many distinguished biologists have sub-
sequently quoted the notion of a distinct 'biological* time with
apparent sanction. The alcoholic who draws on his bottle
irregularly will find that its progress towards emptiness follows
an irregular scale, 'alcoholic time', so that judged by the rate
of emptying of the bottle, 'sidereal' time appears to progress
unevenly. But variation in rate is hardly an occult, or even an
unfamiliar, phenomenon. Like others before him, du Notiy has
gone down clutching a platitude and come up embracing a
metaphysical system.

In almost any other important biological field than that of
senescence, it is possible to present the main theories historic-

6

Introductory and Historical

ally, and to show a steady progression from a large number of
speculative, to one or two highly probable, main hypotheses.
In the case of senescence this cannot profitably be done. The
general theories of its nature and cause which have been put
forward from the time of Aristotle to the present day have fallen
into a number of overall groups, and have been divided almost
equally between fundamentalist theories which explain all
senescence, or treat it as an inherent property of living matter
or of metazoan cells, and epiphenomenalist theories which
relate it to particular physiological systems or conditions. They
are also fairly evenly divided between the various categories of
Baconian idola. It is a striking feature of these theories that they
show little or no historical development; they can much more
readily be summarized as a catalogue than as a process of
developing scientific awareness. To the fundamentalist group
belong, in the first place, all theories which assume the exist-
ence of cellular 'wear and tear' (Abnutzungstheorie) without fur-
ther particularization (Weismann, 1882; Pearl, 1928; Warthin,
1929); the mechanochemical deterioration of cell colloids
(Bauer, Bergauer, 1924; Ruzicka, 1924; 1929; Dhar, 1932;
Lepeschkin, 1931; Szabo, 1931; Marinesco, 1934; Kopaczew-
ski, 1938; Georgiana, 1949); and pathological or histological
elaborations of these, which attribute senescence to inherent
changes in specified tissues, nervous (Muhlmann, 1900, 1910,
1914, 1927; Ribbert, 1908; Vogt and Vogt, 1946; Bab, 1948),
endocrine (Lorand, 1904; Gley, 1922; Dunn, 1946; Findley,
1949; to cite only a few from an enormous literature in which
the endocrine nature of mammalian senescence is discussed,
stated or assumed), vascular (Demange, 1886), or even con-
nective (Bogomolets, 1947). To the epiphenomenalist group
belong toxic theories based on products of intestinal bacteria
(Metchnikoff, 1904, 1907; Lorand, 1929; Metalnikov, 1937),
accumulation of 'metaplasm' or of metabolites (Kassowitz,
1899; Jickeli, 1902; Montgomery, 1906; Muhlmann, 1910;
Molisch, 1938; Heilbrunn, 1943; Lansing, 1942; etc.), the
action of gravity (Daranyi, 1930), the accumulation of heavy
water (Hakh and Westling, 1934) and the deleterious effect
of cosmic rays (Kunze, 1933). There are also general develop-
mental theories which stress the continuity of senescence with

7

The Biology of Senescence

morphogenesis (Baer, 1864; Cholodkowsky, 1882; Roux,
1881; Delage, 1903; Warthin, 1929) or the operation of an
Aristotelean entelechy (Driesch, 1941; Burger, 1954), metabolic
theories introducing the concept of a fixed-quantity reaction
or of a rate/quantity relationship in determining longevity
(Rubner, 1908; Loeb, 1908; Pearl, 1928; Robertson, 1923),
attainment of a critical volume-surface relationship (Muhl-
mann, 1910 etc.), depletive theories relating senescence to
reproduction (Orton, 1929) and finally an important group of
theories which relate senescence to the cessation of somatic
growth (Minot, 1908; Carrel and Ebeling, 1921; Brody, 1924;
Bidder, 1932; Lansing, 1947, 1951). Most of the older theories
have been reviewed, against a background of Drieschian neo-
vitalism, in the textbook of Burger (1954).

The distribution of dates in this catalogue sufficiently indi-
cates the state of the subject. When Francis Bacon examined
the relationship between animal specific longevity, growth-rate,
size and gestation period, he concluded that the available facts
were unfortunately insufficient to support a general theory.
That conclusion remains valid in practically all the instances
quoted, but Bacon's self-denial failed to set a precedent for his
successors. Almost all these theories, judging from the literature,
continue at some point to influence biological thinking: some
can be partially, or even largely, justified by the suitable selec-
tion of instances. Others did not bear critical inspection at the
time they were first formulated, bearing in mind the known
behaviour of cells, and the known discrepancies in longevity
and in rate of ageing between animals of similar size, histo-
logical complexity, and physiological organization. Relatively
few are supported by any body of fundamental experiment.
The devising of general theories of senescence has employed
able men, chiefly in their spare time from laboratory research,
for many years. It seems reasonable to assume that almost all
the mechanisms which might theoretically be involved have
been considered, and if we are to understand what does in fact
occur in a given ageing organism, we now need a combination
of general observation with planned causal analysis in experi-
mental animals.

The main theories of ageing will be discussed in the text.

8

Introduction and Historical

There are, however, a few which should be outlined in greater
detail here — either because they are still of importance, or
because, though untenable, they have a considerable surviving
influence.

The most influential nineteenth-century contribution to this
second category was probably that of Weismann, whose theory
sprang directly from his distinction between germ plasm and
soma. Weismann regarded senescence as an inherent property
of metazoa, though not of living matter, since he failed to find
it in protozoans and other unicellular organisms. Its evolution
had gone hand in hand with the evolution of the soma as a
distinct entity, and it was the product of natural selection, aris-
ing like other mutants by chance, but perpetuated as a posi-
tively beneficial adaptation, because 'unlimited duration of life
of the individual would be a senseless luxury'. 'Death', accord-,
ing to this view, 'takes place because a worn-out tissue cannot
forever renew itself. . . . 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' (1882). This argument both
assumes what it sets out to explain, that the survival value of
an individual decreases with increasing age, and denies its own
premise, by suggesting that worn-out individuals threaten the
existence of the young. It had the advantage, however, of being
an evolutionary theory, and we shall see later that this is the
only type of theory which today seems likely to offer a general
approach to the emergence of senescence in all the groups
which exhibit it. The idea that all somatic cells must necessarily
undergo irreversible senescence was challenged early in the
century by the studies of Child (1915) upon planarians, and of
Carrel (1912) upon tissue culture. The assumption that all
higher metazoa must ex hypothesi exhibit senescence, however,
dies hard, and the fallacious argument based on selection has
been repeated as recently as 1937 (Metalnikov, 1936, 1937).

A considerable number of metabolic theories were based on the
fact that an inverse relationship exists between length of life
and 'rate of living'. On the basis of calorimetric experiments,
Rubner (1908, 1909) calculated that the amount of energy
required for the doubling of weight by body growth was
approximately equal in a number of mammals. The energy

9

The Biology of Senescence

requirement for the maintenance of metabolism, per unit
adult body weight, was also approximately equal between
species. Rubner inferred that senescence might, from these
energy relationships, represent the completion of one particu-
lar system of chemical reactions, depending on a fixed total
energy expenditure. He was obliged to erect a special category
for man, whose energy requirement was found to be far higher
than in laboratory or domestic animals. Loeb (1902, 1908)
attempted to find out whether the temperature coefficient of
this hypothetical reaction was identical with that of general
rate of development. Working with echinoderm eggs at various
temperatures, and using a hatchability criterion to determine
'senescence', if the word can be used in such a highly-
specialized instance, he concluded that the two coefficients were
distinct. The importance of this work has been that its pre-
suppositions have recurred in later studies, where some authors
have based very similar inferences about the relationship of
growth and senescence to a 'monomolecular, autocatalytic
reaction' on the shape and supposed mathematical proportions
of the growth curve. As d'Arcy Thompson pointed out, this
might equally prove the 'autocatalytic' character of growth in
a human population. In fact, with suitable adjustment, curves
based on biological material can be made to provide support
for almost any hypothesis of this kind.

Little need be said of the various toxic or pathological theories
of mammalian senescence. We are really left with five historic-
ally important theories, or groups of observations: the sugges-
tion of Weismann that senescence is evolved, not intrinsic in
all cellular matter; the work of Pearl (1928) which leads to the
conception of a 'rate of living', such that factors which retard
development or reduce metabolism tend in many organisms to
prevent or postpone senescence; the work of Minot (1913,
1908), of which the most important surviving parts are his
relation of senescence to the decline of growth, and his insist-
ence upon its continuous and gradual character and its con-
tinuity with morphogenesis; the experimental studies of Child
(1915), which showed that cellular differentiation and 'sen-
escence' in planarians is reversible, and of Carrel (1912), who
demonstrated that some tissue cells derived from adult animals

10

Introductory and Historical

could be propagated indefinitely in vitro, and finally the theories
of Bidder (1932).

Minot considered that senescence was the direct outcome of
cell differentiation, that differentiated cells, by reason of the
changes undergone, chiefly by their cytoplasm, in the course of
morphogenesis, had become largely incapable of growth or
repair. He believed that the negative acceleration of specific
growth, found in a very wide variety of organisms, and ultimate
senescence, were products of this process, and that the first was
a measure of the second. It followed from this that the rate of
senescence, so defined, must actually be highest in embryonic
life and in infancy, when the rate of differentiation is highest.
Many of Minot's concepts, such as the rigid irreversibility of
cell differentiation, echoed later by Warthin (1929), the in-
capacity of differentiated cells for growth, and the necessarily
increasing liability to senescence of successive cell-generations,
are now disproved or at least impugned. His work, however,
leaves with us the two important concepts of a gradual process
of senescence linked to morphogenesis, and of a relation between
it and the decline of growth-potential. By using negative growth
acceleration and rate of differentiation as a direct measure of
senescence, Minot arrived at the conclusion that the rate of
senescence is highest in foetal, and least in adult, life. This
concept has been widely adopted. Its validity depends upon
the acceptance of Minot's definition; if senescence be regarded,
as we shall regard it, in terms of deteriorative change in the
organism's power of resistance, the idea requires qualification.

A far more important question, which had been latent in the
literature since Ray Lankester (1870) pointed to the apparent
non-senescence offish, was raised by Bidder (1932). With the
exception of Metchnikoff (1904, 1907) who was attempting to
relate longevity to the form of the digestive tract, very nearly
all biological theorists had assumed that senescence occurs in
all vertebrates. This may in fact be so, but if it is not, then
manifestly the general theories of senescence based on degree of
tissue differentiation, irreplaceability of neurones, and other
such systems fall to the ground. Bidder pointed out that there
were several lower vertebrates in which there was no ground
for suspecting that the mortality ever increased with increasing

11

The Biology of Senescence

age, beyond the inevitable increment from accumulation of
evident injuries. He suggested that vertebrate senescence is a cor-
relate of the evolution of determinate growth and of a final absolute size.
Bidder regarded determinate size as a property which had
evolved as a result of the migration of vertebrates to dry land.
He pointed to a number of instances in fish where constant
expectation of life, capacity for growth, and general vigour
appeared to persist indefinitely (Bidder 1925). Bidder's argu-
ment is of importance, and is worth quoting in full.

'Giant trees, cultures of chick cells and of Paramecium,
measurements of plaice and of sponges, all indicate that indefi-
nite grow this natural. Galileo proved it fatal to swiftly moving
land animals, therefore swiftly moving mammals and birds
were impossible until their ancestors had evolved a mechanism
for maintaining specific size within an error not impairing
adequate efficiency. Even without evidence of evergrowing
organisms, we could not suppose that the close correspondence
to specific size, which we see in all swiftly moving creatures of
earth or air, results from mere "senescent" fading-out of the
zygotic impulse to cell division and cell increase. Specific size
is probably most important to birds, with their aeroplane
mechanics stricdy enjoining conformity of scale to plan; but to
men it is most noticeable in man. Only familiarity prevents
marvel at the rarity of meeting a man more than 20 per cent
taller or shorter than 5 \ ft., or of discovering his remains in any
place, or any race, or any epoch. Probably our erect posture
enforces accurate proportions of length to weight, for running.

'Adequate efficiency could only be obtained by the evolution
of some mechanism to stop natural growth so soon as specific
size is reached. This mechanism may be called the regulator,
avoiding the word "inhibitor" so as not to connote a physio-
logical assumption. However ignorant we are of its nature, its
action is traced in anthropometric statistics; a steady diminu-
tion in growth rate from a maximum at puberty to a vanishing-
point in the twenties. That the regulator works through change
in the constitution of the blood is shown by the perpetual divi-
sion of Garrell's chick cells in embryonic plasma, whereas cell
division is ended in the heart of a hen.

12

Introductory and Historical

'I have suggested that senescence is the result of the con-
tinued action of the regulator after growth is stopped. The
regulator does efficiently all that concerns the welfare of the
species. Man is within 2 cm. of the same height between 18
and 60, he gently rises 2 cm. between 20 and 27, and still more
gently loses 1 cm. by 40 or thereabouts. If primitive man at
18 begat a son, the species had no more need of him by 37,
when his son could hunt for food for the grandchildren. There-
fore the dwindling of cartilage, muscle and nerve cell, which
we call senescence, did not affect the survival of the species, the
checking of growth had secured that by ensuring a perfect
physique between 20 and 40. Effects of continued negative
growth after 37 were of indifference to the race; probably no
man ever reached 60 years old until language attained such
importance in the equipment of the species that long experi-
ence became valuable in a man who could neither fight nor
hunt. This negative growth is not the manifestation of a weak-
ness inherent in protoplasm or characteristic of nucleated cells;
it is the unimportant by-product of a regulating mechanism
necessary to the survival of swiftly moving land animals, a
mechanism evolved by selection and survival as have been
evolved the jointing of mammalian limbs, and with similar
perfection' (Bidder, 1932).

Bidder's theory, besides raising the question of senescence as
an effect lying outside the 'programme' imposed by natural
selection, poses the highly important suggestion that there may
be two categories of vertebrates — those whose life span is fixed
as in mammals, and those whose life-span is not fixed. From
the theoretical point of view the establishment of the truth or
falsity of this suggestion might be the key problem in the elucid-
ation of mammalian ageing, since the disproof of almost all the
major existing theories of senescence would follow from the
demonstration that it is not universally present in vertebrates.
This might appear a simple issue of fact, but for reasons which
will appear later, no such demonstration one way or the other
has yet been forthcoming.

Bidder's theory marks the last major attempt to produce a
hypothesis of vertebrate senescence. No significant theory of

13

The Biology of Senescence

the general biology of ageing has appeared since, although its
evolutionary basis has been discussed (Haldane, 1941; Meda-
war, 1952). The decline in abstract speculation about old age
is probably in itself a very good augury for research. Much of
the previous published matter abundantly justified the view of
Bacon that 'the method of discovery and proof whereby the
most general principles are first established, and then inter-
mediate axioms are tried and proved by them, is the parent of
error and the curse of all science'.

In later work, the relation between growth-cessation and
ageing has been generalized to cover the senescence of all
kinds of organisms which have a fixed life span, such as the
rotifers (Lansing, 1947a) in which the mechanism of ageing
may be, and very probably is, quite unlike that which occurs
in vertebrates. The real importance of Bidder's suggestions
lies, however, in the possibility they indicate that mammalian
senescence may be a close evolutionary correlate of certain
investigable mechanisms, such as homoeothermy, which have
evolved with it. The probing of this possibility belongs to the
future.

14

100

o

o

o

90

c£

80

ill

a.

>

70

h-

<

60

H

O

50

z

UL

40

u

30

"" 20
_i
<
D 10

Z

I o

/ /
/ /
/ /

/ /

/ /

/ /

/ /

/ /

-♦— 52-38 TOTAL MALES

/ /
/ /

/ /

•*- 41 52 TOTAL FEMALES

/ /

/ /

^ —

0 10 20

AGE IN YEARS

30

40

50

60

70

80

Fig. 1. — Annual rate of mortality per 1,000 by sex: United States, 1939-41.

100.000

90,000

^ZEHE^r^

__^__

^--^

— ^~

~^_

80.000

«>.^

70,000"

60,000"

£ 50.000
O

VIV

o

o
o

\\ \ \
\\ \ \

oC

\ \ \\

^ 30.000-

\N \ N

u_

° 20.000 *

\v\\

a:

m

\\\

vlUMB

o

o

D O

X^V

AGE

IN

YEARS

^^k>

20

30

40

50

60

70

90

100

Fig. 2. — Number of survivors out of 100,000 born alive, for each race by
sex: United States, 1939-41.

100,000
90,000
80,000
70, 000 -I
60.000

"Dn,w^S^;^^

X

10

AGE IN YEARS
~20 30 ' 40~

50

60

70

80

90

100

Fig. 3. — Number of Survivors out of 100,000 male live births, from recent
life-tables for selected countries.

10 20 30 40 50 60 70
Age at death in years

90 100

Fig. 4. — Frequency distribution of ages at death in a

cohort starting with 100,000 live births, based on the

mortality of white males: United States, 1939-41.

1

l£7> t^J

THE NATURE AND CRITERIA OF

SENESCENCE

1 • 1 Measurement of Senescence

Senescence is a deteriorative process. What is being measured,
when we measure it, is a decrease in viability and an increase
in vulnerability. Other definitions are possible, but they tend
to ignore the raison d'etre of human and scientific concern with
age processes. Senescence shows itself as an increasing prob-
ability of death with increasing chronological age: the study of
senescence is the study of the group of processes, different in
different organisms, which lead to this increase in vulnerability.

The probability that an individual organism which has sur-
vived to time x will die before time x + 1 depends on the rate
of mortality (q) per 1000, meaning the number, out of 1000
individuals living at time x, who have died by time x + 1 . The
force of mortality (fi) is given at any age x by

dn — d .
* = ~n dx = ~dJln-n
where n is the number of individuals which have survived to
age x.

In most organisms, the likelihood of dying within a given
period undergoes fluctuations, often large, throughout the life
cycle. Senescence appears as a progressive increase throughout
life, or after a given stadium, in the likelihood that a given
individual will die, during the next succeeding unit of time,
from randomly-distributed causes; the pressure of the environ-
ment, which it has successfully withstood in the past, it now
ceases to be able to withstand, even though that pressure is not
increased. It is rare that we can determine the vulnerability of

17

The Biology of Senescence

an individual. Our estimate of it is determined statistically,
upon a population. The demonstration of such an increase in
vulnerability is a necessary condition for demonstrating sen-
escence: it is, obviously, only a sufficient condition if selective
mortality from age-distributed external causes is ruled out.
Real populations are subject to mortality both from random
and from age-distributed causes — the variation of exposure rate
throughout life is familiar in man; grown men are subject to
risks which do not affect children, and so on. Differences in
'risk' throughout life have been studied in some other animals,
such as the locusts whose causes of death were analysed by
Bodenheimer (1938) or the gall-fly Urophora (Varley, 1947).
Pearson (1895), in his mathematical analysis of the curve of
human survivorship into five components, attempted to limit
the meaning of 'senile mortality' to one such component, reach-
ing its maximum incidence between 70 and 75 years of age.
This would be an ideal solution if it were practicable, but
Pearson's analysis is artificial in the extreme, and his 'five
separate Deaths' directing their fire at different age groups are
not biologically identifiable. In general, however, a progres-
sively increasing force of mortality and decreasing expectation of
life in a population, if significant variation in exposure rate can
be excluded, is evidence of the senescence of its individual
members. The preliminary test for senescence in an animal
species depends, therefore, on the life-table of an adequate
population sample, studied with suitable precautions against
selective causes of death.

The expected differences in behaviour, and form of life-table,
between populations which age and which do not age are
shown in Figs. 5a,b. In a population not subject to senescence
and exposed only to random overall mortality, the decline of
numbers is logarithmic, and animals die, ex hypothesis from
causes which would have killed them at any age. In a popula-
tion exposed only to death from reduced resistance, due to
senescence, the curve approaches a rectangular form: after a
certain age, animals die from causes which would not have
killed them in youth. In one case the force of mortality is con-
stant, in the second it rises steadily with age. Thus in rats the
force of mortality rises after the ninth month of life in a geo-

18

The Nature and Criteria of Senescence

metrical progression (Wiesner and Sheard, 1934). Real survival
graphs are commonly intermediate in form between the two
ideal contours. Pearl and Miner (1935) distinguished three
main types of observed death-curve, varying in skewness from
the nearly rectangular in organisms with a low standing death-
rate throughout life, but showing a tendency to die almost
simultaneously in old age, to the logarithmic decline char-
acteristic of populations which show no senescence, or which
die out before it can become evident (Fig. 6) . A fourth theoret-
ical type, in which the curve is rectangular but inverse to that

TIME
Fig. 5. (a) — Survival curve at a
constant rate of mortality (50 per
cent per unit time) .

TIME
Fig. 5 (b). — Survival curve of a
population which exhibits sene-
scence.

found in the ideal senescent population, was recognized by
Pearl (1940) as a theoretical possibility; it seems to be realized
in nature among organisms which have a high infant mortality,
but whose expectation of life increases over a long period with
increasing age. This pattern of survival is characteristic of some
trees (Szabo, 1931) but probably also occurs in animals. 'There
may be animals in which the expectation of life increases con-
tinuously with age. This may be so for many fish under natural
conditions. It certainly goes on increasing for a considerable
time. Thus in a species where the expectation of life was equal
to the age, or better, to the age plus one week, no members
would live for ever, but a small fraction would live for a very
long time. A centenarian aware of the facts would pity a child,
with an expectation of life of only a few years, but would envy
a bicentenarian' (Haldane, 1953).
G 19

1000

800 -

600 -

400

200

1000

_?800

Q.

X

£ 600

o
>

> 400

200 -

1000

800 -

600

400

200

NX-.

~

v.

-

GROUP A

-

v

STARVED DROSOPHILA

(WILD)
STARVED DROSOPHiLA

—

-

(VESTIGIAL)

_

PR0ALES

-

*^-^. _

^§£ii.

^

—

V*

GROUP B

-

\\\

\\

ROACH

WILD DROSOPHILA

-

^

\

AUTOMOBILE

MOUSE

-

\

-

\

-

v\\

-

<X:

N.

•^

s.

\

\

\ \

—

\ \

GROUP C

VESTIGIAL DROSOPHILA

•X v

HYDRA

-

AGRIOLIMAX

-

vX<

-

-

s

V

N.

|

1

-100 -50 0 50 100 150 200

PERCENTAGE DEVIATION FROM MEAN DURATION OF LIFE

Fig. 6.— Types of survival curve (from Pearl, after Allee et al., 1949).

The Nature and Criteria of Senescence

Some real animal populations decline in an approximately
logarithmic manner. The 'potential immortality' of individuals
in a population following such a path of decline, an entirely
meaningless phrase which has caused much philosophical agita-
tion in the past, is not more significant as a practical issue than
the 'potential5 meeting of any pair of railway metals at infinity.
No population of organisms which is subject to a constant
overall death-rate contains 'potentially' immortal individuals.
The only advantage which a non-senescent organism possesses
over senescent forms is that the odds in favour of its death
within a fixed period remain constant instead of shortening
with the passage of time.

The human survival curve, in societies possessing developed
medical services and a high standard of living, is intermediate
between the rectangular and log-linear contours, but approaches
the rectangular, with an initial decline due to infant mortality.
Figs. 1-3 and 7 show, first, the comparative curves of mortality
for populations in the present century living under different
conditions of economic and climatic advantage, and second, the
change in form of the life- table for North German populations
between 1787 and 1800. Many life- tables for populations
before the advent of scientific medicine are given by Dublin
(1949). The significance of technical and economic privilege is
nowhere more evident than in the study of life-tables. The
effects of public health upon the life- table are expressed rather
in making it approach more closely to the rectangular shape
than in prolonging the preinflectional part of the rectangle. In
very many organisms, and in man under bad social and medical
conditions, the infant mortality is so large as to obscure all sub-
sequent trends, the curve coming to imitate Pearl's fourth,
inverse rectangular, type (Fig. 6). The terminal increase in
liability to die may also be masked by cyclical variations in
mortality associated with breeding or wintering, but the pre-
sence of such an increase remains an essential requisite for the
demonstration of senescence in an organism. In many senescent
populations, such as the sheep and cavies in Figs. 11 and 12,
the survival curve in adult life is not so much rectangular as
arith-linear, a constant number of individuals dying during each
unit of time, the mortality necessarily decreasing as the supply

21

The Biology of Senescence

of animals decreases, like the companions of Odysseus of whom
Polyphemus ate a fixed number daily.

The breakage rate of crockery or glasswear has sometimes
been used to illustrate the decline of a non-senescent population
(e.g. Medawar, 1952). A life-table for tumblers was actually

100

Fig. 7. — Survival curves of a German population. Hufeland's table ( 1 798)

is based on 'experience' and estimates. Silbergleit's data are based on official

statistics as given in the Deutschen Statistischen Jahrbuch for 1915. Both sets of

data relate to N. Germany (From Vischer, 1947)

constructed by Brown and Flood (1947) — that for annealed
tumblers approaches the curve of constant mortality, though
only roughly, while the decline of a smaller group of toughened
glass tumblers was nearly arith-linear (Fig. 8). Senescence,
however, does apparently occur in tumblers, since abrasions of

22

The Nature and Criteria of Senescence

the lip make subsequent cracking more likely (Brown and
Flood, 1947).

It is convenient to treat survival curves such as those of man

• Annealed glass tumblers
© Toughened glass tumblers

Fig. 8. — Survival curves for cafeteria tumblers.

A = 549 annealed glass tumblers • — •
B = 241 toughened glass tumblers 0 — 0

Time: 1 scale division = 2 weeks for Curve A, 5 weeks for curve B.

(Drawn from the data of Brown and Flood, 1947)

or Drosophila as combinations of the log-linear 'environmental'
curve, found where the standing death-rate is high, with a
terminal rectangular decline due to senescence, since it is evi-
dent that not all those individuals who die in middle life

23

The Biology of Senescence

owe any part of their misfortune to the senile increase in
vulnerability, however early in life this is taken to begin.

Bodenheimer (1938) draws a useful distinction between the
'physiological' longevity of a species — that attained under op-
timal conditions in a genetically homogeneous population and
approaching the longest recorded life-span within the species;
and the 'ecological' longevity, which is the mean longevity ob-
served empirically under given conditions. The ideal rectangular
'physiological' curve postulated by Bodenheimer is a convenient
abstraction, at most, since the genetic and environmental con-
ditions laid down for it cannot in practice be obtained in any
real population. But in some forms the observed life-table in
laboratory culture or domestication approximates to the ideal
rectangular form, and this approximation is closest of all in
some human societies. It is, however, pointless in terms of the
actuarial definition of senescence to pursue a 'physiological' as
opposed to a 'pathological' senescence in most laboratory
animals. If senescence is measured as increased general vulner-
ability, Bodenheimer's 'physiological' longevity represents only
the approximate region in which the rise in the curve of
vulnerability to all assaults of the environment becomes so steep
that even major protection against such assaults is insufficient
to prolong life very greatly. The pattern can be modified and
the apparent physiological longevity increased by removing
specific causes of death — e.g. enteritis and ear disease in old
rats (Korenchevsky, 1949) but the postponement of death
obtainable in this way is itself limited, and argument about
'natural' death, apart from pathological processes, in mammals
is quite otiose.

It is manifestly impossible to demonstrate senescence from
life-tables unless the mortality in early and adult life is suffi-
ciently low, and the number of animals reaching old age is
therefore sufficiently high, for an endogenous increase in sus-
ceptibility to death-producing factors of random incidence to
be evident. Thus wild mice die at a rate which precludes their
reaching old age, but mice kept under laboratory conditions
have a life-table similar to that of Western European human
populations in the year 1900 (Leslie and Ranson, 1940, Fig. 30,
p. 109 ; Haldane, 1953) : not even the most cherished labor-

24

The Mature and Criteria of Senescence

atory population can receive as detailed medical attention as
civilized man, but if such were possible, the life-table of mice
might then approach that for Western European man in
1953.

Organisms which undergo senescence, as judged by the life-
table, also exhibit specific age, meaning an age at death which is
characteristic of the species when living under conditions ap-
proaching Bodenheimer's 'physiological' conditions.

Some of the limitations of the statistical definition of sen-
escence have recently been re-stated by Medawar (1952). It is
obvious that any survival curve can be simulated by judicious,
or injudicious, choice of material. Tables based only on age at
death, a single arbitrary event, are open to serious criticism if
they are used as indices of a continuous process of declining
vitality. The shape of such a curve is a measure of many things,
including the genetic homogeneity of the sample. The incidence
of various risks itself varies between age groups: the statistical
appearance of senescence would, for example, be found in the
life- table of any population offish which was subject to frequent
fishing with a net of fixed mesh size. Selective predation cer-
tainly produces effects of this kind of nature. The increased
force of mortality among men of military age during a war is
not a manifestation of senescence. On the other hand, some
causes of mortality, such as cancer (Rutgers, 1953), have a
curve of incidence which parallels the total curve of mortality.
In employing the force of mortality as an index of senescence it
is essential, as we have seen, to exclude so far as possible external
factors which are not of random incidence in relation to age,
yet this cannot be done with strict logical consistency. In the
case of human life-tables, large secular changes in cause and
incidence of death may occur within an individual life-span,
while constitutional differences in rate of senescence between
individuals ensure that the genetical composition of the sur-
vivors at, say, age 60, is not representative of the whole cohort
under study. These sources of error are, in fact, capable of
avoidance or correction for most practical purposes, but they
must always be recognized in inferring senescence from any
life-table.

Since there is no direct way of measuring the liability of an

25

The Biology of Senescence

individual to die without actually killing it, the statistical defini-
tion of senescence, although it reflects a real process in indivi-
duals, can only be tested upon a population.

It is evident that in applied biology, and especially in
medicine, it is desirable to be able to infer not only the existence
of senescence in a species but the degree of senile change in a
given individual. This estimate must be based on secondary
criteria, and can be made with accuracy only in forms whose
life-cycle, like that of man or of Drosophila, has been subject to
intensive study. The importance of the statistical definition of
senescence is that it implies an obligatory recourse to adequate
population studies. An over-common practice has been to keep
a single specimen, a bird or a bullfrog, for ten or twenty years,
and, when it is found dead, having been so for hours or possibly
days, to describe histological appearances in its tissues in a note
entitled 'Senile change in the nervous system of Passer (or Bufo)\
While senescence cannot be inferred from every life- table in which
the force of mortality rises, neither can descriptions of 'senile*
changes be properly based on single observations upon sup-
posedly ageing organisms belonging to groups whose life-cycle,
in relation to senescence, is not fully known.

In practice, other criteria than the life-table can be applied
to organisms whose life-cycle is familiar, as secondary indices of
senescence; these are distinct from mere measures of chrono-
logical age, based upon the morphology of scales, teeth or
otoliths. Certain factors which are, in effect, direct measures of
vigour or of vulnerability, such as the mortality from burns,
(Ball and Squire, 1949), or even the annual absenteeism from
sickness (Schlomka and Kersten, 1952) follow the general force
of mortality in man. The supposed decline in the rate of wound
healing proposed as a measure of senescence by du Nouy (1932)
was based on grossly inadequate clinical material and is not
supported by later work (Bourliere 1950). Less general criteria
such as skin elasticity in man (Evans, Cowdry and Nielson,
1943; Kirk and Kvorning, 1949), organ weight and relative
organ hyporplasia in rats (Korenchevsky, 1942; 1949), heart
rate in Cladocerans (Ingle, Wood and Banta, 1937, Fig. 39,
p. 146), milk yields in cattle (Brody, Ragsdale and Turner,
1923), tgg production in fowls (Clark, 1940, Fig. 14, p. 63),

26

20-

African Ibis
Night Heron

•

10

O

z
>
>

D
at

LU
CO

z

D
Z

MONTHS

1 —

GO

i~ -o 1

CO
CM

50

100

150

200

Fig. 9.-

Night

-African Ibis ( Threskiornis aethiopicus) . Survival of 2 1 individuals —

sexes combined (London Zoo) • — •
Heron {Nycticorax nycticorax). Survival of 17 individuals — sexes
combined (London Zoo) 0 — 0.

24

18

16

12

8

DAYS
Fig. 10,

100

200

300

400

500

600

700

800

-Orkney Vole (Microtus orcadensis) . Survival of 24 individuals-
sexes combined (London Zoo).

bb j

50

{

40

••

•••

••
••

•

• i

30

bo

c

>

••

20-

>

CO

•-,

•
*••

10-

E
Z

i i i 1-

— > 1 f—

• •

•

e

* •

CO
O
CM

-. 1 9-h—

DAYS 1000 2Q00

Fig. 11.— Patagonian Cavy (Dolichotis patagona). Survival of 55 individuals
— sexes combined (London Zoo).

70

50

t

••.

i

>

D

CO •

10-

MONTHS

12 20 30 100 150

fIG# 12.— Mouflon Sheep {Ovis musimon). Survival of 77 individuals,
beginning at 12 months— sexes combined (London Zoo).

The Nature and Criteria of Senescence

YEARS

Fig. 13. — Irish Wolfhounds. Survival of 67 individuals from 12 months of
age — sexes combined. Line — whole sample, annual totals. Points — 38 in-
dividuals whose exact date of death was known, to scale (data from

Miss D. Gardner).

histological appearances of many kinds, and estimations of
general or special metabolism are of value within sharply-
defined limits, but all are subject to considerable variation apart
from the general senile process. In retarded Gladocera, for
example, where mean life-span is artificially prolonged by post-
poning growth, the heart rate fails to decline before death to
the low levels normally found in old age (Ingle, Wood and
Banta, 1937). Minot (1908), Hertwig and others considered that
a steady decrease in the nucleocytoplasmic ratio was a general
feature of ageing in organisms, a suggestion which has not been
upheld by later work, and which would today require trans-
lation into more precise biochemical terms. Dehydration was
also formerly regarded as a general senile phenomenon. From
a recalculation of older data, however, and from fresh material,

29

The Biology of Senescence

Lowry and Hastings (1952) find that increased hydration, due
perhaps to extracellular oedema, loss of cells, and even gross
pathological causes such as heart failure, is the most consistent
finding in senile mammalian tissue. There is at present no bio-
chemical sign characteristic of 'oldness' in tissues or in cells,
and the search for one may well reflect a fundamental mis-
conception. It is in assessing the relevance of all such criteria to
the main phenomenon of senescence, the decline in resistance to
random stresses, that the statistical approach is essential. All
assertions about senescence based upon pathological anatomy,
or upon general theories which treat it as a single process, are
open to question.

The decline of growth rate throughout life in some or all tissues
appears to be a near-universal feature of metazoa.

'The specific growth rate always falls: living tissue progres-
y sively loses the power to reproduce itself at the rate at which it
was formed. Minot arrived at this generalization, which should
rightly be known as "Minot's Law", from the collation of his
percentage growth-rate curves; and it was he who first recog-
nized that the point of inflection of the integral curve of growth,
and the division it makes between a period of positive and
negative acceleration, is not of critical importance. The pro-
gressive dissipation of "growth energy" which this first law
affirms was thought by Minot to be an expression of the
phenomenon of senescence — "ageing" with its everyday impli-
cations. Senescence is not, in this view, a process which sets in
after a preliminary period of maturation has run its course:
senescence is development, looked at from the other end of life'
(Medawar, 1945b).

The use of this criterion, which is a readily measurable one,
and can be applied to smallish groups of animals with suitable
precautions, as well as to single tissues or organs, implies the
acceptance of Minot's definition of senescence. The definition
is defensible. On the other hand, one of the important points at
issue at the present moment is precisely whether all animals
which show a decline in the specific growth rate — all verte-
brates, and almost all invertebrates — display pari passu an ulti-
mate increase in mortality either as a result of this process, or

30

The Mature and Criteria of Senescence

as a result of some factor which also causes that decline. The
decline is not a measure of senescence in its actuarial sense,
since it does not run parallel with the force of mortality, and
it would not be even an obligatory precursor of senescence if
mortality only increased with age in those animals whose
capacity for renewal or growth in some or all tissues has fallen
to zero. It is with actuarial, deteriorative senescence that we are
here concerned — if senility implied only the decline of growth
rate in man, it would cause little public concern.

It is possible that future work will produce a workable and
justifiable 'direct measure' of senescence in individuals, based
on the time-lag in cell-division of tissue explants derived from
old animals (Gohn and Murray, 1925; Suzuki, 1926; Medawar,
1940). No practical test of this kind has yet been developed,
however — meanwhile the critical observations on the distribu-
tion of senescence in vertebrates have almost all to be made
upon organisms (large fish, crocodiles, tortoises) where life-
table studies are out of the question. In these forms it is rela-
tively easy to observe histologically or by a mating test, the
degree of reproductive power persisting in individuals of known
age. Inferences based on 'reproductive' senescence are there-
fore easy to draw, compared with the insuperable difficulties
involved in measuring the force of mortality in such cases.

Reproductive decline is a very general feature of those vertebrates
which undergo senescence as judged by the increasing force
of mortality: its evidences in various forms include gonadal
changes, loss of secondary sexual characters, cessation of ovarian
cycles, and a fall in sperm production, fertilizing power, hatch-
ability, litter size and viability. These changes follow a time
scale which is different from that of the increase in force of
mortality, however, and which bears no constant relation to
that increase in different species. The gonad often appears to
behave as an 'organism' having its own determinate life-span,
but this is equally true of other structures, such as the thymus.
The limited life of the gonad is in a special category only
because, in terms of evolutionary teleology, the gonad is the
significant part of the organism. Ageing of the whole organism
after a prolonged post-reproductive period is a process which is
realized only by human interference, at least so far as most

31

The Biology of Senescence

species are concerned, and not 'envisaged' by evolutionary
teleology. It could be argued that once gonadal senescence has
become established in a species, eventual somatic senescence is
a rule inevitable from the withdrawal in post-reproductive
life of the selection-pressure towards homoeostasis. A clear
physiological link between the activity of the gonad and the
growth and survival of the animal has been demonstrated in a
few forms (e.g. Daphnia, Edlen, 1937, 1938), although even in
Daphnia, oogenesis continues until death (Schulze-Robbecke
1951). In many vertebrates, however, even total castration has
little or no adverse effect on longevity, and may increase it.
Although senescence of the gonad is, in an evolutionary sense,
the most important form of ageing, it is not self-evident that in
the artificially-protected animal it must always be followed by
generalized somatic senescence, unless the two processes are
causally related. Such an identification reflects, once more, a
human preoccupation. Reproductive function, because of its
ease of measurement, remains at most a justifiable test of con-
tinuing vitality in old animals — fertility at least indicating the
absence of irreversible organ changes in one important system.
Metabolic decline, either measured directly by calorimetry and
manometry, or inferred from reduction in spontaneous activity,
has also been regarded as an index of senescence — often on
theoretical grounds, as representing the accumulation of in-
active 'metaplasm' at the expense of active protoplasm (Kasso-
witz, 1899, etc.) or the completion of a 'monomolecular
autocatalytic reaction' such as that postulated by Robertson
(1923) or Bertalanffy (1941). The decline of heart rate in
Cladocerans (Ingle, Wood and Banta, 1937) has already been
mentioned. The mean resting heart rate in man also tends to
decline throughout foetal and postnatal life. Child measured the
age of hydromedusae by the decline in their rate of pulsation
(Child, 1918). In some invertebrates (planarians, Child, 1915;
hydromedusae, Child, 1918; molluscan adductor muscle, Hop-
kins, 1924, 1930) and in some isolated vertebrate tissues (arti-
cular cartilages, Rosenthal, Bowie and Wagoner, 1940, 1941,
1942; rat blood vessels, Lazovskaya, 1942, 1943; avian muscle,
Glezina, 1939; rabbit muscle, Cheymol and Pelou, 1944; rat
brain homogenate, Reiner, 1947; liver, kidney and heart homo-

32

The Nature and Criteria of Senescence

genates, Pearce, 1936; mouse lymphoid tissue, Victor and
Potter, 1935) 02 uptake has been reported to decline with age.
Calorimetric experiments on the whole mammal indicate a
general decline in heat production with increasing age (Sonden
and Tigerstedt, 1895; Benedict and Root, 1934; Magnus-Levy
and Falk, 1899; Shock, 1942, 1948; Benedict, 1935; Boothby
et aL, 1936; Kise and Ochi, 1934). This decline, however, like
that of growth-energy, is greatest in early life, and relatively
slight in man after the. age of 50 (Shock, 1953). It does not
parallel the senescent increase in mortality. There is also gross
individual variation. Kunde and Norlund found (1927) no
significant decrease in the basal metabolism of dogs up to 12
years of age. In rats, Benedict and Sherman (1937) found a
slight decrease in heat production with increasing age, measured
in the same individuals, but with the onset of senescence the body
weight itself declined, so that the metabolism per unit body
weight appeared to increase. In man Oa uptake per litre intra-
cellular fluid shows no decrease with age (Shock, Watkin and
Yiengst, 1954). A fuller bibliography is given by Shock (1951,
1953). It is not so far possible, in most organisms, to base
intelligible estimates of individual senescence upon changes in
metabolic rate.

1-2 Forms of senescence

Increase in death-rate and decrease in resistance after a
certain age might be expected in a number of model systems.
The curve of failure rate for mechanical devices such as lamp
bulbs, telephone switchboards (Kurtz and Winfrey, 1931), or
radar units bears a superficial resemblance to the mortality
curve of a senescent population, both in cases where all-or-none
failure results from wear or from the passage of time (lamp
filament failure, crystallization of metals, changes in condenser
dielectrics) or where wear is cumulative and inefficiency in-
creases to the point of failure (frictional wear, decline of cathode
emission). The resemblance to biological senescence is closest
in cases where several coincident processes ultimately become
self-reinforcing. The 'death-rate' of motor-cars, plotted by
Griffin (1928) and Pearl and Miner (1935) is closely similar to
that of wild-type Drosophila (Fig. 6).

33

The Biology of Senescence

1-2-1 MECHANICAL SENESCENCE

A few precise analogies to the failure of a non-replaceable
part in a mechanical system are known to occur in organisms.
Deterioration of the waxy epicuticle in insect imagines and of
the teeth in the African elephant (Perry, 1953), the mongoose
(Pearson and Baldwin, 1953), the shrew (Pearson, 1945; Pruitt,
1954) and some large carnivores are examples of strictly
mechanical senescence. Such changes would ultimately kill the
animal. Similar, though less obvious, mechanical changes may
contribute to senescence in other forms. It is probable that the
gradual loss of nephra in the mammalian kidney is an example
of the incidental loss of essential structures, but one which
rarely reaches the point of causing death per se. On the other
hand, the differences between an old cart and an old horse are
sufficiently striking to make the extensive acceptance of 'wear*
as an explanation of senescence, and the resort to mechanical
analogies based on the 'spontaneous slow decomposition' of
explosives (Lepeschkin, 1931) or the behaviour of inanimate
colloids (Ruzicka, 1924; Dhar, 1932) largely irrelevant. 'The
old organism does not contain old colloids, it contains newly-
formed colloids of an old character' (Lansing) . The mean half-
life of human protein is 80 days, of liver and serum proteins
10 days, and that of the carcase proteins 158 days (Bender,
1953). The continuation of high protein turnover in adult life
has been demonstrated by isotope studies (Shemin and Ritten-
berg, 1944) though the turnover of materials such as collagen
decreases almost to zero with increasing age (Perrone and Slack,
1952; Neuberger and Slack, 1953). The weakness of the 'col-
loidal' concept of ageing had been pointed out even before the
discovery of colloids: 'Quoniam vero duplex est duratio cor-
porum: altera in identitate simplici, altera per reparationem:
quarum prima in inanimatis tantum obtinet, secunda in vege-
tabilibus et animalibus; et perficitur per alimentationem' {Hist.
Vitae et Mortis).1

A chemical extension of the idea of 'mechanical' senescence

1 Since there are in fact two ways in which bodies maintain their identity,
the first, which applies only to inanimate objects, is simply by remaining
the same. The second which applies to plants and animals, is by renewing
themselves; and they do this by means of nourishment.

34

The Nature and Criteria of Senescence

could be based more plausibly on the existence of expendable
enzyme systems renewable only by cell division, to explain the
ultimate death of some fixed postmitotic cells (Cowdry, 1952);
this concept will be discussed later on. In all organisms except
those which are capable of total regeneration, mechanical
injury of a more general kind must accumulate with time, but
this process will vary greatly in rate under different environ-
mental conditions. The constancy of the specific age in forms
which senesce is a strong argument against the primacy of
'mechanical' ageing.

1-2-2 'ACCUMULATION' AND 'DEPLETION'

In addition to a limited number of cases in which mechanical
wear normally, or potentially, terminates an animal life-cycle,
most of the other postulated 'causes' of senescence such as the
accumulation of metabolites (MetchnikofT, 1915; Jickeli, 1902)
and the exhaustion of stored irreplenishable reserves, do very
probably contribute to senescence in specific instances. The
very large literature of calcium and pigment accumulation in
the cells of higher animals (reviewed by Lansing, 1951) deals
with changes which are probably reversible consequences,
rather than primary causes, of an underlying senile process.
Lansing (1942) found, however, that reduction in the calcium
content of the medium greatly increased the life-span of
rotifers. A similar increase was produced by a single immersion
in weak citrate solution. Accumulation of calcium with age was
demonstrated in the same organisms by microincineration.
Similar processes are described in plants (Molisch, 1938;
Ahrens, 1938; Lansing, 1942). The 'life' of spermatozoa, though
by no means analogous, has been shown to be prolonged by
chelating agents which bind Cu++ and Zn++ (Tyler, 1953). In
the case of the rotifer, at least, the evidence for an accumulative
element in senescence is fairly strong.

Depletion certainly terminates the life-cycle of some non-feed-
ing insect imagines, especially among Lepidoptera (Norris,
1934; Waloff, Norris and Broadhead, 1947), and possibly other
types of imago (Krumbiegel, 1929a,b). Many animals die or
become more vulnerable, as a result of the depletion or physio-
logical derangement caused by spawning (Orton, 1929). The
d 35

The Biology of Senescence

incidence of parental mortality in molluscs is reviewed by
Pelseneer (1935). Attempts to explain the human menopause
in terms of exhaustion of the supply of ova will be discussed
later. There is no evidence of a 'depletive' senescence in
mammals, unless the decline of growth rate be taken as
evidence of the exhaustion of some hypothetical substance.

1-2-3 MORPHOGENETIC SENESCENCE

The accumulation of injuries presents no biological problem
— it is readily observable in structures such as skin, and the
only serious difficulty lies in assessing its contribution to ageing
in particular structures or animal species.

But in addition to the processes of mechanical or metabolic
senescence, and sometimes affecting the same organisms if they
are protected from these, it is necessary to postulate a further,
morphogenetic senescence to explain the sequence of events
observed in many organisms. This senescence has been con-
sidered to arise directly from the operation of the processes of
cell development which determine the shape and size char-
acteristic of the species and of its organs either through changes
in cell behaviour, or through the effects of divergent processes
of heterogony; it expresses itself as a decline in the capacity to
regenerate or maintain structures or conditions which, during
growth and a post-growing period of variable duration, are nor-
mally regenerated and maintained. Morphogenetic senescence
is a cumulative failure of homoeostasis, affecting the body as a
whole, to which coincident or dependent mechanical failure or
accumulative processes may contribute, but which appears to
be continuous with the processes which control cell-differentia-
tion and regulation. More accurately, it appears to represent
the withdrawal of coordination between these processes, so that
physiological homoeostasis 'falls apart'. It is this form of sen-
escence which characterizes higher vertebrates and is par-
ticularly well seen in man. The chief evidence that this,
morphogenetic, senescence is more that the 'sum of environ-
mental insult' which was formerly invoked to explain it, is the
existence in many organisms of specific age, analogous to
specific size and possibly related to it, which displays little en-
vironmental, but marked inter-race and interspecific, variation.

36

The Nature and Criteria of Senescence

Much information about the behaviour of self-restoring and
self-regulating systems, and a number of important general
concepts, are now available from the study of mechanical
models. These analogies apply, strictly, only to the elucidation
of single components of the process of maintaining physiological
stability; the most important feature of 'cybernetics' and homoe-
ostasis in the organism has no precise mechanical analogy.
This is the fact that the homoeostatic process, the state of
quantitative invariance, or self-restoration, in various physio-
logical systems, is superimposed on qualitative and quantitative
change in the nature of the systems themselves, their specificity,
relative proportions, and function — in other words, upon de-
velopmental change. There is a strong inference that senescence
occurs when these long-term changes, which are probably con-
trolled or initiated largely by the same humoral mediators which
function in day-to-day homoeostasis, pass out of control, or
reach a point beyond which homoeostasis is no longer possible.

This argument ultimately stands or falls by the result of our
study of the phylogeny of senescence. If mammalian senescence
results from morphogenetic processes which ultimately escape
from the homoeostatic mechanisms that operate during adult
vigour, and if, on the other hand, some other vertebrates reach
a state of growing, or self-replacing, equilibrium, even over
limited periods, the problem of understanding mammalian sen-
escence will be very greatly restricted in theoretical scope,
though probably not very greatly simplified in experimental
detail. Such an equilibrium would be most likely to be found
in those forms where differential growth is least evident. The
evidence on this point will be examined later.

1 -3 Senescence in Evolution

Senescence has frequently been regarded as an evolved
adaptation, rather than as an inherent property of somatic
organization. This view, which is reasonably well in accord
with the existing, and very incomplete, evidence of its distri-
bution in phylogeny, was held by Weismann in spite of his
insistence on the contrast between germinal immortality and
somatic mortality. Weismann, however, regarded senile change,

37

The Biology of Senescence

and the limitation of the individual life-span, as a positively-
beneficial adaptation, and his argument is, as we have seen, of
a circular kind.

The theoretical difficulties of devising a system in which short
life is selected as a character of fitness are considerable though
not insuperable.1 In any circumstances where a high number of
generations in unit time has an adaptive value, the Weisman-
nian argument against individual longevity might hold. The
most conspicuous adaptive modifications of life-span in phylo-
geny seem, however, to be chiefly in the other direction. The
development of social insects probably depended upon the
evolution of long-lived sexual forms, and it is very likely that
a similar process may have operated in human phylogeny, in
connection with the development of social behaviour and the
family unit. Not only was the evolution of neurones having a
long potential life a condition for the development of elaborate
learned behaviour and long parental dependence, but, with the
development of rational power and social organization, the
advantages of possessing the experience of even a few long-lived
members was probably very high in any early hominid com-
munity. The social animals, especially man, provide one of the
best examples where longevity depending on factors outside the
reproductive period can theoretically be subject to positive
selection in terms of fitness.

The chief objection to Weismann's idea of senescence as an
adaptive effect is the rarity of its demonstrable occurrence in
nature. In all but the few forms discussed on pp. 108-13,
senescence is a potentiality, not a benefit or a handicap; it is
realized only when we interfere artificially with the animal or
its environment, and it is arguable whether evolution can select
for such potentialities. Bidder, it will be recalled (p. 12), con-
sidered that senescence in mammals was an evolutionarily un-
important 'by-product' of an important positive adaptation, the
limitation of size. It would indeed be possible to attribute senile
change to the accumulation of such by-products outside the
reproductive period. More recently it has been suggested that

1 Ribbands (1953) found an apparent example in worker bees, where
the summer brood could increase its working life by consuming pollen, but
uses it instead to rear additional larvae.

38

The Nature and Criteria of Senescence

senescence is to be regarded not as the positively beneficial
character which Weismann believed it to be, but as a potenti-
ality lying outside the part of the life cycle which is relevant to
evolution. It has certainly been 'evolved', in that the living
system which senesces has evolved, but it has not evolved as a
physiological mechanism. The line of argument which appears
most plausible is that suggested by Medawar (1945, 1952). It
seems probable, for a number of reasons, that except in certain
social animals there can be little effective selection pressure
against senescence as such. Normal population structure in wild
communities of animals, even in the absence of senescence, leads
to a continual preponderance of young reproducing over old
reproducing individuals, sufficient to override the advantage in
number of progeny which arises from a longer reproductive life.
Death from senescence is itself in many species so rare an event
in the wild state that failure to senesce early, or at all, has little
value from the point of view of survival. In many forms the
cessation or reduction of breeding capacity precedes senescence
proper — with certain exceptions in social animals, events occur-
ring in the post-reproductive period are theoretically outside
the reach of selection, and irrelevant to it. A consequence even
more important than the mere failure of evolutionary processes
to operate in favour of the postponement of senescence follows
from the same facts. In view of the constant reproductive pre-
ponderance of young individuals, the postponement of the
action of a harmful genetic effect until late in the reproductive
life is almost equivalent, in selective value, to its complete
elimination: the longer the postponement, the closer the equiva-
lence. The evolutionary 'demon' is concerned only to clear the
part of the life-span in which he works, not the parts which
might be reached if the environment were artificially made
more favourable. This mechanism, by acting to move all ad-
verse genetic effects which are capable of postponement and all
the consequences of divergent but temporarily beneficial systems
into the late reproductive or post-reproductive life, may itself
provide a partial explanation of the evolution of senescence, as
Haldane (1941) has already suggested. The selective equilibrium
reached in man would be expected in this case to be such
that the force of mortality is lowest when reproductive

39

The Biology of Senescence

activity is potentially highest, though the observed lowest level
falls rather earlier than this (10-12 years in males, Greville,
1946).

The selectionist argument which regards senescence as the
decline of evolved survival-power through successive age groups
is most convincing when we apply it to mammals and birds:
among invertebrates, reservations require to be made. In those
which are predominantly seasonal, with a total life-span less
than one year, and which winter as fertilized adults, it is by no
means true that at all times of the year young individuals must
outnumber old in a free-running population. The autumn con-
tingent of overwintering animals will consist of 'old' individuals.
In such forms, the selective advantage of different genotypes
will vary from season to season, and there will be an ultimate
requirement that the adult be capable of living long enough to
overwinter. Forms producing two broods annually will tend to
select fertility in the spring brood and longevity in the autumn,
but with a time-lag of one generation between selection and
potential expression. The mechanism of selection in such a
system must be very complicated.

In mammals some selective advantage would also presum-
ably attach to longevity where older males are polygamous and
younger males compete for the remaining females (deer,
baboons). The solipsist model of selection operating on 'the
individual' can obviously be upset by any selection pressures
introduced into the system by interaction between individuals,
and by community-patterns of ecological behaviour in the
species; the idea of an 'individual' animal unsupported by the
rest of the ecological community in which it lives is in fact
unbiological, and large unpredictable selection pressures affect-
ing the life-span may well arise from such hidden social
relationships.

In spite of this criticism, the theory of senescence as a measure
of declining selection-pressure is important. The declining
evolutionary importance of the individual with age may be
expressed in another way in the 'morphogenetic' senescence
seen in mammals. At the point where a system of differential
growth ceased to be regulated by forces which arose from
natural selection, it would cease to be under effectively direc-

40

The Nature and Criteria of Senescence

tional morphogenetic control, and would resemble an auto-
matic control device which has run out of 'programme'. In any
such system the equilibrium must be increasingly unstable.
These two views of senescence, as accumulation of delayed
lethal or sublethal genetic effects, and as a withdrawal of the
evolutionary pressure towards homoeostasis with increasing age,
are complementary, though probably only partial, pictures of
its evolutionary significance. The concept of senescence as
exhaustion of programme also restores a far greater unity to
our definition of ageing, which includes a great many effects
having little in common beyond their destructive effect on
homoeostasis. All such effects fall within the idea of deteriora-
tion lying outside the 'terms of reference' of each species, as laid
down by natural selection. The 'flying bomb' which failed to
dive on its objective would ultimately 'die' either of fuel
exhaustion, or through wear in its expendable engine. If its
design had been produced by evolution, and its evolutionary
relevance ceased at the moment of passing its objective, or
decreased as a function of the distance flown, both these events
would be outside the programme laid down by the selective
equilibrium, as they were outside the calculation of the design-
ing engineers. Death in such an expendable system may result
from one of many factors, and even, as Bidder recognized, from
the consequence of processes which contribute to fitness during
earlier life, such as systems of differential growth. We shall find
a good deal of gerontology is primarily the study of a living
system's behaviour after its biological programme is exhausted.
The various evolutionary explanations of ageing already com-
bine to offer us some idea of the reasons why this may be so.

41

2

THE DISTRIBUTION OF SENESCENCE

2-1 Character of the Evidence

To find out which animals exhibit an increasing mortality with
increasing age, we should ideally keep large numbers of each
species, or of representative species, from birth to death, under
optimal conditions of captivity. In point of fact, apart from the
impracticability of keeping any significant number of species in
this way, the results would be both artificial and potentially
misleading. It is possible to invent about animal senescence a
paradox rather analogous to the principle of physical uncer-
tainty: it is Virtually unknowable' or, in other words, mean-
ingless to ask, whether certain organisms are 'susceptible to
senescence', because the organism is biologically dependent on
its environment: in the wild state these forms never normally
live long enough to reach senescence, while domestication or
protective interference with the environment brings about
changes in physiology and behaviour which produce effectively
a different organism. The object of the paradox is to point out
the fruitlessness of argument over 'potential' behaviour which
is practically unrealizable. Almost all our detailed knowledge
of senescence comes either from the observation of man, or of
domestication-artefacts such as the laboratory mouse or the
laboratory strains of Drosophila. In the wild state it is most
unlikely that any species of Mus or of Drosophila reaches old age
with sufficient regularity to be subject to study. In most cases
we are creating for study a state which has no part in the life
cycle as it has been shaped by evolution, but is at most a
potentiality. This must be taken into account on every occasion
when theories of the evolution of senescence are being based on
the appearance of senescence in domestic animals.

42

The Distribution of Senescence

In comparative studies of animal senescence we have three
main types of information: single records of the extreme re-
corded longevity of different species under various conditions;
observations on the composition of natural populations,1 and
marking experiments, which give a measure of longevity and
mortality under natural conditions; and life- tables and com-
parable material plotted for animals under domestic and labor-
atory conditions, which indicate the susceptibility of these
species to eventual senescence, either by measuring the force of
mortality, or by measuring subsidiary age-characters such as
reproductive capacity. Studies of senescence in domestic animals
are singularly deficient, since apart from individual pets of
various phyla neither farmers nor laboratories are usually inter-
ested in maintaining their stocks throughout the whole life-
cycle and on into senescence. While there are extensive data on
the early development of almost all animals of economic im-
portance, the senile period is a virtually unworked biometric
field. The literature does not even contain life-tables for repre-
sentative species of each class of vertebrate. There is only one
published life- table for birds in captivity (domestic fowls —
Gardner and Hurst, 1933) and that is incomplete. No life-table
appears to have been published for any reptile or fish, including
common and short-lived aquarium species, or even for dogs or
cats. Apart from man, adequate or partially adequate actuarial
data exist in the literature for the following animals in captivity:

Mice (KoboziefT, 1931; Murray and Hoffman, 1941; Grune-

berg, 1951, etc.) (Fig. 31).
Rats (Slonaker, 1912; Wiesner and Sheard, 1934).
Rattus natalensis (Oliff, 1953).
Voles (Leslie and Ranson, 1940) (Fig. 30).
Sheep (partial) (Kelley, 1939).
Fowls (Gardner and Hurst, 1933).
Limnaea (Gastropoda) (Winsor and Winsor, 1935; Baily,

1931) (Fig. 28).
Agriolimax (Gastropoda ) (Pearl and Miner, 1935) (Fig. 5c).

1 Methods of age determination in fish, molluscs and mammals have been
well reviewed (Trans. N.T. Acad. Scl, 16, no. 6, 1954) and their value
criticized.

43

The Biology of Senescence

Drosophila (Diptera) (Pearl, 1928; Pearl and Parker, 1922,

1924; Gonzalez, 1923; Alpatov and Pearl, 1929; Alpatov,

1931; Pearl and Miner, 1935; Bilewicz, 1953).
Aedes aegypti (Diptera) (Kershaw, Lavoipierre and Chalmers,

1953).
Acrobasis caryae (Lepidoptera) (Pearl and Miner, 1936).
Fumea crassiorella (Lepidoptera) (Mathes 1951) (Fig. 26).
Telea polyphemus (Lepidoptera) (Pearl and Miner, 1935).
Bombyx mori (Lepidoptera) (Alpatov and Gordeenko, 1932).
Tribolium confusum, T. madens (Coleoptera) (Park, 1945;

Pearl, Park and Miner, 1941) (Fig. 36).
Cockroaches (Blatta orientalis, Periplaneta americana) (Rau,

1924; Griffiths and Tauber, 1942).
Bees (Ribbands, 1952, 1953).
Locusts (Bodenheimer, 1938).
Daphnia (Cladocera) (McArthur and Bailey, 1926; 1929a,

1929b; Ingle, Wood and Banta, 1937; Dunham, 1938;

Anderson and Jenkins, 1942).
Latrodectes (Arachnida) (Deevey and Deevey, 1945) (Figs. 34,

35).
Rotifers (Jennings and Lynch, 1928; Lynch and Smith, 1931;

Miller, 1931; Edmondson, 1945a, b; Lansing, 1942 etc.),

Figs. 20, 22, 23).
Hydra (Pearl and Miner, 1935) (Fig. 5c).

There must be others which have been overlooked, especially
in the entomological literature. Some unpublished survival
curves, mostly from populations too small for actuarial treat-
ment, are given in the text figures:

Lebistes reticulatus (teleost) (Fig. 18&).
Night heron {Nycticorax nycticorax). (Fig. 9).
African Ibis (Threskiornis aethiopicus) (Fig. 9).
Orkney vole (Microtus orcadensis) (Fig. 10).
Patagonian cavy {Dolichotis patagond) (Fig. 11).
Mouflon sheep (Ovis musimori) (Fig. 12).
Irish wolfhounds (Fig. 13).

Figs. 9-12 are prepared from data in the records of the London

44

The Distribution of Senescence

Zoo, by kind permission of the Director, and Fig. 13 from data
sent me by Miss D. Gardner.

Maximum longevity records of animals species have a
definite, but limited, use in giving a comparative picture of the
possible longevity in different forms. They can give no direct
evidence of the distribution of senescence, but they can provide
an important test of a number of general theories — those based,
for example, on the exhaustion of neurones (Vogt and Vogt,
1946; Bab, 1948) are difficult to reconcile with the variation
in specific age and potential longevity between closely-related
forms. A large scatter of maximum recorded ages is in itself
suggestive, but not of course demonstrative, evidence of an
indeterminate life-span, except in cases where it is evidence only
of improving cultural methods and better understanding of the
requirements of the animal under laboratory or domestic con-
ditions. For a very large range of species we can readily infer
a 'potential' age which is never attained, either in the wild,
because of accident and predation, or in captivity, because the
animals cannot be kept alive in captivity — the 'potential'
longevity of snakes, chamaeleons (Flower, 1925, 1937) or
mammals of little known habits (pangolins — Flower, 1931) are
cases in point. 'Concerning the length and brevity of life in
beasts, the knowledge which may be had is slender, the obser-
vation negligent, and tradition fabulous; in household beasts the
idle life corrupts; in wild, the violence of the climate cuts them
off' (Historia Vitae et Mortis). With most birds, fully domestic
mammals, hardy reptiles such as tortoises, and man, however,
maximum records can be taken to represent in some real degree
the extreme length of time for which the species, or its hardier
genotypes, can remain self-maintaining if protected from gross
disease or accident. Theories of senescence must fit these data,
or at least not contradict them, to be available as working
hypotheses.

2-2 Maximum Longevities in Animals

Apart from the observations collected by Bacon, which were
remarkably critical and accurate compared with the wildness of
later estimates, the accurate study of animal life-spans virtually

45

The Biology of Senescence

begins with the enormously painstaking studies of Chalmers
Mitchell (1911) and Flower (1925, 1931, 1935, 1936, 1937,
1938 x) in purging a vast body of legendary and anecdotal
material which encumbered the subject. Much of this legendary
material unfortunately persists in other books and papers
{Tabulae Biologicae — Heilbrunn, 1943; Nagornyi, 1948; Ham-
mond and Marshall, 1952; Schmidt, 1952; Wurmbach, 1951)
deriving their data from Korschelt (1922). The scepticism of
Flower's papers was very valuable, in view of the exorbitant
claims made for parrots, elephants and so on, but it seems
probable that birds, in particular, are in fact capable of living
considerably longer than Flower's maximum figures suggest.
Better data may, in time, become available, though the value
of such records is still not sufficiently widely appreciated and
many opportunities must have been lost through failure to keep
track of individual specimens. No recent writer has dealt equally
painstakingly with the longevity of invertebrates.

The data on vertebrate senescence which follow are those of
Flower, except where otherwise stated. Some more recent re-
cords have been added, including a number derived from the
series of longevity studies published by the Penrose Laboratory
of the Philadelphia Zoo in the years prior to 1942 (Duetz, 1938,
1939, 1940, 1942).

2-2-1 MAMMALS

The longest-lived species is man. Elephas indicus is known to
reach 60 years: a few individuals may reach or exceed 70 in cap-
tivity (77? — Mohr, 1951). The only other mammals which are
known to approach or exceed 50 years are the horse, hippopota-
mus (49 years 6 months: 1953 — Ann.Rep.N.Y. zooL Soc, 53, 12),
Rhinoceros unicornis (49 years — Flower, 1931) and probably the
ass (47 years? — Flower MS.). Many larger mammals, including
baboons and other large primates, cats, bears, African elephant,
equines, tapirs, can approach or exceed 30 years (Chimpanzee,

1 Those references marked 'Flower MS.' refer to the card-index of data
and letters from biological workers which Flower was preparing against a
revision of his first mammalian list, and which was uncompleted at his
death. This index is in the library of the Zoological Society of London, and
includes also bird and reptilian records and the skeleton of a list of inverte-
brate longevities.

46

The Distribution of Senescence

39, Tomilin, 1936; Baboon, Papio papio, 27— Duetz, 1938;

P. anubis 30 -\ Krohn, in press; Gibbon, Hylobates lar, 32 -\

Duetz, 1938; domestic cat, 31— Mellen, 1939, 1940; 27, Com-
fort, 1955: Chapman's zebra, 40 — Weber, 1942: Echidna, nearly
40 years — Duetz, 1942). A large group, including almost all
ruminants, many medium-sized herbivores and carnivores,
large bats and the larger rodents (beaver, capybara, the domes-
tic rabbit) have recorded maximum ages between 12 and 20
years (Golden agouti, 15 — Duetz, 1938). The maximum ages
of very many rodents and small carnivores are not accurately
established, since few specimens have been kept, but it is likely
that a very large group among these forms has a potential life-
span approaching ten years. The small Chiroptera certainly
have a much longer life than most mammals of comparable size
— ringed horseshoe bats have been recovered after at least
7 years (Bourliere, 1947). This agrees with their slow rate of
reproduction.

The shortest-lived mammalian group ( < 5 years) includes
rats, mice, voles, and other small rodents, and the small
insectivores. (Rat — 4 years 8 months in a white rat probably
already 1 year old — Donaldson, 1924; laboratory mouse, 3
years 3 months — Kobozieff, 1931; Micromys minutus, nearly
4 years — Pitt, 1945; golden hamster (Cricetus auratus), usually
2-3 years maximum — Bruce and Hindle, 1934; Deansley, 1938;
one specimen in London Zoo, 3 years 11 months — Flower MS;
guinea pig, 7 years 7 months — Rogers, 1950; Blarina, 18 months
— Pearson, 1945 ;Sorexfumeus, 13-14 months — Hamilton, 1940.)
The real life-span of whales has never been established, but it
is almost certainly not more than 30-50 years at the most, and
probably less. The age of maturity of whales has been placed as
low as 2 years (John, 1937). Ruud et al. (1950) found that blue
whales reach sexual maturity in about 5 years — no individual
in their very large sample was apparently older than 12 years,
judged by the baleen pattern. The life-span of dolphins in the
wild appears to be of the same order (15 years — Sleptzov, 1940;
30+, one specimen — Parker, 1933).

Detailed records of many other mammalian species are given
by Flower.

Recent data on the longevity of seals were reviewed by Laws

47

The Biology of Senescence

(1953), upon the basis of tooth sections. Captive records include
Otaria byronia, 23 years, Eumetopias stelleri, 19 (Flower, 1931);
£alophus calif ornianus, 23; Arctocephalus pusillus, 20 (Bourliere,
1951); Phoca vitulina, 19 (Sivertsen, 1941); Halichoerus grypus,
41-2 (Matheson, 1950). In the wild, Callorhinus ursinus has
reached 21 + years (Schaffer, 1950); Mirounga leonina $ 20; ? 18
(Laws, 1953).

The maximum age records of horses and domestic pets are of import-
ance because these animals are the only mammals kept through-
out life in sufficient numbers to give any estimate of the extreme
age for the species. In spite of the likelihood of exaggeration and
mistake, records of domestic pets kept singly, throughout life,
by intelligent witnesses, provide evidence as good as that from
laboratory stocks and sometimes better than that from zoos,
since reliable mnemonic evidence is better than unreliable
documents.

The known range of maximum ages for some domestic species
is given below. The figures in brackets represent unauthentic-
ated claims within the range of possibility, based as a rule on
evidence which can be neither assessed nor dismissed.

Horses certainly exceed 40, may perhaps exceed 50 years —
most higher claims refer to ponies. Smyth (1937) reported a
46-year-old brood mare which foaled for the 34th time at 42
— this case appears authentic. (Horse, 62 — Flower, 1931; jennet,
reputed 60 — Wright, 1936; pony, 54 — Rothschild/^ Flower,
1931; Shetland pony, 58 — The Times, 3/5/44; roan pony, 52 —
The Times, 12/4/44; Iceland pony, 47— The Times, 7/8/34;
many records between 40 and 45.) A zebra has reached 40 in
captivity (Weber, 1942). Asses — probably exceed 40 (47 —
Flower MS. from a press report; but an 86-year-old ass in The
Times, 29/11/37, can hardly be taken seriously). A 48-year-old
mule is reported (Galea, 1936).

Domestic goats, 15 and probably up to 20 years (20 years
9 months — female wild goat, London Zoo).

Carnivora: cats are the longest lived of the small domestic
mammals. Mellen (1940) from questionnaires sent out in
Canada and the U.S.A. obtained these records: gelt males, 21,
21, 22, 23, 24, 24, 25, 28, 31 years; entire males, 23, 24, 26;
females, 21, 21, 22, 31. These were owners' estimates, but at

48

The Distribution of Senescence

least one 31 -year record was well supported. 33 years has been
claimed (Mellen, 1940). Figures for cats in England in recent
years included at least ten apparently authentic cases over 20,
and one gelt male alive at 28 (Comfort, 1955).

Dogs very seldom exceed 18 years, and only exceptionally
reach twenty. In many breeds the limit is far lower. There are
remarkably few claims of greater longevity in the literature
(34 years, Lankester, 1870).

Rodents: the rabbit can almost certainly exceed 15 years.
(10 years 3 months in the laboratory — Tegge 1936; buck 13
years — Barrett-Hamilton 1911; buck, chinchilla x Belgian hare,
11 years two cases; English buck 14 years, both authenticated
— Comfort 1955.) Flower MS. contains a plausible correspond-
ence with the owner of a rabbit (doe) which was said to have
exceeded 18 years and was still alive.

2-2-2 BIRDS

Flower's longest 'incontestable' record in captivity (Flower,
1925, 1938) was 68 years in Bubo bubo. This is probably too low.
Records exceeding 70 years in parrots, swans, and several large
predators given by Gurney (1899), though less fully proven, are
probably substantially correct.

The maximum life-span in birds is not proportional to size.
It is materially longer than in mammals of comparable size and
activity. Many species can live 30-40 years, including small and
active birds such as pigeons (Flower 1938, Fitzinger 1853:
Streptopelia risoria 40 years, Columba livia 30 years, Goura cristata
(J 49, ? 53 years), while even the smaller passerines have a
potential life of 10-15 or more years in captivity (29 years in a
chaffinch — Moltoni, 1947) and ages of this order are occasion-
ally reached even in the wild state (Perry, R. 1953). It has been
properly remarked that

A robin redbreast in a cage
Lives to a tremendous age.

Extensive aviary records are given by Chalmers Mitchell (1911).

49

The Biology of Senescence

TABLE I

MAXIMUM RECORDED LONGEVITIES IN 45 SPECIES OF BIRD

(Flower, 1938)

Eagle owl (Bubo bubo)

Greater sulphur-crested Cockatoo (Cacatua galerita)

Bateleur Eagle (Terathopsius ecaudatus)

Vasa Parrot (Coracopsis vasa)

Condor (Vultur gryphus)

White Pelican (Pelicanus onocrotalus)

Grey Parrot (Psittacus erythacus)

Golden-naped Parrot (Amazona auropalliata)

Australian Crane (Megalornis rubicunda)

Golden Eagle (Aquila chrysaetos)

Adalbert's Eagle (Aquila adalberti)

Blue-and-yellow Macaw (Ara ararauna)

Grey Crane (Megalornis grus)

Leadbeater's Cockatoo (Cacatua leadbeateri)

Caracara (Polyborus tharus)

Chilean Eagle (Geranoaetus melanoleucus)

White-tailed Eagle (Haliaetus albicillus)

Sarus Crane (Megalornis antigone)

Rough-billed Pelican (Pelicanus erythrorhynchos)

Manchurian Crane (Megalornis japonensis)

Asiatic White Crane (M. leucogeranus)

Herring gull (Larus argentatus)

Banksian Cockatoo (Calyptorrhynchus banksii)

Bare-eyed Cockatoo (Cacatua gymnopis)

Western slender-billed Cockatoo (Licmetis pastinator)

Tawny Eagle (Aquila rapax)

King Vulture (Sarcorhamphus papa)

Ceylon Fish Owl (Ketupa zeylonensis)

Cinereous Vulture (Aegypius monachus)

Red-and-blue Macaw (Ara macao)

Griffon Vulture (Gyps fulvus)

American Crane (Megalornis americana)

Californian Condor (Pseudogryphus californianus)

Shoebill (Balaeniceps rex)

Domestic goose (Anser anser domesticus)

Slender-billed Cockatoo (Licmetis tenuirostris)

Canadian Goose (Branta canadensis)

Orange-winged Parrot (Amazona amazonica)

Roseate Cockatoo (Cacatua roseicapilla)

Domestic Pigeon (Columba livia domestica)

Domestic Dove (Streptopelia risoria)

Emu (Dromiceius novae-hollandiae)

Ostrich (Struthio camelus)

Egyptian Vulture (Neophron percnopterus)

Crowned Pigeon (Goura cristata)

50

Tears

Proven

Reported

68

56

69, 80, 120

55

54

52

51

49

73

49

47

46

80

44

43

43

42

60

42

42

42

42

41

41

41

41

44,49

40

40

40

40

40

39

39

38

64

38

117

38

37

36

35

80

34

85

33

47

30

71

30

47

30

35

30

42

28

40

27

40

23

101

16

49,53

The Distribution of Senescence

2-2-3 REPTILES

The longevity of tortoises is one of the few popular beliefs
about animal life-span which is correct, though it has been
exaggerated. There is no clear evidence that the larger species

TABLE II

MAXIMUM RECORDED LONGEVITIES OF CHELONIANS

(Data from Flower, 1937, except where otherwise stated)

Tears

Testudo sumeiri Marion's Tortoise 1 52 +

elephantopus Galapagos Tortoise 100 +

graeca
daudini
hermanni
radiata
gigantea
sulcata
marginata
Terrapene Carolina

Emys orbicularis

Macroclemmys
temminckii

Clemmys guttata
Pelusios derbianus
subniger

Greek Tortoise 102, 105

Daudin's Tortoise 1 00 +

Hermann's Tortoise 90 +

Radiated Tortoise 85 +

Giant Tortoise 68-180

Spurred Tortoise 42

Margined Tortoise 28
Carolina Box-
tortoise

European Pond-
tortoise

Snapping Turtle

Speckled Terrapin
Derby's Terrapin

Sternotherus odoratus Stinkpot Terrapin

Kinosternon subrubrum Pennsylvania
Terrapin

Chelodina longicollis Longnecked
Terrapin

Caretta caretta Loggerhead Turtle

Malaclemmys centrata Diamond-backed

Terrapin
Cuora trifasciata Three-banded

Terrapin
Geoclemmys reevesi

123 +

118+ Dittmars, 1934

129* Oliver 1953

88+* Deck, 1927

65* Edney and Allen, 1951

70-120 Rollinat, 1934

58 + , 47 Conant and Hud-
son, 1949
42 +
41 +
29 + Conant and Hudson,

1949
53 + Conant and Hudson,
1949

38 +

37 +

31+ Conant and Hudson,

1949
33

?40 Hildebrand, 1932

26 +

24+ Conant and Hudson,
1949
* Marked individual recovered in the wild.
51

The Biology of Senescence

are potentially very much longer-lived than some small forms.
The maximum authenticated records include Testudo sumeirii,
152+ (years); T. elephantopus 100 + ; T. graeca, 102, 105;
T. daudini, 100 + ; T. hermanni, 90 + ; Emys orbicularis , 70-120
(Flower, 1925, 1937; Rollinat, 1934; Korschelt, 1931); Terra-
pene Carolina, 118+ (Ditmars, 1934), 88+ (Deck, 1926), 64
(Edney and Allen, 1951), the last two in the wild. The age of
the royal tortoise of Tonga, said to have belonged to Capt.
Cook, and still living, is unsupported by documents, but may
well be authentic.

The longevity of turtles and luths (Parker, 1926, 1929) and
of crocodiles has been assumed, upon a basis of recorded sizes,
to be very great, though the longest captive record of a crocodile
is 56+ years (Flower, 1937). Alligator sinensis has been kept
52 years (Lederer, 1941), and A. mississippiensis 41 years in the
London Zoo (1912-53). The records of snakes are limited by
their poor survival in zoos. (Eunectes murinus, 29 years (Flower,
1937), 28 (Perkins, 1948); Epicrates cenchris 27 (Perkins, 1948).)
Lizards: Anguis fragilis, 33 years (Hvass, 1938), 32 (Flower,
1937), 27 (Thummel, 1938); Sphenodon punctatus, 28+ (Flower,
1937); Heloderma suspectum, 20 (Conant and Hudson, 1949);
Ophisaurus apodus, 11 years 7 months (Conant and Hudson,
1949), 24 years (Perkins, 1948). The maximum life-span appears
to be relatively brief in chamaeleons, but this may simply be due
to failure to thrive in captivity.

2-2-4 AMPHIBIANS

Here again the figures in relation to size and growth give no
very clear evidence that the life-span is sharply determined.
The maximum records are in Megalobatrachus (52+ years —
Flower, 1936; 65+ years — Schneider, 1932) but many small
species are capable of very long life {Triton spp., 35 years —
Smith, 1951; Triturus pyrrhogaster, 25 — Walterstorff, 1928; Am-
phiuma punctatum, 25 — Koch, 1952; Triton marmoratus, 24, 21 —
Wendt, 1934; Pleurodeles waltl, 20— Noble, 1931). Siren, 25 years,
Amphiuma, 26— Noble, 1931; Salamandra salamandra, 24, Bufo
bufo, 36, Hyla coerulea, 16, Rana catesbiana, 15, Xenopus laevis, 15
(Flower 1925; 1936), Rana temporaria 12+ years (Wilson 1950),

52

The Distribution of Senescence

R. esculenta 14 + , 16+, R. temporaria 9+ (Sebesta 1935), Lepto-
dactylus pentadactylus 15 years 9 months (Gonant and Hudson,
1949).

2-2-5 FISH

Seriously acceptable records of longevity in the larger fish are
very few. The longest accepted by Flower are Silurus glanis,
60+ years, Anguilla anguilla, 55, A. chrisypa, 50 (Flower 1935).
Some of the more celebrated legends of fish longevity (up to
170, 200, 300, or 400 years in carp, and 250 years in pike) are
revived by Backmann (1938) and by Wurmbach (1951). 'Wenn
auch diese Angaben hier und da iibertreiben sein sollten', re-
marks Wurmbach, 'so kann doch gar kein Zweifel daran herr-
schen, dass der Karpfen wirklich ausserordentlich alt wird, und
das Alter des Menschen weitaus iibertrifft' — this is quite pos-
sibly true. Many exaggerated estimates have been based upon
size, as extrapolations of the normal mean growth rate for the
species — upon this basis, a 720 kg. sturgeon should be about
200 years old, and occasional examples weighing 1200 to 1600
kg. would be of fantastic antiquity. In no case, however, are
any of these estimates supported by otolith or comparable
studies, and the extrapolation is almost certainly unjustified. It
is a matter of considerable biological importance to get proper
age determination upon exceptionally large specimens of this
kind.

The life-span of small fish is certainly limited in captivity
(Aphya pellucida, 1 year; Lebistes, 1-2; Xiphophorus, 2-3; Mol-
liensia latipinna, 3-4; Betta pugnax, 1J-2 — Wurmbach, 1951). A
few species of Gobius and Latrunculus must be regarded as
annuals, even in captivity (Bourliere, 1946, Meyers 1952). In
this field there is little new information since the seventeenth
century. 'The life of fishes is more doubtful than that of land
beasts, since, living below the waters, they are less observed.
Dolphins are said to live about thirty years; this is obtained by
experiment upon some of them, the tail being marked by cut-
ting; they grow for ten years. In Caesar's fishponds were certain
Muraenae found to have lived to the sixtieth year. Indeed, they
were grown with long use so familiar, that Crassus the orator
mourned for the death of one. The pike, of freshwater fish, is

53

The Biology of Senescence

found to live the longest, sometimes to the fortieth year. But the
carp, bream, tench, eel and the like are not held to live above
ten years. Salmon grow quickly and live not long, as do also
trout; but the perch grows slowly and lives longer. How long
the breath governs the vast bulk of whales and orcae, we have
no certain knowledge; neither for seals, nor for innumerable
other fish' x {Hist. Vitae et Mortis) . Most of these figures are
reasonably congruent with Flower's list.

2-2-6 INVERTEBRATES

Previous lists of invertebrate longevities {Tabulae Biologicae;
Heilbrunn, 1943; Nagornyi, 1948, etc.), apart from the excellent
data collected by Weismann (1891), almost all spring directly
from the opinions of Korschelt (1922). These are based on data
from the older literature, largely unsupported by exact refer-
ences, some accurate, but others highly speculative. The type
of evidence which has got into such lists is well exemplified by
the 15-20 year life-span of the crayfish. This, though probably
correct, appears to owe its origin to an aside by T. H. Huxley
(1880) to the effect that 'it seems probable that the life of these
animals may be prolonged to as much as fifteen or twenty
years' {The Crayfish, p. 32). The large Tridacna may in fact be
the longest-lived invertebrate, in view of high records of age in
much smaller pelecypods, but the literature contains no in-
formation of any description about its life-span, and the rela-
tionship between great size and great age is perpetually being
disproved in other animals. Of a supposedly 18-year-old Helix
pomatia Korschelt writes elsewhere: 'Gewiss hat diese Angabe
von vornherein wenig Wahrscheinlichkeit fur sich, aber als

1 'Piscium vita magis incerta est, quam terrestrium, quum sub aquis
degentes minus observantur. . . . Delphini traduntur vivere annos circa
triginta; capta experimento in aliquibus a cauda precisa; grandescunt
autem ad annos decern. Deprehensae sunt aliquando in piscinis Caesarianis
muraenae vixisse ad annum sexagesimum. Certe redditae sunt longo usu
tarn familiares, ut Crassus orator unam ex illis defleverit. Lucius, ex
piscibus aquae dulcis, longissime vivere reperitur; ad annum quandoque
quadragesimum ... at carpio, abramis, tinea, anguilla et huiusmodi non
putantur vivere ultra annos decern. Salmones cito grandescunt, brevi
vivunt, quod etiam faciunt trutae; at perca tarde crescit, et vivit diutius.
Vasta ilia moles balaenarum et orcarum, quamdiu spiritu regatur, nil
certi habemus; neque etiam de phocis. . . . et aliis piscibus innumeris.'

54

The Distribution of Senescence

unmoglich wird man dieser Langlebigkeit nach dem, was man
von anderen Tieren weiss, nicht bezeichnen diirfen' 1 (1922,
p. 36, footnote).

A proper survey of the longevity of invertebrates can hardly
yet be undertaken — the information is mostly lacking. It seemed
wisest in compiling Table III, which includes a few of the
longest and most interesting invertebrate records, to give not
only the record and source, but the type of evidence upon
which the record is based. In invertebrates which metamor-
phose, length of larval life often depends entirely upon environ-
ment and food, while in other forms adult life can be punctuated
by very long spells of diapause. Figures for these forms should
therefore when possible indicate the circumstances of life.
Larval life-spans have in general been omitted from Table III.
The most reliable records are in all cases those of animals kept,
like Labitte's (1916) beetles or the Edinburgh sea anemones
(Ashworth and Annandale, 1904) under close observation in
captivity. Evidence from growth rings requires very careful
scrutiny. Some purely inferential evidence, as of the age of
termite primaries, is probably reliable. There are also some
surprisingly high records in the wild, especially for pelecypods,
where the method of ageing by rings of growth has been well
upheld by other evidence. The life-span of common inverte-
brates certainly remains a wide-open field for those with
facilities and an unlimited capacity for taking pains, and one
where any reliably-attested information is worth putting on
record.

1 'No doubt these findings have little probability in themselves, but one
cannot dismiss such longevity records as impossible, in view of what is
known of other animals.'

55

The Biology of Senescence

TABLE III

MAXIMUM RECORDED LONGEVITIES OF VARIOUS INVERTEBRATE
SPECIES BY PHYLA

(Ages in years unless specified)
c = information based on specimens in culture or captivity
w = information based on specimens in the wild
g = age estimated by examination of growth rings
h = case history (in parasitic forms)

Porifera

Suberites carnosus

15

c

Arndt, 1941

Adocia alba

9

c

Arndt, 1941

Coelenterata

Actinia mesembryanthemum

65-70

Dalyell, 1848

c

: Korschelt, 1922

Cereus pedunculatus

85-90

c

: Ashworth and

Annandale, 1904
Stephenson, 1935
Warwick, 1954

(personal commn.)

Platyhelminths

Schistosoma haematobium

25

h

Kirkland, 1928

28

h

Christopherson, 1924

Clonorchis sinensis

25

h

Moore, 1924

Gastrodiscus aegyptiacus

9

h

Christopherson, 1924

Taeniorrhynchus saginatus

>35

h

Penfold, Penfold and
Phillips, 1936

Diphyllobothrium latum

29

h

Riley, 1919

'Echinococcus cysts'

56

h

Lawson, 1939*

Dugesia tigrina (= Planaria

maculata)

6-7

c

Goldsmith, 1942

Dendrocoelum lacteum

5

?

Bresslau, 1928-33

Nematoda

Loa loa

15

h

Coutelen, 1935

Wuchereria bancrofti

17

h

Knabe, 1932

Necator americanus

12

h

Sandground, 1936

Rotifera — see Table IV

Annelida

Eisenia foetida
Lumbricus terrestris
Allolobophora longa

3-4§

5-6
5-10

c

\ Rabes, 1901
[Korschelt, 1914

Sabella pavonina

>10

c

Wilson D. P., 1949

Arthropoda

(Arachnida)

'Tarantula' (aviculariid)$

11-20

c

Baerg, 1945

Avicularia avicularia $

>7

c

Didlake, 1937

Tegenaria derhami $

7

c

Savory, 1927

Filistata insidiatrix $

10,11

c

Bonnet, 1935

* See also Coutelen et al., 1950, Davaine, 1877, Wardle and McLeod,
pp. 116-17.

56

1952,

The Distribution of Senescence

Arthropoda

Physocyclus simoni

4

c

Bonnet, 1935

Teutana grossa $

6

c

Bacelar and Frade, 1933

Psalmopoeus cambridgii

5*

c

London Zoo, Flower

MS.

Lasiodera curtior

4i

c

London Zoo, Flower
MS.

(Crustacea)

Astacus

15-25

inference Friedel, 1880

Homarus

50

inference Herrick, 1898,
1911

Leander serratus $

5-6

inference Solland, 1916

Oniscus asellus

4i

O]

Philoscia muscorum

4

c

Porcellio scaber

dilatatus

3|

c
c

> Collinge, 1944

Platyarthrus hoffmansegg

i 5+

c

Armadillium vulgare

4+

cJ

Balanus balanoides

>5

w

Moore, 1934

(Insecta)

Thysanura

Ctenolepisma longicaudata total 7

w

Lindsay, 1940

Ephemeroptera

Cloeon dipterum

imago 4 wks

c

Vane, 1946

Isoptera

Neotermes castaneus $ $

imago >25 yrs

w

Snyder, fide
Howard, 1939

Nasutitermes-physogastric 5 20-40

w

v. Hagen, 1938

'Termite primaries'

60-?

w

Richards, 1953

Lepidoptera

Nymphalis antiopa

imago 12 wks

cl

Calliophrys rubi

imago 6 wks

c

>■ Frohawk, 1935

Maniola jurtina

imago 44 days

cj

Coleoptera*

Blaps gigas

imago > 10 yrs

c1

Timarcha sp.

imago > 5

c

► Labitte, 1916

Carabus auratus

imago 3-1

c)

Dytiscus marginalis

imago < 3

c

Blunck, 1924

Prionotheca coronata

imago 6, 7+

cl

London Zoo, Flower

Akis bacarozzo

imago> 4

cj MS.

Cybister laterimarginalis

imago 5 J

c

Sharp, 1883

Hymenoptera

Apis mellifica $

imago > 5

c

Pflugfelder, 1948

Lasius niger $

imago >19

c

Goetsch, 1940

Stenamma westwoodi $

imago 16-18

c

Donisthorpe, 1936

Formica fusca ?

imago 10+

c

Janet, 1904

?

imago 15+

c

Lubbock /<& Weis-
mann, 1882

sanguinea $

imago 5+

c

Lubbock fide Weis-
mann, 1882

Lasius niger $

imago\ ?
imagoj

c

Lubbock fide Weis-

Formica fusca $

mann, 1882

♦For a discussion of the longevity of beetle larvae.
1935 (Cossus) Linsley 1938 (Stromatium) .

57

see Howard 1939; also Latter

The Biology of Senescence

Echinodermata

Echinus esculentus

>8

w

Moore, 1935

Psammechinus miliaris

>6

c

Bull, 1938

Asterias rubens

(reaches sexual maturity)

5-6

c

Bull, 1934

Marthasterias glacialis

>7

c

Wilson, 1954 (personal

comm.)

Ophiothrix fragilis

>5

c

Zool. Gart., 1930

Mollusca

Patella vulgata

15

wg

Fischer- Piette, 1939

Acmaea dorsuosa

15

wg

Abe, 1932

Gibbula umbilicalis

5

w

Pelseneer, 1934

Trochus niloticus

12

wg

Rao, 1937

Viviparus contectus

5 ?

c}

■ Oldham, 1931

viviparus

9

c

Geyer, 1909

Hydrobia ulvae

>5

c

Quick, 1924

Pila sp.

5

c

Flower, 1922

Haminea hydatis

4

w

Berrill, 1931

Planorbis corneus

6

c

Oldham, 1930

Limnaea luteola

3

w

Seshaiya, 1927

Physa gyrina

12-13 months
14-15 months

c}

- DeWitt, 1954

Abida secale

<3

c

fide Flower MS.

Helix pomatia

6-7

c

Kiinkel, 1916

aspersa

5-6

c

Welch, 1901

(Cepaea) nemoralis

7

c

Brockmeier, 1896

hortensis

9

c

Lang, 1896

nem X hort hybrid

10

c

Cuenot, 1911

(Arianta) arbustorum

5

c

Kiinkel, 1916

'Helix spiriplana'

15

c

Vignal, 1923

Eulota fruticum

5-6*

w c

Kiinkel, 1928

Hyalinia villae

5*

c

van der Horst, 1929

Limax cinereoniger
Geomalacus maculosus

5
. 6*

c}

Oldham, 1942a, b

Achatina zebra

6i

c

Longstaff, 1921

Rumina decollata

12

c

Vignal, 1919

Oxystyla capax

(aestivating)

23

c

Baker, 1934

Ostrea edulis

>12

wg

Orton and Amirtha-
lingam, 1930

Pecten jessoensis

>8

wg

Bazykalova, 1934

Megalonaias gigantea

54,36

wg

Chamberlain, 1933

Quadrula sp.

>30 w, marked

Isely, 1931

20-50

wg

Lefevre and Curtis,
1912

Margaritana margaritifera

> 60 inference

Geyer, 1909 (see also

Israel, 1913, Cuenot,

1911)

Cardium corbis

>16

wg

Weymouth and
Thompson, 1930

Tivela stultorum

20

wg

Weymouth, 1923

Venus mercenaria

>40

wg

Hopkins, 1930

Siliqua patula

14-16

wg

Weymouth, 1931

Mya arenaria

>8

58

wg

Newcombe, 1935, 1936

The Distribution of Senescence

TABLE IV

MAXIMUM RECORDED LONGEVITIES OF ROTIFERS

(Bibliography from Hyman, 1951)

Asplanchna sieboldii

2-3 weeks

c

Tannreuther, 1919

Proales decipiens

12 days

c

Liebers, 1937

sordida

22 days

c

Jennings and Lynch, 1928
Lynch and Smith, 1931

Cupelopagis vorax

40 days

c

Gori, 1925

Euchlanis triquetra

21 days

c

Lehmensick, 1926

Epiphanes senta

8 days

c

Ferris, 1932

Brachionus pala

12-19 days

c

Chu, 1934

Euchlanis dilatata

23 days

c

Liebers, 1937

Keratella aculeata

29 days

c

Kolisko, 1938

Epiphanes brachionus

17 days

c

Kolisko, 1938

Floscularia conifera

18 days

c

Edmondson, 1945

Lecane inermis

14 days

c

Miller, 1931

Philodina roseola

10 days^|

citrina

21 days

megalotrocha

1 7 days >

c

Spemann, 1924

Rotaria macrura

58 days

rotatoria

20-50 daysJ

Callidina sp.

5 months

c

Zelinka, 1891

Adineta vaga

15-22 days >,

barbata

21 days

Habrotrocha constricta
Macrotrachela

34 days

>c

Dobers, 1915

quadricornifera

2 months

Mniobia russeola

30 days J

2-3 Maximum Life-span in Man

Human longevity records are even more notorious than those
of animals. They depend largely on unsupported memory and
tradition in a field where the emotional premiums of exaggera-
tion are high. 'Furthermore, even though this satisfaction and
vanity, of which we have spoken, were absent, yet such is the
peculiar and perpetual wandering of the human Intellect, that
it is more moved and roused by affirmatives than by negatives,
whereas properly it ought to be just to both, nay even, in the
forming of any axiom, the force of the negative instance is the
greater.5 {Novum Organum). King (1911) found in the 1911
census a discrepancy between the size of the 85-90 year-old
age group and that of the next higher group, which was almost
certainly due to exaggeration. William Thorns (1873), founder

59

The Biology of Senescence

of Notes and Queries, and Young (1899) devoted much time to
exposing the pretensions of past supercentenarians. In some of
the 'documented' cases, the life-span of father, son, and grandson
of the same names were apparently conjoined in one record.
Young's greatest authenticated record was a few weeks short
of 1 1 1 years.

The actuarial probability of an individual's exceeding the
age of 150 years, on the life-data of 1939, has been estimated
at (J)50 (Greenwood and Irwin, 1939). Putter (1921) calculated
on a basis of German vital statistics for the years 1871-91 that
ages over 105 were effectively impossible, and that for every
million persons reaching 20 years, the number reaching 109
would be 4-8 x 10-10: 'danach ware es nunmehr wohl an der
Zeit, die Berichte iiber 120, 130, 140, 150 usw.-jahrige dahin zu
verweisen, wohin sie gehoren: ins Reich der FabeP. This
scepticism has proved excessive, especially as regards the popu-
lation-frequency of centenarians (see Freudenberg, 1949). The
existence of supercentenarians cannot be disproved by statistical
means unless the distribution of ages is really continuous, since
ordinary life- tables have no defence against, say, a rare geno-
type with double the normal potential life-span. The number
of persons reaching 100 years is in any event too small for
statistically significant estimates of the rate of increase in the
force of mortality after about 90 years of age. Putter's estimate
was based on the assumption that this increase continued at the
same rate as in earlier life. The relation between observation
and calculation in this part of the life-table is fully discussed
by Greenwood and Irwin (1939).

Subsequent writers have been content to rely on direct
observation, provided that only records supported by proper
documentary evidence are taken seriously (Forster, 1945;
Tomilin, 1938). The minimum requirements are these laid
down by Thorns (1873) — documentary evidence of birth (or
baptism), of death or present age, and of identity. The third of
these, as Pearl (1928) points out, is commonly the key to false
records of extreme age. The best of such evidence, from com-
pulsory birth certification, has been available in England since
1837, and would now be available for records up to 118 years
(1955). By critical standards of comparable severity the greatest

60

The Distribution of Senescence

human age to be authenticated with reasonable certainty has
been said to be 120 years (Fisher, 1923). A considerable num-
ber of cases between 110 and 115 years have also stood up to
examination (e.g. Bowerman, 1939; Backman, 1945; Koren-
chevsky, 1947). The greatest age to be authenticated in Eng-
land and Wales by actual birth certificate, however, is 109
years, while in one case the absence of a birth certificate
indicated an age in excess of 1 1 1 years.

Claims of extreme longevity in particular districts abound.
Metchnikoff investigated statements of this kind in Bulgaria
and the Caucasus. Bazilievitch (1938a, b) led an expedition to
investigate the celebrated longevity of Abkhasians, and ex-
amined several claimants in detail. Two of these were reputed
to be over 130 years old. The evidence (identity papers and
memory of events in the Caucasus during the early nineteenth
century) is given by Bazilievitch in careful detail; much of it is
extremely entertaining, but far from conclusive, although the
subjects were certainly very old men (Bazilievitch, 1938b).1 In
recent years very large numbers of claims to extreme longevity
have been made in Russia (e.g. Rokhlina, 1951; Nagornyi,
1948; Lukyanov, 1952; Nikitin, 1954). Dealing with the figures
in the 1926 census of the U.S.S.R., which showed proportions
of 3-5 and 3-8 centenarians per thousand gross population in
Daghestan and Abkhasia respectively, as to 1-8 per million
among Volga Germans, Tomilin (1938) says 'We must doubt
the factual truth of these figures, since no documentary evidence
of the age of persons who had passed the century mark was
produced.' The analysed distribution of age-groups in the
Abkhasian census shows exactly the same deficiency in the
85-89 and 95-99 year groups, compared with the 90-94 and
100+ groups, which was observed by King (1911) in England.
'Without special documentary evidence of the accuracy of these
age-data, we cannot conclude definitely that the relative num-
ber of persons reaching the age of 100 and over in the general

1 Prof. G. Z. Pitshelaouri, of Tbilisi University, who very kindly showed
me his unpublished data on the longevity of Abkhasians, has found several
subjects whose reputed age exceeds 130 years and is colourably supported
by baptismal registers — one man still living took part in, and accurately
describes, the Crimean war of 1 854-56. I have failed to obtain a paper by
Mishaikov (1929) giving statistics for centenarians in Bulgaria.

61

Men

1940

20 (105)

1941

18 (112)

1942

12 (107)

1943

21 (108)

1044

21 (109)

1045

19 (105)*

1946

22 (105)*

1947

19 (106)*

1948

19 (103)*

1949

27 (104)*

1950

22 (102)*

1951

33 (104)*

1952

24 (105)*

The Biology of Senescence

TABLE V

NUMBER AND MAXIMUM AGES OF CENTENARIANS DYING IN ENGLAND AND

WALES

(Registrar-General's statistics)

Year Number and probable maximum age

Women
102 (108)

91 (108)
79 (108)

92 (106)*
85 (105)*
71 (106)*
94 (108)*
97 (108)*

107 (115)
133 (106)*
131 (107)*
142 (109)*
147 (107)*

* = verifiable by birth certificate.

mass of the population of Abkhasia is really higher than in the
population of Russia' (Tomilin 1938). In America, Nascher's
investigation of John Shell, reputed to be 131, showed him to
be in fact about 100 years old (Nascher 1920). In England and
Wales, the oldest persons dying between 1930 and 1945 appear
to have reached ages of 112 and 109 years (Korenchevsky,
1947). A woman who died at St. Asaph, Flintshire, in 1948
may have reached 115 years, and had certainly reached 111.
Sporadic records of supercentenarians such as Old Parr,
whose tomb in Westminster Abbey credits him with an age of
152 years, whose body was examined by Harvey, and whose
complete lack of documentation was exposed by Thorns
(1873), occur in almost all cultures: a long series of similar
anecdotes is given by Gould and Pyle (1898). The best recent
summary of these often-paraded examples is that of R. T.
Gould (1945). Though in most cases the stories conform closely
to the childhood fantasy of 'going on living for almost always',
they may also indicate that authenticated records do not yet
represent the extreme of human longevity under all conditions.
There is some ground, apart from the absence of critical record
in backward countries, to associate extreme individual longevity

62

The Distribution of Senescence

with a low rather than a very high standard of living throughout
life (Gumbel, 1938), an argument which fortunately has not so
far been advanced to justify starvation as a social policy.
Extreme records in man, occurring in excess of statistical prob-
ability, are chiefly of interest in suggesting that after a certain
age the rate of increase in the force of mortality is not main-
tained, either by reason of selection or from other causes.

24 Distribution of Senescence in Vertebrates

Actuarial senescence is known, or reasonably assumed, to
occur in all mammals, provided they live long enough. It is less
easily recognized, but apparently equally universal, in birds.
There are apparently no satisfactory life-table studies of birds
under domestic conditions apart from a single paper on fowls
(Gardner and Hurst, 1933), but individuals kept as pets cer-
tainly become increasingly enfeebled after an age which is
fairly constant for the species, and the reproductive senescence
of poultry, marked by a steep decline in egg production, is well-
known to farmers (Clark, 1940; Brody, 1945; Fig. 14). This

180

ISO

120

cL
5 90

1X1

a.

60

30

YEAR OF EGG LAYING

10

123456789

Fig. 14. — The decline in egg production in successive years of laying
(drawn from the data of Clark and of Hall and Marble) .

63

The Biology of Senescence

decline can be reduced by mild hypothyroidism (Turner and
Kempster, 1948) The pair of crowned pigeons which lived,
according to Fitzinger (1853) for over 40 years, mated and laid
throughout life, but hatched no offspring after the age of 18 or
20 years (Flower, 1938). Spermatogenesis likewise appears to
decline (Payne, 1952). The life-span of birds is longer in pro-
portion to size and metabolic rate than that of mammals, and
the scatter of age in senescence as shown by aviary records
appears, superficially at least, to be rather greater within a
species.

We are familiar with the ageing of warm-blooded animals
because we keep them. It is among the 'cold-blooded' verte-
brates that the real uncertainty begins. We keep fish, but only
the smaller forms — we do not, apart from zoological gardens
and occasional pet tortoises, keep reptiles; as for amphibia,
Hilaire Belloc wrote uncontrovertibly concerning lonely people
who keep frogs that

by the way
They are extremely rare.

The general assumption that all vertebrates must necessarily
undergo a senescence at least superficially similar to that of
mammals has prejudiced even those biologists who have kept
frogs for long periods, with the result that very little real in-
formation unbiassed by this assumption has been published.
Very possibly the assumption may prove to be correct, but it
cannot be lightly made. It is evident that some senile change,
in the form of an accumulation of injuries, must occur in all
vertebrates with the passage of time, and be reflected in the
force of mortality. But this effect is certainly small and incon-
stant compared with the 'morphogenetic' senescence which
determines the life-span of mammals. It is this morphogenetic
component which we are concerned to detect and estimate in
lower vertebrates. Unfortunately for such a study, the life of
many of these creatures, whether it ends in senescence or not,
is, as we have seen, long enough to make ordinary short-term
laboratory observation useless.

Bidder's opinions on the relation between perpetual youth
and continuing growth have already been quoted (p. 12).

64

The Distribution of Senescence

Three types of growth-pattern are theoretically possible in
vertebrates — growth to a maximum size, ceasing when this is
reached: growth toward a limiting size which is approached
asymptotically: and growth without a limiting size. In the third
of these cases, the specific growth-acceleration can be negative
— i.e. the growth rate continually declines — but it could
theoretically do so in such a way that, given a sufficiently
long life, any final size could be reached. These last two modes
of growth correspond to convergent and divergent series. Thus
in the series

(1) 1 +£+i+J...lim2,
and the series

(2) 1 +*+*+*+*...,

the increment at each term decreases (the specific growth rate
falls), but whereas in (1) the series tends to a limiting size
(specific size), in (2) it does not, and can be indefinitely con-
tinued so that any sum is ultimately attained. The terms 'in-
determinate growth' and 'indeterminate size' have been differ-
ently used by different writers. D'Arcy Thompson wrote, 'It is
the rule in fishes and other cold-blooded vertebrates that
growth is asymptotic and size indeterminate' (1942). If the
growth of an animal is in fact asymptotic, its size is limited by
the sum of the asymptotic series. 'Indeterminate' growth with-
out limit, but with a decline in the specific growth-rate, strictly
follows the pattern of the divergent series. For this reason it
would be desirable, but it is not empirically possible, given real
biological material, to distinguish between 'asymptotic' and
'indeterminate' growth. In both cases the rate of growth
declines with advancing age; but in the second case the potential
size is unlimited.

Distinctions of this kind, however, are based upon the fitting
of equations to points derived by averaging observations upon
populations of animals, and in spite of the real value of such
biometric applications, in the study of growth-curves they have
tended to lose contact with the real behaviour of observable
material. It is possible in practice to distinguish only between
species, or particular populations of a given species, which

65

43 Weeks

Fig. 15 (a). — Growth in length (mm.) of male fish of the genera Xipho-

phorusy Lebistes and Heterandria during the first year of life. Sexual maturity is

indicated by 6* (from Wellensieck, 1953).

44 48 weeks

Fig. 15 (b). — Growth in length of female fish of the genera Xiphophorusy

Lebistes and Heterandria during the first year of life. Sexual maturity is

indicated by '$ (Wellensieck, 1953).

The Distribution of Senescence

continue throughout life to increase tangibly in size, given suit-
able conditions, and forms where the maximum size is reached
relatively early in life, is fixed for the species, and does not
increase further with increasing age even under the most
favourable conditions. The chief obstacle to wide generalization
about the determinacy or indeterminacy of growth in lower
vertebrates, and in other forms such as pelecypods, lies in the
fact that arrest of growth at an apparent specific size can be
brought about by environmental conditions. In some cases
growth can be resumed after such an arrest — in others, appar-
ently, it cannot. Even less is known of the effect of these pheno-
mena on the life-span than is known of the normal ageing of
such forms. Differences are also substantial within each of the
main groups of poikilo thermic vertebrates. In many reptiles and
small fish, continued growth after a relatively early age is no
more evident than in the male rat. In amphibia, 'many species,
particularly some tropical forms, seem to have an absolute
size, which the males soon attain, but this does not hold for
many salamanders, nor for some Northern frogs' (Noble, 1931).
In many cases the male has an absolute size and the female
has not. If enough data were available, the variety of growth-
patterns is more than sufficient to test Bidder's hypothesis —
unfortunately, corresponding data upon age/mortality rela-
tions are almost entirely lacking.

The idea of a 'self-maintaining' vertebrate is not impossible
ex hypothesis It is in fact what we should expect if growth-
cessation is an equilibrium process, if there is no important pro-
cess of differential growth at work, and if there is no qualitative
change in the regenerative power of cells throughout adult life.
It is not self-evident, though it might be true, that an animal
should be obliged to increase in size in order to retain the
power of carrying out running replacements. It seems reason-
able for our purposes to regard an animal of 'indeterminate'
growth as one in which the probability of nursing an individual
to the point at which increase in somatic size has ceased is
infinitely small, and an animal of 'indeterminate' life-span as
one in which the survival rate under favourable conditions is
substantially independent of age, however long a population
of that animal is observed from birth.
f 67

The Biology of Senescence

In all the groups whose susceptibility to senescence is doubt-
ful, there is wide variation in life-cycle and growth-pattern,
which is very probably reflected in differences of their capacity
for age changes. Some aquarium species offish certainly 'age'
as judged by their declining reproductive powers: in the larger
sea fish this has not been clearly demonstrated. The contra-
dictory views offish senescence given to Flower (1935) by two
acknowledged authorities, one on aquarium and the other on
marine ichthyology, and based on small and large teleosts
respectively, are possibly both correct. In other forms there is
an obvious sex difference in growth-maintenance, in longevity,
or in both. The specific age might also be indefinite in animals
which nevertheless became more liable to die, as individuals,
with increasing age. It was implied by Ricker (1945) that fish
might senesce individually, i.e. undergo a waning of vitality and
resistance with age, but that there is no sharp specific age — the
life-span of each individual would be limited by senescence, but
the senile process would reach its critical point at a much more
variable age than in mammals: as if the menopause in human
beings were to occur with approximately equal probability in
any year after the menarche. Such senescence would be real,
but could not readily be detected actuarially.

Very nearly all these problems require abundant new data
to settle them finally. The general evidence of the distribution
of vertebrate senescence which will be given here is both frag-
mentary and equivocal. It does, however, contain some facts
which suggest that Bidder's hypothesis is too simple and that
the manner of growth-cessation, rather than the fact of it, is the
main determinant of the mammalian pattern of senescence.

2-4-1 FISH

The 'indeterminate' growth of fish, on which Bidder based
his hypothesis, has often been discussed (Hecht, 1916; Keys,
1928; Huxley, 1932; Vaznetzov, 1934; Thompson, 1942; Wel-
lensieck, 1953). Many large species of teleosts can continue to
grow throughout life, and the rate of decline of their growth-
rate is considerably slower than e.g. in most reptiles. The locus
classicus of continued growth without evidence of senescence,
actuarial or reproductive, is the female plaice. Here the evi-

68

The Distribution of Senescence

dence supports Bidder in that growth in the male plaice ceases
relatively early, and there is evidence that it has a shorter life-
span than the female (Wimpenny, 1953). On the other hand,
in many small teleosts reproductive senescence is known to
occur, and both the sexes appear to exhibit specific age, in spite
of the fact that growth in the female may continue throughout
life. The reproductive failure of many teleosts with increasing
age is familiar to aquarists. So is the tendency of particular
species to have a limiting age, although there have previously
been no published life-tables for any teleost in captivity by
which this impression could be confirmed. The growth of
some small teleosts has been studied (Felin, 1951; Wellensieck,
1953).

There is a good deal of evidence that small teleosts, and
perhaps teleost species generally, undergo both reproductive
and actuarial senescence comparable to that of mammals. The
most dogmatic assertions on this score are those of pathological
anatomists. On the basis of concretions occurring in the testis
of a single teleost species (Astyanax americanus), Rasquin and
Hafter (1951) hold that the 'appearance of senility changes
shows that the teleosts conform to the common vertebrate pat-
tern of ageing despite a widespread misconception to the con-
trary'. The decline of fertility in some aquarium species pro-
vides more solid evidence in support of this view. Many species
of fish are in any case exposed to a specialized series of fluctua-
tions in mortality associated with reproduction — the difficulties
of treating these fluctuations as a form of senescence in those
species which always die after breeding, such as the male of
Callionymus (Chang, 1951) and the lamprey, are indicated by
observations upon other fish in which there are a limited num-
ber of survivors from each breeding season, and these thereafter
acquire a new lease of life. It is doubtful if any cyclical or
potentially cyclical change in mortality can properly be called
senile. In Callionymus lyra in the wild, the male appears to live
5 and the female up to 7 years. The males disappear, probably
through death, but possibly by migration to deeper water, after
breeding once. Females may first breed in their third, fourth,
or fifth year of life, depending on their rate of growth, and
probably breed more than once (Chang, 1951). In such a case,

69

The Biology of Senescence

the late-developing females would very probably have a longer
total life-cycle.

Studies of wild populations are almost always conducted
under conditions where the standing force of mortality through-
out life is very high, and they therefore give little information
about mortality trends in the latter part of the life-cycle of the
longer-lived forms. Excluding the very high larval mortality,
populations of many species of fish, studied in the wild, show
an age structure and a pattern of death similar to that found
in birds, i.e. a high constant mortality unrelated to age and a
virtually constant expectation of life (Frost and Smyly, 1952
(Fig. 16); Deevey, 1947). Substantial differences in life-span may
be dictated by availability of food organisms of a size suited to
adult feeding, and by competition between the fry of the
observed species and adults of other species. Some populations
of minnows show apparent specific age which is exceeded in
other populations of a closely-related species by a very large
factor (Frost, 1943; Tack, 1940). In Pimephales promelas, Markus
(1934) observed apparent specific size and specific age in all
but a few exceptionally large individuals. This was apparently
due to the fact that there was an overall mortality of 80 per cent
following spawning; the survivors, and individuals which took
no part in breeding during their first year of maturity, con-
tinued growth until the next breeding season. If reproduction
is avoided, life may be prolonged — Bidder (1932) points out
that eels, which, it is believed, normally die after spawning, live
many years in captivity (Flower, 1925). Frost and Smyly (1952)
found considerable differences in growth rates and in the form
of the growth curve between brown trout inhabiting tarns and
those inhabiting Windermere (Fig. 16). The age structure of
the tarn population agreed well with a steady annual survival
rate of 35 per cent between the second and eighth year of life.
In these fish growth had become very slow, whereas in the
Windermere population individual fish were still growing at
7 years upon an approximately linear scale. The ability to con-
tinue growth may depend on attaining a size which makes it
possible to prey on smaller fish. Long-lived fish such as pike
certainly continue to grow measurably for very long periods
(Schloemer, 1936) but the increase in size is associated with an

70

The Distribution of Senescence

increase in the size of the prey taken (Frost, 1 954). Ricker (1945)
comments that 'senile death is an everyday occurrence' in popu-
lation-studies of the Indiana sun fish. This conclusion is, how-

Fig. 16. — Growth of trout in Windermere and the small tarns (Frost and

Smyly, 1952)

ever, based upon the failure of known sources of death (disease
and predation) to account for the disappearance of fish. The
overall mortality rates actually found in marking experiments
were 56 per cent for small and 58 per cent for older specimens.

71

The Biology of Senescence

But in many unfished populations of other species there is a
steady increase in mortality with increasing age and size
(Ricker, 1948).

A great deal of important information upon fish growth was
collected by Schmalhausen (1928) from the data of a number
of Russian workers (e.g. Tereschenko, 1917). In the sturgeon,
growth continues actively throughout at least 30 years of life,
with little decline in rate at sexual maturity (about 15 years).
In the bream, on the other hand, the growth constant shows
a more regular and progressive decline. These fish were found
to mature at about 3 years, and degenerative changes in the

10

0-5

*1,

oTTj « o

& MLgj^'nulu n0 *-*

I I I I I I I I 1 I I I I I I I 1 I I I 1 I 1 I I 1 I I I I

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Age in Years

Fig. 17 (a). — Growth-constant for growth in length of the sturgeon,

Acipenser stellatus, at various ages. Broken line — females: single points —

males: thick line, mean value for males; dashes, mean value for females.

Kx - 0-67, K2 = 0-58 (from Schmalhausen, 1928).

gonad were usually evident from the sixth year on — two definite
stadia could be observed in the growth curve, one following
puberty, and the other following this gonadal senescence, the
growth coefficient settling down to a steady value thereafter
without further decline up to 13 years of age (Fig. 17#, b, c). This
rather closely resembles the pattern reported in the goldfish.

In Xiphophorus and Lebistes the male exhibits sharp specific
size, but the female may continue to grow measurably through-
out life, the pattern of growth differing little from that of the
plaice (Wellensieck, 1953). Yet in these forms previous experi-
ence suggests that there is no striking difference between the
survivals of the two sexes in captivity (Bellamy, 1934). In
Heterandria both sexes reach a virtual limiting size (Wellensieck,

72

The Distribution of Senescence

1953), Fig. 15fl, b. In the goldfish, according to exhibition
breeders, fertility reaches a maximum under aquarium condi-
tions in the third year of life, declining thereafter, and almost
all fish are sterile by the seventh year. Breeding at 10 years is
recorded (Hervey and Hems, 1948). When the reproductive

3000

2500

2000 -

1500

1000

500 -

10

20

11
22

12
21

13 A b ram is drama
26Acipenser stef/afus

12 3 15 6 7 8
2 1 6 8 10 12 11 16

Age in Years

Fig. 17 (b). — Growth in weight of the sturgeon (Acipenser stellatus) and the

bream (Abramis brama). Scale for A. stellatus 1 Russian lb = 100. Scale for

A. brama 1 g. = 1 (after Schmalhausen, 1928).

life is over, however, the fish may improve greatly in condition,
and appear much less sensitive to environmental damage than
before. In exhibition fish the life-span appears to be about
1 7 years, though much older examples are known. The extreme

73

The Biology of Senescence

record of longevity appears to be between 30 and 40 years.
Rate of growth is extremely variable. One specimen, kept in
a six-gallon tank, reached a length of only 4 inches in 25 years
(Hervey and Hems, 1948).

In none of these cases is it clear how large a part of the
potential life-cycle is actually covered by the observed growth
curve. In most fish the rate of growth does in fact decline with
age, though in many the effective reproductive life appears to

300

2-00 -

POO

7 2 3V
Age in YeGrs

10 11 12 13

Fig. 17 (c). — Growth in weight of the Bream (Abramis brama) — annual

increments. The mean growth-coefficients at various stadia (youth,

maturity, postreproductive life) are indicated by transverse lines Klf 2j 3

(from Schmalhausen, 1928).

have ceased long before this decline has produced an almost
stationary body-size. The reproductive decline, moreover,
does not appear to involve any decrease in vigour, and may
actually imply the reverse, in view of the hazards which repro-
duction involves for many fish.

In the small teleosts, it ought to be possible to answer most
of these questions by direct experiment. For this purpose the
guppy (Lebistes) is proving a particularly suitable experimental
animal, both because of the ease with which it can be reared
and handled for purposes of measurement, and because of the

74

The Distribution of Senescence

neatness with which its growth can be controlled by varying
the food intake and living space.

A large population of Lebistes has been kept at University
College under conditions in which each individual was isolated
throughout life, and the dates of birth and death were accur-
ately known. Dr. H. Spurway has kindly allowed me to use her
data on this population; from these, and from fish which I have
kept myself, it has been possible to construct a number of life-
tables for Lebistes under different conditions. These are con-
sistent, and all show a progressive increase in mortality with

100

DAYS

100

500

1000

Fig. 18a. — Survival of 12 male and 40 female guppies (Lebistes
reticulatus) kept individually at 25° C: sexes combined. (From
data provided by Dr. H. Spurway.)

advancing age very similar to that found in mammalian life-
tables (Fig. 18fl). The fish in this sample were not measured
during life, but had certainly reached or exceeded the usual
'specific size'. The growth rate of female Lebistes and the size
at which the growth curve reaches a plateau can very easily be
altered. Spurway has found that these fish could be kept in
individual half-pint milk bottles — by combining restricted space
with restricted diet, I have kept female Lebistes at a length of
about 2 cm. for as long as 600 days. In this state they are

75

The Biology of Senescence

reproductively mature (unlike the rats subjected to retardation
by McCay — p. 149) and capable of resuming growth. There are
'specific sizes' characteristic of each size of container and each
level of nutrition — or, alternatively, of each population-density
in a tank, when a fish is promoted from one such container to
a larger, or when fish are removed from a tank population, a
new plateau is rapidly reached. The curve given by Wellensieck
represents only one such equilibrium, The growth capacity also
appears to decline somewhat throughout life, and there is a

A = half pint milk bottle, restricted food

D = 21b jar, restricted food

O = 7 lb jar, plentiful food

• = large aquarium tank, plentiful food

AGE (days)

100

200

300

400

500

600

Fig. 18£. — Growth of female guppies (Lebistes
reticulatus) in different conditions of feeding and
living-space. The symbols indicate the time of
transference to new conditions. All the fish
were alive at the time of drawing. The growth
pattern of the fish whose life-table is given in
Fig. 1 8a approximated to that of the fastest
group shown in this figure.

practical limit, as might be expected, to the size of guppy which
can be produced at maximum food intake and maximum living
space. The combination of variables in Lebistes, and the fact
that the life-span of the non-growing males is not, upon present
data, grossly different from that of the growing females, suggest
that a great deal about growth and senescence in fish can be

76

The Distribution of Senescence

learned by the collection of actuarial data for guppies subjected
to different programmes of growth: this work is in hand, but it
is in the nature of the subject that life-table making cannot be
hurried, and life-tables under various conditions of growth were
not ready for inclusion here. The growth curves shown in
Fig. 18b are instructive, however, when we compare them with
the closely similar growth-behaviour and pseudospecific size of
wild trout.

It seems probable that there is as much variation in 'sen-
escence' as in growth-patterns among teleosts. Some forms
apparently resemble monocarpic plants, mortality being linked
to reproduction. Some, in captivity, have a life-span determined
by senescence, their mortality increasing with age on a curve
closely similar to that of mammals. Some forms, however, may
conceivably have an effectively indeterminate life-span, and are
not at present known to undergo any form of senescence, repro-
ductive or general, other than the accumulation of injuries,
though this may well mean only that their 'determinate' maxi-
mum, as in wild birds, comes so late in relation to mortality
as never to be reached in practice. The effect of variation in the
growth-rate upon the development of senescence has yet to be
determined.

2-4-2 REPTILES

There are no published reptilian life-tables, but a number of
careful studies of reptilian growth have been made (Sergeev,
1937; Townsend, 1931, 1937; Cagle, 1946). By collating these
with maximum age records, a good deal of significant informa-
tion can be obtained. Sergeev found that while, in all reptiles,
early growth depends on environmental conditions, being some-
times very rapid, and growth rate declines with increasing age,
there are a number of forms where both sexes have an effective
specific size which is reached early in life, and after the attain-
ment of which no further growth occurs. The cessation of
growth in these forms is apparently as definitive as that in
mammals, and its timing does not appear to depend on the
arrival of sexual maturity. There appears to be no close correla-
tion between either of these two patterns of growth and the
length of the life-span.

77

The Biology of Senescence

Among chelonians, both patterns of growth are known to
occur. Continuous growth as a decreasing rate appears to be
general in tortoises, the large species having inherently higher
growth rates throughout. Townsend (1931, 1937) found that
early growth in 100 specimens of the large T. vicina, kept in
captivity, was potentially very rapid, and continued after the
age of sexual maturity (about 20 years of age). Flower (1945)
observed continuing growth in a 39-year old specimen of
T. graeca. The age of sexual maturity in the male Terrapene
Carolina appears to lie between 12 and 15 years (Nichols, 1939).
All these are known to be long-lived forms. On the other hand,
the majority of terrapins exhibit specific size. In Emys Sergeev

18

c

-C

I

YEARS

Fig. 19.— Growth of Emys (Sergeev, 1937).

(1937) found that growth-cessation by the fifteenth year of life
was as complete as in the adult mammal (Fig. 19) although E.
orbicularis, like T. graeca, is apparently capable of living 70-120
years and probably of breeding throughout life (Flower, 1937).
Rollinat, however, on whose observations Flower's records were
based, considered that growth in this form might continue for
30-40 years (Rollinat, 1934). Hildebrand (1932) studied the
longevity and growth of over 1,000 specimens of Malaclemmys
centrata in captivity — an investigation which is the nearest pub-
lished approach to a chelonian life-table, but which was unfor-
tunately continued in detail for only 10 years. He found the age
of maturity much more variable than in mammals, some indi-
viduals being full-grown in 8-9 years, others requiring 12-15.
The oldest specimens in captivity were 21 years old, and

78

The Distribution of Senescence

'showed every appearance of being young animals', but other
wild specimens taken when full-grown had been kept for 20 years
without decline of vitality or reproductive power. Hildebrand
placed the maximum life-span for this species at 40 years or
more, but evidence from other small terrapins suggests that this
may be a considerable underestimate. Although the only aquatic
species which is known to have reached an age comparable with
that of the land tortoises is Emys orbicularis, it would be very
difficult to argue upon the existing evidence that specific size
and determinate age are correlated in chelonians. Contrary to
Bidder's hypothesis, specific size here seems to be an adaptation
to carnivorous life in small pools, while continuing growth is
found in land tortoises and marine turtles (Parker, 1926, 1929).
Crocodiles have also been credited with indeterminate growth
— 'Crocodili perhibentur esse admodum vivaces, atque grande-
scendi periodem itidem habere insignem; adeo ut hos solos ex
animalibus perpetuo, dum vivunt, grandescere opinio fit. . . .
At de aliquo testaceo genere, nihil certi, quod ad vitam ipsorum
attinet, reperimus' * (Hist. Vitae et Mortis) . Claims of longevity
are based on the exceptional size of some specimens. Large
alligators have been observed in captivity to remain for 25 years
in a non-growing state, e.g. Alligator sinensis (Dathe, 1935),
though the difficulties of accurate length-measurement are
evident.

2-5 Distribution of Senescence in Invertebrates

Among invertebrates not only is there a demonstrable variety,
greater than in vertebrates, in the nature of the preponderant
senile process, but we have also the full range, from indeter-
minacy to very sharply defined determinacy of life-span. The
gaps in our knowledge of life-cycles are so large that we cannot
yet picture the distribution of senescence in invertebrate phy-
logeny: papers entitled 'The life-history of . . .' only very
exceptionally include reference to the senescence of the species

1 'Crocodiles are held to be very lively, and to have a notable span of
growth — so that they alone of beasts, so opinion runs, grow so long as they
live. . . . But of any hard-skinned beast, as pertaining to their length of life,
we find nothing certain.'

79

The Biology of Senescence

under study — an extraordinary deficiency, which is a measure
of the equally extraordinary lack of interest in age processes.
It is fairly evident, however, that the distribution both of sen-
escence in general and of any one process of senescence, such
as depletion or mechanical deterioration, is quite discontinuous
in phylogeny. This evolutionary discontinuity is what we should
expect if 'exhaustion of programme' is the common basis of
adverse age changes.

Senescence in some shape or form probably occurs in every
group where the power of regeneration or fissile reproduction
is less than total, or where body-cells are not continuously and
'indeterminately' replaced. Some forms which 'degrow' under
adverse conditions appear to be capable, in all probability, of
unlimited alternate growth and degrowth, at least in the
laboratory, while in a few, such as actinians, the adult can
remain indefinitely in statu quo, though with a changing popula-
tion of cells. Senescence is most striking in forms such as rotifers
where determinacy of cell number is very highly-developed and
the power of regeneration is usually negligible. There do not
appear to be any invertebrate cells (except possibly pelecypod
neurones, of whose longevity and renewability we know little)
which are called upon to remain for 100 or more years in active
function, like a human neurone, or for still longer, like the
neurones of the tortoise. The distribution of senescence in
invertebrates suggests that in spite of the general argument
against the selection of long-lived forms, relatively great longev-
ity is sometimes an evolved adaptation, and that if some cold-
blooded vertebrates are in fact immune to senile change, that,
too, is likewise a specialized mechanism and not a primitive or
an 'inherent' mechanism which has been lost with increasing
somatic complexity.

2-5-1 PORIFERA

Bidder infelicitously cited 'the sea anemone, the bath sponge
and the water-vole' as three organisms insusceptible to sen-
escence. The only serious study of senescence in Porifera appears
to be that of Arndt (1928) who concludes that it does not occur,
although some sponges are fatally disrupted by their own larvae.
Aquarium specimens have an effectively limited life, as in so

80

The Distribution of Senescence

many other groups, but sponges seem ideally able to conform
to Bidder's expectations of them.

2-5-2 COELENTERATES

In hydromedusae, Child (1918) observed a progressive
decrease in metabolism and pulsation rate with increasing size,
which he regarded as evidence of senescence. His work on the
processes of ageing and rejuvenation in hydroids (1915) depends
on the criterion of resistance to cyanide as evidence of 'physio-
logical age' — one which is hardly acceptable in this context.
Child's results with Pennaria, using this test of age, were in any
case less consistent than those he obtained with planarians,
where cyanide resistance rose steadily throughout life (Child,
1915).

Evidence that the life-span of sea anemones is 'indeterminate*
is probably stronger than for any other metazoan group.
Dalyell's (1848) celebrated specimens of Actinia lived for 70
years in captivity without any sign of deterioration. An even
more famous batch of sea anemones were collected 'some years
prior to 1862', and were first identified as Sagartia troglodytes
by Ash worth and Annandale (1904), later by Stephenson as
Cereus pedunculatus (1935). They remained in the aquarium of
Edinburgh University Department of Zoology until 1940 or
1942, when they were all simultaneously found dead. Budding
continued freely throughout life, and the animals underwent
no obvious change during eighty to ninety years of continu-
ous observation (Warwick, 1954, personal communication).
Whether gametogenesis likewise continued throughout life is
not known.

Hydra, The long-standing controversy over the senescence of
Hydra illustrates some of the difficulties of placing a geronto-
logical interpretation on life-tables and histological appear-
ances. Hydra was a favourite organism, earlier in the century, in
the argument over the 'potentielle Unsterblichkeit' of metazoa.
Differences in culture conditions almost certainly account for
the very irregular results obtained.

Early workers (Hertwig, 1906; Boecker, 1914; Berninger,
1910) on this question found it impossible to keep Hydra for long
periods without the onset of 'depression', evidenced by cloudy

81

The Biology of Senescence

swelling and cytolysis. With better cultural methods Goetsch
(1922, 1925) kept individuals of Pelmatohydra oligactis, Hydra
attenuata, and Chlorohydra viridissima alive for 27 months. Goetsch
considered that like the actinians Hydra was capable of remain-
ing indefinitely in statu quo. Gross (1925) working with P. oligactis
failed to keep any individual alive for more than 349 days,
'senescence' being evidenced by irregular and hypertrophic
budding or by the animal becoming smaller and smaller in the
presence of abundant food. 'Senile' changes in Gross's material
began after the fourth month of life. A life- table, drawn from
Hase's (1909) data by Pearl and Miner (1935), extending over
only 148 days, indicates some increase in mortality with age,
but is closer to the log-linear than to the rectangular contour
(see Fig. 6c, p. 20). Hartlaub (1916) had already described
experiments on Syncorinae in which he concluded that the
power of producing gametes was lost relatively early in life,
while that of budding persisted.

David (1925) kept isolation records in cultures of P. oligactis
and satisfied himself that in this form the individual animals
tended to die between 20 and 28 months in approximate order
of individual age — an important observation which has not been
repeated. According to Schlottke, however (Schlottke, 1930),
the material in David's histological sections was heavily parasit-
ized. Schlottke's own observations suggested that all the tissues
of Hydra are continuously replaced throughout life, from a sub-
jacent reserve of interstitial cells. This view is supported by the
work of Brien (1953), who showed by marking experiments that
there is continuous growth in Hydra from before backward, the
marked zone travelling down the animal and being ultimately
rejected at the base: a case, in other words, of 'indeterminate
growth' coexisting with a final specific size.

In colonial hydroids, however, it seems to have been shown
beyond reasonable doubt that the life-span of each hydranth is
physiologically determinate. The resorption and involution of
hydranths was described in full by Huxley and de Beer (1923);
the hydranth shrinks, the gut becomes filled with cellular
debris, and the degenerating material is returned to the colony
by the contraction of the hydranth itself. In Obelia and Cam-
panularia Growell (1953) has now shown that regression takes

82

The Distribution of Senescence

place strictly in order of age, each hydranth having a life of
4 days at 21° G. and 7 at 17° G. When regression is accelerated
by starvation or adverse culture conditions, the age order is
still preserved.

2-5-3 SUNDRY INVERTEBRATES

Existing observations are scattered rather thinly over a num-
ber of groups. Child (1911, 1913, 1914, 1915, 1918) carried out
exhaustive studies upon the regeneration of planarians, and
upon their capacity for de-differentiation, to which subsequent
research has been able to add little or nothing. Here again, as
in Pennaria, he employed the increase in resistance to dilute
cyanide solutions as a criterion of senescence, on the assump-
tion that this change reflected a decrease in metabolic rate.
While susceptibility decreased as a function of age in the grow-
ing animal, planarians kept for several months at a constant
size showed no such increase, and planarians undergoing shrink-
age under adverse food conditions showed a decrease in sus-
ceptibility. Child also demonstrated the 'rejuvenation', partial
or entire, of regenerating fragments of planarians. This further
observation, using the same criterion of resistance to toxicity
(1915), that a gradient of 'rejuvenation' exists in Stenostomum
(Rhabdocoela) during the production of new zooids has been
confirmed by Sonneborn (1930) using direct life- table studies.
Sonneborn's experiments showed that the regenerative effects
of fission were markedly unequal in the two halves, since the
head portions, which required only to regenerate tails, under-
went typical senescence, and died after a limited number of
divisions, while tails, which required to regenerate most of the
body and nervous system, could be propagated indefinitely.

In Aeolosoma (Gligochaeta), Haemmerling (1924) found that
the anterior end of the body appeared to undergo eventual
senescence, new worms being produced from the posterior end.
Stole (1902) had already given a circumstantial histopatho-
logical account of 'senile' death in Aeolosoma as a whole, but the
appearances observed might have resulted from almost any
environmental cause. In JVais (Annelida), Stolte (1924, 1927)
found extensive histological changes with age, with disappear-
ance of the normal zones, degeneration of the visceral ganglia,
G 83

The Biology of Senescence

and the cessation of reserve-cell production from the embryonic
tissue persisting in the posterior end. The significance of these
changes is again obscure, and no attempt was made to deter-
mine actuarially the mortality rates at different ages. Rhab-
docoelians have (Bresslau, 1928-33) been observed to be in-
creasingly susceptible to protozoan parasites the longer they live.

How far the capacity for 'degrowth', which is found in
planarians, is evidence of a potentially indeterminate life-span
is not evident, but it seems likely that forms such as Lineus
(Nemertinea), which revert on starvation over a period of years
to a mass of cells resembling an embryo (DawidofT, 1924), might
be maintained indefinitely in alternate growth and degrowth
until the patience of the investigator was exhausted.

The evidence in fissile worms at present suggests that non-
senescence depends upon fairly active replacement of cells, and
that any organ which fails to take part in the regenerative pro-
cess is liable to undergo senile change. Harms (1949) considered
that the senescence of Serpulids was due primarily to changes
in the nervous system, and rejuvenated old specimens oiProtula
by grafting young heads. Some further work on this subject as
careful as that of Child and Sonneborn would probably be well
worth undertaking.

Morphogenetic loss throughout life of the power of regenera-
tion in a nematode of determinate cell-number was actually
demonstrated by Pai (1928) in Anguillula aceti. Amputation of
the tail with nuclear removal kills the animal at any age. In
young individuals, provided the nucleus is left intact, wound
healing and cytoplasmic regeneration can take place. In mature
animals there is wound closure but no cytoplasmic regenera-
tion, while in senile animals amputation is fatal. In Anguillula
senescence follows a pattern very similar to that of rotifers
(see below) and occurs at about 44 days. The degenerative
cellular changes in ovaries, gut and nerve cells have been des-
cribed: these appear in the two or three days preceding death
Pai, 1928; Burger, 1954).

2-5-4 ROTIFERS

The ageing of rotifers is one of the most spectacular examples
of endogenous senescence in animals. It is also one of the most

84

The Distribution of Senescence

thoroughly-studied, at least from the descriptive point of view.
The life-span varies in different species from a few days to
several months, and each species tends to exhibit very sharp
specific age. After a period of growth, which takes place by
increase in cell size, the nuclear number being fixed, and adult
vigour, rotifers enter a period of senescence, with conspicuous
loss of activity, degeneration of cells, deposition of pigment, and
ultimate death in extension. In some forms the senescent phase
is genuinely post-reproductive, but in the majority it occurs
while egg-laying is still occurring at a diminished rate, and may
be accompanied by the production of malformed eggs, or eggs
of varying size.

The external appearances of rotifer senescence have been
vividly described in several forms (Callidina, Plate 1886; Pleur-
otrocha, Metchnikoff, 1907; Proales, Noyes, 1922; Jennings and
Lynch, 1928; Hydatina, Plate, 1886; Lecane, Szabo, 1935; Miller,
1931; Rotifer vulgaris, Spemann, 1924). The animal becomes
sluggish in behaviour and reaction to jarring, the tissues and
cuticle shrink and become opaque or granular in appearance,
swimming is replaced by creeping, pigment accumulates in the
gut, digestive gland and mastax. The movements of the pharyn-
geal cilia are the last signs of life to persist.

It seems clear that this is an endogenous process of degenera-
tion. A number of attempts have been made to correlate it with
other features of rotifer organization. Plate (1886) considered
that senescence in Hydatina occurred typically when the activity
of the ovary, and the supply of germ cells, failed. This is not the
case in all rotifers, however. In Lecane inermis Miller (1931)
found that the mictic females cease egg laying early in life and
have a relatively prolonged post-reproductive period, while
amictic females show signs of age before the last egg is produced,
and all are dead within two days thereafter; the life-span of
males is even shorter (Fig. 20). In this species, fertilization of
the mictic females does not appear to influence longevity.
Miller attributes the difference in life-span between mictic and
amictic females directly to the difference in fertility, but this is
not fully borne out by her life-tables, the chief difference being
in the longer post-reproductive period of the mictic females. In
Hydatina senta it is the amictic females which are the longer

85

The Biology of Senescence

lived (Ferris, 1932). In both these species, however, the form
having the higher reproductive rate in early life dies younger,
an observation which supports Miller's suggestion that death
results from 'exhaustion'. Egg laying in Apsilus vorax continues
until death (Cori, 1925) and in Proales, appears itself to be
adversely affected by somatic senescence, the egg-substance
failing to enter the eggshell, and eggs of bizarre size and shape
and of low hatchability being produced (Jennings and Lynch,
1928). Old populations of P. sordida consist of two types of

». — Survival curves for males, mictic (M) and amictic (A) females of
Lecane inermis (from Miller, 1931).

senile individuals, some thick and opaque, and others abnorm-
ally transparent, with pigmentation of the gastric glands. There
is considerable individual variation in the length of survival
once senescence is established.

Impairment of function in all the species which have been
studied is so general during senescence that it is not possible to
identify a pace-maker organ in the process, though in some
cases it appears to be the digestive system which first deterior-
ates. The pattern is fully consistent with some or all of the
somatic cells having a fixed survival-time under normal meta-
bolic conditions — a highly important precedent for the study
of other types of metazoan senescence. The senescent change

86

The Distribution of Senescence

depends directly upon metabolism — encysted rotifers can sur-
vive for very long periods (59 years — Rahm, 1923) and display
enhanced reproductive performance on emergence from dia-
pause (Dobers, 1915).

It is particularly interesting that this dramatic senesence in
rotifers accompanies a very strict determinacy of cell number,
a lack of regenerative capacity, and in most species a very

180

Of 2 3 4 5 6 7

Fig. 21. — Growth in length of Philodina citrina (Lansing, 1948).

limited power of repair. Nuclear division after hatching has not
been described in any rotifer. In many forms wound healing is
confined to young animals — older animals die after amputation
(Pai, 1934) but in young Asplanchna brightwelli (Pai, 1934) and
Stephanoceros (Jurszyk, 1926, 1927; Ubisch, 1926) the cytoplasm
of the coronal lobes can be regenerated, as can parts of the
coronal funnel in Cupelopagis (Huhnerhoff, 1931; references
from Hyman, 1951).

87

The Biology of Senescence

The somatic growth of rotifers has been studied by several
workers. Rotifer vulgaris shows little or no change in size through-
out life (Spemann, 1924). The growth curve of Apsilus vorax is
a parabola, with shrinkage before death (Cori, 1925) while in
Philodina citrina growth ceases by the sixth day (Lansing, 1948).
(See Fig. 21.) Lansing also made the striking observation that

1 1~\ r\

-

IOO '

Vs

90«

80-

70-

I

60-

\

50-

\

. PARC NT

^\

.6F,

40-

1

30-

ry x./ v

\

,

20-

I * * '•••4

\

lO-

/• N%-

\

rn

\ *

^ A

30

20 §

10

10

15

20

DAYS

25 30 35

Fig. 22. — Life-span and egglaying of Philodina citrina (Rotifera) over 6
generations in normal culture (from Lansing, 1952).

if these rotifers are propagated in each generation from eggs
laid at or after the fifth day of maternal life, the rate of develop-
ment becomes progressively greater and greater from genera-
tion to generation, and the longevity of the individual less and
less, so that clones propagated in each generation from old
mothers invariable become extinct (Figs. 22, 23, 24). Jennings

88

The Distribution of Senescence

and Lynch (1928) had already noted that the offspring of very
old rotifers are less viable than those of vigorous adults. Lan-
sing's results suggested that the effects of maternal age are
cumulative from generation to generation: they were also
reversible, the eggs laid by young members of such a clone
being capable of giving rise to normally long-lived individuals.
Lansing also found that clones propagated in each generation
from the eggs of very young mothers showed an increase in

35 40

DAYS
Fig. 23. — Progressive decline in life-span of a strain of Philodina citrina
(Rotifera) raised in each generation from eggs laid by old mothers (from

Lansing, 1952).

longevity over the control stock. In Euchlanis triquetra, the
'young' orthoclone could not be maintained, however, because
within a few generations it gave rise largely to male-producing
eggs. Lansing regards his ageing-factor as a product of growth-
cessation, since it appears in the individual animal at the point
where the negative specific acceleration of growth is greatest.

The susceptibility of rotifer eggs to external influences affect-
ing the life-cycle of the progeny has been much studied in forms

89

The Biology of Senescence

which give rise periodically to mictic generations. The literature
on sex-determination in rotifers is reviewed by Hyman (1951).
The longevity of the two types of female differs considerably;
'somewhere in the ontogeny of the females, it must be deter-
mined which kind of egg they are destined to lay. The deter-
mination occurs during the maturation of the egg from which

CO

>
<

Z
<

CL
CO

LU

Z

<

LU

35

30

25

20

15

10

\
\ \

\ \

\ \

%

%
i

«
i

%
«
•
%
%
%
i
i i

4 DAY ORTHOCLONE

II ■

17 "

P Fl 2 3 4 5

ORTHOCLONE GENERATION
Fig. 24. — Life-span of successive generations of Philodina reared in each
generation from the eggs of 4- 11- and 17- day-old mothers (Lansing, 1948).

the female comes, that is, during the last few hours before the
egg is laid' (Hyman, 1951). Both internal and environmental
factors of great complexity appear to operate in different forms.
'The conclusion from numerous researches seems to be that, in
addition to an inherent rhythm as regards male production,
monotony of conditions suppresses mictic females, whereas any

90

The Distribution of Senescence

sudden change, especially of diet and of physiochemical com-
position of the water, induces the appearance of mictic females'
(Hyman, 1951). The reason for the difference in longevity
between Lansing's old and young orthoclones is not, it should
be noted, entirely comparable to that between mictic and
amictic females. It does not seem to represent a difference in
specific age due to a shortening of that part of the survival-
curves which, owing to the low early mortality in rotifers, is
usually horizontal, but rather a 'breaking away' of this plateau
by the introduction of a higher and higher early mortality, the
curve becoming less and less rectangular and more and more
oblique. If conclusions are to be drawn upon the effect of
maternal age upon senescence, this difference is important.

It is possible that the uniform specific age of rotifer popula-
tions is due to depletion. Reproductive exhaustion has already
been discussed. It is also known that the limited regeneration
observed in Stephanoceros takes place at the expense of reproduc-
tion and of somatic growth. Little is known of the metabolic
capacity of rotifers — they apparently store glycogen, but may
be incapable of assimilating carbohydrate (Hyman, 1951). Sud-
den senescence might well represent the exhaustion of a meta-
bolic substrate, or of a non-renewable system. Accumulation has
also been suggested: pigment certainly does accumulate, prob-
ably secondarily to the ageing process. Lansing (1942) demon-
strated the accumulation of calcium in old rotifers, and suc-
ceeded in prolonging their life by immersion in dilute citrate
solution — it is not clear how often this process can be repeated.
A more curious factor influencing the life-span was observed by
Edmondson (1945) in Floscularia conifera, where individuals
growing in aggregation reach twice the length, twice the age,
and a higher level of fertility than solitary specimens.

The peculiarities of rotifer organization are so numerous that
some, if not all, of the mechanisms controlling their longevity
are likely to be peculiar to the group. On the other hand, their
short life-span makes them a suitable object for study, and they
provide an unequivocal example of senescence coupled with
cellular non-renewal which calls for further investigation.

91

The Biology of Senescence

2-5-5 ARTHROPODS

Senescence in arthropods is widespread and probably uni-
versal. Those forms which have wings, jaws, bristies, and other
chitinous tegumentary structures not renewed by moulting are
particularly liable to genuinely 'mechanical' senescence. In the
forms which moult as adults, the time of ecdysis is a particu-
larly arduous one, judged by the mortality, and many of these,
such as Daphnia and large spiders, appear very often to die in
the attempt to carry out a final moult.

' Physiological5 senescence, in the sense in which nineteenth-
century biology used the term, also appears in a convincing
form for the first time in arthropods, since, as MetchnikofT first
pointed out (1907, 1915), a non-feeding imago must be regarded
as expendable from the evolutionary point of view. The evolu-
tion of short sexual life as a modification in some groups is
balanced by the evolution of a very long sexual life in specialized
individuals of other, social, species, as part of the adaptive
development of a group-existence — the longest life-span being
reached in one or both sexual forms among true ants and
termites.

There is no known case of arthropod indeterminacy com-
parable with that of actinians. Growth in most insect imagines
is more or less rigidly limited, although the capacity for con-
tinued cell division persists in varying degrees. According to
Harms (1949), somatic mitosis in many arthropod imagines is
virtually confined to the mid-gut. Mitotic capacity has not1 been
shown to bear any relationship to longevity, except perhaps in
forms producing queens, where the relation of continued repro-
duction to long life might be either a direct example of cause
and effect, or the result of two parallel adaptations. Some
solitary arthropods are capable of very long life (20 years in
tarantulas, Baerg 1945, possibly 50 years in lobsters, Herrick,
1896).

Crustacea. The small Crustacea (Cladocerans, Copepods, Iso-
pods) generally show very sharp specific age. In Daphnia1 this

1 There is a striking lack of unanimity in the literature over the 'normal'
life-span of various species of Daphnia, even when grown in apparently
similar media. Fritsch (1953) has shown that this variation depends to some
extent upon the amount of available pantothenic acid. Where Daphnids are

92

The Distribution of Senescence

appears to be definable in terms of ins tars, D. longispina living
for 19-22 instars, the duration of which depend upon the con-
ditions of culture (Ingle, Wood and Banta, 1937) and D. magna
for 17 instars (Anderson and Jenkins, 1942). Detailed studies
upon factors which retard or accelerate the rate of development
and life-span in Daphnia have been carried out. (Mc Arthur and
Baillie, 1926 seq.; Ingle, Wood and Banta, 1937; Anderson and
Jenkins 1942, etc., see pp. 143 seq.). In view of the availability
of life-tables for Daphnia, the pattern of its normal growth is of

Observed growth
Calculated growth
Growth rate 1st cycle
Growth rate 2nd cycle

r -. i_

60 Days 70

Fig. 25 (a).— Growth of Daphnia magna— first type (Edlen, 1938).

particular interest. Edlen (1938) showed that the growth of
normal daphnids takes place in two cycles, the first levelling off
after three or four instars, and the second coinciding with the
development of the gonad. He found three types of pattern in

fed upon living cultures of protozoans or algae, the food organism itself may
metabolize and remove pantothenic acid. It is highly questionable in view
of Fritsch's findings how far the life -tables obtained by workers using
different culture techniques, or even by one worker at different times, are
comparable. This is unfortunate, as Cladocerans are most useful organisms
to gerontologists — further standardization of culture techniques seems
essential if they are to be used in this way, however.

93

The Biology of Senescence

the growth of individual D. magna. In the majority of specimens
(Fig. 25#) the two cycles of growth followed one another, the
growth-potential in the second cycle falling almost to zero with
increasing age; this fall is accompanied by a decrease in tgg
size and number, and the animal finally dies after a short period
in which growth has almost ceased. In individuals of the second

40

30

tu
if 20

LU

z I-

i

o

z

-

111

-J
CD
co

>

CO

h-

Z
tu

CC

LU

CO

Z
<

X

u

Q

O
O

d.

CO

z

CO

a

LU

• • •

Q

LU

X

- fc

CD

5

3
cC

ID

LU

I

CO
LU

>
u
o

• • 1

• observed growth

"

•
• •

o

o

z

CO

O

c?

LU

0

z

•

•
1 L I . 1

1

... 1 1 1

— i — i i .... i. ..

10

20

DAYS

30

Fig. 25 (b). — Growth of D. magna — second type: growth-phases
superimposed (Edlen, 1938).

type (Fig. 25b) the two growth cycles were superimposed — in
these growth was very rapid, there was no prepubertal 'shelf in
the curve of body size, and fertility was lost early, although the
life-span appeared to be normal. The third type (Fig. 25^)
showed only the first cycle of growth, but no gonadal function
developed, and adult size was not attained. These forms died
early. Edlen considered that the developing ovary exerts a

94

The Distribution of Senescence

hormonal control over growth, and possibly over the mainten-
ance of life-processes generally.

The chief senile changes in Daphnia appear to be in the fat
body, intestinal epithelium and musculature (Schulze-Robbecke,
1951). This author found no evidence of the 'cerebral death*
which was once widely accepted as the general cause of inverte-
brate senescence (Harms, 1926; Muhlmann, 1900, 1911), and
which was described by Walter (1922) in Cyclops. Withdrawal

20

CO
LU

h-

Z 10

-

DIED

1

/

\
\

\

\ -^

1 "'■ «

— r-l • . . . "

- l'

^r — x... ._

* \
/ \
• \

■ \

\

>

Observed growth

Calculated growth

Growth rate

/ 5

' '

s

N
X

— 1 II... 1

~~---r---«.-_i

10

20

DAYS 30

Fig. 25 (c). — Growth of D. magna — third type. One growth-cycle only

(Edlen, 1938).

of ovarian activity does not seem to be the direct cause of
senescence in Daphnia. Schulze-Robbecke's oldest specimens
still showed considerable ovarian activity, and continued to lay
eggs, though in reduced numbers, up to the time of death.
Oocytes at all stages of maturation remained in the ovary to the
last. Schulze-Robbecke attributed the death of Daphnia in old
age to failure of nutrition following degenerative changes in
the gut.

Walter's work on Cyclops (1922) dealt with C. viridis, which
has a life-span of about 9 months, 'senile' change in gut
epithelium and in the cerebral ganglia being evident from the

95

The Biology of Senescence

fifth month. Somatic mitosis in adult life occursr only in the
mid-gut of Cyclops, and in this region 'senile' changes were not
found (Harms, 1949). The most striking degenerative changes
in Cyclops were found in the chromatin of the ganglion cells,
with the appearance of large inclusion bodies suspiciously
reminiscent of virus inclusions or fixation artefacts, and in the
antennules. Gut degeneration occurred later in life, about the
eighth month, and was confined to the anterior gut, mitosis con-
tinuing in other parts of the gut epithelium until the end. It is
difficult to know what connection, if any, these changes have
with the process of senescence.

Needham (A. E. Needham, 1950) has studied the growth
rate of limb regeneration in Asellus aquaticus; growth in Crus-
tacea, on Needham's figures for Asellus and Carcinus, is deter-
minate and the curve sigmoid, the arithmetic rate of growth
rising to a maximum and declining asymptotically there-
after to zero. The geometric rate of growth declines monoton-
ically with age from the outset, and the rate of decline itself
declines with increasing age. The specific regeneration rate
decreases progressively with age, owing to the progressive
increase in the duration of each instar; the rate of decline is
much less than that of the normal growth rate, and itself de-
clines with age. 'In some Crustacea the limiting size is attained
at an age beyond the mean expected life-span. Growth is
indeterminate in Crustacea only in this sense. They are not
potentially immortal.' This investigation illustrates once again
the difficulty of characterizing the growth-behaviour of real
organisms mathematically: the decline of arithmetic growth
may be asymptotic, or tangible size increase may continue;
in Crustacea there is the additional difficulty that growth is dis-
continuous, being interrupted by stadia, which superimpose a
'quantal' effect on the smooth ideal curve. The conclusion of
Needham's studies is that the growth of Crustacea follows a
convergent series, and must cease, presumably, for practical
purposes in any form which lives long enough. But the relation
between such cessation and determinacy of life-span is still
entirely conjectural.

Insects. It has long been recognized that several separate
types of senile change may occur in insects. Mechanical damage

96

The Distribution of Senescence

to the cuticle (Blunck, 1924; Wigglesworth, 1945), depletion of
reserves both in feeding and non-feeding imagines, accelerated
in some cases by reproduction (Krumbiegel, 1929a, b; Bilewicz,
1953), accumulation of urates (Metchnikoff, 1915), deterior-
ation of the nervous system (Hodge, 1894-95; Pixell-Good-
rich, 1920; Schmidt, 1923; Weyer, 1931, etc.) and 'general
senile decay' have all been demonstrated by more or less
satisfactory evidence. The vast majority of holometabolous
imagines give every evidence of having a sharp specific age, and
this is a group in which we are unusually well-equipped with life-
tables. The nature of the processes which limits imaginal life
seems, however, to vary widely, but they have the common
property of being processes operating in a cellular system where
little or no renewal, and no further morphogenetic develop-
ment, are occurring.

One of the best general descriptions of insect senescence is
given by Blunck (1924) for Dytiscus marginalis: he describes the
main signs of advancing age as diminution in activity and
deterioration of the epicuticle, with the growth of colonial
protozoa on the dorsal shield, legs and mouth parts, which the
animal cannot any longer clean effectively. The cleaning secre-
tions seem to be reduced, and the chitin appears brittle, whole
legs or antennae being occasionally snapped off in swimming. If
pygidial gland secretion fails, air enters the subalar air chamber
and the beetle drowns. In the beetles dissected by Blunck, the
gonads had almost disappeared during the third year of life, the
fat-body was increased in size, almost filling the body cavity,
but chalky and full of concretions. In some individuals there
was almost complete atrophy of the wing muscles. The extreme
life-span is under 3 years, females living longer than males:
sexual activity usually ceases in the second year but may persist
in individuals into the third. A number of senile processes,
which may not be mutually dependent, can be detected in this
description. The balance between mechanical, depletive and
'morphogenetic' senescence must vary considerably from species
to species, and even from individual to individual. Blunck's
description is of interest in providing not only an account of
such a mixed senescence, but one of the very few instances
where the 'change in inert structures', so popular with colloid

97

The Biologp of Senescence

chemists investigating senescence, really seems to occur — in the
progressive hardening and weakening of the chitin of Dytiscus
elytra, which Blunck found to be a reliable rough measure of
the age of specimens taken in the wild. On the other hand, a
considerable part of this change, as Blunck himself suspected,
may represent failure to secrete the normal lubricant coat — a
cellular rather than a mechanical deterioration.

In many insects, especially lepidoptera, there is evidence that
the fat-body contains a definite reserve of materials, which are
not replaceable during imaginal life. In females of the moth
Ephestia elutella, longevity and fecundity are both functions of
body weight at eclosion (Waloff, Norris and Broadhead, 1947).
Longevity is also greater in virgin females, possibly owing to the
sparing of reserves through egg-rudiment resorption (Norris,
1933, 1934). Exhaustion of the fat-body is characteristically
found in Ephestia which appear to have died of old age. Norris
(1934) found evidence that the fat-body contains two types of
store, one needed for the maintenance of the ovaries and the
other for the maintenance of life. The second appears to be
supplemented by feeding the imago, but not the first (Norris,
1933). Similar deterioration of the fat-body has been described
as a sign of senescence in Carabus and Drosophila (Krumbiegel,
1929) and Sitodrepa panicea (Janisch, 1924) in which the period
of depletion is apparently hastened by exposure to G02. This
type of 'depletion senescence' is, in fact, in one sense an exten-
sion of morphogenetic senescence, if in the transition from larva
to imago the organism loses the power of synthesis or assimila-
tion of some material which it is able to store during larval and
pupal life. How far depletion of larval reserves is a general
feature of insect senescence it is difficult to say. The non-feeding
or the starved imago is necessarily dependent upon what stores
it has, although Metchnikoff (1915) from a careful study of
Bombjx, favoured an 'accumulative5 rather than a 'depletive'
mechanism to account for imaginal death. Other imagines prob-
ably vary a great deal in their biochemical accomplishments.
Some lepidopteran imagines feed on nectar and are known to
absorb water and sugars. Frohawk (1935) kept Nymphalis antiopa
alive for three months from eclosion by feeding sugar solution.
On the other hand, robust Coleoptera, such as Blaps, are fully

98

The Distribution of Senescence

capable of living on their intake and stores for ten years, while
the mole cricket has been thought able to live much longer.
Activity reduces the life-span: Camboue (1926) greatly pro-
longed the life of butterflies by decapitating them.

The influence of reproduction on life-span is equally variable,
but it often seems to involve inroads upon stored and irreplace-
able reserves. Unmated females of Periplaneta lay fewer eggs
than mated females and live longer (Griffiths and Tauber,
1942). The life-span in both male and female Drosophila is sub-
stantially decreased by mating (Bilewicz 1953). Krumbiegel

%
100

SO

60

kO

20

M m

uk

\|»0

^^

!6

6

[30

39

V

JK

JO

V1

\j

6

t

)AYS

Is

^r-

! .

1-

7

8

II 12

Fig. 26. — Survival curves of 145 isolated virgin females (I) and 44 isolated,
fertilized females (II) of the moth Fumea crassiorella (from Matthes, 1951).

found that the reserves in the fat-body of Carabids decreases
after first copulation, but increases again with feeding (1929).
In the moth Fumea crassiorella Matthes (1951) found that the
longevity of the female was halved by copulation if egg-laying
was allowed, and slightly reduced by it if egg-laying was
prevented (Fig. 26).

The theory of 'cerebral death' (Gehirntod) in insects arises
chiefly from some long-standing work on bees. Hodge (1894,
1895), Pixell-Goodrich (1920) and Schmidt (1923) all described
cerebral degeneration, reduction of cerebral cell-number, and
disorganization of the nervous system as characteristic and
probably causal mechanisms in the senescence of worker bees.
h 99

The Biology of Senescence

According to Hodge, the cell-number in the brain of old
workers was reduced by three quarters. Pixell- Goodrich found
that in diseased, and therefore inactive, workers, the cerebral
architecture was more normal than in healthy workers. Schmidt
attributed the reduction in cell size and cell-number to direct
'wear by use', the amount of work done by the insect being a
fixed quantity. Holmgren (1909) found a similar deterioration
in the supraoesophageal ganglia of old termite primaries: the
brain of old physogastric queens of Eutermes was reduced to
two thirds of the volume usually found in virgin queens.
Another instance of 'Gehirntod' in insects was described by
Hansemann (1914) in Bacillus rossi (Phasmidae).

Quite apart from the fact that they have been indiscrimin-
ately transferred to mammals, these findings themselves have
been open to intermittent criticism.1 Smallwood and Phillips
(1916) were by no means satisfied that the changes in relative
nuclear size described by Hodge in worker bees resulted from
ageing or were in any way pathological. Weyer (1930) regarded
the cerebral ganglion changes as secondary, since the supposedly
senile degeneration appears remarkably suddenly, and only
after evident deterioration in other organs. In a 5-year-old
queen, Pflugfelder (1948) found some disturbances of cerebral
histology especially in the corpora pedunculata, but no sig-
nificant cerebral change in old drones and workers. Rockstein
(1950), however, found a decline in cell-number in the brain of
worker bees from a mean 522 at eclosion to 369 at 6 weeks. The
complex behaviour of worker bees deteriorates suddenly just
before death, moreover.

Schulze-Robbecke (1951) made a careful search for evidence
of 'cerebral death' in Dixippus and Melolontha and found no signs
of it whatsoever, the primary senile deterioration being most
evident in gut and musculature. 'Vielleicht hat v. Hansemann
bereits tote Tiere untersucht, was sehr leicht vorkommen kann,
da bei den Stabheuschrecken infolge ihrer Reaktionstragheit
der Ubergang von den letzten Lebensausserungen zum Tode
nicht ohne weiteres festzustellen ist.' The amount of senile
change described either as a result of fixation-artefacts or the

1 The large literature of neurone loss with increasing age in mammals,
especially in the cerebellum, is reviewed by Andrew (1955).

100

The Distribution of Senescence

sectioning of 'that which dies of itself has yet to be assessed in
the literature.

A considerable amount of work has been done upon the
physiological factors which influence longevity in worker bees.
Winter bees are known to be considerably longer-lived than the
summer brood even when they are kept under similar conditions
of temperature and activity. Maurizio (1946) found that caged
winter bees had a mean survival of 36 days from eclosion com-
pared with 24 days in caged summer bees. The life-span of
summer bees can be prolonged in two ways — by feeding pollen
to caged bees (Maurizio, 1946), or by removing all the sealed
brood regularly from the colony, so that the same bees continue
with brood-rearing throughout life — under these conditions
bees may live as long as 72 days (Moskovljevic, 1939; Maurizio,
1950).

Two factors appear to influence the longevity of workers. One
of these is certainly activity. Ribbands (1950) found that anaes-
thesia with C02 had the effect of causing young bees to begin
foraging earlier than usual: in bees which forage early, expect-
ation of imaginal life is less (30T ±1*2 days), but expectation
of for aging- life is greater (15-0 ±1*2 days) than in late starters
(37-1 ± 0-6 and 10-8 ± 0-8 days). The second appears to be
dietary. 'Winter bees differ from summer bees in the greater
development of their pharyngeal glands and their fat-bodies.
This development results from autumn consumption of pollen,
in excess of the requirements for immediate brood-rearing. In
queen-right colonies in summer, prevention of brood-rearing
can produce similar consequences, and in pre-swarming colonies
temporary interruption of brood-rearing produces conditions
different only in degree.' In all these cases the increased expect-
ation of life is associated both with enhanced development of
the pharyngeal glands and fat-body, and with decreased activity
(Ribbands, 1953). It appears that worker bees have a life-span
which is partially expressible in 'flying-hours', and that this
life-span, and the total output of work per life, can be increased
by increasing pollen consumption (Maurizio, 1950), but summer
bees only increase their life-span in this way if they are deprived
of brood. There is also ground for believing that the activity of
worker bees is reduced by the possession of internal food

101

The Biology of Senescence

reserves. In winter bees, then, absence of brood leads to reple-
tion, which in turn induces both quiescence and inherently
greater longevity at a time when both are beneficial to survival.
The expectation of foraging life decreases in proportion to the
age of the bee when it begins to forage — endogenous senescence
therefore appears to play a part in limiting the life of workers,
and they do not all die from accident alone (Ribbands, 1952).
Whatever the facts concerning 'Gehirntod', this process of sen-
escence appears to contain a major depletive element, com-
bined, in all probability, with an element of mechanical
damage. In this respect the senescence of worker bees conforms
to a pattern which seems to be widespread in insects.

2-5-6 MOLLUSCS

Pelseneer (1934) divides molluscs in the wild into annual
species, pluriennial species with a short reproductive life, and
pluriennial species with a long reproductive life. In some
members of this last group, indeterminacy of life-span cannot be
excluded. Most of the evidence is obtained from wild material.
The combination of patterns appears analogous to that found
in fish. Like fish, molluscs include short-lived forms, forms with
a longer but apparently determinate life, and forms, expecially
among the larger pelecypods, which appear to have no
maximum size.

The annual forms include many nudibranchs (Pelseneer,
1934, 1935) and probably most of the smaller freshwater species
(Paludestrina jenkinsi — Boycott, 1936; Ancylus — Hunter, 1953).
According to Boycott (1936) Planorbis corneus is the only British
fresh water pulmonate which is not normally an annual. Many
of these annual forms die immediately after reproduction. In
Viviparus contectoides (van Cleave and Lederer, 1932) and V.
bengalensis (Annandale and Sewell, 1921) the wild males live
one, and the females up to three years. In captivity Oldham
kept male V. contectus for 4| and female for 5 years (Oldham,
1931), and living embryos were present at the time of death.
Growth, judged by length, ceased in the second or third year
of fife. A number of other forms live for a maximum of 2 or 3
years in the wild, breeding during one or two seasons (Lioplax
— van Cleave and Chambers, 1935; Bithynia — Boycott, 1936;

102

The Distribution of Senescence

Lilly, 1953; Fossaria — van Cleave, 1935; Sphaerium — Foster,
1932; Teredo navalis — Grave, 1928). Specimens of Limnaea
columella kept in captivity under good conditions give a life-
table showing a typical senile increase in force of mortality
similar to that in Drosophila (Winsor and Winsor 1935; Baily
1931). The growth of Limnaea has been studied by Baily (1931)
and Crabb (1929). Among rather longer-lived forms, Pelseneer
(1934) found a complete cessation of shell growth and a decline

200

100 -

i

i

I 2 3 4 5 6
YEARS OF LIFE
Fig. 27. — Egg production of Eulota fruticum (after Kunkel, 1928).

in fertility with age in Gibbula umbilicalis, complete infertility
being general at about 54 months, and the extreme life-span
4J-5 years. In Eulota fruticum, the relation of egg-laying to age
has been determined. According to Kunkel, death takes place
in captivity 'when the germinal glands are exhausted' (Kunkel,
1928). (Fig. 27.) In Physagyrina DeWitt (1954) found a distinct
post-reproductive period, amounting to as much as 49 per cent
of the total life-span in mass culture, or 36 per cent in isolation
culture, the overall mean life-spans being 211 and 143 days
respectively. Szabo and Szabo (1929-36 passim) published a

103

The Biology of Senescence

number of studies upon the 'senescent' changes observable in
the digestive gland and nervous system of Agriolimax: these,
however, were inconstant from species to species, and a life-
table (Fig. 6c, p. 20) constructed from their data by Pearl &
Miner (1935) shows a steady high mortality: it is probable that
none of these slugs reached their maximum potential age.

Very little evidence exists to relate the apparent senescence of
short-lived molluscs in the wild to their growth pattern, or to
their potential life-span in isolation. Van Cleave (1935) and
Hoff (1937) considered that the snails Fossaria and Viviparus con-
tinue to grow throughout life; in these forms, according to van
Cleave, the maximum size which is characteristic of the facies
of any colony is secured by a combination of environmental
effects on the growth rate and an endogenous process of sen-
escence which kills the animal after the completion of its life-
cycle, irrespective of its general somatic growth. In the large
Trochus niloticus Rao (1937) found no evidence of senile mortal-
ity, the upper limit of age being about 12 years in a wild
population, and growth continuing at a decreasing rate through-
out life. On the other hand, in spiral gastropods with an
elaborate lip armature, growth must be effectively determinate
so far as shell size is concerned. It has been suggested that in
Polygyra growth in size ceases at lip-formation, but body weight
and shell thickness continue to increase (Foster, 1936). Przi-
bram (1909) quoted observations by Taylor and de Villepoix
that the gland-cells of the mantle disappear in fully-grown
specimens of Helix aspersa, H. nemoralis and Clausilia perversa.
The growth of molluscs is seasonal, and the development of the
gonad appears in some forms to compete with, or inhibit, body
growth — in oysters, the periods of shell growth occur in each
year before and after spawning (Orton, 1928) while in
Hydrobia ulvae parasitic castration leads to gigantism (Roths-
child, 1935). The life-span of such giants was unfortunately not
recorded. On a small series of Limnaea columella Baily (1931)
found that shell growth ceased at or soon after sexual maturity,
and that the shortest-lived individuals were those with the
highest growth rates. A life- table was constructed for this
species by Winsor and Winsor (1935) (Fig. 28). Species whose
life-cycle rarely exceeds two years may be capable of much

104

The Distribution of Senescence

longer life in captivity. Oldham (1930) kept Planorbis corneus
in active reproduction up to 6 years of age. Many Helicidae,
especially the smaller forms, have been regarded as annuals
in the wild (Lamy, 1933; Pelseneer, 1935): the potential life
of helicids and medium-sized land snails in captivity (exclud-
ing diapause, aestivation and so on), may, however, reach or
exceed 10 years (Rumina decollata, 12 years, 'Helix spiriplana\
15 years — Vignal 1919; Helix pomatia, 6-7 years — Kiinkel 1916;

90

X ■ i T 1

1 1 1 1 I --'

60

-

o

70

z

>

>

60

**

a.

D

H

z

50

\

UJ

40

A

U

QC

UJ

a

30

^V ^

20

\\

10

1 1 ! 1 1

i i niL— -fr-i i

25

50 75

Fig. 28.

100 J25 150 175 200 225
AGE IN DA\S

-Life-span of the pulmonate Limnaea columella at two different
population densities (from Winsor and Winsor, 1935).

6-8 years — Cuenot, 1911; H. hortetisis, 6 years, hybrid H. hor-
tensis x nemoralis 10 years — Cuenot, 1911) while Oxystyla capax
has been revived from diapause after 23 years (Baker, 1934).

Even less is known about the longevity and liability to senes-
cence of most marine gastropods. In Acmaea dorsuosa Abe (1932)
found that growth continued in 15-year-old specimens from
some localities, while in other localities an apparent specific
size was reached at 5 years. Apparent specific size in certain

105

The Biology of Senescence

colonies was also found by Hamai (1937) in Patelloida grata. The
most suggestive evidence of a determinate life-span in limpets
comes from Fischer- Piette's (1939) observations which showed
a definite inverse relationship between longevity, judged by
growth rings, and rate of growth in different stations (Fig. 29).
This strongly suggests that a process of morphogenetic ageing is
occurring at different rates depending on the rate of growth.
In some pelecypods, there might be ground for regarding
growth as genuinely indeterminate. It continues in the oldest

</> 40

LU

cL

\-

LLI
Z

z

I

20

10

UJ

YEARS

I 2 3 4 5 6 7 8 9 10 I! 12 13 14 15 16 17

Fig. 29. — Growth and longevity of Patella vulgata in various stations, showing
the short life of rapidly-growing populations (from Fischer-Piette, 1939).

recorded specimens, and the occurrence of abnormally large
individuals of species whose age can be estimated by growth
rings makes it possible also to estimate the rate of decline of
growth rate by measurement of the intervals between the rings.
Growth continuing actively to the maximum recorded age has
been found in many forms (Cardium, 16 years — Weymouth
and Thompson, 1930; Venus mercenaria, 40 years — Hopkins,
1930; Siliqua, 14-16 years — Weymouth, 1931; Pecten jessoensis,
8 years — Bazykalova, 1934; Mya, 7-8 years — Newcombe, 1935,
1936). The larger fresh-water pelecypods, which have fewer
enemies and are not subject to tidal disturbances, reach even
greater ages. Huge specimens of the washboard mussel, Mega-

106

The Distribution of Senescence

lonaias gigantea (Unionidae) showing 54 and 36 annual rings are
recorded (Chamberlain, 1933) — these molluscs were still grow-
ing, and even larger examples exist. Longevity of this order is
not, however, confined to large species — marking experiments
have enabled some Unionidae (Quadrula) already over 20 years
old to be followed for a further 15 years, and the correspond-
ence of adult growth rings to years confirmed (Isely, 1931).
(See also Coker, Shira, Clark and Howard, 1919-20.) In the
small Margaritana margaritifera no important decline in the
growth rate has occurred at 13 years (Saldau, 1939.) Geyer
(1909) gave this species a life-span of at least 60 years. The
estimate of 100 years quoted by Korschelt from Israel (1914)
is unlikely but not impossible.

The supposed longevity of Tridacna has already been men-
tioned, together with the fact that nothing whatever is known
about its real life-span. The same applies to many large marine
pelecypods, whose probable age can only be discussed when we
know something of their growth rate.

The pelecypods also illustrate the risks of purely ideal and
mathematical representations of growth-pattern. Pseudo-specific
size from environmental causes is common. In Siliqua (Wey-
mouth, 1931), some populations reach an apparent limiting
size, cease altogether to grow thereafter, and die early: this,
like Fischer- Piette's observation on limpets (1939) might sug-
gest that a senile process is at work. Wild limpets apparently
die while in active growth, but those which grow fastest die
earliest (Fig. 29). Other molluscan populations have growth
records which, though fitted for practical purposes to an
asymptotic curve, actually give observed readings in the highest
age groups which lie well above such a curve, and indicate that
in these groups growth is continuing (Weymouth, 1931). There
is an obvious objection to the use of growth rings to measure
age, however, if conclusions are then to be drawn about con-
tinuing growth — arrest of growth lasting for years would leave
no record of itself in this system of notation. The results ob-
tained by the use of the ring method in pelecypods have so far
been reasonably consistent (see Newcombe, 1936). The validity
of growth rings as annual markers requires careful confirma-
tion in each population examined, however (Haskings, 1954).

107

The Biology of Senescence

Hopkins (1930) found that in Venus mercenaria growth was con-
tinuing actively at 20 years. The oldest specimens aged by
growth rings were in general not the largest shells. Some small
examples had reached an estimated age of 40 years, and ap-
peared to have grown abnormally slowly. This observation, like
Fischer- Piette's (1939) and Weymouth's (1931) findings, should
lead to a great deal of caution in regarding the life-span of
any mollusc as indeterminate in the same sense as that of
actinians.

2 6 Senescence in Wild Populations

Senescence as a potential part of the individual life-cycle is,
as we have seen, widespread: in discussing the evolution of
senile processes, however, it is important to know how far it
really occurs in wild animals. The weight of evidence suggests
that senescence in the wild is rare but not unknown. Its com-
monest form is undoubtedly the pseudo-senescence which follows
reproduction, but genuine senescence analogous to that of man
is occasionally reached, at least by individuals, while there are
probably some forms in which it is normally reached. If our
observation of animal life-cycles were confined to small birds
and mammals in the wild, however, we should probably not
recognize senescence as an entity except in ourselves.

2-6-1 VERTEBRATES

Although data from bird and small mammal populations
have perhaps led to an overstatement of the case against
'natural' senescence, old age is undoubtedly a relatively rare or
very rare termination to the life-cycle of vertebrates studied in
the field — as it is for man in societies where medical and
economic conditions are bad. For large numbers of animal
species, the typical curve is one in which a high or very high
infant mortality rate is succeeded by a high adult mortality
rate which does not increase with age. These species, even when
they are capable of senescence, never reach it. This type of
curve has been repeatedly demonstrated in population studies,
(see Lack, 1954). Whereas in voles kept in the laboratory the
survival curve approximates to that of man (Leslie and Ranson,

108

The Distribution of Senescence

1940, Fig. 30), in wild voles (Hinton, 1925, 1926; Elton, 1942)
and in Peromyscus (Burt, 1940) senescence is never observed,
judging from the state of the teeth and bones of recent and
fossil animals. In some populations the vole must be regarded
as an annual (Elton, 1942). Tooth wear is a reliable index of age
in short- tailed shrews, those over 2 years of age being edentulous,

0 8 16 24 32 40 48 56 64 72 80 88 96
Fig. 30. — Smoothed survival curve for the vole, Microtus agrestis in captivity
(from Leslie and Ranson, 1940).

but age limitation by this mechanical form of senescence is more
potential than actual since few survive to exhibit it. They may
survive in captivity up to 33 months (Pearson, 1945).

The log-linear pattern of decline in survivorship is highly
characteristic of birds. It has been demonstrated in the black-
bird, song- thrush, robin, starling and lapwing (Lack, 1 943a, b, c) ,
redstart (Buxton, 1950), American robin (Farner, 1945) and
herring gull (Marshall, 1947). In a series of robins ringed by
Lack (1943a), 1 1 1 out of 144 leaving the nest (77 per cent) died

109

The Biology of Senescence

in the first year. This compares with a maximum recorded age
of 11 years, which is occasionally reached in the wild state.
The succeeding annual mortality was at a steady rate approach-
ing 50 per cent. A survivorship curve for lapwings (Vanellus
vanellus) calculated from 1333 birds is closely fitted by a line
corresponding to a constant mortality of 40 per cent per annum
(Kraak, Rinkel and Hoogerheide, 1940; Lack, 1950). The rates
of mortality for most birds which have been studied appear to
fall between 30 and 60 per cent per annum. Very much lower
figures have been recorded for large sea birds such as cormorants
(Kortlandt, 1942) in which the mortality was found to decline
from 17 per cent before fledging to an annual rate of 4 per cent
between the third and twelfth years. The annual mortality in
one species of albatross (Diomedea epomophora) is only 3 per cent.
Such birds may well live to reach senescence, if their life-span
is 50 years. But considerable evidence has accumulated, chiefly
from ringing studies, to show that the expectation of life of
some wild birds actually increases with age. Although the total
of ringed birds recovered in Europe does not exceed 10,000 per
year, a few individuals are known to have survived for longer
than could be expected if the early mortality were maintained.
R. Perry (1953) gives records of this kind (Redwing — Turdus
musicus, 17+ years; goldfinch — Carduelis carduelis, 16+ years;
meadow pipit — Anthus pratensis, 13 years) all of them in species
which have mean annual survivals of the order of 50 per cent
(Lack, 1950). A ringed starling [Sturnus vulgaris) has been re-
taken after 18 years. The probability of such records being
obtained as a result of chance, bearing in mind how few birds
are ringed, is very low indeed. An almost exactly similar situa-
tion has been observed in the human population of the Punjab,
where, in spite of a very heavy early and adult mortality, very
old individuals are not uncommon, and those who survive
beyond middle life have an expectation of life comparable to
that in Western Europe (Yacob and Swaroop, 1945).

In lizards, the wild mortality rate declines with increasing
age (Sergeev, 1939): this result agrees with the ecological
studies of Stebbings (Stebbings and Robinson, 1946; Stebbings,
1948) on Sceleporus graciosus in the wild. A very high proportion
of the population was found to consist of lizards 6 to 9 years

110

The Distribution of Senescence

old (30 per cent), and there were signs of a decreasing force of
mortality with age. In some cases the decrease may be even
steeper. In some vertebrates the enormous infant mortality
would completely overshadow subsequent trends in any life-
table based upon a cohort at birth: in the mackerel, for
instance (Sette, 1943), survival to the 50 mm. stage is less
than 0-0004 per cent.

There are a certain number of apparent instances where
senescence occurs as a regular phenomenon in wild populations
of animals, both vertebrate and invertebrate, quite apart from
occasional records of 'old' individuals. Murie (1944), from the
examination of the skulls of 608 mountain sheep (Ovis dalli),
constructed a life-table in which the death-rate was minimal
between 1 f and 5 years of age, and climbed thereafter. The
main deaths in old and young sheep appear to have been due
to predation by wolves. The Arctic fin whales studied by
Wheeler (1934) appeared to undergo an increase in mortality
after the fifteenth year of age (in females) ; the apparent increase
may however have been the result of the failure of the older
specimens to return from their winter quarters to the regions
where they can be caught and recorded. A good many larger
carnivores and herding animals probably survive occasionally
into old age in the wild state, though death must as a rule occur
very early in the process of declining resistance. It is evidently
impossible, in population studies, to assume either a constant
mortality with age or a mortality increasing with increasing age,
without some prior evidence of the behaviour of similar forms.

The 'normal' or 'wild' pattern of mortality in man is, of
course, an abstraction, since even man in modern urban society
is, biologically speaking, living 'in the wild', albeit after much
social and behavioural adaptation. Early and primitive human
societies almost certainly resembled in their ageing behaviour
those populations of animals which occasionally reach old age,
and in which the force of mortality shows some decrease during
middle adult life. This is the pattern one would expect in social
animals, where the survival of certain experienced individuals
has probably a positive survival-value for the group, although
in man the adaptation has been expressed in increasing capacity
for abstract thought and social organization, rather than in

111

The Biology of Senescence

increasing longevity per se. Although one may guess that early
man occasionally reached the point at which his powers of
homoeostasis began to fail through age, he must have died
through environmental pressure, like Murie's sheep, very early
in the process. Out of 173 palaeolithic and mesolithic indivi-
duals whose age could be determined, only 3 (all males)
appeared to have been older than 50 years, and none much
older (Vallois, 1937). Palaeolithic man in the Chinese deposits
normally died from violence at a presenile age (Wiedenreich,
1939). In rather more civilized societies, the fall in mortality
with increasing age becomes more evident: according to Lack
(1943a, 1954) the curve of mortality based on the ages given in
Roman funerary inscriptions (Macdonnell, 1913) is much like
that for birds. Hufeland's (1798) and Silbergleit's (Vischer,
1947) figures (Fig. 7) illustrate further stages in the transition
to the rectangular survival curve of modern societies in privi-
leged countries: many other examples have been collected by
Dublin and his fellow actuaries (1949).

2-6-2 INVERTEBRATES

Senescence also occurs in the wild in some invertebrates,
though it is often probably of the type of the 'parental' deaths
of shotten eels. Senescence in one form or another has been
invoked to account for the fixity of size and life-span in some
fresh-water gastropods (Sewell, 1924; Van Cleave, 1934, 1935).
The figures of Fischer- Piette (1939), relating longevity inversely
to growth rate in Patella, also suggest the operation of senescence.
It very probably occurs in the long-lived sexual forms of social
insects, such as termite primaries, and has been found to con-
tribute to the mortality of worker bees (Ribbands, 1952).
Among other insects, Jackson (1940) observed a factor of sen-
escence in tsetse flies (Glossina) occurring only during the rainy
season, when the life-span of the flies is longer. Dowdeswell,
Fisher and Ford (1940) infer the possibility of a decline in the
viability of butterflies (Polyommatus icarus) throughout imaginal
life. The position in insects is considerably complicated by the
existence of specialized overwintering forms. Overwintering
Gerrids show changes in the muscles which appear to precede
natural death — mechanical wear of the rostrum, which occurs

112

The Distribution of Senescence

in old insects, is never far enough advanced to explain their
decease (Guthrie, 1953). Cladocera and Amphipoda, together
with other small crustaceans, tend to exhibit constant specific
age in the laboratory, and may also do so in the wild state. In
a natural population of Corophium volutator (Watkin, 1941) the
mortality in females rose sharply after maturity.

Several genera of rotifers also exhibit well-defined specific age
in culture (Pearl and Doering, 1923; Pearl and Miner, 1935;
Lansing, 1942, 1947a, b, 1948) and almost certainly undergo
senescence with significant frequency in the wild state. A
population of the tube-building Floscularia marked in the wild
with carmine underwent a linear decline, followed by a steep
increase in mortality in the final survivors (Edmonson, 1945):
the curve obtained in this marking experiment was not far
different from those obtained in laboratory populations of
rotifers.

113

SENESCENCE IN PROTOZOA

3-1 Individual Cells

Much theoretical study was devoted during the last century
to the 'immortality' of protozoans, and their insusceptibility to
senescence, following the concepts put forward by Weismann.
It was considered that in unicellular organisms generally, and
in populations of metazoan cells undergoing division without
differentiation, the product of a cell's division is always a pair
of daughter cells having the same age status, and destined each
to lose its identity in another division. This theory makes very
important assumptions about the nature of the copying pro-
cesses which underlie cell division. In the majority of cases to
which it was applied, the assumptions are probably correct,
although there seem to have been no direct experiments
designed to show whether, in a given protozoan population,
the diagram of lineages shows any tendency to segregate the
deaths of individual cells towards its edges, as in a metazoan
genealogy.

Weismann had been impressed (1882) that in protozoa there
is no death because there is no corpse. 'Natural death' of indivi-
duals (often apparently from strictly endogenous causes) does
occur in protozoa, as Jennings (1945) has shown (see below);
and the assumption that there is no unrenewable matter at cell
division is not universally true; in many forms, especially those
producing swarm spores, there may be a substantial corpse, at
least as tangible as the rejected parental shell of the dividing
radiolarian. This is less often demonstrably the case in somatic
cells, and the analogy between strictly acellular organisms and
tissue cells cannot now be whole-heartedly maintained: it is
still generally held, however, that the outcome of a protozoan

114

Senescence in Protozoa

cell division is a pair of rejuvenated and infant cells rather than
a mother and a daughter of different seniority.

The indeterminacy of cell lineages has lately been attacked
with some ferocity, though on grounds of political philosophy
rather than experimental evidence (Lyepeschinskaya, 1950).
The question legitimately arises, however, particularly in cili-
ates, how far the renewal of structures at mitosis is evenly
distributed between the resulting cells. Child (1915) noticed
that in Stentor one of the progeny retains the old, while the other
forms a new, peristome. From experiments he concluded that
this made no difference to the age status of the inheritors, both
being equally 'young'. The criterion of 'youth', however, was
high susceptibility to cyanide poisoning. The critical experi-
ment of making a genealogical table to determine the order of
death of the fission products over several generations on the
pattern of Sonneborn's (1930) Stenostomum experiments does not
appear to have been carried out.

True senescence, and a marked difference in age status
between mother and progeny, certainly appears to occur in
suctorians. Korschelt (1922) noticed this in several forms {Acan-
thocystis, Spirochona, Podophrya, etc.), while in Tokophrya the
parent organism's life-span can be measured, and is increased
by underfeeding (Rudzinska, 1952). In a far greater number of
cases, there are signs that the copying process at division only
produces a new structure additional to one which already
exists, not two new, or one new and one manifestly renovated,
structure. The theoretical interest of this process (in Euglypha)
and its bearing on protozoan 'age' has been noticed before
(Severtsov, 1934). In such cases, either the structure does not
deteriorate with time, or it is maintained continuously during
life, or its possession must ultimately confer a disadvantage on
one or other of the division products.

Whereas in some protozoa organelles, axostyles, flagella and
cilia are visibly resorbed or shed at fission, and new ones pro-
duced for each fission product, in others, especially in ciliates,
maternal organelles, flagella and other structures are shared
between the progeny, being taken over by one daughter cell
while copies are developed in the other. Of two closely-related
species of Spirotrichonympha infesting termites, for example, one
i 115 "

The Biology of Senescence

divides longitudinally in the normal flagellate manner, while in
the other division is transverse, the anterior daughter receiving
all the extranuclear organelles of the parent cell except the
axostyle, while new organs are formed for the posterior daughter.
The axostyle is resorbed (Cleveland, 1938). The possibility that
the 'inheritance' of organelles may modify the age status of the
inheritor certainly merits re-investigation.

3 2 The l Senescence' of Clones

A large part of the literature included in the bibliographies
of senescence deals with the presence or absence of 'ageing' in
protozoan clones. Maupas (1886) appears to have been the first
to draw an analogy between somatic ageing in metazoa and the
behaviour of protozoan populations. He predicted that such
populations would display a life-cycle including a phase analog-
ous to metazoan senescence, and ending in the death of the
population, unless nuclear reorganization by conjugation, or
some similar mechanism, brought about the 'rejuvenation' of
the stock. For many years a vigorous competition was conducted
between proto-zoologists in seeing how many asexual genera-
tions of Paramecium, Eudorina, and similar creatures they could
rear. In the course of this process much nonsense was written
about 'potential immortality', but a great deal was learnt about
protozoan reproduction and culture methods. It became evi-
dent that some clones deteriorate and others, including somatic
cells such as fibroblasts in tissue-culture, do not. Calkins (1919)
in a classical study showed that strains of Uroleptus mobilis kept
in isolation culture without conjugation underwent senescence
characterized by falling-off of growth-potential, degeneration
of nuclei, and ultimate loss of micro-nucleus. These strains ulti-
mately became extinct. Conjugation at any stage of the process,
and probably also endomixis, produced an immediate reversion
to normal, regardless of whether the conjugates came from old
or young isolation strains. Sonneborn (1938) succeeded, by
selection of strains of Paramecium in which endomixis was long
delayed, in breeding a race which no longer exhibited any kind
of nuclear reorganization. These strains invariably died after
4 or 5 months. Rizet (1953) has recently reported similar results

116

Senescence in Protozoa

with an Ascomycete kept in continuous vegetative reproduc-
tion. On the other hand, Belar (1924) maintained Actinophrys sol
in isolation culture, without the occurrence of paedogamy, for
1244 generations over 32 months, and observed no decline in
the rate of cell division. Beers (1929) kept Didinium nasutum for
1384 generations without conjugation or endomixis. Hartman
(1921) kept Eudorina elegans in active asexual reproduction for
8 years. Woodruff's oldest culture of Paramecium aurelia persisted
for over 15,000 generations but was undergoing autogamy. The
conclusion must be that some clones are stable while others are
not.

More light is thrown on this problem by the work of Jennings
(1945) upon clones of Paramecium bursaria. He found that in this
species the life-cycle fell into well-defined phases of growth,
sexual reproduction by conjugation with other clones, and
decline. The length and character of these phases differed sub-
stantially from clone to clone. In the decline phase the death
of individual cells, and especially of the progeny of conjugation
between old clones, becomes very common. The vitality and
viability of the progeny of conjugation, even when the con-
jugant clones are young, varies greatly, and a very high pro-
portion of ex-conjugants normally die. This mortality is highest
among the progeny of conjugation between related clones. Of
20,478 ex-conjugants, 10,800 (52-7 per cent) died before under-
going their fifth successive cell division, while 29-7 per cent died
without dividing at all. Most conjugations produced some non-
viable clones, some weakly clones capable of limited survival,
and a few exceptionally strong clones, some of which appeared
capable of unlimited asexual reproduction. It is from these
strong races that the population of laboratory cultures is
normally obtained.

Jennings concluded as follows: 'Death did not take origin in
consequence of organisms becoming multicellular ... it occurs
on a vast scale in the Protozoa, and it results from causes which
are intrinsic to the organism. Most if not all clones ultimately
die if they do not undergo some form of sexual reproduction. . . .
Rejuvenation through sexual reproduction is a fact . . . yet
conjugation produces, in addition to rejuvenated clones, vast
numbers of weak, pathological or abnormal clones whose

117

The Biology of Senescence

pre-destined fate is early death. The rejuvenating function of
conjugation is distinct from, and in addition to, its function as
a producer of variation by redistribution of genes. Among the
clones produced (by conjugation) there are seemingly, in some
species, some clones of such vigour that they may continue
vegetatively for an indefinite period, without decline or death'
(Jennings, 1945).

Some authors have regarded the increased proportion of
weak and non- viable conjugants of old clones as the outcome of
an accumulation of unfavourable mutations. Comparable effects
(Banta, 1914; Banta and Wood, 1937) have been described in
clones ofDaphnia. This was long since suggested by Raffel (1932)
on the basis of Paramecium experiments. The type of lineal
'senescence' which occurs in Paramecium is in some respects
analogous to the processes which are familiar in inbred stocks
reproducing sexually, from Drosophila to domestic cattle (Regan,
Mead and Gregory, 1947), and described under the general
title of inbreeding depression, but differs from it in that in clones
the accumulation of mutations, rather than the segregation of
existing genes and the loss of the advantages of heterozygy, have
been held to be involved. The mortality among the progeny
of autogamy in Paramecium is directly related to the length of
time during which autogamy has been previously suppressed
(Pierson, 1938). The time scale of the group 'life-cycle' is
modified by a great many physical and chemical agents — on
the other hand, methylcholanthrene, normally a mutagenic
agent, delays the decline of Paramecium clones (Spenser and
Melroy, 1949).

The real mechanism of clonal senescence in Paramecium, how-
ever, has now been brilliantly elucidated by Sonneborn and
his co-workers (in press). It depends, as other workers have
foreseen (Faure-Fremiet, 1953) on the peculiar mechanism in
ciliates whereby the germinal and vegetative functions of the
nucleus are divided between two separate structures. When
Paramecium divides after a sexual process, the new nucleus of
each daughter cell again divides into two. One of these products,
the micronucleus, which reaches the anterior end of the cell,
has the normal diploid number of chromosomes, and is appar-
ently concerned solely with conveying the genotype: it is, in

118

Senescence in Protozoa

other words, the 'germ-plasm'. The other portion, the macro-
nucleus, controls the metabolism of the cell. It becomes highly
polyploid, and at subsequent cell divisions, while the micro-
nucleus divides evenly in the normal manner of nuclei, the
macronucleus distributes its chromosomes at random to the
daughter macronucleus arising from it. Because of the enormous
number of sets which it contains, every cell in the earlier divi-
sions has a fair chance of getting its quota, but with the passage
of time more and more daughters receive an unbalanced set and
a reduced physiological repertoire, and a chromosome once lost
cannot be restored from the micronucleus except by sexual
division — conjugation or autogamy. In the later stages of clonal
senescence even sexual division is affected and abnormal or
non-viable products increase. Sonneborn has shown that this is
not due to the accumulation of mutations, since it can be pre-
vented by periodic autogamy, even though this does not alter
the genotype: it appears to be due to injury inflicted upon the
micronucleus itself through the abnormal intracellular condi-
tions produced by the defective macronucleus. In ciliates the
germ-plasm has to live in the cell where the processes of somatic
maintenance are carried out, and it is therefore unusually
exposed. This is probably a unique situation — it does not even
apply in other ciliates — and the division of function between
vegetative and germinal nuclei is confined to this group. The
existence of presumed cytoplasmic mutations, although there is
no evidence to relate them to metazoan senescence as such,
might be far more relevant to it than studies of protozoan clones.
A kindred subject, that of somatic aneuploidy, is discussed in
6-1-3 (p. 168). It is in any case probably misleading to identify
the decline of protozoan cultures with the metazoan senescence
which it superficially resembles; it is doubtful if analogies can
properly be drawn between acellular organisms and metazoan
cells, and the only relevance of the whole question of 'ageing5
in protozoan clones to ageing in the metazoan body lies in the
the light which it might possibly throw upon the effects of cell
division in renewing expendable enzyme systems. There is no
special reason, upon the present evidence, why the 'senescence'
of Paramecium should continue to figure as extensively as it has
done in treatises devoted to gerontology.

119

The Biology of Senescence

The 'senescence5 of some lines of plants in vegetative propa-
gation apparently depend on the accumulation of exogenous
viruses which hamper vigour (Crocker, 1939) — other agricul-
turally important varieties have been propagated vegetatively
for years or centuries without deterioration. The accumulation
of exogenous viruses itself raises interesting questions in regard
to the possible accumulation of other, endogenous, intra- or
extranuclear self-propagating materials.

Not all senescence or degeneration in clones, however, can be
put down to the peculiarities of protozoa or to the action of
viruses. A striking example of such a degeneration has been
studied at Oxford by K. G. McWhirter, to whom I am much
indebted for his unpublished observations on it. This is the con-
dition called 'June Yellows', which affects strawberry plants
propagated by runners, and impairs the formation of chloro-
plasts. It appears simultaneously in all plantations of a clone,
even when they are geographically separated, and progresses in
jumps, all the plants of the same clonal (but not individual) age
passing synchronously from stage to stage. Usually in the end
the clone dies out. The condition cannot be transmitted to adult
plants by grafting. Transmission to seedlings is ambilinear
through both egg-cell and pollen. In the progeny of crosses
between clones at different stages of degeneration it is matro-
clinous: seedlings of very degenerate 'mothers' deteriorate most
rapidly. As a clone degenerates, the tendency to transmit
'yellows' to its offspring increases. The factor or factors remain
latent in some clones, but 'yellows' may appear after varying
intervals in some of the selfed or crossed seedlings obtained from
these clones, thus showing a latency reminiscent of that of the pre-
sumed oncogenic plasmagenes. This similarity has been pointed
out before (Darlington, 1948; Darlington and Mather, 1949).

The behaviour of this degeneration is like that of a mutation
which is in part cytoplasmically controlled. Such conditions are
characterized by a lag-phase, by simultaneous appearance in all
the members of a clone, non-infectivity, passage through a series
of stable phenotypic stages, and interaction with growth and
reproductive hormones. In some of McWhirter's material,
'yellows' appeared to be aggravated during the flowering period,
although it may occur in seedlings long before flowering,

120

4

THE INFLUENCE OF GENETIC

CONSTITUTION ON SENESCENCE AND

LONGEVITY

4-1 Inheritance of life-span

411 GENERAL

It is evident in any comparison of laboratory stocks that differ-
ences of specific age are to some degree 'inherited' (Pearl and
Parker, 1922; Gonzales, 1923, Gruneberg, 1951, Fig. 31), but
detailed genetic knowledge of the manner of their inheritance
is not plentiful. Much variation in life-span occurs between
inbred lines. This variation is often related to a single heritable
predisposition to die of cancer, renal disease, or some other
single cause: in these cases it is often short life, not long life,
which is capable of genetic selection in the homozygote. Bittner
(1937) showed that in some cases it is possible to transpose the
longevities of strains of mice by cross-suckling. In other cases,
secondary causes, such as restricted capacity for activity in
deformed stocks, affect the life-span. In a stock of mice bred
by Strong (Strong, 1936; Strong and Smith, 1936) longevity
increased the apparent incidence of disease by allowing animals
to reach the cancer age. Two factors appear at first sight to be
involved in inherited longevity — absence of genetic predis-
position to specific causes of death, and a less definite quantity
('vigour') which contributes to Darwinian fitness because it is
usually expressed both in fertility and in longevity. It is by no
means certain that these factors are distinct. 'Vigour' itself may
in fact represent either the covering-up of deleterious recessives
by heterozygosis, or a state of over-dominance, in which the he-
terozygote is inherently more vigorous than either homozygote.

121

The Biology of Senescence

Hereditary factors in human longevity have often been
sought. Pearl and Pearl (1934a, b) found, for instance, that the
summed ages at death of the six immediate ancestors of centen-
arians and nonagenarians were significantly greater than in a
control series of the relatives of individuals not selected for
longevity. 86-6 per cent of long-lived (> 70) subjects had at
least one long-lived parent, while 48-5 per cent of nonagenarians

60-

60

: 40

u

20 -

i i i i I i i i i i I i i i i iir'rff'l | <"i

24 monfhl

Fig. 31. — Survival curves of mice in laboratory culture — breeding females.

Curve A based on 241 dba females, curve B on 730 Bittner albinos, curve C,

on 1350 Marsh albinos (from Gruneberg, 1951).

and 53-4 per cent of centenarians had two such parents, all
these figures being significantly higher than in the control series.
Kallman and Sander (1948, 1949) found that in 1062 pairs of
twins the mean difference in longevity between dizygotic twin
individuals was twice as great as in monozygotics. These and
other studies indicate that longevity is 'hereditary', but un-
fortunately give little light on its genetics. Beeton and Pearson
(1901) studied the longevity records of Quaker families, and
found that the sib-sib correlation of longevity was nearly twice
the parent-offspring correlation, in those individuals who died

122

The Influence of Genetic Constitution

at 2 1 years of age or later, but that there was a far lower sib-sib
correlation between those dying as minors. Haldane (1949) has
pointed out that this is the type of correlation which would be
expected where the heterozygote is fitter in the Darwinian sense
than either homozygote: insofar as natural selection operates to
elimate homozygosis, not to promote it, such fitness must imply
a higher correlation between sibs in an equilibrium population
than between parent and child. In any case, the degree of
parent-child correlation observed by Beeton and Pearson is only
a quarter that between parental and filial statures in comparable
studies.

Dublin and his colleagues (1949) have summarized most of
the historic studies on the inheritance of longevity in man. They
conclude that the popular idea of inheritance as a factor in
longevity is probably correct, that the evidence from actuarial
studies is heavily vitiated by all kinds of environmental influ-
ences, and that the order of advantage to the sons of long-lived
fathers is small compared with the secular increase in life-span
during recent generations. The difference in life expectation at
25 years between those with better and poorer parental lon-
gevity records is between 2 and 4 years — this compares with a
gain of 6-7 years in the general expectation of life at 25 years in
the U.S.A. between 1900 and 1946. Tt may be well, as has been
suggested, to seek advantages in longevity by being careful in
the choice of one's grandparents, but the method is not very
practicable. It is simpler and more effective to adapt the
environment more closely to man, (Dublin, 1949, p. 118).

It does not follow from these considerations that longer life
cannot be obtained in a given population by selective breeding,
and in mice this has, in fact, been done (Strong and Smith,
1936). There may well be single-gene characters where the
homozygote is significantly longer-lived. 'Vigour', on the other
hand, which is a correlate of both longevity and fertility, and
hence of Darwinian fitness, is likely in most cases to be an
expression of heterozygosis, and one would not expect to be
able necessarily to produce abnormally long-lived animals by
inbreeding long-lived parents.

Agricultural genetics, like natural selection, has for the most
part attempted to increase lifetime production averages by

123

The Biology of Senescence

increasing early output of eggs or offspring. Greenwood (1932)
found that the fertility and hatchability of hens' eggs decline
with age of the parent to such an extent that attempts to
improve the stock by breeding from long-lived birds were
economically impracticable. Apart from the obvious difficulty

Fig. 32. — Drosophila subobscura. Strain K. Survival curves of flies raised in

each generation from eggs laid by adults which had passed the thirtieth day

of imaginal life. Compare Fig. 33.

of breeding for long life in any animal with a substantial
post-reproductive period, which involves rearing all the progeny
of large numbers of animals throughout life, the consequence
of inbreeding per se, and the tendency of inbred laboratory
stocks to reach a very stable equilibrium life-span (Pearl
and Parker, 1922) militate against any such experiment. In
Pearl's own experiments (Pearl, 1928) the long- and short-lived

124

The Influence of Genetic Constitution

Drosophila segregates were identified in the Fx by subsidiary,
anatomical characters known to be associated with the desired
lines. In Drosophila subobscura of the structurally homozygous
Kiissnacht strain, which had been in culture for about three
years, and had reached an equilibrium life-span considerably
shorter than that of wild-caught flies, breeding for 8 generations
over 1 year exclusively from eggs laid after the thirtieth day of
parental life produced no significant alteration in mean imaginal
longevity (Comfort, 1953) (Fig. 32).

The inheritance of long life in man is presumably bound up
with the inheritance of 'general health' (Pearson and Elderton,
1913; Pearl, 1927) an element which is not more susceptible to
analysis than 'vigour', though it has been partially described in
terms of response to stress (Selye, 1946). Robertson and Ray
(1920) found that in a population of mice the relatively long-
lived individuals formed a stable sub-group, displaying the
least variation and the highest resistance to disturbing factors.
In such a group the growth-rate tends to be a measure of
'general health', and rapid rather than retarded growth corre-
lates with longevity. In other studies on groups of animals living
under standardized conditions, rate of growth and length of
life have been found to vary independently (Sherman and
Campbell, 1935). The relation between growth-rate and vigour
in a mixed population requires to be distinguished from the
effect of growth retardation by dietary means in a homogeneous
population, when the retarded growers live longer. As McCay
(1952) points out, much early work on the relation between
growth-rate and longevity was vitiated by this confusion in
experimental planning.

4-1-2 PARENTAL AGE

The age of the mother is known in certain cases to modify the
longevity of her offspring. This influence apparently include a
wide range of dissimilar effects, some strictly 'genetic', and
others operating at various stages in the process of embryo-
genesis, or, in mammals, on into lactation. Certain of these
effects appear only in the Fl5 while others, like the factor
described by Lansing in rotifers, which leads to a decreas-
ing life-span in successive generations of clones propagated

125

The Biology of Senescence

exclusively from old individuals, appear to be cumulative and
reversible (p. 88).

The general question of maternal age effects in genetics is
beyond the scope of this book. It has recently been reviewed
(Miner, 1954) in a valuable symposium. In mammals the age of
the mother exerts an influence on the vigour of the progeny
which appears to vary greatly in direction and extent, even
within a species. Sawin (in Miner, 1954) found that in one
strain of rabbits, the early ( < 6 months) mortality was lowest
in the progeny of young mothers, and increased throughout
maternal life, while in another, larger, strain it reached a
minimum in the progeny of mothers 18 months old. These
changes were not correlated with any differences in lactation or
maternal weight. Jalavisto (1950) found evidence that in man
the expectation of life decreases with increasing maternal, but
not paternal, age. The percentage of abnormal offspring is
greatest in litters born to young guinea pigs (Wright, 1926) and
elderly women (Murphy in Miner, 1954). It is possible that the
association of mongolism with high parental age is a reflection
not of increasing foetal abnormality, but a decrease to the point
of viability in an abnormality which, at younger maternal ages,
is lethal (Penrose in Miner, 1954). In some celebrated experi-
ments upon mouse leukaemia, McDowell and his co-workers
have shown that when susceptible males are crossed with
resistant females, the age of onset of leukaemia in the hybrid Fx
is significantly retarded in mice born to, or suckled by, old as
compared with young mothers. At the same time, the mean
longevity of mice which die of causes other than leukaemia is
also greatest in the progeny and nurslings of old mothers
(McDowell, Taylor and Broadfort, 1951). Strong (in Miner,
1954) has described a factor influencing the latent period of
sarcoma production after injection of methylcholanthrene into
mice, which appears, like Lansing's rotifer longevity factor, to be
cumulative — a line derived from seventh to ninth litters in each
generation had a significantly increased and increasing latent
period compared with a line derived from first and second
litters. Unlike Lansing's effect, this increase has not been shown
to be reversible in the progeny of young members of the 'old'
line. There is at present no evidence in mammals of any

126

The Influence of Genetic Constitution

cumulative disadvantage in longevity accruing to 'youngest
sons of youngest daughters'. In this connection Strong has how-
ever stressed, on a number of occasions, the need for further
information on the relation between longevity and cumulative
parental age in human genealogies. Such information is un-
fortunately hard to come by, and no large-scale study has yet
been published. Lansing's effect might well be sought in the
parthenogenetic Gladocera. The age of the mother affects the
rate of development, and probably the longevity, of young
Daphnia. Green (1954) recently found that the size o£ Daphnia at
birth determines the instar in which maturity occurs, the largest
becoming mature earliest. The birth size itself depends upon
maternal age, being highest (in D. magna) in the third brood.
Since the pre-mature phase is the part of the life-cycle in which
most variation occurs, the mature phase being usually of fixed
length, early developers might be expected to be significantly
shorter-lived than late. But R. H. Fritsch, at the Justus Liebig
School in Giessen (unpublished) has compared the longevity
under carefully standardized conditions of successive genera-
tions of Daphnia raised wholly from first, third, and sixth hatch-
ings, and finds no significant trend in any of the orthoclones,
the mean life-span in all remaining at about 30 days.

4-2 Heterosis or Hybrid Vigour

Abnormally long-lived animals can regularly be produced by
crossing certain pure lines, not themselves unusually long-lived,
the effect being greatest in the hybrid Fx and their offspring and
declining rapidly on subsequent inbreeding. This is, in fact, the
simplest method of increasing the specific age in many already
inbred laboratory and domestic animals. Striking examples of
this effect (heterosis) in increasing longevity have been recorded
in mice. Gates (1926) by crossing Japanese waltzing with 'dilute
brown' strains produced a generation which was still actively
breeding at 2 years of age. Comparative life-tables for crosses
exhibiting extreme hybrid vigour do not seem to have been
published. 'Super-mice' produced by heterosis develop pre-
cociously, reach a large size, and remain in active reproduc-
tion much longer than their parents, thereby exhibiting a

127

The Biology of Senescence

combination of rapid growth with increased longevity analogous
to that of the rapid-growers described by Robertson and Ray
(1920). An example of the same effect in Drosophila is shown in
Fig. 33. The greater longevity of goldfinch-canary mules com-
pared with the parent species is apparently well known to
aviarists, and such techniques of crossing are of widespread

100

k^*---» — •"•-...

^■v*— - «... —•—•—*. -^

X\ •••»....._ -• — #

90

\\ "-"••>, v'vvb^k

80

\ *•"•■•■•••.... \

70

-

v 60

>

Vb

~K

\Y %

t 50

■a

E

Z 40

30

\\ \

20

\\

10

I

(

)

10 20 30 40 50 60 70 80 90
Age in days

100

Fig. 33. — Drosophila subobscura — hybrid vigour and longevity. Survival

curves for the inbred lines B and K, and for the reciprocal hybrids between

them (sexes combined) (from Clarke and Maynard Smith, 1955).

agricultural and economic importance when applied to sheep
or to plants.

The existence and magnitude of this effect should always be
borne in mind in the planning of experiments on the life-span
of animals drawn from closed laboratory stocks — such work can
produce very seriously misleading results if unrecognized
heterosis takes place. If an experiment in which the longevity
of generations is compared begins with hybrid progeny, marked
inbreeding depression can shorten the life-span of the succeed-
ing generations if genetical precautions are not taken.

A fuller study of longevity in hybrids might provide useful

128

The Influence of Genetic Constitution

information on the nature of 'constitutional vigour' in relation
to growth-rate. The effect is variously explained. Some of the
possible complications of heterosis in relation to the criteria of
vigour are indicated by the findings of Rutman (1950, 1951),
who compared the rates of methionine uptake in liver slices
from a fast- and a slow-growing strain of rats. The meth-
ionine replacement rate in slices derived from the fast-
growing strain was almost double that in the slow, but the
growth-rate of the slow strain could be made to approach that
of the fast by transposing the litters during suckling, and
appeared to be controlled by a milk-borne factor. Interstrain
hybrids at first showed a growth pattern like that of the mother,
but later followed that of the faster-growing strain.

Although by a very elegant experiment J. and S. Maynard
Smith (1954) have shown that in certain cases at least heterosis
appears to result from orthodox heterozygy, the number of
instances in which cytoplasmic and environmental factors also
appear able to evoke vigour is increasing. This is largely a
reflection of the very heterogeneous character of 'vigour'. Some
years ago Borisenko (1939, 1941) reported an increase in vigour
in the progeny of Drosophila matings where the inbred parents
were reared under different environmental conditions. This
observation does not appear to have been repeated. The ques-
tion of the induction of vigour by non-genic means has since
been most actively investigated by avowed anti-Mendelians
(Kurbatov, 1951; Hasek, 1953, etc.) but by no means all the
positive results come from this school. As far back as 1928,
Parkes observed that mice suckled by rats exhibited an extra-
ordinary overgrowth, which results simply from excessive nutri-
tion. Marshak (1936) found evidence of a maternal cytoplasmic
factor influencing the increase of growth-rate due to heterosis
in mice. The increased vigour in progeny of pure-line ova
transplanted to hybrid mothers (Kurbatov, 1951) is also found
in transplanted foetuses (Venge, 1953). Hasek has claimed
(1953) that when parabiosis is carried out between Rhode
Island and Leghorn embryos in the egg, by an ingenious tech-
nique, the pullets occasionally show even greater vigour than
the progeny of R.I. R. X Leghorn crosses. Without endorsing the
sweeping theoretical claims based by the Russian school upon

129

The Biology of Senescence

'vegetative hybridization' of this type, it seems clear that the
last word has yet to be said upon the nature of induced vigour.
The whole problem is one which might be of considerable
interest to gerontology, since in some cases Vigour' appears
capable of induction post-conceptually, or even post-natally. It
is important to notice, however, that there is no clear evidence
at present to show that the vigour and longevity obtainable by
true heterosis are greater than those existing in wild strains.
Heterosis should be regarded, in all probability, as the restora-
tion of 'wild' vigour, whether by restoring heterozygy or by
other processes, in lines which have lost that vigour through
inbreeding. Whether the results of heterosis can be superior
to those of wildness , in longevity or otherwise, remains to be
demonstrated.

4-3 Sex Differences

In most animals which have been studied, the male sex is
the shorter lived. This is true in organisms as dissimilar as fish
(Bidder, 1932; Wimpenny, 1953), spiders (Deevey and Deevey,
1945, Figs. 34, 35), Drosophila (Alpatov and Pearl, 1929;
Bilewicz, 1953), Habrobracon (Georgiana, 1949), Tribolium
(Park, 1945, Fig. 36); water-beetles (Blunck, 1924) and man.
In exceptional cases the preponderance of male mortality can
be reversed. Thus Woolley (1946) found that in crosses between
dba female and c57 male mice, the virgin females of the Fx had
a mean life of 27 and the males 29 months: in the reciprocal
cross, the females lived 30 and the males 33 months. Males of
Rattus natalensis outlive the females (Oliff, 1953). Darwin (1874)
regarded the shorter life of the male as 'a natural and con-
stitutional peculiarity due to sex alone'. Attempts have also
been made to explain it in genetic terms (Geiser, 1924; Gowen,
1931, 1934). Gowen constructed life- tables for Drosophila inter-
sexes and triploids, and concluded from his results that chromo-
some imbalance in itself exerted an adverse effect on life-span.
In most of the forms where full life-tables have been made, the
bias of mortality against the male follows the rule of greater
vigour in the homogametic sex. McArthur and Baillie (1932)
pointed out that if the lowered vitality of the male was due to

130

SURVIVORS (lx) Curves

for the Black Widow,

known males and

Males females only

SURVIVORSHIP (U) Curves

Black Widow Males

Black Widow Females

DEATH (dx) Curves

IO(HO-60-40-?0 0 +20+40+60+80 +I0O+I2O+WM60H8O
Percentage deviation from mean duration of life

Fig. 34. — Survivorship curves for 82 males and 45 females of the black widow
spider Latrodecte smactans (Fabr). Note that the male curve is shown to five
times the time scale of the female curve (from Deevey and Deevey, 1 945) .

Fig. 35. — Survivorship, death and death-rate curves for the black widow.

Note that the death curves are shown to twice the ordinate scale of the

others (from Deevey and Deevey, 1945).

1000

\ x\

MALESV "N

^FEMALES

500

O

■ >

>

■ a.

D

■ in

i

-H-1-

AGE IN DAYS

■ ' 1 ' ' ■ i ■ ' '

^*~"~~s-=^=-=^

0

1 ' ■ ' 1 ' ' ' 1 ' ' ' 1

III.

0 80 160 240 320 400 480

Fig. 36. — Survivorship curves for male and female Tribolium madens (radix
of 1,000 images) (from Park, 1945).

The Biology of Senescence

greater homozygosis for adverse genes, the effect should be
reversed in those forms where the female is heterogametic —
notably lepidoptera and birds. From the studies of Landauer
and Landauer on fowls (1931) and of Rau and Rau (1914) on
saturnid moths, they could find no evidence of such a reversal.
Adequate life-table studies are still very scarce in these groups.
In crosses, the difference in vigour between homogametic and
heterogametic sexes may certainly be so great that only the
homogametic reaches maturity — thus Federley (1929) found
that in certain interspecific crosses in hawk moths, only the
males survived pupation, though in reciprocal crosses both
sexes survived. Beside these studies, that of Pearl and Miner
(1936) upon Acrobasis caryae, which is one of the few actuarially-
constructed lepidopteran life-tables which have been published,
and an extensive study by Woodruffe (1951) on the survival of
the moth Hofmanophila pseudospretella under different environ-
mental conditions both show a significantly higher female life-
span. Alpatov and Gordeenko (1932) working on Bombyx mori,
found no difference in longevity between unmated males and
females, but a significantly longer male life in mated moths. Re-
examining the results of Rau and Rau (1914) they concluded
that in both Samia cecropia and Calosamia promethea the mated
female had a shorter life-span than the male. This difference,
however, might be due at least in part to the exhaustion of
reserves by more frequent egg-laying in mated females. The
life-span of the female Aglia tau is said to be the shorter
(Metchnikoff, 1907).

Rey (1936) working on the non-feeding imago of the moth
Galleria mellonella, found that the males lived up' to twice as long
as the females, the difference being unaffected by humidity but
exaggerated at low temperatures. He assumes this to be 'the
rule for lepidoptera'.

In poultry, it seems to be established that the female is the
more viable and has the longer reproductive life (Pease, 1947)
and observations such as those of Mcllhenny on wild ducks
(1940), which indicate an increase in the proportion of males
with increasing age, are probably the result of differential risks.
Longer life-span in males is also found in some other birds in
the wild (Lack, 1954). In cyprinodont fishes, some of which have

132

The Influence of Genetic Constitution

an atypical mechanism of sex-determination, evidence is in-
adequate, but Bellamy (1934) found no conspicuous sex differ-
ence in longevity in a small series. An example of longer life in
the male teleost occurs in minnows (van Cleave and Markus,
1929) but this refers to a wild population. It seems altogether
likely that where a sex difference in longevity is observed it
arises from the sum of differences in metabolic rate and be-
havioural pattern — in other words, from physiological sexual
dimorphism. A number of invertebrate metabolic studies sup-
port such a view (Daphnia, Mc Arthur and Baillie, 1929a, b;
Drosophila, Alpatov and Pearl, 1929), by indicating that the
'rate of living' in the male is in fact higher. The degree to
which the inferior vitality of the male mammal results directly
from the action of androgens has been discussed, and the whole
question of male mortality reviewed at length, by Hamilton
(1948). In man, the higher male mortality is present both in
utero and in infancy. At later ages the question is, of course,
complicated by social and occupational factors (Herdan, 1952).
There are as yet no fully satisfactory human data upon the
relative longevity of castrates, though their life-span is certainly
not grossly inferior to that of normals. Many of the highest
recorded ages in cats have occurred in gelt males (p. 48) . The
finding of Slonaker (1930) that castration produces a slight
decrease in rat longevity was based on too few animals to be
significant.

In some instances (Drosophila, Bilewicz, 1953) the life of the
male is still further shortened by copulation. In others (Latro-
dectes, Shulov, 1939-40) the male dies after a determinate short
life-span, whether mated or not. While the mortality of Anglican
clergy in England during the 1930's was only 69 per cent of
the general male mortality, and that of other Protestant clergy
74 per cent, the mortality of Roman Catholic clergy was 105 per
cent (Registrar-General's statistics, 1938). This observation is
complicated by a variety of factors: in rats, however, regular
mating improves the condition and longevity of the male
(Agduhr, 1939; Agduhr and Barron, 1938). This might tend
to support the view that the virtues of 'continence' in man,
vis-a-vis longevity, have been over-praised by interested parties.
After citing the opinion given by the Dutch physician Boerhave

133

The Biology of Senescence

(1668-1 738), who 'recommended an old Burgomaster of Amster-
dam to lie between two young girls, assuring him that he would
thus recover strength and spirits', Hufeland (1798) remarks 'We
cannot refuse our approval to the method.' It would seem by
tradition to be applicable only to the male.

44 Progeria

Although the rate of senile deterioration varies between indi-
viduals, the specific age of genetically homogeneous animal
lines is very stable; even in human populations the range of
apparent variation is not very great, and the few descriptions
of racially-distributed 'premature senility', as in Eskimos
(Brown, Sinclair, Cronk and Clark, 1948) are not actuarially
supported, though such variation, genetic or environmental,
may occur.

Sporadic cases of syndromes having some of the general char-
acters of premature old age occasionally occur in man. It is not
clear how far any of these syndromes can be regarded as
genuine accelerations of the timing mechanisms which deter-
mines senescence. They are apparently pleiotropic genetic de-
fects, occurring commonly in sibs, and are most conveniently
mentioned here. They have been regarded as pluriglandular
endocrine disturbances, but they affect many ectodermal struc-
tures and have the rather generalized character more typical of
an inborn error of metabolism — possibly the deficiency of an
enzyme system.

Infantile progeria (Hutchinson- Gilford syndrome) (Thomson
and Forfar, 1950; Manschot, 1940, 1950) occurs in childhood.
After an apparently normal infancy, the child begins to show
retarded growth, with dwarfism and progressively increasing
physical abnormality. The appearance becomes senile, the skin
atrophic, and there is hypertension with extensive atheroma
and calcification. Death usually occurs from coronary disease
before the thirtieth year. The mental development of these
children may be retarded, but is more typically precocious.
Cataract may occur. The endocrine appearances at necropsy
are inconstant, but pituitary eosinophiles are reported to be
reduced (Manschot, 1940).

134

The Influence of Genetic Constitution

Adult progeria (Werner's syndrome) was first described by
Werner (1904) in four sibs. It bears some resemblance to a
delayed infantile progeria, occurring after growth has been
wholly or partially completed. The subjects are short and of
unusual appearance. The symptoms begin in the third or fourth
decade, with the development of baldness, greying, skin changes,
cataract, calcification of vessels and occasionally of tissues,
osteoporosis, hypogonadism, and a tendency to diabetes (Thann-
hauser, 1945). This seems in general a more promising source of
analogy with normal senescence than does the infantile progeric
syndrome. Extensive bibliographies of progeria are given by
Thannhauser (1945) and by Thomson and Forfar (1950).

Other less generalized conditions with rather similar sympto-
matology have been described. 'Senile' change may be limited
to the extremities (acrogeria) . The main interest of these con-
ditions is in providing examples of mechanisms which may
mimic the deteriorative changes of human old age. The con-
ditions are all rare, and no parallels have been described in
laboratory animals. In infant progeria, pituitary growth-hor-
mone deficiency appears to play some part, though the condition
differs markedly from straightforward dwarfism. The deficiency
of oxyphil cells in some reported cases bears a resemblance to
that which follows castration (Wolfe, 1941, 1943).

Sudden 'senescence* in adults, a great standby of the nineteenth-
century dramatist, is an uncommon endocrine, or possibly
hypothalamic, reaction to severe emotional shock or accident,
which, although not genetic, can conveniently be considered
here because of its superficial affinity with progeria. In the
interest of literary effect, the preliminary phase of sudden bald-
ness is usually not stressed. The hair may fall out within twenty-
four hours, to be replaced when it grows again after an interval,
by white or structurally defective hair — impotence, depression
and cachexia are described as concomitants. The condition is
recoverable, and appears to have more connection with Sim-
monds' disease than with senescence. A case was reported by
Greene and Paterson (1943) in a railwayman who fell from a
locomotive and suffered head injury and severe shock. A few
cases are alleged to have followed intense fear, as in battle. The
pituitary may well be the endocrine chiefly responsible.

135

The Biology of Senescence

4-5 Choice of Material for Experimental Study of Age Effects

Research on the senescence of man and most large mammals
necessarily involves work on genetically diverse populations.
Where closed laboratory stocks are used, or the subject is a
'geneticaP domestic animal such as the mouse or Drosophila,
genetic precautions are essential in ageing experiments, especi-
ally if comparisons are to be made between the life-spans of
different groups or different generations. The size of the effect
which can be produced in such stocks by heterosis has already
been mentioned. The presence or absence of uniformity in
the experimental population is also particularly important in
research involving life- tables, since in many inbred lines the form
of the life-table depends entirely on one cause of death which
is not typical of the species, or even the phylum.

In non-genetical experiments (nutrition, biochemistry, growth-
rate and so on) the choice lies between inbred, hybrid and random-
bred material. Inbred lines commonly have a life-span which
is rather low for the species, and this may be advantage-
ous. Their vigour is often low, though inbreeding depression
is more evident in some species than others. Inbred lines are
often chosen by non-genetical workers for bioassay, in the
belief that they have the advantage of uniformity. This, how-
ever, is not so. Griineberg (1954) has stressed two important
characters of such lines: they cannot be relied upon to remain
constant in their heritable properties with the passage of time,
and may diverge rapidly when split into separate colonies;
and they do not constitute phenotypically uniform material, but
may, on the contrary, be strikingly more variable than Fx
hybrids between strains, and even than random-bred material
(McLaren and Michie, 1954).

Hybrid material, bred in each generation by crossing inbred
lines, suitably chosen, has a number of important advantages
for general work upon ageing. In such crosses the life-span
approaches the maximum for the species under the experi-
mental conditions. Vigour is high, so that 'background' losses
due to temperature change, infection, operative mortality and
accident is much reduced, and variation between individuals is
minimal. This uniformity is itself probably a reflection of

136

The Influence of Genetic Constitution

vigour, in the form of better homoeostasis (Robertson and
Reeve, 1952). Hybrids can, like inbreds, be employed for trans-
plantation experiments. Like inbreds, too, they may all die of
a single cause, and will do so as a rule with greater unanimity
in regard to age.

Random-mated material, when mating is genuinely random,
and not occurring within an already highly-inbred colony or
between such colonies, produces animals, the strongest of which
exhibit a vigour and life-span approaching that of hybrids,
more variable than hybrids, unsuitable for transplantation
experiments, showing a variety of causes of death more closely
resembling that in human populations, and, in general, resem-
bling such populations more closely than hybrid or inbred lines.

The choice of inbred, hybrid, or random-bred material,
when it is not dictated by the fact that no pure lines are avail-
able, will depend upon which of these attributes are most useful.
The type of material must, however, be correctly stated, since
it greatly affects the interpretation of results. Probably the most
valuable approach to the comparative study of ageing, though
not always a practicable one, would be a scheme of research
carried out in parallel upon all three types of strain, in an
animal which is already genetically familiar.

137

5

t^s ~1 t£?">

GROWTH AND SENESCENCE

5-1 'Rate of Living'

The idea of the life-span as a fixed quantity is an old one. In
a great many organisms it has long been recognized that the
contrast, perhaps originally moralistic, between a long life and
a high 'rate of living' had valid biological applications. The
phrase 'rate of living' we owe to Pearl, and it conveys the con-
cept very satisfactorily without making too many assumptions.
In many organisms the life-span, like the rate of development,
is a function of the temperature over a considerable range. In
such forms it appears that a fixed quantity of something, which,
for want of a better term, we have called 'programme', must
run out and be succeeded by senescence. The organism must
pass through a fixed sequence of operations, metabolic or
developmental, the rate of its passage determining the observed
life-span.

The period in which the kinetics of metabolism were being
discovered expressed this 'programme' in directly chemical
terms. Life had an observable temperature-coefficient. Growth,
in the classical conception of Robertson (1923), followed the
same course as a monomolecular autocatalytic reaction. Loeb
(1908) attempted to answer by the determination of tempera-
ture coefficients a fundamental question about the 'rate of liv-
ing' in relation to ageing — what is the nature of the 'pro-
gramme' which has to be fulfilled before senescence begins?
Is it a programme of differentiation, or growth, or maintenance
metabolism, or of all three? Loeb's experiments showed that
the temperature coefficient of the rate of 'ageing' in echinoderm
ova differed greatly from that of their respiration. Later work
has shown the relationship between development and tempera-

138

Growth and Senescence

ture to be too complex for simple estimation of coefficients.
Morphogenesis depends upon a large number of simultaneous
and occasionally contrary processes. We should almost certainly
now be inclined to interpret the programme fulfilled by an
animal during its life-cycle in terms drawn from experimental
morphology and from the study of control systems, rather than
directly from physical chemistry.

The postulation that senescence always accompanies, or
follows, the cessation of growth, which certainly appears to fit
many of the observed facts, we owe originally to the work of
Minot (1908). It is, in fact, no more than a postulation, since,
as we have seen, there may be organisms in which senescence
occurs hand-in-hand with growth, and there are certainly
organisms, such as terrapins, which have a virtual maximum
size but are not known to exhibit senescence. Senescence in
man, judged by the life-table, commences while active growth
is in progress: Minot himself considered that the rate of sen-
escence was actually greatest when growth rate was at its maxi-
mum. If the relationship between senescence and growth-
cessation is real, it might mean ( 1 ) that that which 'causes' the
cessation of growth also causes senescence — implying that
growth-cessation results from an active and inhibitory principle,
(2) that that which no longer grows, senesces, (3) that growth-
cessation and senescence are parallel phenomena, both arising
from the process of differentiation.

The dissociability of growth from development was first
shown by Gudernatsch's researches upon the action of thyroid
in the developing tadpole (1912). Metabolism, measured by
respiration, is dissociable from both. 'The fundamental mechan-
isms are not separable only in thought: on the contrary, they
can be dissociated experimentally or thrown out of gear with
one another' (Needham) . The fundamental problem in relation
to the 'rate of living' lies, therefore, in determining which of
these processes, and in what proportions, make up the essential
sequence of operations through which the organism must pass
before senescence makes its appearance. In its crudest form the
question is: given that these processes, though dissociable, are
normally interdependent, does this organism undergo sen-
escence (1) when it reaches a particular stage of cellular

139

The Biology of Senescence

differentiation, (2) when it has exhausted a particular store of
'growth energy', whatever the nature of such a store, or (3)
when it has carried out a certain stint of metabolism — a life-
span measurable in calories or in litres of oxygen consumed?
It is immediately evident that the programme in real organ-
isms is complex, that since senescence is a diverse process the
pacemaker differs in different forms. In some cases, when
( 1 ) above has been satisfied, the further life-span of the differ-
entiated cells may depend upon their metabolism, as in (3). All
concepts based on 'wear and tear' in neurones or other cells
postulate a similar sequence: loss of regenerative power followed
by mechanical or chemical exhaustion. In the rotifer the normal
sequence of differentiation produces an organism which is
almost incapable of cellular repair, and quite incapable of
nuclear regeneration. The life-span of the adult, once this point
in the programme has been reached, is inversely proportional
to temperature and metabolic rate over a certain range. How
this effect operates we do not know. The encysted adult,
although unable to survive in the complete absence of oxygen
(Rahm, 1923) may pass years in diapause. The life-span of
many larvae can be enormously prolonged by underfeeding or
accelerated by heating: once metamorphosis has taken place
the programme is resumed but still responds to changes in tem-
perature by a change in pace. The longest-lived imagines,
termite queens, do in fact increase in size after eclosion (Harvey,
1934). In mammals it has been postulated that, since the meta-
bolic rate is held steady by various homoeostatic devices, the
essential ingredients of the programme leading to senescence
are growth and differentiation; that growth ceases as a result
of some process or processes of differentiation, and that the
absence of growth is a proximate cause of senescence.

We have already suggested that this is not wholly in accord-
ance with the evidence. The observational test, that no verte-
brate which continues to grow undergoes 'morphogenetic' sen-
escence, and that all vertebrates which cease to grow are sub-
ject to it, does not appear to be satisfied, while the experimental
test, the demonstration that the life-span of an adult vertebrate
can be prolonged by keeping it artificially in continued growth,
beginning after normal size and development have been at-

140

Growth and Senescence

tained, has not been carried out, although there is some evi-
dence that it may be experimentally feasible. We cannot yet
identify any single process which, by its failure, produces the
senile decline of homoeostasis in mammals. It is however pos-
sible to treat the developmental sequence leading to senescence,
in its relation to growth and to differentiation, as an 'integrating
system', of the type employed in various calculating and timing
devices.

The most familiar example of such a system, functioning as a
calculating device, is the taxi-meter. This machine records time
when the taxi is stationary, and distance, or time and distance,
when it is moving. The real taxi-meter does so upon an 'open-
ended' scale, the amount of the fare which can be rung up on the
the dial being theoretically unlimited, since the dials after
reaching £99 \9s. \\\d. return to zero. For the purpose of our
argument, the biological taxi-meter has been adapted by an
anarchist to produce an undesirable result when a particular
fare is reached— say £10; or, more correctly, an increasing prob-
ability of this result as £10 is approached and passed: an increas-
ing impairment of the brakes and steering would be a suitable
device. The meter records one shilling per minute, so long as
the taxi is stationary, and half a crown per mile plus one shilling
per minute so long as it is moving. In this case, if the journey
never begins, the impairment will take place eventually, though
not for a very long time. For an extended biological analogy it
is probably better to take the case in which the conditions of the
impairment reaching a disastrous stage are, first, that the fare
shall reach £10, and second that the taxi shall have travelled
at least a short distance from its starting point.

The question we have to ask is this: does mammalian sen-
escence effectively resemble such an integrating system, in
which differentiation is the higher-scoring and the essential
component, but in which retardation leading to continuance
of growth directly or indirectly delays the point at which sen-
escence appears; or does cessation of growth itself, whether it
arises from some active mechanism of size limitation or through
the attainment of an equilibrium state, directly cause the senile
deterioration? The crude application of the calculating-machine
or the time-fuse analogy has many objections, the chief of them

141

The Biology of Senescence

being that the senile decline in resistance in mammals is not a
sudden process, as it is in the rotifer, but a smooth rise in the
force of mortality beginning at an early stage. Mechanical
timing devices produce as a rule a single event after a fixed
programme, not an increasing probability throughout the pro-
gramme, though this objection does not hold good for analogue
computing systems: it is relatively simple to devise an electronic
taximeter-bomb in which the probability of an explosion in-
creases with the increase of time and distance, or a system in
which the steering of the taxi becomes increasingly impaired
as the 'programme' continues. A far more serious objection is
that in ordinary taxi-meters time and distance are not normally
interlocked, as growth and differentiation, though experi-
mentally separable, are interlocked in the developing animal.
It has been suggested that the two processes are in some degree
mutually exclusive (Bertalanffy, 1933, 1941) — a conception
which goes back to Minot. It seems probable that in most
organisms it is the component of differentiation, not that of
mere growth, which is responsible for senescence.

Analogies, in any case, are mostly of use as teaching-illustra-
tions. In the final analysis, senescence, even if it never reaches
the ideal state of being expressed as a sequence of chemical
reactions and equilibria, must presumably be reducible to a
series of definite processes — such-and-such a mechanism leads
to the loss of dividing-power in such-and-such cells, which then
have a life-span limited by the non-renewability of their enzymes
to so many chemical operations, after which they deteriorate
with the following consequences. We are nowhere near such a
picture of any one senile or developmental process in any
organism, let alone of mammalian senescence or morphogenesis
in general. A certain amount of experimental evidence has,
however, accumulated — enough to indicate the directions in
which further research might profitably be directed.

142

Growth and Senescence
5-2 Experimental Alteration of the Growth Rate

5-2-1 INVERTEBRATES

In many invertebrates, the specific age is easily altered, either
in response to temperature changes, to which it bears a simple
relation, or by retardation of growth through the restriction of
food. The total longevity of insects can be increased either by
underfeeding the larvae, or by keeping any or all of the stages,
from egg to imago, at low temperatures. The same applies to
ticks — starvation will increase the life-span of some species from
a few weeks to 2 years (Bishopp and Smith, 1938). Northrop
(Northrop, 1917; Loeb and Northrop, 1917) kept Drosophila
larvae for varying periods on a yeastless medium to delay
growth and induce stunting: by this means the total life-span
from hatching to death was increased. There is disagreement
whether delayed growth of larvae leads to an increase in the
life-span from eclosion. Northrop found no such increase in
imagines reared from retarded larvae. Intermittent starvation
of the imago of Drosophila shortens its life; retardation of
the larvae of Lymantria also fails to increase the life-span of
the imago, which cannot feed (Kopec, 1924, 1928). Alpatov
and Pearl (Alpatov and Pearl, 1929; Alpatov, 1930) found a
slight increase in imaginal life-span in Drosophila when the
larvae were retarded by development at 18°. This effect was
less evident in males, and appeared to be reversed in some
experiments: where the imagines were kept at 25-28° larvae
reared at 28° gave longer- lived flies than those reared at 18°.
The statistical significance of the differences was in any case
small.

The life-span and final size of Daphnia (McArthur and
Baillie, 1926) and Moina (Ingle, 1933; Terao, 1932 etc.) vary
inversely with the temperature over a considerable range. Like
Drosophila, Daphnia can be markedly retarded either by cooling
or by underfeeding. A detailed study on the effect of retarda-
tion upon specific age and growth in Daphnia was carried out
by Ingle, Wood and Banta (1937). By diluting the medium, it
was shown that starvation of Daphnia for varying numbers of
instars resulted in an increase of life-span approximately equal

143

The Biology of Senescence

to the period of starvation, but that individuals starved only
until the 11th or 17th instar lived longer than individuals
starved throughout life. This prolongation of life was achieved

100
90
80
70

60
50

40
30

20

2 ■

»AWWWW4

Well fed (normal)

Starved until 6th instar

Starved until 9th instar

Starved until 12 th instar

Starved until 15th instar

Starved throughout life

O Mean Longevity

DAYS OF LIFE
1 1 ■

\

15

20

25

30 35

40 45

50 55 60

Fig. 37. — Effect of restricted food upon the longevity of Daphnia longispina
(from Ingle, Wood and Banta, 1937).

by lengthening the duration of each retarded instar, the total
number of instars remaining constant. In this species (D.
longispina) the specific age appeared to lie between the 19th and

144

Growth and Senescence

22nd instars, without reference to the chronological age which
these may represent. (Figs. 37, 38, 39.) In D. magna, Anderson
and Jenkins (1942) found a mean life-span of 960 hours or 17
instars — the number of pre-adult instars varied from 4 to 6 and

Well-fed
(Normal)

Starved until
6th instar

Starved
throughout life

1015

Fig. 38. — Effect of restricted food upon the duration of instars in Daphnia
longispina (from Ingle et al., 1937).

the differences in longevity between individuals represented
differences in the length of the pre-adult period. (See also
Dunham, 1938.) The finding of Fritsch (1953) that the pan-
tothenic acid content of the medium is a major factor in

145

The Biology of Senescence

determining the life-span of Daphnia complicates the interpre-
tation of some of these studies of dietary retardation, however.

540-

/\

520-

\ ^^^

500"

^~~~y€L ^v

480-
460-
440-
420
400

^

// ''0~~c--0— \--,

« ° Well-fed (norma/) \

<=° Starved until the 9th instar \

3 80-
3 60-

• 'Starved throughout life \

ȣ No recordings from 2nd to 5th instars inclusive \

3-40

1 • —

INSTAR ^>

-\ — ' — ' — i — • — ■ — i — ■ — ■ — i — • — ■ — i — ' — ' — i — ' — ' — h

Birth 3 6 9 12 15 18 21

Fig. 39. — Effect of restricted food upon the rate of senile change in the
heart rate of Daphnia longispina (from Ingle et al., 1937).

5-2-2 INSECT METAMORPHOSIS AND SENESCENCE

A great many insects are capable of very long pre-imaginal
life, the duration of which is largely determined by food supply.
The 'rate of living', as a simple quantity treated apart from
morphogenetic processes, does not give an entirely satisfactory
picture of insect development. We might possibly make an
experimental approach to the study of insect senescence on the
following lines. Senescence, of course, occurs in the ordinary
course of events only in the imago. The larval or nymphal
stages must be regarded as a system which is self-maintaining
but which tends towards ultimate metamorphosis. They are
analogous to the young growing period of non-metabolous
metazoa. The question arises how long, if metamorphosis could
be indefinitely prevented, the metathetelic larva would remain
self-maintaining as an equilibrium system. It might presumably
do so indefinitely, or it might ultimately undergo a specialized

146

Growth and Senescence

type of senescence due to the suppression of development or to
imbalance between continued, divergent growth processes; or
it might nevertheless undergo senescence from the same cause,
whatever that cause may be, which limits the life of the imago.
To ask whether a larva or a nymph would senesce if it did not
metamorphose is not entirely idle speculation. The data which
we have on the developmental physiology of Rhodnius, chiefly
from the work of Wigglesworth, make it possible to contemplate
interfering with the development of nymphs. The pre-imaginal
phase of Rhodnius, during which growth takes place, is main-
tained by the so-called juvenile hormone. 'During larval life,
imaginal differentiation is suppressed because in the presence
of the juvenile hormone secreted by the corpus allatum the
intracellular system which leads to the production of larval
structures takes precedence over the system which leads to the
formation of adult structures' (Wigglesworth, 1953b). The
influence of temperature on larval development appears to act
through this system, high temperatures or low Oa tensions
depressing the juvenile hormone and producing prothetely, low
temperatures enhancing its effect and producing metathetely.
This system lends itself particularly well to analysis in terms of
control-mechanisms. The tendency of the cellular system in its
Tree-running' state appears to be towards the imaginal form.
Moulting hormone from a fifth-stage larva will cause a first-
stage larva of Rhodnius to metamorphose (Wigglesworth, 1934).
Second-instar moth larvae will metamorphose to minute pupae
and adults if the corpora allata are removed (Bounhiol, 1938)
and isolated fragments from the integument of newly-hatched
moth larvae tend to pupate (Piepho, 1938). The function of
the juvenile hormone appears to be to moderate or prevent this
free-running tendency, though as a standing bias, not as a
negative feedback. The point to which the free-running system
tends, moreover, is an unstable one, ending in eventual sen-
escence. In the fifth stage Rhodnius nymph the thoracic gland
undergoes very rapid disintegration as soon as metamorphosis
takes place, and the possibility of moulting and cuticular
renewal is thereby lost, from lack of evocator, although the
power of the dermal cells to respond to injected moulting
hormone remains (Wigglesworth, 1953a). Long-term change
l 147

The Biology of Senescence

in this system causes the bias to be overcome at the correct
moment. The homoeostasis achieved by the juvenile hormone
is not absolute, otherwise metamorphosis would never take
place; the metamorphosis-producing hormone ultimately carries
the day. But occasional nymphs of Rhodnius devoid of the
thoracic gland cannot metamorphose, and appear to live for
long periods without senescence. This mechanism offers an
opportunity for the dissection of just such a system of partial
homoeostasis, directed to act as a delay-mechanism, as appears
to underlie so many life-cycles which end in senescence. Work
on insect senescence is in many respects unpromising as a source
of principles which can be extended to the biology of vertebrate
old age; such research is frequently confined to the very special
circumstances which exist in the imago — in other words, to a
system which is already in a time-limited equilibrium. For
measures of interference with the growing organism, however,
and attempts to stabilize the system in its earlier stages, insect
material may prove the most manageable. Any example of
indefinite stabilisation at an immature stage, in any organism,
would be of great biological interest. The degree of drift towards
the unstable state probably varies throughout development in
different insects — Bodenstein has shown (1943a, b) that in
Drosophila early salivary glands implanted in late larvae are not
immediately capable of metamorphosis: 'Whether the organ
discs respond with growth or differentiation depends on a
definite relationship between hormone concentration and organ
responsiveness' (Bodenstein, 1943b).

5-2-3 VERTEBRATES

The possibility of producing a long-lasting but recoverable
delay in mammalian growth and development by underfeeding
first arose from the studies of Osborne and Mendel, (1915,
1916). The work of McCay on rats, which extended the results
obtained by underfeeding upon arthropod growth directly to
mammals, is well known, but still very remarkable. It also still
represents the only successful assault which has ever been made
on the problem of mammalian specific age, which is itself the
key problem of medical gerontology; and the rather exceptional
growth-pattern of rats in no way diminishes its interest. The

148

Growth and Senescence

experiments, first described in 1934 (McCay and Crowell,
1934) extended over years, and are fully reviewed in retrospect
by McCay (1952). Groups of rats were reared on a diet suffi-
cient in all other constituents but deficient in calories, and their
growth thereby retarded. After periods of retardation up to
1000 days, the calorie intake was raised to permit growth. The
animals then grew rapidly to adult size, even though the
longest-retarded group had already exceeded the normal life

100
90
80
70
60
50
40
30
20
10

H^5>Sv

.

^\^\V

\

" \

-

\ \
\ \

<OJ

-

L \ \

* \\

Hi

>

Normal
1934^

^ V

- Ij

\ N^ V

<

V

\ v \i

.^^ Retarded 1943

" z

\ \ Normal \

UJ

va^1943

\\
\\

- a.

\ \ \

-

\ V N

\\

TIME IN
i

DAYS

1 . _S*I ^^

I "*-~ i X i

200 400 600 800 1000 1200 1400

Fig. 40. — Survival curves of normal and retarded male and female rats,
showing the effect of dietary restriction (from McCay, Sperling and Barnes,

1943).

span for the strain, and continued to live to approximately
twice the maximum age reached by unretarded controls
(Fig. 40.) This long survival was accompanied by a decreased
incidence of many chronic diseases, which appeared to repre-
sent a true diminution in senile liability to death from random
causes. The chief specific diminution was in death-rate from
pulmonary diseases and from tumours. Tn general, the retarded
rat remains active and appears young whatever its chrono-
logical age. It is very alert. It tends to go blind in the second

149

The Biology of Senescence

and third year of life. Its pulse rate of 340 beats per minute is
about 100 below normal' (McCay, 1952). The basal metabolic
rate of rats so retarded lay between that of normal young and
normal old animals (Horst et al., 1934). In rats retarded for
850 days, heat production per unit surface area was lower, but
heat production per unit weight higher, than in normal controls
(Will and McCay, 1943). The aorta and kidneys of retarded
rats showed in general a higher level of calcification than those
of controls (Hummel and Barnes, 1938) perhaps on account of
the relatively higher mineral concentrations in the restricted
diet (Barnes, 1942). A further series of experiments in the
dietary restriction of animals which had already reached matur-
ity was unfortunately complicated by the introduction of many
groups of variables (exercise, casein intake, liver supplements,
etc.) — in these experiments, underfeeding produced a significant
increase in life-span compared with fully-fed controls, but the
difference was far less conspicuous than in the retardation of
young growing rats, and the factors which were most important
in determining life-span were those which determined the degree
of body fatness (McCay, Maynard, Sperling and Osgood, 1941 :
Silberberg and Silberberg, 1 954) . This difference was largely
accounted for by the higher incidence of renal disease on a high
protein diet and in obese animals (Saxton and Kimball, 1941);
in contrast to the findings in animals retarded while young, the
incidence of chronic pneumonitis and of tumours was not
reduced by underfeeding in mature animals (McCay, Sperling
and Barnes, 1943; Saxon, 1945). A 33 per cent restriction of
calories, other elements in the diet being unaltered, produces
a significant prolongation of life in male C3H mice (King and
Visscher 1950).

The results of these experiments indicate that at least some
mammals are capable while immature of undergoing prolonged
suspension of growth without any acceleration of senescence.
The suspension is not complete, since deaths occur unless some
increase in weight is allowed. The most important inference to
be drawn from the work would appear to be that senescence
itself is the direct consequence not so much of growth-cessation
as that of the attainment of a developmental stage, the timing
of which is partially, but not wholly, linked to the growth-rate

150

Growth and Senescence

— there being no evidence that starved rats remain 'young*
indefinitely. By 1150 days, moreover, only about half the
retarded individuals were capable of resuming growth (McCay,
Sperling and Barnes, 1943). There is some evidence from later
work that prolongation of the life-span, though in a smaller
degree, can be produced by intermittent dietary restriction with-
out any evident effect on the growth-rate (Carlson and Hoelzel,
1946). Moreover most of the changes which ultimately fix the
specific age appear to have occurred at the time of maturity,
since the increase in longevity obtained by underfeeding, adult
rats is far less (McCay, Maynard, Sperling and Osgood, 1941).

The mechanism of retardation by dietary restriction in grow-
ing mammals is partially known from other studies. Inanition
lowers the gonadotrophic activity of the pituitary: this was
clearly shown by the transplantation studies of Mason and
Wolfe (1930) on female rats, and again in male rats by Moore
and Samuels (1931). In rats, reduction of the protein content
of the diet below 7 per cent produces anoestrus from gonado-
tropin deficiency (Guilbert and Goss, 1932). In the 'pseudo-
hypophysectomy' of malnutrition, pituitary growth hormone
will re-initiate growth of the skeleton and decrease the rate of
weight loss even without increase of food intake (Mulinos and
Pomerantz, 1941). The subject was reviewed by Samuels
(1946). The relationship between this 'pseudohypophysectomy'
and McCay's results is not yet clear, but the effects of restricted
food intake on the pituitary probably play a major part in the
alteration of apparent specific age.

In C3H mice, Garr, King and Visscher (1949) produced
anoestrus by reducing the standard calorie intake by half: at
14 months of age, single cycles were readily induced by admin-
istering dextrose, though the dose necessary to bring this about
varied from 0-15 to 1-0 gm. When the mice were permitted at
the age of 2 1 months to feed at will, and mated, all became
pregnant, and 10 out of an initial total of 17 were alive and
sexually active at the age of 23 months.

We have no comparable observations in man. Malnutrition
can produce gross retardation of puberty (as can disease or
'indirect' malnutrition — the effects of bilharzia are particularly
striking) but such malnutrition is always total, and shortens life.

151

The Biology of Senescence

In McCay's experiments the dietary restriction was confined
to reduction of calories. The undernourished majority in the
world at the present time derive no benefit in longevity from
their circumstances. But it is not impossible, as Edmonds sug-
gested in 1832, and as Sinclair (1955) and McCance and
Widdowson (1955) have repeated, that adult life may be
shortened by the pursuit of excessively rapid growth during
childhood. Human puberty can be accelerated by overfeeding
(Bruch 1941), and there is already evidence that while the
maximum mean height of Englishmen has not increased during
the last century, it is now reached no less than five years earlier
(Morant 1950), and the loss of height with increasing age shows
a parallel advance.

In 1948, Evans, Simpson and Li confirmed with pure growth
hormone Wiesner's (1932) original finding that rats could be
kept in continuous growth throughout life by injections of
pituitary growth hormone. Wiesner had reported some im-
provement in the condition of old male rats under the influence
of growth hormone. The experiment of Evans, Simpson and Li
was not designed to study the effect of growth on longevity, and
they found that continued growth from hormone administra-
tion in rats itself leads to death from an increased incidence of
tumours. The 12 animals in the original experiment of Evans,
Simpson and Li were killed at 647 days for histological pur-
poses. With small doses of the purified hormone, 'drug-resist-
ance5 to the growth-promoting and nitrogen-retaining effects
develops (Whitney, Bennett, Li and Evans, 1948). In dogs, and
cats, continued administration of growth hormone after growth
cessation produces not growth but diabetes, while in others
(man) epiphyseal fusion prevents continued body growth after
sexual maturity. Although the response of rats both to retarda-
tion and to growth hormone is apparently atypical, and cer-
tainly differs from that which might be expected in man, the
possibility exists of comparing in a mammal the effects on rate
of senescence of ( 1 ) retarded growth to the full specific size,
(2) of accelerated growth up to, and beyond, the specific size,
and (3) of growth beyond the specific size, but beginning in old
age. Apart from the complication of tumour production, the
general endocrine effects of growth hormone are likely to be

152

Growth and Senescence

too extensive for simple life-table experiments, but its use as a
tool in work on the specific size-specific age relationship deserves
further thought. The difficulty of such a direct application may
be less than it appears. Moon and his co-workers (Moon et al.9
1952) found that massive administration of growth hormone
(2 mg./day) to mice evoked tumours in only one of the tested
strains. It appears, moreover, that growth hormone alone fails
to induce tumours in hypophysectomized animals (Asling et aL,
1952a, b). The resistance which develops to the heterologous
(ox) hormone used in such experiments may perhaps be sur-
mountable. The idea underlying this kind of investigation was
already present in the work of Robertson (Robertson and Ray,
1919; Robertson, 1923) at a time when endocrinology was
insufficiently advanced to enable it to be realized. The results
they obtained in retarding growth with 'tethelin' were almost
certainly non-specific. Work upon growth hormone in mammals
whose epiphyses do not unite would appear to be one of the
critical experiments in finding out how far growth and develop-
ment are an integrating system tending to senescence at a fixed
point, and how far mere growth, induced by one of many
anabolism-stimulating factors, is capable of reversing or pre-
venting senile change.

Attempts to accelerate mammalian senescence have been sur-
prisingly unsuccessful. While laboratory animals can be pre-
maturely killed by a number of drugs or deficiencies, these do
not in general affect the process of senescence. Experimental
efforts to accelerate ageing in rats with dinitrophenol (Tainter,
1936, 1938) and thyroid (Robertson et al. 1933) or retard it
with thiouracil (Hartzell, 1945) have been uniformly unsuc-
cessful in bringing about any change in the specific age. Petrova
(1946) obtained evidence that induced neurosis at least shortens
the life, if it does not affect the specific age, in dogs: it is signi-
ficant that in man the most effective means of reducing the
apparent rate of senile change, ceteris paribus, are psychological,
social and occupational.

5-3 Growth-cessation and Mammalian Senescence
Mammals in captivity under 'optimal' conditions exhibit
both specific size and specific age, and these vary widely

153

The Biology of Senescence

between related species, and between genetic races of the same
species. The mechanism which determines specific size has long
been believed by some workers to intervene more actively in
mammalian development and to be more selective in its action
on tissues, than the mechanism which leads to the more gradual
decline of growth in some reptiles and fish. In these forms,
according to this view, the die-away curve of growth, which is
generally exponential in relation to body weight, suggests a far
more general process of size-limitation affecting all the tissues
approximately equally, and reaching the virtual limiting size
without much alteration in the general physiology of the animal.
'It is the rule in fishes and other cold-blooded vertebrates that
growth is asymptotic and size indeterminate, while in warm-
blooded animals, growth comes, sooner or later, to an end. But
the characteristic form is established earlier in the former case,
and changes less, save for . . . minor fluctuations. In the higher
animals, such as ourselves, the whole course of life is attended
by constant alteration and modification of form' (D'Arcy
Thompson, 1942). The form of the mammalian, and especially
the human, cycle both of growth and of senescence has fre-
quently been interpreted as an active process of negative feed-
back, which operates unequally, which may contribute to the
relatively sharp arrest of growth at the level represented by the
specific size, but which results in a 'morphogenetic' senescence
depending in turn upon a rather limited number of key physio-
logical changes.

With the hypothetical relationship between growth-cessation
and senescence in mind, a number of attempts have been made
in the past to interpret senile changes in terms of endogenous
'growth inhibitors', whether these are regarded as substances
or as physico-chemical conditions (Baker and Carrel, 1926;
Carrel and Ebeling, 1921; Simms and Stillman, 1936). The
case for such an inhibiting system was stated by Bidder (1932)
in the passage already quoted (p. 12). The nature of influences
determining mammalian organ size is virtually unknown. Some
of these appear to be extra-cellular and inhibitory. In cultures,
e.g. of diatoms, growth may be arrested by the accumulation
of a metabolite (Denffer, 1948). The most primitive types of
morphogenesis, such as that found in hydroids, depend on the

154

Growth and Senescence

acquisition by certain zooids of inhibitory powers over the
development of others (Summers, 1938), although the inhibited
cells retain the potentiality of growth. It is also known that
some 'old5 tissue cells are capable of indefinite growth in
cultivation.

The suggestions implicit in this type of reasoning are tempt-
ing, but there are evidential grounds for caution in postulating
a simple, 'toxic' senescence due to the existence of a growth-
inhibiting senile principle. Such a principle is not readily
demonstrated. Bidder once rashly located it in the pineal gland.
Kotsovsky (1931) attempted successfully to retard the growth
of tadpoles by feeding senile heart muscle — an improbable
tissue for such a purpose — and Grimm (1949) obtained similar
results with senile plasma. Picado (1930) enhanced the growth
of young rats by transfusions of adult plasma. More serious data,
however, exist.

The best experimental evidence concerning growth-limitation
is probably that obtained from studies of mammalian liver.
Although mitotic figures and binucleate cells decrease in mam-
malian liver throughout life, regeneration after hepatectomy
occurs in senile rats, apparently at a rate not much lower, so
far as replacement of cell number of concerned, than in young
adults, though much less than in growing animals (Bucher and
Glinos, 1950). As Minot pointed out (1908) the adult differs
more from the infant than the old from the adult. The time lag
between hepatectomy and maximum mitotic count increases
with age (Marshak and Byron, 1945), thereby paralleling the
difference between the behaviour of tissues from young imma-
ture and young adult donors in tissue culture (Hoffman, Gold-
schmidt and Doljanski, 1937), and confirming the universal
finding of increased growth-inertia, rather than decreased
growth capacity, as the most conspicuous character of cellular
explants with increasing donor age (Gohn and Murray, 1925;
Suzuki, 1926; Medawar, 1940). In regenerating rat liver at all
ages, however, the lag reverts to the value characteristic of
young animals (Glinos and Bartlett, 1951). In young, actively
growing rats the restoration of liver mass after hepatectomy
shows a considerable rebound phenomenon, reaching 145 per
cent of the original weight in 7 days (Norris, Blanchard and

155

The Biology of Senescence

Polovny, 1942). All the general characters of tissue behaviour
during the attainment of specific size appear to be exemplified
in liver. These include (1) negative specific acceleration of
growth, (2) retention of growth capacity after the limiting size
has been attained, as demonstrated either by explants, or, in
this case, following partial removal of the organ, (3) increased
growth inertia with increasing age and (4) 'post-inhibition
growth rebound'. Medawar (1942) stresses the surprisingly wide
distribution of this last effect, which is shown by tissue cultures
(Spear, 1928) and Ambly stoma larvae (Buchanan, 1938) retarded
by cooling, and in rats or mice following brief restriction of diet
(Osborne and Mendel, 1916; Clarke and Smith, 1938; Jackson,
1936).

Much has been made of the decline in rate of wound healing
with age proposed by du Noiiy (1916, 1932) as a criterion of
senescence. This entire theory was based on less than a dozen
uncontrolled cases. Experimental studies suggest that while the
mitotic rate in wounded skin is highest in infancy, little differ-
ence in cell multiplication exists between adult and senile
animals, though here again the time-lag in reaching the peak
mitotic rate increases with age (Howes and Harvey, 1932;
Bourliere, 1950. Delayed healing of skin wounds is not clinic-
ally very evident in old people (Elman, 1953). In male mice
the mitosis curve for skin in situ is bimodal, with peaks in infancy
and again in middle age (Bullough, 1949). One easily-measured
growth-system which shows a steady decline throughout later
life is that controlling fingernail growth (Knobloch, 1951;
Burger, 1954).

Attempts have been made in the specific case of liver tissues
to relate organ size to the existence of a mitotic inhibitor or
inhibitors. Studies on plasmapheresis (Glinos and Gey, 1952)
and parabiosis (Bucher, Scott and Aub, 1950) after partial
hepatectomy have yielded some evidence that a humoral
inhibitor, of the kind envisaged by Carrel and Ebeling (1921)
disappears from circulation after hepatectomy. These observa-
tions, though interesting, could provide a suspiciously simple
picture of the dynamics of growth limitation, and of consequent
senescence.

An opposing view to the humoral school has been suggested

156

Growth and Senescence

by Medawar (1942). Both in whole animals and in specific
organs and tissues the rate of growth declines throughout life.
Medawar points out that it is not self-evident that this decline
is the result of active growth inhibition. 'We are so deeply
influenced by the spirit of Newton's First Law that we tend to
think that whenever a rate falls off, something is actively sup-
pressing it. This is true of rates of motion, but it is not true in
quite the same sense of the rates of a type of change which we
may call changes in probability states. The rate at which heat is
lost from a cooling body is initially high, and falls off as its
temperature approaches that of the environment. The rate at
which the distribution of molecules in a closed diffusion system
tends towards uniformity is likewise rapid at first, and slower
and slower thereafter. In these cases, and in others similar to
them, we are dealing with rates that fall off "of their own
accord", with systems that tend to a certain, most-probable state
at a rate which depends upon how far they have yet to go to
reach it. We may look in vain for inhibitors and controllers:
they are not there. I do not know whether what I have called
the "kinetic picture" of growth will be found to fall within the
domain of statistical mechanics. ... It is simply a picture which
we should keep in mind when thinking of growth processes, lest
we should come to regard the doctrine of growth-controlling
factors as self-evident; which it certainly is not.'

This argument is graphical rather than explanatory, and the
analogy which it contains must be approached with caution. It
is evident that organ size has certain properties of an equili-
brium state, in approaching which the cell number and growth
energy vary after the manner of potential energy in the process
of redistribution. The equilibrating forces, however, mani-
festly arise, on the evidence of explanation, from the organ's and
the cell's surroundings. Mathematically similar systems involv-
ing real energy loss, such as cooling, are in no real sense analog-
ous, since they are examples of a process not subject to further
analysis. The decline in human population-growth is as fair a
comparison. Although morphogenesis is no doubt ultimately
expressible as a redistribution of energy, 'inherent' decline of
rate in approaching a most-probable state is only explanatory, in
the sense of providing a satisfactory regression of causes to the

157

The Biology of Senescence

limit of useful experiment, if 'growth energy' is itself a form of
energy in the physical sense, analogous to heat in a kettle or
electro-chemical energy in a battery — in the hypothetical case
where a population of cells was restricted in growth by exhaust-
ing a particular energy source, employed only for growth
and not for maintenance metabolism, such a system would
apply, and would not only depict but 'explain' the course of
events.

The great value of the approach from probability, as Med-
awar points out, is in preventing a facile assumption that if a
growth-rate declines, this decline must result from the action of
a specific toxin or inhibitor. This does not mean, however, that
in a complex biological system we can avoid asking specifically
what declines, since a decline in rate implies real quantitative
and qualitative change in terms of chemical structure, and the
investigation of these changes is practicable. It appears mani-
fest that the reversion of explanted tissues to active growth is
in fact caused by removal from their previous environment. It
seems at least arguable whether the time lag in multiplication
which characterizes aged explants is inherent in the cell at all.
Simms found that the lag in cell-division of aortic explants
from old fowls can be reduced by a number of non-specific
procedures such as papain digestion, or washing with an ultra-
filtrate of serum (Simms, 1936; Simms and Stillman, 1937).
Such effects might even be purely mechanical. For most pur-
poses it is probably also desirable to regard growth energy less
as a 'store', since, to maintain the analogy, such a 'store' must
be almost immediately 'replenished' after hepatectomy or
explanation, than as a 'space', with walls defined by the con-
tinuously-varying properties of any individual cell in the grow-
ing tissue, and by the continuously-varying properties of the
'environment', in which are included all the adjacent cells of
the same tissue. Such a concept, and, in fact, any concept of
limiting size as an equilibrium process, would seem incidentally
to imply the continuous replacement of any deciduous cells.
The chief criticism of the humoral theories of growth limitation
is their readiness to assume that the limiting factors derived
from the 'environment' can (a) be treated in isolation and
(b) necessarily correspond to substances rather than to physico-

158

Growth and Senescence

chemical states and gradients. That adjacent-cell effects need
not depend upon molecular hormones is well shown by Whita-
ker's work on the mutual orientation of Fucus egg cells through
a simple pH gradient (see J. Needham, 1942). The search for
hormonal substances which can be isolated is abundantly justi-
fiable, but the failure to find them should not be astonishing or
discouraging. There is much evidence (reviewed by Stewart
and Kirk, 1954) to suggest that the 'inhibitors' detected in old
serum by Carrel and his associates were nonspecific materials,
probably including the serum lipoproteins. This is not to say
that such materials do not exert a growth-inhibiting effect in
vivo, or that such an effect is without physiological significance,
but most existing studies certainly support Medawar's conclu-
sion (1942) that there is no simply extractable contact hormone
in adult tissue which directly restrains the growth of cells. The
'inflection5 in the curve of absolute growth (weight/time) is still
occasionally quoted as evidence of active growth-inhibition, but
this is a mathematical fallacy which has been repeatedly ex-
posed (Minot, 1908; Schmalhausen, 1929; Weymouth, 1931;
Medawar, 1945).

The most interesting aspect of this question, in relation to
senescence rather than morphogenesis, turns once again on the
supposed absence of age changes in some reptiles and fish,
though speculation is vain so long as we do not know whether
this absence is real. The growth of the body, and of the indi-
vidual organs, in some of these forms follows much the same
pattern of decline as that described by Medawar in the growth
energy of isolated tissues. If reptiles whose growth declines in
this way, and whose degree of histological complexity is in any
case similar to that found in mammals, do not exhibit sen-
escence, then this general pattern of growth-decline with age,
although it occurs in many mammalian tissues treated indi-
vidually, is not the 'cause' of mammalian senescence. Equili-
brium cessation of growth implies the probability of one-for-one
replacement in tissues which are capable of continuing division,
so that unless some other process intervenes, an organism in the
equilibrium state as regards growth should remain indefinitely
self-maintaining, except for tissues whose degree of differen-
tiation precludes mitotic renewal. This seems a reasonable

159

The Biology of Senescence

depiction of the state of affairs in long-lived cold-blooded
vertebrates (at least there seems to be no good evidence to the
contrary), but it does not appear to obtain in mammals.

We have already suggested that while we might have reason
to expect senescence, or one form of it, in the total absence of
cell division, either in the whole animal or in certain organs,
it is not self-evident why, in order to avoid senescence, an animal
should be obliged to increase constantly its total cell number or
its overall body size. If this were the case, it would suggest,
perhaps, not that growth prevents senescence, but that the
capacity for continued growth reflects a type of morphogenetic
physiology which does not produce senescence.

The possibility exists, then, that vertebrate growth-cessation
might be of two kinds: that some cold-blooded vertebrates may
cease to grow visibly when a cell-population of a particular
size and composition is reached, and that this population there-
after remains substantially static, with replacement of all except
such mechanically irreplaceable cells as neurones, while mam-
malian growth is arrested by a more active process — probably
of differentiation rather than mere mitotic inhibition — affect-
ing a few key points. This would resemble in its effects the
difference between the behaviour of a society which voluntarily
limited its reproduction to replacement level, and one which,
when a predetermined figure was reached, summarily castrated
a vital and hereditary profession. Such a difference, if real,
would explain the apparently less catastrophic effects of growth
cessation upon those reptiles which exhibit virtual specific size,
compared with the rapid post-mature decline in most mammals.
Birds, significantly, occupy a midway position, since it is virtu-
ally certain that all species are subject to senescence in cap-
tivity, though at specific ages considerably higher than those of
mammals of comparable size and activity: their period of
growth, however, is proportionately much shorter. A serious
investigation of the phylogeny of senescence is badly needed.
The hypothesis put forward here would regard it, so far as
mammals are concerned, not as the consequence of general
growth cessation, but of a particular manner of growth cessation,
involving, perhaps, selective non-renewal of certain important
structures and changes in the specificity of the response in

160

Growth and Senescence

others; a genuinely morphogenetic senescence depending upon
alteration of cell-responses, and having evolved, or re-evolved,
within the phylogeny of vertebrates. The mammalian pattern
of ageing, if it differs from that of other vertebrates, would in
this case have evolved as a correlate, though not necessarily a
consequence, of several fundamentally important physiological
processes — homoeothermy, the development of a complex endo-
crine regulation centred in the pituitary, the avian-mammalian
pattern of determinate growth, which is linked with this
development, and the system of immune response and tissue
specificity characteristic of higher vertebrates. The possible
association between size-limitation and homoeothermy is inter-
esting in view of the different relationships between pituitary
and thyroid hormones in the determination of growth which
have been found in mammals and in amphibians (Evans,
Simpson and Pencharz, 1939; Scow and Marx, 1945; Stein-
metz, 1954); at some point in vertebrate evolution, a balance-
mechanism between thyroid and growth-hormone, which leads
to gigantism in the thiouracil-treated tadpole, has become con-
verted into a synergism such as normally operates in the rat or
in man. Unfortunately for any phylogenetic theory, the pattern
in fish appears to resemble that in mammals (Goldsmith et ai,
1944; Hopper, 1950). This subject will be further considered in
a subsequent chapter.

161

6

<*^> r * <^

THE MECHANISMS OF SENESCENCE

All theories of senescence are at present based on unwarrant-
able assumptions, in the absence of concrete answers to the
essential questions of fact. The formulation which would re-
ceive, perhaps, the widest assent, at least in the matter of human
senescence, is that morphogenetic processes lead to the differ-
entiation of cells which have lost the capacity for division, such
as neurones and skeletal muscle fibres, and to a suspension of
division in others, and that processes of 'wear and tear',
chemical, mechanical, or of a degree of biophysical subtlety
depending on the taste of the investigator, thereafter bring
about the decline of some or all of the tissues thus deprived of
the power of self-renewal. This is plausible and probably true.
On the other hand (1) it has been shown already that we do
not know whether all vertebrates, in spite of their apparently
similar degree of histological complexity, are susceptible to
senescence, and there is some ground to suspect that they are
not; (2) no satisfactory technique has been devised for the study
of cell populations in situ, apart from the search for mitotic
figures in sections: we do not, therefore, know the life-span of
any tissue cell in its natural situation; (3) many of the descrip-
tions of senile change in fixed postmitotics, especially neurones,
are based upon the assumption that the life-span of cells
specialized to this extent is limited by their incapacity for
division, as appears to be the case in rotifer and Anguillula cells.
The striking differences in specific age between related species
do not disprove the contribution of cell ageing to general sen-
escence, but they cast a great deal of doubt on any assumption
that the effect of wear and tear upon neurones (Bab, 1948;
Vogt and Vogt, 1946) or any similar process is the prime mover

162

The Mechanisms of Senescence

in determining the senile decline. The powers of self-renewal
possessed by neurones apart from cell division have almost
certainly been under- rather than over-estimated. Neurone
regeneration in adult fish and amphibia can involve actual cell
replacement from a reserve of neuroblasts : 1 in adult birds and
mammals it is usually held to be limited to axon growth (see
Clemente and Windle 1955 for a review of the large literature).
Cell division, however, may not be the only means of nuclear
renewal — the appearance of 'binucleate' neurones in some old
animals has been taken as evidence of a process of reconstitu-
tion (Andrew 1955). Apart from this it is evident from observa-
tion that some neurones are capable of living and remaining in
function for 100-150 years, unless we postulate a system of
'reserve circuits' which has so far no evidence to support it.
The distinction drawn by Weismann between immortal germ
cells and mortal soma still persists in many of these assumptions,
in spite of the growing number of instances where differentiated
somatic cells in invertebrates are thought to give rise to germ
cells, or to structures having the potentialities of germ cells
(Brien, 1953).

6-1 Senescence in Cells

611 'IRREPLACEABLE' ENZYMES

There is no self-evident reason why morphogenetic forces
acting upon cells, and inhibiting their free division, should lead
to their senescence. Mechanical and 'colloidal' interpretations
will not do — they fail to treat postmitotic cells as the dynamic
systems which they certainly are. A theory of 'mechanical age-
ing' in postmitotic cells could, however, be based upon the
exhaustion of specific cell constituents. It is reasonable to ask
how much of the senescence of such cells, if they necessarily
undergo senescence, is due to the existence of 'expendable'
enzymic or other intracellular structures which can be replaced
only at cell division.

1 It has even been claimed that the Purkinje cells of the mammalian
cerebellum, the least likely of all such cells to do so, undergo a cycle of
periodic replacement from such a reserve (Baffoni, 1954). This is surely
either a fundamental discovery or an egregious error.
M 163

The Biology of Senescence

The concept of an expendable 'life ferment' appears to have
originated with Biitschli (1882), although he probably regarded
it simply as a material undergoing distribution from the germ
cells, where it is highly concentrated, to the somatic cells in
which it is increasingly diluted by subdivision. There are two
essentials for our acceptance today of a system in which sen-
escence depends on enzyme exhaustion in postmitotics — we
have to postulate (1) a fixed quantity of enzyme present in the
cell and exhaustible by use, and (2) the existence of an essential
enzyme replaceable only at cell division. The first proviso
appears already to be largely met, since it is known that the
effective life-span of enzyme molecules is finite in terms of mole-
cule-turnover (Mcllwain, 1946, 1949; Theorell et aL, 1951).

The existence of enzyme systems renewed only at cell division
has not, it seems, been demonstrated as such, but with the
single general and large exception of 'hereditary materials',
nuclear and extranuclear, it has not been sought. Some direct
evidence might be derived from the action of known selective
blocking agents upon bacteria of protozoa. It will be evident,
however, that the idea of an 'enzyme replaced only by mitosis'
falls very close to some biochemical models of the gene, which
has been invested, either directly or at one remove, with direct
catalytic properties. Mcllwain (1946) has shown that in some
catalytic systems the number of enzyme molecules per cell is of
the order of unity. The inference from his figures is that if genes
are not themselves molecules acting as catalysts, each gene
during its 'lifetime' (i.e. between one cell division and the next)
produces one such molecule. Mcllwain (1949) also calculated
the life-span per molecule of the nicotinic acid co-enzyme com-
ponent of Lactobacillus arabinosus as representing the production
of 5-8 x 107 mol. lactic acid. Theorell (1951) in tracer experi-
ments demonstrated a very slow turnover of haemoprotein
enzymes, the exception being liver catalase which has a mole-
cule/life of only 4-5 days. The wastage of such systems is due,
presumably, in part to side reactions and non-specific inactiv-
ation, and in part to competitive inhibition or blocking by
metabolites partially resembling the correct substrate. If the
determination of such a single-molecule system were a cause of
cell senescence, and if the catalyst itself were to be identified

164

The Mechanisms of Senescence

with the gene, we would evidently need to postulate a copying
mechanism at mitosis in which inactivation of the catalytic
portion of the system (1) does not interfere with the production
of a copy, and (2) is itself reversible : or, alternatively, one in
which the products of division are two copies, not an original
and a copy. It can, of course, be argued that when a differ-
entiated cell in fact undergoes senescence, we cannot infer
whether any system in it would be renewed by further division.
Its failure to divide, even if that failure is a physiological one,
leading to final differentiation, may be due to the loss of a
copying mechanism. This type of problem has been encountered
already, however, by workers attempting to explain some of the
results of research on adaptive variation in Neurospora and
yeasts. 'Unless gene reproduction and gene action are totally
independent of each other, we have to reconcile the uniformity
in the reproduction of genes with the enormous variation in what
we believe to be their primary products' (Pontecorvo, 1946).
Classical genetics, although they allocate an equal proportion
of nuclear genie material to every cell, have so far given little
direct information about the activity of this material in cells of
different kinds at any time except during mitosis, and a new
category of study ('epigenetics') has had to be coined to cover
this activity. Of the large number of subsidiary copying pro-
cesses which have been inferred from adaptation experiments
and work on anuclear portions of cells some apparently con-
tinue undiminished throughout the intermitotic period. The
power of adaptive enzyme production persists in yeasts rendered
non- viable by X-rays (Spiegelman, Baron and Quastler, 1951).
In a neurone which may remain functioning in man for over
a century, either the enzymic mechanism which maintains cell
metabolism is continuously kept in repair, or it is of a kind
which is almost invulnerable to incidental spoilage by use. The
survival-time of non-dividing cells varies greatly, even between
closely related organisms: thus in Rotifers, the life-span reaches
5 months in Callidina (Zelinka, 1891) and even, perhaps, several
years in certain bdelloids (Murray, 1910). If 'wear' is to be
invoked in these cases where senescence occurs in the presence
of cellular determinancy, then the susceptibility to it must vary
enormously. Other cells — squamous epithelium, for example —

165

The Biology of Senescence

have a function which depends on the progressive change in their
structure and metabolism from formation to complete cornifi-
cation. This implies a process of chemical heterauxesis within
the cell, and all such 'open-ended' systems, if they continue,
must eventually destroy homoeostasis. The development of
histochemical methods of detecting enzymes in cells, and of
selective blocking agents which irreversibly inactivate particular
enzymes, already suggest experiments by which we might learn
something of the limits of the postmitotic cell's power to re-
generate its enzymic complement, and detect long-term changes
in this power.

A special case of limited survival in the non-dividing cell is
provided by the mammalian erythrocyte. This is one of the few
cell-types for which a life-table can be constructed. The form
of the curves obtained by a variety of methods indicates that
the decay of circulating erythrocytes is a true 'senescence', i.e.
that the probability of the destruction of a given cell increases
markedly after it reaches a certain age. There is also, apparently,
an 'infant mortality' among newly-formed red cells to make the
mimicry of a metazoan survival curve even closer.

The cause of erythrocyte 'senescence' has been the subject of
a good deal of study. Although it is probable that the proximate
cause of erythrocyte destruction is a change in the physical
properties of the stroma or the envelope of the cell, there is
considerable evidence that the timing mechanism in this in-
stance is the deleterious effect on other intracellular systems of
the products of one particular oxidation-reduction system, in
which methaemoglobin is formed from haemoglobin. The evi-
dence for this view has been reviewed by Lemberg and Legge
(1949). In this case we are dealing with a specialized, anuclear
cell — in the nucleated avian erythrocyte, haemoglobin syn-
thesis, and possibly other processes of renewal, continue in the
circulation, but the life-span is, curiously enough, very much
shorter (Hevesy and Ottesen, 1945; Hevesy, 1947) than that of
the mammalian red cell.

6-1-2 CELL TURNOVER

It is necessary to point to a widespread impression among
medical writers on senescence that cell turnover in the organs

166

The Mechanisms of Senescence

of the adult animal is virtually confined to such tissues as skin,
and that the 'cause' of senescence resides in the exhaustion
of endocrine cells which have accompanied the individual
throughout life. This may be true of some invertebrates, many
of which have a wholly determinate cell-number throughout,
or in certain organs, such as the corpus allatum of bees (Pflug-
felder, 1948) and the suggestion cannot be refuted in terms of
individual cells in man and other mammals. Such irreplace-
ables may exist, but the idea is in many ways open to the same
criticism as that which attributes ageing to changes in bio-
logically inert colloids, and the body of evidence suggests that
it is equally erroneous. Although the rate of turnover in liver
cells decreases with age, and mitotic figures become few, the
mean mitosis rate in adult rats is such as to double the volume
of the organ in the animal's lifetime, if there were no incidental
wastage.1 This figure suggests that some liver cells may accom-
pany the animals from cradle to grave, but that the majority
do not. Adrenal cortical cells are continuously replaced in the
adult cat (Lobban, 1952). Mitotic activity does, however,
appear to decrease with advancing age (Blumenthal, 1945;
Townsend, 1946; Korenchevsky, Paris and Benjamin, 1950).
The adrenals of old rats show various degenerative changes
(Jayne, 1953). Mitosis varying in frequency with the sexual
cycle occurs in the anterior pituitary (Hunt, 1942, 1943, 1947)
though it may not affect all cell types equally, since the popula-
tion changes in composition with advancing age (Parsons, 1936) .
There is no direct evidence that the power of cell replacement
is lost in any endocrine gland with age, though there may be
more general involutional changes at both cellular and tissue
levels. The pattern of mammalian endocrine cell behaviour is
predominantly one of continual division and replacement,
regulated in level by hormonal influences, and often occurring
in cycles. It is impossible to say at present whether there is a
single key exception to this pattern, but ageing is unlikely to be
so simple a matter as the defection of one type of cell. It is
significant that the syndrome of senescence cannot be produced
experimentally by extirpating any one gland.

The morphological changes in endocrine cells with age have
1 I am indebted to Mr. M. Abercrombie for this figure.

167

The Biology of Senescence

been widely studied, though here, as in all pathological studies
on ageing, no line can be drawn between cause and effect. Such
morphological changes in pituitary cells have recently been re-
examined by Weiss and Lansing (1953) and by Shanklin (1953),
but without any new findings on the rate of cellular replace-
ment. In some glandular organs, such as rat salivary glands,
mitosis becomes both rare and abnormal in pattern after the
end of active body growth, while in senile rats numerous imper-
fect mitoses occur (Andrew, 1953).

6-1-3 SOMATIC MUTATION

If the copying-mechanisms of somatic cells could be shown
to deteriorate, like those of the Paramecium macronucleus, the
further they travel from the germ line, senescence might result
from the fact of cell turnover, not from its cessation. A possible
mechanism for this has been suggested to me by Dr. Helen
Spurway. She suggests that as a result of somatic mutation the
constituent cells of some mammalian tissues may lose their
autarky and become a 'community', in which both function
and the capacity for replacement have undergone distribution.
Such a community would contain both irreplaceable and indis-
pensable members, and would therefore ultimately undergo
senescence.

It is clear that if mammalian tissues exhibit increasingly im-
perfect mitosis with increasing age, resulting in the accumula-
tion of aneuploid cells, this would lead to steady deterioration
of equilibrium. Any difference in liability to senescence between
mammals and lower vertebrates would be explicable, in terms
of this theory, on the ground that mutation rate would be higher,
in all probability, with higher temperatures and chromosome
numbers.

This is a highly ingenious but unproven suggestion. The con-
tentious issue of mammalian aneuploidy generally has been
reviewed recently by Hsu and Pomerat (1953): a large and
even more contentious body of data is reviewed by Sorokina
(1950). Variation in chromosome number has been reported in
human (Andres and Jiv, 1936; Timonen and Therman, 1950;
Therman and Timonen 1951) and pig embryos (Sorokina,
1950). In adult rat liver, Tanaka (1951, 1953) recently des-

168

The Mechanisms of Senescence

cribed wide variation in chromosome number. He found that
cells with 42 chromosomes (diploids) contribute primarily to the
growth of embryonic liver and to regeneration after hepatectomy
in adults, and that growth and restoration were apparently con-
fined to diploids and subdiploids. The idea of increasingly faulty
copying with increasing age, or of an accumulation of faulty
copies, is one best left for verification and assessment to the
cytologists, not all of whom regard the evidence for somatic
aneuploidy as satisfactory. Walker and Boothroyd (1953) have
shown that such 'aneuploidy' is easily simulated by faulty tech-
nique. Spurway's basic suggestion should, however, be suscept-
ible to verification. Clearly any evidence that the copying of
somatic cells deteriorates is of gerontological importance, and
would, if established, restore the relevance to metazoan ageing
of much which was formerly written concerning the senescence
of clones.

6- 14 SPECIFICITY

Campbell and Work (1953) have recently drawn attention to
the significance of the fact that animals cannot in general be
immunized against their own proteins; and they suggest that
the action of the genotype in determining specificity may be
chiefly a negative one, in the prevention rather than the
creation of a specific configuration. It is a matter of extreme
interest that the character of the proteins produced by the
animal body appears to change with age, as judged by calcium
binding power (Lansing, et at. 1949 Hansard et at. 1954,
serological properties (Duran-Reynals, 1940) and amino-acid
composition (Lansing et aL, 1951) and by the appearance of
collagen and elastic fibres in skin (Ejiri, 1936; Gross and
Schmidt, 1948, 1950). The acid-extractable collagen decreases
markedly with age (Banfield, 1952). In general these products
arise, however, not from aged cells, but from the cells of aged
organisms, and the change in specificity has taken place over
several cellular generations. It has occasionally been suggested
that senescence is a manifestation of an 'immune' response to
endogenous hormones (e.g. Picado and Rotter, 1936; Freud and
Uyldert, 1947). A more subtle change in specificity of cell re-
sponse, or of the properties of cellular products, whether it be

169

The Biology of Senescence

interpreted immunologically, or, more probably, morphogenet-
ically, cannot be ruled out. If cells in general or certain cells
could be shown to acquire adaptive resistance to physiological
regulators, as do bacteria to unfamiliar metabolites, from gener-
ation to generation, interesting possibilities would certainly be
opened.

Connection between the processes of senescence and those of
immunology was suggested long since by Metchnikoff, though
in rather a different context (1907). He attributed the senile
atrophy of differentiated tissues, especially neurones, to over-
active phagocytosis, brought on by constant exposure to the
toxic (and antigenic) products of symbiotic and endoparasitic
bacteria: the second part of this theory has received more
publicity than the first. Metchnikoff was also responsible for the
suggestion that many morphogenetic processes in the embryo
are 'immunological' in character, within the wide definition of
immunology implied in his theory of phagocytosis, and that the
defence mechanisms of the adult animal are directly derived
from mechanisms which, in embryonic life, have been primarily
concerned with morphogenesis. The embryo is not exposed as
a rule to exogenous antigens, though its own chemical com-
position is changing. More recent work, in fact, has shown that
exposure to an antigen during embryonic life can lead to a
lifelong inability to react against that antigen (Billingham,
Brent and Medawar, 1953). The possibility that morphogenetic
mechanisms give rise in ontogeny to the defence mechanisms of
the adult, which is evidently true in the case of some mech-
anisms of chemical homoeostasis, is still popular in Russian
research. Whether such effects modify cell responses during the
later phase of development, senescence, is unknown.

Gaillard (1942) has carried out a long series of studies relating
the degree of differentiation which can be produced and main-
tained in tissue explants to the age-status of the press juice in
which they are grown. According to these results, functional
differentiation in endocrine and other explants can be obtained
if they are grown successively in press juices from embryos of
increasing age, while some degree of regression of structure
occurs if the series is reversed, and explants are grown in juices
of decreasing age-order. This process has not so far been followed

170

The Mechanisms of Senescence

into senescence. No doubt if explants could be cultivated in
media exactly simulating the chemical and physical environ-
ment and its changes through all the stages of development, they
would pass through all the normal phases of in vivo histology.
The point of interest to the gerontologist is to know how far it
is possible to maintain a status quo at any point. Analysis of the
power of tissues to mark time developmentally while retaining
function requires more elaborate methods of organ and tissue
culture than are so far available, but the lead given by Carrel
in applying these techniques to age-processes has hardly been
pursued with the vigour it merits. Gey (1952) has recently
referred to the in vitro culture of 'thyroid, parathyroid, adrenal
cells, and the germinal epithelium of the ovary5 even to the
production of follicles, but these findings remain contentious.

6-2 Endocrine Senescence

6-2-1 GENERAL

The early discoverers of hormones were fully convinced that
they had in their hands the key to the prevention of senescence
in man. The fact that little of their enthusiasm persists today is
due very largely to the manner in which the confluence of two
deeply emotive subjects — ageing and the gonad — affected scien-
tific judgement in the early years of the century. The hypotheses
of the rejuvenators were in many respects reasonable, if their
published claims were not. In any discussion of endocrine sen-
escence it is probably worth restating ( 1 ) that organ and tissue
grafting are appropriate and fully respectable techniques for
the investigation of senile change, and that they have given
misleading answers chiefly because the wrong questions were
asked, (2) that the use of hormones in the palliation of senile
changes in man, although it is largely ineffective, was a reason-
able experiment which has not yet been exhausted, and (3)
although gonadal 'senescence' does not 'cause5 somatic sen-
escence (this was self-evident in antiquity, from the life histories
of eunuchs, long before testosterone was found to be unavailing in
reversing general senile decay), it is a highly important model
process, and a relatively accessible one for further study.

171

The Biology of Senescence

Characteristic variations in hormone output occur through-
out the mammalian life cycle; they are of two types — cyclical
and secular. These changes are the biochemical equivalent of
the sudden movements of embryonic tissue which are seen in
speeded-up films of developing organisms — they are part cause,
part effect, and they represent only the outward and visible
manifestation of changes in the quality and quantity of cell
response. The hormones most likely to be linked directly with
the senile process, such as the growth hormone of the pituitary,
cannot be estimated in the intact animal. Of those which can
be so estimated, the group of 1 7-keto-steroids show a decline
which continues with the rise of the force of mortality
(McGavack, 1951; Kirk, 1949; Hamburger, 1948; Robinson,
1948; Hamilton and Hamilton, 1948; Hamilton, Hamilton and
Mestler, 1954 (Fig. 41) ). The concept of an 'adrenopause5
analogous to the menopause seems to have little to support
it at present. Apart from this, there is no single hormonal
change which correlates with senescence, no single endocrine
organ, of those which can be removed without fatal results,
whose extirpation produces the syndrome of senility in mam-
mals; and no hormone or combination of hormones which
is known to produce more than a limited, and apparently
secondary, reversal of senile changes. Ablation of a gland is not
the same thing as its senescence in situ, and in surgical castration
for prostatic and mammary cancer there is evidence that inter-
conversion may take place between adrenal and gonadal
steroids, but no simple hypothesis that senescence is a 'with-
drawal' effect is substantiated by the existing experimental
evidence. It appears that the sequence of developmental changes
in endocrine activity which ends in senescence cannot so far be
made to run backwards by hormone supplements, except in a
very minor degree.

In a long series of studies, Korenchevsky (e.g. Korenchevsky
and Jones, 1947, 1948; Korenchevsky, Paris and Benjamin,
1950, 1953) has tried to show how far hormone supplements can
reverse the senile process, judged by the restoration of the
relative hypoplasia of organs. A great many of the senile
structural changes described in endocrines are closely paralleled,
though at a lesser level of severity, by the changes in structure

172

The Mechanisms of Senescence

which follow gonadectomy, and many of them are reversible
by gonadal hormones. (Korenchevsky et aL, 1950; McGavack,
1951). The peripheral effects of sex hormones in senility, such
as the recornification of the vagina by oestrogen (Loeb, 1944),
are familiar enough. The decrease of mitotic rate and degree
of vacuolation in the adrenal (Townsend, 1946; Blumenthal,
1945) and the increase of collagen and reticulum in the capsule

MC
16
15

14

13
12

II
10
9 ■
8 -
7 •
6 ■
5 ■
4 • \ ><\KIRK,I949

3

2 ■
I -

-HAMBURGER, 1949

HAMILTON AND HAMILTON, 1948 ROBINSON, 1948

YEARS 40 50 60 70 80 90

Fig. 41. — Neutral 1 7-Ketosteroids, 24-hour urinary excretion (Kirk 1949).

and in the parenchyma (Dribben and Wolfe, 1947) which
occur with advancing age are partially reversible by oestrogen,
and more fully reversible by a combination of oestrogen,
androgen and progesterone (Korenchevsky, Paris and Ben-
jamin, 1950). In the pituitary of the senile rat, Wolfe (1943)
found a decrease in eosinophils, but no increase in basophils
vacuolation, like that which follows castration, occurred in the
basophils with increasing age. These changes, particularly the
decrease in eosinophils, are at least partially reversed by testos-
terone propionate (Wolfe, 1941). The senescent changes in

173

The Biology of Senescence

fowl pituitary described by Payne (1949, 1952) were greatly
hastened by gonadectomy. In other words, gonadal failure may
contribute to senescence, but probably does so only when it
occurs at a certain point in the endocrine developmental pro-
gramme.

In some instances, direct estimations of the capacity of senile
endocrines to respond to physiological stimuli have been made.
Solomon and Shock (1950) tested the response of the adrenal
cortex in 27 young and 26 old men to a dose of adrenocortico-
tropic hormone (ACTH) and in 15 young and 13 old men to a
dose of 0-4 mg adrenaline. No difference in eosinopenia was
observed after ACTH, but adrenaline produced a significantly
greater eosinophil depression in the young group. From this it
was inferred that the senile cortex can still secrete 11-17 oxy-
steroids without gross impairment, but that the response of the
pituitary to acute adrenaline stimulation is lower in old than
young subjects. Pincus (1950) likewise found no impairment of
response to ACTH in old men compared with young controls.
But with chromatographic techniques, Rubin, Dorfman and
Pincus (1955) have examined the range of a- 17 ketosteroids
produced at different ages. There is here little difference between
men and women: in both, the greatest decline with age is in
androsterone and aetiocholan-3a-ol-17-one, and the second
greatest in the 5a- 11 -oxygenated steroids. Such studies offer
considerable promise.

The hypophysis is clearly the site of election for 'fundamental'
and all-explaining endocrine changes leading to senescence —
the part which it has played in the provision of such emotion-
ally-satisfying theories follows naturally from the fact that it is
known or credibly suspected to be involved in the regulation of
almost all mammalian processes of homoeostasis. Its proximity
to the hypothalamus enables it to be linked with theories which
locate senescence in the central nervous system. Hypophysial
factors in senescence have also a special importance because of
the relation of the hypophysis to the control of growth. In its
simplest form, starting from cellular exhaustion, the idea of a
primary pituitary senescence drew plausible anatomical argu-
ments from the histological studies of Parsons (1936) and Sim-
monds (1914) on long series of glands from subjects of various

174

The Mechanisms of Senescence

ages, or from more recent work such as that of Payne (1949,
1952) on the ageing fowl.

If there has been a tendency for the existence of the pituitary
gland to serve as a pretext for vagueness of thought concerning
the nature of senescence in mammals — the function formerly
discharged by the pineal in the search for the seat of the soul
— there are also solid arguments for its direct involvement,
certainly as mediator, but possibly also as originator, of senile
processes. Because the pituitary is profoundly concerned with
several processes of homoeostasis, and is involved with morpho-
genetic timing mechanisms like that which initiates puberty, it
is easy to develop hypotheses of pituitary senescence which do
not depend upon an unbiological argument in terms of single
hormones.

The function of trophic hormones appears to be the provision
of one limb of a system of negative feedback, by which the level
of effector-organ secretion is maintained and kept constant. If
senescence be regarded as a continuously self-aggravating dis-
equilibrium (a positive feedback process), then such a process
can be induced in a model control system, normally dependent
upon negative feedback, by several types of change.

Consider a system in which a device A produces a signal
which increases the activity of a second device B, and in which,
at the same time, the activity of B produces a signal which
reduces the activity of device A. The properties of most self-
regulating biological systems can be reproduced in this model
by varying the characteristic of the stimulus A — > B or the
negative feedback B — > A, and the number of stable states of
A or B. If there is no time-lag in either of these processes, the
level of output B will tend to be constant and self-restoring. If
there is an appreciable time-constant in one limb of the circuit,
the system will tend to function as a relaxation oscillator. In
this case, A stimulates B, which does not immediately respond.
Stimulus A — > B continues to increase, reaching a level which
corresponds to an ultimate response in B sufficient to inhibit A
completely. The output of B then declines, permitting A to
recover, and this represents one whole cycle. It is a requisite
for the functioning of such a system that the unmodified output
of A shall tend to increase in the absence of output B, i.e. that

175

The Biology of Senescence

the state of A is inherently unstable. If output A be assumed to
be the pituitary gonadotrophin and output B the gonadal
hormone, then the immediate response of the pituitary to
castration, or senile decline in gonadal response, is of the un-
stable type, though other mechanisms operate later to restore
regulation. Where, as may be the case in the male, the pituitary-
gonad balance operates as a level-control, gradual failure of
B's response would be expected to cause a gradual increase in
output A. In a similar system containing a time-constant, and
therefore behaving cyclically, blocking would be expected to
occur at one point in one particular cycle, with a proportion-
ately greater rise in output A.

The importance of this homoeostatic concept is that in the
simplified model self-regulating equilibrium can be turned to
progressive disequilibrium by several types of change. Decline
in the capacity of either A or B to respond to B — > A or A — > B
will result in a permanently unstable state of A. If the response
of A fails, B will also be driven into maximum output. Decline
in the capacity of B to produce B — > A will induce the un-
stable state of A. Decline in the capacity of A to produce A — > B
will lead to the relapse of B into its stable state of zero output.
This, however, where B's output effects other systems, will
cause disequilibrium in a complex physiology. In addition to
these purely quantitative changes, biological cybernetic mech-
anisms are also capable of exhibiting, and, in development,
characteristically do exhibit, qualitative change in the specifi-
city both of signal and response, which further complicates the
picture.

Although analogies from circuits oversimplifying the reality
of mammalian homoeostasis, they indicate the number of
variables to be considered in studying the senescence of homo-
eostatic systems; and they indicate some of the ways in which
such senescence can be analysed. Evidence suggests that in the
case of the gonad it is the function of B which declines. With
the pituitary trophic hormones, other than the gonadotrophins
and ACTH, in vivo estimation of levels, corresponding to the
measurement of output A, cannot as a rule be carried out. The
action of these homoeostatic systems, moreover, is not expressed
through static components but through a developing organism

176

The Mechanisms of Senescence

in which long-term trends in cell specificity are themselves
controlled by the signals involved in the homoeostatic process —
as if the system A ^ B were fitted in a vehicle whose move-
ment it controlled, but which travelled into a hotter and
hotter environment, thereby upsetting the characteristics of A
andB.

These highly complex hormonal homoeostatic systems are of
the greatest importance in higher vertebrates, although the
problem of 'three-dimensional' homoeostasis, or homoeostasis
superimposed on morphogenesis, is general in all developmental
physiology. The views of Minot (1908) upon 'cytomorphosis'
(differentiation and maturation) as a cause of senescence carry
the very important, and at first sight very probable, inference
that no complex organism, and certainly no vertebrate, can
remain in an indefinitely stable equilibrium. Where growth
processes and differentiation are superimposed on homoeostasis
they are analogous to 'drift' in a control system — on this basis
any system of differential growth must tend to increasing dis-
equilibrium, unless the developmental 'drift' itself modifies the
homoeostatic system to keep them appropriate to the altered
situation.

6-2-2 GONAD-PITUITARY SYSTEM

Senescence of the gonad regularly precedes or accompanies
senescence of the owner in a number of phyla — so much so that
declining reproductive capacity is a token of senescence almost
as valuable as the direct measurement of increased force of
mortality. The relation of gonadal senescence to somatic sen-
escence has clearly much evolutionary interest, since once the
first is complete, in an organism, to the point at which that
organism's contribution to posterity is no longer statistically
significant, any further adverse change in viability is generally
speaking inaccessible to the influence of natural selection, except
in a very roundabout way.

In almost all litter-bearing mammals, a decline in litter size
is characteristic of senescence. The long post-reproductive period
found in women is exceptional in mammals. It probably repre-
sents a genuine biological difference, quite apart from the far
greater perfection of the techniques for keeping human beings

177

The Biology of Senescence

alive. Compared with the more gradual disappearance of ova
in other mammals, the human menopause is unusually complete
and sudden. The range of this phenomenon among primates
is not at present known. Nobody has seen a postmenopausal
monkey (Krohn 1955).

There is some cause for regarding gonadal senescence and
the group of effects which follow senile gonadal withdrawal as
a separate 'senescence' from that of the animal as a whole, since
gonadal supplements can reverse a whole series of subsidiary
senile changes without materially reversing the progress of
somatic senescence judged by survival. In the castrated male
mammal, gonadal hormone supplements may perhaps actually
shorten life. There is no demonstrable androgen deficiency in
normal senile male rats (Korenchevsky, Paris and Benjamin,
1953), nor, probably, in most old men.

Although sex hormones do not rejuvenate the organism in the
manner envisaged by writers such as Voronoff, they produce a
closer approach to 'rejuvenation', covering more structures and
body processes, than do any other hormones which have been
investigated. In addition to their effects upon the secondary
sexual characters, such as beard growth (Chieffi, 1949) or
structure of vaginal epithelium (Loeb, 1944; Allen and Masters,
1948) and upon the genitalia themselves, androgens (Kenyon,
1942; Kochakian and Murlin, 1931; Kochakian, 1937) and
probably also oestrogens (Kenyon, 1942) exert an important
'anabolic' effect with nitrogen retention and increased protein
synthesis, and produce a number of unexpected peripheral
changes, generally in the direction of a restoration of 'young'
structure (Korenchevsky, Paris and Benjamin, 1953). Thus
oestrogens have been stated to produce a striking reversal of the
atrophy of the senile nasal mucosa (Kountz, 1950) and cer-
tainly produce extensive changes in senile skin, with dermal
regeneration (Goldzieher and Goldzieher, 1950; Chieffi, 1950a)
and restoration of elasticity (Chieffi, 1951). That the pos-
sibility of 'rejuvenation by replacement' is limited, even where
the reproductive organs are concerned, is shown by Kirk's
failure (1948, 1949) to restore the phosphatase activity of senile
prostatic secretion with androgens, although this activity is so
restored in young hypogonadal males.

178

The Mechanisms of Senescence

The relation of these changes and their reversal to the general
picture of senescence in mammals remains extremely obscure.
Of the more general processes in which gonadal hormone with-
drawal plays a part, few have been identified with certainty. It
has been suggested, for instance, that the osteoporosis of old age
is a result of the withdrawal of gonadal anabolic hormones
(Allbright, 1947). All theories of gonadal action in senescence
have, however, to accommodate the probability that in mammals
senile change in the force of mortality of apparently typical form, and at
approximately the typical specific age, occurs in both sexes in the absence
of both gonads, regardless of the age at which these are removed.

The mechanisms which fix the timing of puberty, and of the
human menopause, are the most obvious of all mammalian age
processes, and quite the most promising experimentally. The
key problem is to determine whether the timing-factors reside
primarily in the gonadal cells or elsewhere. The application of
transplantation techniques to this question has been reviewed
by Krohn (1955). Unfortunately, it already appears likely that
there are considerable interspecific differences. The ability of
the gonadal cells in situ to respond to gonadotrophin has some-
times been regarded as controlling the onset of puberty.
Pituitary gonadotrophin is detectable in the hypophysis of the
17 cm. pig foetus (Smith and Dortzbach, 1929) and the im-
plantation of glands from 3-month-old rabbits is as effective in
inducing puberty as implantation of adult glands (Saxton and
Greene, 1939). The ovaries of immature rabbits do not respond
to injected gonadotrophin (Hertz and Hisaw 1943, Parkes
1942-44, Adams 1953) On the other hand, it has long been
known that when ovaries are transplanted reciprocally be-
tween young and old animals, it is the age of the recipient
before or after puberty, not that of the ovary, which determines
function or non-function (Foa, 1900, 1901; Long and Evans,
1922) and in some species gonadotrophin readily induces pre-
cocious ovarian and testicular development. Domm (1934) was
able to induce crowing at 9 days of age and treading at 13 days
in cockerel chicks by injections of pituitary gonadotrophins.
The timing-mechanism is stable within a species or a genetic
strain. Human puberty very exceptionally occurs during child-
hood without any obvious pathological cause ('constitutional
n 179

The Biology of Senescence

precocious puberty' — Novak, 1944) and pregnancy has actually
been reported in a child of five (Escomel, 1939). In albino rats,
according to Mandl and Zuckerman (1952) genetic factors
seem to play the major part in determining the age of puberty.
Lorenz and Lerner (1946) likewise found clear evidence that
age of sexual maturation in turkeys is heritable. 'The reactivity
of the gonads may be the most important factor in determining
the time at which sexual maturity actually occurs, but the
factors which affect this reactivity are largely unknown' (Rob-
son, 1947). Change also takes place at puberty in the specificity
of pituitary response — while oestradiol induces pituitary and
adrenal hypertrophy in rats castrated after puberty, it reduces
pituitary and adrenal weight in animals castrated while still
immature (Selye and Albert, 1942).

The end of the reproductive period, as well as the beginning,
is marked by changes in gonadal reactivity. These have led
some writers to regard the human menopause as a form of
depletion-senescence (Swyer, 1954): Hertig (1944) describes the
exhaustion of a 'capital' of ova, which is not increased during
post-natal life, but his findings suggest that the actual meno-
pause precedes the end of all follicular activity. In man and
many other mammals (the only admitted exceptions occur in
Lemuroidea, the galago and the loris), the occurrence of oogen-
esis during post-pubertal life has been doubted — the case
against it has been persuasively put by Zuckerman (1951):
this 'perennial controversy' has been continued by Parkes and
Smith (1953), who found evidence of oocyte regeneration in
rat ovaries grafted after freezing. But it is in any case most
unlikely on existing evidence that the menopause occurs because
the supply of ova is exhausted. Engle (1944) mentions the find-
ing of apparently normal corpora lutea in women of 50: at the
menopause most if not all ova and follicles have normally dis-
appeared. Kurzrok and Smith (1938) found that in the human
ovary, in contrast to the senile ovary of some other mammals,
ova cease to be found, and that this change occurs at or soon
after the menopause. They consider that the postmenopausal
ovary can no longer respond to pituitary gonadotrophin.
Gardner (1952) transplanted ovaries between old and young
rats, and apparently found a greater tendency to malignant

180

The Mechanisms of Senescence

change in old ovaries carried by young hosts. Details of this
study have not yet appeared.

In rats semicastrated during old age, Wiesner (1932) found
a marked reduction in compensatory hypertrophy of the remain-
ing ovary. One major feature of senescence is probably pro-
gressive reduction of the ovarian reserve in terms of hormone
production. Such an effect, rather than the consumption of
ova, may account for the earlier menopause in semicastration
(Masters, 1952). Failure of the ovary to respond to pituitary
stimulation is almost certainly the proximate cause of the human
menopause. (See also Klebanow and Hegnauer, 1949.)

In rare instances a failure of the menopause itself has been
reported — menstruation is said to have persisted in a woman
of 104 (Novak, 1921) — but pathological causes were probably
responsible for some at least of these cases.

By far the most interesting fact from the standpoint of sen-
escence is the striking increase in pituitary gonadotrophin level
at the human menopause, and the comparable but more
gradual increase in senile men (Henderson and Rowlands,
1938) and rats (Lauson, Golden and Severinghaus, 1939). This
not only indicates a change in gonadal tissue reactivity with
age, but it also shows how limited senile processes may provoke
compensatory reactions and further disturb homoeostasis. Ova-
riectomy in certain strains of mice predisposes them to carcinoma
of the adrenal cortex, which can be prevented by oestrogens
(Woolley and Little, 1946). The senile increase in gonado-
trophin closely resembles that which follows castration, though
it develops more gradually (in female rats — Lauson, Golden
and Severinghaus, 1939). Witschi (1952) found that in women
the castrate level of FSH by pituitary gland assay is established
very rapidly after the menopause, and persists for the rest of
life, while in men the rise is far more gradual, the castrate level
being reached only at 70 years and only by a few individuals.
In a majority of cases the hypophyseal FSH content either
remains at the normal adult level or falls, occasionally even
below childhood levels. But in male eunuchs castrated in child-
hood, pituitary gonadotrophin output may remain, from the
time of normal puberty into middle age, at about ten times the
normal level (Hamilton, Catchpole and Hawke, 1944, 1945).

131

The Biology of Senescence

There is no evidence that this staggering increase, maintained
over 40-50 years, has any observable effect on the rate of ageing.

6-2-3 HORMONAL REGULATION OF GROWTH IN VERTEBRATES

In mammals, where growth and differentiation are difficult
to dissociate experimentally, we have abundant evidence of
senescence even in the longest-lived forms. In amphibia, where
there is a clear-cut metamorphosis, and where growth and
differentiation can be manipulated with relative ease, we have
so far no evidence of senescence. We cannot directly find out
whether the life of intact amphibia, the neoteny of the axolotl,
or the gigantism of athyroid tadpoles, ends in senescence, for
the practical reason that axolotls may well be capable of living
for 50, and normal frogs for twelve, fifteen or twenty years.
This conspiracy of circumstances perpetually recurs in the study
of ageing. The large literature of lower vertebrate endocrin-
ology and morphogenesis cannot be brought to bear on the
problem, for lack of actuarial data.

Both homoeotherms and poikilotherms, whether they meta-
morphose or not, tend to pass through an earlier phase of active
growth and a later phase of active reproduction, each char-
acterized by a separate type of endocrine control, and the
second by a relative loss of regenerative in favour of reproduc-
tive capacity. These phases are separated by the operation of a
timing-mechanism which is linked to processes in the juvenile
phase. In mammals, these phases are apparently controlled by
the pituitary growth and gonad-regulating mechanisms suc-
cessively. In lower vertebrates the differentiation-process and
the transition to adult function appears to depend on a pituitary-
thyroid balance. Pituitary growth hormone of mammalian
origin is able to promote the growth offish (Swift, 1954).

The relation between morphogenesis under the influence of
gonadal hormones and loss of regenerative power has special
interest in gerontology. Grobstein (1947) found that when the
gonopodium of poeciliid fish differentiates, under the influence
of androgens, regenerative power is lost: he stresses the analogy
between this process and the loss of regenerative capacity in
the developing anuran limb. Such a change need not depend
upon irreversible loss of cellular capacity to grow — this does

182

The Mechanisms of Senescence

not appear to be the case in amphibian limbs (Borssuk, 1935;
Polezaiev and Ginsburg, 1939) but the physiological loss of
repair-power may be as complete, so far as the intact animal
is concerned, as is the loss of moulting-power in Rhodnius once
the evocator is lost. There is clearly here, as Minot recognized,
a possible mechanism for the induction of senile change.

A certain amount of evidence is available concerning the
hormonal influences which affect protein anabolism, and regu-
late growth in mammals, especially in man. Where these factors
have been studied, they give little support to the idea of a
simple relation between senescence and growth-cessation, and
even less to the conception of a single, 'master', endocrine
inhibitor which can be detached from the general pattern of
progressing developmental change. The pattern which exists in
man has all the complication of a dynamic system where homo-
eostasis coexists with change. Much of the existing information
is provisional, and there are as yet no studies extending into
the period of senescence. It is plain, however, that in man,
and probably in some but not all other mammals, the 'anabolic'
stimulus to form new protein is not the same throughout life.
In adult life it is closely linked to the gonadal cycle. The
extent of the differences in endocrine control of growth between
determinate growers such as man and continual growers such
as the rat has not yet been mapped, and very little is known of
the hormonal control of growth in lower vertebrates. The exist-
ing evidence is quite enough, however, to render any static
conception of growth-cessation in terms of single-hormone
deficiencies untenable. A more accurate picture would perhaps
be obtained by treating pre-pubertal and pubertal life as separ-
ate instars separated by what amounts to a biochemical meta-
morphosis.

The growth of human beings, like that of Daphnia (p. 93),
occurs in two overlapping cycles — one prepuberal, the other
coinciding with puberty. (See Figs. 42, 43.) The prepuberal
cycle has its most active phase during the first six months of
life. This cycle, according to Kinsell (1953), is almost wholly
controlled by the pituitary growth hormone. The puberal cycle
appears to be evoked directly by anabolic steroid hormones
derived from the gonad and adrenal cortex. During both cycles

183

The Biology of Senescence

a minimum output of thyroid hormone is required to maintain
growth and development. At puberty, in response to the
pituitary gonadotropins, the gonads produce steroid hormones
which directly stimulate the growth of bone and of soft tissues.

Fig. 42. — The postnatal growth in weight of male children (kg/years)
(from the data of Quetelet) .

300 ~

"1 §

Puberty

■rj\ ' '

L_

n2\ _ 1
1 1 1 l 1 l 1

i i i i i i i i i

i i

2 3 t 5 6 ? 6 9 10 11 12 f3 14 15 IS 17 18 13 20 21 22

Age in Years from Birth

Fig. 43. — Annual growth increment in boys, from the data of Quetelet
krk4 = growth constants at each period (from Schmalhausen, 1928).

The process of bone growth in man is, however, self-limited,
since the same hormones produce skeletal maturation and
fusion of the epiphyses. There is reason to suspect that they
also inhibit the production of pituitary growth hormone—

184

The Mechanisms of Senescence

probably through a negative feedback system from the level of
protein anabolism. It is particularly interesting to notice that
acromegalic symptoms (McCullagh and Renshaw, 1934) or
frank gigantism (Joedicke, 1919) are occasional sequels to
castration in males — so, however, are polyuria and diabetes
insipidus (Hamilton, 1948). The puberal growth phase in girls
appears to be largely of adrenal origin, since the growth-
promoting effects of oestrogens and of progesterone are less
marked, except in the promotion of Ca and P04 retention,
than those of androgens (Kinsell, 1953). It is generally held that
thyroxin potentiates the action of pituitary growth hormone in
mammals (Evans, Simpson and Pencharz, 1939; Scow and
Marx, 1945) during the prepuberal phase, as well as hastening
differentiation. This does not appear to be the case in anurans,
where thiouracil produces pseudogigantism, and a balance
between thyroid and pituitary has been postulated (Steinmetz,
1954), one evoking differentiation and the other growth and a
'juvenile' condition.

This picture, which requires considerable amplification,
accords reasonably well with the known effects of various endo-
crine deficiencies in producing dwarfism or gigantism in man.
To some extent the appearance of the puberal cycle curtails the
prepuberal by inducing bone maturation. Epiphyseal union and
the change-over to the puberal phase of growth are delayed by
administration of growth hormone (Freud, Levie and Kroon,
1939), as they are in natural gigantism. On the other hand,
abolition of the whole gonadal influence by prepuberal castra-
tion has, at least, no gross effect on the life-span. Here again
the analogy to Edlen's findings in Daphnia is remarkably
close.

Various workers have suggested that mammalian senescence
'is' (or involves) the decline of growth hormone production,
and that it 'is' (or involves) the long-term effect of the pituitary
gonadotrophin on non-gonadal tissues. Insofar as senescence
results from differentiation, this is doubtless true, but the experi-
mental question is rather this — to what extent can the adminis-
tration of one or more 'anabolic' hormones affect the power of
continued homoeostasis in adult animals? It is possible that the
growth hormone itself may be the 'juvenile hormone' of the

185

The Biology of Senescence

mammalian pre-imaginal period. It is a primary stimulator of
protein anabolism and somatic growth. The change from a
protein-building and nitrogen-retaining economy, and the
negative specific acceleration of growth, are two of the most
evident correlates of senescence (Mayer, 1949). The adminis-
tration of growth hormone 'confers strangely youthful propor-
tions on the nitrogen, fat and water components of the body,
even in old animals' (Asling et al., 1952). Change in specificity
of tissue response to growth hormone certainly appears to occur
in some mammals, and this change coincides with the attain-
ment of maturity and the appearance of a fresh anabolism-
maintaining mechanism. The experimental work of Young in
England and Li in America suggests that before a critical time,
injected growth hormone induces only protein anabolism —
after that time, it also induces diabetes. This is the pattern in
man, the kitten (Cotes, Reid and Young, 1949) and the dog
(Campbell et al., 1950) but not in the rat (Bennett, Li and
Evans, 1948) or, apparently, the mouse (Moon et al., 1952)
which respond by continued growth. That the change in speci-
ficity involves endogenous as well as exogenous hormone is
evident from the occurrence of diabetes in association with
spontaneous acromegaly. Evidence for the existence of a separ-
ate diabetogenic principle is not very impressive (Raben and
Westermeyer, 1952; Young, 1953). Complete ablation of the
anterior lobe in adults leads to failure of growth but not, in
general, to other acceptable evidences of senility (in rats)
though this cannot be shown from the life-table.

In experimental studies, even highly purified growth hor-
mone administered to rats produces decreasing effects upon
nitrogen retention and upon growth after repeated administra-
tion (Whitney et al., 1948). These experiments, however, have
invariably been carried out with heterologous (usually ox)
hormones, and, as in the case of antigonadotrophic effects, no
physiological importance can be attached to the apparent
increase in tissue resistance.

Of the other hormones concerned in growth and differentia-
tion, the pituitary thyreotropic hormone appears in most mam-
mals which have been studied (rats — Turner and Cupps, 1938;
rabbits — Bergman and Turner, 1941; mice — Adams and

186

The Mechanisms of Senescence

Mothes, 1945; cattle — Reece and Turner, 1937) to reach a peak
at or about puberty, with a subsequent decline which has never
been followed by assay into old age. The decline of general
metabolism with increasing age, which has been frequently
linked with the decline of growth-capacity as an index of 'phy-
siological ageing', appears to involve both a fall in thyroid
activity and perhaps a decline in cell response, since thyroi-
dectomized rats show no senile decline in heart rate and 02
uptake, and old normal rats are decreasingly responsive to
thyroxin administration (Grad, 1953). The declining heat-
production of ageing human subjects may well be a reflection
as much of muscle atrophy as of thyroid involution.

The results obtained by McCay, using dietary restriction,
could be regarded as the consequences of dietary hypophy-
sectomy. Such a state of affairs interferes with the production
of both growth hormone and gonadotrophin, and its effect is a
general slowing of the 'integrating system' of growth + develop-
ment. The separation of these systems in mammals is a problem
of great interest and considerable practical difficulty. Dietary
retardation greatly postpones, but cannot be kept at such a
level as to prevent, the onset of oestrus (Asdell and Crowell,
1935). McCay, Sperling and Barnes (1943) found that the
capacity of retarded rats to resume growth was ultimately lost
if retardation was prolonged. Apparently if growth is delayed
without differentiation, it may ultimately encounter a block at
the cellular level.

A beginning has been made on the problem of selective inter-
ference with mammalian differentiation by the school of Li and
Evans (Walker et al., 1952; Asling et al., 1952a, b). Hypophy-
sectomy in 6-day-old rats does not arrest the eruption of teeth
or the opening of the eyes, but later sexual and pre-sexual
development is suppressed. Untreated animals ultimately die
from paralysis due to cerebral compression, the brain outgrow-
ing the cranium. Rats which survive the postoperative period
have been maintained in good health by growth hormone sup-
plements. In these supplemented rats, the rate of growth was
only slightly less than that of unoperated controls. Skeletal
development was normal, but adult organ-differentiation and
sexual maturation did not take place. Three such 'metathetelic'

187

The Biology of Senescence

individuals, were kept for 200-300 days in an attempt to pro-
duce gigantism. The end results of this experiment have not yet
been reported in full. It would be interesting to know how long
such animals are capable of living, and what senile changes
they ultimately exhibit.

Selective suppression of gonadotrophin production is not yet
feasible, though recent studies with Lithospermum extracts sug-
gest that the chemical 'dissection' of pituitary effects with
chemical antagonists is not, perhaps, an unreasonable hope
(Wiesner and Yudkin, 1952). The effect upon life-span of
inducing precocious puberty in mammals does not appear to
have been studied: mice, which mature very early, are not
ideal subjects, and an experiment on longer-lived mammals
encounters the familiar practical difficulties.

188

7

CONCLUSION

We have now briefly examined some of the evidence which
requires to be considered, and some of the questions which
require to be answered, in attempting to understand animal
senescence. We have seen, in particular, that many organisms
appear to have been provided by evolutionary selection with a
'programme' of development and function which is directional
and finite, and that progressive loss of the power to remain in
stable function occurs towards the end of that programme.
Weismann suggested that senescence is itself a functionally-
determined item in the programme: it seems more probable
that as the contribution of successive age groups to the next
generation of progeny is reduced by natural causes, so the
selection-pressure declines, and the efficiency of the homo-
eostatic mechanisms with it. The organism ultimately dies of
old age because it is now an unstable system which is pro-
vided with no further sequence of operational instructions, and
in which divergent processes are no longer co-ordinated to
maintain function.

In some cases the system fails suddenly, at a fixed point, after
the pattern of the senescence of rotifers or red blood corpuscles.
Some such cases apparently depend on the existence of cell
constituents renewable only by division. In mammals the
decline of resistance and the rise of the force of mortality are
gradual and smooth, and agree well with the probable shape
of a curve representing the declining efficiency of the evolu-
tionary pressure towards survival at different ages.

Insofar as any general theory of senescence is justified, this
seems at present the most plausible. It is probably as unprofit-
able to discuss the 'cause' of ageing as to discuss the 'cause' of

189

The Biology of Senescence

development. Senescence is a change in the behaviour of the
organism with age, which leads to a decreased power of survival
and adjustment. It is not a single overall process, except in the
evolutionary sense which we have outlined. Various factors in
varying proportions contribute to the senile change in different
species. Among these are the deterioration of irreplaceable
structures; the sum of previous injuries which are imperfectly
repaired; and progressive morphogenetic changes in the nature
and specificity of cell response and organ function. Any or all
of these factors may contribute to senescence in a given species.
Experimental removal of the factor which operates earliest in
the life-span may reveal another subsequent to it, and so on.
There is no conclusive evidence to incriminate cessation of
growth as a 'cause' of senescence, except in cases where cell
division ceases altogether. Senescence is not an 'inherent' pro-
perty of the metazoa, but one which they have on several occa-
sions acquired as a potentiality, probably through the opera-
tion of evolutionary forces directed to other biological ends.
In this respect the senescence of insects and of man is probably
a comparable process only to the same extent that the eyes of
these organisms are comparable structures. It is obvious that
such a conception, while it does not prevent us from ascertain-
ing what factors produce the age-deterioration in a given
species, excludes general physiopathological theories of the
'causation' of ageing as a whole.

Unlike the functional evolution of the eye, senescence is
typically an undirected process — not a part of the programme,
but a weakening of the directive force of the programme, an
escape from co-ordination, combined with the arrears of pro-
cesses which once contributed to fitness but are now running
free. Attempts to invest the programme of morphogenesis with
metaphysical or supra-natural properties (Driesch, 1941; Bur-
ger, 1954) have already been adequately answered by J. Need-
ham (1942), and need not be dealt with here. The idea of sen-
escence as the 'fated' or 'destined' end of the organism, i.e. a
positively-subsistent and ordered process of life-curtailment,
though it is not always the fruit of an avowed vitalism, has
much in common with it. Such treatment of senescence as an
evolved entity, and the idea that it must have developed as a

190

Conclusion

positive character, has almost certainly gained plausibility, like
so much else in the biological literature of old age, from human
preoccupations. The gerontologist, with the prolongation of
human life in mind, is interested in something which is not, as
such, of interest to the evolutionary 'demon', and whose evolu-
tion is in no sense comparable with the evolution of sight. Sen-
escence has no function — it is the subversion of function. On
the other hand, as Huxley (1942) suggests, the evolutionary
process in man has been transferred in the process of cephaliza-
tion from the 'demon' to the operation of conscious intellect.
It should now be possible in our thinking to separate human
goals from the effects of selection, and to renounce the animistic
confusion between them which has influenced the past theore-
ticians of old age. The whole conception of 'senescence', in fact,
belongs to the field of applied science. It embraces a group of
deteriorative effects which we have isolated because they are
deteriorative — in other words, because human beings dislike
them. Some biological thinkers have reduced themselves to
impotence in this field by the cultivation of philosophic doubt
whether senescence is an 'entity' at all. Viewed abstractly it is
not, any more than disease is an 'entity', but the same biologists
will certainly encounter, as they approach their seventieth year,
a sequence of changes which will kill them within a limited time.

Insofar as biology is more than a branch of idle curiosity, its
assignment in the study of old age is to devise if possible means
of keeping human beings alive in active health for a longer time
than would normally be the case — in other words, to prolong
individual life. People now rightly look to 'science' to provide
the practical realization of perennial human wishes which our
ancestors have failed to realize by magic — or at least to investi-
gate the prospect of realizing them. Under the influence of the
study which is necessary to fulfil such wishes, the character of
the wish itself generally changes in the direction of realism, so
that most people today would incline to prefer the prospect of
longevity, which may be realizable, to a physical immortality
which is not, and, pari passu, 'potentielle Unsterblichkeit' is al-
ready disappearing from the biological literature. An analogous
process can be seen in the psychology of individual growing-up.

The objective of prolonging human life is one which can bear

191

The Biology of Senescence

aggressive restatement from gerontologists, particularly at a
time when there are scientists who seek ethical reasons why
human life ought not to be prolonged, at least in communities
of which they are not themselves members. Although it has
much fundamental interest, we have seen that senescence is not
biologically speaking a very satisfactory entity. It appears in
most animals only under artificial conditions, and it would
probably seem to most of us pointless to devote great effort to
so arbitrary a part of development if it were not involved with
a primary human desire. As it is, medicine has always accepted
the prolongation of active and healthy human life in time as
one of its self-evident objects, and this object has only been
seriously challenged in the past two decades by the growth of
pathological forms of anti-liberalism. Gerontology differs from
other fields of medical biology only in the fact that while most
medical research is directed to making the curve of human sur-
vival as nearly as possible rectangular, gerontology is directed
to prolonging the rectangle, and shifting the point of decline
further in time from the origin. The applied character of such
work, and the object it has in mind, would not require emphasis
or defence at a period of culture when they ran no risk of pro-
voking a neo-Malthusian uproar. The beggarly opinions of such
writers as W. Vogt (1949) merit the rebuke of James Parkinson
(1755-1824), that 'if the population exceeded the means of sup-
port, the fault lay not in Nature, but in the ability of Politicians
to discover some latent defect in the laws respecting the division
and appropriation of property'. Postponement of old age, like
all the other advances in human control of environment, must
involve corresponding social adjustments: in the prevention of
presenile mortality, as the graphs in Fig. 3 abundantly indicate,
social, economic and political factors clearly predominate
already. But whatever problems might be raised by future
increases in the human specific age, in this and other fields
medicine can afford to treat protests based upon an interested
misreading of the biology of human societies with the contempt
they deserve, as a compound of illiberal opinions and bad
science. The emotional preoccupation of former workers with
magical rejuvenation did no good to the progress of science,
but it was at least a humane preoccupation.

192

Conclusion

In fact, the social correlates of longevity, which are probably
its most important practical aspects, have been omitted alto-
gether from consideration in this book. It is clear throughout
phylogeny that there is a relation between survival into the
senile period and the existence of a social mode of life. In some
cases longevity has evolved as a prerequisite of social organiza-
tion, in others social organization itself increases the possibility
of survival into old age, while the social group very probably
draws adaptive benefits from the existence of old individuals.
Both these trends appear to be at work in social primates. The
potential life-span in palaeolithic man probably resembled our
own: its realization has been possible through the development
of a complex social and rational behaviour. While therefore it
is legitimate to abstract the idea of an evolutionary programme
in morphogenetic or physiological terms when we discuss the
development and senescence of an individual man or of a
worker bee, in neither case is this 'programme' really detach-
able from the social programme which coexists with it, and
which plays an equally important part in the determination of
selection or survival. The irrelevance of discussing the biology
of individual animals, even of non-social species, divorced from
their ecology, has long been evident. Prolongation of the social
activity and significance of the individual human being almost
certainly leads to a change in the shape of the life-table, others
things being equal. Continuance of active work, retention of
interests, of the respect of our fellows, and of a sense of signi-
ficance in the common life of the species, apparently make us
live longer — loss of these things makes us die young. This is a
result we might have expected, but which we still largely ignore
in practice. How much of senile 'involution' is the effect of the
compulsory psychological and social 'winding-up' imposed on
the human individual by our form of society and our norms
for the behaviour of old people we do not yet know, but it is
certainly a very considerable part, and the most important
measures for the prolonging of useful individual life which come
within the range of the immediately practicable are all con-
cerned with social adjustment. The contrast between the place
of the (relatively few) aged in primitive societies (Simmonds,
1945, 1946) and the relatively many in our own (Sheldon, 1949;

193

The Biology of Senescence

Sanderson, 1949) is particularly striking. In primitive cultures
'important means of security for old people are their active
association with others and assistance in their interests and
enterprises. They are regarded as repositories of knowledge,
imparters of valuable information, and mediators between their
fellows and the fearful supernatural powers . . . The proportion
of the old who remain active, productive, and essential in
primitive societies is much higher than in advanced civilization,
for they succeed to an amazing degree in providing cultural
conditions which utilize the services of their few old people'
(Simmonds, 1946). How little this applies to our own culture
is evident from the studies of Sheldon (1949); other evidence
suggests that although in certain groups (Lehman, 1943), such
as amateur naturalists — or among those who retain, perhaps,
some of the magico-social functions of the primitive elder (poli-
ticians, judges and clergy) the element of social support based
upon continued activity leads to an apparently superior reten-
tion of the capacity for public life, the society of compulsory
retirement, individual privacy and the small family has little
to offer to old people. This is a topic which cannot be pursued
here, but its importance in the social medicine of age is para-
mount at present.

The problem of medical gerontology at the biological level,
however, is to prolong the human life cycle in time, either by
deformation and stretching or by addition, and in particular to
prolong that part of it which contains the period of 'adult
vigour'. Such a problem could theoretically be solved in any of
three ways, bearing in mind the evidence regarding the exist-
ence of a developmental 'programme' — that programme could
be prolonged by the provision of new developmental operations;
or its movement, throughout or in part, could be slowed down;
or active life could be maintained after the expiry of the pro-
gramme by piecemeal adjustment of homoestatic mechanisms
with supplements, medicaments and prostheses of various kinds.

The first of these possibilities, though it is biologically the
most interesting, does not merit discussion at present, at any
rate in relation to man. We do not know enough about morpho-
genesis to interfere with it clinically, except in a few simple
deficiency states, let alone devise and apply a sequence of self-

194

Conclusion

regulating operations in growth or development subsequent to
normal adulthood. The third possibility is that which is already,
and very justifiably, receiving the major part of the energy
devoted to clinical studies upon human ageing. The removal
of successive causes of death should in theory increase the specific
age. It is interesting to notice that there is so far very little
evidence of such an effect. It was once widely believed that
with the removal of 'pathological' causes of death the specific
age would rise very rapidly in man and approach the recorded
maximum. Although research on the diseases of later life has
not yet had time to reach a stature comparable with the life-
saving powers of surgery and epidemiology in early life, it
seems possible that even with increased control over neoplasms
and cardiovascular disease the age at death might only come
to be more and more normally distributed about the present
specific age. Most people who die in old age are found at post-
mortem to have several further pathological processes at work,
beside the one from which they died. If an arbitrary normal
distribution of deaths due to 'pure' senescence were assumed,
the shape of the curve could be inferred from the shape of that
one-half of the existing, positively-skew curve which lies between
the ages of 75 and 100. Such an assumption may well, however,
be false. The smoothness of the curve between those ages is
largely due to manipulation by actuaries, relying on a conven-
tional end to the human survival curve to help them over
statistically insignificant material (Freudenberg, 1949). We
have not reached the limits of the purely prosthetic and sup-
portive treatment of senility, and we are unlikely to do so for
a very long time.

There remains the possibility of prolonging life by slowing
down the movement of the developmental programme. Such
slowing, to be of medical interest, must be compatible with
retention of normal vigour and activity. The mere duration of
individual existence could quite possibly be extended, as John
Hunter proposed, by some form of artificial hibernation,
punctuated by periods of activity: but this possibility is more
interesting to stockbreeders, astronauts and perhaps politicians
than to gerontologists. The prolongation of infancy is unlikely
to be of much interest, unless it is indispensable to some effect
o 195

The Biology of Senescence

in later life. The prolongation of that part of the life cycle
which lies normally between the eighteenth and fiftieth years
of age, even by a small percentage, would come nearest to
fulfilling our objectives. The prolongation of childhood pre-
sents special interest, since, on existing mammalian evidence,
it is quite probably feasible by relatively simple, if heroic,
means such as calorie-restriction. Such a prolongation would have
interesting and far-reaching social consequences, both upon the
family and upon the acquisition of skills: its effects on character-
structure would be even more interesting. It seems very prob-
able that the time-scale of prepubertal development in the
majority of mammals is relatively labile. There is so far no
direct evidence that the same applies to later stages in the life
cycle. Dietary restriction, although it may favour longevity,
does not greatly delay senescence in adult mammals, and it may-
be that the time scale of the adult period, after somatic growth
has ceased, is not susceptible to any major interference without
at the same time destroying normal function. To assess the
possibilities of such interference we require to know how far
'marking time' at each stage of the mammalian developmental
programme is possible, and, if possible, is compatible with
functional health. It is also a matter of practical import whether
the rate of child growth influences the length of the period of
adult vigour in man (Sinclair, 1955). The degree of linkage
between growth, development, and metabolism may vary con-
siderably at different periods of the life cycle, and the bulk of
the work upon their separation has been carried out only in
non-mammalian embryos and larvae. We have to reckon with
the possibility that the post-pubertal mammal behaves like an
imago — that its life-span is closely linked to metabolism, which,
in homoeotherms, is virtually invariable by the methods which
can affect it in invertebrates, and that the fundamental change
which leads to eventual senescence has already taken place at
puberty. In this case, interference with the length of the adult
phase could only be prosthetic.

There remains the possibility that a substantial change in the
specific age, and in the duration of healthy life, might result
from one particular adjustment. This was the hope which led to
the use of sex hormones for purposes of 'rejuvenation', and

196

Conclusion

which was largely disappointed. If such an adjustment is pos-
sible, it is most likely, perhaps, to concern one or more of the
anabolism-promoting substances which maintain growth in the
young animal.

To the question 'Can the effective human life-span be pro-
longed artificially?' the most probable answer, based on all
these possibilities, would appear to be 'Yes'. To the further
question 'By what factor?' no meaningful answer can be given
until we know more of the nature of the predominant processes
which determine human senescence. Supplementary questions
dealing with the degree of reversibility in established senile
change cannot at present be answered at all, beyond the con-
jecture that the morphogenetic programme in man is hardly
likely to be simply reversible in any fundamental sense, but that
the irreversibility of local changes in ageing is at present prob-
ably over- rather than under-estimated.

The only excuse for such speculation is, in any case, the pos-
sibility that it will drive us into the laboratory to ascertain the
facts and to answer the questions it raises, thereby removing
gerontology from the field of 'entelechies' and 'inherent prin-
ciples' into that of intelligible evidence.

In planning research upon a subject such as senescence, it
pays to put to ourselves the questions which most urgently
require to be answered, and then to select from the list in the
likely order of practicability. From the defects which exist in
the evidence, the essential preliminary questions concerning
senescence which we must ask appear to be these.

(1) Does senescence occur in all vertebrates?

(2) In what instances, if any, among vertebrates does sen-
escence coexist with continuing somatic growth, or non-
senescence with fixed size?

(3) To what extent is arrest of developmental processes com-
patible, at different stages of vertebrate ontogeny between
conception and senility, with activity and normal func-
tion? Is the retardation of further development, towards
senility, realizable after puberty in mammals?

(4) To what extent is the artificial induction of somatic
growth possible in the mature vertebrate, and what is its
effect on the specific age?

197

The Biology of Senescence

(5) What are the limits of the power of enzyme-renewal and
physicochemical self-maintenance in fixed post- mitotic
cells?

(6) In the case of single organs such as the mammalian ovary,
how far is the expectation of life of the organ intrinsic in
its stage of development, and what is its life-span in pas-
sage through successive young hosts? What, in more
general terms, is the relation between the physiological
age of individual tissues and the chronological age of the
host animal?

I have singled out these specific questions for attention as
being rather unlikely to be solved incidentally, in the course of
general biological research upon other topics. The great bulk
of the information which is missing on other specific points is
likely to be derived eventually from studies in endocrinology
or morphogenetics which are not undertaken ad hoc; this type
of background research cannot be hurried on, ahead of the
general progress of knowledge, except by the cultivation of
interest in ageing among biologists of all kinds.

Three main types of research are involved in the investiga-
tion of our six preliminary questions — study of the phylogeny
of senescence in vertebrates, study of the correlations and the
experimental modification of growth and development in popu-
lations where the life-span can be concurrently measured, and
study of tissue-environment relationships through the creation
of age chimaeras. The problems of the first of these studies have
already been mentioned. A reliable test of 'senescence' which
correlates with the decline of resistance, does not kill the indi-
vidual animal, and can be related to actuarial senescence by an
intelligible process of reasoning, might offer some solution. The
development of such a test would probably depend, however,
upon the establishment of the part which declining growth-
energy plays in the process of ageing. The time-lag in explant
growth might conceivably give a basis for some such attempt.
Any method of marking tissue cells in situ, to enable their life-
span to be determined like that of red blood corpuscles, would
be a highly desirable advance, and a key to many doors. The
study of growth and development relations, and the whole
group of studies which require to be undertaken in determining

198

Conclusion

the factors which predominate in mammalian ageing, or which
can modify it, encounter a rather different obstacle. The choice
of experimental animals for such work obviously presents great
difficulty, since it is necessary either to work on forms whose
life-span is short compared with that of the investigator, or to
use elderly individuals whose early history has not been fol-
lowed. The complication which this time-factor introduces is
of great importance for the planning of research. Man is by far
the most numerous senile animal, and his life cycle is extremely
well known — even to the point at which we can estimate his
physiological age by inspection; some research on senile men
can be justified ethically, but the gerontological aspects of
laboratory animal-breeding cannot much longer be neglected,
since in many problems no further progress is possible until
mammals of known life cycle, heredity and physiology are
available in quantity. At present the choice lies between experi-
ment on patients, the basing of general conclusions upon the
behaviour of invertebrates and small rodents, and postponing
investigation for several years while a chosen population of
larger mammals completes its life cycle. Failure to deal with
the logistics of this problem now will hinder research in ten
or twenty years time, and that hindrance could be avoided by
forethought.

Another extremely important source of information is likely
to be found in the creation of age chimaeras. In spite of the well-
merited disrepute incurred by much work upon organ grafting
in relation to old age, the resources of modern transplantation
techniques now offer a very tempting range of experimental
possibilities. The reimplantation of stored infant tissues into the
same animal, the cross- transplantation of organs such as ovaries,
or tissues such as portions of skin, and the observation of the
reciprocal influences of host and implant, are technically prac-
ticable and established procedures. Parabiosis between animals
of disparate age but identical genotype, or even between
retarded and unretarded littermates, is another tool of the same
kind. The emphasis of such research is likely to be chiefly
directed to dissecting tissue and somatic factors in the senile
process, but its possible use in experimental, and even later in
clinical, prosthetics is thoroughly well-justified. This is, of

199

The Biology of Senescence

course, not the first occasion in biology when a procedure which
was attempted with enthusiasm and abandoned in disgust has
come back into useful currency after a period of meditation and
study. An equally productive field may well be that of organ
culture, initiated by Carrel, but still comparatively little used.
Apart from such specialized investigations, serious progress
depends on the cultivation of general awareness among bio-
logists of the importance of prolonging their study of every
animal into the senile period, of collecting and publishing life-
tables, especially for cold-blooded vertebrates under good
laboratory conditions, and of seeking confirmatory evidence of
the distribution of senescence in phylogeny. A few years of
propaganda to zoologists in training might bring in a rich
factual harvest later. Much modern research into ageing tends
to be desultory, although the single subjects with which it deals
are important in themselves. We ought to try to devise critical
experiments, and if we destroy more hypotheses than we demon-
strate, this is a subject which can well stand such treatment in
contrast to the speculation which has gone before. The most
desirable condition for progress in gerontology at the moment
is that the exact nature and scope of the problems raised by
senescence should be understood, and the possibility of new
experimental evidence borne in mind, during the planning and
assessment of all biological research, even when it is primarily
directed to other objects. Senescence, like Mount Everest,
challenges our ingenuity by the fact that it is there, and the
focusing of our attention on it is unlikely to be fruitless.

200

BIBLIOGRAPHY

NO TE: Russian references are given under the initial of the author in
English transliteration. Where an alternative title in a language other than
Russian is given by the author, this title is cited, otherwise the title is given
in Russian, followed by an English translation.

Italic numbers in parenthesis following the reference refer to the page on
which that reference is cited. Some additional papers by cited authors are
included without a page reference.

Abe, N. (1932). 'The age and growth of the limpet (Acmaea dorsuosa Gould).'

Sri. Rep. Tohoku Univ. 7, 347. (58, 105)
Adams, A. E., and Mothes, A. M. (1945). 'The thyrotrophic potency of

the pituitaries of albino mice with respect to age and sex.' Anat. Rec. 91,

21. (186)
Adams, C. E. (1953) in Ciba Foundation Symposium — Mammalian Germ

Cells, p. 198. London: Churchill. (179)
Agduhr, E. (1939). 'Internal secretion and resistance to injurious factors.'

Acta medica scand., 99, 387. (133)
Agduhr, E., and Barron, D. H. (1938). 'Further observations on the

increased resistance of mated animals to toxic agents: Medinal.' Arch. int.

Pharmacodyn., 58, 351. (133)
Ahrens, K. (1938). 'Lokaler Nachweis von Kalzium in den Membranen

des £7o^<z-blattes mittels Natriumoleat.' Protoplasma, 31, 508. (35)
Allbright, F. (1947). 'Osteoporosis.' Ann. int. Med., 27, 861. (179)
Allee, W. C.j Emerson, A. E., Park, O., Park, T., and Schmidt, K. P.

(1949). Principles of animal ecology. Philadelphia: W. B. Saunders Co.

(20)
Allen, W. M., and Masters, W. H. (1948). 'Investigation of sexual

rejuvenation in elderly women.' Tenth Conference on Problems of Aging.

Josiah Macey, Jr. Foundation, 1948. (178)
Alpatov, W. W. (1930). 'Experimental studies on the duration of life.

XIII. The influence of different feeding during the larval and imaginal

stages on the duration of life in the imago of Drosophila melanogaster.'

Amer. Nat., 64, 37. (44, 143)
Alpatov, V. V., and Gordeenko, N. A. (1932). AjinaTOB, B.B. h TopffeeHKO,

H.A. 'EjiHHHiie cnapHBamiH Ha npo,n;oji>KHTejiHOCTB >kh3hh

mejiKOnpHflOB.' (Influence of mating on the longevity of silk-moths.)

Zool.ZK 11. (2), 60. (44,132)

201

Bibliography

Alpatov, W. W., and Pearl, R. (1929). 'Experimental studies on the

duration of life. XII. Influence of temperature during the larval period

and adult life on the duration of life in the imago of Drosophila melano-

gaster: Amer. Nat., 63, 37. (44, 130, 133, 143)
Anderson, B. G., and Jenkins, J. C. (1942). 'A time study of events in the

life span of Daphnia magna: Biol. Bull. Woods Hole, 83, 260. (44, 93, 145)
Andres, A., and Jiv, B. V. (1936). 'Somatic chromosome complex of human

embryos.' Cytologia, 7, 371. (168)
Andrew, W. (1953). 'Phenomena of abnormal nuclear division in relation

to the ageing process. J. Geront., 8, 372. (168)

(1955). 'Amitotic division in senile tissues as a probable means of self-
preservation in cells.' J. Geront., 10, 1. (100, 163)
Annandale, N., and Sewell, R. B. S. (1921). 'The banded pond-snail of

India — Vivipara bengalensis.' Rec. Indian Mus., 22, 215. (102)
Arndt, W. (1928). 'Lebensdauer, Altern und Tod der Schwamme.' S.B.

Ges. naturf. Fr., Berl, 23, 44. (80)

(1941). 'Lebendbeobachtungen an Kiesel- und Horn-Schwammen des

Berliner Aquarium.' £ool. Gdrt., Lpz., 13, 140. (56)
Asdell, S. A., and Crowell, M. F. (1935). 'The effect of retarded growth

upon the sexual development of rats.' J. Nutrit., 10, 13. (187)
Ashworth,J. H., and Annandale, N. (1904). 'Observations on some aged

specimens of Sagartia troglodytes and on the duration of life in Coel-

enterates.' Proc. Roy. Soc. Edin., 25, 295. (55, 56, 81)
Asling, C. W., Moon, H. D., Bennett, L. L., and Evans, H. M. (1952a).

'Relation of the anterior hypophysis to problems of ageing.' J. Gerontol.,

7, 292. (153, 187)
Asling, C. W., Walker, D. G., Simpson, M. E., Li, C. H., and Evans,

H. M. (1952b). 'Deaths in rats submitted to hypophysectomy at an

extremely early age and the survival effected by growth hormone.' Anat.

Rec., 114,49. (153, 187)

Bab, H. (1948). 'The process of ageing.' Brit. med. J., i, 1000. (7, 45, 162)
Bagelar, A., and Frade, F. (1933). 'Sur la longevite chez les araignees.'

C. R. Soc. Biol., 113, 523. (57)
Backman, G. (1938). 'Wachtumszyklen und phylogenetische Entwicklung.'
Lunds Univ. Arskrift, N.F. Avd., 2, 34, Nr. 5. (53)

(1944). 'Der Lebenslauf der Organismen nebst kritischenBetrachtungen
zu meiner Wachtumstheorie.' Z- Alter sf or sch., 4, 237.

(1945). Altern und Lebensdauer der Organismen. (Uppsala and Stockholm).
(61)
Bacon, F. (1645). Historia Vitae et Mortis. Dillingen. (34, 45, 54, 79)
Baer, K. (1864). Reden. St. Petersburg. (8)

Baerg, W. J. (1945). 'The black widow and the tarantula.' Trans. Conn.
Acad. Arts Sci., 36, 99. (56, 92)

202

Bibliography

Baffoni, G. M. (1954). 'La citomorfosi degli elementi di Purkinje del

cervelletto.' Ricerc. scient., Rome, 24, 1641. (163)
Baily, J. L. (1931). 'Some data on growth, longevity and fecundity in

Limnaea columella Say.' Biol, gen., 7, 407. (43, 103, 104)
Baker, F. (1934). 'A conchological Rip van Winkle.' Nautilus, Phil., 48,

5. (58, 105)
Baker, L. E., and Carrel, A. (1926). ' Au sujet du pouvoir inhibiteur du

serum pendant la vieillesse.' C.R. Soc. Biol., Paris, 95, 958. (154)
Ball, J. P., and Squire, J. R. (1949). 'A study of mortality in a burns unit.'

Ann. Surg., 130, 160. (26)
Banfield, W. G. (1952). 'The solubility and swelling of collagen in dilute

acid with age variations in man.' Anat. Rec, 114, 157. (169)
Banta, A. M. (1914). 'One hundred parthenogenetic generations ofDaphnia

without sexual forms. Proc. Soc. exp. Biol., N.T., 11, 7. (118)
Banta, A. M., and Wood, T. R. (1937). 'The accumulation of recessive

physiological mutations during long-continued parthenogenesis.' Genetics,

22, 183. (118)
Barnes, L. L. (1942). 'The deposition of calcium in the hearts and kidneys

of rats in relation to age, source of calcium, exercise and diet.' Am. J. Path.,

18,41. (150)
Barrett-Hamilton, G. E. H. (1911). History of British Mammals. London.

(49)
Bauer, E. (1924). 'Beitrage zum Studium der Protoplasmahysteresis, etc'

II. Arch. Mikr. Anat., 101, 483. VIII. Arch. mikr. Anat., 101, 521. (7)
Bazilevitch, I. V. (1938a). Ba3i«ieBHH, H.B.: 'Chhapom HopMajiHon

CTapocTii.' (The syndrome of normal old age.) Proc. Conf. Probl. Old Age,

Kiev, 255. (61)
Bazilevitch, I. W. (1938b). 'Sur 1'a.ge des vieillards centenaires d'Ab-

khasya.' Medits. Zh.f 8, 7. (61)
Bazykalova, A. (1934). Ea3MKajioBa, A.: 'Bo3pacr h TeMn pocTa Pecten

jessoensis Jay.' (Age and growth-rate of P. j.) Bull. Acad. Sci. U.R.S.S.,

2-3, 389. (58)
Beers, C. D. (1929). 'On the possibility of indefinite reproduction in the

ciliate Didinium without conjugation or endomixis.' Amer. Nat., 63, 125.

(117)
Beeton, M., and Pearson, K. (1901). 'On the inheritance of the duration

of life, and on the intensity of natural selection in man.' Biometrika, 1, 50.

(122)
Belar, K. (1924). 'Untersuchungen an Actinophrys sol Ehrenburg. II.

Beitrage zur Physiologie des Formwechsels.' Arch. Protistenk., 48, 371.

(117)
Bellamy, A. W. (1934). 'Life span of Platypoecilus, Xiphophorus and their

hybrids in the laboratory.' Science, 80, 191. (72, 133)
Bender, A. E. (1953). 'Recent advances in protein synthesis.' Lancet, 265,

1142. (34)

203

Bibliography

Benedict, F. G. (1935). 'Old age and basal metabolism.' New Engl. J. med.,

212, 1111. (33)
Benedict, F. G., and Root, H. F. (1934). 'The potentialities of extreme

old age.' Proc. nat. Acad. Set., 20, 389. (33)
Benedict, F. G., and Sherman, H. C. (1937). 'Basal metabolism of rats

in relation to old age and exercise during old age.' J. JVutrit., 14, 179.

(33)
Bergauer, V. (1924). 'Beitrage zum Studium der Protoplasmahysteresis,

etc.' III. Arch. mikr. Anat., 101, 489. VI. Arch. mikr. Anat., 101, 508.

VII. Arch. mikr. Anat., 101, 512. (7)
Bergman, A. J., and Turner, C. W. (1941). 'Thyrotropic hormone content

of rabbit pituitary during growth.' Endocrinology, 29, 313. (186)
Berninger, J. (1910). 'Uber Einwirkung des Hungers auf Hydra.' £ool.

Anz.,36, 271. (81)
Berrill, N. J. (1931). 'The natural history of Bulla hydatis Linn.' J. mar.

biol. Ass. U.K., 17, 567. (58)
Bertalannfy, L. von (1933). Modern theories of development. London:

Humphrey Milford. (142)

(1941). 'Stofrwechseltypen und Wachtumstypen.' Biol. ££/., 61, 510.

(32, 142)
Bidder, G. P. (1925). 'The mortality of plaice.' Nature, Lond., 115, 495.

(12)

(1932). 'Senescence.' Brit. med. J., ii, 5831. (8, 11, 13, 70, 130, 154)
Bilewicz, S. (1953). 'Doswiadczenia nad wplywem czynnosci rozrodczych

na dlugosc zycia u muchy owocowej Drosophila rnelanogaster.' (Influence of

mating on the longevity of D. m.) Folia Biol. Warsaw, 1, 175. (44, 97, 99,

130, 133)
Billingham, R. E., Brent, L., and Medawar, P. B. (1953). 'Actively

acquired tolerance of foreign cells.' Nature, Lond., 172, 603 (170)
Bishopp, F. C., and Smith, C. N. (1938). 'The American dog tick, eastern

carrier of Rocky Mountain spotted fever.' Circ. U.S. Dep. Agric., No. 478,

1. (143)
Bittner, J. J. (1937). 'Mammary tumours in mice in relation to nursing.'

Amer. J. Cancer, 30, 530. (121)
Blumenthal, H. (1945). 'The ageing process in the endocrine glands of the

guinea pig. I. The influence of age, sex and pregnancy on the mitotic

activity and the histological structure of the thyroid, parathyroid and

adrenal glands.' Arch. Path., 40, 284. (167, 173).
Blungk, H. (1924). 'Lebensdauer, Fortpflanzungsvermogen und Alters-

erscheinungen beim Gelbrand (Dytiscus marginalis L.).' <W. Anz., 58,

163. (57, 97, 130)
Bodenheimer, F. S. (1938). Problems of animal ecology. Oxford University

Press. (18, 24, 44)
Bodenstein, D. (1943a). 'Factors influencing growth and metamorphosis

of the salivary gland in Drosophila.' Biol. Bull. Wood's Hole, 84, 13. (148)

204

Bibliography

Bodenstein, D. (1943b). 'Hormones and tissue competence in the

development of Drosophila.' Biol. Bull. Wood's Hole, 84, 35. (148)

(1953). 'Endocrine control of metamorphosis.' Proc. IX Int. Cong.

Entom. Amsterdam, 1951. (148)
Boecker, E. (1914). 'Depression und Missbildungen bei Hydra.' Z°°l- Anz.,

44, 75. (81)
Bogomolets, A. A. (1947). The prolongation of life. Duell Sloan and Pearce.

New York, 1946. (7)
Bonnet, P. (1935). 'La longevite chez les araignees. Bull. Soc. ent. Fr., 40,

272. (56, 57)
Boothby, W. M., Berkson, J., and Dunn, H. L. (1936). 'Studies of the

energy metabolism of normal individuals.' Amer. J. Physiol., 116, 468.

(33)
Borisenko, E. Ya. (1939). EoppiceHKO, E.H.: 'BjiHHHue vcjiobM pa3BHTHH

Ha nocjieftCTBMH HH6pii,n;HHra.' (Influence of conditions of development on

the effects of inbreeding) . Tarovizatsiya, 5-6, 162. (129)

(1941). *K Bonpocy 06 HHSpiiAHHre b >khbothoboactbg.' (On inbreeding

in animal husbandry.) Trud. sel.-Khoz. Acad. Timiryazeva, 5, 5. (129)
Borssuk, R. A. (1935). 'Untersuchung des Verlustes der Regenerations-

fahigkeit der hinteren Extremitat von Rana temporaria.' Arch. EntwMeck.

Org., 133, 349. (182)
Bounhiol, J. (1938). 'Recherches experimentales sur la determinisme de la

metamorphose chez les Lepidopteres.' Biol. Bull. Suppl., 24, 1. (147)
Bourliere, F. (1946). 'Longevite moyenne et longevite maximum chez les

vertebres.' Annie Biol., Paris, 22, 10. (53)

(1947). 'Quelques remarques sur la longevite des petits mammiferes

sauvages.' Mammalia, 11, 111. (47)

(1950). 'Senescence et vitesse de cicatrisation chez le rat.' Rev. med.

Likge, 5, 669 (26, 156)

(1951). Vie et moeurs des mammiferes. Payot, Paris. (48)
Bowerman, W. G. (1939). 'Centenarians.' Trans. Actuarial Soc. Amer., 40,

360. (61)
Boycott, A. E. (1936). 'Habits of freshwater mollusca in Britain.' J. Anim.

Ecol.,5, 116. (102)
Bresslau, E. (1928-33). Turbellaria: in Kiikenthal, W., and Krumbach,

T. Handbuch der Zoologie, II. (56, 84)
Brien, P. (1953). 'La perennite somatique.' Biol. Rev., 28, 308. (82,

163)
Brockmeier, H. (1896). 'Beitrage zur Biologie unsere Susswassermollusken.'

Nchrbl. dtsch. malak. Ges., 57, (58)
Brody, S. (1924). 'The kinetics of senescence.' J. gen. Physiol, 6, 245. (8)

(1945). Bioenergetics and Growth. Baltimore. (63)
Brody, S., Ragsdale, A. C, and Turner, C. W. (1923). 'The rate of

growth of the dairy cow. IV. Growth and senescence as measured by the

size and fall of milk secretion with age.' J. gen. Physiol., 6, 31. (26)

205

Bibliography

Brown, G. W., and Flood, M. M. (1947). 'Tumbler mortality.' J. Amer.

Statist. Ass., 42, 562. (22, 23)
Brown, M., Sinclair, R. G., Cronk, L. B., and Clark, G. C. (1948).

'Some remarks on premature aging in the Eskimos.' Rev. Canad. Biol.,

7, 178. (134)
Bruce, H. M., and Hindle, E. (1934). 'The golden hamster, Cricetus

(Mesocricetus) auratus Waterhouse. Notes on its breeding and growth.'

Proc. zool. Soc. Lond., p. 364. (47)
Bruch, H. (1941). 'Obesity in relation to puberty.' J. Pediat., 19, 365.

(152)
Buchanan, J. W. (1938). 'Developmental acceleration following inhibi-
tion.' J. exp. Zool, 79, 109. (156)
Bucher, N. L. R., and Glinos, A. D. (1950). 'The effect of age on the

regeneration of rat liver.' Cancer Res., 10, 324. (155)
Bucher, N. L. R., Scott, J. F., and Aub, J. C. (1950). 'Regeneration of

liver in parabiotic rats.' Cancer Res., 10, 207. (156)
Bull, H. O. (1934). 'Aquarium observations on the rate of growth and

enemies of the common starfish (Asterias rubens L.).' Rep. Dove Marine Lab.,

3rd series, 2, 60. (58)

(1938). 'The growth of Psammechinus miliaris (Gml) under aquarium

conditions.' Rep. Dove Marine Lab., 6, 39. (58)
Bullough, W. S. (1949). 'Age and mitotic activity in the male mouse, Mus

musculus L.' J. exp. Biol, 26, 261. (156)
Burger, M. (1954). Altern und Krankheit. 2. Aufl. Leipzig: Georg Thiene.

(4, 8, 84, 156, 190)
Burt, W. H. (1940). 'Territorial behavior and populations of some small

mammals in southern Michigan.' Misc. Publ. Mus. Z°°l Univ. Mich., 45,

1-58. (109)
Butschli, O. (1882). 'Gedanke iiber Leben und Tod.' Z°°l Anz., 5, 64.

(164)
Buxton, J. (1950). The redstart. London: Collins. (109)

Cagle, F. R. (1946). 'Growth of the slider turtle, Pseudemys scripta elegans.'

Amer. Midi Nat., 36, 685. (77)
Calkins, G. N. (1919). lUroleptus mobilis Eng. II. Renewal of vitality

through conjugation.' (116)
Camboue, P. (1926). 'Prolongation de la vie chez les papillons decapites.'

C.R. Soc. Biol, Paris, 183, 372. (99)
Campbell, J., Davidson, I. W. F., Snair, W. D., and Lei, H. P. (1950).

'Diabetogenic effect of purified growth hormone.' Endocrinology, 46, 273.

(186)
Campbell, P. N., and Work, T. S. (1953). 'Biosynthesis of proteins.'

Nature, Lond., 171, 997. (169)
Carlson, A. J., and Hoelzel, F. (1947). 'Growth and longevity of rats fed

omnivorous and vegetarian diets.' J. Nutrit., 34, 81.

206

Bibliography

Carlson, A. J. and Hoelzel, F. (1948). 'Prolongation of the life span of

rats by bulk formers in the diet.' J. Nutrit., 36, 27.

(1946). 'Apparent prolongation of the life-span of rats by intermittent

fasting.' (151)
Carr, C. J., King, J. T., and Vissgher, M. B. (1949). 'Delay of senescence

infertility by dietary restriction.' Feder. Proc, 8, 22. (151)
Carrel, A. (1912). 'On the permanent life of tissues outside the organism.'

J. exp. Med, 15, 516. (9, 10)
Carrel, A., and Eberling, A. H. (1921). 'Antagonistic growth principles

of serum and their relation to old age.' J. exp. Med., 38, 419. (8, 10, 154,

156)
Castle, W. E. (1929). 'A further study of size inheritance in rabbits, with

special reference to the existence of genes for size characters.' J. exp.

Zool, 53, 421.
Chamberlain, T. K. (1933). 'Age and shell measurements of two large

specimens of Megalonaias gigantea Barnes.' Nautilus, 47, 29. (58, 107)
Chang, H. S. (1951). 'Age and growth of Callionymus lyra L.' J. mar. Biol.

Ass. U.K., 30, 281. (69)
Cheymol, J., and Pelou, A. (1944). 'La respiration du muscle du lapin

suivant l'age.' C.R. Soc. Biol., Paris, 138, 91. (32)
Chieffi, M. (1949). 'Effect of testosterone administration on the beard

growth of elderly males.' J. Gerontol., 4, 200. (178)

(1950). 'An investigation of the effects of parenteral and topical

administration of steroids on the elastic properties of senile skin.' J.

Gerontol., 5, 17. (178)

(1951). 'The effect of topical oestrogen application on the elastic

properties of the skin of elderly men.' J. Gerontol., 5, 387. (178)
Child, C. M. (1911). 'A study of senescence and rejuvenation based on

experiments with planarians.' Arch. EntwMeck. Org., 31, 537. (83)

(1913). 'The asexual cycle in Planaria velata in relation to senescence

and rejuvenescence.' Biol. Bull. Wood's Hole, 25, 181. (83)

(1914). 'Asexual breeding and prevention of senescence in Planaria

velata.' Biol. Bull. Wood's Hole, 26, 286. (83)

(1915). Senescence and rejuvenescence. Chicago: University of Chicago

Press. (9, 10, 32, 81, 83, 115)

(1918). 'Physiological senescence in hydromedusae.' Biol. Bull. Wood's

Hole, 34, 49. (32, 81, 83)
Cholodkowsky, N. (1882). 'Tod und Unsterblichkeit in der Tierwelt.'

Zool. AnZ., 5, 264. (8)
Christopherson, J. B. (1924). 'Longevity of parasitic worms: the term of

living existence of Schistosoma haematobium in the human body.' Lancet,

p. 742. (56)
Chu, J. (1934). 'Reproduction, life span, growth and senescence of Brachi-

onus.' Sci. Rep. Univ. Chekiang, No. 1. (59)

207

Bibliography

Clark, T. B. (1940). 'The relation of production and egg weight to age

in White Leghorn fowls.' Poultry Sci., 14, 54. (26, 63)
Clarke, J. M., and Maynard Smith, J. (1955). 'The genetics and cytology

of Drosophila subolsana. XI. Hybrid vigour and longevity.' J. Genet., 53,

172. (128)
Clarke, M. F., and Smith, A. (1938). 'Recovery following suppression of

growth in the rat.' J. Nutrit., 15, 245. (156)
Clemente, C. D., and Windle, W. F. (1955). 'Regeneration of severed

nerve fibres in the spinal cord of the adult cat.' J. compar. Neurol., 101,

691. (163)
Cleveland, L. R. (1938). 'Longitudinal and transverse division in two

closely-related flagellates.' Biol. Bull. Wood's Hole, 74, 1. (116)
Cohn, A. E., and Murray, H. A. (1925). 'The negative acceleration of

growth with age, as demonstrated by tissue culture.' J. exp. Med., 42,

275. (31, 155)
Coker, R. E., Shira, A. F., Clark, H. W., and Howard, A. D. (1919-20).

'Natural history and propagation of the freshwater mussels.' Bull. U.S.

Bureau Fish, 37, 75. (107)
Collinge, W. E. (1944). 'Notes on the terrestrial Isopoda (Woodlice).'

Northw.Nat., 19, 112. (57)
Comfort, A. (1953). 'Absence of a Lansing effect in Drosophila subobscura.'

Nature, Lond., 172, 83. (125)

(1954). 'Biological aspects of senescence.' Biol. Revs., 29, 284. (4)
(1955). 'Maximum ages reached by domestic cats.' J. Mammal., 34,

(4) in press. (47, 49)
Conant, R., and Hudson, R. G. (1949). 'Longevity records for reptiles

and amphibia in the Philadelphia Zoological Garden.' Herpetologia, San

Diego, 5, 1-8. (51, 52)
Cori, C. I. (1925). 'Morphologie und Biologie von Apsilus vorax.' Z- w^ss-

Zool, 125, 557. (59, 86, 88)
Cotes, P. M., Reid, E., and Young, F. G. (1949). 'Diabetogenic action of

pure anterior pituitary growth hormone.' Nature, Lond., 164, 209. (186)
Coutelen, F. (1935). 'La longevite de la filaire Loa loa (Guyet 1778) et des

embryons de filaires.' Bull. Soc. Path, exot., 28, 126. (56)
Coutelen, F., Razemon, P., and Biguet, J. (1950). 'La longevite des

echinocoques — etude critique.' Ann. Parasitol., 25, 267. (56)
Cowdry, E. V. (1952). In Lansing, A.I., Problems of Ageing. Baltimore:

Williams and Wilkins Co. (35)
Crabb, E. D. (1929). 'Growth of a pond snail, Limnaea stagnalis appressa, as

indicated by increase in shell size.' Biol. Bull. Wood's Hole, 56, 41. (103)
Crocker, W. (1939). 'Ageing in plants.' In Cowdry, E. V., Problems of

Ageing, 1st edn. (4, 120)
Crowell, S. (1953). 'The regression-replacement cycle of hydranths of

Obelia and Campamdaria? Physiol. <W., 26, 319. (82)
Cuenot, L. (1911). La genise des espkes animates. Paris. (58, 105)

208

Bibliography

Dalyell, J. G. (1848). Rare and remarkable animals of Scotland, ii, ch. 10.

London. (56, 81)
Daranyi, G. (1930). 'Fejlodes, fajfenntartas es 6regedes a termeszetben.'

Term. Tud. KozL, 62, 305. (7)
Darlington, G. D. (1948) .'The plasmagene theory of the origin of cancer.'

Brit. J. Cancer, 2, 118. (120)
Darlington, G. D., and Mather, K., 1949). The elements of genetics.

London: Allen and Unwin. (120)
Darwin, G. (1874). The descent of man. London: John Murray. (ISO)
Dathe, H. (1935). Zool Garten. Lpz., 7, 303 (79)
Davaine, C. (1887). Traite des Entozoaires et des maladies vermineuses de Vhomme

et des animaux domestiqaes. 2nd edn., Paris: Bailliere. (56)
David, K. (1925). 'Zur Frage der potentiellen Unsterblichkeit der Meta-

zoen.' Zool. Anz., 64, 126. (82)
Dawidoff, C. (1924). 'Sur le retour d'une Nemerte, Lineus lacteus, en

inanition, a un etat embryonnaire.' C.R. Acad. Sci., Paris, 179, 1222. (84)
Deansley, R. (1938). 'The reproductive cycle of the golden hamster

(Cricetus auratus).' Proc. Zool. Soc. Lond., p. 31. (47)
Deck, R. S. (1936). 'Longevity of Terrapene Carolina Linn.' Copeia, 160,

179. (52)
Deevey, E. S. (1947). 'Life tables for natural populations of animals.'

Quart. Rev. Biol, 22, 283. (70)
Deevey, G. B., and Deevey, E. S., Jr. (1945). 'A life table for the black

widow.' Trans. Conn. Acad. Arts Sci., 36, 115. (4, 44, 130, 131)
Delage, Y. (1903). Uheredite et les grandes probl ernes de la biologic Paris. (8)
Demange, E. (1886). Etudes cliniques et anatomopathologiques de la vieillesse.

Paris: Bailliere. (7)
Denffer, D. V. (1948). 'Uber einen Wachstumshemmstoff in alternden

Diatomeenkulturen.' Biol. ££/., 67, 7-13. (154)
De Witt, R. M. (1954). 'Reproductive capacity in a pulmonate snail,

Physagyrina Say.' Amer. Nat., 88, 159. (58, 103)
Dhar, N. H. (1932). 'Senescence, an inherent property of animal cells.'

Quart. Rev. Biol, 7, 70. (7)
Didlake, M. L. (1937). In Needham, J. G. (Ed.). Culture methods for inverte-
brate animals. New York: Comstock. p. 244. (56)
Ditmars, R. L. (1934). 'A review of the box turtles.' Zoologica, N.Y., 17,

1. (51)
Dobers, E. (1915). 'Biologie der Bdelloidea.' Int. Rev. Ges. Hydrobiol.,

Ser. 6, suppl. 7. (59, 87)
Domm, L. V. (1934). 'The precocious development of sexual characters in

the male chick by daily injections of thebin.' Anat. Rec, 58, 6. (179)
Donaldson, H. H. (1924). 'The Rat.' Mem. Wistar Inst. Philadelphia, No. 6.

(47)
Donisthorpe, H. (1936). 'The oldest insect on record.' Ent. Rec, 48, 1. (57)

209

Bibliography

Dowdeswell, W. H., Fisher, R. A., and Ford, E. B. (1940). 'The quanti-
tative study of populations in the Lepidoptera. I. Polyommatus icarus Rott.'
Ann. Eugen., Camb., 10, 123. {112)

Draper, C. C, and Davidson, G. (1935). 'A new method of estimating
the survival rate of anopheline mosquitoes in nature.' Nature, Lond., 172,
503.

Dribben, I. S., and Wolfe, J. M. (1947). 'Structural changes in the con-
nective tissue of the adrenal glands of female rats associated with
advancing age.' Anat. Rec, 98, 557. (173)

Driesch, H. (1941). 'Zur Problematik des Alterns.' Z- Alter sforsch., 3,
26. (8, 190)

Dublin, L. I., Lotka, A. J., and Spiegelman, M. (1949). Length of Life:
a study of the life table. New York: Ronald Press Co. {21, 112, 123)

Duetz, G. H. (1938). 'Comments on longevity and average exhibition ages
during the year 1937.' Lab. Rep. zool. Soc. Philadelphia, 66, 31. (46, 47)

(1939). 'Revised tables of maximum exhibition periods for animals in
the Philadelphia Collection.' Ibid., 67, 22. (46)
(1940). Ibid., 68,26. (46)
(1942). Ibid., 70, 23. (46, 47)

Dunham, H. H. (1938). 'Abundant feeding followed by restricted feeding
and longevity in Daphnia.' Physiol. Z°6l-> 11 1 399. (44, 145)

Dunn, C. W. (1946). 'Endocrines in senescence.' Clinics, Phil, 5, 847. (7)

Duran-Reynals, F. (1940). 'Neutralisation of tumour viruses by the blood
of normal fowls of different ages.' Tale J. Biol. Med., 13, 61. (169)

Edlen, A. (1937). 'Experimentelle Wachstumstudien an Daphnia magna.1

Lunds Univ. Arsberdtt (Avd. 2) 34, 1. (32)

(1938). 'Geburt, Geschlechtsreife und Vermehrung in Beziehung zum

Wachstumsverlauf bei Daphnia magna.' Arch. EntwMeck., 137, 804. (32,

93, 94. 95)
Edmonds, T. R. (1832). 'Life tables founded upon the discovery of a

numerical law, etc' London: J. Duncan. (152)
Edmondson, W. T. (1945a). 'Ecological studies of sessile Rotatoria.' Ecol.

Monogr., 14, 15. (44)

(1945b). 'Ecological studies of sessile Rotatoria. II.' Ecol. Monogr., 15,

141. (44, 59,113)
Edney, J. M., and Allen, W. R., (1951). 'Age of the box turtle, Terrapene

Carolina Carolina L.' Copeia, 312, 644. (51, 52)
Ejiri, J. (1936). 'Studien uber die Histologic der menschlichen Haut. II.

Uber die Alters und Geschlechtsverschiedenheiten der elastischen

Fasern.' Jap. J. Derm. Urol., 40, 173. (169)
Elman, R. (1953). 'Surgical problems in the aged.' In Cowdry, E. V.

Problems of Ageing, 2nd edn., 1953. (156)
Elton, C. (1942). Voles, mice and lemmings. Problems in population dynamics.

Oxford: Clarendon Press. (109)

210

Bibliography

Engle, E. T. (1944). 'The menopause, an introduction.' J. clin. Endocrin.,

4, 567. (180)
Escomel, E. (1939). 'La plus jeune mere du monde.' Pr. med., 47, 875.

(179)
Evans, H. M., Simpson, M. E., and Li, C. H. (1948). 'The gigantism pro-
duced in normal rats by injection of the pituitary growth hormone. I.

Body growth and organ changes.' Growth, 12, 15. (152)
Evans, H. M., Simpson, M. E., and Pengharz, R. I. (1939). 'Relation

between the growth-producing effects of the pituitary and the thyroid

hormone.' Endocrinology, 25, 175. (161, 185)
Evans, R., Cowdry, E. V., and Nielson, P. E. (1943). 'Ageing of human

skin.' Anat. Rec, 86, 545. (26)

Farner, D. S. (1945). 'Age groups and longevity in the American robin.'

Wilson Bull., 57, 56. (109)
Faure-Fremiet, E. (1953). 'L'hypothese de la senescence et les cycles de
reorganisation nucleaire chez les Cilies.' Rev. Suisse Zjool., 60, 426. (118)
Federley, H. (1929). Uber subletale und disharmonische Chromosomen-

kombinationen.' Hereditas, 12, 271. (132)
Felin, F. E. (1951). 'Growth characteristics of the poeciliid fish, Platy-

poecilus maculatus? Copeia, p. 15. (69)
Ferris, J. C. (1932). 'Comparison of the life histories of mictic and amictic
females in the rotifer, Hydatina senta.' Biol. Bull. Wood's Hole, 63, 442.
(59, 66)
Findley, T. (1949). 'Role of neurohypophysis in the pathogenesis of
hypertension and some allied disorders associated with ageing.' Amer.
J. Med., 7, 70. (7)
Fisgher-Piette, E. (1939). 'Sur la croissance et la longevite de Patella
vulgata L. en fonction du milieu.' J. Conchyl., Paris, 83, 303. (58, 106,
107, 108, 112)
Fisher, I. (1923). Report on national vitality, its wastes and conservation.

Washington. (61)
Fitzinger, L. J. F. J. (1853). Versuch einer Geschichte der Menagerien des

Osterreichisch-Kaiserlichen Hofes. Vienna. (49, 64)
Flower, S. S. (1922). 'Longevity of molluscs.' Cairo Sci. J., 10, 115. (58)
(1925). 'Contributions to our knowledge of the duration of life in
vertebrate animals.' (46)

T. Fishes.' Proc. zool. Soc. Lond., 247. (70)
'II. Batrachians.' Ibid., 269. (52)
'III.Reptiles.' Ibid., 911. (45, 52)
TV. Birds.' Ibid., 1365. (49)
(1931). 'V. Mammals.' Proc. zool. Soc. Lond., 145. (45, 46, 48)
(1935). 'Further notes on the duration of life in animals.'

T. Fishes.' Prac. zool Soc. Lond., 265. (46, 53, 68)
(1936). TI. Amphibians.' Proc. zool. Soc. Lond., 369. (46, 52)
P 211

Bibliography

Flower, S. S. (1937). 'III. Reptiles.' Proc. zool. Soc. Lond., 1. (45, 46, 52,

78)

(1938). 'IV. Birds.' Proc. zool. Soc. Lond., A, 195. (46, 49, 64)

(1945). 'Persistent growth in the tortoise, Testudo graeca, for 39 years,

with other notes concerning the species.' Proc. zool. Soc. Lond., 114, 451.

(45)
Foa, C. (1900). 'La greffe des ovaires en relation avec quelques questions

de biologie generate.' Arch. ital. Biol., 34, 43. (179)

(1901). 'Sur la greffe des ovaires.' Arch. ital. Biol, 35, 364. (179)
Forster, A. (1945). 'Longevity.' Brit. med. J., ii, 545. (60)
Foster, T. D. (1932). 'Observations on the life history of a fingernail shell

of the genus Sphaerium.' J. Morph., 53, 473. (103)

(1936). 'Size of shell in land snails of the genus Polygyra with particular

reference to major and minor varieties.' Amer. Midi Nat., 17, 978.

(104)
Freud, J., Levie, L. H., and Kroon, D. B. (1939). 'Observations on growth

(chorionotrophic) hormone and localization of its point of attack.' J.

Endocrin., 1, 56. (185)
Freud, J., and Uyldert, E. (1947). 'A new idea about senility.' Acta brev.

need. Physiol, 14, 18. (169)
Freudenberg, K. (1951). 'Die naturliche Lebensdauer des Menschen.'

Z. Altersforsch., 5, 241. (60, 195)
Friedel, E. (1880). 'Die lebenden Wasserthiere auf der Internationalen

Fischerei-Ausstellung zu Berlin im Jahre 1880.' Zool Gart., 21, 323. (57)
Fritsch, R. H. (1953). 'Die Lebensdauer von Daphnia spec, bei verscheidener

Ernahrung usw.' Z- w^ss- Z°ol-> 157, 35. (92, 145)
Frohawk, F. W. (1935). 'Feeding butterflies in captivity.' Entomologist, 68,

184. (57, 98)
Frost, W. E. (1943). 'The natural history of the minnow, Phoxinus phoxinus.*

J. Anim. EcoL, 12, 139. (70)

(1954). 'The food of pike, Esox lucius L., in Windermere.' J. Anim.

EcoL, 23, 339. (71)
Frost, W. E., and Smyly, W.J. P. (1952). 'The brown trout of a moorland

fishpond.' J. Anim. EcoL, 21, 71. (70, 71)

Gaillard, P. J. (1942). Hormones regulating growth and development in embryonic

explants. Paris: Hermann. (170)
Galea, P. H. (1936). 'Longevity of a mule.' Field, 1936, p. 1556. (48)
Gardner, G., and Hurst, H. (1933). 'Life-tables for white Leghorn

chickens in the State of Utah.' Proc. Utah Acad. Sci., 10, 149. (43)
Gardner, T. S. (1946). 'The effect of yeast nucleic acid on the survival

time of 600 day old albino mice.' J. Gerontol, 1, 445.
Gardner, T. S., and Forbes, F. B. (1946). 'The effect of sodium thio-

cyanate and yeast nucleic acid on the survival time of 700 day old albino

mice.' J. Gerontol, 1, 453.

212

Bibliography

Gardner, W. K. (1952). 'Some endocrinological aspects of ageing.' J.

Gerontol, 7, 293. (180)
Gates, W. H. (1926). 'The Japanese waltzing mouse.' Publ. Carneg. Instn.,

337, 83. (127)
Geiser, S. W. (1924-5). 'The differential death-rate of the sexes among

animals.' Wash, Univ. Stud., 12, 73. (130)
Georgiana, M. (1949). 'Longevity of the parasitic wasp, Habrobracon

juglandis Ashmead.' Amer. Nat., 83, 39. (7, 130)
Gey, G. O. (1952). 'Cellular gerontologic research.' J. Gerontol., 7, 294.

(171)
Geyer, D. (1909). Die Weichtiere Deutschlands. Stuttgart. (58, 107)
Gley, E. (1922). 'Senescence et endocrinologie.' Bull. Acad. Med. Paris, 87,

285. (7)
Glezina, O. M. (1939). 'Age changes in oxidation-reduction processes in

the muscle tissue of birds.' Biokhim. %h., 13, 105. (32)
Glinos, A. D., and Bartlett, E. G. (1951). 'The effect of regeneration on

the growth potential in vitro of rat liver at different ages.' Cane. Res., 1 1 ,

164. (155)
Glinos, A. D., and Gey, G. O. (1952). 'Humoral factors involved in the

induction of liver regeneration in the rat.' Proc. Soc. exp. Biol. N.T., 80,

421. (156)
Glover, J. W. (1921). U.S. life tables, 1890, 1901, 1910 and 1901-10, p. 301

(U.S. Bureau of Census).
Goetsch, W. (1922). 'Lebensdauer und Geschlechtlige Fortpflanzung

bei Hydra.' Biol. ZbL, 42, 231. (82)

(1925). 'Beitrage zum Unsterblichkeitsproblem der Metazoen. V.'

Biol. ZbL, 45, 192. (82)

(1940). Vergleichende Biologie der Insektenstaaten. Leipzig. (57)
Goldschmidt, J., Hoffman, R., and Doljansky, L. (1937). 'fitude com-
parative sur la duree de la periode de latence pour la croissance des

tissus embryonnaires et adultes explantes in vitro'. C.R. Soc. biol., Paris,

126, 389.
Goldsmith, E. D. (1942). 'Sexuality in Dugesia trigrina (syn. Planaria

maculata): Nature, 150, 351. (56)
Goldsmith, E. D., Nigrelli, R. F., Gordon, R. S. Charipper, H. A., and

Gordon, M. (1944). 'Effect of thiourea upon fish development.' Endo-
crinol., 35, 132. (161)
Goldzieher, J. W. (1949). 'The direct effect of steroids on the senile

human skin.' J. Gerontol., 4, 104.
Goldzieher, J. W., and Goldzieher, M. A. (1950). 'Effect of steroids on

the ageing skin.' J. Gerontol., 5, 385. (178)
Gonzalez, B. M. (1923). 'Experimental studies on the duration of life.

VIII. The influence upon duration of life of certain mutant genes of

Drosophila melanogaster.' Amer. Nat., 57, 289. (44, 121)

213

Bibliography

Gould, G. M., and Pyle, W. T. (1898). Anomalies and curiosities of medicine.

London: Rebman. (62)
Gould, R. T. (1945). Enigmas. London: Bles (2nd edn.). (62)
Gowen, J. W. (1931). 'On chromosome balance as a factor in duration of

life.' J. gen. Physiol, 14, 447. (130)

(1934). 'The gene in pathology.' Cold Spr. Harb. Symp. quant. Biol., 2,

128. (130)
Grad, B. (1953). 'Changes in oxygen consumption and heart rate of rats

during growth and ageing; role of the thyroid gland.' Amer. J. Physiol.,

174, 481. (187)
Grave, B. H. (1928). 'Natural history of the shipworm, Teredo navalis, at

Wood's Hole, Mass.' Biol Bull, 55, 260. (103)
Green, J. (1954). 'Size and reproduction in Daphnia magna.' Proc. zool. Soc,

Lond., p. 535. (127)
Greene, R., and Paterson, A. S. (1943). 'Sudden senescence.' Lancet, ii,

158. (135)
Greenwood, A. W. (1932). 'The value of progeny in relation to age of

dam.' Harper- Adams Util. Poult. J., 17, 478. (124)
Greenwood, M., and Irwin, J. O. (1939). 'The biostatistics of senility.'

Hum. Biol, 11, 1. (60)
Greville, T. N. E. (1946). United States life tables and actuarial tables, 1939-

41. Bureau of the Census, Washington. (40)
Griffin, C. E. (1928). 'The life history of automobiles.' Michigan Business

Studies (Univ. Mich.), vol. 1. (33)
Griffiths, J. T., and Tauber, O. E. (1942). 'Fecundity, longevity and

parthenogensis of the American roach (Periplaneta americana L.).' Physiol

Zool, 15, 196. (44, 99)
Grimm, H. (1949). 'Wachstumsfordernde und wachstumshemmende Stoffe

im menschlichen Blutserum.' Z> Alter sf or sch., 5, 197. (155)
Grobstein, C. (1947). 'The role of androgen in the declining regenerative

capacity during morphogenesis of the Platypoecilus maculatus gonopodium.'

J. exp. Zool, 106, 313. (182)
Gross, J. (1925). 'Versuche und Beobachtungen iiber die Biologie der

Hydriden.' Biol Zool, 45, 192. (82)
Gross, J., and Sghmitt, F. O. (1948). 'The structure of human skin collagen

as studied with the electron microscope.' J. exp. Med., 88, 555. (169)
Gross, J., and Sghmitt, F. O. (1950). 'Connective tissue fine structure and

some methods for its analysis.' J. Geront., 5, 343. (169)
Gruneberg, H. (1951). The genetics of the mouse. The Hague. (43, 121,
122)

(1954). 'Variation in inbred lines of mice.' Nature, Lond., 173, 674.
(136)
Gudernatsch, J. F. (1912). 'Feeding experiments on tadpoles. I. The

influence of certain organs given as food on differentiation.' Arch.

EntwMeck. Org., 35, 57. (139)

214

Bibliography

Guilbert, H. R., and Goss, H. (1932). 'Some effects of restricted protein
intake on the oestrous cycle and gestation in rats.' J. Nutrit., 5, 215. (151)

Gumbell, E. J. (1938). 'La duree extreme de la vie humaine.' Actualites Sci.
Indust., 520, 1. (63)

Gurney, J. H. (1899). 'On the comparative ages to which birds live.' Ibis,
p. 19-42. (49)

Guthrie, D. M. (1953). Personal communication. (113)

Haemmerling, J. (1924). 'Die ungeschlechtliche Fortpflanzung und

Regeneration bei Aeolosoma lemprichii.' £ool. Jb. (1. Abt), 41, 581. (83)
von Hagen, W. (1938). 'Contribution to the biology of Nasutitermes s.s.'

Proc. zool. Soc. Lond., 108, A, 39. (57)
Hakh, I. W. D., and Westling, E. H. (1934). 'A possible cause of old age.'

tew*?, 79, 231. (7)
Haldane, J. B. S. (1941). New Paths in Genetics. London. (14, 39)

(1949). 'Paternal and fraternal correlations of fitness.' Attn. Genet., 14,

288. (123)

(1953). 'Some animal life-tables.' J. Inst. Actu., 79, 351. (19, 24)
Hall, G. O., and Marble, D. R. (1931). 'The relationship between the first

year egg production and the egg production of later years.' Poultry Sci.,

10, 194.
Hamai, I. (1937). 'Some notes on relative growth, with special reference to

the growth of limpets.' Sc. Rep. Tohoku Imp. Univ. Biol, 12, 71. (106)
Hamburger, C. (1948). 'Normal urinary excretion of neutral 17-keto-

steroids with special reference to age and sex variations.' Acta endocrinol.,

1, 19. (172)
Hamilton, J. B. (1948). 'The role of testicular secretions as indicated by the

effects of castration in man and by studies of pathological conditions and

the short life span associated with maleness.' Recent Prog. Hormone Res.,

3, 257. (133, 172, 173, 185)
Hamilton, J. B., Catghpole, H. R., and Hawke, C. G. (1944). 'Titres of

urinary gonadotropins in old eunuchs.' Anat. Rec, 88, 435. (181)

(1945). 'Titres of gonadotropins in urine of aged eunuchs.' J. din.

Endocrinol., 5, 203.
Hamilton, J. B., and Hamilton, H. B. (1948). 'Ageing in apparently nor-
mal men. I. Urinary titres of ketosteroids and of a and /^-hydroxy keto-

steroids.' J. Clin. Endocrinol., 8, 433.
Hamilton, J. B., Hamilton, H. B., and Mestler, G. E. (1954). 'Ageing in

apparently normal men. II. Androgenic activity of urinary ketosteroids

and of their alpha and beta fractions.' J. din. Endoc. Metab., 14, 139.

(172)
Hamilton, W. J. (1940). 'The biology of the smoky shrew (Sorex fumeus

fumeus Miller).' Zoologica, N.T., 25, 473.
Hammond, J., and Marshall, F. H. (1952). 'The life cycle.' In Marshall,

Physiology of Reproduction. London: Longmans. (46)

215

Bibliography

Hansard, S. L., Comar, C. L., and Davis, G. K. (1954). 'Effects of age

upon the physiological behaviour of calcium in cattle.' Amer. J. Physiol.,

177, 383. (169)
Hansemann, D. v. (1914). 'liber Alterserscheinungen bei Bacillus rossii.'

Sitzb. Nat. Fr. Berlin, 187-191. (100)
Harms, J. W. (1926). Verjungung des Lebens. Senckenberg-Biicher II. Berlin:

Bermuhler.

(1949). 'Altern und Somatod der Zellverbandstiere.' Z* Alter sf or sch., 5,

73. (4, 84, 92, 96)
Harms, W. (1912). 'Beobachtungen iiber den naturlichen Tod der Tiere.

I. Hydroides pectinata.' £ool. Anz., 40, 117.
Harris, J. A., and Benedict, F. G. (1921). 'A biometric study of basal

metabolism in man.' Carnegie Inst. Wash. Publ., 303,
Hartlaub, C. (1916). 'liber das Altern einer Kolonie von Syncoryne.' Wiss.

Meersuntersuch., 11. (82)
Hartmann, M. (1921). 'Untersuchungen iiber die Morphologie und

Physiologie des Formwechsels der Phytomonadinen ( Volvocales) . III.

Mitt. Die dauernd agame Zucht von Eudorina elegans: experimentelle

Beitrage zum Befruchtungs- und Todproblem.' Arch. Protistenk., 43, 7.

(117)
Hartzell, A. (1945). 'Thiourea (thiocarbamide) — adult life span feeding

experiments in rats.' Conts. Boyce Thompson Inst., 13, 501. (153)
Harvey, P. A. In Kofoid, C. A. (1934). Termites and termite control, p. 227.

Univ. California Press. (140)
Hase, A. (1909). 'Uber die deutschen Susswasserpolypen Hydra fusca.' Arch.

f. Rassen- u. Gesellschafts-Biologie., 6, 721. (82)
Hasek, M. (1953). 'BereTaTHBHan rn6pjmii3aijHH huibothmx nyreM

coejjHHeHHH KpoBOo6pam;eHHH b TeHemie 3M6pnoHajibHoro pa3BHTHH.'

(Vegetative hybridization in animals through embryo parabiosis.)

Cesk. Biol., 2, 267. (129)
Haskins, H. H. (1954). 'Age determination in molluscs.' Trans. N.Y. Acad.

Sci., 16, 300. (107)
Heght, S. (1916). 'Form and growth in fishes.' J. Morph., 27, 379. (68)
Heilbrunn, L. V. (1943). An outline of general physiology. 2nd edn. Phila-
delphia and London: Saunders.
Henderson, W. R., and Rowlands, I. W. (1938). 'The gonadotrophic

activity of the anterior pituitary gland in relation to incteased intra-
cranial pressure.' Brit. med. J., i, 1094. (181)
Herdan, G. (1952). 'Causes of excess male mortality in man.' Acta genet.

Basel. 3, 351. (133)
Herrigk, F. H. (1898). 'The American lobster.' Bull. U.S. Fish Comm.,

15, 1. (57, 92)

(1911). 'Natural history of the American lobster.' Bull. U.S. Bureau of

Fisheries, 29, 149. (57)

216

Bibliography

Hertig, A. T. (1944). 'The ageing ovary — a preliminary note.' J. din.

Endocrin., 4, 581. (180)
Hertwig, R. (1906). 'Uber Knospung und Geschlechtentwicklung von

Hydra fusca: Biol. ZbL, 26, 489. (81)
Hertz, R., and Hisaw, F. L. (1934). 'Effects of follicle-stimulating and

luteinizing pituitary extracts on the ovaries of the infantile and juvenile

rabbit.' Amer. J. Physiol., 108, 1. (179)
Hervey, G. F., and Hems, J. (1948). The goldfish. London: Batchworth

Press. (73, 74)
Hevesy, G. (1947). 'Report of Xlth Intl. Congress, Pure and Applied

Chemistry.' Nature, Lond., 160, 247. (166)
Hevesy, G., and Ottesen, J. (1945). 'Life cycle of the red corpuscles of the

hen.' Nature, Lond., 156, 534. (166)
Hildebrand, S. F. (1932). 'Growth of diamond-backed terrapins: size

attained, sex ratio and longevity.' Zoologica, NT., 9, 551. (51, 78)
Hinton, M. A. C. (1925). Proc. Linn. Soc, 138, 18. (109)

(1926). Monograph of the voles and lemmings. London: British Museum.

(109)
Hodge, C. F. (1894-5). 'Changes in human ganglion cells from birth to

senile death. Observations on man and honey-bee.' J. Physiol., 17, 129.

(97)
Hoff, C. C. (1937). 'Studies on the Limnaeid snail Fossaria parva Lea.'

Trans. Illinois Acad. Scu, 30, 303. (104)
Hoffman, R. S., Goldschmidt, J., and Doljanski, L. (1937). 'Comparative

studies on the growth capacities of tissues from embryonic and adult

chickens.' Growth, 1, 228. (155)
Holmgren, N. (1909). 'Termitenstudien.' Kon. Svensk. Vetensk. Akad.

Handl., 44. (100)
Hopkins, H. S. (1924). 'Respiration in the tissues of mollusks in relation to

age.' Anat. Rec., 29, 91. (32)

(1930). 'Age differences and the respiration of muscle tissue of mol-
lusks.' J. exp. ZooL, 56, 209. (32, 58, 106, 108)
Hopper, A. F. (1950). 'The effect of mammalian thyroid powder and thio-

uracil on growth rates and on the differentiation of the gonopod in

Lebistes reticulatus.' Anat. Rec., 108, 66. (161)
Horst, K., Mendel, L. B., and Benedict, F. G. (1934). 'The influence of

previous diet, growth and age upon the basal metabolism of the rat.'

J.Nutr.,8, 139. (150)
Horst, van der. (1929). 'Lebensalter und Schalengrosse.' Arch. Mollusk.,

61,46. (58)
Howard, L. O. (1939). 'Ageing of insects.' In Cowdry, E. V.: Problems of

ageing. London: Bailliere Tindall and Cox. (57)
Howes, E. L., and Harvey, S. C. (1932). 'Age factor in velocity of growth

of fibroblasts in the healing wound.' J. exp. Med., 55, 577. (156)

217

Bibliography

Hsu, T. C, and Pomerat, C. M. (1953). 'Mammalian chromosomes in vitro.

III. On somatic aneuploidy.' J. Morph., 93, 301. (168)
Hufeland, C. W. (1798). Makrobiotik, oder der Kunst das menschliche Leben

zu verldngern. Jena. (22, 112, 134)

(1829). The art of prolonging human life &c. London: Simpkin Marshall.

(Eng. trans, of foregoing.)
Hummel, K. P., and Barnes, L. L. (1938). 'Calcification of the aorta, heart

and kidneys of the albino rat.' Am. J. Path., 14, 121. (150)
Hunerhoff, E. (1931). 'Uber ein bisher unbekanntes Larvenorgan und

die Regeneration bei Apsilus vorax.' %ool. Anz., 92, 327. (87)
Hunt, T. E. (1942). 'Mitotic activity in the anterior hypophysis of female

rates.' Anat. Rec, 82, 263. (167)

(1943). 'Mitotic activity in the anterior hypophysis of mature female

rats of different age groups and at different periods of the day.' Endocrin-
ology, 32, 334. (167)

(1947). 'Mitotic activity in the anterior hypophysis of ovariectomized

rats after injection of oestrogens.' Anat. Rec, 97, 127. (167)
Hunter, W. R. (1953). 'On the growth of the freshwater limpet, Ancylus

fluviatilis: Proc. Zool. Soc., Lond., 123, 623. (102)
Hutt, F. B. (1949). Genetics of the fowl. New York: McGraw-Hill.
Huxley, A. (1937). After many a summer. London: Chatto and Windus.
Huxley, J. S. (1932). Problems of relative growth. London: Methuen. (68)
(1942). Evolution: the modern synthesis. London: Allen and Unwin.
(191)
Huxley, J. S., and de Beer, G. R. (1923). 'Studies in dedifferentiation — IV.

Resorption and differential inhibition in Obelia and Campanularia.' Quart.

J. micr. Scu, 67, 473. (82)
Huxley, T. H. (1880). The Crayfish. Kegan Paul: London. (54)
Hvass, H. (1938). ZooL Garten, LpZ., 10, 229. (52)
Hyman, L. H. (1951). The Invertebrates. Vol. III. McGraw Hill Co.: London

and New York. (59, 87, 90, 91)

Ingle, L. (1933). 'Effects of environmental conditions on longevity.'

Science, 78, 511. (143)
Ingle, L., Wood, T. R., and Banta, A. M. (1937). 'A study of the longevity,

growth, reproduction and heart rate in Daphnia longispina as influenced

by limitations in quantity of food.' J. exp. ZooL, 76, 325. (26, 29, 32, 44,

93, 143, 144, 145, 146)
Isely, F. B. (1931). 'A 15 year growth record in freshwater mussels (Quad-

rula): Ecology, 12, 616. (58, 107)
Israel, W. (1913). Biologie der Susswassermuscheln. Stuttgart. (58)

Jackson, C. H. N. (1940). 'The analysis of a tsetse fly population.' Ann.
Eugen.. Lond., 10, 332. (112)

218

Bibliography

Jackson, C. M. (1936). 'Recovery in rats upon re-feeding after prolonged

suppression of growth by dietary deficiency in protein.' Amer. J. Anat.,

58, 179. (156)
Jalavisto, E. (1950). 'The influence of parental age on the expectation of

life.' Rev. Med. Liege, 5, 719. (126)
Janet, C. (1904). Observations sur les fourmis. Limoges. (57)
Janisch, E. (1924). 'Uber die experimentelle Beeinflussung der Lebensdauer

und des Alterns schadlicher Insekten.' Arb. Biol. Reichsart. f. Land.- u.

FortwirtschafL, 13, 173. (98)
Jayne, E. P. (1953). 'Cytology of the adrenal gland of the rat at different

ages.' Anat. Rec, 115, 459. (167)
Jennings, H. S. (1945). 'Paramecium bursaria: life history. V. Some relations

of external conditions, past or present, to ageing and to mortality of

exconjugants, with summary of conclusions on age and death.' J. exp.

Zool., 99, 15. (114, 117, 118)
Jennings, H. S., and Lynch, R. S. (1928). 'Age, mortality, fertility and

individual diversity in the Rotifer Proales sordida Gorse. II.' J. exp. £00/.,

51,339. (44, 59, 85, 86, 89)
Jickeli, C. F. (1902). Die Unvolkommenheit des Stqffwechsels als Veranlassung

fur Vermehrung usw. Friedlander: Berlin. (7, 35)
Joedicke, P. (1919). 'Ein Beitrag zum eunuchoiden Riesenwuchs.' Z- Ses*

Neurol. Psychiat., 44, 385. (185)
John, D. D. (1937). 'Antarctic whales.' J. Soc. Pres. Fauna Emp., 31, 15.

Jurczyk, C. (1926). 'Zur Regeneration bei Stephanocerus.' Zool. Anz., 67,
333. (87)

(1927). 'Beitrage zur Morphologie, Biologie and Regeneration von
Stephanocerus fimbriatus Goldfuss. %. wiss. £00/., 129, 103. (87)

Kallman, F. J., and Sander, G. (1948). 'Twin studies on ageing and
longevity.' J. Hered., 39, 349. (122)

(1949). 'Twin studies in senescence.' Amer. J. Psychiat., 106, 29. (122)

Kassowitz, M. (1899). Allgemeine Biologie. Vienna. (7, 32)

Kelley, R. B. (1939). 'Female aspects of relative fertility in sheep.' Aust.
vet. J., 15, 184. (43)

Kenyon, A. T. (1942). 'The comparative metabolic influences of testicular
and ovarian hormones in man.' Biol. Sympos., 9, 11. (178)

Kershaw, W. E., Lavoipierre, M. M. J., and Chalmers, T. A. (1953).
'Studies on the intake of microfilariae by their insect vectors, their sur-
vival, and their effect on the survival of their vectors. I. Dirqfilaria
immitis and Aedes aegypti* Ann. trop. Med. Parasitol., 47, 207. (44)

Keys, A. B. (1928). 'The weight-length relation in fishes.' Proc. nat. Acad.
Sci., Wash., 14, 922. (68)

King, H. D. (1915). 'Growth and variability in body weight of the albino
rat.' Anat. Rec, 9, 751.

219

Bibliography

King, H. D. (1939). Life processes in gray Norway rats during 14 years

in captivity.' Amer. anat. Mem., 17, 1.
King, J. T., and Vissgher, M. B. (1950). 'Longevity as a function of diet

in the C3H mouse.' Feder. Proc, 9, 70. (150)
King, W. G. (1911). Census of England and Wales, 7, 46. (59 9 61)
Kinsell, L. W. (1953). 'Hormonal regulation of human growth.' In Protein

metabolism, hormones and growth. Rutgers Univ. Press: New Jersey. (183,

185)
Kirk, E., and Kvorning, S. A. (1949). 'Quantitative measurement of the

elastic properties of the skin and subcutaneous tissue in young and old

individuals.' J. Gerontol, 4, 273. (26)
Kirk, J. E. (1948). 'The acid phosphatase concentration of the prostatic

fluid in young, middle-aged and old individuals.' J. Gerontol., 3, 98.

(178)

(1949). 'The effect of testosterone administration on the acid phos-
phatase concentration of the prostatic exprimate in old men.' Urol.

cutan. Rev., 53, 683. (172, 178)

(1949). 'The urinary excretion of neutral 1 7-ketosteroids in middle-
aged and old men.' J. Gerontol, 4, 34. (173)

(1951). 'Steroid hormones and ageing.' A review. J. Gerontol, 6,

253.
Kirkland, H. T. (1928). 'A case of schistosomiasis presenting some un-
usual features.' J. trop. Med. Hyg., 31, 78. (56)
Kise, Y. and Ochi, T. (1934). 'Basal metabolism of old people.' J. Lab. din.

Med., 19, 1073. (33)
Klebanow, D., and Hegnauer, H. (1949). 'Die germinative InsufBzienz

der alternden Frau.' Z- Alter sforsch., 5, 157. (181)
Knabe, K. (1932). 'Beitrag zur Dauer von Filarieninfektion.' Arch.f. Sch. u.

Tropenhyg., 36, 496. (56)
Knoblogh, M. (1951). 'Fingernagelwachstum und Alter.' £. Alter sforsch.,

5, 357. (156)
Kobozieff, N. (1931). 'Mortalite et age limite chez la souris.' C.R. Soc. Biol,

Paris, 106, 704. (43, 47)
Koch, C. (1952). 'Von meinen altesten Urodelen.' Aquar. Terrar. £., 5, 9.

(52)
Koghakian, C. D. (1937). 'Testosterone and testosterone acetate and the

protein and energy metabolism of castrate dogs.' Endocrinology, 21, 750.

(178)
Kochakian, C. D., and Murlin, J. R. (1931). 'The effect of male hormone

on the protein and energy metabolism of castrate dogs.' J. JVutrit., 10,

439. (178)
Kolisko, A. (1938). 'Lebensgeschichte der Radertiere auf Grund von

Individualzuchten.' Arch. Hydrobiol, 33, 165. (59)
Kopaczewski, W. (1938). 'Probleme de vieillissement: recherches sur les

colloides,' Protoplasma, 30, 291. (7)

220

Bibliography

Kopec, S. (1924). 'Studies on the influence of inanition on the develop-
ment and duration of life in insects.' Biol. Bull. Wood's Hole, 46, 1. (143)
(1928). 'On the influence of intermittent starvation on the longevity of
the imaginal stage of Drosophila melanogaster? Brit. J. exp. Biol., 5, 204.
(143)
Korenghevsky, V. (1942). 'Natural relative hyperplasia and the process
of ageing.' J. Path. Bad., 54, 13. (26)

(1947). 'The longest span of life found in the records of centenarians
in England and Wales.' Brit. med. J., ii, 14. (61, 62)

(1948). 'Effect of sex and thyroid hormones on the process of ageing
in female rats.' Brit. med. J., i, 728.

(1949). 'The problem of ageing. Basic difficulties of research.' Brit. med.
J, i, 66. (24, 26)
Korenghevsky, V., and Jones, V. E. (1947). 'The effects of androsterone,
oestradiol, and thyroid hormone on the artificial premature "climacteric"
of pure gonadal origin produced by ovariectomy in rats. III. Effects on
histologic structure of vagina, uterus, adrenals and thyroid.' J. Geront.,
2, 116. (172)

(1948). 'The effects of androsterone, oestradiol, and thyroid hormone
on the artificial premature "climacteric" of pure gonadal origin pro-
duced by ovariectomy in rats.' J. Geront., 3, 21. (172)
Korenchevsky, V., Paris, S. K., and Benjamin, B. (1950). 'Treatment of
senescence in female rats with sex and thyroid hormones.' J. Geront., 5,
120. (167, 172, 173)

(1953). 'Treatment of senescence in male rats with sex and thyroid

hormones and desoxycorticosterone acetate.' J'. Geront., 8, 415. (172, 178)

Korsghelt, E. (1908). 'Versuche an Lumbriciden und deren Lebensdauer

im Vergleich mit andern wirbellosen Tieren.' Verh. deutsch. zool. Ges.,

p. 113.

(1914). 'Uber Transplantationsversuche, Ruhezustande und Lebens-
dauer der Lumbriciden.' £ool. Anz., 43, 537. (56, 107)

(1922). Lebensdauer, Altern und Tod. Fischer, Jena. (46, 54, 55, 56, 115)
(1925). In Leben, Altern, Tod. Senckenberg-bucher II. Berlin: Ber-
muhler-Verlag.

(1931). 'Uber das vermutliche Alter der Riesenschildkroten.' £ool. Anz.
96, 113. (52) ■
Kortlandt, A. (1942). 'Levensloop, samenstelling en structuur der Neder-

landse aalscholver bevolking.' Ardea, 31, 175. (110)
Kotsovsky, D. (1929). 'The origin of senility.' Amer. J. Physiol., 90, 419.
(1931). 'Allgemeine vergleichende Biologie des Alters.' Ergeb. Physiol.,
31, 132. (155)
Kountz, W. B. (1950). 'Restoration of body function in the aged.' J.

Geront., 5, 385. (178)
Kraak, W. K., Rinkel, G. L., and Hoogerheide, J. (1940). 'Oecologische
bewerking van de Europese ringgegevens van de Kievit ( Vanellus vanellus
L.).' Ardea, 29, 151. (110)

221

Bibliography

Krohn, P. L. (1955). 'Tissue transplantation techniques applied to the

problem of the ageing of the organs of reproduction.' Ciba Foundn. Collo-

quia on Ageing, 1, 141. (178)
Krumbiegel, I. (1929a). 'Lebensdauer, Altern und Tod in ihren Bezie-

hungen zur Fortpflanzung.' S.B. Ges. naturf. Fr. BerL, 94, 1928 (1929).

(35, 97)

(1929b). 'Untersuchungen liber die Einwirkung auf Altern und

Lebensdauer der Insekten ausgefuhrt an Carabus und Drosophila.' £ool.

Jb. (2. Abt.), 51, 111. (35,97,98)
Kunde, M. M., and Norlund, M. (1927). 'Inactivity and age as factors

influencing the basal metabolic rate of dogs.' Amer. J. Physiol., 80, 681.

(33)
Kunkel, E. (1916). Biologie der Lungenschnecken. Heidelberg. (58, 105)
Kunkel, K. (1928). 'Zur Biologie von Eulota fruticum Miiller.' £ool. Jb.,

45,317. (58, 103)
Kunze, — . (1933). Cited by Burger, M. (1954). (7)
Kurbatov, A. D. (1951). KypSaTOB, A.R.: 'H3MeHeHHH HacjieflCTBeHHOCTii

h noBLiuieHne >KH3HeHHOCTH noTOMCTBa nyreM TpaHCiuiaHraijHii hhhhmx

KJieTOK Me>Kfly nopo^aMH kpojihkob.' (Alteration of heredity and increased

vigour of progeny after ovum transplantation between strains of rabbit.)

Adv. mod. Biol. Moscow, 31, 2. (129)
Kurtz, E. B., and Winfrey, R. (1931). 'Life-characteristics of physical

property.' Bull. Iowa Engineering Exp. Station, No. 103. (33)
Kurzrok, R., and Smith, P. E. (1938). 'The pituitary gland.' Proc. Assn.

Nervous and Mental Diseases, 17, chap. xvii. (180).

Labitte, A. (1916). 'Longevite de quelques insectes en captivite.' Bull. Mus.

Hist. nat. Paris, 22, 105. (55, 57)
Lack, D. (1943a). The life of the robin. Witherby: London. (109, 112)
(1943b). 'The age of blackbirds.' Brit. Birds, 36, 166. (109)
(1943c). 'The age of some more British birds.' Brit. Birds, 36, 193,
214. (109)

(1946). 'Do juvenile birds survive less well than adults?' Brit. Birds, 39,
258.

(1950). 'Population ecology in birds.' A review. Proc. Xth Int. Ornith.
Congr. (110)

(1954). The natural regulation of animal numbers. Oxford: University Press.
(108, 112, 132)
Lamy, E. (1933). 'Quelques mots sur la duree de la vie chez les Mollusques.'

J. Conchyliol, 77, 48. (105)
Landauer, W., and Landauer, A. B. (1931). 'Chick mortality and sex

ratios in the domestic fowl.' Amer. Nat., 65, 492. (132)
Lankester, E. Ray (1870). On comparative longevity in man and the lower
animals. London. (11, 49)

222

Bibliography

Lansing, A. I. (1942). 'Some effects of hydrogen ion concentration, total

salt concentration, calcium and citrate on longevity and fecundity in

the rotifer.' J. exp. ZooL, 91, 195. (7, 35, 44, 91, 113)

(1947a). 'Evidence for ageing as a consequence of growth cessation.

Anat. Rec, 99, 579. (8, 14, 113)

(1947b). 'A transmissible, cumulative and reversible factor in ageing.

J. Geront., 2, 228. (113)

(1948). 'The influence of parental age on longevity in rotifers.' J

Geront., 3, 6. (87, 88, 90, 113)

(1951). 'Some physiological aspects of ageing.' Physiol. Rev., 31, 274

(4, 8, 169)

(1952). Problems of ageing (ed.). New York: Williams and Wilkins. (4

88, 89)
Lansing, A. I., Roberts, E., Ramasarma, G. B., Rosenthal, T. B., and

Alex, M. (1951). 'Changes with age in aminoacid composition of

arterial elastin.' Proc. Soc. exp. Biol. N.T., 76, 714.
Lansing, A. I., Rosenthal, T. B., and Kamen, M. D. (1949). 'Effect of

age on calcium binding in mouse liver.' Arch. Biochem., 20, 125. (169)
Lansing, A. I., and Wolfe, J. M. (1942). 'Changes in the fibrillar tissue of

the anterior pituitary of the rat associated with advancing age.' Anat.

Rec, 83, 355.
Latter, O. H. (1935). 'Unusual length of life and curious site of larva of

Cossus ligfiiperda.' Proc. R. ent. Soc, Lond., 10, 41. (57)
Lauson, H. D., Golden, J. B., and Severinghaus, E. L. (1939). The

gonadotrophic content of the hypophysis throughout the life cycle of the

normal female rat.' Amer. J. Physiol., 125, 396. (181)
Laws, R. M. (1953). 'The elephant seal (Mirounga leonina L.). I. Growth

and age.' Falkland Ids. Dependencies Survey Sci. Rep., No. 8. (48)
Lawson, T. C. (1939). 'Echinococcus cysts of the liver of 56 years' dura-
tion.' J. Amer. Med. Ass., 112, 1331. (56)
Lazovskaya, L. N. (1942). 'Age modifications of respiration of blood

vessels.' Biul. Eksp. Biol. Med., 14, 46. (32)

(1943). 'The change in respiration of blood vessels with age.' Biokhimia,
8, 171. (32)
Leathem, J. H. (1949). 'The antihormone problem in endocrine therapy.'

Rec. Prog. Hormone Res., IV, 115.
Lederer, G. (1941). 'Zur Haltung des China-alligators (Alligator sinensis

Fauvel).' ZooL Gart. LpZ., 13, 255. (52)
Lefeire, G., and Curtis, W. C. (1912). 'Studies on the reproduction and

artificial propagation of freshwater mussels.' Dept. Comm. Lab. Bull.

Bureau Fish, 30, 105.
Lehman, H. C. (1943). 'The longevity of the eminent.' Science, 98, 270.

(194)
Lehmensigk, R. (1926). 'Zur Biologie, Anatomie und Eireifung der Rader-

tiere.' Z- wiss. ZooL, 128, 37. (59)

223

Bibliography

Lemberg, R., and Legge, J. W. (1949). Hematin compounds and bile pigments.

New York: Interscience Publications. {166)
Lepeschkin, W. W. (1931). 'Death and its causes.' Quart. Rev. Biol., 6,

167. (7, 34)
Leslie, P. H., and Ranson, R. M. (1940). 'The mortality, fertility and rate

of natural increase of the vole (Microtus agrestis) as observed in the

laboratory.' J. Anim. Ecol., 9, 27. (24, 43, 108, 109)
Liebers, R. (1937). 'Beitragezur Biologie der Radertiere: Untersuchungen

an Euchlanis dilatata und Proales decipiens.' %. wiss. £ool., 150, 206. (59)
Lilly, M. M. (1953). 'The mode of life and the structure and functioning

of the reproductive ducts of Bithynia tentaculata (L).' Proc. malac. Soc. Lond.,

30, 87. (103)
Lindsay, E. (1940). 'The biology of the silverfish, Ctenolepisma longicaudata

Esch. with particular reference to its feeding habits.' Proc. Roy. Soc.

Victoria, 52, 35. (57)
Linsley, E. G. (1938). 'Longevity in the Cerambycidae.' Pan-Pacif. Ent.,

14, 177. (57)
Lobban, M. C. (1952). 'Structural variations in the adrenal cortex of the

cat.' J. Physiol, 118, 565. (167)
Loeb, J. (1908). 'Uber den TemperaturkoefHzienten fur die Lebensdauer

kaltblutiger Thiere usw.' Pflugers Arch., 124, 411. (10, 138)
Loeb, J., and Northrop, J. H. (1917). 'On the influence of food and tem-
perature on the duration of life.' J. biol. Chem., 32, 103. (8, 138, 143)
Loeb, L. (1944) . Hormones and the process of ageing. Harvey Lecture, 36, 228. (173)
Long, J. A., and Evans, H. M. (1922). 'The estrous cycle in the rat and its

associated phenomena.' Mem. Univ. Calif., 6. (179)
Longstaff, J. (1921). 'Observations on the habits of Cochlitoma zebra var.

fulgurata Pfr. and obesa Pfr. in confinement.' Proc. zool. Soc. Lond., 379.

(1921). (58)
Lorand, A. (1904). 'Quelques considerations sur les causes de la senilite.'

C.R. Soc. biol. Paris, 57, 500. (7)

(1929). 'La role de l'autointoxication intestinale dans la production

de la vieillesse.' Clinique, Paris, 24, 205. (7)
Lorenz, T. W., and Lerner, I. (1946). 'Inheritance of sexual maturity in

male chickens and turkeys.' Poult. Sci., 25, 188. (180)
Lowry, O. H., and Hastings, A. B. (1952). 'Quantitative histochemical

changes in ageing.' In Cowdry, E. V., Problems of Ageing. Williams and

Wilkins Co. (30)
Luk'yanov, V. S. (1952). JlyKLHHOB, B.C.: 'O coxpaHemie 3ji;opoBbH h

pa6oTOCnoco6HOCTM.' (The preservation of health and work-capacity.)

Medgiz, Moscow. (61)
Lyepesghinskaya, O. B. (1950). JlenecmiHCKan, O.B.: 'IIpoH3xo>KfleHHe

KJieTOK H3 TKHBoro BemecTBa 0 pojib TKHBoro BemecTBa b opraHH3Me.'

(Production of cells from living matter and the role of living matter in

the organism). Acad. Med. Sci. USSR. Moscow. 2nd edn. (115)

224

Bibliography

Lynch, R. S., and Smith, H. B. (1931). 'A study of the effects of modifica-
tion of the culture medium upon length of life and fecundity in a Rotifer
(Proales sordida), etc' Biol. Bull. Wood's Hole, 60, 30. (44, 59)

MacArthur, J. W., and Baillie, W. H. T. (1926). 'Sex differences in

mortality and metabolic activity in Daphnia magna.' Science, 64, 229. (44)
(1929a). 'Metabolic activity and duration of life. I. Influence of tem-
perature on longevity in Daphnia magna.' J. exp. £00/., 53, 221. (44, 133,

143)

(1929b). 'Metabolic activity and duration of life. II. Metabolic rates

and their relation to longevity in Daphnia magna.' J. exp. £ool., 53, 243.

(44, 133)

(1932). 'Sex differences of mortality in Abraxas-type species.' Quart.

Rev. Biol., 7, 313. (130)
McCance, R. A., and Widdowson, A. M. (1955). 'A fantasy on ageing and

the bearing of nutrition upon it.' Ciba Foundation Colloquia on Ageing, I,

186. (152)
McCay, C. M. (1952). In Lansing, A. I. (Ed.), Problems of ageing. Baltimore:

Williams and Wilkins Co. (125, 149, 150)
McCay, C. M., and Crowell, M. F. (1934). 'Prolonging the life-span.'

Set. Mon., 39, 405. (149)
McCay, C. M., Maynard, L. A., Sperling, G., and Osgood, H. S. (1941).

'Nutritional requirements during the latter half of life.' J. Nutrit., 21, 45.

(150, 151)
McCay, C. M., Sperling, L. S., and Barnes, L. L. (1943). 'Growth, ageing

and chronic diseases and life span in the rat.' Arch. Biochem., 2, 469. (149,

151, 187)
MgCullough, E. P., and Renshaw, J. F. (1934). 'The effects of castration

in the adult male.' J. Amer. med. Ass., 103, 1140. (185)
Macdonnell, W. R. (1913). 'On the expectation of life in ancient Rome

and in the provinces of Hispania and Lusitania, and Africa.' Biometrika,

9, 366. (112)
McDowell, E. C, Taylor, M. T., and Broadfort, T. (1951).

Carnegie Inst. Wash. Tear Book, 50, 200. (126)
McGavack, T. H. (1951). 'Endocrine patterns during aging.' Amer. J. Int.

Med., 35,961. (172, 173)
McIlhenny, E. A. (1940). 'Sex ratio in wild birds.' Auk, 57, 85. (132)
McIlwain, H. (1946). 'The magnitude of microbial reactions involving

vitamin-like compounds.' Nature, Lond., 158, 898. (164)

(1949). 'Metabolic changes which form the basis of a microbiological

assay of nicotinic acid.' Proc. Roy. Soc. B, 136, 12. (164)
McLaren, A., and Michie, D. (1954). 'Are inbred strains suitable for bio-
assay?' Nature, London, 173, 686. (136)
Magnus-Levy, H., and Falk, E. (1899). 'Der Lungengaswechsel der Men-

schen in den verscheidenen Alterstugen.' Pflugers. Arch. ges. Physiol.

(Physiol. Suppl.), 314. (33)

225

Bibliography

Mandl, M., and Zuckerman, S. (1952). 'Factors influencing the onset

of puberty in albino rats.' J. Endocrinol., 8, 357. (179)
Manschot, W. A. (1940). 'Een geval van progero-nanie (Progeria von

Gilford).' Nederl Tijdschr. Geneesk, 84, 3374. (134)
(1950). Acta Paediatr. Stockh., 39, 158. (134)
Marinesco, G. (1934). 'Nouvelle contribution a l'etude du mecanisme de

la vieillesse.' Bull. Acad. Med. Paris, 111, 761. (7)

(1934). 'fitudes sur le mecanisme de la vieillesse.' Rev. Soc. argent. Biol.,

10, 355. (7)
Markus, H. G. (1934). 'Life history of the blackhead minnow (Pimephales

promelas): Copeia, 116. (70)
Marshak, A. (1936). 'Growth differences in reciprocal hybrids and cyto-
plasmic influence on growth in mice.' J. exp. £ool., 72, 497. (129)
Marshak, A., and Bryon, R. L. (1945). 'The use of regenerating liver as

a method of assay.' Proc. Soc. exp. Biol., N.T., 59, 200. (155)
Marshall, H. (1947). 'Longevity of the American herring gull.' Auk, 64,

188. (109)
Mason, K. E., and Wolfe, J. M. (1930). 'The physiological activity of the

hypophyses of rats under various experimental conditions.' Anat. Rec,

45, 232. (151)
Masters, W. H. (1952). 'The female reproductive system.' In Cowdry,

E. V. Problems of ageing. Williams and Wilkins Co. (181)
Matheson, C. (1950). 'Longevity in the grey seal.' Nature, Lond., 166, 73.

(48)
Matthes, E. (1951). 'Der Einfluss der Fortpflanzung auf die Lebensdauer

eines Schmetterlings (Fumea crassiorella) .' %• vergl. Physiol., 33, 1. (44, 99)
Maupas, E. (1883). 'Contribution a l'etude morphologique et anatomique

des Infusoires cilies.' Arch. £ool. exp. gen. S, 2, vol. 1.

(1886). 'Recherches experimentales sur la multiplication des Infusoires

cilies.' Arch. £ool. exp. gen. S. 2, 6. (116)
Maurizio, A. (1946). 'Beobachtungen iiber die Lebensdauer und den

Futterverbrauch gefangen gehaltener Bienen.' Beih. Schweiz. Beinenztg.,

2, 1. (101)

(1950). 'Untersuchungen iiber den Einfluss der Pollennahrung und

Brutpflege auf die Lebensdauer und den physiologische Zustand von

Bienen.' Schweiz. Beinenztg., 73, 58. (101)
Mayer, J. (1949). 'Definition and quantitative expression of ageing.'

Growth, 13, 97. (180)
Maynard Smith, J., and Maynard Smith, S. (1954). 'Genetics and cyto-
logy of Drosophila subobscura. VIII. Heterozygosity, viability and rate of

development.' J. Genet., 52, 152. (129)
Medawar, P. B. (1940). 'The growth, growth-energy and ageing of the

chicken's heart.' Proc. Roy. Soc. B, 129, 332. (31)

(1942). 'Discussion of growth and new growth.' Proc. R. Soc. Med., 35,

500. (156, 157, 159)

226

Bibliography

Medawar, P. B. (1945). 'Old age and natural death.' Modern Quart., 1,

30. (4, 39, 159)

(1952). An unsolved problem of biology. London: H. K. Lewis. (14, 22, 25,

39)
Mellen, I. (1939). A practical cat book for amateurs and professionals. Scribner,

N.Y. (47)

(1940). The Science and Mystery of the Cat. Scribner, N.Y. (47, 48, 49)
Metalnikov, S. (1936). 'L'evolution de la mort dans le regne animal.'

Rev. gen. Sci. 31 Jan. 1936. (9)

(1937). La lutte contre la mort. Paris: Gallimard. (7, 9)
Metghnikoff, E. (1904). The Nature of Man. London: Heinemann. (7, 11)
(1907). The prolongation of life — optimistic studies. London: Heinemann.

(7, 11, 85, 92, 132, 170)

(1915). 'La mort du papillon du murier — un chapitre de thanatologie.'

Ann. Inst. Pasteur. (1915), p. 477. (35, 92, 97, 98)
Meyers, G. S. (1952). 'Annual fishes.' Aquarium J., 23, 125. (53)
Miller, H. M. (1931). 'Alternation of generations in the rotifer Lecane

inermis Bryce. I. Life histories of the sexual and non-sexual generations.'

Biol. Bull. Wood's Hole, 60, 345. (44, 59, 85, 86)
Miner, R. W. (1954). Editor. 'Parental age and characteristics of the off-
spring.' Ann. N.T. Acad. Sci., 57, 451. (126)
Minot, C. S. (1908). The problem of age, growth and death; a study of cyto-

morphosis, based on lectures at the Lowell Institute, March 1907. London. (8, 10,

29, 139, 155, 159, 177)

(1913). Moderne Probleme der Biologic Jena. (10)
Mishaikov, D. (1929). (Title unknown.) Trimesechno Spisanie Glavnata

Direkt. Statist., 1, 153, 174. (61)
Mitchell, P. G. (1911). 'On longevity and relative viability in mammals

and birds, with a note on the theory of longevity.' Proc. zool. Soc, 1, 425.

(46, 49)
Mohr, E. (1951). 'Lebensdauer einiger Tiere in Zoologischen garten.'

Zool. Gart. Lpz., 18, 60. (46)
Molisgh, H. (1938). The longevity of plants. New York: N.Y. Botanical

Garden. (7, 35)
Moltoni, E. (1947). 'Fringuello vissuto in schiavitu per ben 29 anni.' Riv.

ital. Orn., 17, 139. (49)
Montgomery, T. H. (1906). 'On reproduction, animal life cycles and the

biological unit.' Trans. Tex. Acad. Sci., 9. (7)
Moon, H. D., Simpson, M. E., Li, C. H., and Evans, H. M. (1952).

'Effects of pituitary growth hormone in mice.' Cancer Res., 12, 448. (153,

186)
Moore, C. R., and Samuels, L. T. (1931). 'Action of testis hormone in

correcting changes induced in rat prostate and seminal vesicles by

vitamin B deficiency or partial inanition.' Amer. J. Physiol., 96, 278.

(151)

ft 227

Bibliography

Moore, D. (1924). 'Note on the longevity of Clonorchis sinensis.'' U.S. Public

Health Reports, 39, 1802. (56)
Moore, H. B. (1934). 'The biology of Balanus balanoides. I. Growth rate in

relation to size, season and tidal level.' J. Mar. Biol. Ass. U.K., 19, 851.

(57)

(1935). 'A comparison of the biology of Echinus esculentus in different

habitats.' J. mar. Biol. Ass. U.K., 20, 109. (58)
Morant, G. M. (1950). 'Secular changes in the heights of British people.'

Proc. Roy. Soc. B., 137, 443. (152)
Moskovljevic, V. (1939). Bee World, 20, 83; 21, 39-41. (101)
Muhlmann, M. S. (1900). Uber die Ursache des Alters. Wiesbaden: Bergman.

(1911). 'Das Altern und der physiologische Tod.' Samml. anat. physiol.

Vortr., 1,455. (7, 8)

(1924). 'Meine Theorie des Al terns und des Todes; zugleich zur

Abwehr.' Virchows Arch., 253, 225. (7)

(1927). 'Wachstum, Altern und Tod. Uber die Ursache des Alterns

und des Todes.' Ergeb. Anat. EntwGesch.; Anat. Abt., 27, 1. (7)
Mulinos, M. G., and Pomerantz, L. (1941). 'Hormonal influences on

weight of adrenal in inanition.' Amer. J. Physiol., 132, 368. (151)
Murie, A. (1944). The wolves of Mount McKinley. U.S. Dep. Int. Nat. Park

Service, Washington. (Ill)
Murray, J. (1910). 'Antarctic Rotifera.' British Antarctic Expedition, 1907—

1909, I (3), 41. (165)
Murray, W. S. (1934). 'The breeding behaviour of the dilute brown stock

of mice (Little dba).' Amer. J. Cancer, 20, 573.
Murray, W. S., and Hoffmann, J. G. (1941). 'Physiological age as a basis

for the comparison of strains of mice subject to spontaneous mammary

carcinoma.' Cancer Res., 1 , 298. (43)

Nagornyi, A. V. (1948). HaropHLiii, A.B.: 'OrapeHne h npuftJieHHe jkh3hh.'
(Old age and the prolongation of life) Sovietskaya Nauka, Moscow. (46,
54, 61)

Nasgher, I. (1920). 'A noted case of longevity — John Shell, centenarian.'
Amer. Med., 15, 151. (62)

Needham, A. E. (1950). 'Growth and regeneration rates in relation to age
in Crustacea.' J. Geront., 5, 5. (96)

Needham, J. (1942). Biochemistry and morphogenesis. Cambridge: University
Press. (159, 190)

Neuberger, A., and Slack, H. G. B. (1953). 'The metabolism of collagen
from liver, bone, skin and tendon in the normal rat.' Biochem. J., 53 , 47. ( 34)

Newgombe, C. L. (1935). 'Growth of Mya arenaria in the Bay of Fundy
region.' Canad. J. Res., 13, 97. (58, 106)

(1936). 'Validity of concentric rings of Mya arenaria L. for deter-
mining age.' Nature, Lond., 137, 191. (58, 106, 107)

228

Bibliography

Nichols, J. T. (1939). 'Data on size, growth and age in the box turtle,

Terrapene Carolina.' Copeia, 14. (78)
Nikitin, V. N. (1954). Hhkhthh, B.H.: 'JJojirojieTKe.' (Longevity.) Set. &

Life, Moscow, (8) 27. (61)
Noble, G. K. (1931). The Biology of the Amphibia. New York: McGraw Hill.

(52, 67)
Norris, J. L., Blanchard, J., and Polovny, C. (1942). 'Regeneration of

rat liver at different ages.' Arch. Path. (Lab. Med.), 34, 208. (155)
Norris, M.J. (1933). 'Contributions toward the study of insect fertility.

II. Experiments on the factors influencing fertility in Ephestia kuhniella

Z: Proc. zool. Soc. Lond., 903. (98)

(1934). 'Contributions toward the study of insect fertility. III. Adult

nutrition, fecundity and longevity in the genus Ephestia.' Proc. zool. Soc.

Lond., p. 334. (35, 98)
Northrop, J. (1917). 'The effect of prolongation of the period of growth on

the total duration of life.' J. biol. Chem., 32, 123. (143)
du Nouy, P. L. (1916). 'Cicatrization of wounds.' J. exp. Med., 24, 461.

(26, 156)

(1932). 'Une mesure de l'activite physiologique.' C.R. Soc. Biol., Paris,

109, 1227. (20, 156)

(1936). Biological Time. Methuen, London. (5)
Novak, E. (1921). Menstruation and its disorders. New York: Appleton. (181)
(1944). 'The constitutional type of precocious female puberty, with a

report of nine cases.' Amer. J. Obst. Gynec, 47, 20. (179)
Noyes, B. (1922). 'Experimental studies on the life history of a rotifer

reproducing parthenogenetically (Proales decipiens).' J. Exp. %ool., 35,

222. (85)

Oldham, C. (1930). 'Fecundity of Planorbis corneus.' Naturalist, 111. (58)
(1931). 'Note on V(iviparus) contectus.' J. Conch., 19, 179. (58, 102)
(1942a). 'Autofecundation and duration of life in Limax cinereoniger?
Proc. malac. Soc, Lond., 25, 9. (58)

(1942b). 'Notes on Geomalacus maculosus' Proc. malac. Soc. Lond., 25,
10. (58)
Oliff, W. D. (1953). 'The mortality, fecundity and intrinsic rate of natural
increase of the multimammate mouse (Rattus (Mastomys) natalensis
Smith) in the laboratory.' J. Anim. Ecol., 22, 217. (43, 130)
Oliver, J. A. (1935). 'Young Billy Johnson's old box turtle.' Animal King-
dom, 56, 154. (51)
Orton, J. H. (1928). 'On rhythmic periods of shell growth in Ostrea edulis
with a note on fattening.' J. mar. Biol. Ass. U.K., 15, 365. (104)

(1929). 'Reproduction and death in invertebrates and fishes.' Nature,
Lond., 123, 14. (8, 35)
Orton, J. H., and Amurthalingam, C. (1930). 'Giant English Oysters.'
Nature, Lond., 126, 309. (58)

229

Bibliography

Osborne, T. B., and Mendel, L. B. (1916). 'The resumption of growth
after long-continued failure to grow.' J. biol. Chem., 23, 439. (148)

(1916). 'Acceleration of growth after retardation.' Amer. J. Physiol.,
40, 16. (148, 156)

Pai, S. (1928). 'Die phasen des Lebenzyklus der Anguillula aceti Ehr. und
ihre experimentellmorphologische Beeinflussung.' £. wiss. <oo/., 131,
293. (84)

(1934). 'Regenerationsversuche an Rotatorien.' Sci. Rep. Univ. Che-
kiang, 1. (87)

Park, T. (1945). 'Life tables for the black flour-beetle, Tribolium madens
Charp.' Amer. Nat., 79, 436. (44, 130, 131).

Parker, G. H. (1926). 'The growth of turtles.' Proc. Nat. Acad. Sci. Wash.,
12, 422. (52, 79)

(1929). 'The growth of the loggerhead turtle.' Amer. Nat., 63, 367.
(52, 79)

Parker, W. R. (1933). 'Pelorus Jack.' Proc. Linn. Soc, 1933, p. 2. (47)

Parkes, A. S. (1929). 'Note on the growth of young mice suckled by rats.'
Ann. Appl. Biol, 16, 171.

(1942-44). 'Induction of superovulation and superfecundation in
rabbits.' J. Endocrinol, 3, 268. (179)

Parkes, A. S., and Smith, A. U. (1953). 'Regeneration of rat ovarian tissue
grafts after exposure to low temperatures.' Proc. Roy. Soc. B, 140, 455.
(180)

Parsons, R. J. (1936). In Medical papers dedicated to H. A. Christian. Balti-
more. (167, 174)

Payne, F. (1949). 'Changes in the endocrine glands of the fowl with age.'
J. Geront., 4, 193. (174)

(1952). 'Cytological changes in the cells of the pituitary, thyroids,
adrenals and sex glands of the ageing fowl.' In Lansing, Problems of
ageing. (64, 174)

Pearce, J. M. (1936). 'Age and tissue respiration.' Amer. J. Physiol, 114,
255. (33)

Pearl, R. (1927). 'On the distribution of differences of vitality among indi-
viduals.' Amer. Nat., 61, 113.

(1928). The rate of living. New York: Knopf. (4, 7, 8, 10, 44, 124)
(1940). Introduction to medical biometry and statistics, 3rd edn. Penn-
sylvania: W. B. Saunders Co. (19)

Pearl, R., and Doering, C. R. (1923). 'A comparison of the mortality of
certain lower organisms with that of man.' Science, 57, 209. (113)

Pearl, R., and Miner, J. R. (1935). 'Experimental studies in the duration
of life. XIV. The comparative mortality of certain lower organisms.'
Quart. Rev. Biol, 10, 60. (19, 33, 43, 44, 82, 104, 113)

(1936). 'Life tables for the pecan-nut case bearer (Acrobasis caryae,
Grote).' Mem. Mus. Hist. nat. Belg., 3, 169. (44, 132)

230

Bibliography

Pearl, R., Park, T., and Miner, J. R. (1941). 'Experimental studies on
the duration of life. XVI. Life-tables for the flour beetle Tribolium con-
fusum Duval.' Amer. Nat., 75, 5. (44)

Pearl, R., and Parker, S. L. (1924). 'Experimental studies on the dura-
tion of life. IX. New life-tables for Drosophila.' Amer. Nat., 58, 71. (44)

(1922). 'Experimental studies in the duration of life. II. Hereditary
differences in duration of life in line-bred strains of Drosophila.' Amer. Nat.
56, 174. (44, 121, 124)

Pearl, R., and Pearl, R. de W. (1934a). The ancestry of the long-lived.
London: H. Milford. (122)

(1943b). 'Studies on human longevity. VI. Distribution and correla-
tion of variation in the total immediate ancestral longevity of nonagen-
arians and centenarians in relation to inheritance factor in the duration
of life.' Hum. Biol, 6, 98. (122)

Pearson, K. (1895). 'Mathematical contributions to the theory of evolu-
tion. II. Skew variations in homogeneous material.' Phil. Trans. Roy. Soc,
186, Ser. A, 343. (18)

Pearson, K., and Elderton, E. M. (1913). 'On the hereditary character
of general health.' Biometrika, 9, 320. (125)

Pearson, O. P. (1945). 'Longevity of the short-tailed shrew.' Amer. Midi.
Nat., 34, 531. (34, 47, 109)

Pearson, O. P., and Baldwin, P. H. (1953). 'Reproduction and age struc-
ture of a mongoose population.' J. Mammal., 34, 436. (34)

Pease, M. (1947). 'How long do poultry breeding stock live?' J. Ministr.
Agric, 54, 263. (132)

Pease, M. S. (1928). 'Experiments on the inheritance of weight in rabbits.'
J. Genet., 20, 261.

Pelseneer, R. (1934). 'La duree de la vie et Page de la maturite sexuelle
chez certains mollusques.' Ann. Soc. zool. Belg., 64, 93. (58, 102)
(1935). Essai d'ethologie zoologique. Bruxelles. (36, 102, 105)

Penfold, W. J., Penfold, H. B., and Phillips, M. (1936), 'A survey of the
incidence of Taenia saginata infestation in the population of the State of
Victoria, etc' Med. J. Australia, 23rd Year, I, 283, (56)

Perkins, C. B. (1948). 'Longevity of snakes in captivity in the U.S.' Copeia,
217. (52)

Perrone, J. C, and Slack, H. G. B. (1952). 'The metabolism of collagen
from skin, bone and liver in the normal rat.' Biochem. J., 49, lxxii. (34)

Perry, J. S. (1953). 'The reproduction of the African elephant, Loxodonta
africana: Phil. Trans. Roy. Soc. B, 237, 93. (34)

Perry, R. (1953). 'Some results of bird ringing.' New Biol, 15, 58. (34,
49, 110)

Petrova, M. K. (1946). IleTpoBa, M.K.: 'O pojin (JjymjHOHajiHoii ocJia6-

JieHHOft KOpH TOJIOBHOrO M03ra B B03HHKOBeHHHH paSJIH^HBIX naTO-

JiorinecKHX npoijeccoB.' (Functional weakening of the cerebral cortex in
the aetiology of various disease states.) Moscow. (153)

231

Bibliography

Pflugfelder, O. (1948). 'Volumetrische untersuchungen an den corpora

allata der Honigbiene {Apis melliftca) .' Biol, tyl, 67, 223. (57, 100,

167)
Pigado, T. C. (1930). 'Effets des injections de serum homologue sur la

taille et croissance des animaux.' Ann. Inst. Pasteur, 44, 584. (155)
Picado, T. C, and Rotter, W. (1936). 'Precipitines anti-glandes endo-

crines et longevite chez quelques especes de vertebres.' C.R. Soc. Biol.,

Paris, 123, 869. (169)
Piepho, H. (1938). 'Uber die Auslosbarkeit uberzahliger Haiitungen und

vorzeitiger Verpuppung an Hautstiicken bei Kleinschmetterlingen.'

Naturwiss., 26, 841. (147)
Pierson, B. F. (1938). 'Relation of mortality after endomixis to the prior

interendomitotic interval in Paramecium aurelia.' Biol. Bull. Wood's Hole,

74, 235. (118)
Pincus, G. (1950). 'Measures of stress responsivity in younger and older

men.' Psychosom. Med., 12, 225. (174)
Pitt, F. (1945). 'Breeding of the harvest mouse in captivity.' Nature, Lond.,

155, 700. (47)
Pixell-Goodrich, H. (1920). 'Determination of age in honey bees.' Quart.

J. micr.Scl, 64, 191. (97, 99)
Plate, L. (1886). 'Beitrage zur Naturgeschichte der Rotatorien.' Jena £.f.

Naturwiss., 19, 1. (85)
Polezaiev, L. V., and Ginsburg, G. I. (1939). 'Studies by the method of

transplantation on the loss and restoration of the regenerative power in

the tailless amphibian limbs.' C.R. Acad. Sci. U.R.S.S., 23, 733. (182)
Pontecorvo, G. (1946). 'Microbiology, biochemistry and the genetics of

microorganisms.' Nature, Lond., 157, 95. (165)
Pruitt, W. O. (1954). 'Ageing in the masked shrew, Sorex cinereus cinereus

Kerr.' J. Mammal, 35, 35. (34)
Przibram, H. (1909). Experimental-Zoologie. Vienna. Vol. II, p. 126. (104)
Putter, A. (1921). 'Die altester Menschen.' Naturwiss., 9, 875. (60)

Quick, H. E. (1924). 'Length of life of Paludestrina ulvae.' J. Conchol., 17,
169. (58)

Raben, M. S., and Westermeyer, V. W. (1952). 'Differentiation of growth

hormone from the pituitary factor which produces diabetes.' Proc. Soc.

exp. Biol., 80, 83. (186)
Rabes, O. (1901). 'Uber Transplantations-Versuche an Lumbriciden.' Biol.

Zbl., 21,633. (56)
Raffel, D. (1932). 'The occurrence of gene mutations in Paramecium

aurelia.' J. exp. ^ool., 63, 371. (118)
Rahm, P. G. (1923). 'Beitrage zur Kenntnis der Moosfauna.' Z- a^S-

Physiol, 20, 1. (87, 140)

232

Bibliography

Rao, H. S. (1937). 'On the habitat and habits of Trochus niloticus Linn, in
the Andaman Seas.' Rec. Indian Mus., 39, 47. (58, 104)

Rasquin, P., and Hafter, E. (1951). 'Age changes in the testis of the teleost,
Astyanax americanus.' J. Morph., 89, 397. (69)

Rau, P. (1924). 'The biology of the roach, Blatta orientalis Linn.' Trans.
Acad. Sci. St. Louis, 25, 57. (44)

Rau, P., and Rau, N. (1914). 'Longevity in saturnid moths and its relation
to the function of reproduction.' Trans. Acad. Sci. St. Louis, 23, 1. (132)

Reege, R. P., and Turner, C. W. (1937). 'The lactogenic and thyrotropic
hormone content of the anterior lobe of the pituitary gland.' Univ. Mis-
souri Agr. Exp. Sta. Res. Bull., 266, 1. (186)

Regan, W. M., Mead, S. W., and Gregory, P. W. (1947). 'The relation of
inbreeding to calf mortality.' Growth, 11, 101. (118)

Reiner, J. M. (1947). 'The effect of age on the carbohydrate metabolism
of tissue homogenates.' J. Geront., 2, 315. (32)

Rey, P. (1936). 'La longevite des Galleria adultes, males et femelles.' C.R.
Soc.Biol., 121, 1184. (132)

Ribbands, C. R. (1950). 'Changes in the behaviour of honey-bees following
their recovery from anaesthesia.' J. exp. Biol., 27, 302. (101)

(1952). 'Division of labour in the honey-bee community.' Proc. Roy.
Soc. B, 140, 32. (44, 102, 112)

(1953). The behaviour and social life of honey bees. London: Bee Research
Association Ltd. (38, 44, 101)

Ribbert, H. (1908). Der Tod aus Alter sschwdche. Bonn: Cohen. (7)

Richards, O. W. (1953). The Social Insects, p. 188. Macdonald: London.
(57)

Rigker, W. E. (1945). 'Natural mortality among Indiana blue-gill sun-
fish.' Ecology, 26, 111. (68, 71)

(1948). 'Methods of estimating vital statistics of fish populations.'
Indiana Univ. Publ. Sci. Ser., 15, 1. (72)

Riley, W. A. (1919). 'The longevity of the fish tapeworm of man, Diphyllo-
bothrium latum.' J. Parasitol., 5, 193. (56)

Rizet, G. (1953). 'Sur Pimpossibilite d'obtenir la multiplication vegetative
ininterrompue et illimitee de l'Ascomycete Podospora anserina.' C.R. Acad.
Sci., 237, 828. (116)

Robertson, F. W., and Reeve, E. C. R. (1952). 'Heterozygosity, environ-
mental variation, and heterosis.' Nature, Lond., 170, 286. (137)

Robertson, T. B. (1923). The chemical basis of growth and senescence. Phila-
delphia. (8, 32, 138, 153)

Robertson, T. B., Dawbarn, M. C, Walters, J. W., and Wilson,
J. D. O. (1933). 'Experiments on the growth and longevity of the white
mouse, II.' Aust. J. exp. Biol. med. Sci., 11, 219. (153)

Robertson, T. B., and Ray, L. A. (1920). 'Experimental studies on growth.
XV. On the growth of relatively long-lived compared with that of rela-
tively short-lived animals.' J. biol. Chem., 42, 71. (125, 128, 153)

233

Bibliography

Robertson, T. B., and Ray, L. A. (1919). 'Experimental studies on growth.

XI. The influence of pituitary gland tissue, tethelin, egg lecithin, and

cholesterol upon the duration of life in the white mouse.' J. biol. Chem.,

37, 427. (153)
Robson, J. M. (1947). Recent advances in sex and reproductive physiology. Lon-
don: Churchill. (180)
Rockstein, M. (1950). 'The relation of cholinesterase activity to change in

cell number with age in the brain of the adult worker honey bee.' J. Cell.

comp. Phys.,35, 11. (100)
Rogers, J. B. (1950). 'The development of senility in the guinea pig.' Anat.

Rec, 106, 286. (47)
Rokhlina, M. L. (1951. PoxjiHHa, M.JL: 'IlyTL c ftOJirojieTHio.' (The road

to longevity.) Pravda, Moscow. (61)
Rollinat, R. (1934). 'La vie des reptiles de la France centrale.' Paris.

(51, 78)
Rosenthal, O., Bowie, M. A., and Wagoner, G. (1940). 'Metabolism of

cartilage (bovine) with particular reference to the effects of ageing.' J.

Amer. med. Ass., 115, 2114. (32)

(1941). 'Studies in the metabolism of articular cartilage. I. Respiration

and glucolysis of cartilage in relation to age.' J. cell. comp. Physiol., 17,

221. (32)

(1942). 'The dehydrogenetic ability of bovine articular cartilage in

relation to its age.' J. cell. comp. Physiol., 19, 333. (32)
Rothschild, M. (1935). 'Gigantism and variation in Peringia ulvae Penn.

caused by infection with larval trematodes.' J. mar. Biol. Ass. U.K., 20,

537. (104)
Roux, W. (1881). Der Kampfder Teile im Organismus. Leipzig. (8)
Rubin, B. L., Dorfman, R. I., and Pincus, G. (1955). ' 1 7-ketosteroid

excretion in ageing subjects.' Ciba Foundation Symposia on Ageing, 1, 126.
Rubner, M. (1908). 'Probleme des Wachstums und der Lebensdauer.'

MittGes. inn. Med., Wien, 7, 58. (8, 9)
Rudzinska, M. (1951). 'The influence of amount of food on the reproduc-
tion rate and longevity of a Suctorian (Tokophrya infusionum).3 Science,

113, 10-11. (115)

(1952). 'Overfeeding and life-span in Tokophrya infusionum.9 J. Geront.,

7, 544. (115)
Rutgers, A. J. (1953). 'Mortality by cancer as a function of age.' Experi-

entia, 12, 470. (25)
Rutman, R. J. (1950). 'A maternal influence on the incorporation of

methionine into liver protein.' Science, 112, 252. (129)

(1951). 'The inheritance of rates of methionine uptake by rat liver

protein and relations to growth.' Genetics, 36, 59. (129)
Ruud, J. T., Jonsgard, A., and Ottestad, P. (1950). 'Age studies in blue

whales.' Hvalrdd. Skr., 33, 1. (47)

234

Bibliography

Ruzicka, V. (1924). 'Beitrage zum Studium der Protoplasmahysteretischen
Vorgange (Zur Kausalitat des Alterns).' Arch. mikr. Anat., 101, 459.
(7, 34)

(1929). 'Beitrage zum Studium der Protoplasmahysteresis, etc. (Zur
Kausalitat des Alterns).' Arch. EntwMeck. Org., 116, 104. (7)

Saldau, M. P. (1939). Caji^ay, M.IL: 'TeMn pocTa npoMbicjiOBbix

mojijiiockob HeKOToptix paiiOHOB eBponeiiCKOH qacTH CCCP.' (Growth rate

of commercially valuable molluscs in some districts of the European part

of the USSR.) Bull. Inst.freshw. Fish., Leningr., 22, 244. (107)
Salmon, T. N. (1941). 'Effect of pituitary growth substance on the develop-
ment of rats thyroidectomised at birth.' Endocrinology, 29, 291.
Samuels, L. T. (1946). 'The relation of the anterior pituitary hormones to

nutrition.' Rec. Prog. Hormone Res., 1, 147. (151)
Sanderson, W. A. (1949). [Report of British Association Symposium on

social and psychiatric aspects of ageing.] Nature, Lond., 163, 221. (194)
Sandground, J. H. (1936). 'On the potential longevity of various helminths,

with a record from a species of Trichostrongylus in man.' J. Parasit., 22,

464. (56)
Savory, T. H. (1927). The biology of spiders. Sidgwick and Jackson: London.

(56)
Sawin, P. B. (1954). In Miner, R.W. (Ed.) 'Parental age and characteristics

of the offspring.' Ann. N.T. Acad. Sci., 57, 451. (126)
Saxton, J. A. (1945). 'Nutrition and growth and their influence on

longevity in rats.' Biol. Sympos., 11, 177. (150)
Saxton, J. A., and Kimball, G. G. (1941). 'Relation of nephrosis and other

diseases of albino rats to modifications of diet.' Arch. Path., Chicago, 32,

951. (150)
Saxton, J. H., and Greene, H. S. N. (1939). 'Age and sex differences in

hormone content of the rabbit hypophysis.' Endocrinology, 24, 494. (179)
Schaffer, V. B. (1950). 'Growth layers on the teeth of Pinnipedia as an

indication of age.' Science N.T., 112, 309. (48)
Schloemer, C. L. (1936). 'The growth of the muskellunge, Esox masquinongy

immaculatus (Garrard) in various lakes and drainage areas of Northern

Wisconsin.' Copeia, 185. (70)
Schlomka, G., and Kersten, E. (1952). 'Uber Moglichkeiten einer statis-

tichen Altenscharakteristik auf Grund von Morbiditatszahlen.' £.

Altersforsch., 6, 306. (26)
Schlottke, E. (1930). 'Zellstudien an Hydra. I. Altern und Abbau von

Zellen und Kernen.' Z- mikr. -anat. Forsch., 22, 493. (82)
Schmalhausen, I. (1928). 'Das Wachstumsgesetz und die Methode der

Bestimmung der Wachstumskonstante.' Arch. EntwMeck. Org., 113, 462.

(72, 73, 74, 184)

(1929). 'Zum Wachstumstheorie.' Arch. EntwMeck. Org., 116, 5677.

(159)

235

Bibliography

Schmidt, H. (1923). 'Uber den Alterstod der Biene.' Jena Z-f- Naturwiss.,

59, 343. (97, 99)
Schmidt, W.J. (1952). 'Einiges uber das altern der Tiere.' Z> Altersforsch.,

6, 344. (46)
Schneider, K. M. (1932). 'Zum Tode des Leipzigen Riesensalamanders.'

Zool. Gart. Lpz., 5, 142. (52)
Schulze-Robbecke, G. (1951). 'Untersuchungen iiber Lebensdauer,

Altern, und Tod bei Anthropoden.' Zool Jb., 62, 366. (32, 95, 100)
Scow, R. O., and Marx,W. (1945). 'Response to pituitary growth hormone

of rats thyroidectomized on the day of birth.' Anat. Rec, 91, 227. (161,

185)
Sebesta, F. (1935). 'Beobachtungen uber das Alter der Raniden.' Bldtt. Aq.

Terrar., 46, 115.
Sekera, E. (1926). 'Beitrage zur Kenntnis der Lebensdauer bei einigen

Turbellarien und Susswassernemertinen.' Z°°l Anz., 66, 307.
Selye, H. (1946). 'General adaptation syndrome and diseases of adapta-
tion.' J. clin. Endocrin., 6, 117. (125)
Selye, H., and Albert, S. (1942). 'Age factor in responsiveness of pituitary

and adrenal folliculoids.' Proc. soc. exp. Biol. N.T., 50, 159. (180)
Sergeev, A. (1937). 'Some materials to the problem of the reptile post-
embryonic growth.' Zool ZK 16, 723. (77, 78)

(1939). [? title]. Probl Ecol. Biotsen., Leningr., 4, 276. (110)
Seshaiya, R. V. (1927). 'On the breeding habits and fecundity of the snail

Limnaea luteola Lk.' J. Bombay nat. Hist. Soc, 32, 154. (58)
Sette, O. E. (1943). 'Biology of the Atlantic mackerel (Scomber scombrus) of

North America. I. Early life history including the growth, drift and mor-
tality of the egg and larval population.' Bull. U.S. Fish. Wildl. Serv. Fish.,

50, 147. (Ill)
Severtsov, A. N. (1939). CeBepu,0B, A.H.: ' Mop^ojionmecKne 3anoHO-

MepHOCTH aBOJiioijHOHHoro npoijecca.' (Morphological laws of the

evolutionary process.) Sorena, 3, 23. (115)
Sewell, R. B. S. (1924). 'Observations on growth in certain molluscs and

on changes correlated with growth in the radula of Pyrazus palustris.* Rec.

Ind. Mus. Calcutta, 26, 529. (112)
Shanklin, W. M. (1953). 'Age changes in the histology of the human

pituitary.' Acta anat., Basel, 19, 290. (168)
Sharp, D. (1883). 'Longevity in a beetle.' Entom. monthly Mag., 19, 260.

(57)
Sheldon, J. H. (1949). Social medicine of old age. Oxford University Press.

(193, 194)
Shemin, D., and Rittenberg, D. (1944). 'Some interrelationships in

general nitrogen metabolism.' J. biol Chem., 153, 401. (34)
Sherman, H. C., and Campbell, H. L. (1935). 'Rate of growth and length

of life.' Proc. Nat. Acad. Sci., 21, 235. (125)

236

Bibliography

Shock, N. W. (1942). 'Standard values for basal oxygen consumption in

adolescence.' Amer. J. Dis. Childhood, 64, 19. (33)

(1948). 'Metabolism in old age.' Geriatrics, 1, 232. (33)
(1951). A classified bibliography of gerontology and geriatrics. U.S.A.: Stan-
ford University Press. (4, 33)
Shock, N. W., Watkin, D. M., and Yiengst, M.J. (1954). 'Age changes in

renal function and basal oxygen consumption as related to total body

water.' Fed. Proc, 13, 136. (33)
Shulov, A. (1939-40). 'On the biology of Latrodectes spiders in Palestine.'

Proc. Linn. Soc, Lond., 309. (133)
Silberberg, M., and Silberberg, R. (1954). 'Factors modifying the life

span of mice.' Amer. J. Physiol, 177, 23. (150)
Simmonds, L. W. (1945). The role of the aged in primitive society. Yale Univer-
sity Press. (193)

(1946). 'Attitudes toward ageing and the aged: primitive societies.' J.

Geront,, 1, 72. (193,194)
Simmonds, M. (1914). 'Uber embolische Prozesse in der Hypophysis.' Arch.

f path. Anat., 217,226. (174)

(1914b). 'Zur Pathologie der Hypophysis.' Verh. dtsch. path. Ges., 17,

808. (174)
Simms, H. S. (1936). 'The effect of physiological agents upon adult tissues

in vitro.' Science, 83, 418. (158)
Simms, H. S., and Stillman, N. P. (1936). 'Substances affecting adult tissue

in vitro. II. A growth inhibitor in adult tissue.' J. gen. Physiol., 20, 621.

(154) ^

(1937). 'Substances affecting adult tissue in vitro. III. A stimulant (the

'A' factor) in serum ultrafiltrate involved in overcoming adult tissue

dormancy.' J. gen. Physiol., 20, 649. (158)
Sinclair, H. M. (1955). 'Too rapid maturation in children as a cause of

ageing.' Ciba Foundation Colloquia on Ageing, I, 194. (152, 196)
Sivertsen, E. (1941). 'On the biology of the Harp Seal, Phoca groenlandica

Ersel.' Hvalrdd. Skr., Oslo, No. 26. (48)
Sleptzov, M. M. (1940). 'Determination de l'age chez Delphinus delphis L.'

Bull. Soc. Nat. Moscow, 49, 43. (47)
Slonaker, J. R. (1912). 'The normal activity of the albino rat from birth

to natural death, its rate of growth and the duration of life.' J. Anim.

Behav., 2, 20. (43)

(1930). 'The effect of the excision of different sexual organs on the

development, growth and longevity of the albino rat.' Am. J. Physiol.,

93, 307. (133)
Smallwood, W. M., and Philips, R. L. (1916). 'The nuclear size in the

nerve cells of bees during the life cycle.' J. comp. Neurol., 27, 69. (100)
Smith, M. (1951). The British amphibians and reptiles. London: Collins.

(52)

237

Bibliography

Smith, P. E., and Dortzbach, C. (1929). 'The first appearance in the

anterior pituitary of the developing pig foetus of detectable amounts of

the hormones stimulating ovarian maturity and general body growth.'

Anat. Rec, 43, 277. (179)
Smyth, N. (1937). 'Longevity in horses.' Cavalry J., 27, 101. (48)
Sollaud, E. (1916). 'Recherches sur la biologie des 'Palemonides' des cotes

de la France.' Rec. Fonds Bonaparte, 1, 69. (57)
Solomon, D. R., and Shock, N. W. (1950). 'Studies of adrenal cortical and

anterior pituitary function in elderly men.' J. Geront., 5, 302. (174)
Sonden, K., and Tigerstedt, R. (1895). 'Die Respiration und der Ge-

sammtstoffwechsel des Menschen.' Skand. Arch. Physiol., 6, 1. (33)
Sonneborn, T. M. (1930). 'Genetic studies on Stenostomum incaudatum n.sp.

I. The nature and origin of differences in individuals formed during

vegetative reproduction.' J. exp. ZG°l-> 57, 57. (83, 115)

(1938). 'The delayed occurrence and total omission of endomixis in

selected lines of Paramecium aurelia.' Biol. Bull. Wood^s Hole, 74, 76.

(116)

SOROKINA, M. I. (1950). CopOKHHa, M.H.: '06 H3MeHHHBOCTH XpOMOCOMHOrO

KOMiuieKca b TKaHeBHx KJieTax TenjiOKpoBHLix jkhbothbix.' (Variability
of the chromosome complex in tissue cells of warm-blooded animals.)
Bull. Acad. Sci. URSS., 6, 97. (168)

Spear, F. G. (1928). 'The effect of low temperature on mitosis in vitro.'
Arch. exp. Zellforsch., 1, 484. (156)

Spemann, F. W. (1924). 'Uber Lebendsauer, Altern und andere Fragen der
Rotatorien-biologie.' %. wiss. gool., 123, 1. (59, 85, 88)

Spenser, R. R., and Melroy, M. B. (1949). 'Studies of survival of uni-
cellular species. I. Variations in life expectancy of a paramecium under
laboratory conditions.' J. nat. Cancer Inst., 10, 1. (118)

Spiegelman, S., Baron, L. S., and Quastler, H. (1951). 'Enzymatic
adaptation in non-viable cells.' Fed. Proc, 10, 130. (165)

Stebbins, R. G. (1948). 'Additional observations on home ranges and
longevity in the lizard Sceleporus graciosus.' Copeia, 20, (110)

Stebbins, R. C., and Robinson, H. B. (1946). 'Further analyses of a
population of the lizard Sceleporus graciosus gracilis.' Univ. Calif. Publ. Z°ol-,
48, 149. (110)

Steinmetz, G. H. (1954). 'Some effects of thyroxine and antithyroid com-
pounds on tadpoles and their relation to hormonal control of growth.'
Physiol. Zool., 27, 28. (161, 185)

Stephenson, R. A. (1935). British sea anemones. London: The Ray Society.
(56, 81).

Stewart, D. C, and Kirk, P. L. (1954). 'The liquid medium in tissue
culture.' Biol. Revs., 29, 119. (159)

Stolc, A. (1902). 'Uber den Lebenszyklus der niedrigsten Siisswasser-
annulaten und uber einiger sich anschliessende biologische Fragen.'
Bull, intern. Acad. Sci. Bohime, 1902. (83)

238

Bibliography

Stolte, H. A. (1924). 'Altersveranderung bei limicolen Oligochaten.' Ver-
hand. Deutsch. Zool. Ges., 29, 43. (83)

(1927). 'Studien zur Histologic des Altersprozesses.' Z- wiss. Zool., 129,
1. (83)

(1937). 'Gestaltung, Zeichnung und Organabbau unter dem Einfluss
normaler und 'alternden' gonaden bei Polyophthalmus pictus (Polychaeta).'
Z wiss. Zool., 150, 107. (33)
Strong, L. C. (1936). 'Production of CBA strain inbred mice: long life

associated with low tumour incidence.' Brit. J. exp. Path., 17, 60. (121)
Strong, L. C, and Smith, G. M. (1936). 'Benign hepatomas in mice of

the CBA strain.' Amer. J. Cancer, 27, 279. (121, 123)
Summers, F. M. (1938). 'Form regulation in Zoothamnion alternans.' Biol. Bull.

Wood's Hole, 74, 130. (155)
Suzuki, Y. (1926). 'A study of the resistance of animals by the tissue

culture method.' Mitt. allg. Path. Sendai, 2, 191. (31, 155)
Swift, D. R. (1954). 'Influence of mammalian growth hormone on rate of

growth offish.' Nature, Lond., 173, 1096. (182)
Swyer, G. I. M. (1954). Reproduction and sex. London: Routledge. (180)
Szab6, I. (1931a). 'The three types of mortality curves.' Quart. Rev. Biol.,
6, 462. (19)

(1931b). 'Absterben und Entwicklung.' Biol, gen., 7, 163.
(1932a). Elettartam is oregedes. Budapest.

(1932b). 'Das Alterspigment bei einigen Schnecken, untersucht an
iiberlebenden Ganglienzellen.' Arb. wig. Biol. Forsch. Inst., 5, 38.

(1935). 'Senescence and death in invertebrate animals.' Riv. Biol. ,19,
377. (7)
Szabo, I., and Szabc5, M. (1929). 'Lebensdauer, Wachstum und Altern,
studiert bei der Nacktschneckenart Agriolimax agrestis.' Biol, gen., 5, 95.
(103)

(1930a). Todesursachen und pathologische Erscheinungen bei Pulmon-
aten.' Arch. Molluskenk., 62, 123. (19, 103)

(1930b). 'Vorlaufige Mitteilung uber die an der Nacktschnecke
Agriolimax agrestis beobachteten Altersveranderungen.' Arb. ung. Biol.
Forsch. Inst., 3, 350. (103)

(1931a). 'Todesursachen und pathologische Erscheinungen bei
Pulmonaten IL Hautkrankheiten bei Nacktschnecken.' Arch. Molluskenk.,
63, 156. (103)

(1931b). 'Histologische Studien uber den Zusammenhang der ver-
schiedenen Alterserscheinungen bei Schnecken. I and II.' Z- vgl. Physiol.,
15, 329 and 345. (103)

(1934). 'Alterserscheinungen und Alterstod bei Nacktschnecken.' Biol.
Zbl., 54, 471. (103)

(1936). 'Histologische Untersuchungen uber den Zusammenhang
zwischen Langlebigkeit und Fortpflanzung.' Z0°l- Anz. (103)
Szabo, M. (1935a). 'Pathologische Erscheinungen bei Schnecken.' Allattani
Kozl. T., 32, 132. (4)

239

Bibliography

Szabo, M. (1935b). 'On a culture method for the rotifer Lecane inermis Bv.
together with some notes on the biology of this animal.' Arb. ung. Biol.
Forsch. Inst., 8. (85)

Tack, E. (1940). 'Die Ellritze (Phoxinus laevis Bg.): eine monographische

Bearbeitung.' Arch. Hydrobiol, 37, 321. (70)
Tainter, M. L. (1936). 'Dinitrophenol in diet, on growth and duration of

life of the white rat (sic): Proc. Soc. exp. Biol, 31, 1161. (153)

(1938). 'Growth, life-span and food intake of white rats fed dinitro-
phenol throughout life.' J. Pharmacol., 63, 51. (153)
Tanaka, T. (1951). 'A study of the somatic chromosomes in various organs

of the white rat (Rattus norvegicus) especially in regard to the number and

its variation.' Res. Genet., 2, 39. (168)

(1953). 'A study of the somatic chromosomes of rats.' Cytologia, 18,

343. (168)
Tannreuther, G. (1919). 'Studies on Asplanchnia ebbesborni with special

reference to the male.' Biol. Bull. Wood's Hole, 37, 194. (59)
Tegge, M. S. (1936). 'Length of life of a rabbit.' Science, 84, 575. (49)
Terao, A. (1931). 'Change of vitality with age as based on the living unit

of the organism. I. Oxygen consumption in the daphnid Simocephalus

exspinosus.' Proc. Imp. Acad. Tokyo, 7, 23. (143)

(1932). 'Duration of life in the water flea Moina macropa Straus, with

special reference to rate of living.' J. Imp. Fish. Inst., 27, 63. (143)
Terao, A., and Tanaka, T. (1930). 'Duration of life in the water flea

Moina macropa Straus in relation to temperature.' J. Imp. Fish. Inst., 25,

67. (143)
Teresghenko, K. K. (1917). 'La breme (Abramis brama) de la region du

Volga et de la mer Caspienne.' Trav. Lab. Ichthyol. Astrakhan, 4, 2. (72)
Thannhauser, S. J. (1945). 'Werner's syndrome (progeria of the adult)

and Rothmund's syndrome: 2 types of closely related heredofamilial

atrophic dermatoses with juvenile cataracts and endocrine features; a

critical study with 5 new cases.' Ann. int. Med., 23, 559. (135)
Theorell, H., Beznak, M., Bonnighsen, R., Paul, K. G., and Akeson, A.

(1951). 'Distribution of injected radioactive iron in guinea pigs and its

rate of appearance in some hemoproteins and feuritins.' Acta chem.

Scand., 5, 445. (164)
Therman, E., and Timonen, S. (1951). 'Inconstancy of the human somatic

chromosome complement.' Hereditas, 37, 266. (168)
Thoms, W.J. (1873). The longevity of man: its facts and fictions. London. (59,

60, 62)
Thompson, D'Arcy W. (1942). On growth and form. New ed., p. 199. Cam-
bridge University Press. (65, 68, 154)
Thomson, J., and Forfar, J. O. (1950). 'Progeria (Hutchinson-Gilford

syndrome). Report of a case and review of the literature.' Arch. Dis.

Childhood, 25, 224. (134, 135)

240

Bibliography

Thummel, E. (1938). 'Lebensdauer einer Blindschleiche.' £ool. Garten. Lpz.,
10, 153. (52)

Timonen, S., and Therman, E. (1950). 'Variation of the somatic chromo-
some number in man.' Nature, London, 166, 995. (168)

Tomilin, M. I. (1936). 'Death of the oldest chimpanzee in captivity.'
Science, 83, 103. (47)

Tomilin, S. A. (1938). Tomhjihh, C.A.: 'C/raTHCTHKa flOJirOBenHOCTH.' (The
statistics of longevity.) Proc. Conf. ProbL Old Age, Kiev, p. 247. (60, 61, 62)

Townsend, C. H. (1931). 'Growth and age in the giant tortoise of the
Galapagos.' Zoologica, N.T., 9, 469. (77, 78)

(1937). 'Growth of the Galapagos tortoise, Testudo vicina from 1928
to 1937.' Zoologica, NT., 22, 289. (77, 78)

Townsend, F. (1946). 'Ageing processes in the endocrine glands.' J. Geront.,
1,278. (167, 173)

Turner, G. W., and Cupps, P. T. (1938). 'The thyrotropic hormone of the
pituitary in albino rats during growth, pregnancy and lactation.' Endo-
crinology, 24, 650. (186)

Turner, G. W., and Kempster, H. L. (1948). 'Mild hypothyroidism main-
tains egg production with advancing age.' Poult. Sci., 27, 453. (64)

Tyler, A. (1953). 'Prolongation of the life span of sea urchin spermatozoa
and improvement of the fertilization reaction, by treatment of sperm and
eggs with metal chelating agents.' Biol. Bull. Wood's Hole, 104, 224. (35)

Ubisgh, L. (1926). 'Untersuchungen uber Bau, Funktion, Entwicklung und
Regeneration der Rense des Weibchens von Stephanoceros eichorni.' Z> wiss.
Zool., 127, 590. (87)

Vallois, H. V. (1937). 'La duree de vie chez Phomme fossile.' C.R. Acad.

Sci., 204, 60. (112)
van Cleave, H. J. (1934). 'Length of life-span as a factor in regulating

populations.' Ecology, 15, 17. (112)

(1935). 'The seasonal life history of an amphibious snail, Fossaria modi-

cella, living on sandstone cliffs.' Ecology, 16, 101. (103, 104, 112)
van Cleave, H. J., and Chambers, R. (1935). 'Studies in the life history of

a snail of the genus Lioplax.' Amer. Midi. Nat., 16, 913. (102)
van Cleave, H. J., and Lederer, L. C. (1932). 'Studies on the life cycle of

the snail Viviparus contectoides.' J. Morph., 53, 499. (102)
van Cleave, H. J., and Markus, H. C. (1929). 'Studies on the life cycle of

the blunt-nosed minnow.' Amer. Nat., 63, 530. (133)
Vane, F. R. (1946). 'Longevity of mayflies.' Northw. Nat., 21, 252. (57)
Varley, G. G. (1947). 'The natural control of population balance in the

knapweed gall-fly (Urophora jaceana) .' J. Anim. Ecol., 16, 139. (18)
Vasnetzoff, V. V. (1934). 'Versuch einer vergleichenden Erforschung des

Wachstums der Cypriniden.' £00/. ZK 13, 540. (68)

241

Bibliography

Venge, O. (1953). 'Studies of the maternal influence on the growth in

rabbits.' Acta agric. Scand., 3, 243. (129)
Victor, J., and Potters, J. S. (1935). 'Studies in mouse leukaemia: pre-

leukaemic changes in lymphoid metabolism.' Brit. J. exp. Path., 16, 243.

(33)
Vignal., L. (1919). 'Observations sur les Rumina decollata? Bull. Soc. £ool.

Fr., 44, 115. (58,105)

(1923). 'De la duree de la vie chez Y Helix spiriplana Olivi.' J. Conchyl.,

67, 262. (58, 105)
Vischer, A. L. (1947). Old age, its compensations and rewards. London:

Allen and Unvvin. (22, 112)
Vogt, C, and Vogt, O. (1946). 'Age changes in neurones.' Nature, Lond.,

158, 304. (7, 45, 162)
Vogt, W. (1949). The road to survival. London and New York. (192)

Walker, B. E., and Boothroyd, E. R. (1953). 'Chromosome numbers in

somatic tissues of mouse and man.' Genetics, 39, 210. (169)
Walker, D. G., Asling, G. W., Simpson, M. E., Li, C. H., and Evans,

H. M. (1952). 'Structural alterations in rats hypophysectomised at six

days of age and their connection with growth hormone.' Anat. Rec, 114,

19. (187)
Waloff, N., Norris, M. J., and Broadhead, E. C. (1947). 'Fecundity and

longevity of Ephestia elutella Hubner.' Proc. R. ent. Soc, Lond., 99, 245.

(35, 98)
Walter, E. (1922). 'Uber die Lebensdauer der freilebenden Siisswasser-

cyclopiden und andere Fragen ihrer Biologic' £ool. Jb. (Syst. Abt.),

44, 375. (95)
Walterstoff, W. (1928). 'Triton (Cynops) pyrrhogaster 25 Jahre.' Bldtt.

Aquar. Terrar. Kde., 39, 183. (52)
Wardle, R. A., and MgLeod, J. A. (1952). The zoology of tapeworms.

Minneapolis: Univ. Minnesota Press. (56)
Warthin, A. S. (1929). Old age, the major involution; the physiology and

pathology of the ageing process. New York; Hoeber. (5, 7)
Watkin, E. E. (1941). 'The yearly life cycle of the amphipod Corophium

volutator.* J. Anim. Ecol., 10, 77. (112)
Weber, R. (1942). 'Hohes Alter verscheidenen Tiere im Diisseldorfer

Zoologischen Garten.' gool. Gart., 14, 208. (47)
Weidenreigh, F. (1939). 'The duration of life of fossil man in China and

the pathological lesions found in his skeleton.' China Med. J., 55, 34.

(112)
Weismann, A. (1882). Uber die Dauer des Lebens. Jena. (7, 9, 54, 114)

(1891). 'The Duration of Life.' In Essays upon Heredity. Oxford. (54)
Weiss, J., and Lansing, A. I. (1953). 'Age changes in the fine structure of

anterior pituitary of the mouse.' Proc. Soc. exp. Biol., N.T., 82, 460.
(168)

242

Bibliography

Wellensiek, U. (1953). 'Die Allometrie verhaltnisse und Konstruktions-

anderung bei dem kleinsten Fisch im Vergleich mit etwas grosseren

verwandten Formen.' Zool. Jb. (AnaL), 73, 187. (66, 68, 69, 72)
Wendt, A. (1934). 'Verjugung eines Marmormolches.' Bldtt. Aquar. Terrar.,

45,281.
Werner, C. W. O. (1904). Uber Katarakt in Verbindung mit Sklerodermie. In-
augural Dissertation: Kiel. (135)
Weyer, F. (1931). 'Cytologische Untersuchungen am Gehirn alternden

Bienen und die Frage nach dem Alterstod.' £. 2&Bforsdi.i 14, 1-54. (97,

100)
Weymouth, F. W. (1923). 'The life history and growth of the Pismo clam.'

Fish. Bull., 7, State of Calif. Fish and Game Commission, Sacramento.

(58)

(1931). 'The relative growth-rate and mortality of the Pacific razor

clam (Siliqua patula Dixon) and their bearing on the commercial fishery.'

Bull. U.S. Bur. Fish., 46, 543. (58, 106, 107, 108, 159)
Weymouth, F. W., and Thompson, S. H. (1930). 'The age and growth of

the Pacific cockle (Cardium corbis, Martyn).' Bull. U.S. Bur. Fish., 46,

633. (58, 106)
Wheeler, J. F. G. (1934). 'On the stock of whales at South Georgia.' Dis-
covery Rep., 9, 251. (Ill)
Whitney, J. E., Bennet, L. L., Li, C. H., and Evans, H. M. (1948).

'Effect of growth hormone on the N2 excretion and body weight of adult

female rats.' Endocrinology, 43, 237. (152, 186)
Wiesner, B. P. (1932). 'The experimental study of senescence. Brit. med.

J., ii, 585. (180)
Wiesner, B. P., and Sheard, N. M. (1934). 'The duration of life in an

albino rat population.' Proc. R. Soc. Edinb., 55, 1. (19, 43)
Wiesner, B. P., and Yudkin, J. (1952). 'Inhibition of oestrus by cultivated

gromwell.' Nature, Lond., 170, 474. (188)
Wigglesworth, V. B. (1934). 'Physiology of ecdysis in Rhodnius prolixus

(Hemiptera) II. Factors controlling moulting and metamorphosis.' Quart.

J. micr. Sci., 76, 269. (147)

(1945). 'Transpiration through the cuticle of insects.' J. exp. Biol., 21,

97. (97)

(1953a). 'The thoracic gland in Rhodnius prolixus (Hemiptera) and its

role in moulting.' J. exp. Biol., 29, 561. (147)

(1953b). 'Hormone balance and the control of metamorphosis in

Rhodnius prolixus (Hemiptera). J. exp. Biol., 29, 620. (147)
Will, L. C, and MgCay, C. M. (1943). 'Ageing, basal metabolism and

retarded growth.' Arch. Biochem., 2, 481. (150)
Wilson, D. P. (1949). 'Notes from the Plymouth aquarium.' J. mar. Biol.

Ass. U.K., 28, 345.
Wilson, M. A. (1950). 'Duration of life in Rana temporaria Linn.' Brit. J.

Herpetol, 3, 66. (53)

r 243

Bibliography

Wimpenny, R. S. (1953). The plaice. London: Arnold. (69)

Winsor, C. P., and Winsor, A. A. (1935). 'The longevity and fertility of

the pond-snail Limnaea columella* J. Wash. Acad. Sci., 25, 302. (43, 103,

104, 105)
Witschi, E. (1952). 'Gonadotropins of the human hypophysis, particularly

in old age.' J. Gerontol, 7, 307. (181)
Wolfe, J. M. (1941). 'Effects of testosterone propionate on the structure

of the anterior pituitaries of old male rats.* Endocrinol., 29, 969. (135,

173)

(1943). 'The effects of advancing age in the structure of the anterior

hypophysis and ovaries of female rats.' Amer. J. Anat., 72, 361. (135, 173)
Woodruffe, G. E. (1951). 'A life-history of the brown house moth, Hqf-

mannsphila pseudospretella Faint.' Bull. ent. Res., 41, 529. (132)
Wooley, G. W., and Little, C. C. (1946). 'Prevention of adrenal cortical

carcinoma by diethylstilboestrol.' Proc. nat. Acad. Sci. Wash., 32, 239.

(181)
Woolley, G. (1946). In Hamilton, J. B. (1948). Rec. Prog. Hormone Res.,

3, 257. (130)
Wright, M. N. (1936). 'The oldest jennet?' Field, 1936, p. 1556. (48)
Wright, S. (1926). 'Effect of age of parents upon characteristics of the

guinea pig.' Amer. Nat., 60, 552. (126)
Wurmbach, H. (1951). 'Uber Wachstum und Altern der Fische.' £. Alters-

forsch., 5, 277. (46, 53)

Yagob, M., and Swaroop, S. (1945). 'Longevity and old age in the Pun-
jab.' Brit. med. J., 2, 433. (110)

Young, F. G. (1953). 'Growth hormone and diabetes.' Rec. Prog. Hormone
Res., VIII, 471. (186)

Young, T. E. (1899). On centenarians and the duration of the human race. Lon-
don: Lay ton. (60)

Zelinka, C. (1891). 'Studien uber Radertiere — III.' £. wiss. Zool., 53,

323. (59, 165)
Zondek, B., and Aschheim, S. (1927). 'Hypophysenvorderlappen und

Ovarium.' Arch. Gynaek., 130, 1.
Zuckerman, S. (1951). 'The number of oocytes in the mature ovary.' Rec.

Prog. Hormone Res., 6, 63. (180)

244

GENERAL INDEX

Abkhasia, 61-2
Absenteeism, 26
Acceleration

negative, of growth, 11

of senescence, mammals, 153
Accumulation, 35, 91, 98, 154
Acrogeria, 135
Acromegaly, 185, 186
Actinians, 80, 92

Activity, and longevity, 101, 193-4
Adaptation, senescence as evolved,

9,38
Adrenal, 167, 173, 174 ff., 185

cells, cultivation, 171

mitosis in, 167
Adrenalin, 174
Adrenocorticotrophic hormone, 1 74,

176
Adrenopause, 132
Age

evidence of, 45, 59 ff.

maternal, effects, 88, 125 ff.

specific, 25, 36-7, 68, 70, 143, 162
in mammals, 148
in man, 195
inheritance, 121
Aggregation, in rotifers, 91
Agouti, 47
Albatross, 110
Alligator, 79
America, 15, 16, 62
Amphibia

growth, 67

longevity, 52, 53, 182
Amphipoda, 113

Anabolic hormones, 183, 185, 197
Androgens, 178
Aneuploidy, somatic, 168-9
Annelids, 56, 83

Antigen reaction, 170

Ants, 57, 92

Arachnids, 44

Arthropods, 56-7, 92-102

Asexual reproduction, 1 16-17, 83 ff.

Ass, 46, 48

Autocatalytic reaction, 10, 32, 138

Axolotl, 182

Bacon, Francis, 3, 8, 34, 45, 53-4,

59
Bacteria, 7, 164, 170
Baldness, 135
Basal metabolism, age changes, 32,

33
Bats, 47
Bears, 46
Beaver, 47

Bees, 38»., 44, 99 ff., 112, 167
Beetles, 55, 57, 97
Bidder, G. P., 12-14, 38, 41, 64, 67
Bilharzia, 151
Biochemical criteria, of senescence,

30
'Biological' time, 5-6
Birds, 63, 132

life-span, 49 ff., 109-10

ringing studies, 1 1 0

specific size, 12, 160

wild mortality, 1 1 0
Blackbird, 109
Bone maturation 185
Bream, 54, 72-4
Bulgaria, 61, 6\n.
Burns, mortality, 185
Butterflies, 99, 112
'By-product', senescence as, 12-14,
38-9

245

General Index

Calcium accumulation, 35, 91, 150, Collagen, 173

169
Cancer, 25
Capybara, 47
Caracara, 50
Carnivores, 47, 48, 111
Carp, 53, 54

Castrates, longevity of, 133
Castration, effects, 32, 172, 181,

185
Cat, 46, 47, 48, 133, 152, 167,

186
Catalysts, exhaustion, 163-4

genes as, 164
Cattle, milk yield, 26
Caucasus, old inhabitants, 61
Cavy, 21, 28, 44

Cells (see under cell-types and
organisms)
life-span of, 80, 142, 162 ff.
red blood, 166
senescence of, 163 ff.
turnover of, 167, 198
Cell-lineages, indeterminacy, 115
Centenarians, 60-1
'Cerebral death', 95, 99, 100
Chaffinch, 49
Chamaeleon, 45, 52
Chelonians, 51-2, 78-9
Chick, tissue in culture, 12
Childhood, growth, 152, 196

prolongation, effect, 196
Chimaeras, age-, 199-200
Chimpanzee, 46
Chitin, 97-8
Chromosome number, 118-19, 168—

169
Ciliates, 115, 119
Cladocera, 26, 29, 32, 44, 92, 93,

113, 127
Clergy, mortality of, and mating,

133
Clones, senescence in, 116 ff., 169
Cockatoo, 50
Cockroach, 20, 44
Coelenterates, 56, 81
Coleoptera, 44, 57

age changes in, 169

turnover of, 34
Colloids, 34

age changes in, 2, 7, 97
Communities, of cells, 168
Condor, 50

Conjugation, protozoan, 116-17
Copepods, 92
Cormorants, 110
Corpora lutea, 180

pedunculata, 100
Corpus allatum, 167
Cosmic rays, 7
Crane, 50
Crayfish, 54
Cricket, mole, 99
Criteria of senescence, 26-30
Crockery, breakage rate, 22
Crocodiles, 31, 52, 79
Crustaceans, 92 ff., 11?
Cyanide resistance, 81, 83, 115
Cybernetics, 37, 175-6
Cyprinodont fish, 132-3
Cytomorphosis, 177
Cytoplasmic factors

in clonal senescence, 120

and vigour, 129

Daghestan, 61

Daphnids, 32, 92-5, 127, 143-6

Death, 'cerebral', 95, 99, 100

natural, of protozoa, 114
Deer, 40

De-growth, 80, 84
Dehydration, 29
Depletion, 35-6, 91, 98, 102
Diabetes, 186
Diatoms, 153
Differentiation, cellular, 10, 11, 140,

142, 162
Digestive tract, 1 1

of arthropods, mitosis, 91, 96
Di-nitrophenol, 153
Diptera, 44
Diseases, age-linked, 3
Divergent series, 65

246

General Index

Dogs, 33, 49, 152, 153, 186
Dolphins, 47, 53
Domestic animals, 42, 45
Dove, 50
Ducks, wild, 132

Eagle, 50

Ear disease, in old rats, 24

Echidna, 47

Echinoderms, 10, 58

Eels, 53, 54, 70, 112

Egg production, in fowls, 26, 63

Eggs, of Daphnia, 94 ff., 127

fowls', hatchability, 124

parabiosis between, 129

of rotifers, determination in, 85 ff.
Elephant, 34, 46
Elasticity, of skin, 26
Emu, 50
Endocrines

cells, turnover of, 167

mitosis in, 167

and senescence, 7, 171 ff.

senile, histology, 167-8

senile, response to stimuli, 1 74
Endomixis, 116-17
Energy, rate of use, 10

'growth', nature of, 158

requirements, human, 10
England and Wales, centenarians,

62
Entelechy, 8, 190
Enteritis, in old rats, 24
Enzymes, expendable, 35, 163-4

haemoprotein, turnover, 164

renewal in cells, 163-4, 166
Ephemeroptera, 57
Epigenetics, 165
Epiphyseal union, 185
Equilibria

in cell-replacement, 37, 157 ff.

hormonal, maintenance, 175
Equines, 46
Erythrocytes, 166
Eskimos, 134

Evolution of senescence, 9, 10, 31,
32, 37 ff., 161, 190, 191

Exhaustion, reproductive, 2, 69, 99,

132
Experience, selective value, 13, 111
Experimental material

choice, 136-7

supply of, 199
Explosives, ageing of, 34

Family, human, development, 38
Fat-body, deterioration, 98, 101
'Ferment, life', 164
Fingernails, growth of, 156
Fish, 11, 12, 19, 25, 31, 132, 133

growth pattern, 66-77, 154, 159

hormone action in, 161, 182

life-span, 53-4

life-table, Lebistes, 75

reproductive exhaustion, 2, 70
Fitness, Darwinian, 121, 123
Flowering, death after, 2
Follicle-stimulating hormone, 81
Force of mortality, 17, 18, 25
Fowls, 26, 43, 63, 129, 132
Frogs, 52, 64, 67, 182
Fruit-fly, 20, 42, 124-5, 130, 143

Galago, 180
Gall-fly, 18
Gastropods, 43, 104

fresh-water, 112

marine, 105-6
Genes, and catalysis, 164-5
Genetics (genetic)

of ciliate clones, 116 ff.

control of life-span, 121 ff.

of 'June Yellows', 120

timing of puberty, 1 80

of vigour, 127 ff.
Germ-cells, 163
Gerontology, 3, 192, 194
Gerrids, 112
Gibbon, 47
Gigantism

in castrate males, 185

in Hydrobia, 104
Goat, 48
Goldfinch, 110

247

General Index

Goldfish, 72-3

Gonad, limited life of, 3 1 -2

ageing of, 173-4, 177 ff.
Gonadotropin, pituitary, 179, 181,

187, 188
Gonopodium, 182
Goose, 50

Grafts, 171, 179, 180, 199
Gravity, 7
Growth, 8, 10, 12, 14, 150, 153 ff.

asymptotic and indeterminate, 65

cessation, 67-8, 139, 141, 160

of Crustacea, 96

of Daphnia, 92

of Emys, 78

offish, 71 ff.

hormones and, 182

human, 153, 182 ff., 196

of Lebistes, 74

of molluscs, 105 ff.

at puberty, 183

rapid, effect on longevity, 152,
196

retarded, effect on longevity,
187 ff.
Growth curve, 10, 159
Growth hormone, pituitary, 152,

153, 161, 182, 184 ff., 187
Growth potential, decline of, 11,

30-1
Growth rate, experimental alteration

in invertebrates, 143-8

in fish, 74-5

in mammals, 148 ff.

in man, 152, 196
Growth-rings, 55, 107
Guinea pig, 47, 125, 126
Gull, herring, 50, 109
Guppy, 74-7

Hamster, golden, 47
Harvey, William, 62
Hatchability, 124
Hawk-moth, 132

Health, general, inheritance, 125
Heart rate, in Cladocera, 26, 29, 32
in man, 32

Heat production, decline, 33, 187

Helicids, 105

Herbivores, 47

Heredity, and longevity in man,

122 ff.
Heron, night, 27, 44
Heterauxesis, endocellular, 166
Heterogametic sex, vigour of, 1 30
Heterosis, 127-30, 136
Heterozygote, fitness of, 123, 129
Hibernation, artificial, 195
Hippopotamus, 46
Homoeostasis, 36-7, 148, 175-7
Homoeothermy, 14, 161
Homogametic sex, vigour of, 130
Hormones (see individual hormones)

and growth regulation, 182 ff.

in insect development, 146

and senescence, 171 ff.
Horse, 46, 48
Hufeland, 22, 133
Hutchinson-Gilford syndrome, 134
Hybrids

use in research, 136

vigour of, 127-30, 136
Hydranths, 82
Hydration, increased, 30
Hydroids, 81-3
Hydromedusae, 32, 81
Hymenoptera, 57
Hypophysectomy, 153, 187
Hypophysis, 174
Hypoplasia, organ-, in rats, 26

Ibis, African, 27, 44

Imago, insect, 34, 35, 92, 97, 98, 196

'Immortality'

of germ-plasm, 5

'potential', 21, 116, 191
Immunology, and senescence, 170
Inbred animals, variability, 136
Inbreeding depression, 118, 127 ff.
Infant mortality, 21
Inflection, of growth-curve, 159
Inhibitors, growth-, 154-5
Injuries, accumulation of, 35, 36, 64
Insectivores, 47

248

General Index

Insects, 57, 92, 96 ff., 112-13, 143,

146 ff.
social, evolution of, 38
Integrating systems, 141, 187
Invertebrates, 40, 46, 54-7, 79 ff.,

' 112 ff, 143 ff.
Involution, 5
Irreversibility, of cell-differentiation,

11
Isopods, 92
Isoptera, 57

'June Yellows', 120
Juvenile hormone, 147, 185

Ketosteroid excretion, levels, 1 72 ff.

Lamp bulbs, failure rate, 33

Lamprey, 69

Land, migration of vertebrates to, 12

Lapwing, 109, 110

Larvae, insect, senescence in, 146-7

Lepidoptera, 44, 57, 98, 132, 146-7

Leukaemia, in mice, 126

Life, individual, prolongation of

human, 191 ff.
Life-tables, 21 et passim
Limpets, 106, 107, 112
Literature, reviews of, 4
Litter size, decline of, 177
Liver, regeneration of, 155-6
Lizards, 52, 110-11
Lobsters, 92
Locust, 18, 44
Longevity

of amphibia, 52-3

of birds, 49-50

as evolved adaptation, 80

of fish, 53-4

human, 59 ff

inherited factors in, 121

of invertebrates, 54-9

of mammals, 46-9

maximum, 45-6, 48-9

physiological and ecological, 24

of reptiles, 51-2

Loris, 180
Luths, 52

Macaw, 50

Mackerel, 111

Males, preponderant mortality in,

130 ff.
Malnutrition, in man, 151
Malthusianism, 192
Mammals, growth cessation in,
153 ff.

longevity of, 46-9
Man

heredity and longevity, 122 ff.

higher male mortality in, 1 33

maximum life-span, 59 ff.

prehistoric, 112

primitive, and old age, 193-4

specific size in, 12-13

'wild' mortality patterns in, 111
Material, experimental

choice of, genetic, 136-7

supply of, 198
Maternal age, effects, 91, 125 ff.
Mating

effect on longevity, 99, 132, 133

test, 31, 32
Mechanical senescence, 34-5, 92,

97, 163
Menopause, 36, 177-8, 180-1
Metabolic decline, 9-10, 32-3
Metabolic theories, 9-10
Metabolites, accumulation, 35, 154
Metamorphosis, insect, 146 ff.
Metaplasm, 7, 32
Metathetely, 147, 187
Metazoa, senescence of, 9
Methionine, uptake rate, 129
Methylcholanthrene, 118, 126
Mictic females, rotifers, 86, 90, 91
Milk yields, 26
Minnows, 70, 133
Minot's Law, 30
Mitosis, 167 ff.

in arthropods, 92, 96
Molecules, catalytic, genes as, 164
Molluscs, 20, 36, 58, 102-8

249

General Index

Mongolism, 126

Mongoose, 34

Morphogenesis, 8, 10, 11, 64, 139,

162, 170
'Morphogenetic' senescence, 36-7,

97, 140
Mother, age of, 125 ff.
Moths, 98, 99, 132
Motor cars, 'death rate', 20, 33
Mouflon sheep, 28, 44
Moulting hormone, 146, 147
Mouse, 20, 24, 33, 42, 43, 47, 121,

122, 125, 126, 127, 129, 130,

151, 186
harvest, 47

multimammate, 43, 130
Mule, 48
Mussel, 106

Mutations, unfavourable, 118
somatic, 168

Nematodes, 56, 84
Nemertines, 84
Nephra, loss of, 34
Neurones

exhaustion, 45, 140, 162

life-span, 38, 80, 163

loss of, 100

regeneration, 163
Neurosis, induced, 153
Nucleocytoplasmic ratio, 29
Nucleus, in cilia tes, 118-19
Nudibranchs, 36, 102

Oestrogens, effects of, 178
Oligochaeta, 83
Oogenesis, postpubertal, 180
Organ size, determination, 154

weight, in rats, 26
Organelles, 'inheritance' of, 115-16
Osteoporosis, 179
Ostrich, 50

Ova, exhaustion of, and menopause,
180

tissue culture of, 171

transplantation, 129
Ovary, 177 ff.

Owls, 50

Oxygen uptake, 33

Oysters, 104

Palaeolithic man, mortality, 112

Pangolins, 45

Pantothenic acid, 92-3, 145

Parabiosis, 129, 156, 199

Paramecium, 12, 118-19, 168

Parasites, 56, 84

Parkinson, James, 192

Parr, Old, 62

Parrots, 46, 49, 50

Passerines, 49

Pathological theories, of ageing, 10

Pathology, in human old age, 195

Pelecypods, 54, 55, 67, 80, 102, 106-

107
Pelican, 50
Perch, 54

Pets, domestic, longevity, 48, 64
Phagocytosis, 170
Phosphatase, 178
Phylogeny of senescence, 160-1
Pig, 179

Pigeons, 49, 50, 64
Pike, 53, 70
Pineal gland, 155, 175
Pipit, meadow, 110
Pituitary gland

age changes in, 167-8, 173

mitosis in, 167
Plaice, 12, 68-9
Planarians, 9, 10, 32, 83
Plants, senescence, 4, 120

monocarpic, 2, 77
Plasmagenes, 120
Platyhelminths, 56
Polyploidy, of meganucleus, 119
Ponies, 48

Porifera, 12, 56, 80-1
Post-inhibitory rebound, 155-6
Potentialities, selection for, 38
Predation, selective, 25
Press juice, effect on explants, 1 70
Primitive man, 111, 112, 193-4
Probability states, and growth, 157

250

General Index

Progeria, 134 ff.

adult, 135

infantile, 134
'Programme', 41, 138, 139

exhaustion of, 41, 80
Prolongation of life, 195 ff.
Protein, half-life of body-, 34

changes in, with age, 169, 183
Protozoa, 9, 114 ff.
Puberty

growth at, 185

mechanism initiating, 179

pituitary factors in, 180

precocious, 179, 188

senescence consequent on, 150,
196
Punjab, longevity in, 110
Purkinje cells, 163n.

Quakers, 122

Queens, insect, 57, 92, 100

Rabbit, 32, 47, 49, 125, 179
Radar units, failure rate, 33
Rat, 18, 24, 26, 32, 43, 47, 76, 129,
130, 133, 148-51, 152, 153, 155,
167, 180-1, 186, 187
Rate of living, 9-10, 138 ff.

of mortality, 17
Rates of change, decline, 157
Redstart, 109
Redwing, 110
Regulators, of specific size, 12-13

of growth, 153 ff.
Rejuvenation, 178, 196

of planarians, 83
Reproduction, 8, 31, 99, 132, 133

of cellular enzymes, 165, 169

decline with age, 31-2

of organelles, 115-16
Reptiles, 51-2, 67, 77-9, 159
Resistance, declining, senescence as,

11
Reversibility, of senescence, 9-10,

83, 178, 196
Rhabdocoelians, 84
Rhinoceros, 46

Rings, growth, 55, 105, 106-7
Risk, variation in, 18, 21, 25
Robin, 49, 109
Rodents (see also individual species),

47,49
Roman funerary inscriptions, 112
Rotifers, 14, 35, 44, 59, 80, 84-91,

113, 125, 140, 165
Ruminants, 47
Rush, Benjamin, 3
Russia, 61, 129, 170

Salamander, 67

Salmon, 54

Sea anemones, 47-8, 80, 81

Seals, 47-8, 54

Selection, natural, 9, 37 ff., 189 ff.

against senescence, 39, 40
Self-maintaining vertebrates, 67,

159 ff.
Senescence (see under names of
organisms and systems)

contributory factors, 190

criteria of, 26-30

definition of, 1, 4, 7

distribution, 42 ff.

evolution, 9, 10, 31, 32, 37 ff., 161,
190, 191

forms of, 33 ff.

measurement, 17 ff.

physiological, 92

rate of, 1 1

sudden, in man, 135

theories regarding, 7 ff., 189

as undirected process, 190

in wild populations, 108 ff.
Senile change, in individual, 26
Senile mortality, 18, 195
Senility, premature, 134 ff.
Serpulids, 84

Sex differences in longevity, 130 ff.
Sexual activity, effect on life-span.

99, 132, 133
Sexuality, 2
Sheep, 21, 43

mouflon, 28, 44

mountain, 111

251

General Index

Shoebill, 50

Shortness of life, as fitness character,

38
Shrew, 34, 109
Sib-sib correlation, of longevity in

man, 122-3
Simmonds' disease, 135
Size

determinate specific, 12, 38, 67,
70 fT., 154 et passim

and growth, in fish, 53, 65
Skin elasticity, 26
Slugs, 104
Snails, 104, 105
Snakes, 45, 52
Social animals, 38, 40, 1 1 1
Social aspects, of human longevity,

192-3
Soma, evolution of, 9
Spermatogenesis, decline of, 64
Spiders, 44, 56, 92, 130, 131
Sponges, 12, 56, 80-1
Standard of living, and longevity,

63
Starling, 109, 110
Starvation

effect on life-span, 143 ff.
Stresses, random, resistance to, 30
Strawberry plants, clonal degenera-
tion, 120
Sturgeon, 53, 72
Suctorians, 115
Sunfish, Indiana, 71
Supercentenarians, 60-1
Survival curves, 18

human, 3, 21, etc.

types, 19, 20

(see under specific organisms)
Swans, 49

Tadpoles, 139, 155, 161
Tapirs, 46
Tarantulas, 56, 92
Taxi-meter, 141
Teeth, wear, 34, 109
Teleosts, 44, 68, 69, 74, 77
Telephone switchboards, 33

Temperature

coefficient, of growth and develop-
ment, 10, 138

effect on larvel development,
147

effect on life-span, 143
Tench, 54
Termites, 55, 57, 92, 100, 112,

140
Terrapins, 51-2, 78-9
Test, of individual senescence, 31,

198
Testis, 69
Testosterone, 171
Theories, of senescence, 7 ff., 162,

189
Thiouracil, 153, 161
Thrush, 109
Thymus, 31

Thyreo trophic hormone, 186
Thyroid, 64, 149, 153, 161, 171, 183,

185, 186, 187
Ticks, 143

Time, 'biological', 5-6
Tissue culture, 9, 10-11. 171, 200
Toads, 52
Tortoises, 31, 45, 51-2

growth, 78, 79
Toxins, as cause of senescence, 10,

154 ff., 170
Trees, giant, 12

survival, 19
Trout, 54, 70, 71, 77
Tsetse flies, 112

Tumblers, breakage rate, 22, 23
Turkeys, 180
Turnover, of cells, 166-7, 168

of enzymes, 164

of protein, 34
Turtles, 51, 52, 79
Twins, longevity studies, 122

Underfeeding, effects, 143 ff.
Unionidae, 107

Vertebrates, 11

types of life-span in, 13

252

General Index

Vigour, 121

environmental factors in, 129
genetic factors in, 121 ff.
hybrid, 127-30, 136
Viruses, in plant clones, 120
Vole, 27, 43, 44, 47, 80, 108-9
Vulnerability, increase in, 17, 18
Vulture, 50

Weismann, August, 9, 37-8, 114,

163, 189
Werner's syndrome, 135
Whales, 47, 54, 1 1 1
Wild populations, senescence, 108 ff.
Wolfhounds, Irish, 29, 44
Woman, post-reproductive period,

177
Wound healing, rate of, 26, 156

Warthin, 5

Water-heavy, 7

Water-beetles, 130

Wear and tear, cellular, 2, 7, 140,

162
Weight, organ, in rats, 26

Yeasts, 165
'Yellows, June', 120
Young, reproductive preponderance
of, 39

Zebra, Chapman's, 47, 48

253

INDEX OF GENERIC NAMES

Abida, 58
Abramis, 72, 73
Acanthocystis, 1 1 5
Achatina, 58
Acipenser, 72, 73
Acrobasis, 44, 132
Actinia, 56, 81
Adineta, 59
Adocia, 56
,4«fes, 44
Aegypius, 50
Aeolosoma, 83, 84
Agriolimax, 20, 43, 104
v^j-, 57
Alligator, 52, 79
Allolobophora, 56
Amazona, 50
Amblystoma, 156
Amphiuma, 52
Ancylus, 102
Anguilla, 53, 84, 162
Anguis, 52
^4n^r, 50
Anthus, 110
^Ajya, 53
^4/?w, 57
i4/w£/Kj, 86, 88
Aquila, 50
i4ra, 50

Arctocephalus, 48
Arianta, 58
Armadillium, 57
Asellus, 96
Asplanchna, 59, 87
Astacus, 57
Asterias, 58

^50>fltt<2#, 69

Avicularia, 56

Bacillus, 100
Balaeniceps, 50
Balanus, 57
#£to, 53
Bithynia, 102
5/a/tf, 57, 98
Blarina, 47
5/fltta, 44

Amtgyx, 44, 98, 132
Brachionus, 59
Branta, 50
£«fo, 49, 50
£«/b, 26, 52

Cacatua, 50
Callidina, 59, 85, 165
Callionymus, 69
Calliophrys, 57
Callorhinus, 48
Calosamia, 132
Campanularia, 82
C«ra^, 57, 98
Cardium, 58
Carduelis, 110
Caretta, 51
Cepaea, 58
Cereus, 56, 81
Chelodina, 51
Chlorohydra, 82
Clausilia, 104
Cloeon, 57
Clonorchis, 56
Columba, 49, 50
Coracopsis, 50
Corophium, 113
Cojj'wj-, 57
Cricetus, 47
Ctenolepisma, 57

254

Index of Generic Names

Cuora, 51

Cupelopagis, 59, 87
Cy bister, 57
Cyclops, 94

Daphnia, 32, 92 ff., 118, 127, 133,

143 ff., 183, 185
Dendrocoelum, 56
Didinium, 1 1 7
Diomedea, 110
Diphyllobothrium, 56
Dixippus, 100
Dolichotis, 28, 44
Dromiceius, 50
Drosophila, 20, 23, 26, 33, 44, 98, 99,

103, 118, 124, 125, 128, 129,

130, 132, 133, 143
Dugesia, 56
Dytiscus, 57, 97 ff.

Echidna, 47
Echinus, 58
Eisenia, 56
£&/te, 46
£m>tf, 51, 52, 78, 79
Ephestia, 98
Epicrates, 52
Epiphanes, 59
Euchlanis, 59, 89
Eudorina, 116, 117
Euglypha, 115
£«/ote, 58, 103
Eumetopias, 48
Eunectes, 52
Eutermes, 100

Filistata, 56
Floscularia, 59, 91, 113
Formica, 57
Fossaria, 103, 104
Fmcmj-, 159
Fwrn^a, 44, 99

GaZ/ma, 132
Gastrodiscus, 56
Geoclemmys, 51

Geomalacus, 58
Geranoaetus, 50
(Zftfafa, 58, 103
Glossina, 1 1 2
Gobius, 53
Gowra, 49, 50
Gy/w, 50

Habrobracon, 130
Habrotrocha, 59
Haliaetus, 50
Haminea, 58
/Mx, 54, 58, 104, 105
Heloderma, 52
Heterandria, 66, 72
Hqfmanophila, 132
Homarus, 57
Hyalinia, 58
Hydatina, 85
/£«/ra, 44, 81 ff.
Hydrobia, 58, 104
fly/a, 52
Hylobates, 47

Keratella, 59
AWwjto, 50
Kinosternon, 51

Lactobacillus, 164

Larztf, 50

Lasiodera, 57

Lasius, 57

Latrodectes, 44, 131, 133

Latrunculus, 53

Leander, 57

L«£wto, 44, 53, 66, 72 ff.

Lecane, 59, 85

Leptodactylus, 53

Licmetis, 50

Limax, 58

Lim/arca, 43, 58, 103, 104, 105

Lineus, 84

Lioplax, 102

Lithospermum, 188

Z,oa, 56

Lumbricus, 56

Lymantria, 143

255

Index of Generic Names

Macroclemmys, 51
Macrotrachela, 59
Malaclemmys, 51, 78
Maniola, 57
Margaritana, 58, 107
Marthasterias, 58
Megalobatrachus, 52
Megalonaias, 58, 106
Megalornis, 50
Melolontha, 100
Micromys, 47
Microtus, 27, 44, 109
Mirounga, 48
Mniobia, 59
Moina, 143
Mollienisia, 53
Muraena, 53
Afyfl, 58, 106

jVaw, 83
Nasutitermes, 57
Necator, 56
Neophron, 50
Neotermes, 57
Neurospora, 165
Nycticorax, 27, 44
Nymphalis, 57, 98

Ofo/ia, 82
Oniscus, 57
Ophiothrix, 58
Ophisaurus, 52
Ostrea, 58
Ow, 28, 44, 111
0^(y/fl, 58, 105

Paludestrina, 102

Pa/tt'o, 47

Paramecium, 12, 116 ff., 168

Passer, 26

Patella, 58, 106, 112

Patelloida, 106

Pttfeii, 58, 106

Pelicanus, 5

Pelmatohydra, 82

Pelusios, 51

Pennaria, 81, 83

Periplaneta, 44, 99
Peromyscus, 109
Philodina, 59, 87 ff.
Philoscia, 57
PAoca, 48
PAjva, 58, 103
Physocyclus, 57
Pi/fl, 58
Pimephales, 70
Planorbis, 58, 102, 105
Platyarthrus, 57
Pleurodeles, 52
Pleurotrocha, 85
Podophrya, 115
Polyborus, 50
Polygyra, 104
Polyommatus, 112
Porcellio, 57
Prionotheca, 57
/Vofl/w, 20, 59, 85 ff.
Protula, 84
Psalmopoeus, 57
Psammechinus, 58
Pseudogryphus, 50
Psittacus, 50

Quadrula, 58, 107

Pana, 52, 53
flaWitf, 43, 130
Rhinoceros, 46
Rhodnius, 147, 148, 183
Rotaria, 59
Jblgfc 85, 88
Rumina, 58, 105

•Safo/Za, 56
Sagartia, 81
Salamandra, 52
Sflmtfl, 132
Sarcorhampus, 50
Sceleporus, 110
Schistosoma, 56
Si/ifwa, 106
Sitodrepa, 98
Sore*, 47
Sphaerium, 103

256

Index of Generic Names

Sphenodoriy 52
Spirotrichonympha, 115
Spirochona, 115
Stenamma, 57
Stenostomum, 83, 115
Stentor, 115
Stephanoceros, 87, 91
Sternotherus, 51
Streptopelia, 49, 50
Stromatium, 57
Struthio, 50
Sturnus, 110
Suberites, 56

Taenia, 56
Tegenaria, 57
7<?/£tf, 44
Terathiopsis, 50
7>r«fo, 103
Terrapene, 51, 52, 78
TbWo, 51, 52, 78
Teutana, 57
Threskiornis, 27, 44
Timarcha, 57

Tivela, 58
Tokophrya, 115
Tribolium, 44, 130, 131
Tridacna, 54, 106
Triton, 52
Triturus, 52
Trodm, 58, 104
7ttrcfo.y, 110

Uroleptus, 1 16
Urophora, 18

Vanellus, 110
Fmhj, 58, 106, 108
Viviparus, 58, 102, 104
Fa/for, 50

Wuchereria, 56

Xenopus, 52
Xiphophorus, 53, 66, 72

Zalophus, 48

257