© International Epidemiological Association 2001
International Journal of Epidemiology 2001;30:928–934
Printed in Great Britain
REVIEW
Fetal experience and good adult designa
Patrick Bateson
Keywords
Adaptation, design, development, evolution, low birthweight, thrifty phenotype,
maternal induction, alternative lives
Accepted
1 June 2001
Biology is the study of complicated things that give the
appearance of having been designed for a purpose. This thought
has been articulated most clearly in recent years by Richard
Dawkins,1 but it has been expressed in various ways for 200
years. A traveller finding a watch on a mountain path would
not fail to attribute the quality of its design to human agency. A
great British naturalist and theologian, William Paley2 regarded
the design he saw everywhere in nature as proof of the existence
of God. These days, the design to which Paley referred would
instead be attributed by most biologists to blind Darwinian
evolution. The form and behaviour of individuals vary within the
same species and, in any given set of environmental conditions,
some individuals may be better able to survive and reproduce
than others because their distinctive characteristics are particularly well suited to those conditions. If their characteristics are
inherited, then an ever increasing number of individuals in the
population will be better adapted to that environment than was
previously the case.
Plant and animal breeders have known for years how to
select artificially the characteristics which they prized for one
reason or another. Darwin called the blind evolutionary process
leading to the appearance of good design ‘natural selection’.3
His phrase was popular in the 19th century because it suggested
an agent for evolution. These days many biologists prefer to
focus on the processes. What generates variation in the first
place? What leads to differential survival and reproductive success? What genetic and environmental factors enable individual
characteristics to be replicated in subsequent generations? While
each of these questions raises separate issues, it is worth keeping in mind the idea of the evolutionary outcome—apparent
design.
Environmental triggers
After a fire on the high grassland planes of East Africa, the recently
hatched grasshoppers are black instead of being the normal pale
a Based on ‘Design for Living’, an address given at the First World Congress
on Fetal Origins of Adult Disease, Mumbai, India, 3 February 2001.
Sub-Department of Animal Behaviour, University of Cambridge, Cambridge,
UK.
Correspondence: Professor PPG Bateson, The Provost’s Lodge, King’s College,
Cambridge CB2 1ST, UK.
yellowish-green. Something has switched the course of their
development onto a different track. The grasshopper’s colour
makes a big difference to the risk that it will be spotted and
eaten by a bird as the scorched grassland may remain black for
many months after a fire. So matching its body colour to the
blackened background is important for its survival.
The developmental mechanism for making this switch in
body colour is automatic and depends on the amount of light
reflected from the ground. If the young grasshoppers are placed
on black paper they are black when they moult to the next
stage.4 If they are placed on pale paper, however, the moulting
grasshoppers are the normal green colour. The grasshoppers
actively select habitats with colours that match their own. If
the colour of the background changes they can also change their
colour at the next moult to match the background, but they are
committed to a colour once they reach adulthood.
Turtles and crocodiles and some other reptiles commit
themselves early in life to developing along one of two different
developmental tracks and like grasshoppers, they do so in
response to a feature of their environment. Each individual starts
life with the capacity to become either a male or a female.5 The
outcome depends on environmental temperature during the
middle third of embryonic development. If the eggs from which
they hatch are buried in sand below 30°C, the young turtles
become males. If, however, the eggs are incubated at above
30°C they become females. Temperatures below 30°C activate
genes responsible for the production of male sex hormones and
male sex hormone receptors. If the incubation temperature is
above 30°C, a different set of genes is activated, producing
female hormones and receptors instead. It so happens that in
alligators the sex determination works the other way around,
such that eggs incubated at higher temperatures produce males.
(In humans and other mammals, by contrast, the sex of each
individual is determined genetically at conception; if it inherits
only one X sex chromosome it becomes male.)
Each grasshopper and turtle starts life with the capacity to
take one of two distinctly different developmental routes—
becoming green or black, male or female. A particular feature of
the environment determined the path taken by the individual
for the rest of its life, and once committed, the individual cannot
switch to the other route. Once black as an adult, the grasshopper cannot subsequently change its colour to green, just as
a male turtle cannot transform itself into a female.
928
FETAL EXPERIENCE AND GOOD ADULT DESIGN
The broad pattern of an individual’s social and sexual behaviour may also be determined early in life, with the individual
developing along one of two or more qualitatively different
tracks. Many examples are found in the animal kingdom. The
caste of a female social insect is determined by her nutrition
early in life. The main egg producer of an ant colony, the queen,
is part of a teeming nest in which some of her sisters care for her
offspring, others forage, yet others clean or mend the nest, and
finally other sisters specialize in guarding it.6
The sexual behaviour of some primates can also develop
along two or more distinctly different tracks. An adult male
gelada baboon, for example, will typically defend and breed
with a harem of females. After a relatively brief but active
reproductive life, he is displaced by another male and never
breeds again. To position himself so that he can acquire and
defend a harem, the male must grow rapidly at puberty. He
develops the distinctive golden mane of a male in his prime and
becomes almost twice the size of the females.7 However, when
many such males are present in the social group, an adolescent
male may adopt a distinctly different style of reproductive
behaviour. He does not develop a mane or undergo a growth
spurt. Instead, he remains similar in appearance and size to the
females. These small males hang around the big males’ harems,
sneakily mating with a female when the harem-holder is not
paying attention. Since the small, sneaky male never has to
fight for females, he is likely to have a longer, if less intense,
reproductive life. If he lasts long enough he may even do better
in terms of siring offspring than a male who pursues the alternative route of growing large and holding a harem. It is believed
that these two different modes of breeding behaviour represent
two distinctly different developmental routes, and each male
baboon must commit himself to one or other of them before
puberty.7
In each of the above cases, the individual animal starts its life
with the capacity to develop in a number of distinctly different
ways. Like a jukebox, the individual has the potential to play a
number of different developmental tunes. But during the course
of its life it plays only one tune. The particular developmental
tune it plays is triggered by a feature of the environment in
which the individual is growing up—whether it be the colour
of the ground, the temperature of the sand, the type of food, or
the presence of other males. Furthermore, the particular tune
emanating from the developmental jukebox is adapted to the
conditions in which it is played.8 Many other examples can
be given.9,10 The juke-box analogy has its drawbacks because it
implies the tune is pre-formed somehow; like everything else,
the expressed phenotype has to develop.11 The analogy with
‘programming’12 is even worse because it implies that the environmental trigger contains the instructions for the phenotype
that will be expressed. The term ‘imprinting’13,14 has a somewhat similar defect. Perhaps, the best general term for the processes is an old one used in development biology for many years,
namely ‘induction’.
The implication of many of the phenomena described here is
that environmental induction provides a prediction about the
conditions of the world which the individual will subsequently
inhabit. In mammals the best route for such a forecast may be
via the mother. Vole pups born in the autumn have much thicker
coats than those born in spring; the cue to produce a thicker coat
is provided by the mother before birth.15 The value of preparing
929
in this way for colder weather is obvious. Maternal forecasting
by induction is likely to be very important in human biology.
Human development
Is it the case that people, like grasshoppers or baboons, are
conceived with the capacity to play a number of qualitatively
different developmental tunes—in other words, to live alternative lives? Each of us started life with the capacity to live many
different lives, but each of us lives only one. In one sense
individual humans are obviously bathed in the values of their
own particular culture and become committed by their early
experience to behaving in one of many possible ways.
Differences in early linguistic experience, for example, have
obvious and long-lasting effects. By the end of a typical high
school education, a young American will probably know about
50 000 different words.16 The words are different from those
used by a Russian of the same age. In general, individual humans
imbibe the particular characteristics of their culture by learning
(often unwittingly) from older people. When environmental
conditions induce a particular developmental route in animals,
the mechanisms involved are likely to be different; learning
may not enter into the picture at all. Is it possible that some
aspects of human development are triggered by the environment, as though the individual were a jukebox? Was each of
us conceived with the capacity to develop along a number of
different tracks—to live a number of distinctly different sorts of
life? And does the environment trigger the particular developmental track that each of us follows?
The now famous series of studies, led by David Barker,
assessed people across their entire lifespan from birth to death.17
This work has lent strength to the suggestion that human
development may also involve environmental cues that prepare
the individual for a particular sort of environment. Those men
who had had the lowest body-weights at birth and at one year
of age were most likely to die from cardiovascular disease later
in life. Those born as the heaviest babies enjoyed a much reduced
risk of dying from cardiovascular disease. It was only half the
average for the group as a whole, whereas the risk for the smallest babies was 50% above average (in other words, three times
greater than that for the largest babies). Individuals who had
been small babies were also more likely to suffer from diseases
such as non-insulin dependent diabetes and stroke in adulthood.
The analogies with rats are striking. When pregnant mother rats
are given restricted diets, their offspring are smaller and when
given plenty of food they become much more obese than the
offspring of mothers given an unrestricted diet.18
How could a link have arisen between a man’s birthweight
and his physical health decades later? The evidence pointed to
a connection with the mother’s nutritional state. Women with
poor diets during pregnancy had smaller placentas, and 40 years
later their offspring had higher blood pressure (a risk factor
for cardiovascular disease and stroke).17 But the links with
maternal nutrition went much further back than pregnancy.
Measurements of the mothers’ pelvises revealed that those who
had a flat, bony pelvis tended to give birth to small babies. These
small babies, after they had grown up, were much more likely
as adults to die from stroke.17 The implication was that poor
nutrition during their mother’s childhood affected the growth of
her pelvis which, in turn, curtailed the growth of her offspring
930
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
during pregnancy which, in turn, increased her offspring’s risk
of stroke and cardiovascular disease in adulthood.
The evidence took a surprisingly long time to be accepted
given that nobody had doubted that environmental conditions
early in development have a significant impact on many other
aspects of human biology, including size. People are getting
bigger.19 For decades now, the average height of men and
women in industrialized countries has been steadily increasing.
Although some of the height differences between people are
due to genetic differences, the general trend for average height
to increase is almost certainly due primarily to improvements in
nutrition and, to a lesser extent, health. Hence, successive generations of the same family have grown taller despite having a
similar genetic make-up.
The average size of men in the UK has been increasing at
about one centimetre per decade.20 In Russia, where the improvements in nutrition have occurred more recently, the rate of
increase in height lagged behind Britain but has been at almost
three times the rates found in Britain in the last few decades.21
That means an increase of 3 cm a decade. In the US the trend
towards ever taller offspring in successive generations, which
started earlier than in Britain, has levelled off in recent years.
These findings suggest an upper limit on the effect of nutrition
on human height. It is noteworthy that the effects of improved
nutrition on height have been much more pronounced in men
than in women.20
The average age of puberty (as marked by the sudden slowing
of growth that occurs shortly after puberty) has declined more
rapidly in girls than it has in boys.19 This sex difference in
the changing age of puberty has therefore worked in the opposite
direction to that of height, where men have become relatively
taller. Why has the age of puberty in girls apparently responded
more rapidly to nutritional improvements than it has in boys?
It may not be necessary to invoke an adaptive explanation. Girls
develop sexually earlier than boys, so it could be that when
nutritional conditions are bad, girls are simply forced to develop
later and the gap between boys and girls is thereby reduced.
Needless to say, another hypothesis accounts for the sex difference in adaptive terms.
In a rich environment well supplied with food, females are
able to start having children earlier in their lives. Females who
responded to the nutritional conditions by starting to reproduce
earlier would have more offspring than females who did not
respond to those cues. Individuals who were equipped with
the developmental flexibility to adjust the timing of their sexual
maturation in response to environmental conditions would be
at an advantage. This would set in train an evolutionary trend
to establish a jukebox-like developmental response to the quality
of the environment. The developmental rule would be: if conditions are good, become sexually mature early; but if conditions
are poor, delay maturity. The developing male, on the one hand
must therefore strike a balance between starting to breed as
early as possible and developing the physical capacities to cope
with competition from other males. A developing male who
channelled all his resources into accelerating his sexual development, at the cost of remaining smaller and weaker, would be
less well equipped to compete with other males. Without the
physical size and maturity, a sexually precocious male would
be unable to compete with adult males, and would suffer in the
process of trying. An optimum balance is struck when the young
male has grown to a reasonably large size before he becomes
sexually mature. This adaptive analysis fits the observation that
good conditions have led to a larger acceleration in sexual
development in females than in males.
These associations between environmental conditions, body
sizes and mating systems raise some provocative questions
about humans. Is the type of marriage system found within a
culture correlated with the difference in size between men and
women in that culture? And is monogamy more common in
societies living in harsh environments where food is in short
supply? At first glance, the answer is disappointing, since the
ratio of male to female height is roughly the same in both monogamous and polygamous human societies. However, a clear
difference does emerge when monogamous societies are divided
up into those in which the marriage system of monogamy is
imposed by ecological conditions, and those found in more
affluent conditions where it is socially imposed, in the sense
that it is a requirement of the culture. Men and women living
in harsh environments where life is a struggle, such as the high
Arctic and the edges of deserts, are more similar in size than
those living in easier circumstances, and they are more likely to
be monogamous.22 This supports the adaptive hypothesis that
in poor conditions, a monogamous man caring for the offspring
of one woman is more likely to have surviving children and
grandchildren than a polygamous man who has children by
many women.
The thrifty phenotype
Poor maternal physique and health are associated with reduced
fetal growth, with consequences for the offspring’s later health.
The question arises, then, as to whether these connections make
sense in adaptive terms. Could it be that, in bad conditions, the
pregnant woman unwittingly signals to her unborn baby that
the environment which her child is about enter is likely to be
harsh? (Remember that we are thinking here about what might
have been happening tens of thousands of years ago as these
mechanisms were evolving in ancestral humans.) And perhaps
this weather forecast from the mother’s body results in her
baby being born with adaptations, such as a small body and a
modified metabolism, helping the child to cope with a shortage
of food. This hypothetical set of adaptations has been called the
‘thrifty phenotype’.23,24 Perhaps these individuals with a thrifty
phenotype, having small bodies and specialized metabolisms
adapted to cope with meagre diets, run into problems if instead
they find themselves growing up in an affluent industrialized
society to which they are poorly adapted. That, at least, is the
hypothesis.
People who grow up in impoverished conditions tend to have
a smaller body size, a lower metabolic rate and a reduced level
of behavioural activity.14 These responses to early deprivation
are generally regarded as pathological—just three of the many
damaging consequences of poverty. But they could also be viewed
as part of a package of characteristics that are appropriate to the
conditions in which the individual grows up—in other words,
adaptations to an environment that is chronically short on food,
rather than merely the pathological by-products of a bad diet.
Having a lower metabolic rate, reduced activity and a smaller
body all help to reduce energy expenditure, which can be crucial
FETAL EXPERIENCE AND GOOD ADULT DESIGN
when food is usually in short supply.14 If environmental conditions are bad and likely to remain bad, individuals exhibit
adaptive developmental responses to those conditions.
Now this conjecture might well be regarded as offensive. It
could be seen as encouraging the rich to look complacently
at their impoverished fellow human beings, by arguing that
all is for the best in this best of all possible worlds (as Voltaire’s
Doctor Pangloss would have had it). Merely to assert that every
human develops the body size, physiology, biochemistry and
behaviour that is best suited to their station in life would indeed
be banal. The point, however, is not that the rich and the poor
have the same quality of life. Rather, it is that, if environmental
conditions are bad and likely to remain bad, individuals exhibit
adaptive developmental responses to those conditions. To put it
simply, they are designed to make the best of a bad job.
Defending nutritional targets
It is worth bringing in here some more animal work. It has been
known for many years that if a pigeon is deprived of both food
and water, and then given a choice, it will alternate between
choosing food and water.25 The animal efficiently compromises
between meeting a variety of target requirements. Exactly the
same point applies to the various constituents of food. Stephen
Simpson and his colleagues have suggested how the animal
defends a combination of food constituents that provides the
best balance for itself.26 For simplicity, suppose that the target
diet for an individual is 50% carbohydrate and 50% protein.
If it is given a choice of two diets one of which contains 25%
protein and 75% carbohydrate and the other 75% protein and
25% carbohydrate, then the animal will switch its preference
from one diet to the other. The overall effect is that it gets 50%
protein and 50% carbohydrate.
If the animal is given a single diet containing the wrong
combination relative to the target balance, it will reduce its
intake to minimize the risk of imbalance. To be more accurate,
many animals will do so—but with exceptions that are highly
relevant to the matter in hand. A particularly interesting case is
the desert locust—one of the most famous cases where a single
species exists in various phenotypes, the extremes of which are
a solitary phase and a gregarious phase. The expression of
the phenotype depends on population density.27 The phases at
their extremes are strikingly different in their behaviour, appearance and physiology28—so striking, indeed, that at one time
they were treated as different species.
The solitary form is a shy, cryptic animal only flying at night.
Typically, it lives at low densities and avoids other locusts except
to mate. However, when the population starts to become more
dense, the appearance and the behaviour start to change, taking
several generations to complete the full transition to the other
form. The fully formed gregarious phase looks strikingly different. These animals aggregate with each other. They form enormous swarms that fly in daylight. These swarms can contain
hundreds of millions of insects per square kilometre and consume
hundreds of thousands of tons of vegetation per day. These
terrible plagues can extend over 57 countries and cover more
than 20% of the land surface of the Earth.27
The two forms differ in the way they eat. The gregarious form
is a generalist, consuming a wide variety of vegetable matter of
different compositions. The gregarious form can expect to balance
931
its diet because it is on the move through a heterogeneous environment, eating everything it comes across. The solitary form
has little opportunity to rebalance and has to be more careful.
Simpson and his colleagues29 have shown what happens
when the two phases are given an unbalanced diet. The solitary
form utilizes its diet more efficiently and with less cost to itself.
It incorporates a higher proportion of the protein it ingests into
its body. Furthermore, it is more likely to survive until the end
of its period of growth.
Obviously this intriguing story cannot be generalized lock,
stock and barrel to the human case. But it does raise important
questions about human behaviour and physiology that may be
relevant to why small babies are more likely at the other end of
their lives to die from heart disease in an affluent environment.
The people with the thrifty phenotype may be like the gregarious locust in the sense that, when they are given an unbalanced
diet, they fail to regulate properly and do not achieve their dietary targets so readily. It would be possible to test this possibility
in a non-invasive way using the elegant techniques of Simpson
and Raubenheimer.26
Babies with thrifty phenotypes growing up in an affluent
environment may be effectively poisoned and therefore operate
sub-optimally. Some recent research found a strong correlation
between cognitive abilities and low birthweight.30 Rather than
treating this as further evidence of pathology, it is possible that
these people have been polluting themselves with an inappropriate diet. The effects of a diet to which they are not well
adapted could have non-specific effects. These people with thrifty
phenotypes could become lethargic with shorter attention spans
and perform poorly in cognitive tests in an affluent environment. Here again, the hypothesis can be tested.
Why don’t individuals adapt to local conditions in their
own lifetimes? The image of the adaptive landscape used by
Sewall Wright31 in evolutionary biology may be helpful here.
His thought was that in the same environment individuals with
different gene combinations might be equally well adapted (on
equally high mountains using his image), but that going from
one mountain to another entailed a loss of fitness. Engineers
and economists dealing with optimization problems often find
local optima, knowing full well that better solutions can be
found. In the context of the evolutionary adaptive landscape, an
organism may reach the top of one mountain. While it might
be beneficial to cross over to a higher mountain, getting from a
low mountain top to a higher one, involves going downhill
before climbing once again. The same image may be used in development. Once a phenotype is fully formed, it may be difficult
to switch to another phenotype that has become more beneficial
because of a change in local conditions.
A number of important conclusions have emerged from
studies of sensitive periods in development.8 Although the use
of a common term like ‘sensitive period’ should not imply a
common underlying mechanism, it does draw attention to the
basic fact that all forms of experience are not equally important
at all stages in development. A body, once built, is difficult to alter.
Making fundamental changes to mature behaviour patterns
or personality traits will similarly take time, resources and quite
possibly support from others. Adults have important tasks to
carry out, such as feeding and caring for their family, and cannot
readily dissolve themselves and re-construct their behaviour
without others to care for them during the transition phase.
932
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
Accumulating evidence
The thrifty phenotype hypothesis has led to a re-examination
of what happens to people unlucky enough to have been born
into unusually harsh environments, such as those found in a
war. Towards the end of the Second World War, the German
occupation forces in The Netherlands cut off the food supplies
coming into the country. The population of much of The
Netherlands suffered severe food shortages for six months.
Babies born to mothers who suffered particularly badly from
starvation during the final three months of their pregnancies
were born with low birthweights. When these babies grew up,
their capacity to deal with high levels of sugar was markedly
reduced, as would be expected if they were adapted to a world
containing little sugar. One undesirable consequence for these
individuals was an increased risk of developing diabetes, since
they had actually grown up in a much richer, post-war environment.32 Another was that women who had been fetuses in the
last third of pregnancy during the famine gave birth to children
with lower birthweights than normal.33 This second-generation
effect is important when considering the big differences in heart
disease between adjacent countries.
The siege of Leningrad (now St Petersburg) by the German
army in the Second World War lasted from the autumn of 1941
until the end of January 1944 and was one of the most gruelling
sieges in history. The Germans, having invaded the Soviet
Union in June 1941, approached Leningrad and almost completely encircled the city by September 1941, cutting off nearly
all of its supply lines. The resulting starvation and disease, combined with German shelling, killed 650 000 of its inhabitants in
1942 alone. What happened to people who were born during
the siege of Leningrad? Fifty years later, three groups of people
were compared: those whose mothers had suffered from malnutrition while they were pregnant during the siege; a second
group born in Leningrad just before the siege began, who
experienced starvation during their infancy; and a third group
born at the same time but outside the siege area.34 No differences were found in the incidence of heart disease or diabetes
between those whose mothers had been starved during pregnancy and those who were themselves starved in infancy. This
result is still consistent with the thrifty phenotype hypothesis,
since the nutritional state of people in Leningrad after the
Second World War was generally poor compared with people
living in the West. It could be argued that the mothers’ nutritional weather forecasts, made to their unborn children during
the siege, were on the whole reasonably accurate. Furthermore,
although some people among all three groups grew into obese
adults, those whose mothers had been pregnant during the
siege had higher blood pressure than obese people in the other
groups. This suggests again that they were less well adapted to
a world markedly different from that of their fetal life—a world
in which food was in rich supply.
The thrifty phenotype hypothesis suggests a novel explanation for another observation that has long puzzled nutritionists.
People living in France have significantly less ischaemic heart
disease than people living in northern European countries. The
difference is particularly noticeable when France is compared
with Finland. The two countries have comparable average intakes
of dietary cholesterol and saturated fat, yet the mortality from
heart disease in Finland has for decades remained about four
times higher than that in France for both sexes. From 1986 to
1994 the average standard death rate was 354 per 100 000 men
per year in Finland and 94 in France.35 It seems unlikely that
such a large difference can be explained simply in terms of the
different methods of classifying heart disease in the two countries.
Nor does a difference in general health seem plausible as an
explanation. Indeed, over the 1986–1994 period the average
standard death rate from infectious diseases and parasites
was 8 per 100 000 men per year in Finland and 12 in
France.35
Popular explanations for the differences between Finland and
France have focused on nutritional differences such as the
intake of red wine.8 However, in the context of maternal forecasting, an intriguing possibility is raised by the recent histories
of the two countries. Until the end of the Second World War,
Finland was a very poor country, but since then it has acquired
a high standard of living. If a poor forecast during pregnancy
prepares the developing offspring for a poor environment, while
a good forecast prepares the individual to cope with a fatty diet
later in life and the effect may carry over two generations, heart
disease in Finland should start to fall in about 2015. What about
France? After the Franco-Prussian War, the French were
concerned that they were becoming a third class military power
and wished to produce stronger children.36 Accordingly, the
French government supplemented the rations of pregnant
mothers about 50 years before any other country introduced
such measures. Allowing 20 years for the 1880 batch of babies,
which received a forecast of affluent conditions, to grow up and
another 50 years for them to reach the age when heart disease
is likely to strike, the incidence of heart disease should have
started to decline in 1950 and continue to fall as the second
generation effects started to kick in. With the improvement of
nutritional conditions in other parts of Europe, the mothers’
forecasts will increasingly correspond to reality and heart
disease should drop to the low levels reported in France.
If the thrifty phenotype hypothesis is correct then the massive
social experiment of the one-child policy in China could produce
ill effects, arising from a mismatch between body and environment. The only-children resulting from the Chinese government’s
policy tend to be bigger, heavier and better nourished. But
many of these children have been born to mothers who had low
birthweights and thrifty bodies. If the thrifty phenotype idea is
correct, the children may be at great risk from the diseases of
affluence.
By the same logic, people who grew up in good environments
may be at greater risk during periods of prolonged famine than
those who were born as low birthweight babies. And perhaps
children born to affluent parents are more likely to suffer
adverse effects if they adopt rigorous diets in adulthood. In
German concentration camps, anecdotal evidence suggests that
the strong died while at least some of the weak survived.37
After the capturing of the Sixth Army at the end of the siege of
Stalingrad in 1943, chances of survival of the German prisoners
of war was low. Beevor reports: ‘The first to die were generally
those who had been large and powerfully built. The small thin
man always stood the best chance’.38
The public health implications of the thrifty phenotype hypothesis are profound. From 1970 to 1997 the change in nutritional
intake has been dramatic. In the tables of available data given
in the United Nations report on development, 38 countries
FETAL EXPERIENCE AND GOOD ADULT DESIGN
have gone down, but 106 have gone up.39 Three-quarters of
the countries of the world have increased their average energy
intake per capita. Of those that have gone down, some are
countries where people have started to gain control of overeating, some are due to the breakdown of economic order as in
the former Soviet bloc and some represent the terrible effects of
deteriorating local conditions. Analysis of trends in the various
supposed consequences of having a thrifty phenotype in an
affluent environment is vastly complicated by the widespread
use of new drugs that control blood pressure, for example. This
might explain the declining incidence of coronary heart disease
in most developed countries despite the increasing incidence of
obesity. If it remains possible to sustain the heightened standard
of living which has been demonstrated across the world, then
the secular trends will continue and the maladaptive consequences disappear when eventually maternal forecasts become
accurate. Meanwhile, public health objectives would be, first,
to encourage people who were born with low birthweights to
eat a healthy diet. Second, moves should be made to supplement during pregnancy the diets of those small mothers whose
children are likely to grow up in an affluent environment.
Supplementing the diets of those mothers whose children are
likely to remain in a thrifty environment would be counterproductive.
Conclusion
The general point is that humans, like many other animals, are
capable of developing in different ways and, in stable conditions,
their characteristics are well adapted to the environmental
conditions in which they find themselves. The cues for the way
in which the individual develops is provided during sensitive
periods early in development. The mechanisms are largely to be
discovered.40,41 Generally such systems of developmental plasticity
work well, but in a changing environment they generate poorly
adapted phenotypes because the environmental forecast proved
to be incorrect. The argument may be generalized to the effect
of a human mother on her unborn child in the last three months
of pregnancy. The condition of the mother in late pregnancy
may be taken to provide a forecast about the state of the
environment in which the child will grow. This then determines
the pattern of growth and the metabolic pathways of the child.
If the mother’s forecast was wrong, the consequences for the
child can be dire. The hypothesis is that the under-nourished
mother gives birth to a child who is small in size and is adapted
to a thrifty environment. Conversely a well-nourished mother
gives birth to a large baby who is adapted to an affluent
environment.
Clearly, extreme malnourishment has pathological effects.
Kwashikor resulting from extended periods of protein deficiency
is seriously damaging to the individual. However, it is illuminating to treat the human body as the exquisitely well-designed
thing that it is—capable of responding to a very wide range of
environmental conditions. Instead of treating the body’s responses
to early deprivation as invariably pathological, they should be
viewed as part of a package of characteristics that are appropriate to the conditions in which the individual grows up—in
other words, adaptations to an environment that is chronically
short on food. This is not an empty Just-So Story, but a hypothesis that suggests many new lines of enquiry.
933
References
1 Dawkins R. The Blind Watchmaker. London: Longman, 1986.
2 Paley W. Natural Theology: or, Evidences of the Existence and Attributes
of the Deity, Collected From the Appearances of Nature. London: Faulder,
1802.
3 Darwin C. The Origin of Species. London: Macmillan, 1859.
4 Rowell CHF. The variable coloration of the Acridoid grasshoppers.
Advances in Insect Physiology 1971;8:145–98.
5 Bull JJ. Sex determination in reptiles. Q Rev Biol 1980;55:3–21.
6 Wilson EO. The Insect Societies. Cambridge, MA: Harvard University
Press, 1971.
7 Dunbar RIM. Reproductive Decisions: An Economic Analysis of Gelada
Baboon Social Strategies. Princeton: Princeton University Press, 1985.
8 Bateson P, Martin P. Design for a Life: How Behaviour Develops. London:
Vintage Paperbacks, 2000.
9 Caro TM, Bateson P. Organisation and ontogeny of alternative tactics.
Animal Behav 1986;34:1483–99.
10 Lott DF. Intraspecific Variation in the Social Systems of Wild Vertebrates.
Cambridge: Cambridge University Press, 1991.
11 Oyama S, Griffiths PE, Gray RS. Cycles of Contingency: Developmental
Systems and Evolution. Cambridge, MA: MIT Press, 2001.
12 Horton TH, Stetson MH. Maternal programming of the fetal brain
dictates the response of juvenile siberian hamsters to photoperiod—
dissecting the information-transfer system. J Exp Zool 1990;(Suppl.4):
200–02.
13 Patel MS, Vadlamudi SP, Johanning GL. Overview of pup in a cup
model—hepatic lipogenesis in rats artificially reared on a highcarbohydrate formula. J Nutr 1993;123(Suppl.S):373–77.
14 Waterlow JC. Mechanisms of adaptation to low energy intakes. In:
Harrison GA, Waterlow JC (eds). Diet and Disease in Traditional and Developing Countries. Cambridge: Cambridge University Press, 1990, pp.5–23.
15 Lee TM, Zucker I. Vole infant development is influenced perinatally
by maternal photoperiodic history. Am J Physiol 1988;255:R831–38.
16 Pinker S. The Language Instinct. New York: Morrow, 1994.
17 Barker DJP. Mothers, Babies and Health in Later Life. Edinburgh:
Churchill Livingstone, 1998.
18 Jones AP, Friedman MI. Obesity and adipocyte abnormalities in offspring
of rats undernourished during pregnancy. Science 1982;215:1518–19.
19 Eveleth PB, Tanner JM. Worldwide Variation in Human Growth.
Cambridge: Cambridge University Press, 1990.
20 Huh DL, Power C, Rodgers B. Secular trends in social class and sex
differences in adult height. Int J Epidemiol 1991;20:1001–09.
21 Dubrova YE, Kurbatova OL, Kholo ON et al. Secular growth trend in
two generations of the Russian population. Hum Biol 1995;67:755–67.
22 Alexander RD, Hoogland JL, Howard RD et al. Sexual dimorphisms
and breeding systems in pinnipeds, ungulates, primates, and humans.
In: Chagnon NA, Irons W (eds). Evolutionary Biology and Human Social
Behaviour. North Scituate, MA: Duxbury, 1979, pp.402–35.
23 Hales CN, Barker DJP. Type 2 (non-insulin-dependent) diabetes mellitus:
the thrifty phenotype hypothesis. Diabetologia 1992;35:595–601.
24 Hales CN, Desai M, Ozanne SE. The thrifty phenotype hypothesis:
how does it look after 5 years? Diabetic Med 1997;14:189–95.
25 McFarland DJ, Sibly R. ‘Unitary drives’ revisited. Animal Behav 1972;
20:548–63.
26 Raubenheimer D, Simpson SJ. Integrative models of nutrient balanc-
ing: application to insects and veterbrates. Nutr Res Rev 1997;10:151–79.
27 Simpson SJ, McCaffery AR, Hägele BF. A behavioural analysis of
phase change in the desert locust. Biol Rev 1999;74:461–80.
28 Pener MP, Yerushalmi Y. The physiology of locust phase polymor-
phism: an update. J Insect Physiol 1998;44:365–77.
934
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
29 Simpson SJ, Raubenheimer D. Assuaging nutritional complexity: a
geometrical approach. Proc Nutr Soc 1999;58:779–89.
35 Eurostat. Yearbook ‘97: A Statistical Eye on Europe. Luxembourg: Office
for Official Publications of the European Communities, 1997.
30 Richards M, Hardy R, Kuh D et al. Birth weight and cognitive function
36 Law M, Wald N, Stampfer M et al. Why heart disease mortality is low
in the British 1946 birth cohort: longitudinal population based study.
Br Med J 2001;322:199–203.
in France: intrauterine nutrition may be important. Br Med J 1999;
318:1477–78.
31 Wright S. Genic interaction. In: Burdette WJ (ed.). Methodology in
37 Steinberg P. Speak You Also: A Survivor’s Reckoning. London: Allen Lane,
Mammalian Genetics. San Francisco: Holden Day, 1963, pp.159–92.
32 Ravelli ACJ, van der Meulen JHP, Michels RPJ et al. Glucose tolerance in
adults after prenatal exposure to famine. Lancet 1998;351:173–77.
33 Stein Z, Susser M, Saenger G et al. Famine and Human Development.
New York: Oxford University Press, 1975.
34 Stanner SA, Bulmer K, Andres C et al. Does malnutrition in utero
determine diabetes and coronary heart disease in adulthood? Results
from the Leningrad siege study, a cross-sectional study. Br Med J 1997;
315:1342–49.
Photography: Mary Shaw
2001.
38 Beevor A. Stalingrad. London: Viking, 1998.
39 Programme, UND. Human Development Report 2000. New York: Oxford
University Press, 2000.
40 Waterland RA, Garza C. Potential mechanisms of metabolic imprint-
ing that lead to chronic disease. Am J Clin Nutr 1999;69:179–97.
41 Chapman C, Morgan LM, Murphy MC. Maternal and early dietary
fatty acid intake: changes in lipid metabolism and liver enzymes in
adult rats. J Nutr 2000;130:146–51.