Imperial College London Centenary
About Imperial
About ImperialContacts/getting hereAlumniResearchCoursesAbout this site
Select your text size  for this site here: Small Text Normal Text Large Text Extra Large Text

Note: Some of the graphical elements of this site are only visible to browsers that support accepted web standards. The content of this site is, however, accessible to any browser or Internet device.


New research confirms that natural selection is acting on the current human population

See also...
-Ian Owens' home page

For Immediate Use
Monday 23 April 2001

New evidence suggests that natural selection is leading women to have their first child at earlier ages. This is shown to be an inherited evolutionary change that is taking place despite the influence of social factors such as religion and education.

The findings, by a team of British, Australian and American scientists, show for the first time that both genetic and social factors, such as religion and education, are having a profound effect on the timing of human reproduction. These factors are influencing human evolution to a greater extent than at any time in the pre-history of humans.

Writing in the latest edition of the journal Evolution (1), the researchers reveal that women who reproduce earlier in their lives have higher Darwinian fitness (2). They show that this trend is being passed on from generation to generation through natural selection - the process where successful genes pass on to successive generations through evolution.

These findings are different from previous studies of pre-industrialised populations (3), where the age at women start their periods is the best predictor of the number of children that they will eventually produce.

The differences in behaviour are probably due to changes in society that occurred along side industrialisation, such as freely available contraception for women and better healthcare and nutrition for children.

Dr Ian Owens of the Department of Biology and Biochemistry at Imperial College, an author on the paper, described the significance of their research:

"The most important finding is that changes in society, such as freely-available birth control for women and eradication of several important childhood diseases, which have taken place in the last 20 to 30 years will probably lead to genetic changes in humans through evolution.

"We can say this because the sorts of factors that we found that are associated with reproduction now aren't the same ones that people had found in pre-industrial populations."

The findings were made after an exhaustive three-year analysis of questionnaire data from 2,710 pairs of identical and non-identical twins held in the Australian Twin Registry (ATR) (4).

After removing the influence of differences in religion, education and the effect of the baby boom, the researchers still found heritable genetic differences in three key 'life history' traits. These are the age at which women:

  • started their periods
  • had their first baby
  • reached menopause

Each of these traits was found to be genetically associated with 'fitness' - the ability of an organism to reproduce itself because it is well adapted to its environment.

Out of the three traits it is the age that women had their first baby, which is most significantly associated with reproductive fitness. This means that genes influencing an early age of reproduction in the female population will become more common. In other words, women's genes will predispose them to start reproducing earlier.

Dr Owens believes this work has identified a 'genetic phenomenon' that is likely to have many implications in terms of how human behaviour will change over the next few hundred years.

· The influence of social factors on inherited genetic change

Evolutionary biologists have long agreed that the life history of modern humans has been moulded by evolution due to selection that occurred in the past.

Hotly debated however, is whether human life history traits are still under selection, and whether changes in human culture have led to new forms of selection.

Part of the twin data analysis aimed to discover the effect that social, psychological and historical factors had on the number and timing of children born to the 2,710 pairs of twins studied.

The researchers found many of the variations in the threetraits were controlled by social factors such as religion and education (5). For example, Roman Catholic women had 20 per cent higher reproductive fitness than other religions. University educated women had 35 per cent lower fitness than those who left school as early as possible.

"I was staggered by the results we got," said Dr Owens. "When we decided to control for these factors, I wasn't expecting anything to come out of it. I thought, 'let's just run with the analysis'. But there was a massive difference in the number of children born to families with a religious affiliation. Many of the Catholic twins we studied had an average family of five children, where other families were having only one or two children.

"We also found that mothers with more education were typically having just one child at an older age. Their reproductive fitness was much lower than their peers who left school as early as possible. Again, and again, our analyses for these two factors came back with the same results."

The influence of religion and education in family size may seem an obvious finding - but what the scientists found really astonishing was that after controlling for these social factors, genetic changes were influencing the three life traits studied.

"Even after we controlled for these social factors, there was still lots of genetically heritable genetic variation in the three life history traits. This is a really unexpected finding."

However, he cautions against linking this work with the possibility of a eugenic programme for selective human breeding.

"Looking to the future, I would expect to pick up genetic changes within the ten generations (6) since industrialisation. However, what this work doesn't indicate or find, is a genetic marker for human reproduction - so you can't breed for early reproduction from our data. All the traits that we have examined are controlled by interactions between the environment and many genes."

The future work aims to understand more fully, the contribution psychological factors make, says Dr Owens. "We also want to repeat our experiments using twins databases elsewhere, to really put our results into a 'western world' context," he said.

· Twin analysis

The Australian Twin Registry (ATR) is one of the largest twin databases in the world. It is managed by Professor Nick Martin, another author on the paper, and has taken 30 years to build up. It is normally used to study inherited diseases, but evolutionary biologist Dr Owens was the first to suggest that the data could be used to indicate changes in human evolutionary behaviour.

To calculate whether any of these attributes were genetically inherited, the team analysed lifestyle questionnaires previously gathered by the ATR. The register contains details for 30,000 twins. Of these, 2,710 pairs of identical and non-identical satisfied the criteria for inclusion in the team's research.

The data were then assessed using the 'classical twin method'. This method scores traits on a scale that ranges from wholly genetic through to wholly environmental.

The researchers placed these scores into special 'structural models' on a computer, which allowed a number of different influences to be compared within a rigorous statistical framework. The results revealed any correlations between life history traits and reproductive fitness, removing the effect of social and psychological factors.

For further information, please contact:

Dr Ian P.F. Owens
Department of Biology and Biochemistry
Imperial College at Silwood Park
Berkshire SL5 7PY, UK

Taslima Khan
Science Information Officer
Imperial College Press Office
Tel: +44 20 7594 6712

Notes to editors:

1. Paper: "Natural selection and quantitative genetics of life-history traits in western women: a twin study". Evolution Vol. 55: No. 2, pp 423-435.

Available online by clicking here

Authors: K. M. Kirk (a), S. P. Blomberg (a, b), D. L. Duffy (a), A. C. Heath (c), I. P. F. Owens (d), N. G. Martin (a).

(a) Genetic Epidemiology Laboratory, Queensland Institute for Medical Research, Brisbane, Queensland 4029, Australia
(b) Department of Zoology and Entomology, University of Wisconsin-Madison, Wisconsin 53706-1381, USA
(c) Department of Psychiatry, Washington University School of Medicine, St. Louis, Missouri 63110, USA
(d) Department of Biology and Biochemistry, Imperial College at Silwood Park, Ascot, Berkshire SL5 7PY, United Kingdom

2. 'Fitness' is the term that describes the ability of an organism to reproduce its-self because it is well adapted to its environment.

3. In this work, pre-industrial means the years before the twentieth century.

4. Subjects for this study are female twins aged 45 years and over, who were originally recruited for participation in one or more health and lifestyle studies from the Australian National Health and Medical Research Council Twin Registry (ATR), a volunteer register begun in 1978. This register has more than 30,000 twin pairs enrolled and in various stages of active contact. The twin registry is managed by Professor Nick Martin, who is in charge of the genetic epidemiology group at the Queensland Institute for Medical Research in Brisbane, Australia.

The first health and lifestyle questionnaire study, conducted in 1980-1982, was sent to all available twin pairs aged over 18 at that time (that is, born prior to 1965), and responses from 3,808 pairs of twins and 567 singles were received. The second study, conducted in 1988-1990 and focusing on those twins who responded to the original survey, obtained follow-up data from 6,327 individuals. A further questionnaire, also concerned with health and lifestyle issues, was mailed between 1993 and 1995 to all eligible twins over the age of 50 and registered with the ATR, including, but not limited, to those who had participated in the other two health and lifestyle studies. In each of the three questionnaires, the study participants were asked to provide information on the number, sex, and dates of birth and death of their children and the timing of three life history events: age at menarche, age at first reproduction, and age at menopause.

5. To estimate the role of maternal and/or cultural effects, information on each individual's religious affiliation and level of education was also collected. Educational background was represented by six categories (£7 years, 8-10 years, or 11-12 years; post-secondary apprenticeship or certificate; post-secondary technical or teaching diploma; and university degree or higher). Religious affiliation was collapsed into four categories (no religion, non-Catholic Christian [e.g. Protestant, Orthodox, Evangelical, or Fundamentalist], Catholic Christian, and other religion [non-Christian]).

To estimate fitness of individual twins, the researchers needed to be confident that each twin studied had completed their reproduction. Since many of the subjects were surveyed at multiple time points, it was possible to empirically determine an age at which this could be assumed to have occurred.

6. In general terms, a generation represents 21 years.

7. Imperial College of Science, Technology and Medicine is an independent constituent part of the University of London. Founded in 1907, the College teaches a full range of science, engineering, medical and management disciplines at the highest level. The College is the largest applied science and technology university institution in the UK, with one of the largest annual turnovers (UKP339 million in 1999-2000) and research incomes (UKP176 million in 1999-2000). Web site at