Original cartoon by Alexandra Martin
Source: Original cartoon by Alexandra Martin
Public domain image by BurgererSF of a painting in the National Museum in Warsaw.
The Ill-Matched Couple by Lucas Cranach the Elder (1550)
Source: Public domain image by BurgererSF of a painting in the National Museum in Warsaw.

Whenever a prominent rich man weds a much younger nubile bride  —  dismissively called a “trophy wife”  — a media storm usually ensues. An extreme case is the 1994 marriage of 89-year-old oil billionaire J. Howard Marshall to 26-year-old Anna Nicole Smith, a former Playboy Playmate. Yet in the Western world such unions have seemingly been rare for ages. The title of Lucas Cranach the Elder’s 1550 painting The Ill-Matched Couple aptly conveys public condemnation of old men marrying far younger women. The young wife’s hand discreetly dipping into her husband’s purse hints at the hard-nosed financial interest commonly attributed to such a “gold-digger”. In fact, the vast majority of marriages in Western society have only minor age differences, with grooms mostly 2-5 years older than their brides. Trophy wives 20 years younger or more are striking outliers. And the antithesis of a woman marrying a distinctly younger “toy boy” is even rarer. Such marriages face the loss of female fertility imposed by menopause at around 50 years of age. While no such abrupt limit exists with ageing husbands, evidence is mounting that ill-matched couples face reproductive problems that gradually increase with male age.

//onlinestatbook.com/). Project Leader: David M. Lane, Rice University.
Ages of male and female marriage partners compared. Most points lie below the oblique red line indicating partners of the same age, because husbands tend to be a few years (1-10) older than their wives. Women more than 20 years younger than their husbands are “trophy wives”, while men more than 20 years younger than their wives are “toy boys”.
Source: Author’s diagram based on a scatter plot in the public domain, from Online Statistics Education: A Multimedia Course of Study (http://onlinestatbook.com/). Project Leader: David M. Lane, Rice University.

Testes and testosterone

Writers often imply that human male fertility is practically unlimited and that men aged 80 and beyond can still father children with ease. However  —  in stark contrast with the intense attention devoted to menopausal cessation of fertility in women  —  relatively little research has been conducted on age-related changes in reproductive capacity in men. Countering this omission, in several publications published in the 1980s, Larry Johnson and colleagues examined relationships between men’s age, testicular weights and daily sperm production. One 1984 paper compared results for two groups of men, aged 21-50 and 51-80, respectively. Daily sperm production declined with age and was significantly lower, by about 25%, in the older group. Interestingly, entire testis weights showed no appreciable change with age. But closer examination revealed that the tunic of the testis (its protective outer layer) thickened with age and represented almost a third more of testis weight in older men. So testes of men aged 21-50 have notably more sperm-producing tissue  —  more bang for the buck. In a 1989 overview, Johnson reported that circulating levels of certain hormones (LH, FSH, estradiol) are higher in older men, whereas testosterone levels and the capacity of the testis to secrete testosterone when stimulated are reduced. Other age-related changes in human testes include reduced sexual activity and impaired sperm production sometimes resulting in complete absence of sperm.

Scatter plots redrawn from Eskenazi et al. (2003).
Plots of semen volume and sperm concentration for men of different ages, showing lower values especially during the last two decades (60-80). Note that the logarithmic scale used for sperm count compresses a steep decline.
Source: Scatter plots redrawn from Eskenazi et al. (2003).

A 2003 report by Brenda Eskenazi and colleagues further explored the association between semen quality and men’s age. These investigators conducted a cross-sectional study with 97 healthy men aged 22-80 years. Ejaculates were assessed for semen volume, sperm concentration, total sperm count and various indicators of sperm motility. Continuous declines with no threshold effect occurred with all measures. After allowing for interactions between variables, ejaculate volume was approximately halved between the ages of 20 and 80, while numbers of progressively motile sperms declined by almost 5% per year. Importantly, the proportion of men with abnormal semen volume, sperm concentration and sperm motility increased significantly across the age range. Impairment of male fertility is known to occur if sperm concentrations fall below about 20 million per millilitre of semen, and increasing numbers of men dropped below that level, particularly after age 60.

In his 1989 overview, Larry Johnson recorded an association between age-related reduction in sperm production and declining testosterone output accompanied by decreasing responsiveness of testes to stimulation. Peter Ellison and colleagues provided an interesting cross-cultural perspective on this change after measuring testosterone levels in saliva samples from men in different populations: Boston, Congo, Nepal and Paraguay. A progressive decline with age occurred in all cases. However, at age 20 average testosterone levels showed marked differences between populations, highest for Boston and lowest for Paraguay. But the rate of decline was steeper with high starting values, so by age 80 men reached a similarly low level in all populations.

Mutations and chromosomal abnormalities

Adapted from a figure in Crow (1997).
Graphs showing changes in two congenital conditions due to dominant mutations (Apert syndrome; achondroplasia) with father’s age. (Actual data indicated in blue; smoothed curves shown in red.)
Source: Adapted from a figure in Crow (1997).

It has long been recognized that spontaneous mutations in the fundamental genetic material, DNA, occur far more frequently in sperms than in eggs and multiply as a man ages. This occurs mainly because sperm production involves far more cell divisions than egg production. Because more cell divisions have taken place with sperms produced by older men, mutations are expected to accumulate with age. By contrast, no such age effect is anticipated in women. So it is reasonable to ask, as geneticist James Crow did in a 1997 review paper, whether proliferation of mutations in sperms poses a health risk. Advances in molecular genetics permitted direct determination of the parental source of specific mutations, leading to the discovery that certain developmental defects are predominantly attributable to mutations in the father’s sperms. These include Apert syndrome (characterized by malformations of the head, hands and feet) and achondroplasia (a common cause of dwarfism). Crow showed that such mutations become more frequent with male age as predicted and do so faster than the number of cell divisions: Curves fitted to frequencies of Apert syndrome and achondroplasia plotted against paternal age accelerate upwards. Crow concluded: “This suggests that the greatest mutational health hazard in the human population at present is fertile old males.”

In addition to DNA mutations, anomalies can also occur in chromosomes. Perhaps the most widely known human example is Down syndrome, with an additional copy of chromosome 21 in cell nuclei of affected individuals. This is a relatively common congenital abnormality affecting approximately one in a thousand newborns. A striking increase in occurrence of Down syndrome with maternal age is a staple example in genetics text-books. However, little attention has been given to effects of ageing on chromosomal aberrations in human sperms. In a novel approach, in 2001 Elza Maria Prestes Sartorelli and colleagues published a study in which hamster eggs were individually fertilized in vitro with semen samples from five men aged 23-39 and seven aged 59-74. Subsequent examination of chromosomes revealed that numerical and structural aberrations were significantly more common with semen from older donors.

Miscarriage and birth defects

Age-related increase of mutations and chromosomal aberrations in human sperms have implications for miscarriage, as Elise de La Rochebrochard and Patrick Thonneau demonstrated in a 2002 paper. Departing from the customary approach, they considered both maternal and paternal influences. Over 3,000 planned pregnancies ending in birth or miscarriage were assessed in a retrospective study of women aged 25-44 in four European countries. After adjusting for factors such as reproductive history, these researchers confirmed previous reports that miscarriage risk was higher for women over 35. But with such older women the risk increase was much greater when the male partner was aged 40 or more. In a later study published in 2005, the same authors reviewed 19 epidemiological reports examining effects of paternal age. All studies that analyzed the relationship between paternal age and miscarriage identified a greater risk with men aged 35 years or more. Other studies reported evidence for an influence of paternal age on late fetal deaths.

A 2003 report by Harry Fisch and colleagues specifically examined the relationship between paternal age and Down syndrome. Previous studies of possible paternal age effects yielded conflicting results, so Fisch and colleagues examined a much larger sample. They culled data for over 3,400 cases of Down syndrome from 1983-1997 from the New York State Department of Health’s congenital malformations registry. No parental age influence on Down syndrome was seen at ages below 35 years. Age effects were most pronounced when maternal age exceeded 40 years, with a risk of Down syndrome six times higher than for partners younger than 35. Starting out from the apparent change in Down syndrome rates with maternal age, Fisch and colleagues estimated that, for mothers aged 40 years or more, about half was due to the paternal contribution. Fisch and colleagues concluded: "It is our belief that increased paternal age as well as maternal age may be responsible for a wide variety of health problems in children. This effect has been underestimated and warrants further research."

Frequency of Down Syndrome (number of cases per 1000 births) with increasing maternal age. Correction for paternal age greatly reduces the association with mother’s age. Indeed, for maternal ages above 40 the effect of father’s age is particularly pronounced.

Source: Adapted from a figure in Fisch et al. (2003).

Fertility preservation through young men

Much has been written about the steadily increasing age at first birth for women in Western society. One suggested counter-measure is “fertility preservation” in which eggs harvested from a woman in her prime reproductive years are used for in vitro fertilization at some later date. (See my blog post Beating the Biological Clock with Eggs in the Freezer, posted April 11, 2016.) But remarkably little has been said about parallel problems with the age-related progressive decline in men’s reproductive health. Yet partners age in tandem. So we need to consider a radically different approach to fertility preservation, with collection and storage of semen samples from men in their 20s for later use by women who delay childbearing. It is far simpler and less invasive to collect semen samples rather than eggs, and the straightforward procedure of artificial insemination could be used (at least initially) to procure conception at the chosen time. Then again, of course, there is always the option of bypassing the medics altogether by seeking out a Toy Boy  ………


Crow, J.F. (1997) The high spontaneous mutation rate: Is it a health risk? Proceedings of the National Academy of Sciences USA 9:8380-8386.

de La Rochebrochard, E. & Thonneau, P. (2002) Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study. Human Reproduction 17:1649-1656.

de La Rochebrochard, E. & Thonneau, P. (2005) Paternal age: are the risks of infecundity and miscarriage higher when the man is aged 40 years or over? Revue d’Épidemiologie et de Santé Publique 53 Spec. No. 2:2S47-42S55.

Ellison, P.T. (2008) Energetics, reproductive ecology, and human evolution. PaleoAnthropology 2008:172-200.

Eskenazi, B., Wyrobek, A.J., Sloter, E., Kidd, S.A., Moore, L., Young, S. & Moore, D. (2003) The association of age and semen quality in healthy men. Human Reproduction 18:447-454.

Fisch, H., Hyun, G., Golden, R., Hensle, T.W., Olsson, C.A. & Liberson, G.L. (2003) The influence of paternal age on Down syndrome. Journal of Urology 169:2275-2278.

Johnson, L. (1986) Spermatogenesis and aging in the human. Journal of Andrology 7:331-354.

Johnson, L. (1989) Evaluation of the human testis and its age-related dysfunction. Progress in Clinical & Biological Research 302:35-60.

Johnson, L., Petty, C.S. & Neaves, W.B. (1984) Influence of age on sperm production and testicular weights in men. Journal of Reproduction & Fertility 70:211-218.

Lian, Z.-H., Zack, M.M. & Erickson, J.D. (1986) Paternal age and the occurrence of birth defects. American Journal of Human Genetics 39:648-660.

Sartorelli, E.M.P., Mazzucatto, L.F. & de Pina-Neto, J.M. (2001) Effect of paternal age on human sperm chromosomes. Fertility & Sterility 76:1119-1123.

You are reading

How We Do It

Masturbation: Self-Abuse or Biological Necessity?

A dispassionate view of the natural history of self-gratification.

Bigfoot Parents Have Small Brains

The cautionary tale of Australia’s peculiar incubator birds

A Sperm’s Obstacle Course to the Egg

Men and women have quite different tactics for sperm numbers