The menopause clearly marks the end of the reproductive phase in the female, but no comparable event of sudden discontinuity in fertility occurs in the male. Successful paternity in man has been recorded at the age of 94 (Seymour, Duffy & Koerner, 1935). This difference between the sexes is rather surprising, as the male does not seem to age in a fundamentally different way from the female and as, moreover, the death rate for males is higher in all age groups, than for females. This may be related to the presence of only one X chromosome, the eventual defects of which cannot be compensated for by the activity of the homologue. The Y chromosome does not appear to bear vital genes, as it is not essential for life, XO individuals being perfectly viable, in distinction to the YO genotype. Moreover in the absence of male inducers, the phenotype is female, suggesting that the male phenotype is a more complex differentiation. For whatever reason, the lower viability of the male remains a striking fact: males represent only 35% of the age group over 75 years. The persistence of reproductive function into old age is therefore the more surprising. This does not mean, however, that a gradual decrease in sexual activity and testicular function in the ageing male does not occur. From adolescence onwards, there is a continuous decline in sexual interest, arousal and activity, without a sudden discontinuity in any age group.

The human menstrual cycle is characterized by a series of dynamic hormonal relationships that result in the regular maturation of ovarian follicles, ovulation and corpus luteum formation. As methodology has changed and improved, it has become possible to examine these events in greater detail. The simple analysis of menstrual cycle length, when done with care in a large number of women over long time spans, has revealed much about the fundamental biological rhythm of the menstrual cycle. Of particular interest are the predictable changes in cycle length and variability that occur at different stages of reproductive life (Treloar et al., 1967; Vollman, 1977). Basal body temperature records add another dimension to the analyses of menstrual cycle length because they permit estimation of the length of the pre- and post-ovulatory phases of the menstrual cycle and provide some index of the integrity of corpus luteum function.

A postal questionnaire was sent out in September 1977 to all women aged between 40 and 60 years in Ede, the Netherlands. The design of the study was similar to that used by Jaszmann et al. in 1967 in the same locality. The response rate was 66·6%. Data were analysed using theprobit analysis, a segmentation analysis and by applying chi-squared tests to cross tabulations. The median age at menopause was 51·7 years, 3 months later than in 1967. When, however, women who were taking oral contraceptives at the time of the survey were excluded the median age at menopause was found to be virtually the same as 10 years previously. The effect of factors associated with early or late menopause was studied. The results are discussed. The concept of a secular trend is briefly commented upon.

This paper discusses some of the things that demographers know about fertility at later ages of reproduction in developed countries. The nature of the information available through, for example, vital statistics is often a product of history and of administrative requirements, rather than the result of a well-based scheme. Surveys are frequently used to fill the gaps in major sources, but no survey appears to have been carried out specifically on the topic under discussion here. The available information will be discussed under four broad headings: (1) the levels and the trends of late fertility, as compared with fertility at younger ages; (2) the specific characteristics of mothers giving birth at later ages; (3) the specific characteristics of the corresponding births, and the risks for pregnancies at older ages; (4) the question of whether these births are less wanted than others.

This paper focuses upon the social factors regulating sexual behaviour and fertility in the later years of the female reproductive period. Information from India, Africa and Europe is presented on traditional beliefs and constraints concerning the cessation of procreation and the modification of sexual relations in middle age. Data showing the diversity of attitudes and practices in India, Bangladesh, Nigeria, the USA and Australia are discussed. Findings are that, in all the areas examined, coital frequency declines with advancing age although the timing and extent of the decline varies. However, the extent to which procreation is expected to cease prior to the onset of biological infecundity varies markedly even between neighbouring cultures at the same level of development. In Western cultures, for personal reasons, wives expect that childbearing should cease well before the onset of infecundity and can state an exact age at which this should occur. In those African and Asian cultures where there is a social limit to continued reproduction, this is most commonly defined in terms of a life-cycle stage attained by the family as a whole; chronological age is rarely of any importance.

The biological and social factors which may interact to determine the level and pattern of fertility during the last decades of the reproductive life are complex, and because of the paucity of data, the mechanisms regulating fertility at late ages must in part remain speculative. This review focuses on two major questions. First, why is there a decline of fertility prior to the menopause? Second, why does reproduction cease at the menopause and what factors influence the age at which this occurs?

The United States Food and Drug Administration and the International Planned Parenthood Federation recommend that women over the age of 40 should utilize forms of contraception other than the pill. This decision was reached as a result of the 1975 clinical papers by Mann and his associates (Mann et al., 1975; Mann & Inman, 1975) and the epidemiological reports of the Royal College of General Practitioners (1977), Mann, Inman & Thorogood (1976) and Vessey, McPherson & Johnson (1977). Several authoritative bodies believed it prudent to suggest that the age limit for oral contraceptive use be lowered to 35, and even to 30 years (Anon, 1977; Planned Parenthood memorandum, 1977). The investigations of Inman & Vessey (1968) point to a precarious rise in mortality figures for women over age 35 on oral contraceptives.

There are many factors which affect the choice of contraceptive practice in middle age. Co-existing disease is only one and in some ways is the least important. Childbirth itself poses a considerable risk to the health of the mother in as much as maternal mortality rises steeply with age especially after 45 years (Fig. 1); similarly perinatal mortality increases with age (Fig. 2). Together these two factors make the need for careful contraception all the more important. Fertility is reduced with age and this unfortunately lulls some couples into a false sense of security and they may take unnecessary risks.

Potts: There are differences in the risks associated with hormonal contraception in relation to smoking, obesity and previous patterns of thromboembolic disease. In a study carried out in Bangkok and Oxford (Chumnijarakij & Poshyachinda, 1975) comparing the incidence of thrombosis, as measured by 125I-labelled fibrinogen, after hysterectomy or cancer surgery, there was a higher incidence of postoperative thrombosis, measured by objective criteria, in Oxford (12–35%) than in Bangkok (1·7–3·8%). The argument could be made by inference that Asian women have less risks with the pill, although there are no data to prove it either way.

Until recently there has been a paucity of factual data on the effects of the climacteric upon women's sexuality. The basic questions here are: do changes occur in sexual behaviour, sexual interest or responsiveness during the climacteric years? If so, what are the variations and how are they caused?

Two variables used extensively as a measure of sexuality are coital frequency and orgasmic frequency. Kinsey et al. (1953) showed that the degree of sexual activity remained fairly constant in unmarried women up to 55 years of age. By contrast, sexual activity in unmarried men declined gradually from puberty onwards. For married men and women they found that the frequency of sexual outlets declined during the life span. Kinsey et al. emphasized that the decreasing frequency of sexual intercourse and orgasm in marriage does not prove that the sexual capacity of the woman is influenced by her own ageing but could instead be a result of her husband's ageing. These authors were of the opinion that there is little evidence of any ageing in the sexual capacity of the female until late in her life.

Among the conflicts of modern life is that faced by many educated women between the desire to pursue a professional career and the demands of their potential maternal role. Some resolve this by deferring having children. This deferment, coupled with an increased awareness of birth defects generally and of the problems faced by older mothers, leads to an increasing number of queries to genetic advisory services, family doctors, obstetricians, and family planning centres, as to the specific risks in reproduction by older parents. The appreciable risk of death in childbed of earlier days is no longer with us, thanks to modern obstetric techniques. But the child morbidity due to congenital and genetic disorders, already appreciable (Roberts, Chavez & Court, 1970; Roberts, 1975) and increasing both proportionately and absolutely, and the evidence that some of these increase in incidence with age, mean that there is a real basis for the enquirers' concern. The following review of the ways in which parental ageing affects the incidence of such disorders, and of the mechanisms that appear to be responsible, places the increased risks in perspective.

Two American ethnic (sub-cultural) groups of Cuban and Jewish women of comparable socio-economic status were studied in New Jersey, USA in order to show how each of their cultures influences the climacteric. A wide spectrum of cultural variables, associated with each of these ethnic groups, was obtained through a questionnaire about self-image, reproductive history, attitudes toward the menopause, climacteric symptoms, the role of the physician during the climacteric and behaviour, attitudes and feelings about sex. A further differentiation of these variables, associated with each specific ethnic group, yielded an ethnic profile from which ethnic group membership could be predicted.

There is no need to have a workshop to show that fertility falls with age. Nothing works so well as it gets older. One has only to look at the progression of one's teeth, a very visible manifestation of the process of ageing, to make one wonder what goes on inside the reproductive system!

Probably there is no evolutionary disadvantage to having reduced fertility with age because by the time that there is a significant reduction in fecundity the genetic pool which will be added to subsequent generations has already been added. In biological terms, there may be a positive advantage in reduced fertility in older women, since the human infant is dependent on its parents for such a relatively long time and there may be an advantage in the mother surviving for 7, 8, 9 or 10 years after her youngest child is born. In looking for a teleological explanation of declining fertility with age and of a menopause, this is one possible explanation.