The purpose of the research reported here is to examine the connection between contraception and those aspects of a woman’s position that are related to her marriage. The research was conducted in two villages among the Pare of northern Tanzania where a shift from hoe cultivation as primary occupation to wage labour has brought about major changes in social relations. The major hypothesis is that a change from a ‘traditional’ marital union to a ‘companionate’ marriage is instrumental in the acceptance of contraception and in lowering fertility. The latter type of relationship between marital partners is related to the status of women. The research methodology consisted of a combination of an ethnographic study, demographic surveys and in-depth interviews. Findings show that approximately half of the women in this community ever used contraception. Of current users, a third are sterilized and half are using a modern reversible method. The determining factor for using modern reversible methods is the nature of the conjugal union.

Contraceptive failure rates for modern methods including sterilization are reported to be high in China, but little is known about the consequence of contraceptive failure and characteristics of women who decide to have an abortion if a contraceptive failure occurs. Using 6225 contraceptive failures from the 1988 Chinese Two-per-Thousand Fertility Survey, this study examines the resolution of contraceptive failure and assesses the impact of some women’s sociodemographic characteristics on the decision to terminate contraceptive failure in abortion. This study has three important findings: (1) The abortion rate was 50·1%, 75·3% and 80·2% for IUD, condom and pill failures, respectively; (2) The abortion rates differed by contraceptive method and women’s social and demographic characteristics. In particular, a woman with just one child was most likely to have the contraceptive failure aborted; (3) Some women experienced repeated abortions because of contraceptive failure. The results suggest that abortion was a backup method if contraception failed in China and the correlates of aborting an unwanted pregnancy reflect the strong impact of the Chinese family planning programme.

Due to their effect on maternal testosterone levels, sons are said to have reduced maternal longevity in pre-industrial humans. This analysis, using information from a Flemish agricultural village in the 18th–20th centuries, confirms the presence of a negative effect of sons on maternal longevity. However, the effect is mainly observed for mothers belonging to the least privileged social group and for sons surviving their fifth birthday. Both findings make the above-mentioned biological explanation relative. However, a plausible alternative, social interpretation is male-dominated intra-household resource competition. It is reasonable to assume that only sons above a certain age are able to claim a serious amount of resources and that competition is strongest within the least privileged social group.

The Old Order Amish are a healthy and well-nourished natural fertility population, so that the timing of births is not influenced by behaviours to limit family size, undernutrition or disease. The present study examines the monthly distribution of 8160 births occurring between 1920 and 1991 in the Geauga Settlement in north-east Ohio, USA. The monthly distribution of births in the Geauga Settlement is bimodal, with a major peak extending from August to October, a minor peak in February, and a major trough from April to June. This pattern is almost identical to the pattern found in the US in 1943. The monthly distribution of first births appears to be influenced to some extent by a highly significant seasonal pattern of weddings. The pattern of births in the Old Order Amish is consistent with the hypothesis that the spring trough in US births is at least partially caused by a decrease in coital frequency and/or a decrease in fecundability as a result of hot summer temperatures but is not consistent with the hypothesis that the fall peak in US births is primarily due to an increase in coital frequency during the Thanksgiving and Christmas holiday seasons.

The compact spot-check method was applied to elucidate the daily time and space use patterns of 121 adults from a Bangladeshi village in which the authors had previously found more skin manifestations due to arsenic poisoning in males than females. The analysis of 2178 spot-check data, which were obtained over a net period of 2 days, highlighted that farming and cash-earning activities were almost exclusively conducted by males. This is probably due to the Muslim Bangladeshi norm for females to be ‘sedentary’ in or around their own dwellings. Discussion focuses on the advantages of the compact spot-check method and the possibility that males’ larger energy expenditure and longer exposure to sunlight are responsible for their severer skin manifestations.

The relation between individual trait differences, social mobility and social structure is central to social biology. Because genetic variance underlies phenotypic variance in some of these traits, for example IQ, several mechanisms determine the population variance. Polygenic inheritance is the basic mechanism. Social mobility and assortative partner choice distribute the trait variance within generations. This feedback circle is constrained by sociological conditions at several levels of analysis. Fundamental to this theory of social assortment is the relation between social–biological traits and social class on the one hand, and these traits and social mobility on the other hand. The focus here is on the relation between social class, social mobility and cognitive ability. The National Child Development Study is drawn upon, including the last follow-up (1999–2000). By approaching this relationship through various methods, both social–biological and sociological aspects of this research question can be assessed.

Bangladesh has been passing through a crucial phase of fertility transition. The level of fertility declined dramatically during the early 1990s without any remarkable improvement in socioeconomic and health status, and then remained constant at a high level of 3·3, despite the increased use of contraception. Such fertility transition can be traced to variations in one or more of the proximate or direct determinants of fertility. This paper critically analyses the fertility levels in Bangladesh with a view to exploring the possible explanations of fertility decline in the 1990s and then its stabilization. The main focus of the study is to examine the role of the major proximate determinants of fertility in bringing about the change in fertility level in Bangladesh. The data for the study come from a series of nationally representative surveys over the period of 1975 to 1999–2000. The analysis indicates that fertility has temporarily ceased to decline in recent years due to the ‘tempo’ effect of high past fertility, but in general a declining trend in fertility is underway. The analysis suggests that the fall in fertility is consistent with the underlying trends in most important proximate determinants of fertility. In recent years contraception has emerged as the highest fertility reducing factor in Bangladesh and its effect is greatest in middle and older age groups. Although until the early 1990s postpartum lactational infecundability was the most important and strongest fertility reducing factor in Bangladesh, in recent years its fertility inhibiting effect has gradually decreasing owing to the declining trend in the lactational infecundability period. The analysis reveals that although the fertility reducing effect of the marriage pattern is increasing, its effect is offset by the declining trend in the lactational infecundability period. A review of these two variables suggests that their effect cannot be raised much for prevailing socioeconomic and cultural reasons, and any future reduction in fertility in Bangladesh may be largely dependent on increased use of effective birth control methods.

Elsewhere the author has suggested that adolescent and adult male homosexual orientation is, in some cases, causally associated with sexual or quasi-sexual experience in childhood (James, 2004). Here it is argued that the available data on men raised by same-sex parents cannot validly be interpreted as supporting or refuting this suggestion.