Questionnaire data were obtained from the homes of 467 boys and girls in special education classes for educable mentally retarded children and minimally brain injured children which allowed six social class factors to be assessed. For four variables the scores for the two disability groups were significantly different, indicating for the minimally brain injured children a more favourable parental occupation, parental education, and social position, and indicating for the educable mentally retarded children a larger number of children in the family. No differences were apparent between the groups when considering the birth order or the month of birth of the children in the project. The influence of sex as an independent variable was not significant in this research.

In Australia, Belgium, Denmark, England and Wales, Holland, Italy, New Zealand, Norway, Scotland, Sweden and Switzerland there was a substantial decline in dizygotic twinning rates during the 1960s. In Portugal, Spain and Japan there was a lesser decline and in the US there was little change. In many of the countries for which evidence is available, it seems that these declines all started at about the same time in the late 1950s. It has been suggested that the recent decline in Scotland occurred mainly in middle class women; in contrast it is suggested here that the earlier decline in the US during the years 1933–58 was more characteristic of lower class women. For England and Wales and for Scotland, the recent decline (unlike that in the US 1933–58) occurred about equally in young and older women, so that the causes of the declines in the US in 1933–58 and in other countries at present may be different. In the absence of hard evidence, it is speculated that the current declines may be due to hormonal or pesticidal substances widely used in agriculture.

Monozygotic twinning rates have shown no similar trend so that pregnancy wastage does not seem to be implicated.

The patterns of growth of various components in the populations of selected cohorts in Australia are compared with the corresponding information for selected calendar years. The use of three-dimensional diagrams demonstrates essential differences between the age structures for cohorts and those for the entire population and shows the way in which the two can be related.

This study indicates that tracing of persons through their lifetimes in this way can lead to a greater understanding of trends in calendar year populations, as well as providing a closer link between the historical viewpoint and statistical analysis. This approach is of particular importance in Australia where the pattern and timing of demographic events have produced interesting differences between trends in cohort age structures and calendar year populations.

The attachment of the infant to the mother can be seen in open parkland when an infant is set free for exercise and play. In an area of safe space, a mother has little cause to restrain her child, nor, owing to other attractions around, to entertain or otherwise divert him. In order to deduce which of the women and infants make up kinship pairs, the observer postulates criteria for recognizing a pair; attachment is then conceived of as factors which identify a mother and her infant in field conditions.

The infant detaches himself from his mother but he maintains proximity to her, particularly around the age of 2 years, when his proximity- and contact-seeking are at their height. Yet, at this age, the infant shows most alertness to remote events. From 2 years old onwards, the presence of age-mates competes for the infant's proximity. Hence, the statistical data presented below is from observations which were made when the likelihood of interruption by neighbouring people was minimal.

Proximity to the mother can be defined as the child's being at rest beside her, or as his playing in her vicinity, or as a pattern of exploratory locomotion with the mother the centre of the territory he covers. In each of these aspects, the behaviour of 1-, 2- and 3-year-olds is distinguished by typical distances from the mother.

The infant's responsiveness to unexpected sights or sounds is inferred from the orientation responses he makes to them. In his postures and the distances he takes up from the mother, his feeling-states of alertness, fear, anger, uncertainty, fatigue and conflict are expressed; when thus activated, he tends to relate himself to his mother.

A study of first attendances at FPA clinics in the defined population of a London Borough showed that the population served was not dissimilar by social class from an appropriate sample of the population of the Borough. Almost two-thirds of the women were prescribed the contraceptive pill at first visit, and the other third were almost equally divided between the diaphragm and the intra-uterine device as a contraceptive method. Use of the pill was inversely proportional to age, whereas the use of the cap and IUD increased with age at first attendance. Women prescribed the pill or IUD tended to continue with the method, whereas 11% of women who initially used the diaphragm changed over to the pill in approximately a 1-year follow-up period. A large proportion of women, 21%, attended the clinics only once, but this was shown to be due, for the most part, to factors which were outside the control of the clinic.

The distribution of the interval between (say) first and third confinements is the product of the distributions of the intervals between first and second and second and third confinements provided these two confinement intervals are independent. In this paper we use some specially provided NSW data to consider the question of the independence of the intervals between various confinements. It is shown that there is some indication of a positive association between confinement intervals, but that this association is not very great.

It seems that women whose first-born children are very heavy (9 lb and more) or very light (less than 5 1b) have more subsequent children than other women. This finding does not appear to arise from any of the possible artifacts discussed, and it could have some bearing on the present secular increase in the height and weight of the population.

Analysis of skin pigmentation, by reflectance spectrophotometry at standard wavelengths, of husband/wife pairs of Sikhs living in Britain shows positive correlations between spouses in pigmentation of forehead, but not of inner upper arm. Several possible explanations are considered, the most likely appearing to be positive assortative mating in respect to exposed pigmentation.

There is some evidence, mostly of an epidemiological nature, that coital rates may be positively associated with dizygotic twinning. It is suggested that this link is causal, in which case an explanation becomes available for some curious—and hitherto unexplained—features of dizygotic twinning rates by maternal age and legitimacy.

The common belief that childlessness is positively associated with instability of marriage derives from official statistics which appear to show higher divorce rates for childless than for fertile couples. The official findings are a procedural artifact, however, and the relationship shown is certainly exaggerated and possibly spurious.

The appropriate strategy in determining whether divorcees are relatively infertile is to find a population with disrupted marriages and measure its fertility experience. In doing so it is necessary to take note of the definition of fertility, the remarriage factor, and (most importantly) the duration of marriage. The official statistics use the conception of legal duration of marriage. Since this ignores separation before divorce, its use exaggerates the infertility of divorcees, as does failure to exclude remarriages. De facto duration of marriage (wedding to separation) is a superior statement of opportunity to conceive, and calculations on this basis eliminate or even reverse the fertility differential between divorcing and stable couples. Local figures are used to illustrate this effect, and the finding is supported by evidence from a sample of marriages ending in legal proceedings lesser than divorce. It is concluded that the alleged relationship between childlessness and instability of marriage is probably either non-existent or the reverse of that normally assumed, and that in any case measurement of the net overall effect of childlessness does not provide a helpful datum. An alternative strategy of research is suggested, which would seek patterns of effect rather than net overall effect, taking heed of relevant variables, and also considering all forms of marital breakdown including marriages in which cohabitation continues only with disharmony.

A prevalent feature of American society is the strong desire to live in areas of low population density and close to nature, a phenomenon which can perhaps be most appropriately termed a ‘rural ideal’.* It is paradoxical that this phenomenon has been little studied and totally disregarded in quantitative research, for it apparently has a long-standing and central place in American history (Schmitt, 1969; White & White, 1962). The ideal, which some would argue conforms to a basic biological need of the human species (e.g. Iltis, Loucks & Andrews, 1970), may help to explain and predict such social phenomena as migration to the suburbs (Schmitt, 1969; Strauss, 1961), the growth and prevalence of recreational activities which bring the individual into closer contact with nature and/or into lesser contact with other people (e.g. camping, fishing, hunting, visits to national parks) (Schmitt, 1969), and rising rates of deviant behaviours such as neuroses and violence with increasing population density (Ehrlich & Ehrlich, 1970).

This is the report of a 5-year follow-up survey of 350 couples married in Hull in 1965–66 and first interviewed shortly after marriage. It is based on the results of a second set of interviews conducted in 1970–71. The couples' initial intentions on family-building and birth-spacing are compared with their fertility experience during the first 5 years of marriage. Changes in patterns of contraceptive usage during the period are explored and both family-building and contraceptive practice are examined in relation to those variables which were found to influence the original intentions as expressed in the first interview.

In England and Wales in 1938–68 the sex ratio of children liveborn or stillborn to women aged 45+ years was lower than the sex ratio among children born to women at any other age. This strengthens the suggestion that the high sex ratio reported by Takahashi among children born to elderly women in Japan is a spurious consequence of false registration.

Birth weight correlation in full sibs is estimated from data for 4229 sibships. It is shown that, when it is necessary to adjust individual observations before the calculation of familial correlations, the genetic implications of the method of choice should be examined. It is shown that full sib correlation increases from 0·47 to 0·52 for unadjusted birth weights when the weight of the first born is excluded. After adjustments for sex, parity, gestation and changes in maternal weight a correlation of 0·56 is obtained.

An analysis of the monthly distribution of births in two areas of Matlab Thana, East Pakistan, indicates that there is a seasonal variation in births different from what would be expected by chance. The highest proportion of births occur in the last three months of a year and the lowest proportion between May and July. Investigation into some of the environmental and social factors which might contribute to the seasonal pattern revealed the following: mean minimum monthly temperature 9 months before birth was inversely related to the number of births; all occupations had seasonal patterns different from what would be expected by chance and the business and mill-and-office occupations had distributions significantly different from each other; the distribution of births for all pregnancy orders was different from chance and the distribution for first order pregnancies was significantly different from those for third and fourth or higher orders.

The familiar and well-established negative association between a woman's age at marriage and family size has received somewhat less attention in recent years. No doubt this is because, with the reduction in fertility of the past century, the observed differentials appear less striking and seem less significant. Age at marriage is a less obvious determinant of family size than in the past. Glass & Grebenik (1954), commenting on the data produced by the 1946 Family Census, emphasized the decreasing importance of the association both absolutely and relatively. Contrasting the completed fertility of the 1900–09 and the 1925 marriage cohorts in Great Britain (Table 1), they argued that, though for all, married women the relative influence of age at marriage was unchanged, the figures for fertile marriages only (those where the wife had at least one child) appeared to indicate ‘that age at marriage and fertility were not quite so tightly linked for the more recent group’. And they added ‘this is the kind of development which would be expected with the increasing spread of family planning’.

In the second and third quarters of 1970, 155 women applied for abortion to two of the three consultant gynaecologists at the Royal Northern Hospital, London. Of the applicants, 100 were single, forty-two married and thirteen widowed, divorced or separated (WDS). The single women came disproportionately from larger than average families, and fifty-eight were from broken or unhappy homes. Of the total sample, 39% were Church of England and 30% Roman Catholic. By occupation, the largest group was of semi-skilled or unskilled workers (thirty-seven) and the smallest one of professional women (seven). Less than half were Britishborn and one-third were from the Commonwealth. More than half had never attempted to use any method of contraception. The 155 patients between them had had fourteen previous abortions and twenty-five illegitimate children.

Of the 155 women, seventy-one single, twenty-three married and eight WDS obtained an abortion, this proportion of single women being considerably higher than the national average. So also was the proportion of women in the age range 20–34 years. Slightly more women who were Britishborn and C of E by religion obtained abortion than were represented in the sample as a whole. A number of clinic and hospital details were recorded, such as stage of pregnancy at first hospital interview and at abortion, delay in securing hospital appointment and bed. A follow-up, about 18 months later, showed that nearly all the women were glad they had had an abortion; very few had experienced serious depression, feelings of guilt or religious problems afterwards. Most of the women who needed it were now using birth control, mostly the pill. Three patients became pregnant again within a year of their termination. One had a miscarriage and one (a schoolgirl) had a second termination. The third applied for a second abortion, but was refused.

Fifty-three women, approximately one-third, did not have a hospital abortion, but only fifteen of these were categorically refused; the others had spontaneous or private abortions, changed their minds after interview or failed to appear for the operation. Of the fifteen, seven were refused because the pregnancy was too far advanced. Two girls were refused because they had had previous unwanted pregnancies, but had not used contraception on this occasion; one of these was 19 years old and already had had two abortions and an illegitimate baby. The remaining six were refused for various reasons.

Some general conclusions are drawn from the survey and some recommendations made.