The Institute's thirtieth annual symposium was held on 23 & 24 September 1993. The twelve papers will be published in full as a single volume in association with Macmillan Press.
Professor Ellison’s team of researchers has made daily collections of saliva samples from women in various parts of the world. The samples have been analysed for progesterone and oestradiol, which reflect important quantitative and qualitative aspects of human fecundity. Oestradiol has an important role in maturation of the follicle and the success of fertilisation while progesterone is crucial to successful implantation and maintenance of pregnancy.
This research allows investigation of variation in ovarian function over a wide spectrum. There are three main sources of variation:
The three levels of variation analysed are:
There is very little random variation in an individual’s ovarian function. Variation between women is three times greater.
Age variation is very consistent within and between populations, showing a steady decline after the age of thirty. Even where there is a significant variation in levels between populations, the pattern of decline is very similar.
There is a great deal of information from Western societies on the inhibitory effects of vigorous exercise (eg by professional athletes) and anorexia on ovarian function. These tend to be extreme, rather artificial effects. In communities subject to natural cycles in energy balance and labour demand there are more gradual, seasonal variations in ovarian function, mirrored by seasonal peaks and troughs in births. It is therefore apparent that suppression of ovarian function is not a pathological condition but a designed and adaptive response to changing ecological and economic circumstances.
As noted in the previous paper, ovarian function is affected by nutrition. What is the mechanism and what are the key nutritional factors?
Energy balance is not the only relevant factor. For example, among women of the same body weight and the same level of calorie intake, vegetarians have a much higher level of menstrual irregularities than omnivores.
Experiments with rhesus monkeys show very short term variation in luteinising hormone in response to food restriction (eg missing one daily meal). However, much more research is needed to elucidate general mechanisms and the roles of body weight and fat level and distribution. Although a lot of work has been done with athletes, it is not yet clear how relevant this is to the sort of work done by women as part of their daily routine.
Breastfeeding is the most natural of contraceptive methods and is totally reversible. It produces well-spaced and healthy children and healthy mothers. However, the degree of postponement of menses and ovulation varies and so doctors do not regard it as a reliable method.
In order to investigate the causes of this variation, breastfeeding women in Edinburgh were given diaries and their urine was sampled regularly. It emerged that supplementary feeding is a crucial factor - unnoticed by mothers, babies reduce suckling frequency and fertility resumes.
Prolactin is the endocrinal stimulant for milk production but, although its level builds up during pregnancy, it is only effective when levels of oestradiol and progesterone fall after birth. Every time the baby suckles, prolactin is released. Only 20% of milk in the breast is released directly by suckling; but suckling - or even the expectation of suckling - stimulates oxytocin production which causes expulsion of the rest of the milk.
During the normal menstrual cycle, follicle maturation is stimulated by luteinising hormone (LH). LH is released from the pituitary gland stimulated by pulses, at intervals of one to four hours, of gonadotrophin releasing hormone (GnRH) from the hypothalamus. How suckling inhibits GnRH release is not known, but there may be direct neural communication to the hypothalamus.
Breastfeeding offers a woman 98% protection from pregnancy while she is:
Unfortunately, few women suckle to six months; but there is only a 2.9% risk of pregnancy in relying on lactational amenorrhoea alone for up to six months.
In western societies, wealthy people have fewer children than poor people. This seems anomalous to an evolutionist. In other societies the position is reversed. Two kinds of evolutionary explanation are available:
It is certainly possible to devise a model in which there is a trade-off between the number of one’s children and their wealth so that, in the long run, the parent whose children are many loses out in competition with parents whose children are fewer but wealthier. In a very simple model in which
an optimal strategy might well be one in which, within certain bounds, wealthier parents have fewer children. However, once these unrealistic constraints are relaxed, it is impossible to find circumstances in which the graph of wealth against optimal number of children has a peak. It is therefore unlikely that the observed negative correlation between fertility and wealth is an evolutionary equilibrium.
Dr Voland described the demography of an agricultural community bordering the North Sea during the 19th century. The population could be divided into farmers, smallholders and the landless. The striking feature was the difference between the farmers and the landless in family size and the marital and reproductive success of children, which seemed to be largely determined by the pattern of wealth inheritance. Particular features of note were:
Many modern populations, particularly in the industrialised nations, exhibit fertility far below levels seen in historical populations or in modern populations in less developed countries. This observation is difficult to accommodate in traditional evolutionary theory which assumes that natural selection will favour those reproductive strategies that maximise long term fertility. The optimum number of offspring should trade off quality and quantity in such a manner that, over several generations, fertility is maximised. The only way of explaining a strategy that does not maximise offspring numbers is to postulate that children in large families are less fertile so that the strategy maximises the production of grandchildren. The authors therefore postulated the following model:
They set about testing this by interviewing 7170 men in Albuquerque, New Mexico. The most popular family size was two for Anglo-Saxon interviewees and three for Hispanics. If the model is correct, the optimum number of children ought to be two or three. However, when a polynomic regression of number of children against number of grandchildren is produced it is pretty close to being a straight line rather than the hump at two or three that the model would predict. Thus the optimum strategy ought to be to have as many children as possible - there must be something wrong with the model. A tentative explanation is that, long term fertility being hard to estimate, human reproductive strategy has been to maximise the competitive capacity of children in the labour market. In the long history during which this psychology evolved, this would have been a good strategy for maximising long term fertility but in modern conditions, where there is scarcely any limit to the amount that can be invested in children’s upbringing to improve their employment potential, it favours suboptimally small families.
The second half of this report will appear in the next issue of the Newsletter.