There is virtually no natural selection for longevity in humans because reproduction usually ceases at 45 for women and is much reduced for men after 60. A significant improvement in longevity has occurred in the last 10 to 15 years because of advances in medicine and surgery, which allow heart and cancer patients to survive longer in developed countries. However, a recent trend in the same countries, is for people to become obese at an early age through availability of cheap food, and lack of exercise. This will counteract the present trend for increased longevity when the current generation of young people reaches old age. These trends are less evident in the poorer countries of Africa and South America because of a higher incidence of Aids and other diseases, malnutrition and lack of medical care.

Returning to countries like Great Britain, there is a popular misconception that, although the average age of death has increased gradually since about 1800, the age when most people died from the middle of the 19th century was around that average age, which was about 50. Averages, particularly the arithmetic mean, are often misleading. If the range (i.e. the spread of ages when people die) is large, with more than one peak, then the mean disguises the fact that most people die either in infancy (before 12) or in old age (after 65). This was certainly true of the time when my great grandfather's children were born (1846 - 1868). He had 14 children, 4 died before the age of 13, and 2 in their early twenties, but the rest survived into old age. A typical example of this effect can be seen in the children of Charles Darwin (1809 - 1882). His 10 children died at the following ages: 3 weeks, 18 months, 10 years and 65, 75, 77, 77, 79, 84 and 93 years. Thus, although the average age of death was 56.2 years, no one died between the ages of 10 and 65. The incidence of child mortality had started to decline towards the end of the 19th century. In my own family, my grandfather's 12 children were born between 1878 and 1899. None of them died in childhood, and apart from two sons (my uncles), who were killed in the First World War (at 21 and 38), the average age of death of the remaining 10 was 73.0 years for the males and 79.1 years for the females. These figures are similar to the 1994 statistics for life expectancy in the U.K. at 73.9 for males and 79.7 years for females. (World Book Encyclopaedia, 1999 edition.) Prior to 1800, and in depressed areas after the industrial revolution, it was certainly true that only a minority of people survived to 70 and to live beyond 50 was considered fortunate. This is also true in many African countries today.

The ability to have children at an advanced age in the past has been confined to men; but recent advances in gynoecology have enabled several women to have children in their 50's and 60's. Advancing reproductive age has the effect of increasing the generation interval. An extreme example of long generation intervals was quoted in the Guinness Book of Records (1991 edition). Baroness Elliot of Harwood was an active member of the House of Lords until her death in 1992 at the age of 89. Her grandfather was born in 1786, her father in 1823 and she was born in 1903. Thus, the generation intervals were 37 and 80 years respectively, with a mean of 58.5 years, and the interval between her grandfather's birth and her death was 206 years. It is quite likely that this record has been exceeded in the past when both a father and a son, with younger wives, had children in old age.

From the point of view of genealogical records, another interesting statistic that could be calculated for each family, in addition to the generation interval, is the average age of death. An analysis of peaks of mortality within families in the past would be enlightening, both from the genetic and historical points of view. The influenza epidemic in 1919, when a lot of young people died, would be picked up in some families by this kind of analysis.

One of the disadvantages of having children in old age is that the children lose their parents when they, the children, are relatively young. Another, more serious problem, is that having children later in life allows more time for harmful mutations to accumulate in the sex cells of the parents, particularly the fathers, where active cell division during spermatogenesis continues into old age. This in turn increases the risk of harmful genetic conditions appearing in their children and descendants. Jones (1996) suggested that this ageing effect could have been responsible for haemophilia being transmitted through Queen Victoria to the royal houses of Europe. Since haemophilia does not appear to have occurred in previous royal generations, it is possible that a mutation occurred in the sex cells of Victoria's father, Prince Edward, who was 51 when Victoria was conceived.