In animal breeding theory the generation interval is a crucial statistic.The shorter the generation interval, everything else being equal, the faster the rate of genetic progress per annum resulting from selection.

The generation interval for a family is defined as the average age of the parents when the children were born. This is worked out for each parent separately and the mean is then taken of these two figures. e.g. If the father's age at each of the births was 22.3, 24.4, 26.7 and 28.4 years, and the mother's age was 20.1, 22.2, 24.5 and 26.2 years, then the father's generation interval is 25.45 years, the mother's is 23.25 years and the overall average is 24.35 years. The average generation interval for a whole population at any time is the weighted average of these family figures. Family generation intervals for each family can be calculated if enough information is available. The average generation interval is usually about 30 years.

The actual generation intervals and their mean for a direct line refers only to those nominated individuals in the direct line, not the whole family. The generation interval is the age of the parent when the relevent child is born. This is obtained by subtracting the birth date of the parent from that of the child in each case. To find the average generation interval for a direct line, it is not necessary to know the date of birth of every member of the line, just the dates for the first and last members. For example, in my own family, one of my ancestors, William Butterfield, was born in 1610 and is 10 generations back from me in the direct line. Since I was born in 1928, the time interval between the two birth dates is:- 1928 - 1610 = 318 years. If this is divided by the number of connecting links between us it gives the average generation interval for the line. i.e. 318/10 = 31.8 years.

Human generation intervals can vary widely (theoretically from about 12 to 90). Some of the causes of variation are given below:-

1.      They are often longer in male-male-male line successions than in female-female-female ones because males remain fertile longer than females.

2.      Another reason for variation is differences in parity; i.e. in large families the generation interval within a direct line can be affected by whether the person is the first or last born.

3.      The generation interval of a particular family depends on the age of the parents when the first child is born. At the present time there are a lot of unplanned teenage pregnancies on the one hand, and on the other married couples often wait until their late twenties or thirties before starting a family.

4.      Total family size also affects the family generation interval. The standard family in Victorian times was 12, which meant that the mother was usually in her mid-forties when the last child was born. Modern families rarely exceed 4 children.

5.      The family generation interval is also affected by the lengths of the spaces between each birth and whether or not they are spread evenly over the fertile period.

6.      Different marriage laws and customs in other countries can have a bearing on the generation intervals. e.g. Child marriage and polygamy.

Non-synchronised Generations

When an older person, usually a male, marries a younger one, it can put the generations out of phase in different families. e.g. A man of 40 marrying a woman of 20 would make him a contempory of the woman's parents.

This can also happen when husband and wife, although the same age, are of widely different parities. e.g. My mother was the youngest of 12 children and was born when her father was 51 whereas my father was the eldest of 4 and was born when his mother was only 19. This puts all the earlier generations out of kilter; so that on the two sides of the family, ancestors of the same generation (relative to the nominated person) are not contemporary and conversely, contemporary relatives are from different generations. In the above example, my maternal grandmother (1857-1938) was a friend and contemporary of one of my paternal great-grandfathers (1856-1941).

When related individuals of different generations marry and have children, this also throws the generations 'out of step'. In this case it may be difficult to decide in which generation some relatives belong. The following is a method sometimes used to number and identify the generations:

The starting point is the central or nominated person in the research, who is given the number 0. Thus, all ancestral generations will have negative numbers and descendant generations positive ones. After determining the generation numbers of all ancestors and descendants (and their siblings) it will be possible to calculate forward to more distant collateral relatives from appropriate common ancestors. The diagram below illustrates this method:-

Figure 62 Assignment of Generation Numbers

The nominated person is L (generation 0).

Therefore, if there has been any inbreeding, and the parents on the two sides of the family are from different generations, several individuals in the family tree may have two generation numbers depending on which direct line is followed back to the common ancestor. e.g.

Figure 63 Indeterminate Generation Numbers

Generation numbers are in red. The above is what happens when a person marries his/her first cousin once removed.

In the extreme example of illegal incest between closely related persons of different generations, generation numbers would cease to have any meaning.