72   Life and Letters of Francis Galton

which he has changed the real names. It is of interest as showing a case in which inoculation was followed by smallpox and death, and a second case in which one person had smallpox three times, both being phenomena of which the possibility raised heated controversy in the 18th century.

Now if we remember that we can hardly form less than 15 principal disease groups nor fewer than 10 age groups, and that we have two parents, it will be seen that we require to divide our material to start with into 300 categories. It would be of little service, if we are to reach really definite conclusions, unless we had 50 to 100 parents in each of these groups, that is to say, records of 15,000 to 30,000 of the first generation, and it might be hoped five times as many in the second generation, a total say of 100,000 to 200,000 recorded deaths, and we must assume these cases to be in a fairly uniform social class and with a fairly like environment. Probably it would be best to work with one social class, and weed out cases having very differentiated environments. Galton had]. 60 usable family records, with an average of 75 individuals, so that he might hope to reach 12,000 records of disease and perhaps 6000 of deaths. Actually he had only about 2000 causes of death recorded, which might correspond to some 300 groups of two parents and five children. On the average this would give for each special age group and each special disease group about two parents of the same sex, mustering ten children of two sexes and all ages, from which to determine whether and to what extent a parent dying at a given age of a specific disease influences the offspring dying of that or other specific diseases at given ages*. The problem is one of probabilities and we shall not have data enough to answer it. Suppose a man to die of cancer between 65 and 75; then we may further be supposed to know the chance that a man of 35 to 45 will die of cancer, but how are we to determine, supposing the latter man is son of the former, whether the relationship in disease is one of chance or heredity? We can only do it, if we have enough pairs of fathers and sons like the above to calculate from the observed frequency the probability of sons dying round the given age of cancer, and to determine if it differs significantly from the probability of deaths in general from cancer of men between the ages of 35 to 45, whether they have or have not a cancerous parent. I have enlarged on these points, because the measurement of heredity in disease is a fundamental problem of eugenics, but its complexity in the general form is rarely recognised ; the labour and great cost of such investigations are in most cases prohibitive. Galton spent £500 in getting his Family Records, but although inheritance of disease was to be an essential part of it, he obtained practically nothing of value. Thus he writes in this connection

'° I had hoped even to the last moment, that my collection of Family Records would have contributed in some small degree towards answering this question, but after many attempts I find them too fragmentary for the purpose. It was a necessary condition of success to have the complete

life-histories of many Fraternities who were born some seventy or moreyears ago, that is during the earlier part of this century, as well as those of their parents and all their uncles and aunts. My

* It is most important to bear this age factor in mind, as the relative proportions of the diseases of which an individual may die vary in life from age to age.