An engaging biography of Francis Galton, heavy with the many amusing Galton anecdotes we all know (a sober analysis of the inefficacy of prayer which drew furious attack; recording people fidgeting during lectures or average attractiveness of women on the street; constructing devices to keep himself awake). Gillham devotes much space to Galton's youthful travels and African expedition and to his fingerprinting work, less to the weather mapping, but that's reasonable inasmuch as those are the most exciting to read about and anyone can understand & appreciate that, even if I have to say that in the long run, Galton's work on the source of the Nile, as ancient a mystery as it may be, was infinitely less important than his other work like twin studies.

What is much more interesting to me is the almost as lengthy discussion of Galton and other biologists' attempts to come up with a mechanistic model of how evolution & heredity could work which explained both simple Mendelian traits but also more complex breeding phenomenon like continuous traits, regression to the mean, and occasional throwbacks. This account of the dispute between the 'Mendelians' and 'biometricians' probably strikes most readers as deeply tedious and perplexing, but I found it interesting and enlightening as most histories of statistics tend to discuss briefly Galton's inventions of correlation & regression and then skip forwards 10-20 years to when Karl Pearson has made many contributions and the stage has been set for R.A. Fisher, ignoring the interregnum, so I didn't really understand what went between. Gillham helps in that respect, although in general his statistical explanations are poor enough and confused enough that I wondered if he understood the issues at all. (I assumed he was a historian, but looking up his biography, he apparently is even a geneticist, so he really ought to be able to do better. One is probably better off looking to Stigler for accounts of things like the Quincunx.)

Aside from being obscure, he often leaves out critical details; for example, two or three times in the account of the debate, he quotes someone coming close to the insight that would resolve it, but Gillham doesn't explain what that insight was or how R.A. Fisher would push the insight through, so I suppose you simply have to already know that Fisher's insight was that the Mendelian view was correct but that with a large number of Mendelian genes, the Central Limit Theorem shows that they will manifest as a continuous phenotype, and the Mendelian traits were simply the extreme where there are only a handful or one relevant gene. This omission is unfortunate because it's a huge flaw in the Mendelian-affiliated eugenicists as it meant that their pedigrees of things like 'feeble-mindedness' were effectively useless since they were discretizing badly a continuous trait† they were often unable to measure accurately in the first place (no accurate IQ tests yet). Another example would be mentioning that Wissler's analysis ended Cattell's mental testing program without mentioning Wissler prompted Spearman to find the general factor (and indeed, some of the sensory testing like reaction time have shown a correlation with intelligence). Some of the criticisms that Gillham quotes approvingly are either ignorant or stupid - for example, that Shakespeare's parents were undistinguished and thus evidence against heritability, which ignores that his father was a wealthy trader & smuggler who had been elected mayor (even if one discounts the Shakespeare arms as due to the son) and his mother descended from the notable Arden family, and would be a poor counterargument even if it were true since base rates alone imply that a large fraction of great men will be of humble origins simply because there are so many humble people that it overcomes their far lower per capita chance of success (as implied by the precis of Hereditary Genius that Gillham gives). In addition to occasionally repeating ridiculous arguments, it's unfortunate Gillham doesn't survey any of the later Fisher & Wright development of behavioral genetics which bore out so many of Galton's inferences. Still, I think I have to give Gillham credit for being as fair as he was in 2001, and it overall is an excellent biography.

† Yes, I know that many cases of severe mental retardation are due to single mutations and so might be Mendelian, but they would be irrelevant from an eugenic perspective since they tend to not reproduce in the first place, while the eugenicists were concerned about the poor in general.