Estimating F-statistics: A historical view.

Characterizing the genetic structure of populations is of importance to evolutionary biology, to human disease gene mapping and to forensic science. Sewall Wright introduced a set of "F-statistics" to describe population structure in 1951 and he emphasized that these quantities were ratios of variances. Responding to uncertainty over the best way to estimate F-statistics, Weir and Cockerham published a method-of-moments set of estimators in 1984 (Evolution 38:1358-1370). This paper continues to be widely cited, with over 7,000 citations to date. Some background to the publishing history of the Weir and Ccckerham paper is given here, along with subsequent developments and a discussion of current uses of Wright's F-statistics.