Statistical Investigations 337
Galton confines his attention to the data for age, height and weight, and remarks
"It seems to me better not to speak at present of the attractive and numerous problems that might be solved by a wider range of inquiry; because if we confine the attention of those we 'ask to few and simple questions, we are far more likely to have them well and thoroughly
answered, than if we had issued a more ambitious programme." (p. 310.)
Anthropometric measurements were soon taken at a number of schools and in some schools anthropometric laboratories established. From the schools they spread to the Universities (Cambridge, 1884; Oxford, 1908; London, Galton Laboratory, 1920). But on the whole there has been a tendency to take in routine fashion a few superficial measurements, and not use the anthropometric laboratory as a means of solving definite problems, physical or mental. They might still be of value if a little inspiration were thrown into their work and psychic or dynamic qualities measured rather than superficial static characteristics. One result of the proposal was those returns from the public schools, upon which Galton based his paper on the weight and height of boys in town and country schools discussed on our p. 125.
Another somewhat slender paper of this period is entitled: "Excess of Females in the West Indian Islands from documents communicated to the Anthropological Institute by the Colonial Office'." This paper gives statistics showing the excess of females in most of the West Indian colonies, although there is an excess of male births. The anomaly is partly due to mortality following dissipation in the young of the male sex, but more extensively to adult male emigration. The whole topic might now be rediscussed with fifty years additional statistics, and would not be without interest. As Galton remarked in 1874 each of the West Indian Islands is an individual social experiment, and each therefore deserves the pains of a separate and thorough statistical investigation.
The collection of statistical data was, however, not the only point that Galton had in view ; he sought to make statistical theory simple and of easy application, and he risked the possibility that loss of refinement might -involve decreased accuracy and a drawing of over hasty conclusions. His "Proposed Statistical Scale" was first given at a Royal Institution Friday evening discourse on February 27, 1874. He followed the lecture up by a letter to Nature on March 5, 1874'. Hiss communication embraced the idea of "ranks," and the whole theory of ranks has been developed from this origin. It is easy to recognise that it is often less difficult to place two persons in order as to the intensities they possess of any physical or psychical character than actually to measure those intensities. A trained schoolmaster can "rank" his class for intelligence with very considerable accuracy. If a number of individuals be placed in order of their intensity for any-character, they are arranged according to Galton on a "statistical scale" (S.S.). The grade of any individual is then determined by the percentage of the whole population who stand above that individual on the statistical scale. The middle man-or the man who would stand half-way between the two middle men if there were two-was later said
'Jour. Anthrop. Inst. Vol. iv, pp. 136-7, 1874. 2 Vol, ix, p. 342 (abstract of lecture, p. 344).