furnished by the 'Dictionary' by the help of portraits, more especially those in the National Portrait Gallery. I have information on this point concerning 334 of the eminent persons on our list. In classifying by pigmentation I have relied in the first place on the eye-color, but have allowed hair-color a certain influence in modifying the class in those cases in which there was marked divergence between the two in lightness or darkness. I have sorted the eminent persons into three classes, according as their eyes were unpigmented (blue), highly pigmented (brown), or occupying an intermediate position (combinations of blue with yellow, orange or brown). This intermediate class has necessarily been large, and I have comprised within it three subdivisions: a fair medium, a dark medium, and, between these two, a doubtful medium. Then the question arose as to how the results thus obtained might be conveniently formulated, so as to enable us to compare the different groups of eminent persons. I finally decided to proceed with each of these groups as follows: The doubtful medium persons in each of these classes were divided equally between the fair medium and the dark medium; then two-thirds of the fair-medium persons were added to the fair class, the remaining third to the dark class, and, likewise, two-thirds of the dark medium were added to the dark class, the remaining third to the fair class; the five classes were thus reduced to two, and, on multiplying the fair by 100 and dividing by the dark, we obtain what may be called an index of pigmentation. This method of notation is really simple, and is quite sufficiently accurate for the nature of the data dealt with; it will be seen that by its use an index of 100 means that fair and dark people are equally numerous in a group, while indices over 100 mean an excess of fair persons, and indices under 100 an excess of dark persons. I have been able to obtain the index of pigmentation in the cases of ten groups, the remaining groups being too small to permit of assured results. I present them, with their index of pigmentation, in the order of decreasing fairness: Men of science, 150; Artists, 108; Lawyers, 100; Sailors, 100; Soldiers, 83; Statesmen, 83; Poets, 78; Men of letters, 67; Divines, 43; Actors, 20.
It will be seen that the range is considerable; but I believe we may have considerable confidence in the results, and the more so since they are not altogether unexpected, for (although I do not wish to assume that these phenomena are entirely explicable by race) it is certainly true that men of science and artists tend largely to come from fair districts of the country, divines and actors from dark districts. The fact that allied classes tend to fall together—soldiers and sailors, poets and men of letters—also gives confidence in the reality of the relationship thus brought out. It may be noted, as a fact probably not without significance, that the more active and unemotional classes tend
