Personal Identification and Description   169

to do with the attractiveness or otherwise of any particular pattern. I do not believe that Galton has quite plumbed the possible depths of the action of sexual selection in this matter. Touch is one of the least studied, and therefore the least understood of the sexual factors. The question is not that of sensitivity in the producer of a sensation, but of the feelings excited in the recipient. It is not, perhaps, probable, but it is still possible, considering how large a part touch plays in courtship, that the shades of feeling excited by it may be associated with finger-print pattern. Those who straight-away dismiss any slender possibility in this direction have hardly the true measure of our present scientific ignorance, and probably do not realise how much greater a part touch plays in the sensitory life of the female than of the male.

Galton holds that there must have been complete promiscuity of matings, or as it is now called, panmixia, with regard to these patterns, and that consequently they ought to have hybridised. I cannot see that this argument is any more valid than the argument that iris-colours ought to hybridise. It is true that both iris-colours and finger-prints- do blend under certain rare physiological conditions that we do not yet understand, but I can see no necessity for a universal rule which anticipates that blending must follow hybridisation. The mere fact that the individual can have finger-prints of various patterns suggests hybridisation, and it seems to me that the question of racial differentiation in finger-print frequencies wants renewed investigation, starting very nearly from the point where Galton left it (see our pp. 140,193-4).

We next turn to the question of natural selection and here we read

"As regards the influence of all other kinds of natural selection, we know that they cooperate in keeping races pure by their much more frequent destruction of the individuals who depart more widely from the typical centre. But natural selection is wholly inoperative in respect to individual varieties of patterns and unable to exercise the slightest check upon their vagaries. Yet, for all that, the different classes of patterns are isolated from one another, through the rarity of transitional cases, just as thoroughly, and just in the same way, as are the genera of plants and animals." (p. 22.)

In the words I have italicised Galton seems to me to have departed from his usual cautious restraint in the matter of dogma, and some suspicion may be thrown on his conclusion from his own data. On p. 21 of the memoir are given measurements of the core and the number of ridges in loop finger= prints of the left and right thumbs. From this it appears that the right thumb exceeds the left thumb in these measurements and in the number of ridges. Is this relation reversed in left-handed persons? Nowadays we know that the finger-print types are not scattered at random among the digits, there is association between individual digit and individual type. Can it be that there is any reversal of this association in left-handed persons? We do not know; but if it should prove to be so, the first step would have been taken to show a relation between finger-print pattern and manual efficiency. It is never safe to dismiss all relationship of a character to natural selection because we cannot for the moment see any link between the character and fitness.

P e III   22