| Obs.: | 5·5 | 4·8 | 0·64 | 0·59 | 0·50 | 0·74 | ||||||
| Calc.: | 6·0 | 4·2 | 0·67 | 0·51 | 0·48 | 0·68 |
The correspondence between the more mediocre cases is much closer than these, and very much closer than between the extreme cases given in the table, namely, the values that 5 per cent fall short of, and 95 exceed. These are of course less regular, the observed instances being very few; but even here the observations are found to agree respectably well with the proportions given by calculation, which is necessarily based upon the supposition of an infinite number of cases having been included in the series.
As the want of agreement between calculation and observation must be caused in part by the paucity of observations, it is worth while to make a larger group, by throwing the six series together, as in Table XXXIII., making a grand total of 965 observations. Their value is not so great as if they were observations taken from that number of different persons, still they are equivalent to a large increase of those already discussed. The six series of observed values were made comparable on equal terms by first reducing them to a uniform PE and then by assigning to M, the point of departure, the value of 0. The results are given in the last column but one, where the orderly run of the observed data is much more conspicuous than it was before. Though there is an obvious want of exact symmetry in the observed values, their general accord with those of the calculated values is very fair. It is quite close enough to establish the general proposition, that we are justified in the conception of a typical form of loop, different for the two thumbs; the departure from the typical form being usually small, sometimes rather greater, and rarely greater still.
I do not see my way to discuss the variations of the arches, because they possess no distinct points of reference. But their general appearance does not give the impression of clustering around a typical centre. They suggest the idea of a fountain-head, whose stream begins to broaden out from the first.
As regards other patterns, I have made many measurements altogether, but the specimens of each sort were comparatively few, except in whorled patterns. In all cases where I was able to form a well-founded opinion, the existence of a typical centre was indicated.
It would be tedious to enumerate the many different trials made for my own satisfaction, to gain assurance that the variability of the several patterns is really of the quasi-normal kind just described. In the first trial I measured in various ways the dimensions of about 500 enlarged photographs of loops, and about as many of other patterns, and found that the measurements in each and every case formed a quasi-normal series. I do not care to submit these results, because they necessitate more explanation and analysis than the interest of the corrected results would perhaps justify, to eliminate from them the effect of variety of size of thumb, and some other uncertainties. Those measurements referred to some children, a few women, many youths, and a fair number of adults; and allowance has to be made for variability in stature in each of these classes.
The proportions of a typical loop on the thumb are easily ascertained if we may assume that the most frequent values of its variable elements, taken separately, are the same as those that enter into the most frequent combination of the elements taken collectively. This would necessarily be true if the variability of each element separately, and that of the sum of them in combination, were all strictly normal, but as they are only quasi-normal, the assumption must be tested. I have done so by making the comparisons (A) and (B) shown in Table XXXIV., which come out correctly to within the first decimal place.
Table XXXIV.
| Right Thumb. | Left Thumb. | ||
| (a) | Median of all the values of KL | 12·5 | 10·1 |
| (b) | Median of all the values of NB | 10·1 | 8·9 |
| (A) | Value of a/b | 1·24 | 1·11 |
| ( Median of all the fractions KL/NB | 1·15 | 1·10 | |
| (c) | Median of all the values of AN | 4·6 | 4·6 |
| (d) | Median of all the values of AH | 4·4 | 3·3 |
| (B) | Value of c/d | 1·05 | 1·40 |
| Median of all the fractions AN/AH | 1·08 | 1·36 | |
It has been shown that the patterns are hereditary, and we have seen that they are uncorrelated with race or temperament or any other noticeable peculiarity, inasmuch as groups of very different classes are alike in their finger marks. They cannot exercise the slightest influence on marriage selection, the very existence both of the ridges and of the patterns having been almost overlooked; they are too small to attract attention, or to be thought worthy of notice. We therefore possess a perfect instance of promiscuity in marriage, or, as it is now called, panmixia, in respect to these patterns. We might consequently have expected them to be hybridised. But that is not the case; they refuse to blend. Their classes are as clearly separated as those of any of the genera of plants and animals. They keep pure and distinct, as if they had severally descended from a thorough-bred ancestry, each in respect to its own peculiar character.