72 
On the hihcrliance of the Finger-Print 
(7) Assortative Mating. 
In this connectiun it would be as well to point out that in the data we have 
there is no tendency for like to marry like and that the assortative mating 
coefficient for type of finger-print is nil. There are only 177 cases where the 
finger-prints of husband and wife are given, and the 0- becomes "0615 and "0536 for 
left and right first finger respectively and the correction for the number of cells is 
■0904 for both coefficients ; accordingly the 0- is insignificant. 
TABLE IX. 
First Right Fiii<jer 
First Left Finger 
Husband 
Husband 
Wife 
Arch 
S. Loop 
L. Loop 
Comp. 
Whorl 
Totals 
Arch 
S. Loop 
L. Loop 
Conip. 
Whorl 
Totals 
1 
S 
6 
1 
23 
1 
14 
3 
3 
4 
25 
Arch 
1-82 
9-49 
5-07 
1-17 
5-46 
2-82 
10-88 
3-53 
1-84 
5-93 
(; 
Ifi 
3 
17 
7-2 
9 
23 
9 
3 
17 
61 
S. LiKip 
5-69 
29-59 
15 83 
3-66 
17-08 
6-89 
26-54 
8-62 
4-48 
14-47 
4 
1:^ 
,s 
1 
10 
35 
3 
13 
4 
2 
7 
29 
L. L(i.)i. 
2-77 
14-44 
7 71 
1-78 
8-31 
3-28 
12-62 
4 10 
2-13 
6-88 
4 
2 
6 
1 
1 
3 
3 
15 
Couip. 
•47 
2-47 
1-32 
•31 
1-42 
1-69 
6 -'53 
2-12 
1-10 
3-56 
3 
19 
4 
s 
41 
6 
20 
8 
2 
11 
47 
Whorls 
3-24 
16-91 
9-03 
2 08 
9-73 
5-31 
20-45 
6-64 
3-45 
11 15 
1 Totals 
1 
14 
73 
39 
9 
- 42 
177 
20 
77 
25 
13 
42 
177 
The black figures give the independent probability and it can be seen quite 
easily that the differences between the independent probability and the actual 
figures are small and the signs indiscriminate ; there are no well marked positive 
differences along the diagonal and it is fairly obvious that there is no evidence of 
assortative mating in finger-prints. It will be clear therefore that assortative 
mating had no infiueiice on our results. 
(8) Probable E.rplanation of tJie low values of the coefficients of inheritance in 
the case of Finr/ei'-Prints. 
Galton's material for parental heredity limits us to the consideration of what 
happens in the case of the forefingers of the two hands. But this limitation com- 
pels us to discuss only the case of inheritance from like finger to like finger. Now 
let us suppose that the finger-prints of both parents were entirely whorls, but that 
the offspring had composites on the two forefingers and the remaining fingers all 
whorls. There would clearly here be a marked inheritance of very anomalous dis- 
tribution, but an investigation which confined itself to the forefingers would fail or 
largely fail to measure its intensity. If we deal solely with like fingers, we overlook 
