432 
Association of Finger-Prints 
although the differences are small, ranging only up to 1'8 while four-fifths of 
them are less than 1, the consistency of the results suggests a slightly closer 
relationship between a pair of digits on the same hand than between the coi're- 
sponding pair on opposite hands. This view is further supported at a later stage 
of this paper. (See Remark {d) on Tables 14-16, p. 450.) 
One further comparison is of interest, namely, the measure of relationship 
between the various digits on a centesimal scale. It should be noted, however, 
that while Galton's means are based on loops and whorls only, omitting arches 
from his three groups, mine are based on small loops, large loops and whorls, 
omitting arches and composites from my five groups ; also Galton gives no results 
for those combinations which include the little finger. 
TABLE 6e 
Approximate Measures of Relationship between the various Digits, 
on a Centesimal Scale. 
Couplet 
Galton* 
Waite 
Means 
Eight 
Left 
Thumb and fore-finger 
„ middle finger ... 
„ ring finger 
Fore and middle finger 
„ ring finger... 
Middle and ring finger 
24 
27 
39 
60 
23 
52 
18 
21 
21 
40 
33 
44 
27 
23 
24 
46 
33 
48 
Right and left thumbs 
61 
54 
,, fore- fingers ... 
48 
43 
„ middle fingers 
43 
47 
„ ring fingers ... 
65 
64 
For the reasons given above we could hardly expect that these readings would 
be even approximately equal, but for all that, the same general relations are seen 
to hold good in the two sets of results. 
It is convenient at this stage to summarize a few of the most important 
points which have been brought to light in the foregoing pages. These are : 
(a) A greater divergence of types in the right hand than in the left. 
(6) A clustering of the same type in the hands of an individual. 
(c) The uneven distribution of the various types in the different fingers, 
especially the almost entire absence of ulnar loops except in the index. 
* Finger Prints, p. 129, Table VIII. 
