346 
A Study of the Hand 
correlation of the same bones in different digits of the same hand ; (iii) the 
correlation of different bones of the same digit ; and (iv) the correlation of the 
same bones of the same digits in right and left hands. To have correlated every 
bone of every digit of both hands would have been an almost insuperable task, 
and further would not have led to any very important results. What we had in 
view was the threefold problem : — to what extent do the parts of the same digit 
fit eacli other, to what extent ilo the digits of the hand fit each other, and to what 
extent do the two hands fit each other. 
The remarkably high degrees of correlation between the parts of the hand and 
of one hand with the other indicated by our first study are amply verified when 
the individual parts of the skeleton are dealt with. The hand is a most highly 
correlated mechanism, and given one long bone of one digit, the range of variation 
occurring in any other long bones of the same or the other hand is wonderfully 
small. It is hard to hold any other view than that this degree of fitting is the 
result of selection for physical use. It is striking to compare the high correlations 
obtained for the parts of both English and German female hands with the 
correlations which we find for bones of the other chief organ of man's supremacy, 
the head. Skill in the use of the head and hand has been man's chief source of 
success, but while the instrument of physical superiority is a highly correlated 
mechanism, the seat of mental superiority, the skull, is probably the least correlated 
portion of the human body ! Of course, if we could in any way subject to measure- 
ment and correlate the soft parts of the head, the organs of sense, and in particular 
the folds and commissures of tlie brain we might find high degrees of relationship. 
We should expect sensory and mental fitness to depend upon high degrees of 
correlation between the parts of the sensory and mental organs. But it still 
remains a noteworthy fact that the bony parts of the skull are on the average not 
correlated with even a third of the correlation of the parts of the hand, and this 
fact alone seems to account for the small apparent relationship between intellectual 
ability and measurements on the head. We should expect to find the parts of 
the organism on which intellectual efficiency depends highly correlated like the 
parts of the hand on which physical efficiency depends ; the absence of high 
correlation in the parts of the skull suggests that it is not chiefly upon its case 
that brain efficiency turns. 
(3) Considering first the absolute size and variability of the parts of the hand 
we shall adopt the following notation: _R = right hand, Z = left hand, = meta- 
carpal bone, = first or proximal phalanx, s = second or middle phalanx, c? = third 
or distal phalanx, the subscripts 1, 2, 3, 4, 5 will refer to the thumb, index, middle, 
ring and little finger respectively. Thus: Rd^, Lp^... would refer to the distal 
phalanx of the thumb of the right hand, and the proximal phalanx of the ring 
finger of the left hand, and so on. 
In Table I. are given the means, standard deviations and coefficients of 
variation of the bones of both hands. In each case the number on which the 
constants are based is stated. 
