244 
A First Study of the Burmese Skull 
narrower foramen magnum both in its absolute measurement (fmb) and in relation 
to its length (fmbjfml). 
Burmans and Dravidians (Maravars). The number of characters which are 
available for comparison are somewhat fewer in number in the case of the Maravar, 
and, as already remarked, we have only data for male skulls at our disposal. 
As in the case of the Hindus, we are confronted with considerable divergencies 
from our Burman constants, and frequently in the same direction. 
Again we find a smaller capacity (C) and height (H) and a much smaller maxi- 
mum breadth (B), while the maximum length (L) and minimum forehead breadth 
{B') are approximately the same; again a reduction in all the cephalic indices 
(BjL, HjL, BjH); also in the horizontal circumference (U), zygomatic breadth {J) 
and height of nose (NH), though to a greater extent than in the Hindu; the width 
of the nose (NB) is also reduced, but not quite so much as in the Hindu, so that 
the nasal index of the small nose of the Maravar does not differ significantly from 
that of the Burman's large one. The orbit {O/, 0^ too is small in comparison with 
the Burman's and flatter; still flatter than the Hindu orbit. Unhke the Hindu, 
the Maravar has a foramen magnum which is shorter {fnil) than the Burman's as 
well as being narrower (fmb), so that its proportions do not differ materially from 
those of the latter. 
In examining columns of means such as are supplied by the comparative tables 
here discussed, it is possible to gain some idea of the extent to which a cranial 
character of one race is modified in another race. But is it possible to answer from 
them the much larger and more important question of how nearly the various races 
characterised by the multiplicity of racial constants are related together? One 
may certainly gain some vague appreciation of relative affinities between the 
various races, as here one gathers that the Burman seems more nearly related to 
the Malayan and Chinaman, than either to the Hindu or the Maravar. 
But to which of the first is he more closely akin? to which of the last? 
Would it be possible to get a measure of the relationship by taking a large 
number of characters in combination instead of considering each individually? 
And if so, could we by this means test whether the larger group and the two 
smaller groups of our Burmese series belong indeed to one and the same popula- 
tion? or whether we were justified in considering them distinct? 
These are such problems as must often have troubled other craniologists, and 
I have to thank Professor Pearson for providing a provisional solution*, in the 
form of a coefficient which shall express numerically to a first approximation the 
degree of relationship existing between any two racial groups. 
Before passing, however, to the consideration of this coefficient, there is one 
feature in our comparative tables of which I have reserved mention until now; 
* Provisional in the sense that a more complete definition will onlj' be possible when wider study 
of the type constants of races provides material for the determination of inter-racial variabilities and 
correlations. 
