M. L. TiLDESLEY 
237 
The material from which our female means have had to be obtained is much 
too slender to base any definite conclusions upon, and is only interesting in so far 
as it may confirm the results obtained for males in the comparisons which follow. 
Another weakness which has to be admitted is that this material is probably 
heterogeneous, being drawn from widely separated districts in China. The cata- 
logues frequently omit to state from which part of China the skulls were obtained: 
47 out of our 102 <^ skulls are simply given as Chinese. Of the rest, 25 are from 
S.E. China, and 16 from the West Indies — presumably settlers from the S.E. 
rather than the north. 12 are from Shanghai and Ningpo district, and only two 
are from as far north as Pekin. 
The females are distributed in much the same proportions: 5 from unknown 
districts, 4 from South China, 1 from Batavia, 1 from Shanghai. Thus we can say 
that the large majority of those whose origin is known are from the southern part 
of China. If the unknowns which form less than half our collection are divided in 
somewhat the same proportions between north and south, our means will be ap- 
proximately those of the South China type, which is the one we need for comparison 
with our Burmese. And the standard deviation of the cephalic index BjL for males 
which is 4-52 ± -29 encourages us to believe that the sample is not more hetero- 
geneous than our Malayan sample. 
Hindus. We turn now to find a Caucasian neighbour from the west. Here again 
the difficulty arose that in many cases no details beyond the term Hindu, which, 
as is well known, is not racial but rehgious, and includes an agglomeration of types 
of non-Caucasian origin, as well as highly diversified Caucasian types. The term 
"Hindu" was therefore still less satisfactory than the bare term "Chinese," and 
the material to hand had to be rigorously sorted out. 
The final table has been confined to skulls from the North-Eastern part of 
India, mainly Bengal, but including a few from round Benares, Allahabad being 
our western limit; further, all from within this area which had not belonged to 
persons of the Hindu rehgion were struck out as being more likely to contribute 
heterogeneous elements; I also omitted individuals belonging to such castes as I 
could discover by means of Risley's Tribes and Castes of Bengal to be of Dravidian 
origin or admixture. Again testing the homogeneity of my material by means of 
the standard deviation of the cephalic index, I obtained the value 4-53 ± -38 for 
(J Hindus, so that judged by this single constant the heterogeneity is of the same 
order as that of the Chinese sample.' 
In this Laboratory there is a short series of skulls of Hindus (9 cS and 6 $) 
from Bengal, from the Lower Ganges. These I have measured and pooled with the 
rest. 
The total stands at 69 ^ and 39 ? Hindus and is made up as follows: Males, 
Royal College of Surgeons 36, Barnard Davis' collection 23, Munich 1, Biometric 
Laboratory 9. Feynales, Royal College of Surgeons 8, Barnard Davis 23, Munich 1, 
Gottingen 1, Biometric Laboratory 6. 
