8 The Ethnology of India. 
vaised the position of such tribes. There is, I think, a good deal 
to suggest that during the various invasions of Southern India by a 
succession of Northern ‘ Yavanas,’ small tribes of these latter may have 
taken up their position in difficult parts of the country, and there, 
amalgamating with the aboriginal tribes, have formed hali-breed races 
of much robber-like and semi-military energy. 
Before going farther, I would suggest the following as especially 
deserving the attention of those who are willing to aid in a popular 
way in classing the various tribes and castes in India. 
1. Physical appearance. ‘The three main types, Caucasian, Mon- 
golian, and Negro or Negrito, are well-known. In India we have, in 
the extreme North, the finest and purest Caucasian type, the handsomer 
and more open form of that which we know as the Jewish cast of 
countenance ; fine head and features, high brow and nose, long beard, 
tall, lithe, powerful figure, colour generally light. Throughout India, 
we have this type modified and subdued by every variety of straight and 
snub nose and plebeian features, much as in Hurope, and with a dark 
skin unknown in Europe. Sometimes the skin becomes very black, 
and the lips are thick and protuberant; there may be marked the 
infusion of blood of the Negrito type which probably modifies the 
higher phase of the Caucasian type, even when actual Negrito features 
cannot be traced. 
This Negrito type we find in India not accompanied with the 
muscular form of many Africans, but in a small slight race. The 
principal points to be marked, in addition to colour, are the lips, already 
alluded to, shape of face, nose, and eyes, presence or absence of a 
considerable beard, character of the hair, 
Among the Thibetans and Nipalese we have extreme examples of 
the type which I call generically Mongolian. The eyes particularly 
there is no mistaking. The hair is straight. The colour is yellowish, 
but never dark. 
2. language is liable to disturbances, and has been perhaps too 
much taken as a sure guide, but it is always of great assistance, and 
in 19 cases out of 20 tells a true tale. In practice I think that no 
considerable philological acquirements are necessary to enable an ob- 
server to make most useful observations of a language quite unknown 
to him, if he can only get the rudest interpreter. There are certain 
