I think this system is the one that most commends itself from its clearness, but there are hardly two writers on ethnology who keep to the same classification.
Agassiz classifies by realms, and has eight divisions.
The Indian races with which we have now to deal are distributed, generally speaking, as follows:—
Caucasian.—(Homo japeticus, Bory and Fischer). Northerly, westerly, and in the Valley of the Ganges in particular, but otherwise generally distributed over the most cultivated parts of the Peninsula, comprising the Afghans (Pathans), Sikhs, Brahmins, Rajputs or Kshatryas of the north-west, the Arabs, Parsees, and Mahrattas of the west coast, the Singhalese of the extreme south, the Tamils of the east, and the Bengalis of the north-east.
Mongolians (H. Scythicus), inhabiting the chain of mountains to the north, from Little Thibet on the west to Bhotan on the east, and then sweeping downwards southerly to where Tenasserim joins the Malay Peninsula. They comprise the Hill Tribes of the N. Himalayas, the Goorkhas of Nepal, and the Hill Tribes of the north-eastern frontier, viz. Khamtis, Singphos, Mishmis, Abors, Nagas, Jynteas, Khasyas, and Garos. Those of the northern borders: Bhotias, Lepchas, Limbus, Murmis and Haioos; of the Assam Valley Kachari, Mech and Koch.
The Malays (H. Neptunianus) Tipperah and Chittagong tribes, the Burmese and Siamese.
Now comes the most difficult group to classify—the aborigines of the interior, and of the hill ranges of Central India, the Kols, Gonds, Bhils, and others which have certain characteristics of the Mongolian, but with skins almost as dark as the Negro, and the full eye of the Caucasian. The main body of these tribes, which I should feel inclined to classify under Fischer's H. Polynesius, have been divided by Indian ethnologists into two large groups—the Kolarians and Dravidians. The former comprise the Juangs, Kharrias, Mundas, Bhumij, Ho or Larka Kols, Santals, Birhors, Korwas, Kurs, Kurkus or Muasis, Bhils, Minas, Kulis. The latter contains the Oraons, Malers, Paharis of Rajamahal, Gonds and Kands.
The Cheroos and Kharwars, Parheyas, Kisans, Bhuikers, Boyars, Nagbansis, Kaurs, Mars, Bhunyiars, Bendkars form another great group apart from the Kolarians and Dravidians, and approximating more to the Indian variety of the Japetic class.
Then there are the extremely low types which one has no hesitation in assigning to the lowest form of the Polynesian group, such as the Andamanese, the jungle tree-men of Chittagong, Tipperah, and the vast forests stretching towards Sambhulpur.
On these I would now more particularly dwell as points of comparison with the rest of the animal kingdom. I have taken but a superficial view of the varieties of the higher types of the human race in India, for the subject, if thoroughly entered into, would require a volume of no ordinary dimensions; and those who wish to pursue the study further should read an able paper by Sir George Campbell in the 'Journal of the Asiatic Society' for June 1866 (vol. xxxv. Part II.), Colonel Dalton's 'Ethnology of Bengal,' the Rev. S. Hislop's 'Memoranda,' and the 'Report of the Central Provinces Ethnological Committee.' There is as yet, however, very little reliable information regarding the wilder forms of humanity inhabiting dense forests, where, enjoying apparently complete immunity from the deadly malaria that proves fatal to all others, they live a life but a few degrees removed from the Quadrumana.