tribes, the low castes of Southern India who had migrated thither from the submerged continent and the Tamil speaking Aryan Brahmans, but only the high class Tamils—the Vellalas and the Chetti castes _who were more or less brown complexioned, fairly civilized, of good physique and of martial habits like the Semitic or Iranian tribes of North-Western Asia. These people, we presume, are now represented by the Todas of the Nilgiris, though there had been on the plains a complete fusion with the aboriginal races and the later Aryan immigrants, as the proverb says,
கள்ளன் மறவன் கனத்ததோ ரகம்படியன் மெள்ள மெள்ள வெள்ளாளன்.
(A Kallan became a Maravan, the Maravan became an Agambadiyan, and the Agambadiyan became a Vellalan.)
Further, the mental and physical characteristics of the Brahuis as described by Mr.D. Bray agree so well with those found in the literature of the early Dravidian Tamils, that one will be justified in regarding both as ethnically related to each other. Thus, we see that this theory is supported by philological as well as ethnological evidences, and we cannot observe any contradiction between them. The Brahuis must, therefore, be regarded as the rear guard in the Dravidian migration and the Todas its van-guard. We may say that the connection between Brahui and Tamil is so great that no other inference than that of the ethnic relationship between the two peoples seems possible, in spite of Dr. Grierson's assertion that the Brahuis
