of Bali. Regarding Nāga worship, I may remark in passing, that the story of Behulā commemorates how the new-comers in the lands of the Angas and the Vangas had to accept and venerate the religious cult of the original inhabitants. We can see from the account we now obtain of Bali, that the name of the common ancestor of the tribes under review was not the creation of a fancy of the Aryans. It has been stated in the previous section, that those who proceeded to Southern India from Bengal, and its neighbourhood, had Nāga for their totem, and we have now seen that Bali is still worshipped in the Southern Presidency. The cumulative effect of the whole evidence is in favour of this supposition, that the original inhabitants of Bengal were by race and habits allied to those who are now designated as Dravidians.
The Vangas who were always connected with the Puṇḍras and the Suhmas, must have occupied the tract of country which lay to the east of our modern Burdwan Division. The fact that the Pāṇḍavas conquered Vanga after subverting the Puṇḍras, and led their victorious soldiers to Suhma after devastating Vanga, supports this position fully (M. Bh., Sabhā, XXX, 23-25). We find also in the Raghuvamśa, that Raghu conquered the Vangas after finishing his task with the Suhmas, and planted his victorious banner in the midstream of the Ganges. The popular notion that Vanga, as described by Kālidāsa, should be identified with the modern Eastern Bengal, is erroneous. To clear up the point, I have first to note that in all old records we get the Vangas in close proximity to the Puṇḍras and the Suhmas; we may then refer to the historical fact, that when Suhma lost its old name and became a sub-province with the name Daṇḍabhukti, it became a Bhukti or sub-province of Banga. The Tirumalai inscriptions decide this point clearly and