of several distinct rivers without any attempt at arrangement of any kind.
If then my inference be correct, the name Hiranyavaha and its Greek rendering Erranoboas belong to the Gandak; Condochates would be the name of the Chhota Gandak river.
The physical characteristic of size of the great Gandak agrees with the Greek accounts, which make it the third river in India, inferior only to the Indus and the Ganges. The Gandak is indeed a great river, and, unlike its southern rival, it is not a river that shrivels up in the hot weather. The Son in the dry months is a very small river, or rather discharges a small volume of water, and only becomes mighty for a short time annually; it is not navigable, except in floods; in short, its essential characteristics are those of a mountain torrent, and as such it cannot bear any sort of comparison with the perennially great rivers, the Ganges and the Indus; but the Gandak, independent of its draining a larger basin than the Son, is fed by the eternal snows of the Himalayas, and never even in the driest mouths dwindles down to insignificance. It is always navigable in the driest seasons as far up as Baggah, or almost to the foot of the Siwâlik hills (see Rennell’ s memorandum and map of inland navigation), and may justly bear comparison with the Ganges and the Indus.
It thus appears that physically the Son cannot be held to represent the Greek Erranoboas, and whatever weight may attach to my arguments regarding the right of the great Gandak to the name Hiranyavaha, the physical inability of the Son can in no way be bettered by a decision for or against it; so that the Son must be abandoned, whatever other river may be adopted instead, and there is no other river that can fulfil the requirements but the Gandak. To sum up, then, I infer that the Sona of the Greeks is the modern Son; that the Condochates of the Greeks is the modern Chhota Gandak or Gandaki, joining the Ganges a little above Hajipur; and that the Erranoboas is the Hiranyavaha or the great Gandaki river, the modern Gandak.
Hitherto all proofs of the identification of Pâṭaliputra with Patna have been based on historical grounds. Nothing, however, in or about Patna has been discovered which could with certainty be pointed out as a relic of Pâṭaliputra. This last link in the chain of evidence I have been enabled to supply.
In one of his letters, General Cunningham informs me that
