________________
MAY, 1929)
ALEXANDER'S CAMPAIGN ON THE N.-W. FRONTIER
17
But a recognition of this fact will not necessarily invalidate the location of Embolima at amb. Arrian's narrative shows that it took Alexander two marches from Embolima to reach the neighbourhood of Aornos. Hence even if the above location is accepted we may still look for Aornos higher up the Indus in that area comprising the tracts of Ghôrband, Chakesar, and Puran, to which the consideration fully set forth above point as the ground most likely to have been sought by the population retreating from Upper Swât. It should, however, be remembered that the identification of Amb with Embolima (Sanskrit Ambulima) rests so far solely on the identity of the modern name with the first syllable of the ancient one, and that the assumed apocope of fully three syllables at the end of the latter is more than can easily be accounted for by the rules governing the phonology of modern IndoÅryan vernaculars. If Embolima were to be looked for farther up the river the position occupied by Kabalgram, a large village at the mouth of the fertile Páran valley and a centre of local trade, might suggest itself on topographical grounds as a likely site.
Ever since my visit to Mahâban in the autumn of 1904 had furnished conclusive evidence against the location of Aornos on that range, I he kept in view " the possibility of our having to look for Aornos higher up the great river."43 But it was only in 1919, after the return from my third Central-Asian expedition and after prolonged labours on the results of the second, that my attention was drawn in a definite fashion to ground where a likely solution of the problem could be hoped for. The right bank of the Indus and all the adjacent territory to the west of it had, indeed, remained as inaccessible as before. But fortunately work on the maps reproducing the surveys carried out during my three Central-Asian expeditions brought me in 1918-19 into close contact with the late Colonel R. A. Wauhope, R.E., at the Trigonometrical Survey Office, Dehra Dun.
The personal knowledge which this highly accomplished officer of the Survey of India had gained of that ground during the survey work conducted by him on the left bank of the Indus during the Black Mountain expeditions of 1888 and 1891-2 furnished me with a very valuable clue. From high survey stations then established on the Black Mountain range, and again during the occupation of the Chagharzai, Nandihâr and Allahi tracts, Colonel Wauhope had ample opportunities for becoming familiar with the general features of the hills on the opposite side of the Indus valley all the way between the Hassanzai country, above Amb, and Chakêsar. Being a sound classical scholar all his life, he was interested in the question of Aornos, and what he had observed at the time had led him to form the belief that a position corresponding to that described by Alexander's historians was more likely to be found on the spurs descending steeply to the Indus opposite Thâkôt in Nandibar than anywhere else. But as an experienced topographer he rightly recognized also that a definite location could be hoped for only by close examination on the spot.
The spurs just referred to are the easternmost finger-like offshoots of the range which trends with a due easterly bearing and a total length of close on 20 miles from the Swat-Indus watershed above Manglawar and Chårbågh to the Indus. On the opposite side the river there passes the mouths of the Nandihår and Allâhî valleys. From the available Survey of India maps, including Sheet No. 43 on the scale of 2 miles to the inch, it was seen that the range may be roughly described as dividing the valleys of Ghôrband and Chakêsar: that its crest rises to triangulated heights between 9,265 feet in the west and 7,011 feet in the cast; and that round its eastern foot the Indus flows in a wide bend. Little else could be made out from the map, based as it necessarily was for this ground on sketches made from a distance, on native route reports and the like.
My first endeavour, made in 1922 after a rapid visit to Agrûr and the Indus banks facing Amb, had been to secure access to the ground just indicated from the tribal territory of Nan. dihår on the opposite side of the river. But by the time I was able to renew the attempt
13 Seo Archaeol. Survey Report, N.W.F.P.' (1905), p. 30.