22 ALEXANDER'S CAMPAIGN IN THE PANJAB [Chap. I, Sec. ii]
behind it, rising to heights above 1,300 feet and cleft by innumerable ravines, would offer, even after the bridge-head had been taken, we may realize that the road crossing the river at Jhelum, convenient for traffic towards the plain of the Panjab in peace-time, would not be the best line to follow for an invader from the north-west as long as armed resistance were to be met on the left bank.
We are justified in believing that Alexander was in possession of all the local information available through the ruler of Taxila, his ally and the immediate neighbour of Poros. He would thus be in a position to decide upon the choice of a route for his advance which was less likely to favour his opponent in the defence of the river crossing. We shall see that there is such a route leading to the Hydaspes much farther down which was repeatedly used by another great invader of India, and the earliest after Alexander of whose advance to the river we have definite historical knowledge. The choice of a route farther south was all the more advisable for Alexander, as it was bound to draw Poros farther away from the territory of Abisares, the ruler of the outer hills, for whose armed assistance he was hoping.35
Before turning to the ground lower down the river in search of the true scene of Alexander's great exploit, I may add a few words regarding the route leading from Taxila via Rawalpindi to Jhelum. It is the shortest in the direction of such old centres as Sialkat and Lahore, and its use is attested from medieval times onwards. But before the Grand Trunk Road was constructed it was by no means the easiest route. Of this I had occasion to convince myself on a long excursion which I made from Jhelum on November 23rd along the old line of the route. It leads up the narrow winding defile ( Fig. 1) in which the Kahan river has cut its way through the eastern branch of the Salt Range past the great fortress of Rohtas, which the Emperor Sher Shah constructed in A.D. 1542-3 in order to guard this difficult exit into the riverine plain against the warlike Ghakkar tribe holding the north-eastern portion of the Salt Range. Thence the old shâhi road, marked by ruined Mughal sarais and other remains,36 could be traced over much broken ground to the Bakrala pass leading across the western branch of the Salt Range. The gorges on either side of the pass, and even more perhaps the extensive network of deep cut ravines furrowing the plateau to the north, must have offered serious impediments to traffic before the Grand Trunk Road was alined here.
35 Cf. Curtius, Histor. viii. xiv. I.
36 While following the line of this `Imperial Road' northwards, we carne at a distance of about 2 miles from Rohtas upon a conspicuous old mound in the fields of Madina. It measures about 90
yards in diameter and rises to some 12 feet above the field level. Among the plentiful broken pottery covering it pieces with stamped relief decoration of an old type, like Mad. 8 (Pl. I ), were numerous.