Sec. ii] TO WARAWT AND UP THE GULF COAST 229
I had to decide upon a return to the coast. Aware of the difficulties to be encountered on a return to Galehdar, I insisted on arrangements which would take us across the coastal range by a new and as yet unsurveyed route.
Fortunately Sohrab Khan was able to propose a route which, though very difficult even for donkeys, would allow us to reach the mouth of the Gabandi valley without having to pass either that part of the Warawi tract that had thrown off his authority or to enter the Tarakuma area. But it cost him protracted negotiations with his obstreperous people lasting all through the next day before some arrangement for the transport of our impedimenta could be secured, on terms extortionate in a measure, yet not altogether amounting to blackmail. It was evident that but for Sohrab Khan's persistent exertions `public feeling' at Warawi would have chosen to exploit the opportunity for plunder even in preference to blackmail. The whole day's trying proceedings afforded an interesting experience of local authority exercised under the prevailing tribal conditions.
Next morning fresh trouble and delay occurred before the baggage could be sent off on thirty-five donkeys with their loads reduced to minimum weights. For mounting our whole party two mules of the Khan had to suffice, while twenty of his armed followers were to assure protection ( Fig. 50) . For 6 miles the route led across the wide valley to the hamlet of Hôzi at the foot of the coastal range. Then followed a troublesome ascent in a steadily narrowing gorge where the donkeys, lightly laden as they were, could be taken only with difficulty over slippery rock faces or through the deep pools formed by a small stream at the bottom of the boulder-filled ravine. At a point known as Chah-igul, some 4 miles above Hôzi, where the ravine widened just enough for tents to be pitched, we halted for the night, while pickets guarded the nearest rocky heights.
The ascent in the morning was at first easier, leading for a couple of miles over sharply tilted limestone ridges. Then, descending from their rocky crest by a steep, serpentine track, into a small side valley known as Burm, we came upon unmistakable proof of this route having seen regular traffic in old times, in the shape of a much-decayed cistern and a kiln for burning lime such as had been used for cementing its walls. This evidence was confirmed on the trying climb up the winding gorge by which the pass across the range is gained, as the steep path, where it ascends along narrow rock ledges, had been widened by slabs solidly set in mortar. Much of this old masonry has fallen away. But even before time had affected this piece of early engineering, the route can never have been practicable for any but lightly laden donkeys, mules, or ponies.
After we men on foot had gained the narrow gap in the crest of the limestone