Abd er Rahman had been minutely examining the whole desert towards the south-west from the top of the hill. Suddenly he touched my arm and drew my attention to two ’alems (landmarks) lying in the distance.

I looked through my glass in the direction that he indicated, but could see no ’alem at all. However, as he persisted they were there, we went down to the bottom of the hill to look for them.

For a long way from the foot of the hill, the whole surface of the desert was covered with loose slabs of sandstone rock. Abd er Rahman led us across this up to a little pile of three stones about a foot high that, with the keen sightedness of the bedawin, he had spotted from a distance of some two hundred yards, although it lay on the ground so covered with loose slabs of stone that I had not been able to see it myself, even when pointed out by Abd er Rahman, until I got within a few yards of it.

These little heaps of stone, sometimes only a few inches high, are placed at intervals along the desert roads to act as landmarks to those who use them. Occasionally, instead of being placed on the road itself, they are erected on a hill, or rising ground, close by it. The bedawin, even if unacquainted with the district, will often travel great distances, relying for their guidance on the ’alems erected along the roads by previous travellers.

Some hundred yards farther on we found the second ’alem that Abd er Rahman had seen. It consisted of a similar pile of stones. I took a bearing along the line of the two and then we proceeded to march along it. But the road proved to be very bad going, the camels slipped and tripped over the loose stone slabs, till once or twice I thought one would be down. But after a time we got on to easier ground, and began to make better progress.

In making a compass traverse, it is, I believe, usual to estimate the speed at which the caravan is travelling for each section of the road. I personally found this method so unsatisfactory that, after many attempts, I at length was forced to abandon it and to keep my route book on a method of my own, which I found to give much better results.

I assumed a uniform speed of two and a half miles per hour, which is about the rate of a caravan of loaded camels over normal ground. Then, after having passed over an unusually difficult section of the road, where I knew we had not been marching up to our standard speed, I estimated the amount of time we had lost, and entered it in the route book as a “halt,” to be deducted from the amount of time actually occupied in crossing it. I found that a compass traverse, booked in this way, not only fitted considerably closer to the astronomical positions I found, but the actual plotting of the traverse itself was very much simplified—and the risk of errors in consequence much reduced—by having a uniform speed to work upon.

Shortly before we camped for the night we crossed a very faint old road running almost due east and west. These old roads—of which we found a large number—remain visible for an extraordinarily long time, where they happen to run over certain kinds of desert.

We found the going over the plateau unusually bad. Not only had we to cross large areas of kharafish (sharp, sand-eroded rock), but we repeatedly came across a particularly obnoxious form of it known, I believe, as sofut—a type of erosion consisting of knife-edged blades of sandstone standing up two or three inches above the ground, which proved to be a severe trial to the soft-footed camels, who tripped and staggered along, uttering the most melancholy groans.

Another type of surface we had occasionally to cross was that known as noser. At first sight this appeared to be a perfectly level expanse of hard crusted sand. But appearances were deceptive. The sand was only a few inches thick and overlay a bed of stiff clay which, under the influence of the great summer heat, had cracked into fissures often a foot or more wide and extending for several feet down into the ground. The smooth sandy surfaces showed no trace of these chasms. But if the heavy camels, while walking over the ground, happened to place one of their feet over a fissure, they immediately broke through the weak crusted sand, and stumbled forward into the hole below, on more than one occasion coming right down and throwing their loads. Fortunately we had no worse casualties; but strained sinews, and even broken legs are by no means uncommon from this cause.