of Hunza. I do not think I ever met a greater ruffian than Ghazan Khan. He not only used every means short of force to obstruct our passage, such as preventing supplies reaching us, but threatened our liberty unless we handed over to him a fort called Chaprot, which of course we had no authority to do, Chaprot being in recognized Kashmir territory.

We were almost the first Europeans to travel right through his country, and when we reached Hunza-his capital and residence-we received a most inhospitable welcome from him. In fact, we were virtually his prisoners. He told us there were only three courses open to us: either to deliver Chaprot, or to go back to Gilgit, or to try to go on without his assistance. We naturally refused these alternatives, and insisted that he had got to help us, or at least to feed us. This annoyed him greatly, and he even considered whether it would not be simpler to kill us and report our death by avalanche. Eventually he let us go on, but gave us little help or supplies. However, by bribing his officials and buying direct from the villagers en route we successfully managed to make our way across the Hindu Kush by the Kilik Pass, which, though a snow pass over 15,000 feet, is not really difficult if crossed before sunrise while the snow is still fairly hard.

On the whole the Hunza route is not one to be recommended to ordinary tourists, but most young and active Englishmen feel an immense attraction to countries or places which have been little visited, and there is no greater satisfaction than that which they experience in the successful accomplishment of such explorations. I at least always look back on my visit to Hunza and Kanjut with the most lively pleasure.

Dr. T. G. LONGSTAFF: It has been a very great pleasure to me to hear Sir Francis Younghusband and Sir Edmund Barrow speaking of Hunza and of the Kanjuti raiders; the result of these first visits and of subsequent contact with other British officers has been to turn those raiders into some of the best frontier militia we have. I think the most interesting of all my visits to the Himalaya was that which fell to my lot, when I was being paid to do what I formerly had to pay to do, while I was serving with those same raiders of Hunza and Nagar in the Gilgit Frontier Militia during 1916 and 1917.

Sir Francis Younghusband has raised a very intriguing problem in connection with the Shaksgam, and there are one or two points that I think should be put on record at such a meeting as this, which is a sort of official résumé of our knowledge of that particular bit of country, and to which I feel that he has not quite done full justice-needless to say, to himself!

Sir Francis is the only European who has been in the upper part of the Shaksgam Valley, but the problem of the Saltoro Pass is by no means a new There is an obscure reference to it in the 'Tarikh-i-Rashidi' of Mirza Muhammad Haidar, written in the sixteenth century, and I think there is also a reference to it by Mir Izzet Ullah. Those references however are not illuminating, and it is to that accomplished and rather neglected traveller, G. T. Vigne, that we owe our first definite information. In 1835 he approached the problem from the south side, while in 1889 Sir Francis attacked it from the north. You will see from the old Survey of India map that before Sir Francis went to that region the country to the north of the Karakoram range was unsurveyed-it is just left as a white blank space; and the main axis of the range is projected southwards in a great angular bastion which projects down to below Vigne's Saltoro Pass at the head of the Bilafond glacier. When Sir Francis' work came to be plotted it was brought down to fill up that great blank wedge beyond the (hypothetical) water-parting. But in 1909, before

I started off to follow Vigne's footsteps, I spent a most delightful week with Sir Francis in the Residency at Srinagar, when he was kind enough to get out his maps and notebooks of 1889 and go through them with me. According to his route-book it was impossible to be certain of the longitudes, for these depended on the correctness of his identification of K, from the north, but as regards his latitudes there was no earthly reason for doubt, and yet on the official map his route had been stretched far down to the south of his observed position to fill up that awkward gap. Now I want to show you a slide: Sir Francis will no doubt recognize his own handwriting. You will see on the sketch the dotted line which represents the outline of the main axis of the Karakoram as laid down on the Indian Atlas sheet. Sir Francis drew this on the day when we had really come to grips with the problem, finally saying in effect, "I was right and it is the map which is wrong; the main axis of the Karakoram is not that shown on the maps, it is here." The man who had been there himself was right: if his original notes and maps had been accepted the mistake that was made of extending the Karakoram axis so far to the south would never have been made, and I should never have had the credit for the little discovery I made in 1909.

Following up Vigne's route, with the late Arthur Neve of Srinagar and with Captain Morris Slingsby of the Frontier Force, who so gallantly laid down his life in 1916, we crossed the pass at the head of the Saltoro Valley, which Vigne had been to the foot of, now called the Bilafond La, and saw the remains of stone shelters used by former native travellers on this route; after another day's march we got out on the north side on to a great glacier which nobody knew anything about. The lower part of that glacier was not unknown, for Colonel Henry Strachey ascended its foot in 1848, but no one imagined that it was 45 miles in length. We followed it up just far enough to see the end of it, and at its head saw a gap in the ridge, which I called "Younghusband's Saddle" on my sketch-map because it corresponded in latitude with his observed position. In Khapalu Rajah Shir Ali Khan had told me that there were two routes through the Saltoro Valley, one to Yarkand and one to Nubra : the same tale was told to Vigne in 1835. The finding by the Workmans in 1912 of an old bivouac farther on beside the main Siachen glacier convinces me that the tradition is correct, and that the main range was formerly crossed just about here. I want you to appreciate the accuracy of Sir Francis' original work and to realize that our chief discovery in 1909 was that the water-parting was just where he said it was! There on his sketch is K2; here is Broad Peak and the Gusherbrums; here is the latitude of the Urdok glacier; then the axis runs east through Teram Kangri and on into the Rimu peaks explored by Sir F. De Filippi in 1914, and thus on nearly due east to the Karakoram Pass, all just as the lecturer predicted. There is one omission: for though the Aghil range, which he alone has crossed, is correctly shown articulating with the Karakoram axis on the north, yet at the actual junction there is an easy pass, which you saw in the last photograph which Sir Francis exhibited. This is the pass discovered by Major Wood* and Mr. Spranger, which you will realize is really beyond and to the north of the main Karakoram range, and which must lead directly to the headwaters of the Shaksgam River, into

May I here record my belief that Major Wood is quite correct ('Survey of India Report on the De Filippi Expedition, 1914' (Dehra Dun, 1922), p. 5) in concluding that the plate opposite p. 640 of my paper in the Geographical Journal for June 1910 is not Teram Kangri but De Filippi's 7391-7388-metre group at the head of the southern branch of the Rimu glacier.-T. G. L.

which the Urdok glacier drains. Incidentally Major Wood found another pass over the tail of the Aghil range a few miles farther to the north-west, with traces of recent travellers. As yet we do not know exactly where these trails crossed the main range: but soon the riddle will be solved, and when it is, then I am willing to bet that the rough sketch which Sir Francis Younghusband drew on a bit of paper in the Residency at Srinagar in 1909 will correspond very well with the reality as revealed by the most modern theodolite in 1926.

The PRESIDENT: The hour is late. I do not know whether Sir Francis would like to say anything in reply to what has been said. You will agree with me that he, Sir Edmund Barrow, and Dr. Longstaff have succeeded in showing that there is still a very considerable deal to do in the Shaksgam country. There lies a problem of extraordinary perplexity and interest which awaits any future explorer fortunate enough to be able to get into that particular part of the Himalaya. You will, I am sure, join with me in giving cordial thanks not only to Sir Francis Younghusband (though chiefly to him) but also to those who have followed him.

NOTE BY THE SECRETARY ON THE POSITION OF THE URDOK GLACIER

Some years ago Sir Francis Younghusband presented to the Society his original field book containing all his astronomical observations made on the journey in 1889 when he crossed the Aghil Pass into the Shaksgam Valley and ascended the Urdok glacier, but was unable to reach the col at its head. His observations include pairs for latitude by north and south stars, altitudes for local time, and an observation for azimuth on a peak which he took to be K2. His orderly kept a compass traverse. All these observations were turned over to the Survey of India at Dehra Dun, and were worked up by a computer in the Survey.

When the Duke of the Abruzzi was preparing his expedition to the Karakorum the Survey of India plotted for him a diagram showing the route of Sir Francis in the Shaksgam as deduced from his observations, placed with reference to the outline copied from the map of Dr. J. Jacot Guillarmod in *Six mois dans l'Himalaya.' The following are the latitudes and heights from the observations of Sir Francis:

The computer at Dehra Dun seems to have concluded that nothing could be made of the observations for azimuth and longitude.

The Duke of the Abruzzi's mapping showed that the Guillarmod map was seriously wrong in the position of Staircase Peak in relation to K2: it was placed too far east: and this affected the plotted position of the ShaksgamUrdok route in longitude. But the latitudes seemed entitled to every confidence: they showed that the col above the head of the Urdok glacier must be close to latitude 35° 40′.

When therefore Mrs. Bullock Workman's surveyor Mr. Peterkin deter

mined the latitude of what she called the Turkestan La as 35° 40', she wrote a passage which has often been taken to mean that she believed the Turkestan La was the head of the Urdok glacier. I was under this impression myself when I remarked to Sir Francis about the time of the Meeting reported above, that it seemed clear from the Workman photographs that the glacier they looked down from the Turkestan La was not the main Urdok but a western affluent.

A recent re-reading of Mrs. Bullock Workman's passage seems to me to show that we had misinterpreted it. The passage is as follows:

"The col we stood on forms a semicircle and ends in the bergschrundfestooned wall visible in the foreground. Directly below the col a sharp drop occurs, say of 2000 feet (it is difficult to estimate height from above a wall). Below, this wall shades off into a short crevassed glacier, which, as an affluent, joins a wide trunk-glacier flowing north-west towards its tongue. We saw well only the upper part rising south-east towards its source behind the group of peaks above mentioned. From the edge of the col the end of the triangular mountain-mass discovered from the Indira col was seen, and the main glacier appeared to take its downward direction along the base of these mountains.

"It is probable that this large glacier flowing north-west joins the Gusherbrum stream seen from the Indira col beyond the triangular mountainrange, or that both end in the same valley near together. The latitude, 35° 41′ 20′′, of the point reached by Sir Francis Younghusband on the Urdok glacier would about correspond to that of this col. After consultation with him there appears to be but one conclusion possible—that this is the glacier he ascended in 1889 when in search of the Saltoro pass, and named the Urdok. The col he saw culminating the Urdok is probably a ridge of the mountain-group seen by us from the east Siachen col [Turkestan La]."*

The ambiguous words are in the middle of the second paragraph: that of this col." The reference cannot be to the Indira: it has been taken as to the Turkestan La: the last sentence makes it probable that "this col" does not refer to either of the cols visited by Mrs. Bullock Workman; and that her interpretation of the photographs was really the same as ours, but that she had obscured the argument by quoting the latitudes unnecessarily.

Some months ago, with the permission of Sir Francis, I tried to see if I could deduce anything from his observations for longitude, which had been rejected by the Survey of India; and as a preliminary I re-computed the latitudes, mostly meridian altitudes of Altair south, and extra-meridian altitudes of Polaris north. The latter required of course the sidereal time, to be derived from the altitudes of the Sun, and a deduced rate for the one watch which had survived to that date. At Shahidulla Sir Francis had recorded that the index error of his sextant was zero. It soon became apparent that in the Shaksgam it was about 3'. There was a discordance of 5' or 6' between the latitudes from north and south stars, pretty constant; and the means of the pairs agreed within a few seconds with those derived by the Survey of India. The agreement was as close as could be expected, considering the uncertainty of the sidereal time.

A recomputation of the local times with the deduced index-error did little,

*Two Summers in the Ice-wilds of Eastern Karakoram, by Fanny Bullock Workman, pp. 186 et seq.

however, to improve the concluded rates of the watch; the Sun-observations for time were all in the morning; the index-error was uncertain; and I was compelled to agree with the computer of the Survey of India, that no differences of longitude could be deduced with any advantage. But the observations for latitude seem unassailable, and it will be surprising if they are not confirmed some day by precise survey.

The Survey of India plotting shows the snout of the Urdok glacier a little away from the Shaksgam river, as described by Sir Francis. But it makes the route up the glacier run south-westerly to Camp II., thence a little south of south-easterly to Camp IV., and thence south. The authority for this curve in the glacier must be the compass-traverse of the orderly, which cannot now be found in the records of the Survey of India or in the Intelligence Department. The triangulation chart compiled by the Survey of India in 1922 gives a different outline of the glacier, which is evidently influenced by the misinterpretation of the Bullock-Workman narrative mentioned above. A. R. H.

T

AIDHAB

G. W. Murray, Survey of Egypt

HE hackneyed comparison of desert life with seafaring still holds good-long periods of monotony are broken by events as surprising as a shipwreck or a fire at sea. Last December on the Red Sea coast of the Sudan, I came across the jetsam of a great catastrophe of this sort. On a flat and waterless mound, in wet weather almost an island, separated by 10 miles of flat desert from the majestic wooded peaks of Jebel Erba, lie the ruins of 'Aidhab, a mediæval city of some importance.

It is perhaps mentioned by the tenth-century geographer, Ibn Haukal, who says that the produce of the gold-mines in the Wadi 'Alaqi was carried to a certain castle on the sea-coast called "Assab," or "Assat." (The reading is corrupt.)

Certainly during the period of the Crusades the land route to Mecca through Sinai and Northern Arabia became unsafe for pilgrims from Spain and North Africa, and they preferred to ascend the Nile as far as Sus or Aswan, from both of which places roads could be followed through the desert to 'Aidhab, a port on the African coast of the Red Sea opposite Jidda. Makrīzi indeed says † that for 215 years from A.H. 450 (A.D. 1058) the Mecca caravans followed this route, and that "it counted formerly among the most important ports of the world, because the ships of India and Yemen disembarked their merchandise there."

The Mecca pilgrims were not, however, safe even at 'Aidhab, for in 1182 Renaud de Chatillon, amongst other piratical exploits, sacked the town, massacred an entire caravan, and destroyed sixteen ships