UC-NRLF $C IS Tfl? ACCOUNT OF THE SURVEY OPERATIONS EASTFRN TURKISTAN, 1873-74, IIV CAPTAIN H. TROTTER, R, E. , f\j^>i>» }v WKiroi'.wtf^Pi'j THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA PRESENTED BT PROF. CHARLES A. KOFOID AND MRS. PRUDENCE W. KOFOID ACCOUNT OF THE SURVEY OPERATIONS IN CONNECTION WITH THE MISSION TO YARKAND AND KASHGHAR IN 1873-74. BY CAPTAIN HENRY TEOTTER, E.E., K DBPUTY-8UPBBINTBNDBNT, GBBAT TBIGONOMBTBICAL 8UBVEY OF INDIA. EXTRACTED FROM THE VOLUME OF REPORTS SUBMITTED TO THE GOVERNMENT OF INDIA BY SIR DOUGUS FORSYTH, C.B., K.C.S.I.. IN CHARGE OF THE MISSION. CALCUTTA : PRINTEt) AT THE FOREIGN DEPARTMENT PRESS. 1875. OOaSTTEN^TS. Page. Introductory 1 Road to Yarkand 3 Description op routes between Ladakh and Turkistan . 9 Excursions in the neighbourhood of Kashgar 17 Excursion to the Pamir steppes and Wakhan 29 The Oxus below Wakhan . 44 Return to Yarkand via the Great Pamir 49 On the construction op the map accompanying Report 54 Meteorology 59 Magnetism 61 Geographical Appendix — Section A. — Latitudes 1 „ B, — Longitudes 33 „ C. — Heights 45 „ B. — Alphabetical List op Latitudes, Longitudes, and Heights . 67 „ E. — Magnetic Observations 75 „ F. — Meteorological Observations 79 „ G. — Routes 123 Map of Eastern Turkistan In cover. Plan op Yarkand . , To face page 9 Khotan „ „ 15 „ Kashghar „ „ 17 Meteorological Diagrams . „ „ 60 ivi3l.'^8«6 Digitized by the Internet Archive in 2008 with funding from IVIicrosoft Corporation http://www.archive.org/details/accountofilurveyoOOtrotrich CHAPTER VII. GEOGRAPHICAL EEPOET, When the Government of India had decided to send a diplomatic Mission to the Atalik Ghazi of Kashghar it was determined to appoint an officer of the Indian Survey Department to accompany the expedition as Geographer ; and Captain Henry Trotter, Royal Engineers, of the Great Trigonometrical Survey of India, was selected for the post. The preliminary arrangements were left by Mr. (now Sir Douglas) Forsyth entirely in the hands of Captain Trotter, subject to such advice as he might receive from Colonel Walker, R.E., the Superintendent of the Great Trigonometrical Survey. The only restrictions insisted on were that everything was to be arranged for mule carriage, and the survey baggage was to be limited to three mule loads, also that the services of two khalasies (carriers) only could be allowed to assist generally in the work. Abdul Siibhan, a Sub-Surveyor in the Topographical Survey Department, was subsequently permitted to accompany Captain Trotter to act as recorder and general assistant. Two of the Great Trigonometrical Survey " Pundits"* with their assist- ants were also placed at Captain Trotter's disposal. It has been found convenient to arrange this Chapter in the shape of a General Report by Captain Trotter, to which is added an appendix shewing in some detail the results of the observations, astronomical, meteorological, hypsometric, and magnetic, taken by that officer and his assistants. The appendix includes some observations by Captain Biddulph on the Ling-zi- thung plains and on the road to Maralbdshi, and also contains detailed accounts descriptive of the various routes followed by members of the Mission, as well as of others compiled from native information, principally by Dr. Bellew. CAPTAIN TROTTER'S REPORT. Introductory. The first point to be decided was as to the instruments and equipment to be taken, and this was an anxious matter ; it was impossible to say what sort of a reception we should meet with in Yarkand, and whether I should be allowed to use openly any survey instru- ments at all ; I had also to bear in mind the, to me, totally new condition, that my instruments would have to be packed and carried on mules, and taken over the highest passes in the world. (In the Indian Survey Department delicate instruments such as theodolites, &c., are always carried by men, and even in the survey party attached to the Abyssinian Expedition this rule, I believe, was never departed from.) This condition imposed the necessity of taking only moderate sized instruments and such as were not likely to be injured by violent shakes and jars. Fortunately among the instruments of the department there was one that had already done good service at Magdala, viz., a 6-inch Transit Theodolite, with micrometer eye-piece, by Troughton and Simms. On Colonel Walker's recommendation I took this as my mainstay for astro- nomical observations, and I may here add that I have used it constantly throughout my absence from India, and have been very much pleased with its performance. A few slight * The term by which it has been customary to designate natives employed by the Great Trigonometrical Survey Department on Trans-Frontier Explorations. A ( 2 ) alterations having- been made in the fittings, it was carried safely in one of our leathern mule tnmks for more than 3,000 miles over I suppose some of the most diflScult roads in Asia, without receiving the slightest injurj'^, or having at any time been unserviceable. For its size it is a most perfect instrument. I had, however, also to consider what I should use in case of secrecy being necessary, and for this of course there was nothing like a sextant, so I provided myself with a 6-inch sextant by Troughton and Simms, as well as a small pocket sextant by the same maker, taking with them the ordinary mercurial artificial horizons. I may allude to the fact that Captains BidduJph and Chapman both spent a short time at the head-quarters of the Great Trigonometrical Survey for the purpose of practising astronomical observations in anticipation of the probability of our party separating in different directions in Eastern Turkestan. It was arranged that the former should take with him a sextant and the latter a theodolite, a sister instrument to my own. There were, therefore, in camp duplicates in case of any accident happening to my own instruments. I also provided myself with a very small light theodolite for use on high peaks (where it would be unsafe to attempt to carry the large instrument) and for traversing along roads if opportunity should occur. Besides the instruments already named I had a supply of prismatic and pocket compasses (I may mention that owing to breakage and accidents I at one time ran short of pocket compasses, and I was much indebted to Colonel Gordon during the Pamir trip for the loan of a very good little instrument), and a small light plane-table, which I had specially made to fit on to my theodolite stand. A good hand telescope was fitted to the same stand which also served, when necessary, with a slight adaptation, for a Hodgkinson's Actinometer belonging to the Royal Society and lent to me by Mr. Hennessey of the Great Trigonometrical Survey for the service of the expedition. Colonel Roberts, the Acting Quarter Master General, kindly gave me an old astronomical telescope which had been in use for many years in the Quarter Master GeneraFs Department. This telescope was presented to the Dadkhwah of Yarkand on our return to India. It was proposed that I should take a complete set of instruments for observation of the magnetic elements — intensity, dip, and declination ; but considering the great bulk and delicacy of these instruments, as well as the time that would have been occupied in making the necessary observations — time which I could not expect to be able to snatch from more important duties — I determined to take the dip circle only, a small instrument and one not occupying very much time to observe with. Observations for declination (variation) I was able to manage with my theodolite, with sufficient accuracy for practical purposes. With regard to chronometers and watches for astronomical purposes, it was decided that I should only take pocket chronometers, and as the Survey Department could only provide me with one good one, I had to order two from England, a gold one by Bent, and a silver watch by Brock, a maker strongly recommended by the Royal Geographical Society. These watches, I found while crossing the Himalayas, could not be depended on for very accurate results, as the sudden and enormous changes of temperature combined with other causes to make the rates very irregular (although all were professedly compensated for temperature), and for that portion of our journey I consider it better to rely upon the Pundit's pacing, checked by latitude observations, rather than on differential longitudinal observations depending on these watches. During trips in Turkestan,however, where there was considerably less variation in temperature, the results are much more satisfactory, and in my excursion towards Ush Turfan the resulting positions in longitude depend entirely upon chronometric differences of time, as also in great measure do those between Kashghar and Panjah in Wakhan. For meteorological observations and for determinations of height I decided to run the risk of taking with me mercurial mountain barometers, feeling that the greater confidence that would be placed in results deduced therefrom would more than compensate for the risk of loss by breakage ; I procured two from Bombay and one from Calcutta, and on the whole I am glad that I did so, as I succeeded in safely transporting all three over the Himalayan ranges, having been able to get them carried by men on foot. On the other side of the mountains I was not so ( 3 ) fortunate. At Sanju one of them was blown down in front of my tent by a sudden violent gust of wind, and the very next day another, which I had entrusted to the care of Dr. Belle w, was, on account of the length of the march, given to a horseman to carry : the horse fell in cross- ing a river and No, 2 was smashed. The third survived, and regular observations were taken with it throughout the winter in Yarkand. It too was broken on the return journey, its carrier falling with it in a stream. I was also provided with numerous aneroid barometers, hypsometers and thermometers. The latter I had specially made to order in England, as none that I could procure in India were graduated low enough to register the minimum temperature to be expected in the higher ranges of the Himalayas. As was to be anticipated in a journey like ours, very many of these have been broken ; loose horses getting at night amongst the tent ropes, and in the extreme cold weather even inside the tent, have much to answer for. While at Dehra Dun ^ prior to starting I was occupied in making myself familiar with the instruments I was about to take with me, and in practising the observations I should probably chiefly be dependent on ; in drawing up and getting lithographed portable and compact forms for registration and computation of observations, and other miscellaneous preparatory work. I prepared a large number of sheets with all the most northerly points fixed by the Great Trigonometrical Survey on the frontier of India projected thereon, as well as the latest determinations of the Russian survey; in order that by whichever route we might go or return, or wherever we might wander, I might lose nothing for want of previous prepara- tion. These charts were not of so much service to me as I had hoped, as wherever there was a chance that they might be utilized, the vicissitudes of the climate, and the rapidity with which we had to travel, invariably intei*posed to prevent my making full use of them. Colonel Walker also designed, and had prepared and photozincographed, a star chart, projected on a new principle, showing only stars of the first three magnitudes, nautical almanac stars being distinguished from all others. I found this chart a very valuable practical guide while observing. On Sir Douglas Forsyth's application to Colonel Walker four of the Great Trigonometri- cal Survey " Pundits," or rather two of the old Pundits with two assistants, were attached to the Mission, as it was hoped that an opening would occur for the despatch of these men from Eastern Turkestan across the Gobi Desert and through Thibet to Hindustan. It was not deemed advisable however to employ them thus, and when the Mission advanced from Yarkand to Kashghar it was necessary to leave them in Yarkand. Permission was given, however, for one of them to follow me to Kashghar, and he did useful work, of which more hereafter. The remainder were employed in Yarkand during the winter in taking meteoro- logical observations. Boad to Yarkand. For a few marches from Leh, in every direction, the country has been carefully and correctly surveyed and mapped in former years by parties of the Great Trigonometrical Survey under Major Montgomerie, R.E., but between this rigorously executed survey (bounded on the north by the head waters of the Nubra and Shy ok, and on the north-east by the Ling-zi- thung plains) and the table-lands of Turkestan, lie vast tracts of mountainous country, parts of which, through the enterprise, zeal, and energy of Messrs. Shaw, Hayward, and Johnson, have been mapped with tolerable accuracy, while other parts have probably never yet been traversed by man, certainly not by geographers. It was my object to weld together as far as possible the existing materials into a harmonious whole and to add whatever I could to existing data. It had been decided that the Mission should proceed to Shahidula in two parties, the head- quarters going by the old Karakorum route, whilst a detached party, consisting of Captain Biddulph (in command). Dr. Stoliczka, and myself, was directed to proceed via Changchenmo by the route by which the former Mission returned from Yarkand in 1870, and as we had * The Head-quarters of the Great Trigonometrical Survey. ( 4 ) ■ several days' start of the main party it was hoped that we might be able to discover some alternative route by which that line of road might be shortened and difficulties avoided. The delay of the Hadji Turrah Sahib, the Yarkand Envoy, in Constantinople made it necessary for the advanced party to halt at Leh until authentic news should arrive of his departure. Authority to advance was not received until we had been there for more than a fortnight, and we left it finally on the 12th September with orders to join the main party at Shahidula on the 20th October. This unfortunate though unavoidable delay not only deprived us of so much time for prosecuting geographical investigation, but postponed our departure to so late a season that inclement weather proved a serious hindrance to our advance, and a still greater impediment to me in carrying out the programme I had laid down for myself. As the best arrangement I could make, one Pundit with an assistant was left at Leh to proceed with the head-quarters camp ; the other one with a better instructed assistant, capable if necessary of doing independent work, was to accompany our own party, and in the probable event of our separation was to be attached to Captain Biddulph, the assistant being attached to Dr. Stoliczka. Abdul Subhan, the Native Surveyor, or " Munshi^' as he is generally called, was to accompany me as recorder and general assistant, and to be sent out with the plane-table should opportunity occur. It might have been expected that the presence of such a large party would have facilitated arrangements for carrying on work ; but the exact contrary was the case, as the demands on the limited resources of the country were so great that it was found impossible to send out any detached parties, the Native Surveyors were therefore obliged to accompany the main camps, to march when they marched, and halt when they halted ; and as the marches are arranged for the convenience of travellers and not of Pundits, some of tiiem were found uncommonly stiff and difficult to get through before dark. As the Pundits were in pairs a great part of the way and thus able to divide the work, the ground was got over with a fair amount of accuracy : and checked and corrected by the latitude observations taken both by themselves and myself on the outward and return journeys, the routes are certainly laid down with an amount of accuracy not hitherto attained. I should explain that in making my plans I was guided by the peculiar nature of the survey work generally done by these natives. This consists of a traverse survey, the angles of which are measured with a prismatic compass and the distances determined by the number of Pundits' paces. These paces have a slightly different unit of length, which is generally determined at the close of operations by comparing the total amount of northing or southing as shewn by the traverse, with the true corresponding distance as determined by the difference of latitude between the starting and closing points. The Pundits are all able to take latitude observations with a sextant and are instructed to do so wherever opportunity occurs. It is obvious that the accuracy of the survey depends upon their being able to keep up a continuous measure of the road ; any break in it would ruin the work. Hence the necessity, if possible, of their working in couples, so that they may relieve each other in the pacing, especially where, as in the present case, they were obliged to accompany the large camps and could not select their own halting places. The days were getting short, and if darkness once overtook a man before he had concluded his work, there was every probability of his whole survey being spoiled. Having thus arranged for the two main lines of road to be laid down with consider- able accuracy, I was free to devote myself to what I considered a very important matter, viz., the fixing accurately the correct positions of certain points on the line of march. I wished to do this either by triangulation in continuation of the Great Trigonometrical Survey system of triangles, or by running with the small theodolite a very careful traverse of the road. With the Munshi's help I trusted to be able to carry this traverse up to Shahidula, a point whose position it was very important to determine with accuracy. I also hoped occasionally to place myself in posi- tion on the plane-table by means of certain trigonometrical points which were fixed years ago by the Survey Department in advance of the accurate detailed survey. Many of these points are in the main Karakorum and Kuen Luen ranges ; some of them in the heart of the terra incognita ( 5 ) before alluded to ; and had I had more time at my disposal, and had the weather been more favourable, I might have done very valuable work. As it was, owing- to the antagonism of the elements my diary shows one almost continuous succession of disappointments, most dis- heartening under the circumstances that it was the beginning of the journey, and that I did not know but that circumstances might prevent any work being done after entering Yarkand territory. Climbing hills at the great elevation we were then at was very hard work, and of course occupied considerably more time and labour than similar ascents at a lower level ; and in nine cases out of ten when one did arrive at the top of a high hill, snow and clouds entirely obscured both distant and neighbouring ptaks. This cloudy weather combined with the necessity of regulating halts and marches according to the places where supplies had been laid out, soon made it evident that it was useless to attempt a continuation of the triangulation. The length of some of the marches and the shortness of the days made the execution of a careful traverse, as impossible as the triangulation, and after some very hard work, I reluctantly came to the conclusion that nothing could be done by myself (i.e., in addition to astronomical work), but to make what use I could of the plane-table. Even with this but little was done owing to the extremely unfavourable state of the weather ; but I fort.unately succeeded in fixing my position satisfactorily at two or three places on the road to Shdhidula, the most northerly point where I did so was at Chibra, south of the Suget Pass. Throughout the journey the cold was so intense that even the Bhots"^ who were with me used, on arrival at the top of a hill, to lie down in hollows or crouch behind stones in order to avoid the bitter blast. Under these circumstances, satisfactory work could not be expected, and although I kept my own health in a wonderful manner, I had the misfortune to knock up more or less nearly every man who accompanied me. From the 24th September, the day on which we reached Gogra, until the 17th October on arrival at Suget, I was never at a lower level than 15,500 feet, and during the whole of that period the thermometer seldom rose as high as freezing point (33° P.), whereas at night the minimum would vary from zero to 26° below zero.f From 26th September to 8th Octo- ber I was never below 16,300 feet, my highest camp being at Dehra KompasJ 17,900 feet above the level of the sea.§ Snow was frequently falling throughout the whole of this period, and for three days was the only substitute for water, for both man and beast. Captain Biddulph, who travelled by a more easterly route than Dr. Stoliczka and myself, was living at even a greater elevation. On the 13th of October Dr. Stoliczka and myself reached Aktagh, where we joined the head-quarters camp. It had originally been arranged that Captain Biddulph, accompanied by a Pundit, was to cross the Karakorum line of road and explore the country to the west of Aktagh towards Kufelong. Dr. Stoliczka, accompanied by the other Pundit, was to have crossed from the Karatagh Lake by a new route to the Karakash River, but his illness, as well as the weak state to which the camp followers were reduced from lengthened exposure to cold and hard work, made it necessary to alter these arrangements. I was now directed to go up the Karakash River, and endeavour to find the road which was believed to exist between some point up the river and the Karatagh Lake, and which it was supposed might turn out a good alternative route. I returned from this expedition, the details of which are given in another place, on the 20th October, and left the following day with a party under the orders of Colonel Gordon for S^njii with instructions to await Sir Douglas Forsyth^s arrival at that place. As Shahidula was the first point where we struck the Atalik's dominions and met his people, I briefly give the result of survey operations up to that point. * Bhots are inhabitants of Ladakh. t i.e., 58° below freezing point. J So called from having been used as a camping ground by a former Survey Officer or Kompas (compass) irala, the native designation for all surveyors, § It was the hardships encountered while traversing this elevated region that brought on the illness which subsequently cost Dr Stoliczka his life. B ( 6 ) One Pundit and his assistant accompanied the head-quarters camp and were kindlj looked after by Captain Chapman, who himself took some astronomical observations along the road. They ran a route survey from Leh, vid the Khardung Pass, up the Nubra Valley to Changlung, thence by the Saser Pass to Saser, from which place the Pundit proceeded by the winter route up the Shyok River and by the Eemo glacier to Daulat Beguldi, while his assist- ant took the summer route by Murghi and the Dipsang plains to the same spot. Thence they proceeded by the regular road over the Karakorum Pass to Aktagh, from which place they carried their traverse down the Yarkand River for three marches to Kirghiz Jangal, returning thence to Shahidula by the Kirghiz Pass. Kishen Sing, the Pundit, accompanying the advanced party commenced his route survey at Chimray, two marches east of Leh. At Zingral, the next halting place, his assistant diverged from the main road, going over the Kay La (Pass) and joined us again at Tankse. The Pundit went with the main camp over the Chang-la.^ From Tankse we all proceeded to Gogra, whence the Pundit was detached to accompany Captain Biddulph, who went over " Cayley'sf Pass'^ and the Ling-zi-thung plains, considerably to the east of the road by which the former Mission returned from Yarkand in 1870, which road, however, he rejoined at Kizil-Jilga, thence following the Karakash in all its bends down to Shahidula. Captain Biddulph took numerous observations for height on his line of march, generally using one of the mercurial barometers for that purpose. The Pundit kept up a continuous route survey the whole way and took frequent astronomical observations for latitude. Both Captain Biddulph's and Pundit Kishen Sing's observations will be found in the Appendix to this Chapter. This Pundit's assistant, aided by the Munshi (as soon as I became convinced that a theo- dolite traverse was impracticable), carried a route survey along the road I myself followed, ix.y the one by which the former Mission returned from Yarkand, by the Changlung-Pangtung Pass. This road skirts the west edge of the Ling-zi-thung plains and striking the Karakash River near its head, follows the course of that river until it turns off suddenly to the north-west, a point a little beyond Khush Maidan; thence the road passes via, the Karatagh Pass and Lake to Aktagh. From Aktagh it goes over the Suget Pass from which place I sent the Route Surveyors up in a north-west direction to cross the hills in front and stike the path passing from the Khirghiz Pass to Shahidula. My object in detaching them by that road was to enable the Munshi to fix himself in position by some of the survey peaks on the Karakorum away to the west. He had one fine day and succeeded in doing so, but at the cost of frost-bitten fingers, from which it took him a considerable time to recover. Throughout the march I made astronomical observations with my theodolite which have been reduced (in duplicate) in the head-quarters office of the Great Trigonometrical Survey since my return to India. They, together with my fixings by the plane-table, as well as my astronomical work on the return journey, and the Pundit's own observations, form the basis on which the whole of the Pundit's traverses have been built up. To this frame I have added such material as is available from the maps of the Trigonometrical Survey and of Messrs. Johnson, Shaw, and Hayward. The whole combined form a map more accurate and complete than anything yet published, and should, for geographical purposes, as far as the actual lines of road are concerned, leave but little more to be desired. Descriptions of the routes traversed by various members of the Mission will be found in the Appendix, Section, Routes. As regards this early portion of our journey the only new contributions I can give to science and geography are the results of a boating expedition on the Pangong Lake, and an account of the excursion, already referred to, which I made from the neighbourhood of Shahidula to try and discover an alternative road on to the Karatagh plain. The Pangong Lake district has been described at considerable length by Captain Godwin Austen in the Royal Geographical Society's Journal for 1867, and the additional * 17,590 feet. [ f 19,280 feet. ( 7 ) information I can now supply refers to the depth of the lake, an interesting subject of enquiry, and one which has, I believe, never been investigated with reference to this or any other of the Himalayan Lakes. A portable India rubber boat, which I had procured from England in the hope of ultimately floating it on Lake Lob, was the means which enabled Captain Biddulph and myself to make a section across the bottom of the lake. We arranged to halt a day for the purpose at Lukong. The soundings were taken by Captain Biddulph with a fishing line, which I had carefully measured and marked before starting. We fortunately had a quiet day, and owing to the entire absence of wind and current there was not the slighest difficulty in getting these soundings most accurately. Starting from the sandy shore at the west end of the lake we made for the island, lying about two miles off and situate nearly equidistant from the two sides of the lake. At 100 yards from the shore {N.B. — The horizontal distances are only rough estimations) the depth was 55 feet, the bank sinking gradually; 150 yards further on, the depth was 93 feet, and 200 yards further \\% feet; at 550 yards more, depth 130 feet, the maximum depth reached between shore and island. As we approached the latter the water grew rapidly shallow. At a distance of 400 yards from it, there was only 50 feet of water, and at 250 yards only 14 feet, from this point a shelving sandy bottom stretched up to the island, which consisted of a mass of rocks, about 150 yards in length and considerably less in breadth, of irregular shape, and extending in a direction parallel to that of the lake, viz., from north-west to south-east. It was composed mostly of calcareous tuhfa, and in no place rose to more than four feet above the surface of the lake. The rock was very brittle and jagged, and in many places was covered with masses of shells, of which I brought away specimens : these shells appeared to me to have become only recently untenanted, but they were pronounced by Dr. Stoliczka to be many centuries old and to be fresh-water specimens. The island or rather islands (a short distance from the main rock in a south-east direction is a long sandbank rising only a few inches above the water) are submerged during heavy storms, for we found many fragments of wood, weeds, and even cattle dung, which had evidently been washed ashore from the mainland. Although the water was beautifully clear we looked in vain for fish, and with the exception of a species of bug, of which myriads were swimming about, we failed to see any animal or signs of life of any description. This is the more curious, as in a small stream which flows into the lake near Lukong there is an abundance of fish. The temperature of the water, which was decidedly brackish, was 55° F.; its color, a very pure blue where deep, and green where shallow. From the island I pulled to the north shore of the lake, which lying under precipi- tous cliffs I expected to find much deeper. The water deepened out gradually to 107 feet at 800 yards from the island, and at about half way across, say half a mile from shore, there was a depth of 136 feet; at 250 yards from the mainshore we found 142 feet; at 100 yards, 114 feet; at 50 yards, 95 feet; at 30 yards, 80 feet; at 20 yards, 50 feet; at 10 yards, 12 feet. The boat^ 12 feet long, was very well adapted for work of the kind, and in the absence of wind I was, without violent exertion, able to pull it along, carrying one passenger, at the rate of nearly, if not quite, four miles an hour, and this at an elevation of 14,000 feet above the sea. In fact the exertion required was considerably less than would be needed for walking at the same pace. The banks of the lake, which is about forty miles in length, showed evident signs of the water having formerly stood at a much higher level than it does at present : and there can be little doubt but that the valley along which the road passes from Tankse up to the lake, was, at no very distant period, its main outlet ; for although there is now a low pass, about two miles from the head of the lake, yet it is not much more than 100 feet above the present level of the water, and is moreover chiefly formed by detritus washed down from two side ravines, and of very recent formation. With reference to my short excursion up the Karakash in search of a new road, I left Suget on the morning of the 17th October, and was accompanied as far as Balakchi (9 miles) by Drs. Bellew and Stoliczka, who were paying a visit to the jade mines. After leaving them I marched on for two and a half hours to Gulbashem, where I met Captain Biddulph on his way down the Karakash to Shahidiila. My syce (groom) and guide, the only man in our united camps at Suget, who professed to know of the existence of the road I was now searching for, had led ( 8 ) me to believe that at about one kos (^ miles) from Gulbashem, by turning up a lateral ravine on the south, a journey of 3 kos would bring me to a very low and easy pass, with an almost imperceptible ascent, from the top of which we were to look down upon the Karatagh plains ; and the evening we were at Gulbashem he pointed out a neighbouring spur, beyond which, he said, the road turned off. What then was my annoyance when the next day we did not reach the turning until after a long and difficult march of 13 miles up the Karakash River, and even then, according to his account, the pass was further off than he had stated it to be the day before. I was extremely vexed at thus partially losing a day, for my camp was so far behind (I having arranged for a short march hoping to get up to the pass and back before night) that there was no time to move it much further on that day, and I had to halt at the entrance of the valley leading to the pass. I had now only two days left in which to visit the pass and return to Shahidiila, as it had been arranged that I should have to be there on the 20th so as to be ready to leave with Colonel Gordon on the 21st. Starting early in the morning of the following day and quitting the Karakash River (at a point 12,500 feet above the sea), we went up a broad open ravine, running south for 2^ miles, to a point where it divides into two branches of which we followed the eastern for about half a mile up a steepish ascent to a point where this also divides into two smaller ravines with a steep spur running down between them. These two ravines were bounded externally by precipitous banks several hundreds of feet in almost perpendicular height. There was apparently no practicable path along the lower portion of these defiles, but the road zigzags up the spur running down between them, and then follows the left hand one. The top of this first ascent was about 2,500 higher than our camp on the Kara- kash River. From the top of the zigzag, the road up which, though steep, was good and practi- cable for laden ponies, we reached a more open country and the road now followed a broad grassy ravine with a gentle but steady rise. I followed this for seven or eight miles rising to about 16,800 feet, and there was still a stretch apparently of several miles, of gently undulating ground in front. As it was getting late in the day, and there was no time for me to go further, I climbed up a hill from which I obtained a view of the water-shed. I sent on the Pundit (the one who had been accompanying Captain Biddulph, and whom I had brought back from Gulbashem) with instructions to follow the ravine to the water-shed, and to go beyond and fix by intersection either the Karatagh Lake or the hill at Tamba camping ground between Karatagh and Aktagh, and, if possible, to return by Aktagh and rejoin the Head-Quarters camp at Shahidula. Pie succeeded the following day in passing the crest which turned out to be not more than a couple of miles from where I had left him, and in fixing by intersections from a distance the position of the hill before alluded to ; but the guide having become seriously ill and no one else knowing the country, and the whole of the Karatagh plain being several inches deep in snow, the Pundit was obliged to return via Gulbashem rejoining the camp after I had left with Colonel Gordon. 1 returned myself that same evening to my camp at Fotash in the Karakash River, not reaching it, however, till late at night and getting two or three falls on the road, which, although tolerably good by daylight, was in certain places, especially at the foot of the steep zigzag, by no means easy in the dark. I do not think that this route is likely ever to come into general use, for although it is perhaps a better road than that between Shahidula and Karakorum Brangsa {t-id the Suget Pass), yet it is much longer. From Shahidula to Brangsa the road vza the Karakash is at least 15 miles longer than the one by the Suget Pass. Should, however, the road via Kizil-jilga and Karatagh come into frequent use, I believe the Fotash route might be employed with advantage, as there is very little difference in length of road, and grass and fire-wood are to be found in abundance all the way from Shahidula up the Karakash River, and for Uoo miles up the (Fotash) ravine, after which there is plenty of grass all the way to the pass as well as any amount of Boortsee.* There were numerous tracks, on the higher * Boortsee is a small plant with large woody roots which grows wild in large quantities, and is in many places the only fuel ohtainahle by travellers. ( 9 ) ground, of both hiang (wild horse) and wild yak, * a good evidence of the excellence of the grass. The road has evidently been occasionally in use, as it is marked in places where it might easily be lost in the snow by small stones placed in an upright position here and there on large rocks. My guide told me that he had only once travelled by it, when accom- panying a very small caravan of not more than five or six ponies, on which occasion, there being a great deal of snow on the Suget Pass, through which they feared they would be unable to force their way, they had resort to this alternative route to the Karakorum. I have never met another man who was acquainted with this road, and its existence appears generally unknown to the Bhots of Ladakh. I returned the following day to Shahidiila (33 miles) visiting en roide the jade mines between Gulbashem and Balakchi. These have been described at length by the late Dr. Stoliczka. It was deemed advisable, that from Shahidiila onwards all open survey or display of instru- ments on the road was to cease. Permission was given, however, for one of the most experienced and wary of the Pundits to take observations quietly with a small pocket com- pass, with which he carried his route survey up to Yarkand. I also made occasional observa- tions at night for latitude, so that a rough but tolerably correct survey of the road was obtained. On our return to India these restrictions were found to be unnecessary, and one of the Pundits proceeded from Yarkand via Sanjii and Shahidiila, and without any attempt at concealment paced and re-surveyed the road carefully. "When we left Yarkand for Kdshghar the Pundits were directed to remain behind, but one of them was permitted to follow a few days after, and made a very fairly accurate route survey up to Kashghar, the others had to stay in Yarkand until we left the country and returned (with the exception before mentioned of the man who went by Sanju and Shahidiila) with the advanced baggage party under Tara Sing, f via the Kugiar and Karakorum route. They carried a route survey from Yarkand up to Leh". The ground between Karghalik and Kiilunaldi (on the Yarkand River below Kufelong) had never before been surveyed. The Pundit who followed us to Kashghar did some good work ; besides making an excursion with Tara Sing to some of the neighbouring bazaars, whose position he was enabled to fix, he accompanied a party of the Mission on an excursion to the north of Kashghar in the Altyn Artysh Districts, making a traverse survey of the road. He also went with Colonel Gordon^s party to Sirikol (Tashkurghan), and returned to Yarkand by himself, diverging from the route by which he had come at Chehil Gombaz, whence he proceeded by the direct road to Yarkand, a distance of upwards of 100 miles, over ground that has never previously been sur- veyed. It was arranged for him to return to India ^7^a Khotan, a journey he accomplished most successfully. Prior to leaving the country he paid a visit to the Sorghak Gold Fields in latitude north 36°39'51" longitude 82°42' east of Greenwich, about 160 miles (by road) to the east of Khotan.* Returning thence to Kiria he found his way back to India by Polu, Noh, and the Pangong Lake, a route running from one and a half to two degrees to the east of the most easterly route we have hitherto possessed, viz., that traversed in 1865 by Mr. W. H. Johnson in his journey to Khotan. Details of the route followed by the Pundit are given in the Appendix, as are also his observations for latitude and height. Description of routes between Ladakh and Turkestan . It is not proposed to give here a detailed description of the lines of route followed by the members of the Mission on their travels between Ladakh and Eastern Turkestan, { but a brief comparison of the various lines of road between the two countries may not be uninter- esting. * A yak is a species of mountain ox which only lives at great altitudes, and is much used for the carriage of merchandize over snow and ice. On ice they are far more sure-footed than any other beasts of burden. t Treasurer to the Mission. X Details of the roads traversed by the members of the Mission will be found in Section G. of the Appendix to this Chapter. C ( 10 ) For practical purposes these routes may be divided into threCj viz. — The Karkkorum route with variations (leading to Yarkand). The Changchenmo route with variations (leading to Yarkand and Khotan) . The Rudokh (or Changthang)'^ route (leading to Khotan). The Karakorum route may be subdivided into two, the Zamistani or winter and the Tabistam or summer road, and although these have a few marches and camping grounds in common, and cross the great water-shed between India and Central Asia at the same point, the Karakorum Pass, yet they diverge from each other throughout the greater part of their courses. As a rule it may be laid down that the winter road passes wherever possible along and over the beds of rivers, which in the cold season contain but little water, and are generally frozen over : these streams, which form no obstacle in winter, are often impassable torrents in summer. It is therefore no matter of surprise that in spite of the intense cold and hard- ships of a winter journey the merchant often selects that season for his travels. The first great obstacle to be encountered after leaving Leh, both in summer and winter, is the well known Kailas range, which is said to run in one unbroken line from the sources of the Indus to the junction of that river with the Shy ok. To the north of Leh this range divides the drainage of these two rivers, and is one of the most formidable obstacles to be encountered by the traveller to Turkestan. In winter it is crossed by the Digar Laf (17,930 feet above sea level), a very difficult pass, in crossing which it is necessary to employ either yaks or men for the carriage of goods. A party of the Mission went over it in June, and even then there was snow lying on the top while ice and snow combined to render the passage difficult along a distance of some miles. The summer road crosses the Khardung or Leh Pass, J almost north of Leh, and is 17,900 feet in height. This Pass also is impracticable for laden ponies, and is so difficult that late in June on our return journey from Yarkand, after descending the Nubra river, it was deemed advisable to go a long detour via the Digar Pass in order to avoid the still more formidable obstacles on the Khardung. This made the journey from Sati to Leh 42 miles instead of 29. After crossing the Kailas range and entering the Shyok valley the traveller has now before him the great Muz-tagh§ or Karakorum Range. In the winter by following the narrow, winding, and difficult valley of the Shyok river he reaches the Karakorum Pass, a distance of 114 miles ; in the course of this portion of the journey the frozen sm'face of the stream has to be crossed no less than 36 times. In winter this can easily be done, as it is generally bridged by snow and ice, but in summer owing to the floods caused by the melting glaciers an entirely different route has to be adopted, and instead of ascending the Shyok the traveller descends that river to a short distance below Sati and then ascends the Nubra river, a large tributary fed from glaciers in the same mountain mass that supplies the Shyok. The Shyok is crossed in boats near Sati, where in the summer it is a very large and rapid river. Passengers and goods are carried over in boats, while the baggage animals are made to swim across. Many of the latter are drownedfin crossing. Ascending the Nubra valley, one of the most fertile and richly cultivated in Ladakh, the traveller goes as far as Changlung (10,760 feet), almost the highest village in the valley, and situated about 40 miles above Sati. The merchant generally takes this bit very easily, advancing by short marches of ten miles each, in order to make the most of the supplies of grain and excellent lucerne grass, both of which are here obtained in abundance. The caravans from Yarkand often halt a week at Panamik (a large and flourishing village a few miles below Changlung) to feed and rest the baggage animals after the hard work and scant fare that they hav e had on the journey. It is here that on the outward journey the real difficulties of the march commence. Instead of following one stream right up to its source in the Karakorum P ass, as is done in the winter route, the traveller has first of all to cross a * Changthang, in the Thibetan language means ' northern plain." t " La " is the Thibetan word for " Pass." J Sometimes also called " Laoche La." § In Turki " Muz-tagh " means " Ice Mountain " and " Karakorum " is the equivalent of " Black gravel." ( 11 ) very high and precipitous hill just above Changlung village. The road ascends by a zigzag and rises rather more than 4^000 feet in a length of about five miles^ the stiffest bit of ascent on the whole journey to Yarkand. After reaching the top of the Karawal Pass (so called from a karawal or outpost erected many years go at this spot to enable the Ladakhis to defend their country from invasion from the north) the road descends into the Saser stream and then passes up it to the Saser La, a pass over a mighty ridge covered with snow and glaciers which runs down from the great mountain mass forming the eastern extremity of the so-called Karakorum Range and separates the waters of the Nubra"^ from those of the Shyok. This pass (17,820 feet) is one of the most difficult on the whole road, and is rarely, if ever, free from snow, while the road passes through, over, and alongside of glaciers for many miles.f The road from the top of the pass follows the bank of a stream which enters the Shyok River at Saser Polu, a halting place on the winter road. The Shyok is here crossed with difficulty, as is proved by the fact that two Ladakhis were drowned there when returning from laying out sup- plies for our return journey. The road now ascends a tributary stream on the left bank of the Shyok crosses a low pass^ and at Murghi Camp joins another stream which flows from the Dipsang plains into the Shyok River. It was at this point, at a height of 15,200 feet, that the late Dr. Stoliczka breathed his last, after having traversed the Karakorum Pass and the perhaps still more trying Dipsang plains which rise to an elevation of about 18,000 feet above the sea. The intense cold of this bleak and dreary waste prevents this route from being adopted in winter, during which season the caravans follow the Shyok River from Saser Polu up to Daulat Beguldi (Turki for " Daulat Beg died," an appropriate name for so desolate a spot) . This camp, which is situated in the north-west corner of the Dipsang plain, marks the junction of the winter and summer routes, which unite here, and cross the Karakorum Pass 11 miles above the camp, continuing together a distance of 40 miles further to Ak-tagh. The Karakorum Pass, though 18,550 feet above the sea, is by no means so formidable an obstacle as is generally sup- posed. It is always free from glaciers, and in summer from snow. The ascent on both sides is gentle, and the road good, so that, although it forms the water-shed between Hindustan and Central Asia, it is less of an obstacle to the merchant than the Digar, the Khardung, the Saser or the Sanju Passes. From it the road passes along the Karakorum stream (one of the head- waters of the Yarkand River) to Ak-tagh, traversing the comparatively open ground on the west of the Karatagh J plain. At Ak-tagh the roads again diverge, the winter route con- tinues down the Yarkand river, which is crossed 18 times between Ak-tagh and Kulunaldi,§ a distance of 74 miles. At the latter place this road ascends the range that was called by Hay- ward the western '^Kuen Luen,*' and crosses it by the Yaugi-Diwan (or "New Pass," 16,000 feet) into the Tiznaf River, which it follows for 41 miles to Chiklik. The road is here taken over one of the northern spurs of the Kuen Luen by the "Tupa"|] or Ak-Korum Pass (10,470 feet), whence it descends along the banks of a gently sloping stream to Kugiar, a considerable village (containing 400 or 500 houses) on the borders of the plains of Eastern Turkestan, and 41 miles distant from Karghalik, a large town situated at the junction of the Zamistani {via Kugiar), and the Tabistani {via Sanju) routes. It was by the Kugiar road that the Mission returned to India. The road had been closed for several years previously by order of the Yarkand authorities owing to the risk to which travellers were exposed of being plundered and sold into slavery by the wild Kanjud robbers (of Hanza and Nagar), who coming down from their fastnesses to the north of Bunji and Gilgit used to render the whole valley of the Yarkand * At the head of the Nubra valley a road passes over the main Kardkorum chain by the Chorbut Pass and descends into the Yarkand river at Khufelong. It was formerly much used by the Baltistan merchants, but is now rarely employed. It is probably not less than 19,000 feet high, and is always closed for at least nine months in the year, and is at no times practicable for laden animals. t On the return journey of the Mission several hundreds of coolies were employed for some weeks in preparing the road over this pass. X " Karatagh"=" Black Mountain." § " Kulunaldi"=" the wild horse died." II In Turki Tupa means "hill," and Ak-Korum "white gravel." ( 12 ) River from Kulunaldi up to Ak-tagh^ utterly unsafe for travellers or merchants unless in large parties and well armed."'^ It was in the month of June that the Pamir party returned by the Kugiar route some- what too late in the season to traverse it with safety, and considerable danger was incurred from the daily increasing floods of the Tiznaf River, which after noon used to come down with such force as frequently to close the road. At this season also the southern slopes of the Yangi- Diwan (Pass) are very difficult to traverse and somewhat dangerous, as the recently dead bodies of numerous baggage animals seen by us on the return journey too surely testified. The floods of the Tiznaf are probably worse in June and July than at any other time of the year, as after that period the snow on the lower mountains has nearly all been melted. The larkand River, on the other hand, above Kulunaldi, being fed more generously by glacier streams is more difficult later on in the hot weather. We found that although there was a much larger body of water in the Yarkand than in the Tiznaf River, yet in the former the bed was broad and level, and was crossed without difficulty ; whereas in the Tiznaf the bottom is narrow and generally composed of large stones and boulders which render its passage very difficult.f The road crossed it nearly 20 times in one march, or about once in every linear mile of its course. A month earlier in the season (May) the river was frozen and was ascended by an advanced party of natives without difficulty. 42. Returning to Ak-tagh, the point of divergence of the two routes, the summer road passes thence over a spur of the Kuen Luen by the Suget, a tolerably easy pass (17,610 feet), from which the road descends along a winding stream to the Karakash river which it strikes a few miles above Shahidiila. J At Shahidula the Karakash river winds through the Kuen Luen Range. § The road follows along it for some 20 miles, and occasionally crosses it. In summer its passage is effected by merchants with considerable difficulty. The Karakash flows in the direction of Khotan, and between the river and Yarkand lies a formidable spur from the Kuen Luen, which has to be crossed. The traveller, if he be here unfettered by political obligations, has the choice of three roads before him, viz., by the Kilik, the Kilian, and the Sanju passes.- Traders are seldom or never allowed to use the former which is said to be the easiest and * Note. — These robbers, apparently from fear of the Kashgbar Amir, have of recent years ceased to infest this road, but it is reported that since the return of the Mission from Yarkand, the Kunjudis have attacked a nomadic tribe called Phakpos, who inhabit numerous valleys on the west bank of the Tiznaf river. The road by which these robbers advance must pass over numerous glaciers, and crosses the Karakorum range by the Shing- shal Pass, a short distance to the west of the Shigar or Muztagh Pass. The road from Shingshal descends the Kum stream and joins the road from the Muztagh Pass at a distance of one and a half marches to the north of the latter. After three short marches more the Yarkand River is reached at Dahn-i-Bazar Darah, three short marches below Kulunaldi (on the same river), a frequently used halting place on the road between Karakorum Pass and Kugiar. The Shingshal Pass is said to be easier than either the Chhorbut or the Shigar Passes, and is at times passable by laden horses. The Muztagh Pass (which was estimated by Godwin Austen at 18,400 feet, and by the » Or natives of Baltistan, a mountainous Schlagentweits at 19,000 feet) road lies for a great distance over glaciers, district inhabited by Shiah Musulmans, and and is diiRcult and dangerous. It is occasionally used by the Baltis,^ Ijmg to e nor -west o a a - . ^^^ \ia,\Q a colony in Yarkand, and who traverse this pass when return- ing thence to their own country. t On one occasion during the return journey, when I had gone on a couple of days ahead of Colonel Gordon's party so as to have more time for survey, I had, in order to insure security from water, placed my chronometers in my pockets instead of in the mule trunks where they were usually carried. It was the first time that I had done so, and as ill-luck would have it, I twice got parted from my horse in deep water while searching for a ford, and had to swim for my life with my chronometers in my pocket. On the same occasion my horses and baggage animals were cut off from all supplies by the floods, and were for more than 36 hours without tasting food. X Note. — At Shahidula is a small fort which during the time of the disturbances in Eastern Turkestan (which resulted in the accession of the present King) was occupied by a detachment of the Maharaja's troops from Kashmir. These were subsequently withdrawn and the place is now generally recognized as belonging to the Kashgbar ruler. The Kirghiz of Sanju have of late years constantly occupied the Karakash valley up as far as the great bend above Sora, and occasionally ascend some of the valleys to the south, leading up to the Karatagh plain ; in many of these valleys there is abundance of grass and wood. § Dividing it according to Hayward's nomenclature into Eastern and Western Kuen Luen. ( 13 ) shortest ; it follows the course of the Toghra, a considerable stream which enters the K^rdkash nine miles below Shahidula. The floods of this stream in hot weather often detain travellers a considerable time on its banks. The Kilik Diwan (Pass) is crossed in the 3rd or 4th march from Shahidula, and after going- over another low pass the road joins the Kugiar route at Beshterek, one dajr's march to the south of Karghalik ; little is known of this road, but it is said that grass and wood are to be found at every stage. It was once much used by the Baltistan merchants who are settled in Yarkand. Nearly three miles below where the Togra-su enters the Karakash River is the fort of Ali Nazar, where the Kilian road leaves the Karakash valley and passes up an open ravine in a north-west direction. This road is sometimes used in the summer as an alternative to that over the Sanju Pass ; it is somewhat higher than the Sanjii Pass, but, although imprac- ticable for laden horses, is not so difficult to traverse. The Kilian pass is crossed in the second day after leaving the Karakash. The road follows the stream from the 'pass for four marches when it debouches into the Turkestan plain at the village of Kilian, two marches to the south of Bora on the road between Sanju and Karghalik.^ The third and most frequented road from Shahidula is via Sanju. It leaves the Karakash 20 miles below Shahidula at Mazar Abii Bakar, from which place the road ascends to the summit of the Sanju (also happily named "Grim^') pass which, although not more than 16,700 feet above sea level, was decidedly the most difficult obstacle encountered by the mission on the road to Ydrkand.f Its summit is never free from snow and ice, and is impassable by laden ponies. Yaks have always to be used and are collected from all quarters for the passage of a large caravan. From the pass the road descends to the Sanju or Sarikia River, which it follows to the large and scattered village of Sanju, on the borders of the Great Turkestan Plain. Occasionally in the hot season the S^njii River is so flooded in its lower course as to become impassable, in which case a detour is made by a road which crosses a small spur by the Chuchu Pass (11,800 feet), after which it follows the Arpalek stream to near Sanju. Thence a good and level road leads to Yarkand, a distance of 122 miles, and meets the Kilian route at Bora, and the Kilik and Kugiar routes at Karghalik. Returning now to the Changchenmo route from Leh to Turkestan, on this also the KaiUs range has to be crossed, but further to the east than on the Karakorum route. The road ascends the Indus for 20 miles, and then goes up a tributary stream for 13 miles to Zingral, from which place the range may be crossed either by the Chang La (17,600 feet), or the Kay La (17,900 feet). By the former and easier road of the two it is 23 miles from Zingral to the large village of Tankse, situated on one of the tributaries of the Shyok River. By the Kay La foot passengers shorten the road by some six miles. The roads over both passes, although free from glaciers, are very dfficult ; and it is usual, although not absolutely necessary, to employ yaks in carrying goods across. Tankse is the last place on this road where supplies are procurable, and is, by the shortest route, 350 miles from Sanju, the first large village encountered in Turkestan. For the whole of this distance supplies of grain, both for men and horses, have to be carried, and at a great many halting places neither grass nor fire-wood is procurable. From Tankse after passing Lukong at the head of the Pangong Lake, the road crosses a lofty mass of mountains, by the Lankar or Marsemik La (18,400 feet), a very high but in summer by no means a difficult pass. It is free from glaciers, and generally clear of snow during the summer and early autumn. Descending into the Changchenmo valley and crossing the stream, a tributary of the Shyok, the road ascends a minor stream to a point eight miles beyond Gogra, from which there is a choice of three different roads all leading on to the Ling-zi-thuug| Plains. The most westerly path ascends the Changlung Pangtung Pass (18,900 feet), crosses the corner of the plateau and descends into a deep ravine running along the stony and very difficult bed of a stream§ (which ultimately finds its way into the Shyok River), ascends again, and skirts the * In former years the Kilian would appear to have been the most frequented route, but it is now little used. t Several mules were lost here, although their loads had all been transferred to yaks. t or Ak-sai Chin. § The march down this ravine was one of the most trying encountered during the outward journey. D ( 14 ) western border of the gently undulating Ling-zi-thung Plain), in traversing which the traveller crosses, almost without knowing it, the water-shed between India and Central Asia. After passing the water-shed the road crosses a small stream, one of the head waters of the Karakash, and then goes over a spur (Kompas La) 18,160 feet in height and descends into the bed of Karakash River, which it strikes, at an elevation of 17,400 feet above the sea and follows to Kizil Jilga. The portion of the road between the Changlung Pass and Kizil Jilga is perhaps the most trying part of this route. The great elevation and consequent bitter cold is much aggravated by frequent snow and a piercing wind which blows from morning to night ; the long dreary marches cause one to arrive, after dark, at camps where there are scant supplies of fuel and no grass; occasional ice beds block up the whole road, one of these extends for three miles down the Karakash River ; all combine to try most severely both man and beast. At Kizil Jilga the road just described joins an alternative road (taken by Captain Biddulph on the outward journey), which, leaving the usual route a few miles north of Gogra, crosses the Changlung Barma Pass (19,300 feet) on to the Ling-zi-thung plains, along which it passes at a still higher elevation than the western road. It descends into the Karakash River at Kizil Jilga ; the greater elevation makes this road perhaps even more trying than the western route. The third route from Gogra before alluded to leaves the Changlung valley 8 miles above Gogra and the Ling-zi-thung plain may be reached by either the Changlung Barma or the Changlung Yokma Pass a little further to the east, and of about the same elevation; This is the pass taken by Mr. (now Sir Douglas) Forsyth in his first mission to Yarkand. By it, the road followed by Captain Biddulph (striking the Karakash River at Kizil Jilga) may be joined, but a more northerly route passing over a succession of elevated plains was taken by the former mission, and the Karakash River was met a few miles above Sora at the sudden bend that the river takes when its course is turned towards the west (in north lat. 35° 55') by the Kuen Luen Range. From this point the road followed the River to Sh^hidula. In addition to the intense cold the principal objection to all three routes skirting or passing over the Ling-zi-thung (also called Aksai Chin) is the extreme elevation at which the traveller has to remain for so many marches : the cattle are exhausted by this, and too frequently suffer in addition from the pangs of hunger and thirst. These difficulties nearly brought the first mission to Yarkand to a disastrous end, and the same causes have proved, and will probably continue to prove, sufficient to deter the experienced merchant from following this road. The older, shorter, and better known route by the Karakorum is likely always to be preferred by the merchant even in summer, whereas in winter an attempt to traverse the Ling-zi-thung plains must almost always result in disaster. From Kizil Jilga the road follows the Karakash River to Chong Tash (or " Great Stone") . From this point the eastern variation, taken by Captain Biddulph, follows the Karakash River right down to Shahidula, a distance of 166 miles, while the western or more direct road is only 113 miles in length, and although in the latter there are two high passes viz., the Karatagh (17,700) and the Suget (17,600) to be crossed en route, yet they are neither of them difficult ones. The Suget Pass may be avoided by going over the lower and still easier pass of Fotash by which the Karakdsh River is struck one march above Gulbashem. In the circuitous line from Chong-tash down the Karakash, the road is bad, but there is the advantage of plentiful supplies of grass and fuel which are almost altogether wanting on the Karatagh line. The Ling-zi-thang routes meet the Karakorum summer route at Aktagh or at Shahidula according as the western or eastern variation is adopted. At the angle formed by the Karakash River above Sora, when turned by the Kuen Luen range, the traveller can proceed to Khotan direct (a distance of 160 miles or 1 1 marches) by crossing the Kuen Luen Range by the Yangi or Elchi Diwan (crossed by Mr. Johnson in his journey to Khotan in 1865), and estimated by him at 19,500 feet in height; after passing this there is another formidable glacier pass, the Naia Khan (height 18,659 accord- ing to Johnson) which has to be crossed before reaching the plains. The Elchi Diwan is said to be open for only three months in the year. ( 15 ) On the Karakash River above Fotash is a camping- ground called Sumgal, from which Robert Schlagintweit crossed the Kuen Luen range by the Hindu-tash Pass, estimated by him at 17,379 feet high. At the top of this Pass is a glacier much crevassed and extremely steep. It is a long and difficult march from its foot, to the village of Bushia, where are numerous tents and caves occupied by Kirghiz, and where supplies can be obtained in large quantities. It is eight marches thence to Khotan and the road is described as bad. The road by the Hindu-tash Pass can only be used by foot passengers. From all accounts the ordinary trade route between Khotan and Lddakh in former years was, as at present, via the Sanjii and Karakorum Passes. The road from Khotan follows that to Y^rkand as far as Zanguia, whence a road goes to Sanju village direct. Another road from Shahidiila to Khotan lies down the Karakash River, and, going over an easy pass, emerges at Duba,"^ a large village said to lie about 20 miles to the south-west of Pialma (on the Khotan and Yarkand road.) The road down the Karakash can only be used in mid winter. We now come to consider the extreme eastern route, via the Chang-thang or " Northern plain." Of this road we have a survey by Kishen Sing Pundit, one of the most important geographical results secured by the mission. f Details will be found in Appendix, Section G., and the road itself is shown on the map accompanying this report. A traveller from Leh to Khotan might follow the route by the Pangong Lake, along which the Pundit travelled, but he would more probably take a short cut from Lukong to the Mangtza Lake, following the ordinary Changchenmo route to Yarkand as far as the point where that road leaves the Changchenmo valley. Passing up the latter, he would make his way eastward to its head, where an easy pass is known to exist leading on to the high table land beyond. By adopting this road he would save forty miles over the more circuitous road by Noh. From Mangtza the road lies over a series of high plateaux varying from 16 to 17,000 feet in height, crossed here and there by low ridges which rise somewhat irregularly from the surface of the plain which contains numerous lakes, most of them brackish. In latitude 35° 7' north the Pundit crossed at a height of but little more than 17,000 feet the water-shed of a snowy range, which may perhaps be the true eastern continuation of the Kuen Luen. From the north of the pass the Kiria stream takes its rise ; the road follows down it as far as Arash (16,000 feet), but again ascends to the Ghubolik plain, which (17,000 feet above the sea) connects the snowy range just alluded to with another somewhat lower range to the north. This last ridge is a buttress of the vast Thibetan plateau, and in descending the Polu stream from the Ghubolik. At Diwan % (17,500 feet) to Polu, a distance of 28 miles (including windings), there is a fall of about 9,000 feet. Polu is a small village in the Khotan district and from it Khotan (or Ilchi) city may be reached either by the direct road (by Chihar Imdm) which skirts the feet of spurs from the elevated plateau above, or the traveller may proceed down the stream to Kiria by the route followed by the Pundit. Throughout the whole of the road from Khotan to Leh traversed by the Pundit fuel was abundant everywhere, and there was only one stage where there was not a good supply of grass. These facts would indicate the line as one well adapted for the native merchant, to whom time is of no great value. As far as I can learn however from enquiry it never has been used as a trade route on a large scale, the chief reason I believe being fear of the Chang-pas§ or Tagh-lik, wandering tribes of Tartars, nominally subject to the Chinese officials at Gartokh and Rudokh, but probably practically only so far subject to them that they would abstain from committing violent aggression on parties travelling under the protection of those authorities. * Duba is shown on Klaproth's map as a large place about half way between Zawa and Sanju, t The only previous account we have of this road is one derived from native information supplied by Mr. E. B. Shaw, and which was published in the proceedings of the R. G. S., No. III. of 1872. This account agrees remarkably well with that given by the Pundit, and every march can be followed on the large scale map I have before me as I write. + Or " Sulphur Horse Pass" so called from its being used by the Polu people when bringing sulphur to Khotan. Sulphur is excavated in large quantities from the ground near the lake in the Ghubolik plaiu. § Chang-pa in Thibetan means Nord-man while the Turki name for the same people is Taghlik, i.e. Mountaineer. ( 16 ) Habibula, who was elected King of Khotan when the Chinese were turned out of the country, sent messengers to try and open up this route in 1864. They were seized by the Chang- pa and compelled to return to Khotan with the threat that any subsequent explorers would be put to death. The inhabitants of Keria and Polu go as far south as Ghubolik to procure sulphur. They also go west of this towards the head of the Yurung-Kash (or Ilehij River where they search for gold and jade, but it would appear that although the Khotanese claim the country up to Lake Yeshil Kul, the head of the Kiria River, as their boundary, yet practically from fear of the Chang-pas they never go quite so far to the south. On the other hand the Chang-pas who probably have equal reason to fear the Turks from the plains, would appear not to wander further north than Rikong Chumik, the ridge to the north of which separates their grazing grounds from plains on the north, through which flows a considerable stream, passed by the Pundit, asserted by his guide to be the head of the Yurung-Kash River."^ It would thus appear that owing to the mutual hostility of the two races there is a large tract of neutral ground which is never occupied by one or the other, extending from Rikong Chumik to Ghubolik ; here the Pundit saw large herds of yak, antelope, and jungle sheep [oves ammon), which had apparently never been scared by the sight of man. Near Rikong Chumik were the remains of numerous huts ; others were frequently seen along the road, but fortunately for the Pundit, he did not meet or see a single human being between Ghubolik and Noh, a distance of 244 miles, a circumstance which enabled him to complete his route survey up to Nohf without interruption. The newly acquired knowledge of this road may perhaps lead to important practical results, but not until our relations with the Chinese Empire, and their too independent subordinates in Thibet, are placed on a more satisfactory footing than they are at present. It is apparent by combining the results of this survey with other information collected by the Survey Pundits during the past few years, that a road exists between the plains of Hindustan and Turkestan which entirely avoids the territories of the Maharaja of Cashmere, and which in the summer months may be traversed without once crossing snow, or without encountering one really difficult pass, such as we know to exist on the Karakorum and Changchenmo routes. Leaving the plains of India at the ancient city of Najibabad (between Hurdwar and Moradabad), the starting point of the old Royal Road stated by Moorcroft to have crossed these same mountain systems, a good road about 210 miles in length, and only crossing one low pass,J leads to the Niti Pass (16,676 feet high) over the main Himalayan range. Descending from the Niti Pass, due north into the Sutlej valley, and crossing that river at Totling (12,200 feet) by the iron suspension bridge still existing (said according to local tradition to have been constructed by Alexander the Great) , and crossing by the Bogo La (19,210 feet) into the Indus valley at Gartokh (14,240 feet), the road would then follow that river to Demchok.§ Thence it would go oVer the Jara Pass due north to Rudokh and Noh, and by the newly surveyed route to Polu and Khotan. Estimating the distance from Najibabad to the Niti Pass at 210 miles, thence to Noh at 275, and from Noh to Khotan {via Keria) 446 miles, we have a total distance of 931 miles * In the map which has been prepared for submission with this report I have not shown this stream as flowing into Yurung-Kash, but I think it not at all improbable that it may find its way through a gap which I have left in the Kuen Luen (just between the letters E. and N. of Luen). I would have inserted it, but it hardly appears consistent with Mr. Johnson's statements as to what he saw when ascending these Kuen Luen peaks in 1865, although, on the other hand, the fact that the river he crossed at Karangolak was a very large and rapid stream would indicate that it probably came from a considerable distance ; knowing also as a fact how the Karakash cuts through the same range at Shahidula and how extremely difficult it is to form an accurate idea of any mountain range when viewed from a single point, I am inclined to regret that I did not show this stream in my map as the head waters of the Yurung-Kash or Eiver of Khotan. t Prom Noh he tried to get to Eudokh, but was not permitted to do so ; in fact the inhabitants tried to com- pel him to return by the way he had come, and it was with great difficulty that he at last got pennission to go to Leh direct. Anticipating a search by the first people he should encounter, he had, when nearing the village of Noh, concealed his instruments and papers in a bush. He was duly searched, but of coui'se nothing was found, and he afterwards succeeded in again getting possession of his valuables. In Thibet the great difficulty en- countered by persons entering in disguise is always on the frontier, where the examination is very strict. When once allowed to pass into the interior of the country there is little to fear. X The Langar Pass 6,500 feet high which is on the 3rd day's march from the plains. § A more direct route existfs from Totling vid Dankhar to Demchok. ( 17 ) between Najibabad and Khotan, and this even might be considerably shortened by taking the direct road from Polu to Khotan. [The ancient Eoyal road probably followed the above to the suspension bridge at Totling, and thence to Rudokh and Noh, whence a road now exists which passes via the head of the Changchenmo valley and Nischo into the Ling-zi-thung plains, down the Karakash river and over the Sanjii Pass to Sdnju (or Sarikia)* which is half way between Yarkand and Khotan.] Summarizing our knowledge of the lengths of the various physically practicable routes from Hindustan to Turkestan we find that the distances are : — Miles. From Amritsir to Leh via Kawul Pindi and Srinagar ... ... ... 635 >j >» to „ via Kangra ... ... ... ... ... 622 „ „ to „ t^id Sealkote and Cashmere ... ... ... ... 575 From Leh to Yarkand OT'd Ling-zi-thung and Karakash Eiver ... ... ... 584 „ „ „ via Changchenmo and Karatagh ... ... ... 527 „ „ „ ««d Karakorum Pass and Sanju (summer route) ... ... 445 „ „ „ wd Karakorum and Kugiar (winter route) ... ... 472^ „ ,, „ via Noh, Polu, and Khotan ... ... ... ... 839 „ „ Khotan w'd Karakorum and Sanju ... ... ... ... 415 ], „ „ vid Ling-zi-thung and Elchi Pass (Mr. Johnson's route) . . . 437 „ „ „ md Noh, Polu, and Kiria ... ... ... ... 637 „ Amritsar to Yarkand by the road followed by the Mission, i.e., via Eawul Pindi, Srinagar, Leh and the summer Karakorum route ... ... ... 1,080 „ Najibabad to Khotan rid the Niti Pass and Western Thibet ... ... 931 At some distant day it is not impossible that the last named road may form the highway to Turkestan, but as long as Europeans are rigorously excluded from Western Thibet we cannot hope that this consummation will be realized. Excursions in the neighbourhood of Kdshghar. During the winter at Kashghar I was permitted to make two excursions in the neigh- bourhood, both of which have enabled me to add something to our geographical knowledge. The second trip was over ground, which, as far as I am aware, has never "hitherto been explored, and is very incorrectly represented on existing maps. During the first of these trips, which occupied us from the 31st December to 10th January, Dr. Stoliczka and myself, under the orders of Colonel Gordon, visited the Russian frontier at Lake Chadyr Kul, about 110 miles north-west by north of Kashghar. We had hoped, from the extreme point reached by us, to have struck off to the Terekty Pass on the east, and to have returned by the Terekty Forts to Kashghar. Unfortunately difficulties were placed in the way of our doing this, and we had to return to Kashghar by the same road that we went. Prior to starting, permission was given by the Amir for me to use my instruments on the road, and I may here mention that from this time forward during the whole of my stay in Kashghar territory I was at liberty to use openly what instruments I chose. Of course a certain amount of caution was necessary. Many of the officials with whom I came in contact were doubtless very suspicious as to what it all meant, yet in no case did any one attempt to hinder my taking observations or notes, although in many cases they endeavoured to neutraHze the value of my work by giving me false information on geographical subjects. I allude to this * The three points that have indicated this as the line of Eoyal Eoad are : — ^ 1st. — Moorcroft's statement that the road started from Najibabad and emerged in the Turkestan plains at Sarikia (which I identify with Sanju) halfway between Yarkand and Khotan. 2,nd. — The existence of an iron suspension bridge at Totling said to have been constructed by Alexander the Great {vide Major Montgomerie's Eeport on Trans-Himalayan explorations made during 1867). Zrd. — The statement made by Muhammed Amin, "Punjab Trade Eeport, Appendix IVA." that — "the old route taken by Moghul conquerors from Tashkend towards CMna passed through the Aksai Chin. Traces of it are still seen." E ( 18 ) matter^ once for all, as one which g-ave me much trouble and annoyance during the whole of my stay and travels in Eastern Turkestan.* It must not be supposed, however, that because I was given permission to use instruments I have been able to turn out very accurate surveys of the countries traversed. The rapidity with which we have always travelled has made it impossible for me to do more than carry on a con- tinuous route survey, checked by frequent astronomical observations taken at night ; and even this is sometimes meagre and incomplete owing to the intense cold which we experienced throughout almost the whole of our travels, which made even the handling of a prismatic compass at times an impossibility ; this, coupled with the shortness of the winter days, the occasional exccessive length of the marches, many of them through snow, and the necessity on these trips of always cutting down both the baggage and the limited establishment of camp servants with which I originally started, must be held to excuse any incompleteness in the maps that I furnish. During our first excursion the marches we made were as follows : — From Yangi-Hissar (Kashghar) to — 1. Besak (Upper Artysh District) 2. Chung Terek ... 3. Chakmak Forts 4. Balghun Basbi 6. TurgatBela ... 6. To Turgat Pass and hill above Chadyr Kul and back to (5) 7. Back to Chatmak Miles. 26 20 20 10 15 32 25 and back to Kashghar by the same road. I suceeded, with no little difficulty, in keeping up a continuous route survey, and took observations for latitude at four points on the line of march, the most northerly being at Turgat Bela (north lat. 40° 23' 53") on which occasion, while observ- ing, the thermometer stood at 10° below zero (Fahrenheit), and an intensely bitter wind was blowing. Later on the same night the thermometer fell to 26°, while inside the akoee, (Kirghiz tent) where we slept, it was as low as 8^°, a temperature hardly adapted for carrying on an elaborate Survey.f We left Yangi-shahr (the new city of Kashghar) and, going northwards, crossed the River Kizil by a good wt)oden bridge. At a distance of 5^ miles we passed on our left the old city of Kashghar, beyond which we crossed the River Taman by another bridge. This stream passes immediately to the north of the town, and joins the Kizil at a short distance to the east, the two forming the Kashghar Darya. At the time we passed there was but little water in either stream, that little being frozen, so that it was impossible to form any idea of the size of the vast mass of water that must come down in the summer time. The left bank of the Taman is covered by tanneries and cemeteries ; the road runs nearly north and enters a narrow lane between two mud walls, on either side of which are enclosed gardens, fields, and hovels. These continue for some four miles, when the road emerges on to an open stony plain forming a very gently rising slope up to a small spur from a low range of hills running nearly due east and west, through a gap in which, formed by the river Artysh, the road passes. On the north side of the range is the wide and fertile valley of the Artysh, a name given to * I may mention that in Kashghar I had been questioning a sepoy, who professed to know all about the Alai and adjacent country, on the subject of the supposed double issue from Lake Karakul. He positively assured me that the waters from it flowed west into the Oxus. A few days subsequently, when talking on the same subject, he assured me with equal confidence that he had seen the place, and tnat its waters flowed east to Kashghar. He subsequentlj' admitted that he had never been within 50 miles of the lake ! t It may be imagined that taking observations in the open, to stars, with the thermometer standing below Zero, is not a very pleasant occupation. After handling the instrument for a short time, sensation, so far as one's fingers are concerned, ceases, and during a set of observations it is necessary to rush frequently into the adjacent tent to restore circulation over a fire. The recorder, on such occasions nurses the hand lantern with great care, and although the ink is placed inside the lantern, yet it would freeze on the pen between the lantern and the paper. I was eventually obliged to allow a pencil to be used on such occasions. My faithful Madras servant " Francis " also experienced no little difficulty in getting the lamps to burn properly. The oil becomes very thick from the cold. The air holes had to be carefully enlarged for high altitudes, so that while admitting more air, they might still be small enough to prevent the high winds which were frequently blowing, from extinguishing the light. ( 19 ) the whole district^ which consists of several small townships scattered over the valley, in one of which, Besak, some five miles beyond where we crossed the river, we put up for the night. From Besak our road lay for a few miles over fields lying in the broad Artysh valley, but we soon entered that of the Toyanda River, which flows from the Turgat Pass. This stream divides into two branches at the place where it debouches into the Artysh plain — the upper one flows nearly due east, and is extensively used in irrigating the fertile valley ; the south or main branch flows into the River Artysh,"^ which passes along the south side of the valley, and after being joined by the Toyanda stream, cuts through the hills to the south at the gap alluded to in the preceding paragraph. On entering the Toyanda valley, here about two miles wide, we may be said to have fairly entered the Thien Shan mountains, the hills we had traversed on our previous day^s journey being an isolated ridge. In marching up this open valley we had in view on our left the sharp serrated edges of the Ming-yol Hill, a prominent object in the panoramic view from the roof of the Embassy buildings in Kashghar; in front of us lay a range of snow-covered peaks also visible from Kashghar ; these formed part of a small range running parallel to the main chain (east to west). We passed the old Chinese outpost of Teshek Tash, or Khitai Karawal, and a little beyond it the village of Tupa, (or Tapii) near which place through a large ravine on the left, is a road said to come from Kizil-boya, a fort near the head of the Kashngar River. A little further on through a broad open valley we reached the picturesque camping ground of Chung Terek, a Kirghiz village, where were a number of akoees pitched for our reception. From this place the scenery gets much bolder and the road passes between precipitious hills rising to a height of some 3,000 feet above the valley, through which a march of 20 miles brought us to the Chakmak Forts ; the road goes steadily up hill, a gentle and regular ascent which continues all the way up to the Turgat Pass, and is passable by laden camels even in " mid-winter.^' Eight miles short of Chakmak we came across the " Mirza Terek," " Past Kurghan," or '' lower fort,'' a carefully constructed work, which would prove a serious obstacle to an advancing foe. Here as is the case at Chakmak, the overhanging heights are so precipitous and inaccessible that it would be almost impossible for an enemy to effect a lodg- ment. The road across the Russian frontier by the Turgat Pass is good, and the slope easy. The road right up to the crest of the Pass was entirely free from snow.f On the slopes near the Pass is an almost inexhaustible supply of grass. There are two roads over this range of hills converging on a point a few miles north of the Chakmdk forts — one from the Suyok Pass, two days' journey in a north-west direction, is little more than a path, and cannot be traversed by horsemen ; but the other from the Turgat Pass, about 30 miles to the north of the junction (Suyok Karawal), is now the main caravan road between Kashghar and the Russian settlement of Alm^ ti (Fort Vernoye), and may be said to be practicable all the year round, although somewhat more difficult perhaps in summer, when there is much more water in the River Toyanda, which has to be crossed some forty times in the course of the journey. The Suyok Pass is stated on Russian authority to be 12,800 feet above sea level. A fort called Yagachak, covers some road in the direction of the Pass, west of Chakmak, but the accounts of its position were so vague and discordant, that I was unable to fix its position even approximately. A road along a ravine about half-way between Chakmak and the Past Kurghan was said to lead to it. Along a ravine lying to the south of the Chakmak forts a road runs across the hills, connecting them with the Terekty Fort, nearly due north of Kashghar. It lies on the shortest road between the Naryn Fort (Russian) and Kashghar via the Bogushta and the Terekty Passes. For 25 miles above Chakmak, the road took us along the course of the frozen stream, passing through volcanic rocks, to Turgat Bela, a little short of which the nature of the coun- try alters, and the precipitous hills are replaced by gently undulating gra-ssy slopes abounding * The Artysh River is said to rise near the Terek Diwan, on the road between Kashghar and Khokand. t In January. ( 20 ) with the " Ovis Poll" (Ovis Argali of the Russians) .* The weather was now intensely cold ; one of our party got his fingers frost-bitten from the cold contact of his rifle, and when I stopped for a few seconds on the top of a ridge to get a view of the country, and to record the reading of my aneroid, my hands and feet became entirely numbed. From Turgat Bela (at an elevation of 11,030 feet above the sea), we rode to the Chadyr Kul Lake, and back to camp the same evening (about 32 miles). Starting early in the morning with the thermometer several degrees below zero, we rode 13 miles to the Pass up a gentle ascent through the broad and open valley, until within a mile of the crest, where the slope though still very easy, is somewhat steeper, there being a rise of about 400 feet in the last mile. On the left of our road was a range of lofty, bold, precipitous peaks, running while near the pass from north-east to south-west, but subsequently in a more westerly direction. The height of these peaks varied from 13,000 to 15,000 feet. On our right were low undulat- ing hills extending away eastward as far as we could see. On reaching the pass (12,760 feet) we did not immediately see the lake, but had to advance for about three miles in a northerly direction, when we came suddenly into full view of the whole lake and the range of mountains beyond, a magnificent panorama. There are two nearly parallel ranges of mountains, the Turgat (sometimes called Koktaw — in Russian maps "Kashghar Daban^^) on which we stood, and the Tashrobat to the north, both portions of the Thien Shan range, which westward, like the Karakorum eastwards, seems to lose its identity and merges into several comparatively unimportant chains of which it is imposssble to say which is the main one. The Chadyr Kul lies between these two ridges, and, as far as one can learn from Russian sources, there is no drainage out of it, but several small streams run into it. Their maps include the lake within their boundary, which they place on the crest of the southern or Turgat range, the peaks and passes of which are of about the same average height as of the northern range. The Kashgharees (in Kashghar) claim the lake, and maintain that the Tashrobat range forms the true boundary, but their officials on the spot appeared to take a different view, and maintained that the lake was the boundary. The Ak-sai River, which rises a few miles east of the lake and between the Tashrobat and Kok-tan ranges, flows into East Turkestan, while the Arpa, which flows from a corresponding position near the west end, finds its way into the Syr Darya. This would indicate the lake itself as a good natural boundary, although it must be remembered that the Ak-sai plains to the east, the head waters of the Ak-sai River, which afterwards becomes the Kokshal, are undoubtedly occupied by Kirghiz subject to Russia. The lake is about fifteen hundred feet below the pass, which would give the former an elevation of 11,300 feet, a result agreeing very nearly with that arrived at by the Russians. From the undulating nature of the low hills to the east of the pass, it was impossible to judge of the direction of the range. Of course from a single view of the lake and the mountains beyond it, it was impossible to form any accurate idea as to their size, but according to the Russian maps the lake is of oblong shape, about 14 miles in length, and 5 or 6 in breadth at its widest part ; its greatest length being from west by south to east by north. From where we stood about three miles north of the pass, the east extremity of the lake bore a little to west of north, while the Tashrobat Pass as pointed out by our guides lay about 17° further to the west. The lake was covered with ice, and the sleet which lay on the surface made it difiicult to distinguish the edge of the lake from the nearly level plain by which it is surrounded, and which was covered with a white saline efflorescence. A single horseman near the edge was the only living object visible, a curious contrast to the other side of the pass, where within a few miles of the crest, we had seen a herd of several hundred Cossack ponies grazing at the foot of the pre- cipitous hills before alluded to. * These extensive grassy slopes, somewhat resembling the English downs, are a very curious feature of the country, and not only attract the Kirghiz as grazing grounds for their cattle, but are equally sought after by the large herds of Guljar, in one of which Dr. Stoliczka counted no less than eighty-five. ( 21 J The caravan road which we had followed from Kashghar lay across the plain in front of us. Beyond it is the Taishrobat Pass about the same height as the Turgat, but somewhat more difficult. A traveller who had crossed it in March told me that the road was then very bad, and difficult for equestrians, but I think his account must be somewhat exaggerated, as the camel caravans from Almati traverse it without much difficulty, and the Russians do not write of it as a difficult pass. Between the Naryn Fort and Kashghar, a distance of 180 miles, there are only these two passes — both about 13,000 feet in height. There is a third pass, the Ak Cheta, between the At-bashi River and the Naryn Fort on the Naryn River, but this is, I believe, sometimes avoided by following the Naryn to its junction with the At-bashi, and then proceed- ing up the latter river to Tashrobat.. When we visited the country early in January there was no snow on the ground, but we were singularly fortunate, for a traveller two months later in the year complained of a good deal of snow, while Baron Osten Sacken wrote on a former occasion that his party suffered much from cold and snow in July. A shorter and more direct road between Naryn and Kashghar is that over the Ak Cheta, the Bogushta, and Terekty Passes, stated by Captain Reinthal to be not more than 134 miles in length, or eight days' journey. The passes, though all about the same height, i.e., between 12,000 and 13,000 feet, are more difficult than on the ordinary caravan road via Tashrobat and Turgat Bela. The Bogashta Pass is sometimes closed in winter. It is covered on the Kashghar side by the Terekty Fort. We never had an opportunity of visiting this fort, which lies, as far as I could make out, nearly due east of Chakmak and due north of Kashghar, and although we must have passed within a few miles of it during a subsequent trip in the Artysh districts, my guides studiously avoided pointing it out, and actually, on one occasion even denied its existence. The distance between Fort Naryn and Vernoe (Almati), a military district centre, with large garrison and supplies, is 180 miles by the shortest road, which goes over three passes, all between 12,000 and 13,000 feet in height. I had hoped that we should have been able to return to Kashghar, over the undulating plateau to the east of the Turgat Pass, and by the Terekty Fort, but we had now to retrace our steps to Kashghar by the road we had come. A notice of the return journey is therefore unnecessary. Whilst our party under Colonel Gordon was visiting the Chakmak Forts, another member of the Embassy, Captain Biddulph, paid a visit to Maralhashi on the direct road to Aksu. A description of his journey will be found elsewhere. During my absence Kishen Sing Pundit, was despatched in company with Sirdar Tara Sing (Treasurer to the Mission), on a visit to Khanarik and Kizil-boia, large villages lying to the south-east of Kashghar. The Pundit carried on a traverse survey wherever he went, which has thrown some light on the intricate maze of rivers and canals which irrigate the villages that are thickly scattered over the whole of the ground visited by him. My second excursion was to the north-east of Kashghar. The Amir having granted per- mission for a visit to the Artysh districts, I was enabled to accompany Sir Douglas Forsyth and party during their stay there, and on their return to Kashghar, I made a rapid journey in company with the late Dr. Stoliczka towards Ush Turfan. Unfortunately on this trip, after leaving the head-quarters party the weather was much against us. Bitter cold was accompanied by snow and clouds, which combined to conceal the rocks and hills from both Dr. Stoliczka and myself, still, as the ground traversed is, as far as I am aware, entirely new to geographical science, a short account ought not to be uninteresting. We left Yangi-hissar (Kashgar) on the 14th February for Bu Miriam Khana, a village about 11 miles north-east of the old city of Kashghar. The first three miles of one road lay across a cultivated and well irrigated plain, and brought us to the banks of the Kizil or Kashghar River, at a place where it was easily f ordable ; after another four miles we reached the large village of Awat, near which large quantities of salt are collected and taken to the Kdshghar market. Four miles of level plain brought us to Bu Miriam, where we learned that our baggage animals had, by mistake, taken the road to Ostyn (Upper) Artysh, and that we had no chance of seeing them that day. We accordingly pushed on to Altyn (Lower) Artysh, where we found com- fortable quarters and a good dinner, provided by our host, the Hakim of the district. Both P ( 22 ) were welcome, as our own things did not come in till next morning", much to my special annoyance, as the chronometers had all run down in the night, a great misfortune, as I had been very anxious to determine a good travelling rate for them, and with that end in view, had taken very careful time observations before departure from Kashghar. About a mile from Bu Miriam, we crossed the small river coming from Ostyn Artysh, the upper part of whose course I have traced in an earlier portion of this narrative. It (or rather what small portion remains after irrigating the large and fertile village of Beshkerim, which we passed on our left) falls into the Kashghar river near Khush Toghrak, about 25 miles to the east of where we crossed the stream. Five miles further on the road traverses the same low range of hills which, south of Ostyn Artysh, is pierced by the Artysh stream. This ridge, composed of clay and shales, is several hundred feet in height at Ostyn Artysh, but gradually gets lower and lower as it runs eastward, until it dwindles into nothing, and gets lost in the level plain a very short distance to the east of where we now crossed it ; a few miles further on, after crossing a small stream supplied from springs on the west, we reached the village of Altyn Artysh, a march of 32 miles. This village partakes of much the same character as Ostyn Artysh, and, indeed, nearly all the villages I have seen in East Turkestan, consisting of a number of small hamlets, scattered about the plain, at intervals from each other varying from a quarter of a mile to a mile. Each hamlet consists of a number of scattered farm-houses, each farm having its separate irrigation canal, its trees, its fields, and out-houses, and forming the residence of a family containing generally from four to a dozen souls. In a central position is the bazar, with long rows of stalls on both sides of the road, somewhat resembling that of an Indian village, but absolutely untenanted except on the weekly market day. In its neighbourhood the Hakim, who generally owns a somewhat better house than his neighbours, administers justice. Sepoys, if the village be important enough to contain any, are generally quartered near the residence of the Hakin^. The valley in which the two Artyshes are situated runs from west to east, and is through- out about eight miles in breadth ; bounded on the south by the low ridge of hills before mentioned which comes to an end south-east of Lower Artysh, it is confined on the north by another and somewhat higher range, which extends eastwards from Teshek Tash on the Chakmak road, to nearly opposite the termination of the southern range, when bending towards the north-east it runs away towards Kalti Ailak, another large group of villages about %% miles east by north of Altyn Artysh. The valley opens, where loosened from its bounds on the south, into the large desert plain which forms part of the one vast plateau of Eastern Turkestan. The one difficulty, in all this country, is want of water, and one cannot help admiring the ingenuity with which the inhabitants have made the best use of the scanty supply of this precious fluid. Where there is a sufficiency the country is one close net-work of irrigar- tion channels, and in the spring, in these places, one unbroken mass of tress and verdure testifies to the excellence of the system. In the Artysh valley there is water in moderation, and, as far as I could learn, nearly every drop, in the spring and summer, is used in irrigation. In the winter, one sometimes comes across tracts of marshy land, but these are generally caused either by springs which rise in the neighbourhood, or by leakage from canals in autumn, at which time the water is no longer required for irrigation, and the saline nature of the soil causes breaks down and consequent leakage, which it is not considered worth while to repair until the following spring. In Altyn Artysh, I was informed that there were in all about 3,000 houses forming the following hamlets : — 1, Meshak ; 2, Sborchi \ 8, Takyun ; 4, Langar ; 5, Kichingiz ; 6, Mai or Tater ; 7, Kij ja ; 8, Bayamat ; and 9, Kukila. It is more thickly inhabited than other parts of the surrounding country, as it forms the seat of the District Government (which includes under it Kalti Ailak and other villages) . It is well watered, but the population being large, it barely produces grain sufficient for its own consumption. This deficiency is, however, made up from the neighbouring village of Kalti Ailak, where there is plenty of good rich soil, and a smaller proportionate population, due to an occasional want of water the supply of which is often insufficient to irrigate the whole of the lan'ds. The headman of Kalti Ailak bitterly complained to me, that where there was plenty of w^ter gpod land was deficient, ( 23 ) and where little water was met it was often the reverse. This village contains about 1,000 houses, divided into the following petty districts : — Kurghan (the chief centre), Golok, Khush Toghrak, Kuyok, and Jainak. There are two streams which enter the Artysh valley, the Toyanda before described, and the Bogoz River; a branch of the former irrigates the villages of Beshkerim and Bu Miriam^ where the greater part of the water is absorbed ; a small remnant however flows eastward, and in favorable seasons finds its way to Khush Toghrak, a southern hamlet of Kalti Ailak, where it mingles with the canals from the Kashghar liiver, employed to irrigate that village. The north branch of the Toyanda River is probably almost all expended in irrigating the fields of Upper Artysh, but it is possible that a small quantity may find its way down to Lower Artysh, or at all events may help to form the supply for certain springs which issue from the ground west of that village. The main water-supply, however, for the latter village is derived from the Bogoz River, which rises in the Chakmak range of hills, about 30 miles to the north, but derives a large portion of its water from hot springs a few miles north of the village. On the 17th of February we started for Tangitar, making a march of about 20 miles in a northerly direction. Following the banks of the River Bogoz, a narrow but somewhat rapid stream, easily fordable, we reached after three miles the range of hills forming the north boundary of the valley. Here on a small isolated mound stand the ruins of an old Chinese fort J a mile beyond this the stream divides, the left (west) branch is the main one and comes from the snows ; the temperature of its water was 42°, while that of the right hand one coming from the warm springs before mentioned was 57°. Our road followed the east branch ; a path along the other goes to Chung Terek in the Toyanda valley, distant about 33 miles. Continuing our road along a ravine passing through the range of hills (which here have a breadth of about three miles from north to south), we at last emerge on to another extensive plain extending like that of Artysh from west to east, and about six miles in breadth from north to south. On the west it was bounded by the hills above Chung Terek, and extended along eastward, as far as one could see, for several miles, merging into the open plain, where the ridge to the south comes to an end. On our right, " near where we entered the valley, is the village of Argu, said to contain 800 houses, but from its appearance I should not have judged it to hold half that number. Its water-supply is derived partly from springs, partly by irrigation from the Bogoz River. The road crosses in a north-west direction over a perfectly bare, stony plain, which continues away on the left as far as the eye can reach ; one or two houses only near the bank of the river break the monotony and barrenness of the land- scape, neither grass nor wood being elsewhere visible. After six miles we enter a gap through which the Bogoz River issues from another range of hills, also running from west to east. Here is another Chinese Karawul"^ in good preservation. The road passes to the north along the Bogoz valley through the hills for about nine miles, to Tangitar, 5,800 feet above sea level. The valley was in places of considerable width, and contained much wood and grass, as a natural consequence of which numerous Kirghiz encampments were scattered over it. We passed successively those of Buabi, Bulak, Kuktam, and Jai Ergiz. On our right were some very precipitous hills, forming the ends of spurs running generally from west-north-west to east-, south-east. Our camp at Tangitar, after a march of 20 miles, was at the entrance of a defile, where two small fortifications are perched up on rocks commanding the south entrance. If larger, they might possibly be of considerable use for purposes of defence, but as they cannot hold a garrison of more than 20 men, they could only be useful in keeping in check badly armed Kirghiz or bands of robbers. They are built on the limestone formation which here com- mences, the hills through which we had hitherto been marching having been composed of clay and gravel. The situation of such forts, both here and in other parts of the country, are, I think, con- vincing proofs of the fact that the Chinese in their dealings vsdth the Kirghiz and other robber tribes, nearly always acted on the defensive, and did not attempt to hold the hilly tracts, or claim sovereignty over them. They apparently used to content themselves with posting strong * Karawul is a Turki word signifying " outpost." ( 24. ) guards on their frontier inside the lines of hills^ which they appear generally to have given over entirely to the wandering tribes. The same facts apply to the hills on the west of the great Turkestan plain, where the line of fortified posts along their base was considered the boundary. This forms a striking contrast to the policy of the present Ruler, who keeps all these tribes in subjection, has disarmed them, and has replaced the former anarchy by peace and quiet. On the 19th February we continued our way up the stream, here called Tangitar,"*^ through a very narrow defile somewhat difficult to traverse on account of its being filled with ice. After marching a mile or so along this due north, the valley opens, and through a stony ravine on the right comes the main stream from the north-east, while opposite to it, on the left, is an open ravine along which a road is said to go to the Terekty Fort, which / believe lay about 10 miles off in a north-west direction. Our own road continued due north for a while, when it edged round to the east over a spur, on rounding which we discovered that we were on the borders of another large open valley, the third we had entered since leaving Kashghar. The view from this spur was very fine ; in front of us lay a vast open valley bounded on the north by the snow covered Chakmak range of hills, which, visible about 40 miles off on our left, above the forts of the same name ran in a bold irregular outline from west-south-west to east-north-east, the crest of the range passing about 16 miles to our north, and running away eastward as far as we could see, apparently getting lower and lower as it did so. The broad grass covered valley before us was about 6,000 feet above the sea, and ran parallel to the crest of the hills. Along the middle of it is a low broken ridge running in the same direction dividing the valley into two parts. Small, bare, bold isolated hills also dotted the plain, which was nearly level, draining slightly towards the south. "We halted, after a short march of only 10 miles, at Tughamati, a camping ground situated in the plain/ along which we continued the following day in a direction east by north for about 15 miles to another Kirghiz camping ground, called Bash Sogon (head of the Sogon) . The road was so level that it was almost impossible to say where we crossed the water-shed which divides the Bogoz basin from that of the Sogon River. The latter has, at this time of the year, its chief source in springs near our camp, but as well as the Bogoz, it must in the hot season get a good supply of water from the snowy range to the north. From a high hill to the south of our camp, I obtained a fine view of the low ranges to the south, bat to the north I could not see over the snowy range. The general run of the hills to the south, was from west by south to east by north. It was formed by a succession of nearly parallel ridges starting abruptly from the plain in front, and dying out gradually as they approached the east. A few miles north-west of Bash Sogon, is a largish village called Arkala, near which large numbers of ponies, sheep, and cattle were grazing. We also saw signs of cultivation, which is occasionally carried out in years when there is sufficient water-supply from the melting snow. Throughout the plain there is a good deal of grass and low jungle, and near the camp I saw some small deer (kik), whilst others of the party had good sport in hawking partridges and hares. The nights we spent on this plain were very cold ; at Tughamati the thermometer outside the akoee fell 20° below zero and stood at 16° below zero when I rose in the morning. This great cold was, I think, in great measure attributable to the presence of saline matter in the soil, for our elevation was not much over 1,200 feet above that of Kashghar, where the corresponding minimum was very much higher. The drainage of the east portion of this large valley runs into the Sogon River, but the supply of water from the hills is apparently very small, owing I presume to the very moderate snow fall. The river, after it emerges into the plains north of Kalti Ailak, wastes away and leaks through crevices in the stony ground, and the Hakim of the latter place assured me that wells had been sunk, but had proved to be of no use, so that the whole of the water from the Sogon runs to waste, if at least we except the small quantity used by the Kirghiz higher up. This diminution in the size of rivers as they descend, is one of the chief characteristics of the country, and occurs in all minor streams that have come under my notice. Of course much of this is due to irrigation, which necessarily carries off large quantities of water, but the stony soil has also much to answer for ; on the other hand the frequent appearance of large springs, giving * Tangitar signifies " narrow defile." ( 25 ) considerable supplies of water, and often issuing- from the open plains at long distances from the mountains, may account in great measure, if not fully, for the water thus lost in its early infancy. On the 20th, our march lay in a south-east direction, following the circuitous course of the Sogon through some low hills, for about 15 miles, to Ayok Sogon (foot of the Sogon), a Kirghiz encampment situate at the east of another small plain, covered with grass and jungle and the abode of numerous Kirghiz. This camp is near the direct road from Kashghar to Ush Turfan, and it was here arranged that Dr. Stoliczka and myself should leave the main party, and push on in the direction of tJsh Turfan ; as far as the limited time and commissariat at our disposal would permit. It was stipulated however that we were not to go beyond the limits of the Artysh district. The marches from Kashghar to tJsh are as follows : — K4shgar Altyn Artysh Kalti Ailak Kyr Bulak Jai Tiipa Ui Bulak Tigarek Akchi Kuyok Tokai Safr Bai Karawul Ush Tiii-faa Total Miles. ..." 22 ... 22 ... 33 ... 20 ... 27 ... 17 ... 19 ... 22 ... 22 ... 22 ... 16 ... 242 Cross the Belowti Pass between Tigarek and Akchi. Eoad from Safr Bai to Bedul Pass across the head of the Naiyn Eiver to the Zauku Pass, and thence by Karakul to Issighkul. From Jai Tupa to tJsh there is said to be an alternative road- Jai Tupa. Pichan. Piklik, over Pass to Guljar Bashi. Kashghar Tokai. Kotan Serik. TJsh Turfan This road is somewhat longer than the other, and strikes the Kokshal Eiver a few miles east of Akchi. Leaving Ayok Sogon after an early breakfast on the 21st, we passed for a mile over the plain in a south-east direction, and struck the main road; then, turning east went up a ravine, through some hills across a low pass (5,670 feet), and found ourselves on the western edge of another of these large characteristic level plains, 15 miles across from north to south, where we entered it, and extending away eastward further than we could see. It was bounded on the north by our old acquaintance, the Chakmak range, and is probably a continuation of the Tughamati valley, which apparently narrowed considerably to the east of our camp at Bash Sogon. The main range was here following a more northerly direction than when we had last seen it, but the peaks were involved in snow and clouds, from which they never emerged during the whole of our trip. On the north, at a distance of about 15 miles, was the Kirghiz village of Karghil, the only habitation visible. Shortly after entering the plain we passed through what proved to be the commencement of a very large forest, composed almost exclusively of poplar trees (toghrak), and a small shrub called " balghun." The poplars were stunted in growth, and although evi- dently in a natural state, they bore the appearance of having . been pollarded. As timber I should not think the wood would be of much value, but it would furnish Kashghar with a plentiful supply of firewood, when the more convenient stocks in its neighbourhood have been exhausted. Passing along in a north-east direction, a low range of hills at a distance of about three miles bounded the plain on the south. At about 12 miles from our last camp, still traversing forest, we passed on our left the camping ground of Kyr Bulak — inhabited in the summer by Kirghiz, but now untenanted — onwards we pushed our way over a most monotonous flat, and through the bare bleak stems of trees, until after six miles, we came to a slightly rising ground G ( 26 ) called Btmg Jigda Buldk,* where a little water was oozing from the ground indicating a spring which, with the presence of a " jigda " or wild olive tree, gave the place its name. On over the plain, which in summer would have been pretty enough, but now was dried up and desolate. The forest ceased within about two miles of our camp at Jai Tupa, which was marked by a clump of trees standing conspicuously on an eminence above the plain. We reached it about dusk, after a march of fully 32 miles, through a very heavy sandy road which so delayed the mules carrying our baggage, that they did not arrive till eight o'clock the next morning, having stopped over night, exhausted in the jungle, about five miles short of our camp. Fortunately, we found an old Kirghiz Musjid, in which we went dinnerless to bed, protected, however, from the wind, and from the snow which fell during the night. The officials at the head-quarters' camp had assured us that we should find Kirghiz and supplies at this place, but there were neither one nor the other, and the Diwan Begi,t who accompanied us, spent his whole ni^ht (after his day's ride) in going over to the village of Karghil beforementioned, and hunting up Kirghiz, with whom he returned about daybreak, bringing supplies for man and beast, both of whom had fasted for at least 24 hours. It snowed all the morning, but about noon we pushed on about five miles in a north-east direction to a Kirghiz camp called Jigda, where we obtained further supplies. Snow and clouds prevented our seeing any of the hills around. The forest recommenced about half way between Jai Tiipa and Jigda. The following day (23rd) we pushed on for 22 miles to Ui Bulak, having obtained from the Kirghiz two or three camel loads of grain and other supplies for our future consumption, as we were told we should not come across any more habitations. Just before starting we felt a slight shock of an earthquake, the only one I have noticed during our stay in Turkestan. Our general direction was now north-east. About two inches of snow lay on the ground, and more was constantly falling. After five miles, we saw a low ridge on our right, running parallel to the road, at a distance of about six miles. At its base in what was apparently the lowest part of the valley, was a strip of forest, a portion of the large one that extends right away to beyond Kjt Bulak, a distance of at least 32 miles. Although long, this forest is comparatively narrow, varying, as far as I could judge, from half mile to two or three miles in breadth. The southerly ridge beforementioned is said to extend easterly to Kalpin (about 15 tash) X a village between tJsh Turfan and Maralbashi, and to be about five tash from the latter place. At Kalpin, hke other ranges that I have traced, it gets lost in the level plain. Our road now lay through low jungle (balgun) with little or no grass, and at about 12 miles from camp, we reached the limit of the plain and ascended a low spur running from the main range ; ground bare and stony. Following this spur in a north-east direction we crossed into an open ravine, about half a mile broad. Ascending it for a short distance we arrived at our camp, near which there was a good deal of grass and plenty of fire-wood. Thermometer at night down to zero. The next day was fortunately very fine, for we had a hard though interesting journey before us. Leaving our servants and baggage ponies behind at Ui Bulak, Dr. Stoliczka and myself continued our journey in a north-east direction, ascending the ravine, for about eight miles the road way very stony, and some inches deep in snow. Near the head of the ravine we crossed a low pass on a spur from the main range. Descending on the other side we crossed the lower slopes of the main range, passing along which for two or three miles, we came upon another large plain about six miles broad lying between two long spurs. On the further side of this plain, at Tigarek, we had the good luck to come unexpectedly across a Kirghiz encampment, belonging to tJsh Turfan. Leaving the " Dah-bashi,"§ who accompanied us as escort, to make preparations for our dinner, we procured a Kirghiz guide and started to try and reach before dark the Belowti Pass, which is on the main range that separates the drainage of the ground we had been traversing from that of the Aksai or Kokshal river, which, rising east of Chadyr-kul, flows nearly due east to tJsh Turfan and Aksu. A march of nine * In Turki Suldk is the equivalent for spring. t The designation of the official who was deputed to make arrangements for our party. + A tash here is taken at five miles, but in many parts it scarcely exceeds four miles. See note to Route XII of Section G. of Geographical Appendix. — T. D. F. § Or " Comniapder of ten (soldiers)." ( 27 ) miles, i,e., three miles in a north-east direction across several low projections of a spur running south, then four miles of steady ascent up a ravine to the north, followed by a sharp pull of two miles in a direction 35° east of north, brought us to the Belowti Pass, the goal for which we had been striving. From the pass itself, which is about 11,500 feet above the sea, no view was to be had, but by ascending a hill to the west, some 300 feet above it, I got a very fine view of a portion of the snowy range on the opposite side of the Kokshal river ; one peak, nearly due north, stood out conspicuously, of no very great height however, its elevation being only 2|° from where I stood. The range appeared to run nearly due east. Parallel to it at its base lay the deep valley of the Kokshal, apparently about eight miles to the north of where I stood. The road from the pass leads down a steep ravine, at first nearly north-east, and then with a north-westerly course to the river. The position of the next camp, Ak-chai, on the big river, was pointed out to me, bearing 10° east of north, but the man who was with me could not give me a good idea of the direction of IJsh Turfan or Aksu. Ak-chi, the first halting-place to the north of the pass, is a Kirghiz camp close to the point where the road from the pass strikes the river. About five miles below it is Kokshal, a large Kirghiz village, between which and tJsh Turfan, a distance of about 90 miles as far as I could learn, are numerous Kirghiz encampments, all under the orders of the Hakim of tJsh Turfan. It was a party of these Kirghiz whom we had had the good fortune to encounter on the south of the Belowti Pass."^ Near Kokshal, the alternative road from Jai Tupa, before alluded to, joins the river which takes its name from the village. The road is said to be shorter and easier than the one we had followed, but for two days there is no fire-wood. One march above Kokshal (or three marches according to another account) is the fort of Kara Bulak, above which the Kirghiz subjects of the Amir are not pllowed to pass, the ground above being held by the Russian Kirghiz, who in their turn are not allowed to cross the frontier eastward. All these Kirghiz are, I believe, of the same tribe, but being under different rulers are to a certain extent hostile ; at all events they are not allowed to communicate with each other. From Safr Bai, about 38 miles to the west of iJsh, is a road leading to Issigh-kul, by the Bedal and Zaiki Passes. The former of these is on the boundary between Russia and Kashgharia. There are said to be about 500 Kirghiz families in the Kokshal valley, and about 350 in the valleys north and north-east of Artysh. The Kokshal valley is exceedingly rich in pasture. Its upper waters (the Aksai) were first occupied by the Russians about 10 years ago. On the range on which I stood there were no high peaks visible, probably none more than 1,000 feet above the pass; the ground on both sides was undulating and grassy, very much resembling that to the east of the Turgat Bela Pass in the same range. It was evident that this range had, as it advanced eastward, become considerably lower, both with regard to its peaks and its water-shed. Like the smaller ranges at its base and parallel to it, I believe it to get lower still, as it goes further east, and at last to be lost in the plains near Aksu. The sun set while I was at the top of the pass ; the thermometer stood at 5° F, with a cold wind blowing, so I was glad to go back to the Kirghiz camp at Tigarek, which we reached at 9 p.m., having made good use of the only fine day we had during our trip. As it was, snow began to fall immediately after we reached the camp. The next day we returned to our standing camp at Ui Bulak j the road was three inches deep in snow, and more falling, accompanied by a bitterly cold wind; next day back to Jigda, 23 miles; weather much the same; next day 25 miles to Kyr Bulak, to which place akoees and supplies had been brought for us from the village of Karghil, 15 miles off. The following day we marched 29 miles to the village of Kalti Ailak. The ground we had been traversing is marked on our maps " the Syrt,'' and is represented as a high table-land. I took some pains to ascertain the limits of the district bearing this name, but could not arrive at very satisfactory conclusions. " Syrt " in Turki means " the back,^' and is therefore necessarily applied to somewhat elevated lands. The Hakim (Governor) of Artysh included under this designation the whole of the highland districts about Sonkul and • From them we obtained both food and shelter. ( 28 ) Chadyrkul ; another authority referred the name more particularly to the plains at the head of the Aksai River. The Kirghiz living- in the districts we had passed through seemed to be entirely ignorant of the name, and did not recognize it ; but after my return, on asking the Hakim of Kalti Ailak the whereabouts of the Syrt, he immediately replied that I had just come from it, and that the name was applicable to the whole district between Artysh and Ush Turf an ; on his evidence I think the name may remain on our maps where it is. The country can, however, by no means be considered as a high table-land rising immediately above the plains of Turkestan ; it should rather be represented as a series of parallel mountain ranges, running, as a rule, from west to east, each one decreasing gradually in height, from the main ridge on the north to the lowest on the south ; each subsidiary range also decreasing in height as it goes eastward. Between these ranges and running parallel to them, are extensive level plains, very little higher than the plateau of Eastern Turkestan, but gradually rising towards the north and sloping down towards the east. Thus the Tughamati Plain, about 45 miles north of Kashghar, is about 2,000 feet higher; while the Jai Tiipa plain, the same distance east of Tughamati, is only 1,000 feet higher than Kashghar. The combined effect is to give a general slope to the south-east. These large plains have in most cases much grass and fuel, though but httle water. From the Sogon eastward we came across no flowing stream. What water is derived from the very moderate annual snowfall seems to percolate into the earth, moistening it generally, and issuing in various places in the form of springs, near which are usually to be found Kirghiz encampments. In the Tigarek plain, at the foot of the Belowti Pass, there are, I believe, no springs ; and although there is good grass, the only time of the year in which the plain can be tenanted by the Kirghiz, is that at which we happened to visit it, the sole substitute for water for them- selves and flocks being the actual snow, which was then lying on the ground. In the Jai Tiipa valley there are water-courses running from north and east, but the supply of water is so precarious that the Kirghiz told me that it was only after years in which there was a more than average snowfall, that they attempted any cultivation at all, and under the most favor- able circumstances the extent is extremely limited. There appeared to be no outlet through the hills surrounding this valley, in the lower portion of which lies the forest before alluded to. The moisture in the soil would seem to be sufficient to nourish these stunted trees. Much of the ground in the plain is covered with saline efflorescence, and from near Jai Tupa itself large quantities of crystallized salt are collected and despatched to Kashghar. The Kirghiz who inhabited the country in the time of the Chinese appear to have led a more jovial life than at present. Under no master^ they used regularly to levy black-mail from passing travellers and merchants at every camping ground; and as prompt payment always ensured a safe passage, there was seldom much difficulty in collecting their dues. Under the strict rule of the Amir they are now disarmed, and are comparatively poor, as they dare not venture on any of their old tricks. A single sepoy of the King's, selected from among themselves, is stationed in each encampment, and is responsible for the good conduct of its members ; an annual present of a choga, a certain amount of grain, and remission of taxes is the remuneration he receives from the State. The Kirghiz pay as taxes annually one sheep in 40, one sheep for every two camels, and one-tenth of the agricultural produce (when there is any). In these parts horses or ponies are scarce. Nature aids the inhabitants in their poverty by a plentiful supply of a plant called locally kuruk or teric, a kind of millet which grows wild and from which they make a preparation called " talkan" corresponding to the Ladakhi suUoo, which they eat uncooked mofstened with a little water. I tried some, and found it to be not unlike Scotch oatmeal, and, as it may be had for the picking it may be looked upon as a bountiful gift of Providence to these otherwise poverty stricken people. Our march from Kyr Bulak to Kalti Ailak was for a great part of the way down the Sogon River. A karawul, garrisoned by a few sepoys is situated where the river enters the hills south of the Ayok Sogon plain. The valley occasionally widens out into small grassy flats. After a time, the river is left (it goes off in a south-east direction and is, as before explained, soon swallowed up by the thirsty gravelly soil) and the road traverses some very bleak and desolate broken ground without a scrap of vegetation or sign of life. After passing through these hills and then over a few miles of flat stony desert we reached Kalti Ailak. ( 29 ) We spent a night there in the residence of the Hdkim and went the following day to Khush Tograk, its southern township, about eight miles to the south-east. After arrival there we pushed on two or three miles to the Kashghar River, which we tried to cross in order to shoot in some jungle at the other side ; but the ice was now breaking up and was so dangerous that our conductors would not venture to take us over, although the head-quarters party had crossed over a few days before on the ice without the slightest difficulty. Next day we continued our return journey, and forded the river several miles higher than where we had attempted its pas- sage the day before. It took us nearly an hour to cross the river, the combined water, ice, and mud making the passage so difficult that our Turkestani attendants had to strip off their four or five superfluous suits of clothing and go to the assistance of our baggage animals, who, after a good deal of plunging and floundering, at last got across without accident. Between Khush. Tograk and the river there are, at this time of the year, extensive swamps, caused in the manner I have before described. Near this place the waters of the Artysh and Kashghar River mingle together ; but in the hot and irrigating season the whole of the water from the Artysh River is said to be expended before reaching the junction, and the Khush Tograk village is then exclusively watered from the Kashghar River. In the early winter when the ice first begins to form, it partially blocks up the streams and the mass of ice growing larger and larger, great frozen lakes are often formed where in summer there is merely a rapid stream of water. This makes it impossible in winter (the season of our travels) to form any accurate idea of the real size of the streams. After passing the river and crossing a few miles of salt waste we came to Faizabad, a large village on the road between Maralbashi and Aksu. On this march I saw, for the first and only time in Turkestan, large numbers of geese and duck, all flying eastward. It was market day in Faizabad, and the crowd attending the bazar was about as large and dense as that I had previously seen at Altyn Artysh, from which circumstance I should infer that the population of the surrounding district is about the same. On the following day (3rd of March) we returned to Kashghar, a march of 37 miles over a perfectly flat country, the road winding almost the whole way through a populous and well cultivated district. We passed successively the scattered villages of Sheaptal, Sang, and Yanduma, every village as usual composed of several hamlets, each with its separate name. At Sheaptal it was market day, but it was too early in the day to be able to form any estimate of the poj^lation attending, though on the march we met crowds of people thronging to it. The road crosses several large canals which leave the southern branch of the Kashghar River several miles above Kashghar and irrigate the whole of the ground south, south-east, and east of the city. One of them bears the name of Yamunyar, and possibly a portion of its waters comes from the river of the same name, which, rising in the Little Karakul Lake flows past Opal and Tashbalig, where it divides into numerous branches and canals, some of which probably intermingle their waters with canals from the Kashghar river. The two together form a net-work of rivers and canals which it is nearly impossible to unravel, and which is moreover constantly changing almost from day to day. From the time of leaving the head-quarters camp at Bash Sogon, the weather was most unfavourable : snow and clouds prevented my seeing the hill tops by day or the stars by night. This lasted until our return to Ui Bulak, 26 miles west of the Belowti Pass. At Ui Bulak, Faizabad and intermediate stations I was more fortunate and was able to secure good star observations, both for latitude and time. These, combined with a rough compass survey which I made of the whole road, have enabled me to map it with a fair amount of accuracy, although the distance traversed during our absence from Kashghar, viz., 340 miles, was accomplished in little more than a fortnight. Excursion to the Fdmir Steppes and WaJchdn. Shortly after our return to Kashghar from the Artysh Districts a party, under the orders of Colonel Gordon, consisting of Captain Biddulph, the late Dr. Stoliczka, and myself, was sent via Sarikol (Tashkurghan) to Wakhan, and I was instructed to take what advantage ( 30 ) I could of siTch opportunities as might offer for the increase of our geographical knowledge. The primary object with which the Mission was despatched necessitated a very rapid out- ward march, and the difficulty of arranging about supplies compelled our return with nearly equal rapidity, giving no time or opportunity for making detours or excursions off the road : with the exception of halts at Panjah in Wakhan, the furthest point westward reached by the Mission, and at Tashkurghan, where we were compelled to halt for the purpose of resting our cattle, and one day at Ak-tash for the same purpose, our journey was merely a rapid conti- nuous march from beginning to end. I am induced to make these remarks at the outset, as I have seen paragraphs in the newspapers, and notably in the telegraphic reports of the London I'w/z^if to the effect that "the Pamir has been completely surveyed/^ and other similar statements which are apt to mislead the public and induce them to expect a great deal more than has been, or could possibly have been, accomplished under the circumstances. What I have been able to perform in the way of actual survey chiefly consists of fairly complete sets of astronomical observations, which have enabled me to fix with considerable accuracy the positions of the more important places along our line of march. These places have all been connected by a route survey, executed as carefully as circumstances would permit. I also succeeded in getting good observations with boiling point thermometers and aneroids on all the passes and at all our camps, which, combined with simultaneous barometric readings at Leh should furnish very trustworthy determinations of height. Observations for magnetic dip and declination were made .at Sarikol, and for declination only at Panjah. Owing to the necessity for cutting down baggage, servants, and camp followers to the lowest possible limit, I left both my survey khlassies behind in Yarkand, as also all photographic apparatus. From Yangi-Hissar as far as Tashkurghan I had the advantage of the Pundit's assistance, and he with the Munshi paced the whole road up to that point. The Pundit being a Hindu was not taken beyond Sirilcol, but Colonel Gordon obtained permission (from Hussan Shah, the Governor of Sirikol,) for him to return to Yarkand via Chehil Gombaz and the Charling River. From Panjah the Munshi was despatched on a special exploration, to be hereafter described. Up to that point I had the advantage of his services as a recorder for astronomical work. On the return journey to India the late Dr. Stoliczka kindly took his place, and recorded for me on seveifcl occasions, the last being only a few days before his death. Before going into the details of my own reconnaissance I may, perhaps with advantage, notice the mistaken ideas ^ which most geographers have held, at all events until very recently, of the nature of the mass of mountains and high table-lands which separate the provinces of Eastern and Western Turkestan. The labours of the Russian Venuikof, who taking the writings of the illustrious Humboldt for his basis, and working on to them the cleverly constructed but mischievous forgeries of Klaproth, have thrown back the geography of this region into almost inextricable confusion, from which even the recondite researches of Colonel Yule and Sir Henry Rawlinson have hardly yet rescued us. The vague statements of ancient travellers such as Huen Tsan and Marco Polo, who scarcely imagined when they penned their writings, the keen interest with which they would centuries later be studied and criticized, have added to the difficulties of forming a clear and correct idea of the country. The ideas I had myself formed before my visit were vague in the extreme, but perhaps not very much more so than those of others who knew a great deal more about it. Such different descrip- tions as the following are difficult to reconcile : — " The Pamir plain extends 1,000 lif from east to west and 100 li from south to north."— Huen Tsdn. " The centre of the plateau is " Saryk-kul" out of which there should issue, according to all accounts, the Jaxartes, Oxus, and a branch of the Indus. This plateau, which affords excel- lent pasturage, extends round the lake for a distance of six days' journey in circumference, and it is said that from this elevation all the adjacent hills appear below the observer." — ^urnes. * Derived from incomplete and discordant information. t A Zi is about one-fifth of a mile. ( 31 ) " For twelve days the course is along this elevated plain, which is called Pamir." — Marco Polo. " The hills and mountains that encircle Lake Sirikol * give rise to some of the principal rivers in Asia. From a ridge at its east end flows a branch of the Yarkand river, one of the largest streams that water China, while from its low hills on the northern side rises the Sirr or river of Khokand, and from the snowy chain opposite both forks of the Oxus as well as a branch of the River Kumir are supplied.^'' — Wood. In the last extract, I see how an excellent, careful, -and reliable observer like Wood falls into error directly he trusts to what he hears ; and I must say that from my own experience I have little confidence in geographical information extracted from the inhabitants of Central Asia, unless from trained and educated men who are accustomed to take notes of what they see. I feel it incumbent on me therefore to discriminate clearly between what I have seen, and what I have heard, and with that end in view I propose first briefly describing that portion of the country which has actually come under my notice. I may then perhaps hazard a few remarks on what I have heard. We started from Kashghar on our journey to Panjah on the 17th March, reaching Yangi- Hissar (36 miles to the south) on the following day. Halting a couple of days to make prepara- tions for our journey, we left on the 21st, starting by the same route by which the " Mirza" went to Kashghar from Cabul in 186,8-69. Our first march was to Ighiz-yar, eighteen miles; crossing the low broken sand hills that ran down to Yangi-Hissar from the mountains on the west, we followed for about thi*ee miles the direct road to Yarkand ; then crossing the Yangi- Hissdr stream at Karabash village we passed in a southerly direction for five miles, over a flat salt waste, to the large but scattered village of Sugat, which it took us nearly half an hour to traverse. On the sandy tract a number of people were digging and collecting a very inferior kind of fire-wood, which is carried off on donkeys to Yangi-Hissar, where fuel is very scarce and dear. From Sugat none of the large mountains on the west were visible on account of the haze ; the ridges before mentioned, and another low sandy ridge running parallel to, and on the other side of the Yarkand road, were all that could be seen. Sugat is situated on a slope, and our road lay up the bed of a water-course, one of many coming from the Kinkol stream, whose banks we were about to follow for several days' journey. When we passed there was but little water in the stream, and what little came was eagerly swallowed up by the thirsty soil. The villagers, on the approach of spring, were commencing to plough their lands. As the summer advances the heat increases, and with it the water-supply from the melting snow, which comes at the time when most wanted by the husbandmen for their early crops. I was informed that in summer a very small quantity of water trickles from the irrigation canals through a large ravine (which, surrounded by much broken groimd, is formed at the foot of the Sugat village) and joins the Yangi-Hissar river. The chief source of supply of the river is from a number of springs situated about six miles west by south of the town from which it derives its name. I visited them the day we halted at Yangi-Hissar on our journey to Kashghar ; they are surrounded by several villages, Kargoi, Konaf and Yangi Salip, and others. At Kona Salip the bed of the river is dry, and is formed by numerous short ravines meeting there. A little lower down, at Kargoi, the banks are fifty feet deep, and a considerable quantity of water bubbles out of the ground ; fresh springs issue for a considerable way down the river, so that by the time it passes south of Yangi-Hissar there is a considerable body of water in the stream, whose bulk is also augmented by drainage from canals supplied from the River Kusau which is said to rise in the Kizil Art mountains in the neighbourhood of Tagharma Peak. The temperature of the canals from this stream was 42°, while that of the river (from the springs) was 47°. The Yangi-Hissar river, after receiving accession to its waters from the Kusan, flows eastward, and is said to lose itself in the desert near the villages of Keltarim and Chakar. The river shortly after its issue from the mountains divides into four artificial branches or canals, the Pasin, the Parach (or Kusan river), the Sailik, * Wood's " Victoria Lake." t In Turki Kona means old, and Yangi means new. ( 32 ) and the Tibiz^ the last and southern of these it is whose waters sui'round the head of, and afterwards mingle with, the river of Yang-i-Hissar. There is also no doubt that the waters of the northern branches mingle with, if indeed they are not identical with, the Yamunyar river (also called Tasgun, Khanarik, Oi Kubok, and a host of other names) . The latter river issues from or near the smaller of the two Lakes Karakul (of which more hereafter) and passes from the hills near the villages of Tash-balig and Opal. The whole of the country south of Kashghar is cut up by one net-work of canals mingled in such confusion that nothing but a careful survey can lead to a clear comprehension of them"^ and moreover day by day they alter, and I have often seen one canal (in appearance more like a large river than anything else) eating its way rapidly through the soft soil into another one. These changes are constantly going on, and a map constructed now would be of but little value fifty years hence. After leaving the village of Sugat the road follows for nine miles, along the edge of a water-course through a stony plain, a narrow border of green showing signs of a scant cultivation. This brings us to our first halting-place, the good-sized village of Ighiz-yar, two miles short of which we pass on the left a conspicuous isolated conical hill with a zyarat (tomb) at top. Before reaching the village a few low hills come in sight on the west, being the ends of the low spurs coming down from the Kizil Art mountains, the first portion of these mountains that we had seen since leaving Kashghar. With our usual ill luck, from the day of our departure, the characteristic Eastern Turkestan haze entirely obscured all view of the lofty mountains on the west, preventing the possibility of forming any opinion as to the shape and direction of the spurs from the main ridge. From Kashghar on a clear day we have often seen the outlines of these hills standing out against the sky, but the distance, to the crest of the range, 70 miles, was much too great to permit of the intervening ranges or spurs being visible. I often longed to make a nearer acquaintance with them, but no opportunity ever occurred for doing so. Fortunately on the upward journey to Kashghar, the Pundit, who followed some days behind us, had some clear days, and was able to fix very satisfactorily the positions of several of the peaks of the main range. On the second day we marched in a south-west direction, for 18 miles, to Aktala (white plain) . The first four miles were up a gently rising stony plain, almost entirely destitute of vegetation, and extending to the entrance of the Kinkol valley, which runs between two spurs of which we had caught a glimpse the previous day. At a distance of two miles up the valley the road passes the foot of an old extensive fortification called " Khatt (lower) Karawul" constructed by the Chinese on the left bank of the river, to defend themselves against incursions from the Kirghiz marauders from the Pamirs and the Alai. It is built on a commanding position running along a spur which nearly closes up the entrance of the valley. A garrison consisting of only a few sepoys, attests the fact that the Amir^s rule has reduced these tribes to order and obedience. Two miles further up is Kichik (small) Karawul, where a road along an open ravine on the left bank leads direct via Opal to Kashghar, and a Kirghiz footpath along a ravine on the opposite side leads to Yarkand. * A great source of difSculty in investigating the courses of rivers in Eastern Turkestan is their nomencla- ture, every portion of a stream having a different name, derived from the nearest village, by which alone is it known to the neighbouring inhabitants. In addition to these purely local designations travellers generally name the rivers after the different large towns situated on their banks, while cosmopolitans have occasionally general names which they apply to a river throughout its whole course, but which are perhaps unknown to the inhabitants of the country. Most rivers are also occasionally known by names expressive their color as Kara-Su, Kizil-Su, or Kok-Su, or Ak-Su (Turki, for black, red, blue, and white rivers), terms which may be seen broadcast in almost any^ map of Central Asia. No river of Eastern Turkestan carries the same name from its cradle to its grave in the big Tarim Gol or Ergol, which swallows up all the rivers of Eastern Turkestan except those that lose them- selves in the desert before they reach it. The final end of the Tarim still remains, and I fear must remain, a matter of mystery. It is generally supposed to flow into Lake Lope (Lob, or Luf ) ; but I have recently heard, on what I considered fairly trustworthy evidence, that at about 25 miles south of its junction with the Karashahr River, i.e., about 65 miles south of Kola (Koila or Kurla), it disappears in the sand near the village of Lop (Lop being I believe a Sanskrit word signifying disappearance). It is further said to reappear in the shape of a large navigable stream at the Chinese city of Saju (P Suchan). I should myself think that it more probably reappears in the marshes and lakes which are believed to exist to the eastward and south-eastward of the still somewhat mythical Lake of Lop. Perhaps Mr. Prjevalski will some day enlighten us on this matter from the east. ( 33 ) Two miles higher is the Ghijak ravine, along which a road runs to the Alai and Khokand by the Karatash and Kizil Art Passes''^. Another seven miles up the stream brought us to our halting-place at Aktala. As we ascended, the increasing bulk of water was very appreciable, partly owing to the snow melting under the increased temperature, but also doubtless in some measure due to a large quantity being lost in the gravelly soil, which, as I have before had occasion to mention, frequently absorbs much water, that would otherwise profitably be employed in irrigation. We had crossed the river shortly before it enters the plains, and although its banks are there several feet in height, the natural bed immediately afterwards opens out on the stony plain and much water must be lost. The river divides into two main branches, the Odelang to the north, and the Ghalchak to the south. These irrigate the villages on the Grand Trunk Road between Kudok and Yangi- Hissar. Above Ghijak the road closes in and we get several very picturesque views as we advance. The scenery is veiy bold, hills several hundreds of feet in almost perpendicular height enclosing the narrow valley. Fuel, water, and grass are in abundance. At Aktala the valley opens and the river branches into two streams of nearly equal size. Some miles up the westera branch is the Kirghiz village of Chumbaz, by which a path is said to lead to the Kaskasu Pass. Our third day's march was to Sasak Taka, 131 miles up stream, by a bad stony road through a very bold defile : there was plenty of wood and grass, which also appeared to be very abundant in some of the lateral valleys. About 2 4 miles short of camp, a deep ravine comes in from the east, and the main stream turns to the south-west, retaining that general direction until arrival at the Kaskasu Pass, which we crossed on the 25th March. At five miles above Sasak Taka a stream joins from the south-east called Kinkol, and gives its name to the river lower down. On this day's march (fourth from Yangi-Hissar) we passed numerous Kirghiz camps containing altogether as many as 30 or 40 tents or Akoees ; amongst their tenants I encountered some who came originally from the neighbourhood of Almati. Although passing along a valley bounded on both sides by spurs from lofty hills which rose some thousands of feet above us, the scenery was not so bold as where we had passed through on the previous day ; the slopes were more gentle, and numerous grassy valleys entered on both sides, up which we saw many camels, yaks and sheep. On the fifth day (25th March)i) we left camp with the thermometer at 7° F. starting early in order to avoid the slush and water to be expected later on in the day from the melting snow. At first we had a bitterly cold west wind, which however moderated after an hour or so, and we had a very fine day for crossing the Kaskasu Pass. The ridge which we had to cross is a spur from the Kizil Art mountains, and separates the drainage of the Kinkol River flowing towards Yangi-Hissar, from that of the Charling River, which goes to Yarkand. For two miles after starting our direction was a little north of west, and then for nearly three miles up to the pass nearly soiith of west. The road leaves the ravine (which is steep and inaccessible) and winds up the side of the valley, passsing round the head of the ravine, and over a flat ridge, at the end of which commences a steep descent of about five miles. The height of the pass is 12,930 feet, and although the length of march was only 10 miles, the baggage ponies did not arrive in camp until late in the afternoon, owing to the slippery descent on the south side, where our loads had to be transfer- red to yaks. There was a great deal of snow on the pass, as well as on the grassy slopes on * This road is said to cross two passes before it reaches the Kizilart Diwan, and to pass, by Kichik (little) Karakul, a small lake probably not more than four or five miles in circumference, from which a road leads to Chong (great) Karakul five marches off and probably four or five days' journey in circumference. This road is said not to cross any large river but to pass mostly over high table land ; abundant supplies of grass and fuel exist throughout, and I was told that camels can traverse it the whole year round. It was formerly used by merchants going from Yarkund to Khokand but is now closed. A direct road from Yangi Hissar joins it at one day's march from Ghijak at a place called Karatash. Another road leads from Opal (about 30 miles south- west from Kashghar) to the Kizilart Pass and Alai. This road was recently used by an Envoy sent from Karatigin to Kashghar one of whose suite was wounded in an encounter with the Alai Kirghiz, who are subject to Khokand. The portion nearest to Kashghar is often used as an alternative road to Khokand when the Terek Pass is closed. ( 34 ) either side. The view was very limited, and the deep snow prevented my leavings the road. On our return journey, about five weeks later, the snow was all melted, and there was no necessity for employing yaks. The camp at Chehil Gombaz was at the junction of the streams coming from the Kaskasii and Torat"^ Passes, the two forming the Charling River, along which a direct road goes to Yarkand, now closed against traffic " by order." On the spur between the two streams is an old Chinese building called Khitai Shahr (Chinese city), a sort of square redoubt built on the steep slope of the hill, presenting a very curious appearance, and reminding one of the perspec- tive pictures of Caesar's camp, in ancient editions of that author's Commentaries. It was said to have been formerly held by a detachment of Chinese, posted to watch the Charling passage to Yarkand. The direct road from Chehil Gombaz to Yarkandf is 133 miles in length. The first march is to Tashkerim, a camping ground 19 miles down the Charling stream. From this place a path crosses the hills to the north, joining at Kinkol, the road we had ourselves followed. The Yarkand road continues down stream for 15 miles to Khaizak passing the villages of Bagh (30 houses), Kiok-tash (8 houses), Mirgul (25 houses), and Joya (15 houses). Between Chehil Gombaz and Bagh (the highest village in the valley) are numerous Kirghiz tents, the grassy valley affording an abundant pasturage to large herds of sheep and cattle, which remain in the valley in the cold weather, but are driven up to higher grazing grounds in the summer. Leaving the Charling stream at Khaizak, the road crosses two low spurs by the Kara Diwan and Kizil Diwan (on which there was no snow in April), and then descending to the bed of the Kizil streamj passes over plain and through desert § to Yakirak Kurghan, from which place to Yarkand 23| miles further on, is a rich, thickly populated, and fertile plain. The Charling and Tashkurghan Rivers unite at Khusherab, five or six miles below Khaizak ; the united stream then flows nearly due east, and is said to be met by a still larger river the Raskam (from the Karakorum Pass), at Kosherap, about 20 miles south-south-west of Yakirak. On the sixth day (26th), we made a short march of only eight miles to Pas Robat, crossing the Pas Robat or Torat (horse's sweat) Pass, which divides the drainage of the Charling River from that of the Tangitar, which also flows into the Yarkand river. The ascent was steep, and the descent still more so, the slope of the valley being 16° for a distance of about two miles. The height of the pass is 13,130 feet, the rise from Chehil Gombaz being about 3,000 feet, and the fall to Pas Robat about 4,000. While we were on the top the sky was cloudy and a fall of snow obscured the peaks to the north. On the return journey, however, I ascended a hill north of the pass and had a good though limited view in every direction. The ground rapidly rises towards the north and north-west, peaks rising to a height of as much as 4,000 feet above the pass, i.e., to over 17,000 feet. The mountains eastward visibly decreased in height as they approached Yarkand. On my 2nd visit the hills near the pass were covered with fresh, low, short, grass. About half way down to camp we came upon a number of willow trees (Turki, Suget), which continued in greater or less quantities down to the foot of the hill. In descending the stream we came upon some very thick river deposits having in places a thickness of 300 feet, and containing large boulders of syenite. The rock in situ was composed of the same materials, as that through which we had been passing for several days, viz : shales and slates, A stream coming from the north-west had a temperature of 42°, while the temperature of the air was only 24°. On the seventh day we continued our march up the Tangitar (Pas Robat) River ; after five miles we passed on the left bank the Yambulak stream leading to the pass of the same name, 14 miles off, situate in a direction a little north of west. Our own path lay along the main * Torat or " horse's sweat." t This road was traversed hy Kishen Sing Pandit from whom the information contained in this paragraph is derived. X Nearly dry in winter, but a large torrent when the snow is melting OQ the hills above. § Called " Shaitan Kum " or " Devil's sands." ( 35 ) stream for about five miles more to Tarbasbi, passing- tbroug-b a narrow and dangerous defile. Tbe road was execrable, and we experienced great difficulty and delay in getting -our baggage through. This defile would be quite impassable for field guns, and a few deter- mined men might in places defend it 'against an army. The road often runs along the bed of the stream, which contains large boulders and deep holes of water. In the winter it is probably easier to traverse, but at the time of our passage we had the double difficulties of ice and water to contend with. As far as I could learn the river is never entirely frozen over, on account of the numerous hot springs which issue from the limestone rocks forming- its side walls. One of these had a temperature of 125°, and the vegetation in its immediate neighbourhood was much in advance of that lower down the stream, and showed signs of ap- proaching spring. In the summer this road is said to be rendered quite impassable by the floods from melting snow ; the alternative route lies up the Yambulak River. On the eighth day our road lay alongside the Tangitar stream, which, from Tarbashi, ascends a gentle slope, bounded on both sides by undulating snow covered hills. The valley rises very gradually for about nine miles, up to an almost imperceptible water-shed (14,480 feet high), by which we reached the Chichiklik plateau, a broad elevated valley whose drainage passes south, through a somewhat narrow defile, to the Tashkurghan river. On the plateau close to the water-shed were two small frozen lakes. The summer road before alluded to which goes up the Yambulak stream enters the Chichikhk plain by the Yambulak Pass about four miles to the north of where we crossed the water-shed. On the opposite side of the valley, which was between four and five miles wide, in a south-west direction from where we entered it, is the Pass of Kok Mainak, on high spur running dowti from the Kizilart mountains. By this pass is the shortest road to Tashkurghan, but on our outward journey it was so deep in snow that we were obliged to take the alternative route down the Shindi valley. It is the Kok Mainak Pass that is called " Chichiklik^' % the " Mirza,'' '' Fyz Bux," and other travellers, but the correct name as given by the Kirghiz who lived in the neighbourhood is, I believe, as I have given it. Our road lay down the stream. After the first two or three miles, where the slope was very easy, the valley narrowed, and the road became exceedingly steep and difficult, passing for several miles through a succession of rocks and boulders At 10 miles below below the lakes we came to our camp at Balghun, shortly before reaching which the valley had opened considerably although surrounded on both sides by lofty mountains. The following day (29th), we descended four miles to the junction of the Shindi with the Tashkurghan (or Sarikol or Taghdumbash) R-iver,^ just above the Sarikoli village of Shindi, in^ habited by Tajiks, and containing about 15 houses, situated in a small well cultivated valley, about two miles long- by one broad. Our road now lay up the Sarikol river, but I descended it for about four miles to fix the direction in which it flowed away (south-east by east) f ; returning to the junction I crossed the main stream with some difficulty owing to the rapidity of the current, and continued along its right bank, where the road passes through a very wild defile of crystalline rocks, forming almost perpendicular banks about 2,000 feet in height, through which the river winds its way with a most tortuous course. At about 10 miles above the junction we emerged on the north-east corner of the Tashkurghan or Sarikol plain. The road by which we travelled is only open in winter, as in summer the large mass of water in the Sarikol River makes it impassable. The route by the Yambulak and Kok Mainak Passes is then used. On entering the Sarikol valley we strike the junction of the Tagharma stream with the main river. The former comes from the plain of the same name on the north-west, and has been incorrectly described by a former traveller as the main source of the Sarikol River. When we saw it there was but little water coming down (temperature 38°), although in summer there is considerably more, but the small size of the water-course, and the evidence of * Sometimes also called Yarkand River. t From Shindi a path goes down the river, but is only practicable during two or three months in winter, ( 36 ) the inhabitants of the district, all tended to show that the river which flows down the Sarikol valley from the Kanjiid mountains, and through the Taghdumbash Pamir, is undoubtedly the main stream. At the point I crossed, just above the junction with the Tag-harma stream, it the river was 15 yards in width, with an average depth of li feet, and a velocity of four miles an hour. Two miles beyond in a southern direction brought us to the village of Chushman, leaving only five miles for our next day^s march (the* 10th from Yangi-Hissar) into Tashkurghan, the chief town or rather village of the Sarikol valley. Between Chushman and Tashkurghan (both on the left bank of the Taghdumbash River) we passed the large village of Tiznaf. It is this village that has caused much confusion to geographers by giving its name to the river, which is frequently called the Tiznaf in its lower course, and is often confounded with another river of the same name which rises on the north side of the Yangi-Diwan Pass and flows past Karghalik. On approaching Tashkurghan (where we halted two days to rest our cattle), while passing up the valley I saw at its upper end some high peaks occasionally emerging fi-om the clouds, but before I could get to camp they had disappeared, never to be seen again during our stay in the valley, or on our return journey, a great disappointment to me, as it is possible they were peaks in the Muz-tagh range, fixed by the Great Trigonometrical Survey. During our halt at Sarikol I took a set of magnetic as well as the usual astronomical observa- tions, and some careful azimuthal bearings with theodolite to a large mass of snowy peaks called Muz-tagh, situated to the north of Tashkurghan. These are identical with Hayward's Taghalma mountains, which are visible from Kashghar, the highest peak of which I determined by accordant trigonometrical measurements from Kashghar and Yapchan, to be 25,350 feet above the sea. The general outline of the Tashkurghan valley towards its head, was fixed by bearings taken from different points on a line across it. On the return journey I was able, by ascending the ridge that separates this plain from that of Tagharma on its north (and by making a detour through the latter on the way to our camp at the foot of the Kok Mainak Pass) to lay down the borders of the northern plain with considerable accuracy. Practically the two form one large plateau divided in the middle by a low range of hills through which flows the • Tagharma River. The Tashkurghan plain extends southwards from the dividing ridge before mentioned, right up to the foot of the Kanjud passes in the Mnz-tagh range, con- stituting in its southern portion, the Taghdumbash Pamir. The Sarikol valley may be said to have an average width of about four miles ; it is bounded on the east by the snowy range of Kandar or Kandahar ;* on the south-west and south are the Taghdumbash mountains ; on the west the Shindi mountains; north-west the Bir-dash, which also forms the western boundary of the Tagharma plain, to the east of which lie the Muz-tagh (or Tagharma) and the Chichiklik mountains. The Tagharma plain extends from the dividing ridge for about 12 miles in a north-north-west direction; it is only two miles in width immediately north of the ridge, but soon increases in an easterly direction to as much as 10 miles; it then narrows, being nearly closed up by spurs running down from the Bir-dash mountains on the west and the Muz-tagh on the east. About 10 miles west of this point is the Bir-dash Pass, over a range which divides this plain from another similar one running nearly parallel to it, ^7^z., the Ak-tash or Ak-su. Opposite the Bir-dash Pass the Sarikol plain again widens and extends, gently undulating, for some eight or ten miles further in the same direction.f According to statements of the Kirghiz it continues right up to the Kizil Art Pass, which separates it from the Alai, and the valley of the Surkhab River, the most northerly tributary of the Oxus. The height of the valley above sea level may be taken at Tagharma at about 10,500 feet, and I doubt whether it is very much higher in any part of its course. The drainage of the southern portion passes through the Tagharma plain into the Sarikol River ; * Over these mountains is a road to Tarkand, wliich. descends into the Tung valley and after passing down it for a march or two, crosses the Arpatalek Mountains, and enters the Turkestan plain near Kosherap. t Thus far I myself saw, from the ridge dividing the Tagharma from the Tashkurghan Plain. ( 37 ) and in a somewhat central position on its east edge, I would place the lake of Kichik Karakul,* about three short days' march north of Tashkurghan. Further north again, south of the Kizil, Art Pass, is the larger Lake Karakul, from which a stream is said to flow westward into 'lAJL the M^tghabi river. I have shown in my map, what I consider the approximate positions / of the various lakes and mountain ranges in those regions, but I of course cannot guarantee I the accuracy of anything off our own Kne of march. The Tagharma plain presented a very lively spectacle : fully 100 Kirghiz akoees were within view, scattered about in different parts of the valley ; their tenants, of the Sark or Syok tribe, being subjects of the Amir of Kashghar. Open, grassy, well watered, and speckled all over with camels, yaks, horses, sheep, and goats it formed a pleasant sight after the wilds through which we had been wandering, and was a striking contrast to the Tashkurghan valley, which looked by comparison a picture of desolation, owing to the numerous uninhabited villages and tumble down houses with which it is covered. The water from the warm springs which issue in numerous places from the earth, causes the young green grass to rise (in April) in great profusion. Formerly in the south-east portion of the Tagharma valley, at Kila-i-Tagharma or Besh Kurghan (the five forts) there were about 50 houses inhabited by Tajiks, under Sarikol. Their history has been a sad one. I got into conversation there with an old man, who told me that nearly each fort had its history. In the principal one, some thirty years ago, resided Mahomed Alum, the Hakim of Sarikol. He was attacked by a number of Andijanis from the north, himself and many of his followers killed, and the remainder carried into slavery ; my informant with only one or two others escaping into the neighbouring hills. At the fort where I was standing, fifteen years later, the Kanjudis had made a raid from the south, and had kiUed or carried into slavery the whole of the inhabitants. Two young men standing by me had been carried off in this very raid as children, and sold as slaves in Yarkand, where they had been released shortly after the accession of the Atalik to power, but they had only within the last month been allowed to return to their homes, where four Tajik families now represent the fifty that had formerly lived there. They were doing their best, with the help of some of the neighbouring Kirghiz, to put their fields into order, and I there saw, for the first time in my travels, the yak yoked to the plough. There is much culturable ground, and it is to be hoped that this recommencement of cultivation on a small scale is only the prelude to a larger. In the time of the Chinese rule, such was the insecurity in these parts, that the inhabitants of Sarikol dared not wander far from their villages, for fear of being seized and carried off either by the Kirghiz from the Alai, or by their neighbours of Kanjud ; now they tell me that if a man drops his w/iip in the middle of the plain, he will find it there if he looks for it a year afterwards. This is a favourite saying amongst the people of Eastern Turkestan, which I have heard more than once employed to describe the sense of security enjoyed under the present reffime. On our return to Yarkand we passed along the south edge of the Tagharma plain. The direct distance from Tashkurghan to the foot of the Darschatt ravine leading to the Kok Maindk Pass is about twelve miles; thence to the pass itself (15,800 feet) is six miles, by a very difficult and stony road. The pass is four miles from the small lakes on the Chichiklik plain. The fort of Tashkurghan, said to be of very ancient date, and to have been founded by ,-7 Afrasiab, the King of Turan, has been described by former travellers who had a better oppor- tunity for inspecting it than we had. The " Takhsobai," or Governor, evinced so great a dis- inclination to receive our visit there, that we had to content ourselves with inspecting it from a distance. The part at present inhabited is apparently of modern construction, and built of * The waters from this lake are said to form the Yamunyar Eiver, which flowing through the Chakar Aghil defile eastward, under the name of Gez River, enters the plains under the name of Yamunyar, and, as before mentioned, divides into several branches near the villages of Tashbalig and Opal and irrigates a great portion of the country south of Kashghar. / ( 38 ) stones and mud, but there were in places remains of " roughed " stone facings on the sides of the rock on which the fort is built In its neighbourhood are numerous fragments of broken wall, but I could not recognize any continuous line marking out its former limits. Taking a hint from Sir Henry Rawlinson's writings, I kept on the look-out for Buddhist remains, but could see nothing. The Pandit (who accompanied us as far as Sarikol), seemed to think that the custom which prevails throughout the whole country between Turkestan and Wakhan of heaping up skulls and horns of sheep and wild animals at the different ziarats or tombs, was a relic of ancient Buddhism. I believe that the custom is common through- out the whole of Central Asia. It certainly is so in Ladakh and Eastern Turkestan. From Tashkurghan to Panjah there are two roads commonly used by merchants ; the first, over the little Pamir, was followed by us on the outward journey, is generally used in winter ; the second is over the great Pamir and is used in summer. The latter is the easier road, but passes over much higher ground than the former and is impassable for caravans in winter, on. account of the deep snow lying on it. A third alternative road is by the Taghdum- bash Pamir* at the head of the Sarikol valley. It lies high, and in midwinter is deep in snow : in former years it was much used by the Bajaori f merchants, who used to go from Badakhshan to Yarkand by the Taghdumbash and Tung valley roads, thus keeping at the greatest possible distance from the Alai Kirghiz, whom they seem to have feared more than they did the Kunjudis. On the Great and Little Pamir routes the first two marches, i.e., to the west foot of the Neza Tash or Shindi Pass, are common to both lines of road, which meet again opposite the village of Zang at the junction of the two large streams which form the Panjah E-iver, the most southerly branch of the Oxus. Leaving Tashkurghan on the 2nd April, our road lay nearly due west for four miles, up a stream which issues from the Shindi mountains through a narrow and difficult defile. J The water, which even thus early in the season flows in considerable quantity, combines with the rocky nature of the bed of the river, which has to be crossed and recrossed in numerous places, to make the road exceedingly difficult for laden horses. After passing four miles up the defile we reached a camping ground called Jangahk situated in a well wooded open valley, about two miles long by half a mile broad. It is often used as a halting-place by travellers, but we continued up the valley to Kanshubar, a march altogether of 16 miles. Shortly before reaching camp we passed numerous hot springs. We pushed on the next day in a south-west direction until we reached the foot of the Neza Tash Diwan, where we turned westward, and passing over a low spur continued our way up the valley, entering a large basin with lofty mountains towering above us on both sides, very bold and precipitous, and of a very peculiar and striking ferruginous colour. § A stiff pull through the snow to the top of the pass (14,915 feet above the sea level) and we were standing on the water-shed between Eastern and "Western Turkestan. I had been given to understand that we should here come in view of the " Pamirs" and was somewhat surprised at seeing in front of me nothing but a long range of low red colored hills about ten miles distant, a portion of which to the right was pointed out to me as the Great Paxuir, and another on the left as the little Pamir. Nothing was visible but an irregular mass of hills whose serrated tips did not appear to rise more than 1,000 feet above the Pass on which I was standing. In front lay a large valley running in a northerly direction which subsequently turned out to be that of the Aksu River, the principal source, as now appears of the Oxus||. The apparent continuity of the range in front of us was, as we shall hereafter see, a delusion ; the hills really form the ends of broad transverse ranges, running in a westerly * It is said that Alif Beg fled from Sarikol by this route. t The district of Baiaor or Bajaur lies to the west of Swat, and its inhabitants are well known as enterpris- ing traders. ■*■ The rocks forming this defile were composed of gneiss. These mountains, Dr. Stoliczka informed me, were composed of triassic limestone. Which name is perhaps derived, as Venoikof suggests, from Ak-su. ( 39 ) direction, and separating the various Pamir valleys, which were concealed from our view by the low hills in front. Descending from the Neza Tash Pass a march of a few miles in a westerly direction, through heavy snow, brought us to our camp at Kogachak, which is about three miles above the junction of the stream from the pass, with the Ak-su River. On the following day (3rd from Tashkurghan) crossing the spur between the two streams we descended into the valley of the Ak-su, a little north of Ak-tash, * at an elevation of 12,600 feet above the sea. We continued south for six miles up the valley, which was here about two miles broad, and deep in snow. In front of us was a fine range of snow covered peaks, running in a direction a little south of west, forming the southern boundary of the little Pamir, which is really the upper portion of the Aksu valley. The latter as we advanced, gradually turns round south-west by south which direction it retains up to and beyond the lake of little Pamir. The Little Pamir is generally considered to commence near where we entered the Ak-su stream, and consists of a long, nearly level, grassy valley, varying from two to four miles in breadth and enclosed on either side by ranges of snow covered hills sloping down rather gently towards it. Its length from east to west is about 68 miles. The Great Pamir, and all other Pamirs are as far as I could learn, of precisely similar character. The ground intervening between the Great and Little Pamirs is fiUed up with lofty mountains of tolerably uniform height and without any very conspicuous peaks, the hills to the west near the junction of the two main branches of the Panjah River being perhaps the highest. Our first halt in the Little Pamir was at Onkul, after a march of 25 miles for a great part of the way over snow, and with such a very bitter wind blowing in our faces, that it was almost impossible to hold an instrument in one^s hand. After entering the long straight reach above the turning, near Ak-tash, several large open valleys are passed on the north, where the hills are comparatively low and undulating, those on the south side being generally much higher. Our second day's march (4th from Tashkurghan) through this Pamir took us along an almost level road for 24 miles. As on the previous day, snow covered mountains lay on both sides as we advanced, and there was a great deal of snow in the valley itself which varied in breadth from 2 to 3^ miles. There was often a good deal of saline matter in the soil, and where this was the case the snow generally melted long before it did so elsewhere. Our camp was on the north edge of the Little Pamir Lake, which has been given by recent travellers the very different names of Barkut Yassin, Chalap, and Gez Kul or Goose Lake (Turki, Oi-kul) . I made repeated enquiries as to its proper name, and found that the Wakhis generally call it the Kul-i-Pamir Khurd, or lake of Little Pamir, while the Sarikolis and Yarkandis give it the name of Oi-kul. As some doubts had been expressed, as to the supposed double exit from this lake, I was naturally very anxious to determine the point, and in ascending the valley on this day's march I took at some twenty different points, observations with aneroid barometers to determine, if possible, the exact water-shed, which from previous accounts I had fully expected / to find at the ea^st end of the lake. The ground, however, was so level for several miles there being a rise of only 230 feet in the 24 miles between Onkul and the lake, that the ( aneroid was not sufficiently delicate for the purpose, and although I walked for a considerable distance on the frozen stream to enable me to satisfy myself on the subject, I arrived in camp on the banks of the lake re-infecfd. The following morning I walked over the lake to its east end, which from a little distance off appeared entirely closed, but on walking round the head to make certain, I was soon undeceived by coming across a very narrow outlet, about nine paces across, and only a few inches deep, all ice of course. I then walked several miles down the stream (east) until I became fully convinced that its bed did slope to the east and drain into the Ak-su. This result being contrary to what I had anticipated, I then rode to the west end of the lake to see whether (as has always been supposed) a stream issued from that end also. I left my horse and started on foot to go round its head ; the ice at this * Three miles distant from Kogachak. ( 40 ) end, instead of being firm and strong, as at the other, was very brittle and would not bear my weight, so I had to wade through the heavy snow and slush on its banks. I soon came across a warm spring, from which water was decidedly flowing due east. A little further on I encountered a frozen stream, on going along which westward the barometer showed that I was walking up hill. I advanced still further, hoping to get completely round the head of the lake, so as to be quite certain that there was no outlet draining westward, but the walking in the deep snow at so great an elevation had completely exhausted both myself and the man who was with me, and it was with some difficulty that I got back to my horse, and hurrying on with no guide but the tracks in the snow left by the rest of the party, it was with great difficulty that I reached camp, 20 miles from the lake, shortly after dark. On the return journey, the Ressaldar came back by this road, and, according to a promise he had made me, rode completely round the head of the west end of the lake up to the foot of the steep mountain rising on the south side. The snow was then all melted, and water was flowing into the lake from the two sources I have just described, and nothing was flowing out. He then went to the east end, whence a stream was flowing towards Ak tash, so this problem has been solved in a-somewhat unexpected manner. The lake has only one outlet, and that eastward, and its waters flow into the Ak-su, afterwards the Murghabi, which joins the Oxus near Wamar, and is in all probabiliti/ the longest branch of the Oxus. I have tried hard to discover the true onward course of this Ak-su River. On our return journey we struck it some 14 miles north-north-west of Ak-tash. It flows thence in a northerly direction for 12 miles and then turns off out of sight north-north-west. It is said to flow in a northerly direction for two marches (say 40 miles) from Ak-tash, after which it either joins, or becomes, the Murghabi River changing its course westward and flowing through the Sariz Pamir to Shighnan. It passes through Bartang, a district of Roshan and joins the river Oxus just above Kila Wamar, the chief town of Roshan. The Little Pamir Lake is 13,200 feet above the sea level. It lies from south-west by west to north-east by east, and for a length of 3^ miles is from 1 to 1^ miles in width; it narrows considerably eastward, where, for about 1^ miles it is nowhere more than a few hundred yards in breadth. Nearly opposite the south-east corner, in a side ravine is a large glacier which drains into the Aksii stream shortly after the latter emerges from the lake. The road passes along the north side of the lake and crosses the watershed two miles beyond the west end at a height of not more than 150 feet above the margin of the lake. Other parts of the watershed, which is nowhere well defined, are probably still lower. The descent beyond is somewhat rapid ; we passed on our left a small stream which rises near the watershed, and takes the drainage of the hills to the south-west of the lake. We went for about seven miles down an open valley, (crossing several small streams flowing down large open ravines on the north), and reached some deserted Kirghiz huts and tombs called Gombaz-i-Bozai, close to where a large stream, the principal affluent of the Sarhadd branch of the Oxus, comes in from the south-east. This river has it source in the Kanjiid mountains on the west side of the Karachunkar Pass"^ which crosses the Shindi or Pamir Range, south of the Neza Tash Pass, and separates the Taghdumbash drainage from tlie head waters of the Oxus. After passing Gombaz our path lay on the right bank of the Sarhadd stream, where we met with a constant succession of steep ascents and descents. The regular path had often to be quitted in order to avoid drifts of snow, which in places lay very deep. In the winter, when the stream is completely frozen over, its hard surface makes a capital road, which is always used by travellers. We passed at a bad season of the year, too late to be able to keep to the ice with safety as it was now breaking up, and yet before the show on the upper road was melted. Later on in the hot weather, the lower road becomes altogether impracticable, as it is impossible * The road before mentioned which was once much frequented by Bajaori merchants crosses the Karachunkar Pass. It is now but little used. ( 41 ) to cross the then swollen river. "^ After a while our road left the main valley which makes a detour to the left and ascends a gentle slope to a low pass, crossing- which the path returns along another broad valley to our camp at Langar. This point has been considered the end of the Pamir, but I should rather be inclined to consider Gombaz-i-Bozai as the true ending. This would reduce its length from Aktash to about 56 miles. Our next day's march (6th from Tashkurghan) was to Daraz Diwan, a distance of 15 miles; the road soon struck the main valley and continued along its northern side over a constant succession of ascents and descents, passing occasionally through snow in deep patches. We saw on the hill side a large number of juniper trees, and in some of the side ravines were birch trees and wild roses. In fact, wherever water trickled down there were signs of vegetation, but everywhere else the hills were bare. In one or two places the road descended to the river bank ; in places the stream was entirely frozen over, the water flowing underneath, elsewhere it was altogether clear of ice. At two or three such places I estimated the breadth to be about 40 feet, depth 2 feet, and velocity 2i miles per hour, temperature of water 35°. Before reaching camp was a very steep descent, having a fall of over 1,000 feet, which it only took a quarter of an hour to walk down. The river is here called by various names, Kanjud, Sarhadd, Panjah, and Hamun. The last name I have heard more than once, and it is of course the same as " Amu.'^ Wakhan seems to be but little better off than Turkestan in the numerous names borne by the same stream. Our seventh day's march (8th April) was at first, as hitherto, on the right bank of the stream, the road crossing high spurs by very steep ascents and descents (which lead me to suppose that this was the road followed by Marco Polo) . There are three roads used at different times of the year, one (in midwinter) on the surface of the frozen stream, a second which we followed, occasionally along the stream, but which generally passed over spurs, and a third, much higher up, and avoiding the stream altogether. We passed several small tributary streams and between the fifth and seventh miles we had to cross the main stream many times where it passed through very steep hills. We crossed generally over ice and snow bridges. At last we emerged into a large open gravelly plain watered by several streams and soon arrived at the village of Sarhadd (head of the boundary), the highest inhabited village of the Wakhan valley, and situated about 11,000 feet above the sea. The march was only 11 miles, but difiicult. We were here met by Ali Murdan Shah, the eldest son of the Mir of Wakhan, who had marched out from Panjah to meet us. On the 8th day (from Tashkurghan) we took a very short march of only four miles to the large village of Patuch or Patur. From this day forward, in order to avoid all cause of suspicion, I took no observations on the road, but accompanied the rest of the party on the march. We were now a large detachment, as we were always escorted by the Mir Bachcha and his somewhat ragged following. This march was, while it lasted, the most trying I have ever experienced, owing to the intense bitterness of the cold wind and drifting snow which blew in our faces the whole way. From Patuch to Kila Panjah, the residence of the Chief of Wakhan, there is not much of geographical interest to notice. At Yiir, 15 miles west of Patuch, a very difficult pathway crosses the mountains to Chitral, and at Vost, about seven miles short of the junction of the two Pamir streams, there is a small fort which covers the entrance of a valley up which another footpath leads to Chitral. The road from Patuch to Panjah, about fifty miles in length, lay along the valley of the Sarhadd stream, sometimes on one side of it, sometimes on the other. The valley was bounded on both sides by lofty and generally precipitous mountains, of whose height it was impos- sible to form any idea, as their tops and the greater part of their sides were always wrapped in clouds and mists. It was perhaps fortunate for me that I was unable to use my instruments, as I know nothing more disheartening to a surveyor than proceeding for days down a valley under such circumstances. Villages were scattered all along the road on both sides of the stream. In the whole distance from Sarhadd to Panjah there are probably about 400 houses, and their corner turrets, like those in the Sarikol valley, are evidence that the inhabitants have not fallen upon much easier times than their neighbours of Sarikol. The houses are not so good as those of Turkestan, and are apparently especially designed to keep out the wind, which seems always to be blowing violently either up or down the valley, generally speaking from west in the * The road by the great Pamir is then adopted. ( 42 ) mornings, and from east in the afternoon. On entering* a house one generally passes through the stables, containing two or three horses or cows, after which one traverses a long winding narrow passage, which leads to the centre of the house which is generally very small and dirty. In the centre is a fire-place, a kind of globe-shaped stove, about 2^ feet in diameter, made of mud, and open in front for the passage of air and fire-wood. Above, is a hole in the timber roof for ventilation. The roof is dome-shaped, supported on cross-beams resting on timber uprights which surround the central fire-place, and help to support the side apartments which all open inwards towards the fire and to one another. Here the different members of the family reside. The larger portion of the house is given up to the females, who, somewhat bashful but good humoured, appear to have a very good idea of keeping the men of the household in decent subjection. The males all wear brown woollen chogas or cloaks of country make ; pubboes or boots of the same kind as are worn by the Ladakhis ; loose trowsers of the same material as the coat; and a generally scanty cotton turban, the almost universal colour of which is blue and white. The women, who are not over good-looking, but are pleasant and matron-like, dress very much like the men, and have long plaits of hair falling down the side of their heads. There is no artificial modesty or attempt to conceal their faces. In a cottage where we took refuge the females remained present the whole time we were there, and made some most excellent barley bread for us, kneading the flour into a cake which they plastered into the inner wall of the oven ; after frequent turning a capital result was secured. Their physiognomies are very divergent, most of them have Jewish noses, but one boy I saw with a most perfect Greek profile. Tbey all age verj- early, and attribute their grey hairs to the poverty of the country. The men seem affectionately disposed towards the females, always handing them fruits, sweetmeats, or any little trifles we might happen to give them. They are all poor ; money and ornaments seem almost unknown, and hardly anything is seen in their houses that is not the produce of the country. At Sarhadd the temperature of the water was 3£°. At Yur I found it to be 40° while the stream was about 60 feet in width, one and a half feet in depth, with a velocity of two and a half miles an hour. Near Yur we passed a large stream on the right, and another on the left bank. Throughout the valley there was much jungle wood, and some old coarse grass, but up to date (April 10th) we saw no signs of spring, neither trees budding, nor grass sprouting. The road throughout is very stony, but not otherwise bad. In many places the tributary streams have brought down immense quantities of stones and debris, which threaten to block up the main stream. This debris generally spreads in a fan-like shape, from where the tributary stream opens into the main valley, and causes the river to flow round the base of the fan. At Babatangi, the valley, which from Sarhadd had varied from three miles to one mile in width, is confined by precipitous mountains to a breadth of about one-third of a mile. It soon opens again however, and shortly after leaving Sas (about 13 miles above Panjah) it enlarges considerably, and gradually opens into a considerable plain, being joined a few miles above Panjah by the valley containing the stream from Upper Pamir. Before the junction the Sarhadd stream passes for several miles through rather dense jungle composed of red and white willows. On our march into Zang (near the junction of the streams) we crossed the river of the Great Pamir here about 30 feet wide, one foot deep, with a velocity of about three miles per hour. It was very considerably smaller than the river we had followed from Sarhadd. Where the streams meet the valley is about three and a half miles wide and almost entirely covered with jungle. It narrows gradually towards Panjah, where it has a breadth of two miles. The height of Panjah above the sea I found to be but little more than 9,000 feet. The vegetation in the valley was very backward, much thrown back doubtless by the violent winds which tear up and down with a bitterness difficult to imagine unless it has been felt. The grass was beginning to show signs of sprouting in the middle of April, and the cultivators were then commencing to turn up the soil preparatory to sowing. The Oxus River flows on the north side of, the valley, and on its left bank is Panjah, between which and the mountain range to the south, a distance of nearly two miles, the ground is almost com- pletely covered by fields, irrigated by a stream which issues from a large ravine on the south, and is derived from a large glacier which entirely blocks up the valley in which it is situated, ( 43 ) and whose foot merging into a snow bed, is not more than 1,000 feet higher than the Oxus valley. At the head of the ravine containing this glacier are some snowy peaks, about six miles to the south,"'^ which I estimated to be between 17,000 and 18,000 feet in height ; they appeared to be on spurs of the Hindoo Koosh Range. It was most annoying being shut up at the bottom of a deep valley, and unable to get a nearer view of these peaks, but there was no help for it, the ravines entering the main valley from north and south were generally inacces- sible, the one on the south being, as I before mentioned, blocked up by an enormous glacier, which was quite impassable, while those on the north are almost vertical chasms which looked as if the mountain had been split up by an earthquake. During our stay at Punjab, J. ascended the mountains to the north to a height of about 3,000 feet above the valley only to find that I was on the lower portion of a much higher range behind, which obscured all view further north, while the hills to the south of the Oxus were so high that they intercepted the view of any peaks on the main range of the Hindoo Koosh that might otherwise have been seen beyond ; in fact I could see very little more than from the ground below. On the only other fine day that we had during our stay at Panjah I went down the valley for about 12 miles, but saw little more than one or two peaks of the range to north. Pan j ah itself is, or rather was, built on five small hillocks, hence perhaps its name,t and I have no doubt in my own mind that the river takes its name from the place, and not the place from the river. These five hillocks are situated near each other on the left bank of the stream, the largest is covered by a fort, the residence of the Mir Fateh Ali Shah, and most of his followers, the other is of nearly equal size, covered by houses, and surrounded by a strong wall ; on two others are small fortified buildings, while on the fifth there are nothing but ruins and graves. These fortified buildings (in one of which resides Alif Beg, ex-ruler of Sarikol) from their near proximity to each other, and commanding situation, form a position of con- siderable strength, and might hold out against an attacking force for some time if artillery were not brought against them. The Mirs of Wakhan have more than once held out in this stronghold against the forces of the Ruler of Baddkhshan to which country tliey are subject. The whole population of Paujah perhaps does not exceed a hundred and fifty souls. The district of Wakhan has been described by former travellers. It comprises the valleys containing the two heads of the Panjah branch of the Oxus, and the valley of the Panjah itself, from the junction at Zung down to Ishkashim. The northern branch of the Panjah has its principal source in the Lake Victoria in the Great Pamir, which, as well as the Little Pamir belongs to Wakhsin, the Ak-tash River forming the well recognized boundary between Kashgharia and Wakhan. Both of the Pamirs were thickly inhabited by Kirghiz in former years, subject to Wakhan, but they are now unoccupied, the constant feuds * Their exaxjt diatance I was unable to determine, as they could only be seen up the ravine, which has too narrow to permit of a base being measured across it of sufficient length to enable an accurate estimate to be made of the distance of the peaks. t "Panj" is the Persian for "five." One possible derivation of the word Panjah is given above. Some authorities would derive the word from the five rivers which are supposed to form the head waters of the river on which Kila Panjah stands. There are two objections to this theory : — Is^. — It is contrary to the custom of Turkestan to narqe a place after a river, and to a hundred cases that I know of where the converse holds good, i.e., a river is named after a place on its banks, I do not know a single instance of a place being named after a river. 2nd. — The word is usually pronounced Panjah, which is nearer in sound to the Persian word " Pinjah" or fifty. The true origin of the word I believe to be from the Panjah or palm (of the hand) of Hazrat Ali (the son- in-law of Muhammad). In a building on a small hijl about 2 miles to the south of Kila Panjah is a stone bearing the impress of a hand. Local tradition says that when this country was in the hands of the Zar-dushtis, or atash-parast (fire worshippers) the people were converted to the religion of Muhammad by a visit (in the spirit) from Hazrat Ali, who left his mark on the stone as thus described, which is an object of religious veneration in the neighbourhood. At Bar Panjah in Shighnan is a similar mark over which the Fort " Bar Panjah," " over the Panjah," has been built. Possibly this tradition has something in common with that which attributes the derivation of the word Pamfr to " Pa-e-Mir," i.e., the foot of the Mir Hazrat Ali. I would myself be inclined to derive the word from "Pam," the Kirghiz word for roof, and "yer," which is both Turki and Kirghiz, for " earth" corresponding to the Persian word " Zamin." Bam-i-dunya or " roof of the world" is a name by which the Pamir is well known. ( 44 ) between the Shighnis, the Wakhis, the Kirghiz of the Alai, and the Kanjudis^ having rendered the country quite unsafe. The highest inhabited village in the northern valley is Langar Kisht, only a few miles above the junction^ and on the right bank of the stream. The Sarhadd valley (the southern branch) is inhabited from Sarhadd downwards, and there are villages scattered along both banks of the Panjah river down to Ishkashim. Wakh^n formerly contained three " sads" or hundreds, i.e., districts, containing 100 houses each — 1st. — Sad-i-Kila Vost or Sarhadd extending from Langar to Hissar. 2nd. — Sad Sipanj* from Hissar to Khandut. Srd. — Sad Khandut from Khandut to Sad Ishtragh. To these three Sads has recently been added that of Sad Ishtragh, which I believe only became a portion of Wakhan in recent times. It lies between Khandut and the State of Ishkashim.t Abdul Subhan estimates the number of houses in Sad Ishtragh at 250, and allowing 100 for each of the other districts this gives a total of 550 houses, with a popula- tion of about 3,000 souls. The Oxus helow Wakhdn. I have now to deal with one of the most interesting geographical problems of the day, viz., the probable course of the Oxus or Panjah from the point where it leaves Wakhan to where it emerges in the plains north of Said village on the frontiers of Kolab, where it has been seen and described by our countryman. Lieutenant Wood. My assistant, Abdul Subhan, left us at Kila Panjah (the most westerly point reached by the European members of the Mission), and followed the course of the Oxus through Wakhan for 63 miles to Ishkashim, thence turning fffftS' northwards he followed the same river for nearly a hundred miles, passing successively through f(y^\/ the districts of Gharan, Shighnan, and Roshan, countries which have hitherto been known to us hardly even by name. From his report I have obtained the information following. The small State of Ishkashim forms, together with Zebak, one of the numerous petty feudal States tributary to Badakhshan. The present ruler of both these small districts is Shah Abdul Rahim, a Syud of Khorassan, who was placed in power by Muhammad Alum Khan, the present Governor of Balkh. It is said that the hereditary Chief of the country, Mir Hak Nazar, was ejected in order to make room for Abdul Rahim. The present territory of Ishkashim extends for about 16 miles to the north of the village of the same name, which now contains about forty houses, and consists, as is generally the case in those parts, of numerous scattered farm houses. There are small villages throughout this district on both banks of the Oxus ; Sumchun and two others oi^ the right bank, and Yakh-duru and Sar-i-Shakh on the left. These belonged to Sad Ishtragh, which was once a separate principality, but is now a district of Wakhan. The road from Ishkashim runs along the left bank of the river up to six miles beyond Sar-i-Shakh, where the river is crossed by an easy ford. In the month of May the water flowed in a single stream, which was 3^ feet deep and about 200 yards in width. In summer it is impossible to cross the river at this point, and a very difficult path leading along the left bank is followed. Down to this point the valley is open, four or five miles in breadth and richly cultivated. The ford marks the boundary between Ishkashim and the district of Kueheh Gharan or " narrow caves,^' which has been for centuries famous for its ruby mines. The Gharan country extends along both banks of the Oxus for about twenty- four miles, and was once upon a time rich, flourishing, and populous. Remains of large villages exist on both banks, and bear witness to the oppression that has been exercised by successive Governors of Badakhshan. The fields near these deserted villages are now culti- vated by the inhabitants of the neighbouring districts of Ragh (the chief town of which is Kila Masnuj) and Sar GholamJ which are at a distance of a long day^s journey on the further side of a range of hills, running parallel to and on the left bank of the river. Barshdr * Sad-i-Panjah. \ Written in the vernacular Shikasliem. ( 45 ) or " above the river," situated four miles below the ford, is the first of these large deserted villages. A little beyond it a large stream enters the Oxus from the east deriving its name, the Boguz, from a -village of some 30 houses situate ten miles up the stream. From this village a road goes to the Shakhdarah district of Shighnan. Near the junction of the Boguz with the Panjah the road crosses to the left bank of the river. Nearly opposite to Barshar is a ravine by which a road goes over the Aghirda Pass to Faizabad, the chief town of Badakhshan. It is said to be open all the year round. Throughout the remainder of the Gharan district numerous ruins are passed on both sides of the stream, the largest of which, Shekh Beg, on the right bank, must formerly have contained about 200 houses. On the same side of the river some sixteen miles below Barshar are the celebrated ruby mines, once the source of considerable wealth to the Rulers of Badakhshan, but now apparently nearly exhausted. These mines have until lately always been worked for the immediate benefit of the Governors of Badakhshan. At the present time some 30 men are employed there under the orders of a few sepoys belonging to Muhammad Alum Khan, the Governor of Turkestan, who sends the produce to the Amir of Cabul. It was said that during the past year one large ruby about the size of a pigeon's egg was found and sundry smaller ones, the whole of which were sent to the Amir. The rubies are found in one very large cavern to which there are three entrances, situated about 1,000 feet above the river, and about a mile up the hill side ; the task of excavating appears to be not unattended with risks, as three workmen were recently killed, having fallen from the rocks while searching for the precious stones. There is a peculiar kind of soft white stone which is found imbedded in the harder rock and in this the rubies are found. In former years the inhabitants of Gharan who worked these mines paid no taxes and held their lands rent-free. The twenty men who are now employed at the mines have to furnish their own food as well as that of the guard, and also to provide lights, torches, and implements for working. The numerous deserted villages prove the possession of the mines to have been a curse rather than a blessing to the inhabitants of the valley, who have from time immemorial been under the direct rule of tiie Chief of Badakhshan. Above the mines is a small village called Koh-i-Lal or " Ruby Mountain," and about one mile below them on the opposite bank of the river is the large deserted village of Shekhbeg whose ruined houses are built with stone and lime.* A small river enters at Shekhbeg on the left bank, and four miles up it lies the village of Gharan Bala said to contain about 100 houses, invisible from the river. A few miles below Shekhbeg on the right bank of the Panjah is the village of "Garm Chashma" (hot springs) where a large stream of warm water joins the main river. On the banks of this stream the Munshi saw 20 or 30 men employed in washing the sand for gold. They were Badakhshis, and farmed the washings of the Gharan district for Rupees 20 Of per annum paid to the Ruler of Badakh- shan. It is only within the last two years that gold has been found in this district. Three miles beyond this is the Kuguz Parin,J the boundary between Gharan and Shighnan. The road throughout the Gharan district lies along the banks of the Panjah, and is in places very difficult to traverse. The valley near Barshar contracts to about one mile in width, and the road runs over large boulders alongside the river, which flows between nearly perpendi- cular banks; the stream is narrow and swift being not more than 200 feet across, and is almost a continuous succession of rapids. Throughout the district the Panjah valley is nowhere much more than a mile in width, and is confined by very precipitous mountains ; the river is everywhere deep. In Gharan apricots of very large size and fine flavour are produced ; these are held in great repute in Badakhshan. Apples and pears are met with in abundance ; and but little grain is grown. There is abundance of grass and fuel to be found at the various camping grounds throughout the valley. The water of the Panjah is rarely or never used for * The villages in these countries are usually buUt of stone and mud. t About twenty pounds sterling. X Kuguz Parin in Shigbai dialect means " boles in the rock." Tbe Persian equivalent is " Rafak-i- Somakb." M ( 46 ) irrigation or for water mills. In the hot weather, oxen, horses, and sheep (for which the country is famous) are driven up side valleys to the tops of the mountains for grazing, returning to the valley in October in splendid condition. Kuguz Parin consists of a tunnel passing through a mountain. On the south side, the road rises by a winding stone staircase, for a height of about 200 feet, to the mouth of the tunnel, which is excavated through solid rock, and is about 100 paces in length, and so narrow and low that it is impossible for a loaded horse to traverse it. The tunnel is said to have been constructed some three hundred years ago. Where the road emerges on the north side the path is so narrow that a projecting mass of rock often precipitates animals into the foaming torrent beneath. The river is here about 150 yards in width, and flows some 500 feet below the mouth of the tunnel.''^ The Shighnis boast of this place as the natural safeguard of their country, and call it their " father." From Kuguz Parin the Oxus flows through the country of Shighnan, which extends for a distance of sixty miles down to the Darband Tower on the frontier of Roshan. This tower is situated on a high rock standing over the river, towards which it presents a perpendi- cular scarp of about 150 feet. The water beneath is very deep. The roadway winds roimd the tower,t and the ascent on both sides is very steep and difficult. The Shighnis call this place their " mother." It is a common saying in the country that if ever there should be a quarrel between Shighnan and Roshan, whichever State first seizes this tower will keep posses- sion of both countries. The river is here barely a gunshot across, and there is no path whatever on the other side. This country of Shighnan would appear to be richer and of much more importance than "Wakhan and other districts of Badakhshan with which we are acquainted. From Kuguz Parin to Darband Tower there are numerous villages scattered along both banks of the river. J These are surrounded with gardens, orchards and well cultivated lands. The chief town. Bar Panjah,§ is on the left bank, and with its suburbs probably contains about 1,500 houses. The palace is inside the fort, and is built of stone ; the windows have shutters outside as in Cabul and Cashmere. The fort itself is square, each side being about 500 paces in length. The walls are very strong, and about 40 feet high, built of clay, stone, and wood. There are five loopholed towers, but these contain no big guns. There is a garrison of about 400 soldiers, who are mostly armed with swords manufactured in the country itself, and with guns, said to be made by the Kirghiz, similar to those which are supplied to the Kashghar troops viz., heavy rifled weapons which are fired resting on the ground, the muzzle being supported on a prong attached to the barrel of the rifle. Lead and all the materials employed in the manufacture of gunpowder are found in the country. The valley at Bar Panjah is about four miles wide and contains a great many houses and gardens. The river runs in numerous channels separated by jungle covered islands. Short punt shaped boats, similar to those in Central India, are used at the ferry. In July and August, when there is much water in the river, all travellers have to cross at Bar Panjah, to the other bank, the road on the left bank being then impracticable. In its passage through Shignan the Oxus receives two considerable affluents on the left bank, the Shewa and Vacherv River. The former is crossed by a good bridge, and was about 25 yards in width, and unfordable, when the Miinshi passed in May. It flows from a lake in the Shewa Pamir, a favored pasture ground much frequented by herds of horses, sheep, and cattle from Badakhshan. The owners of these flocks are said to make payments to the King of Shighnan for the right of grazing there. The Vacherv River is about the same size as the Shewa stream, and joins the Panjah to the south of Bar Panjah. Along it lies a much frequented road from Shighnan, over the Shewa Pamir, to Faizabad. * This portion of tlie route is not improbably the Tangi Badascani of Benedict Goez. t At the tower was a guard of soldiers from Wamur, who examines the passports of all travellers. X The names of these villages and the distances apart will be found in the Appendix. § Or " above Panjah " so named from having been built originally over a stone similar to the one at Kila Panjah, which was supposed to bear the impress of the Panjah or palm of Hazrat Ali. ( 47 ) On its right bank the Oxus receives one very large river, the Suchan, formed by two large streams, the Shakhdara and the Ghund, which unite about half a mile before joining the Panjah. The two branches are of about equal size, and the united stream is about two-thirds of the size of the main river, which continues to be called the Ab-i-Panjah. The Suchan stream enters a few miles south of Bar Panjah. The valley opens opposite the junction to a width of about four miles, forming a beautiful well cultivated plain, with a good deal of pasture land, generally covered with horses and cattle from Bar Panjah, which place forms a most picturesque addition to the landscape, situated as it is on a white rock surrounded by trees and gardens, which extend uninterruptedly a distance of about two miles north of the fort. Both the Shakhdara and the Ghund Rivers have numerous villages on their banks. On the former at two days' march from Bar Panjah is the large fort of Rach^ the residence of the Governor (Hakim) of the Shakhdara district, which is said to contain about 500 houses. The Ghund valley, the chief place on which is Charsim, is said to contain about 700 houses. Roads lie up both these valleys to the Pamir steppes. The Pamir at the head of the Ghund valley goes by the name of Bugrumal, and is possibly a continuation of, if not identical with, the Ahcliur Pamir. The direct road to Kashghar up this valley is said to be a much easier road than that by Tashkurghan. At Sacharv, nine miles north of Bar Panjah, in the Shighnan valley, the river narrows and becomes turbulent and the road is very bad. Sixteen miles further on is the Darband Tower before described. Beyond it lies the territory of Roshan, a dependency of Shighnan, and ruled by the same King, Yusuf Ali Khan. Two and a half miles beyond Darband is the junction of the Murghab River with the Panjah. This is the river I have already traced from its source in the Lake of Little Pamir under the name of Ak-su. It is also said to carry away water from Lakes Karakul* and Rang-kul. The Panjah valley, which at Darband is very narrow, rapidly widens to five miles, and would be fit for cultivation, but that the ground is frequently flooded by the Murghab River. The Munshi crossed the river about two hundred paces above its junction with the Panjah ; the stream was in three channels,, and the torrent was so rapid that most of the horses lost their footing. The Panjah stream was very clear, but the Murghab was red, thick, and muddy. The volume of the latter was considerably larger and its velocity greater than that of the Panjah.f From bank to bank the width of the river bed is about one and a half miles, and of this at least one mile was covered with water. The passage was effected with great difficulty. In the summer floods the water is said to extend from mountain to mountain, a distance of not less than five miles ; it can then only be crossed by boats. This river is generally called the Murghab, but it is also known by the name of the Darya-i-Bartang, so called from the district of that name through which it flows. Three miles below the junction, on the right bank of the now united rivers, which still bear the name of Panjah, is Wamur, the chief town of Roshan. Wamur is a flourishing place ; a large FortJ about the same size as that of Bar-Panjah, is surrounded by several hundred houses and orchards. Fruits and grain grow in abundance, and the soil is very fertile. The Munshi remained several days at Kila Wamur, where the King of Shighnan was residing.§ He was enabled to visit thence the FatUa Sang,\\ which together with the ruby mines are described in the Sir^ as the two sights of Badakhshan. It is situated about three and a half miles down the river, underlying the mountains. He extracted * I am myself by no means sure as to whether this river does really receive any water from Lake Karakul. t This statement of the Munshi is confirmed by other sources of information which I possess. J With a garrison of about 200 men. § The King generally spends the winter months at Kila Wamur returning for the summer to Bar Panjah. II " The wick stone," probably asbestos. ^ The " Sir" is a book written by Moulvi Imam Afzal, Khorassani. ( 48 ) some fragments from the rock in situ and brought them away. They consisted of a sort of soft fibrous stone which can be twisted into the shape of a wick, and when satur- ated with oil will burn almost for ever."^ From the Fatila stone he went two miles further down the river to the village of Pigish, the fiu'thest point reached by him. At this point the Oxus, which from Ishkashim, a distance of about 100 miles, had been flowing due north, takes a sudden bend to the west, and going in that direction for a few miles turns apparently to the north. The Roshan territory is divided into three districts — the Waraur on the right bank embracing the upper portion of the Oxus valley, and containiug about 800 houses. The district of Pa-e-Khoja lies on the left bank of the Oxus, below the turn to the west before alluded to, and is said to contain about 1,000 houses. It is at a long day's journey below Wamur. This district is inhabited by Khojaks^ who pay no tribute, but give their services as soldiers in time of war. The third district is that of Bartang,J which lies up the river of that name, and is said to contain about 500 houses. The direct approach to this district from the Panjah valley is very difficult, owing to the precipitous defiles through which the river passes ; so that the most frequented road between Wamur and Sii-ich Fort, the chief place of the Bartang district, lies by the somewhat roundabout way of the Ghund valley. The Munshi gives the following particulars about Shighnan : — " The country of Shighnan and Roshan is sometimes called Zujan (or two-lived) ; its climate and water being so good that a man on entering the district is said to have come into possession of two lives. The inhabitants state that their country is called Lubnan in the Gulistan of Sheikh Sadi of Shiraz, and that it is by this name that their country is known in Persia. Sheikh Sadi writes that Hhere is one good Mussulman in Lubnan' (Ek-i- az-Sulhai Lubnan)." " In time of war, the two countries combined can produce 7,000 armed men, which allowing three men from each two houses would give a present total of about 4,500 for the number of houses in Shighnan and Roshan together.'' " The family of the Shah-i- Shighnan originally came from Persia. The first arrival from that country (said to have taken place about from 500 to 700 years ago) was the ' Shah-i- Khamosh,' who was a Syud and a fakir. The country was at that time in the hands of the Zerdushtis, a very powerful and learned race. The Shah commenced to teach these people the Koran. There were already at this time Mussulmen in the neighbouring country of Darwaz, A.H. 665, and on the arrival of the Shah-i- Khamosh many people flocked thence into Shighnan. In about ten years' time he had converted large numbers of the people, and a civil war commenced which ended in the Shah-i-Khamosh wresting the kingdom from Kahakah, the then Governor of Shighnan and Roshan, under the Zerdushtis, the seat of whose Government was in Balkh. After another ten years the whole of the people were converted to the Shiah religion. The tomb of Shah-i-Khamosh now exists at Bar Panjah. Every Thursday people meet to worship there." " The Chinese during their occupation of Kashghar used to pay to the surrounding countries a kind of subsidy, in return for which the States to whom the payments were made used to guarantee to keep the roads open, and safe for merchants. For this service the Shah-i-Shighnan used to receive an annual payment of ten Yam boos ;§ the ruler of Sarikol used to receive six ; the Kanjudis four ; and the ruler of Wakhan three." " At one period it is said that Wakhan and Darwaz and all the surrounding States were under the rule of the King of Shighnan." The Munshi did not succeed in bringing back much information about the course of the Oxus below Wamur. The furthest point down the river reached by him was Pigish, a village * Probably asbestos. t i.e., whose ancestors are Syuds on one side only. J Or " Above the Narrows." § A yamboo is a large piece of silver valued about Eupees 170 or seventeen pounds sterling. ( 49 ) four or five miles below Wamur. About five miles beyond Pigish on the right bank, is the village of Bar Roshan, on the frontier of Roshan. At one day's march beyond this, also on the right bank, is Waznud, the frontier village of Darwaz. Between Waznud and Bar Roshan the Pa-e-Khoja valley before alluded to enters the Panjah valley on the south. Five marches along the Panjah beyond Waznud is Kila Khilmb, the chief town of the coiintry of Darwaz. The road to it along the river is very difficult, and impassable for laden horses, the valley being very narrow, and the banks of the river very precipitous. Kila Khumb is said to lie in a northerly direction from Wamur and can be reached in three days by a short summer road, which lies up the small stream which enters the Panjah (right bank) at Wamur, The boundary between Darwaz and Shighnan is the water-shed at the head of the W^mur ravine. ^ The country of Darwaz possesses villages on both banks of the river Panjah. At Khumb itself the fort is on the right bank, but some houses are on the left bank of the river. Below Darwaz is said to be the country of Khatlan, the chief town of which is Kolab. Part of this information, which was supplied by the Miinshi as the result of enquiries made by him, is directly contradictory to the existing ideas of the geography of these regions, and I was at one time disinclined to place much reliance on it. In plotting on paper the Munshf s route survey, it was found that the course of the Panjah river lies much to the north of the position assigned to it in existing maps. I was at first unable to reconcile this with what little authentic information we have, either from English or Russian sources, but further consideration, and study of the subject, has to a certain extent cleared the matter up. While at Simla, immediately after my return to India, I was examining some old documents in the Quarter Master GeneraFs Office, and lighted on a paper of considerable importance. It was a copy of a document well known to students of Central Asian Geography, viz., the route from Khokand to Peshawur, by the Shahzada Sultan Mahomed, an Envoy who came from Khokand to India in 1854. The peculiarity of this individual copy is that it contains a marginal note that six of the halting places on the route, viz., " Faizbad, Naruk, Tootkul, Buljuwan, Kulab, and Surchushma, are on the banks of the River Hamoon, which is called Panj by the natives.^' In the margin was a note, bearing the initials of no less a person than the present Lord Lawrence stating that the Hamoon was the same as the Oxus. The significance of the document consists in the fact that Lord Lawrence (who was then Chief Commissioner for the Punjab) was the person who originally took down the statements of the Khokandian Ambassador, who, during his stay at Murree, " lived for some weeks within a few yards of the Chief Commissioner's residence, and had frequent and intimate intercourse with him." It is evident that the document I had lighted on was a copy of the original statement as taken down by Lord Lawrence, whereas all other versions that I have seen, of the same route, omit the note that the six places abovementioned are on the banks of the Oxus. The deduction that I made from this note was, that the Surkhab River probably joined the Oxus or Panjah somewhere above Faizabad and Naruk, and consequently that the latter river took a very considerable detour to the North, as is shown on my Preliminary Map. There were many arguments in support of this view, which it is now unnecessary to enter upon, as one of the Trans Frontier explorers, " the Havildar,'' has just returned from a visit to Kolab and Darwaz, and although there has not yet been time to plot the details of his work, sufficient is known to prove that the note to which I have alluded must be erroneous, and consequently the portion of my map which was mainly founded thereon is also erroneous. The true course of the Oxus will, I believe, be found to occupy a position intermediate between those shown on my map and on all preceding ones. Return to Yarkdnd via the Great Pamir. Our return route to Yarkand lay up the north branch of the Panjah River, which flows westwards from Wood's (Victoria) Lake through a portion of the Great Pamir. Leaving * Four miles above Wamur on this stream is a mine from which a rich iron ore (kurch) is obtained. At Bar Koshan also iron is found in large quantities. N ( 50 ) Panjah on the 26tli of Api-il (the day previous to which was the first warm day we had since leaving Yang-i-Hissar^ the thermometer in the shade going up to 74° and in the sun to 99°) we made a short march of only six miles to Langarkish (9,350 feet), the highest inhabited spot on the road up to the lake. We passed on the left the villages of Zang and Hissar, between which is a hot spring"^ (temperature 120°) enclosed in a stone building and said to possess valuable curative properties, for the sake of which the old Mir occasionally visits the spot. I may note that hot springs are of frequent occurrence in these mountains ; some near Patur in the Sarhadd valley have a temperature of about 160°. These springs have a sensible influence on the temperature of the rivers they flow into, a fact which tends to neutralize any argument (such as that used by Wood) that the relative elevation of the sources of the two branches of the Oxus, may be estimated from the temperature of the streams at their junction. Where the two Pamir streams meet opposite Zang, the united river was about 40 yards wide and one and a half feet deep, with a velocity of three and a half miles per hour. This measurement was taken at 11 a. m. at which period of the day" the river had not attained its full size and velocity. In the hot weather at Panjah it cannot be forded, but is crossed by rafts made of skins. Close by the village of Hissar (or Asshor) on a small isolated rocky hill, is the ancient fort (or kila or kalhai) of Zanguebar, which I examined in hopes of finding some relic of Zoroastrian worship. The ruined walls had, within memory, been used as dwelling-houses by the inhabitants of the neighbouring village, but 1 could discern no relics of antiquity, except fragments of a surrounding wall, and an arch formed by large slabs of stone resting, on either side, on solid rock. At Langar Kish, a very picturesque village, a fair sized stream from the north joined the main stream, passing through one of those characteristic fissures I have before alluded to. I tried to ascend it, but was very soon stopped by enormous boulders lying in the bed of the stream which flows between perpendicular rocky banks. From this village we had to take all our supplies for the return journey to Sarikol, and as collecting sufficient even for a rapid journey was found to be a matter of considerable difficulty, we had here reluctantly to give up all idea of halting on the road or making any detour for exploration. Our first march from Langar Kish was about 18 miles to Yumkhana (also called Jangalik) . The road follows the right bank of the river rising above it in several places as much as 1,000 feet. From both sides occasional small mountain streams help to swell the waters of the main river. We passed on our right several ruined huts formerly occupied by Kirghiz, who many years ago abandoned this part of the country. The descendants of the men who accompanied Wood on this same journey, driven away by the insecurity of life and property, are now many of them quietly settled, hundreds of miles away, in the neighbourhood of Kilian and Sanju, under the rule of the Amir of Kashghar. Not a single Kirghiz, I was given to understand, remains even under the nominal sway of the Mir of Wakhan. As we advanced the valley opened somewhat, and the mountains on the south appeared to decrease in height, radiating from a pointed peak situated between the two branches of the Panjah River. After a time we came to the Ab-i-Zer-i-Zamin, a stream flowing from the north-west through banks 1,000 feet in height. We had to descend to the bed of the stream, cross and ascend the opposite side, and then traverse a plain, formed by a broad terrace at the foot of the range on our left, and situate about 1,000 feet above the bed of the Oxus. Four miles after passing the Zer-i-Zamin River we reached our camp, where some springs and rich soil had combined to produce a profusion of grass and fire-wood. From our tents we had a very fine view down the valley, seeing in particular one very prominent snowy peak, probably 20,000 feet in height, situated near the head of the glacier opposite Panjah. Next day we continued along the right bank of the river, passing, after five miles, the Ab-i-Matz, along which * Curiously enough a cold spring with a temperature of 60° F. issues from the ground within a few feet of the hot one. ( 51 ) is the summer road to Shighnan * from the head of the Wakhan valley. This road crosses the Joshmgaz, a very high and lofty pass closed by snow throughout the winter and spring, and proceeds down the Shakh Darah (valley) to Kila Rach, the residence of the Hakim o£ the Shakh Darah District of Shighnan. From Rach a road continues down the stream to Bar Pan j ah. On our own road, two miles beyond Ab-i-Matz, is Boharak, an occasional halting place of caravans, stated by our guide to be the commencement of the Great Pamir. Here, the valley, hitherto half a mile across, widens into a large flat open plain, one and a half miles in width, said to have abounded in former years with the magnificent Pamir sheep (Ovis Poli). Of these we saw nothing but bones and skulls. Severe murrain has within the last few years carried off not only nearly the whole of the wild sheep, but also the ibex. Six miles beyond Boharak was our camp at Yol Mazar (road-side temple), two miles short of which is a large stream joining the river on its left bank, and of equal bulk with it. Near the camp a smaller stream entered on the right bank. I ascended this for some distance and found an open grassy valley in which there were some huts in ruins, and some obvious traces of former cultivation ; it was doubtless once the residence of Kirghiz. At our camp, which was at an elevation of about 12,000 feet above the sea, there was plenty of fire-wood and grass, this was the highest point in the valley at which good fire-wood was found, although further up and throughout this Pamir there was abundance of " boortsee" and grass. Two inches of snow fell at night, but the morning, though cold, was fine. We were now fairly in the Great Pamir ; the grassy valley, about a mile broad, was bounded by terraces formed by low spurs coming down in gentle slopes from the mountain ranges on both sides. On the 29th April we continued our march along the Pamir to Bilaor Bas. The road was excellent throughout, as in fact it was the whole way, from Panjah to Ak-tash, although at starting there are numerous steep ascents and descents. This day's march was along the right bank of the river, through a grassy plain, bounded on both sides by low undulating hills. The valley gradually widens, but the flat grassy portion is nowhere much more than a mile in width, the ascent was steady, and the road everywhere first-rate. Shortly before reaching camp we passed on our left the Ab-i-Khargoshi which flows from and through the Khargoshi plain, beyond which, at a day's journey from camp, is the Alichur Pamir, which nominally belongs to Wakhan, but practically to Shighn^n.f In it lies a small salt lake " Tuz-kuF' from which no water flows, and beyond which the drainage goes to Shighnd,n. Two days' march from this lake, i.e., three days from our camp, the Alichur stream is said to fall into the Murghabi. The Alichur Pamir is reported to be higher but smaller than the Great Pamir, and to possess roads going in every direction. On the 30th we continued along the Great Pamir for 20 miles to Mazar Tupa, the plain getting gradually wider and wider as we advanced, until a breadth of six miles is attained. The valley is not so well defined, as that of the Little Pamir, where steep mountains bordered the plain on both sides. Here low spurs from the mountain ranges north and south run into and are hardly to be distinguished from the plain. The mountains on the south are consider- ably higher than those on the north, the former rise to about 5,000 feetj and the latter to about 2,500 feet above the river bed, giving absolute heights of 18,000 and 15,500 feet, respectively. The next day five miles of very gentle ascent brought us to the west end of Wood's Victoria Lake, which, like its sister in the Little Pamir, was supposed to have two outlets. Of that to the west there could be no doubt; through a channel some 12 paces wide, a little stream 6 inches deep, and with a velocity of 2^ miles an hour, emerged from under the ice with which the lake was covered, and flowed steadily westward. The temperature of the water was 38°, and * Vide Appendix. Section Eoutes. t i.e., according to the statement of the Wakhis who accompanied us. J Elevations of hills were measured with an " Abney's clinometer," which I always used to carry in my pocket. Their positions were fixed by the intersections of compass bearings taken at different points on the line of march ; the heights of the latter were determined hypsometrically. ( 52 ) _ it was thus evident tliat the lake was partially supplied from warm springs. A few wild fowl were congregated near this end of the lake doubtless waiting for the rapidly approaching warm weather to melt the ice and enable them to proceed with their parental duties. The lake runs nearly due east and west, is about ten miles long, and nowhere more than two miles in breadth. The valley in which it lies is, opposite the lake, about four miles broad. The height of the hills to the north I estimated at 3,000 feet above the level of the lake, while those on the south were at least 2000 feet higher. The only name by which the lake is well known to natives is " Kul-i-Pamir Kalan,'' i.e., lake of the Great Pamir. I have once or twice heard it called " Airan KuV or buttermilk lake. To avoid confusion, and to make as little possible change in existing nomenclature, I purpose calling it "Kul-i- Pamir Kalan,^^ or "Victoria Lake," the last name being the one originally bestowed by its discoverer, Lieutenant Wood. Our camp, which was about two miles east of vits head was called by the " Wakhis " Sar-i-kul (head of the lake), a camp in a corresponding position at the lower end being called " Bun-i-kul " (foot of the lake). This may account lor the other name erroneously given to it by Lieutenant Wood (Sir-i-kol). ? After reaching camp, a distance of 16^ miles, I went to the head of the lake to investigate ' its drainage and determine its limits (for from a little distance off it was impossible to discriminate between the ice and snow on the lake, and the snow on shore) . I was soon convinced that all the water from the hills at the east end drained into tbe lake, which therefore like its neighbour in the Little Pamir has but one outlet, although in the former case the water flows west, and in the latter east. To the East of the lake the valley opens out, and forms a large basin which extends ten or twelve miles from West to East, and six miles from North to South. At the lower portion of this basin, surrounding the head of the lake, is a great deal of marshy ground formed by the drainage which enters from numerous side valleys,^ coming from the hills on the South. At the time of our visit this marsh was covered with snow and ice ; but later on in the season, when the snow is melting on the surrounding hills, there is much water, and the place is said to become the favoured breeding place of thou- sands of geese. Our march from Sir-i-kul lay along the Northen side of the valley, the whole of which was deep in snow, and was so level that I experienced considerable aifflculty in determining the correct position of the water-shed, which was crossed at a distance of twelve miles from the east end of the lake and at a height of 14,320 feet. A frozen stream here comes down from the North, divided into two portions by a low ridge of gravel, one flowing eastward into the Aksu River, the other westward into the lake. Eastward from the water-shed the Great Pamir valley contracts. We followed down a rivulet which, shortly before passing the camp at Shash Tupa, joins a considerable stream coming down a broad valley from the North. The name of our camp was derived from the " Shash Tupa" or " six hills" by which it is surrounded, and between each pair of which roads issue to different parts of the Pamir steppes. Our road from Shash Tupa lay for nearly eight miles due north on the right bank of the stream, and then continued down it for ten miles in a north-east direction to the camp " Dahn- i-Isligh."t Oil our left we passed three broad open ravines, containing streams coming from the west ; one of them was nearly as large as the river we were following, and before joining * "Up one of these valleys is a road ac^ross the hills to Langar in the Great Pamir. One good day's march takes the traveller over the Warram Kotal (Pass), another half day to Langar, and another half day to Sarhadd (horse marches). This is the road by which a very short time ago Jehandar Shah, the Ex-Mir of Badakhshan, when attacked by the Cabul troops, fled, accompanied by several hundred followers, to Yassin. His shorter route from Panjah would doubtless have been up the Sarhadd valley, but anticipating that he would be intercepted on that line, he made the long detour above mentioned, passing through uninhabited country the whole way from Langar Kish, and striking the Little Pamir at a considerable distance above Sarhadd, instead of having to fight his way up to tbe latter place. t •' Mouth of the Isligh." ( 53 ) it passed through a plain some six miles long and two broad. At Dahn-i-Isligh the river is joined by two more streams, the Kizil Robat coming from the south-east, and the Karasu from the west, both of which pass through broad grassy valleys. The ground is very open, and may be traversed in almost every direction. Two or three miles north-east of our Camp the Great Pamir terminates, having extended for a distance of some 90 miles from Boharak. From Dahn-i-Isligh I took a path which follows the Isligh stream, until it emerges into the Aksu plain ; this road is somewhat circuitous, and the rest of the party took a shorter line, going over a low pass, and rejoined the main stream about 1 6 miles from our starting point. The path I followed is rarely used by travellers ; in summer it is quite impassable on account of floods. When I went down it (in April) the ice was breaking up, and travelling was somewhat dangerous, as the river had to be crossed many times. The hills on the north are very precipitous, and in places rise nearly perpendicularly to a height of some 2,000 feet above the river bed. Where the two paths unite, the valley opens, and down it a good road leads to the Aksu plain, which is crossed diagonally in a S.-E. direction. Prior to reaching our camp at Ak-tash we had much difficulty in crossing the Aksu River,' which was much swollen by melting snow. On this march (37 miles in length) I had the good fortune to shoot an " Ovis Poli,'' the only one that has fallen to the rifles of our party. At Ak-tash we rejoined the road we had followed on our outward journey, and returned by it, to Tashkurghan and Yarkand making the slight variations in our route, to which I have already alluded. It appears from the foregoing narrative that although the name Pamir has been inaccurately employed as a generic term covering the whole of the elevated mass lying between the Hindu Kush and the mountains of Khokand^ yet it is rightly applied to some of the steppes which occupy a large portion of this region. These steppes would appear to be a series of broad undulating grassy valleys, formed on the surface of an elevated plain, by lofty ridges running more or less parallel to the equator. The general slope of the plateau is from east to west. Its eastern portion is gently undulating, and comparatively flat, while its western edge merges into spurs, which slope down gradually to the west, and are separated by bold and precipitous defiles. On the east the Pamir steppes are bounded by a transverse ridge, which has been appropriately termed the Pamir range by Pundit Manphul. This ridge runs in a direction from south-south-west to north-north-east and is the true watershed between Eastern and Western Turkestan ; at the Neza Tash Pass where we crossed it, the watershed is very clearly defined; the ridge was seen trending as far north as latitude 38° 15' ; it appeared to sink gradually, and I was informed by Kirghiz that it eventually subsided to the level of the Kizil-art plains a little short of the Great Karakul (lake) in which vicinity a difference of level of a few feet may probably determine the flow of water, either into the Sea of Aral in Western Turkestan, or into the semi-mythical lake of Lop, on the confines of China. To the east of the Pamir range there is an extensive plateau, which stretches from the Muztagh range of the Himalaya mountains up to the South Khokand range — the Trans-Alai ' of Fedchenko — in the parallel of 39^°. Portions of it are designated in order from south to north as the Taghdumbash Pamir, the Sarikol or Tashkurghan valley, the Tagharma, and the Kizil-art plains. This plateau is in turn bounded on the east by the range to which Hay ward gave the name of the Kizil-art, the name by which it is known to the inhabitants of Kashghar, and which runs nearly parallel to Pandit ManphuFs Pamir range. Fedchenko has questioned the existence of the Kizil-art range m the following words : — " Hayward's researches seem to point to a meridional range to the west of Kashgar, but he only saw these mountains in the distance, and covered entirely with winter snow, which is very misleading as regards direction. Therefore his statement regarding a meridional Kizil-art range with steep easterly declivities appears to me very untrustworthy. When you have the ends of a chain facing you they appear, when covered with snow and seen en face to form a consecutive chain running in a direction perpendicular to the line of sight of the beholder.^' But all the information I have obtained decidedly corroborates Hayward's views, which are aJso shared by Mr. Shaw. I have every reason to believe that the magnificent line of snowy peaks which is viewed from Kashghar, constitutes a meridional chain of mountains, instead of being composed O ( 54 ) of the tail ends of a series of longitudinal chains. It is broken through nearly at right angles by the Yamunyar river, which brings down the drainage of the Little Karakul Lake and the contiguous portion of the Kizil-art plain, just as the Kuen Luen and many of the Himalayan ranges are broken through by rivers whose sources are in the upper table-lands. The positions of several peaks of the Kizil-art range were fixed by numerous bearings, taken from points along the road between Yarkand and Kashghar, and the four most conspi- cuous ones, embracing a length of 52 miles were found to lie almost exactly in one straight line having a direction of about 30° west of the true meridian. The most southerly and the highest of these, the Tagharma peak "^ of Hayward I ascertained trigonometrically to be 25,350 feet above sea level, while two others are at least 22,500 feet high. From the Tagharma peak southwards the range diminishes very much in height. On our return journey we crossed the Chichiklik mountains (which may be considered as a continu- ation of the same range) at the Kok-Mainak Pass at an elevation of 15,670 feet; whilst further south the same mountains are pierced by the Tashkurghan river at a height of about 10,000 feet. Little is known of the range further south, but it would seem to be a connecting link with the Himalayan ranges so that the old Chinese geographers, who did indeed link together the " Bolor" and the " Karakorum^^ under the common name of " Tsung Ling" or " Onion mountains" were not far wrong in their ideas, I am inclined to agree with Mr. Fedchenko in considering the Pamir steppes, within the limits by which I have defined them, to be a portion of the Thien Shan. At all events they present a very similar physical formation, the main feature of which is the existence of ranges situated on a high table-land, and running more or less east and west. We have already seen that in the only portion of the Thien Shan system visited by us, i.e., to the north and north- east of Kashghar, the mountains consist entirely of parallel ranges having an easterly and westerly direction, and that the elevated plain on which they are situated rises rapidly higher and higher as it advances northwards. It is not always easy to detect the parallelism of these ranges. On the expedition to Chadyr Kul, where we continuously ascended the bed of the Toyanda stream, I did not fully realize the fact, and it was only after our subsequent journey towards Ush Turfan, where I had an opportunity of penetrating and crossing no less than four of these ranges, that I was convinced that this southern portion was of the same physical configuration as other portions of the Thien Shan as portrayed on the Russian maps. Fed- chenko, proceeding apparently solely on the basis of this theory of the parallelism of ranges, has shown in his last map the country north-east of Kashghar t in much the same way as I have myself done, and he would doubtless have been much gratified, had he lived, to find his theories so soon verified. An examination of the map accompanying this report will show the ideas I have myself formed of the ground lying between the Great Pamir and the Alai plateau, which last has been visited by M. Fedchenko. The position and extent of the Great and Little Pamirs have been accurately laid down, and it is hoped that the mapping of the ground between them and the Alai will be found to be not very far from correct ; the geographical detail shown is . the result of careful study. On the construction of the Preliminary map accompanying this report. The positions of all places in Eastern Turkestan, and Wakhan^ that have been visited by members of the Mission, depend upon the astronomically fixed position of the Yangi-shahr or new city of Kashghar, for full details of which the appendix. Sections A. and B., may be consulted. * The altitude and bearings of this peak I measured with great care, with my theodolite, from both Yapchan and Kashghar, and I thus obtained two independent results of 25,364 feet and 25,328 feet. f M. Fedchenko was never there, and, as far as I am aware, the Russians possess no survey of the ground to the north-east of Kashghar. ( 55 ) 6' 8" + 6' 8" 76° 12' 55" 1° 3' 0" The final positions in longitude of Yarkand and other important places have been deter- mined as follows : — The true longitude of KASHGHAR (Yangi-Shahr) is ... ... ...' 76° 6' 47" The difference of longitude between Kashghar and Yangi-Hissar as determined by Pandit Kishen Sing's pacing, corrected from latitude observations, is — On outward journey ... ... ... -|- 0° 6' 15" On return journey ... ... ... +0° 6' 0" Giving for longitude of Yangi-Hissar The difference of longitude between Yangi-Hissar and Yarkand, determined in the same manner — By outward journey is ... .'.. ... 1° 3' 0" By return journey ... ... ... 1° 4' 25" On the outward journey the survey was carried along the direct road, about 75 miles in length, and over a perfectly level country, whereas on the return journey the road followed a circuitous line of 180 miles, over one snowy pass and very rough ground. The first value is therefore accepted in preference, viz. Giving a final value for YARKAND (Yangi-Shahr) of .._. ... 77° 15'' 55" which is 0° 3' 5" in defect of the astronomically determined value of the same place. I have determined to accept the value as deduced from Kashghar in preference to the independent results arrived at from observations to the moon. # Again, the final longitude of Yangi-Hissar (as above) is . . . 76° 12 The difference between Yangi-Hissar and Tashkurghan by Pandit's pacing corrected for latitude is ... 63' 25" The difference ascertained chronometrically by Captain Trotter is ... ... ... ... 54' 23 Giving a final value for TASHKURGHAN of ... ^ 75° 19' 1" which is 4' 59" in defect of the value obtained from one night^s observations to the moon at the same place. The longitude of Kila Panjah (Wakhan) was determined chronometrically — 1. On outward journey, from Tashkurghan ... 72° 44' 18" 2. On return journey, from Igliiz-yar (near to and con- nected with Yangi-Hissar by a traverse survey) .. . 72° 46' 40" Givingafinallongitudefor KILAPANJAHof... 72° 45' 29" Whilst the observations for absolute longitude at the same place give a result of ... ... _ ... 72° 45' 30P aod a fourth entirely independent result obtained by Captain Trotter's route survey, corrected for latitude is 72° 44' 10" The mean result obtained chronometrically is adopted for the final position. The wonder- fully accordant results at Kila Panjah, although highly satisfactory, must perhaps to a certain extent be regarded as fortuitous, but the admirable rates obtained for the watch employed in the chronometric determinations, a silver lever watch by Brock of London specially made for explorations, are worth recording* and ought to give results in the accuracy of which great confidence may be placed. * Travelling rates obtained hy Captain Trotter for Brock's lever watch. No. 1602, during journey from Yangi-Hissar to Kila Panjah, and return journey to Yarkand. "J 55" mean 53' 54" SlAGB. Yangi-Hissar to tala. Ak- Ak-tala ghan. to Tashkur- Dates. 18th to 22nd March 22nd to 31st March No. of days from which rate was determined, Rate per diem gain- ing in seconds of time. -f 6. + 6. 1 BE1IA.BKS. (1) Rate obtained by comparing difference of observed times with difference of longitude as derived from Pandit's pacing, corrected for latitude. (2) Ditto Ditto. ( 56 ) I am much gratified to be able to state that after all my computations were completed and the details of routes transferred for the first time on to a correct g-raticule^ my position of the west end of Victoria Lake (the extreme east point visited by Wood in his travels) was latitude 37° 27' north and longitude'^ 73° 40' 38", which is practically identical with the independent determination of the same point by Lieutenant Wood which is given at page 233, new edition of Wood's Oxus, with essay by Colonel Yule, London, 1872. I will now indicate how the positions of points on the road between Leh (Ladakh) and Yar- kandhave been determined. The position of Ak-tagh (2nd camp) was fixed by myself in lat. 36° 0' ll'' and long. 78° 6' 20". It was the converging point of three different route surveys (by Pandits) starting from fixed points on the south, and is in the neighbourhood of a hill above Chibra whose position was satisfactorily fixed by intersection (on the plane-table) of several rays from trigonometrically fixed peaks of the Karakorum. The position of Ak-tagh in long- itude with regard to these peaks may be looked on as correct within a mile, and its position in latitude is undoubtedly correct within a few hundred feet. From this point three traverse lines have been carried by different Surveyors to Karghalik, which, when corrected and adjusted on the proper parallel (37° 53' 15"), had a maximum diver- gence of 3| miles, the mean of the three values gives a position in (true) t longitude of 77° 25' 30." Between Karghalik and Yarkand I had also two independent traverses, i.e., on both outward and return journey, which differed from each other in the resulting longitude of Karghalik by less than a mile. The mean of these two when referred to the value of Yarkand as determined from Kashghar places Karghalik in longitude 77° 28' 30.^' A mean between Travelling rates obtained by Captain Trotter for Broch's lever watch, Sfc, Sfc. — (Concluded). SlASB. Dates. No. of days from which rate was determined. Rate per diem gain- ing in seconds of time. Reuabks. Yangi-Hissar to Wa- khan and back to Ighiz-yar. Kogachak to Ak-tash. . . Kashghar to Ighiz-yar 18th March to 18th May. 3rd April to 5th May 15th to 18th May... 61 32 3 + 6. 1 + 6. 7 + 6. 6 During these 61 days almost an entire circuit was made. The diiference of longitude between Yangi Hissar and Ighiz-yar, viz., V 45" only, was determined by Pandit's pacing. During these 31 days a smaller circuit was made ; the difference of longtitude between Kogachak and Ak-tash is 1' 35". In both these circuits allowance has been made for the station- ary rate (+ 7"8) obtained during our halt in Wakhan. Rate obtained in same manner as (1) & (2). It should be noted that my watches and chronometers were always carried in a small box that I had specially made for them, carefully packed in cotton wool, and inserted in the middle of a large leather mule trunk, packed with clothes. They were thus kept at a tolerably uniform temperature and escaped in great measure the jei-ks and shakes they would otherwise have been exposed to. Of my pocket chronometers, having a regular chronometric escapement, one by Peter Birchall, London, No. 1096, was well suited for astronomical observations, keeping excel- lent time when stationary and beating half seconds very audibly. It was always used by me in my astronomical observations, but it required very careful handling, as a violent jerk was apt to make it gain several seconds suddenly. A third watch, a pocket chronometer, by Dent, unfortunately got out of order before the Pamir trip, but I had found that while travelling, neither its rate nor that of Birchall compared favorably with that obtained from Brock's watch. It is perhaps needless to add that my watches were daily carefully compared together, and also both before and after observations of stars. An omission to do this on a single occasion prevented my getting a chronometric value for the differences of longitude between Yangi-Hissar and Kashghar. * The position in longitude in the " Preliminary map" differs slightly from this, as the latter had to be prepared prior to the completion of the computations. t True, i.e., depending on the most recent determination of the longitude of Madras. All the Indian Survey maps are based on the astronomically determined position of the Madras Observatory. Recent observations have shown that the old value, that is the one adopted by the Survey Department, is about 3 miles too much to the east. In my map I have been compelled to make allowance for this, and have shifted three mUes to the west, the whole of the positions in Northern India taken from the existing maps. ( 57 ) this and the value previously deduced from the south gives 77° 27' 0" which has been assigned as its final position. The smallness of the amount of the adjustment necessary to connect my own work, depending on my own astronomical observations at Kashghar, and that depending on the Indian Survey derived from the astronomically fixed position of Madras is a gratifying proof of the general accuracy of the work. This sketch would be incomplete without a few lines as to my connection on the north with the Russian Survey, which appears, I think, equally satisfactory with the above. The only position in the Amir of Kashghar's dominions in Eastern Turkestan astronomically fixed by the Russians is that of Kdshghar. This was done in 1872, the year prior to our own visit, by Colonel Scharnhorst of the Mission under General Baron Von Kaulbars. A comparison of results is given : — Position of Yangi-Shahr (Kashghar) determined by English Mission, 1873 — Latitude ... ... 39° 24' 26" North. Longitude ... ... 76° 6' 47'' J^.ast of Greenwich. Position of Yangi-Shahr (Kashghar) determined by Russian Mission, 1872 — Latitude ... ... 39° 24' 1 6" North. Longitude ... ... 76° 4' 42" East of Greenwich. As the quarters occupied by the British Mission, where the observations were made, lies outside and to the east of the fort, while those occupied by the Russians were in about the same latitude and nearly one mile. to the west of the fort, the difference in longitude is reduced to about one mile, our latitudes being practically identical. I would have wished to take the mean between the two as the final position of Kashghar, but as our stay there was of much longer duration than that of the Russians, and I had opportunities of taking many more observations than they did, I prefer leaving my own values intact.^ The slight discrepancy now noticed disappears on the road between Kashghar and Chadyr Kul, the only line of survey common both to the Russians and ourselves, and along which I carried a rough traverse survey in which the distances were estimated by the time occupied on the line of march. Prior to my departure from India Colonel Stubendorff, of the Russian War Office, had sent to Colonel Walker, the Superintendent of the Great Trigonometrical Survey, the positions of a number of points in Russian and in Khokandian territory that had been astronomically determined by Russian officers. Amongst them was the north-east corner of Lake Chadyr Kul. Bearing this in mind, when at the most northerly point on the road reached by us, I took a bearing tangential to the east end of the lake, which lay nearly due north at a distance of about three miles from us. On my return to India when I plotted in my work from my own astronomical position of Kashghar, I found that by adopting the Russian value of the east end of the lake, viz., latitude 40° 43' north, our positons in longitudef of the same point exactly coincided. In determining the position of Khotan I have made use of Pandit Kishen Sing's route from Karghalik to Khotan, and thence via Keria back to Ladakh. As a result of this route survey our previously accepted value of the longitude of Khotan has been altered by more than thirty miles. It may appear bold to make this extensive change in the position of a place that has been visited by a European explorer (Mr. Johnson), but the route survey executed by this Pandit is so consistent, and the plotted results agree so closely with the observed latitudes throughout the whole of his work, that I have no hesitation in accepting it as correct. I may further add that I have been in communication with Mr. Johnson on the subject, and that he freely admits the possibility of a large error in his longitude of Khotan. ♦ Since the above was written Colonel Walker has heard from Colonel Stubendorff that the Russian astronomical observations at Kashghar wbich were taken by Colonel Scharnhorst were referred to the most northern angle of the Yangi-Shahr, a position almost identical in latitude with my own, and differing by two-fifths of a mile only in longitude. Colonel Stubendorff mentions that the Eussian observations depend on the eclipse of the sun on the 6th June 1872, and that corrections for error in the lunar tables have not been applied. This last remark applies to my own observations also. — H. T. t 75° 24' East of Greenwich. ( 58 ) He states that in commencing his reconnaissance from the Kuen Luen Mountains (which he carried on with the plane-table only), one of the three trigonometrically fixed points on which his work was based, turned out subsequently to have been incorrectly projected on his board. This, together with the doubt that must always exist when rapidly passing through an unknown country as to the identity of the different peaks visible from the line of march, is quite sufficient to account for the discrepancy. In my preliminary map I have assigned to Khotan a longitude of 79° 59' instead of 79° 26,' the position it has recently occupied on our maps. About its latitude there can be no doubt. Mr. Johnson took several observations there with a 14-inch theodolite and obtained a mean result of 37° 1' 35", whilst from Kishen Singes observations with a sextant extending over nearly a month we have a mean result of 37° 7' 36". The points east of Khotan, i.e., Keria and the Sorghak gold fields, are derived from Kishen Sing's route survey, combined with his latitude observations. We also have from the same source a complete survey for the first time of the road via Polu to Noh, and thence to Leh. As a specimen of the accuracy of this Pundit's work I may mention that when the road from Karghalik to Pal, a distance of 630 miles, was plotted out on the scale of 2,000 paces to the mile, without any correction or adjustment whatever (although 4^° were added to each magnetic bearing in order to allow for magnetic variation) starting from my own value of Karghalik, the plot closed at Pal (fixed by the Great Trigonometrical Survey) almost absolutely correct in latitude and only eight minutes out in longitude, and in no single portion of the whole route, which passes over elevations exceeding 17,000 feet in height, did the plotted value differ by as much as three miles from his own observed astronomical latitude {vide Appendix Section A.) . Of this discre- pancy of eight minutes in longitude it is possible that a portion may be due to error of position in the starting point (Karghalik), but it may be noted that the amount is no more than would be accounted for by an error of 14° in the assumed value of magnetic variation. It is not to be supposed that such accuracy is generally attainable, but in the present case, although the surveyor laboured under certain disadvantages from the absence of inhabitants, yet there were the compensating advantages that he was under no necessity for concealment ; he was there- fore able to take and record bearings when and where he pleased. As regards the work executed to the north-east and east of K^shghar; the position of Maralbashi, on the road to Aksu, was fixed in latitude by Captain Biddulph [vide Appendix Section A.), and its position in longitude is roughly determined by a few bearings, and esti- mated distances taken by him on the road from Kashghar. On the road to Ush Turfan I carried on a rough route survey wherever I went, and took observations for latitude and obtained chronometric determinations of longitude as far as Ui Bulak, in latitude 40° 26' north and longitude 77° 36' east. Thence by route survey I got a deter- mination of the position of the Belowti Pass ; calculating from this the probable position of Ush Turfan I place it about three-quarters of a degree to the east of the position given it in the last edition of Colonel Walker's Turkestan map. On examining the latest Russian map (1873) it appears that the position of Ush Turfan has been recently altered, and placed very near where I would myself locate it. I have therefore in my map adopted the last Russian values of Ush Turfan, Aksu, and all places to the east. It will be found that the cities of Aksu and Kuldja are more than twenty miles to the east of the places assigned them in all but the most recent maps. The details inserted to the north of the map are taken almost exclusively from the Russian topographical map of Central Asia (corrected to ] 873). The portion of ground to the south of Khokand, visited by Mr. Fedchenko, is derived from various maps purporting to be by that distinguished traveller, amongst others, one recently sent by Madame Fedchenko to Colonel Walker, differing materially from all others that I have seen. For the country between the Alai, visited by Mr. Fedchenko on the north, and the scene of our own explorations in the Pamirs on the south, the map is compiled from all the limited sources available* which have * Including the route survey carried on by Abdul Subhan from Panjah to Kila Wamur, the chief town of Roshan, and also including a sketch map prepared by Colonel Gordon and Captain Biddulph representing their joint ideas of the geography of the Pamfr. ( 59 ) been bound together to the best of my ability. I am by no means yet satisfied with the result, and one of my first labours, when I have finished this report, will be the preparation of a map on a larger scale of the Pamir regions, when I doubt not that further considerations will induce some changes in the map as it at present stands. Most of the details to the south of the map, with the exception of those portions north of Leh that have been traversed by members of the Mission, have been taken from the last edition of Colonel Walker's Map of Turkestan, but all the positions in the latter have been shifted three minutes to the west in longitude in order to allow for the most recently determined value of the longitude of Madras, viz., 80° 14' 19.5'" East of Greenwich. In the portion of country traversed by Members and Attaches of the Mission use has been made of all the material collected by them. The maps of Messrs. Shaw and Hayward have also been called into requisition. The reductions of the astronomical observations, and the computations of heights, have all been made in the Office of Colonel Walker, R.E., the Superintendent of the Great Trigonometri- cal Survey, in whose Office also, the map compiled by myself has been drawn and photozin- cographed. A large amount of work has been got through in a moderate space of time, and I am deeply indebted to Colonel Walker for the facilities he has given, and to Messrs. Hen- nessey, Keelan, and Wood in the Computing Office, and Messrs. Atkinson and Sindon in the Drawing Office for the assistance afforded by them in their several departments. Meteorological Observations. Whilst on the march I always took readings of thermometers, and barometer or boiling point thermometer, at our camps, and on high passes, and at other places of interest. These were taken chiefly for the purpose of determining the height above sea level of the stations of observation and where used for this object are shown in Appendix C. Where they are not required for this I have not published them, as isolated observations at different places, taken under constantly varying conditions, are not of much use to the meteorologist. While I was at Leh Mr. R. B. Shaw, the British Joint Commissioner, commenced a regular meteorological registry, which has since been continued under the superintendence of Captain Molloy, the recorder being the Native Doctor attached to the dispensary there. At my special request Mr. Shaw kindly took extra barometrical observations at the hours of 9 a.m. and noon, whilst I was on the journey to, and during my residence and travels in. Eastern Turkestan, these being considered the most likely hours at which I should myself be able to take barometrical observations for height on passes and in camp. I have thus throughout the whole of my absence from Leh got almost simultaneous readings at the fixed Observatory of Leh, whose height has been accurately determined by the Great Trigonometrical Survey. This circumstance combined with the use by myself of mercurial barometers, enable me to compute the height of the various passes and halting-places with an amount of accuracy superior to anything yet attainable. It is hardly necessary to add that I have made at Leh, both on the outward and return journeys, numerous comparisons between my own instruments and those in use at the Leh Observatory, and that my own were previously compared with the standards at Dehra Dun (and some of them at Kew) . While on the march to Y^rkand I succeeded in taking numerous sets of observations with a Hodgkinson's actinometer. I took these at the special request of Mr. J, B. N. Hennessey of the Great Trigonometrical Survey, who supplied me with an instrument belonging to the Royal Society. They were chiefly taken at considerable altitudes, but owing to cloudy weather the Chang La (Pass), 17,600 feet above sea level, was the only very high point at which I was able to take sets extending over a period of several hours in the middle of the day. These actinometric observations have been handed over to Mr. Hennessey (now in England) for reduction, and they ought to give very interesting results, which will probably be communicated to the Royal Society. ( 60 ) At Leh in Ladakh advantage was taken of the presence o£ the Pundits to get a series of continuous hourly observations of the barometer. These extended over a period of six days during which there was no break in the observations. The height of the Observatory above the sea level (11,530 feet), coupled with the extreme dryness of the air of Ladakh, and its position in the interior of a large continent, combine to render the determination of the diurnal curve of considerable value.* In the diagram accompanying the vertical scale is ten times that of the natural scale, the exaggeration being necessary in order to show clearly the curve. The actual barometer readingsf have been corrected for Index Error, and reduced to a temperature of 32°, before being projected on the diagram. The curve indicates two maxima, viz., at 1 a.m. and at 8 a.m., and minima at between 2 and 3 A. M. and between 4 and 5 p. m., which differ considerably from results obtained in other parts of the world. The daily maximum and minimum is very much more clearly marked than the nightly one. At Yarkand also, during the winter, sets of continuous hourly observations were taken on the 20th, 21st, and 22nd of each month. Several of these sets have been reduced to a curve, which I have also shown in the same diagram. J It so happens that on the days that were selected for hourly observations at Yarkand there was almost always a steady fall in the barometer, as will be seen by a glance at the monthly curves in the other plate. In order to allow for this constant fall, the effect of which is to distort the true daily curve, I have applied proportionate corrections, so that the dotted line represents the true diurnal curve. The mean of six days^ hourly observations, viz., on 20th and 21st of December, 20th and 21st of January and 20th and 21st of February, have been employed in constructing the carve. It will be observed that at Yarkand the night maxima and minima are much more clearly marked than at Leh, but that there is much less difference between those of the day and night. The maxima occur at 10 a.m. and 11 p.m., the minima at 4 a.m. and 3 p.m. At Yarkand, where the Pandits passed the winter, meteorological observations were commenced on the 19th November and continued without a break until the 15th March. They consisted of the readings at 9 a.m., noon, 3 p.m., 6 p.m., and 9 p.m., of a mercurial barometer, an aneroid barometer, dry and wet bulb thermometers, and direction of the wind {N.B. — There was no rainfall, but a little snow fell in March) ; also the maximum and minimum temperatures in the shade during the 24 hours. At Kashghar observations were commenced on the 12th December, but were not so complete or regular as those at Yarkand, as I had fewer observers to assist me, and I was myself absent for two periods, viz., from 31st December to 11th January, and again from 15th February to 3rd March. Observations were continued up to the 16th March, and generally consisted of readings of two aneroid barometers at the hours of 9 a.m., noon, and 3 p.m , and occasionally at 6 p.m. Readings of thermometer (dry) and direction of wind were taken at the same hours, besides the maxima and minima during the 24 hours. Readings of the wet bulb thermometer were also taken during the latter half of February and March. In addition to these a series of hypsometric observations were taken, with the object of deter- mining the relative heights of Yarkand and Kashghar. The whole of these observations are shewn in the Appendix, Section G., to which the reader is referred. I have prepared (plate 2) a set of curves showing the connection of the barometric wave between the stations of Kashghar, Yarkand, Leh, and Dehra Dun (at the foot of the southern slopes of the Himalayas) . The curve represents the height of the corrected read- ings of the barometer at 9 a.m., during the four months for which I was able to obtain data, with the exception of Dehra, where 9 a.m. readings not being forthcoming I have taken the observations recorded at 10-30 a.m. * The Schlagintweits took hourly observations at Leh dm-ing the day, and interpolated values for the night hours. The results thus obtained cannot have anything like the same value as those derived from observations taken throughout the 24 hours. t The instrument employed was a mercurial barometer by Casella. J These diagrams have been drawn by Mr. Keelan of the Great Trigonometrical Survey, who has also rendered me great assistance in the preparation of the Appendices of this report. * >« 1^ ^1 ^1 ^1 ' ^^ ( 61 ) Magnetic Observations* As has been mentioned in a previous portion of this report it was decided not to take from India a complete set of magnetic instruments. I took with me however a dip circle by Barrow with which observations for inclination were taken at Leh (in Ladakh)^ Chagra (in Ladakh), Ycirkand^ Kashghar^ and Tashkurghan (in the Sarikol District). It has been laid down as an axiom by General Sabine, the great authority on matters of magnetism, that "the value of each new magnetic station is directly proportional to its distance from those where obser- vations have already been made/^ and I may therefore hope that the record of results, vide Appendix, Section E., may prove of considerable value, as I am not aware that any magnetic observations have ever been taken within a very considerable distance of any of the three last named stations. The rules laid down by General Sabine were rigorously adhered to in taking the observations. Observations for magnetic variation (declination) were taken at, and have been computed out for sixteen stations in, Ladakh, Turkestan, and AYakhan. The station furthest to the north-east was at Ui Bulak (latitude 40° 20' and longitude 77° 36') where the variation was 5° 40' east; the extreme western station was at Kila Panjah (latitude 37° longitude 72° 45') where the variation was 4° 17' east. Details of the results, which are very consistent inter se, are given in the Apj)endix. The instrument employed in the determination of declination was the six-inch transit theodolite, which has been described in the Appendix, Section B. It was fitted with a first-rate magnetic needle. Comparisons taken at Leh and at Dehra both prior to the departure of the Mission, and subsequent to its return, prove that no alteration has taken place in the position of its magnetic pole. The object observed was generally Polaris. In many instances, however, the sun, either near sunrise or sunset, was made use of; at important stations, such as Leh and Kashghar, the mean of several different independent determinations has been taken. I cannot conclude this report without alluding to the sad loss we have all sustained by the recent and sudden death of our much lamented comrade. Dr. F. Stoliczka. Having been in almost daily intercourse with him from the day of leaving the Punjab on our outward journey (through the Changchenmo, Chakmak, Artysh, and Pamir trips) up to the day of his death, and being naturally especially attracted to him as working always cordially with him to add my mite to the field of science, I most bitterly regret his loss. It is hard to think that he should not have been spared to give to the world the results of his laborious investigation and scientific research, and although he has left valuable notes behind him, yet owing to his unequalled knowledge of Himalayan geology there is probably no man living as competent as he was to do full justice to them ; and it is unlikely that any one will go through his valuable zoological and other collections with the same minute care and attention that he would himself have bestowed upon them. I have special reason to regret the absence of his experience and advice while preparing my own report, in writing which I had confidently looked forward to receiving the benefit of his assistance. (Sd.) Henry Trottee, Capt, R. E. Q GEOGRAPHICAL APPENDIX SECTION A. ( 2 ) Abstract of Observations for Latitude on road from LES t s Place of Obsebvation. o G a) 03 Astronomical date. Observer. Instrument observed with. Object observed. Chang La or Sakti Pass... 3 15th Sept. 1873 Capt. Trotter... Theodolite Sun (C/.Z.) Lukong Vil. Pang-gong- chu 7 19th Sept. 1873 5J )> 20th „ Kalian Sing ... Kishen Sing ... Sextant No. r> 44 44 44 44 a Piscis Australis (Fomal- haut) a Aquilse (Altair) a Piscis Australis ( Fomal- Jiauf) Sun (CT.i.) 19th „ Capt. Trotter... Theodolite j> j> Chagra 8 20th Sept. 1873 Kalian Sing ... Sextant No 44 a Piscis Australis (Fomal- haut) a Ursse Minoris (Polaris) a Aquilae (Altair) 2 let Kishen Sing ... J, 5» 44 44 Gogra Camp 12 24th Sept. 1873 25th „ 23rd „ 24th „ Kalian Sing ... Kishen Sing... 31 J! >> 3> 5? 8 8 44 44 44 n Aquilse (Altair) a Piscis Australis (Fomal- haut) a Aquilse (Altair) a Piscis Austrsilia (Fomal- haut) a Ursae Minoris (Polaris) 'Bhao (Shuraal- lung) Camp. 13 26th Sept. 1873 27th „ Kishen Sing ... 5> 44 44 a Aquilse (Altair) 13 Orionis (Eigel) 'S bo Changlung Nischu Camp. 15 27th Sept. 1873 Kishen Sing ... >> ... 44 44 a Aquilae (Altair) a Ursse Minoris (Polaris) 3 o P3 Ling-zi-t h an g plain, camp on. 16 28th Sept. 187S 29th „ Kishen Sing ... )J 44 44 44 a Aquilae (Altair) a Ursse Minoris (Polaris) (3 Orionis (Bigel) W Sumna Camp (E. ^ of Kizil jilga). 18 30th Sept. 187? \ Kishen Sing .. >> 44 44 a Aquilae (Altair) a Ursae Minoris (Polaris) ( 3 ) to YARKAND via Chang- Chenmo and Shdhidulla. Elem ENTS Double a distanc ind€ Ititudes or zenith es corrected for 'X and level errors. USED IN COM- PUTATION OF EEFBACTION". Deduced Latitudes. Remabes. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith north. o / II lytchea. Degrees. O / II o / n 1 II Z. D. 30 47 40 15-8 44 34 4 49 34 4 49 In observations with the sex- tant the stars have been invariably taken on the D. A. 51 27 20 18-2 51 34 1 meridian. In observations with the theodolite the 129 51 5 27 10 18-2 18-2 51 51 33 59 55 34 12 stars have been taken on the meridian except where an asterisk has been affixed 5» Z. D. 114 32 38 17 20 12 18-2 180 51 61 34 23 33 59 57 34 6 to the name of the star (in column 6). Where this 'Occurs the numbers in the column of Remarks are the D. A. 51 17 20 17-6 33 34 5 .4 computed means of the true local times of observa- tion used in the computa- » 70 55 50 176 33 34 5 6 tion for latitude. The » 128 55 40 17-6 33 34 4 40 34 4 59 numbers in brackets in columns 7, 10, and 11, denote the number of pairs D. A. 5> 128 50 128 60 24 47 23 41 10 10 40 170 170 170 170 44 44 44 44 34 20 24 34 20 4 34 21 1 34 22 49 (face left and face right) of observations taken. The corresponding figures in column 7 give the corrected mean zenith distance. J> 71 27 30 17-0 44 34 21 1 34 21 4 The theodolite employed is described in the Geographi- cal Appendix, Section B. D. A. 128 7 163 44 34 28 59 >} 94 21 10 163 44 34 29 13 34 29 6 D. A. 127 55 53 34 34 33 • » 71 56 33 j> ;> 34 35 32 34 35 3 D. A. 127 29 33 34 47 45 )• 72 23 13 7> 3> 34 48 51 » 93 47 23 34 46 10 34 47 54 D. A. 126 32 43 35 16 9 ;> 73 19 43 )) » 35 17 9 35 16 39 ( 4 ) Abstract of Observations for Latitude on road from LER Place of Obseevation. 03 Astronomical date. Observer. Instrument observed with. Object observed. fKotajilga Camp. TO 0^ Ovc3 P3^ SumzumlungPa. ^Dunglung or Shinglung. Kizil-jilga on bank of Karakash River. Chung Tash ^ fDaktod Karpo Sumdo. P3 J;2 j Dungnagu Camp o «s r P^ Sora Camp 19 22 26 27 28 30 32 33 26th Sept. 1873 » j> 28th Sept. 1873 29th 3rd Oct. 1873 Capt. Trotter... Capt. Trotter. Capt. Trotter. 5th Oct. 1873 1st „ 2nd 5th 8th Oct. 1873 9th Oct. 1873 10th 12th Oct. 1873 Kalian Sing . Kishen Sing. Capt. Trotter.. Kishen Sing Kishen Sing. 13th Oct. 1873 Kishen Sing 14th Theodolite Theodolite Theodolite Sextant No. 44 8 „ 44 „ 44 » 44 „ 44 Theodolite Sextant No. 44 „ 44 „ 44 Sextant No. 44 „ 44 „ 44 Sextant No. 44 „ 44 » 44 a Ursse Minoris {^Polaris)* a UrsEB Minoris {Polaris)* y Aquilse a Aquilse (Altair) a Piscis Australis {Fomal' haut) a Pegasi (Markab) y Cephei /3 Ceti a Ursse Minoris (Polaris) a Aquilse (Altair) a Orionis /3 Ceti a Ur. Min. (Polaris) Sun U. L. a Aquilse (Aliair) a Orionis a Cauis Majofis (Sirius) /3 Orionis (Eigel) a Orionis a Canis Majoris (Sirius) . /3 Ceti a Ursse Minoris (Polaris) a Canis Majoris (/SmMs),.. ( 5 ) to YARKAND via Chang- Chenmo and Shdhidulla. — (continued.) Elements Double altitudes or zenith distances corrected for index and level FSED IN COM- PUTATION or EEFEACTION. Deduced Latitudes. Remabks. errors. Baro- Then By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith north. O 1 II Inches. Degrees. O 1 II / n O 1 II H. M. 8. Z. D. 54 24 1(1) 180 16 34 29 22(1) Mean time = 10 25 37 54 23 24(1) 180 16 34 29 27(1) 34 29 25 10 31 14 Z. D. 54 19 16(1) 15-9 9 34 40 39(1) 9 49 42 54 18 23(1) 15-9 9 34 40 45(1) 9 52 57 24 22 39 159 10 34 41 23 26 8 38 15-9 10 34 41 8 34 40 59 Z. D. 65 26 25 160 -3 35 10 13 20 38 5 160 —3 35 9 53 „ 41 45 30 160 —5 35 9 38 35 9 51 D. A. 71 59 16-5 18 35 20 33 73 27 30 16-5 18 35 21 4 „ 126 25 3 165 18 35 19 58 „ 124 6 53 16-5 18 35 19 56 71 59 33 16.5 18 35 20 16 73 28 3 16-5 18 35 21 21 35 20 42 Z. D. 41 16 37 170 34 35 36 56 35 36 56 D. A. 125 33 33 35 45 45 „ 123 15 13 ... ... 35 45 47 ■>. 75 25 13 ... 35 45 52 35 45 48 D. A. 91 37 3 35 51 22 „ 123 3 23 ... • • . 35 51 43 75 13 43 ... ... 35 51 37 35 51 34 D. A. 70 39 33 36 22 74 49 23 • •• • >• 36 2 36 1 9 74 56 33 36 12 ( 6 ) Abstract of Observations for Latitude on road from LBH Place or Obseetation. a t> c 03 Astronomical date. Observer. Instrument observed with. Object observed. -2 '- 'Jung Chidmo Camp 1 35 15th Oct. 1873 16th Kishen Sing ... » Sextant No. 44 „ 44 „ 44 44 (3 Ceti a Ursse Minoris (Polaris) 3 Orionis (Rig el) a Orionis Route a sh River » „ 44 a Canis Majoris (Sirius) f^^ 1 Gulbashem 38 17th Oct. 1873 18th 17th „ Kishen Sing ... Capt. Trotter... Sextant No. 44 „ 44 44 Theodolite /3 Ceti a Ursse Minoris (Polaris) a Canis Majoris (Sirius) ... a UrsEe Minoris (Polaris)* "Shorjilga 39 10th Oct. 1873 Capt. Trotter... Theodolite ... a Ursse Minoris (Polaris) Karatagli Pass ... 41 11th Oct. 1873 Capt. Trotter... Theodolite ... Sun (U. L.) Potash Pass (near Traverse Station 41A 19th Oct. 1873 Capt. Trotter... Theodolite ... Sun (U. L.) l" No. 44.) Aktagh (1st camp) 43 13th Oct. 1873 Nain Sing ... Sextant No. 7... a Ursse Minoris (Polaris) o Q 14th Capt. Chapman Theodolite ... Sun (U. L.) Chibra Camp 43A 15th Oct. 1873 Kalian Sing ... Sextant No. 8... » 8... /3 Ceti a Ursse Minoris (Polaris)... Suget Camp 47 17th Oct. 1873 Kalian Sing ... Sextant No. 8... ,, 8... j8 Ceti a Ursse Minoris (Polaris) ^Suget Pass 45 16th Oct. 1873 Capt. Trotter... Theodolite ... Sun (L. L.) Shahic iuUa Camp 48 22nd Oct. 1873 21st Kalian Sing ... „ Nain Sing Sextaut No. 8... „ 8... ,, 7... ,, 7... ft Ceti a Ursse Minoris (Polaris) » » » ... a Canis Minoris (Procyon) ( 7 ) to YABEAND yri^ Chang-Chenmo and Shdhidulla. — (continued.) Elements Double altitudes or zenith distances corrected for index and level USED IN COM- PUTATION OF EEFEACTION. Deduced Latitudes. Remabes. errors. Baro- Ther. By stars north By sun or star?; Final latitudes meter. Faht. of zenith. south of zenith. north. O t It Inches. Degreet. O / II 1 II O / II D. A. 70 19 43 36 10 19 75 8 33 36 11 36 90 59 3 36 10 14 „ 122 26 3 36 10 24 74 36 43 36 10 9 36 10 32 D. A. 70 6 13 19-4 25 36 17 5 ' 75 22 13 194 25 36 18 26 74 23 3 19-4 18 36 17 1 R. M. S. Z. D. 52 42 13(3) 190 25 36 17 19(3) 36 17 28 Mean time == 8 36 36 Z. D. 53 47 56(3) 16-5 6 35 41 2 35 41 2^3; Mean time = 7 27 27 Z. D. 43 30 58 15-7 30 35 42 54 35 42 54 Z. D. 45 58 4 17-4 40 35 56 31 35 56 31 D. A. 74 43 20 17-2 12 35 59 • Z. D. 43 54 37 16-6 25 35 59 6 35 59 3 D. A. 70 26 30 156 25 36 6 46 75 20 15-6 25 36 7 38 36 7 12 D. A. 70 3 40 lS-9 25 36 18 20 75 23 40 189 25 36 19 10 36 18 45 Z. D. 45 22 5 157 30 36 9 53 36 9 53 D. A. 69 52 30 195 30 36 23 55 75 35 20 19-5 30 86 25 2 D. A. 75 37 20 195 30 36 26 4 „ 118 17 10 19-5 30 36 24 48 36 24 57 ( 8 ) Abstract of Observations for Latitude on road from LEH Place of Obseevation. to QJ S >j „ ... » 8 a Ursse Minoris (^Polaris) Giazgia Camp 48B 29th Oct. 1873 Kishen Sing ... Sextant No. 44 a Ursae Minoris (Polaris) 30th 15 ... „ 44 a Canis Minoris (Procyon) 24th Nain Sing » 7 a Canis Majoris (Sirius) » » » » 7 a Canis Minoris (Proeyon) Tam Langar 51 25t h Oct. 1873 Capt. Trotter Theodolite ... a Ursae Minoris (Polaris)* » j> Nain Sing ... Sextant No. 7 " . J' » 5> JJ j> „ 7 a Canis Minoris (Procyon) Khewaz Langar 51A 26th Oct. 1873 Nain Sing ... Sextant No. 7 a Canis Minoris (Procyon) Sanju Bazar 52 1st Nov. 1873 Capt. Trotter Theodolite ... a Ursse Minoris (Polaris)* 28th Oct. 1873 Nain Sing ... Sextant No. 7 " . " . " 29th „ }) ... » 7 a Canis Majoris (Sirius) ' 30th » „ 7 a Ursse Minoris (Polaris) Khushtagh Village 52A 2nd Nov. 1873 Capt. Trotter Theodolite ... a Ursae Minoris (Polaris)* 5> » Nain Sing Sextant No. 7 JJ ^ JJ „ J> » j> „ 7 a Canis Minoris (Procyon) Oi Toghrak Village 53 3rd Nov. 1873 Capt. Trotter Theodolite ... a Ursae Minoris (Polaris)* Boira Village 54 3rd Nov. 1873 Capt. Trotter Theodolite ... a Ursae Minoris (Polaris)* 4th Nain Sing Sextant No. 7 /3 Ceti » JJ » » 7 a Ursas Minoris (Polaris) Karghalik Bazar 55 6th Nov. 1873 Capt. Trotter Theodolite ... a Ursse Minoris (Polaris)* 26th May 1874 99 * * * JJ a Librae 6th Nov. 1873 Nain Sing ... Sextant No. 7 a Ursse Minoris (Polaris) ' >» 5J 3> • " ' JJ 7 /3 Orionis (Riyel) • » 5> 99 ... „ 7 a Orionis 29th Mar. 1874 „ 7 Sun (CT. L.) " » J> ... „ 7 Jupiter 30th „ J> „ 7 Sun (U. L.) ( 9 ) to YABKAND via Chang-Chenmo and ShdUdula. — (continued.) Elements Double altitudes or j distances correctec index and leve jenith I for USED IN COM- PUTATION or BEFRACTION. Deduced Latitudes. Remabes. errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith, s south of zenith. north. 1 II Incket. Degrees. O 1 II O ' '/ 1 ■ II D. A. 69 26 30 204 25 36 36 59 „ 76 1 20 20-4 25 36 38 36 37 30 D. A. 76 12 43 20-2 32 36 43 46 „ 117 36 33 202 32 36 45 7 73 25 50 199 30 36 45 37 „ 117 36 19-9 30 36 45 24 36 44 59 R. M. S. Z. D. 51 52 38(2) 21-9 32 36 52 5(2) Mean time = 9 19 12 D. A. 76 30 20 211 30 36 52 29 „ 117 23 10 211 30 36 51 51 36 52 4 D. A. 117 1 20 22-5 37 37 2 47 37 2 47 Z. D. 51 30 22(3) 24-3 19 37 11 4(3) Mean time = 9 15 57 D. A. 77 8 245 25 37 11 12 72 33 50 24-5 25 37 11 52 37 11 17 „ 77 7 30 245 25 37 10 58 Z. D. 51 15 34(3) 24-7 21 37 21 40(3) Mean time = 10 22 46 D. A. 77 28 30 23-4 33 37 21 34 „ 116 25 50 23-4 33 37 20 33 37 21 5 Z. D. 51 12 15(3) 24-8 16 37 30 20(3) 37 30 20 Mean time = 8 58 7 Z. D. 51 2 23(3) 25-2 21 37 36 55(3) Mean time = 11 8 11 D. A. 67 25 50 24-4 33 37 37 29 37 37 19 „ 78 10 24-4 33 37 37 22 Z. D. 50 47 19(3) 261 22 37 53 9(3) Mean time =952 „ 53 23 35 23-6 70 37 53 23 D. A. 78 32 25-3 54 37 53 20 „ 78 35 30 253 54 37 52 19 „ 119 1 25-3 54 37 53 2 „ 111 26 50 24-9 40 37 53 43 „ 110 49 20 24-9 40 37 53 39 „ 112 13 50 24-9 40 37 53 30 37 53 15 ( 10 ) Abstract of Observations for Latitude on road fromLEH Place of Obsebvation. Posgiam Bazar YARKAND. At Elchi Khana near the centre of the Yangi-Shahr or new city of Yarkand. g ca 56 57 Astronomical date. 7th Nov. 1873 27th Mar. 1874 8th Nov. 1873 11th „ 12th „ 27th „ 9th Nov. 1873 10th „ 26th „ 30th „ 5th Dec. 1873 )» J) 13th „ 14th „ 16th „ j> » 18th „ » j» 19th „ 20th „ 21st Observer. Capt. Trotter Nain Sing Capt. Trotter 22nd 27th 5J >t 5> 6th Jan. 1874 li'th 13th 17th 19th 28th 30th >j 20th Mar. 1874 Nain Sing Instrument observed with. Theodolite ... Sextant No. 7 7 7 Theodolite Sextant 'No. Object observed. a Ursae Minoris (Polaris)* a Canis Minoris (Froci/on) Jupiter a Ursae Minoris (Polaris)* Sun (L.L.)* a Cephei e Pegasi a Aquarii a Ursae Minoris (Polaris) a Canis Majoris (Sirius) a Canis Minoris (Procyon) a Hydrse a Tauri (Aldeharan) (5 Orion is (Kir/el) ft UrsBB Minoris /3 Ceti " '.,. Jupiter a Ononis (Rigel) a Canis Majoris (/Sfm'ws)... Sun(Z7.i.) a Ursae Minoris (Polaris) a Orionis a Canis Majoris (/Smws)... a Canis Minoris (Procyon) a Leouis (Regulus) Sun(Z7. i.) a Tauri (Aldeharan) Suu(f/:i.) a Hydrse o Leonis (Regulus) a Ursae Minoris (Polaris) Sun (U. L.) Jupiter a Virginis (Spica) /3 Ursse Minoris a Ursae Minoris (Polaris) 13 Orionis (Rigel) a Orionis a Canis Majoris (Sirius)... Sun (U.L.) ( 11 ) to YARKAND via Chang-Chenmo and Shahiduld. — (Concluded.) Elements Doable altitudes or zenith distances corrected for index and level USED IN COM- PUTATIOIS OF EEFEACTION. Deduced Latitudes. Remabks. errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith. north. 1 '1 Incheg. Degrees. 1 II 1 '1 O / II H. M. 8. Z.D. 50 25 21(3) 26-3 29 38 11 53(3) Mean time =: 10 4 44 D.A. 79 8 30 25 4 35 38 11 32 „ 114 42 40 25-4 35 38 12 11 „ 110 1 25-8 51 38 12 1 38 11 54 Z. D. 50 13 58(1) 26-5 28 38 24 48(1) ^ • 56 9 37(5) 26-4 47 Mean time = 10 45 23 37 58 26-4 40 38 24 55 38 25 16(5) „ =12 2 29 29 6 39 26-4 40 }.38 24 57 39 20 19 264 40 38 24 59 50 21 50(3)1 26-5 35 38 24 48(3) 38 25 5 50 13 21(4) 265 31 38 24 40(4) J Mean time = 11 34 „ = 9 14 50 51 D.A. 79 36 20 26-2 40 38 25 28 79 36 40 26-2 40 38 25 38 70 7 30 26-2 22 38 25 4 „ 114 16 50 >) 22 38 25 5 86 58 30 5> 22 38 25 7 „ 135 43 40 1J 22 38 23 52 86 30 10 » 22 38 25 3 46 14 10 J> 22 38 24 59 46 14 10 )> 22 38 24 59 „ 65 51 10 J> 22 38 24 55 „ 105 9 40 J) 22 38 24 57 86 29 JJ 22 38 25 36 > 70 7 30 » 22 38 24 59 56 53 » 32 38 25 16 79 35 40 )) 22 38 25 16 „ 117 55 50 M 22 38 25 36 70 7 )> 22 38 25 14 „ 114 16 10 J) 22 38 25 22 „ 128 21 56 50 10 „ 135 40 40 22 32 22 38 24 55 38 25 18 38 25 22 '38 25 8 Final latitude of Yarkand 38° 25' 2"-5 56 52 30 » 32 38 24 9 86 58 50 » 22 38 24 51 „ 128 21 30 » 22 38 24 39 . „ 74 10 J> 18 38 25 16 58 43 5> 28 38 25 22 - „ 60 6 40 JJ 28 38 24 57 60 45 20 >J 28 38 25 11 „ 104 18 20 JJ 18 38 25 6 82 12 30 >J 18 38 24 37 46 14 10 JJ 18 38 25 7 74 9 40 JJ 18 38 25 5 86 30 JJ 18 38 25 „ 117 56 10 JJ 18 38 25 24 70 7 JJ 18 38 25 6 „ 103 19 40 JJ 49 38 25 15 • { 12 ) Abstract of Observations for Latitude on road from LUH Place of Obseevation. i s a 'J Nain Sing ... ,, 5» „ J» ••• Sextant No. » » 7 7 7 7 7 7 7 7 Saturn a Piscis Australis {Fomal- haut) a Ursai Minoris (Polaris) a Canis Majoris (^SiW^^)... a Ursse Minoris ( Polaris) Jupiter a Virginis (Spica) ft TJrsae Minoris Tutialak Camp (Pang- dong-su.) 59D 7th Oct. 1873 15th May 1874 Nain Sing ... » » Sextant No. 7 7 7 a Aquilse {dltair) Jupiter a Ursae Minoris {Polaris) Sauser-polu... 59F 8th Oct. 1873 13th May 1874 14th 20th June 1874 Nain Sing ... „ Capt. Trotter . . Sextant No. >5 Theodolite 7 7 7 7 a Canis M.a]oris {Sirius)... Jupiter a Ursse Minoris (Polaris) (3 Ursse Minoris Sun (Centre) Bruchse 83 17th June 1874 Capt. Trotter. . >5 ... 3> ... Theodolite » a Serpentis ... ^ Ursse Minoris B Ophiuchi ... /3 Ursse Minoris Khtimdan Camp 60 12th May 1874 Nain Sing ... Sextant No. 7 a Ursse Minoris (Polaris) Giapshan Kizil 60A 11th May 1874 Nain Sing ... Sextant No. 7 Jupiter a Virginis (Spica) a Ursae Minoris (Polaris) Daolatbeg-uldi 61 11th Oct. 1873 Nain Sing ... Sextant No. 7 /3 Orionis (Bigel) Balti or Karakorum polu 62B. 10th May 1874 Nain Sing )> ••• Sextant No. 7 7 7 Jupiter a Virginis (Spica) a UrsjB Minoris (Polaris) ( 13 ) to YAHKAND via KaraTcoram and Kugiar. Double altitudes or zenith distances cor- rected for index and level errors. Elements used in com- putation oe refeaction. Deduced Latitu] [)E3. Rbmaeks. Baro- meter. Ther. Faht. By stars north of zenith. By sun or stars south of zenith. Final Lati- tudes north. / D. A. 49 54 72 21 40 10 Inche». 207 20-7 Degree». 44 44 O / II 34 47 45 O 1 II 34 46 34 / // 34 47 10 D. A. 67 29 49 36 30 20 203 20-3 44 44 34 55 50 34 55 42 72 38 77 5 67 9 119 14 89 8 100 29 20 40 30 40 30 30 20-3 20 3 20-2 20-2 202 20-2 44 44 50 50 50 50 34 56 23 34 55 40 34 54 23 34 55 40 34 56 38 34 56 1 34 55 43 D. A. 127 6 119 6 67 18 30 50 17-4 17 4 17-4 30 35 35 35 26 34 59 31 35 46 35 17 D. A. 76 52 119 1 67 24 100 45 Z. D. 11 35 40 10 53 171 17-3 17-3 17-3 16-6 21 30 30 30 42 35 2 59 35 2 15 35 2 25 35 2 58 35 3 4 35 2 43 • Z. D. 28 15 43 5 38 27 „ 39 35 33 58 27 35 16-6 16 6 166 166 42 42 42 42 35 4 23 35 4 16 35 5 6 35 5 39 35 4 51 D. A. 67 34 17-0 30 35 8 1 35 8 1 D. A. 118 30 „ 88 24 „ 67 53 40 ]6-7 16-7 16-7 19 19 19 35 17 30 35 17 58 35 18 14 35 17 54 D. A. 92 35 10 16-2 15 35 22 16 35 22 16 D.A. 118 16 88 9 68 5 30 40 40 16-5 16-5 165 17 17 17 35 23 49 35 24 41 35 25 24 35 24 26 4x ( 14 ) Abstract of Observations for Latitude on road from LBS. Place of Obsbevation. Astronomical date. Observer. Instrument observed with. Object observed. Nain Sing's Camp near Kara-korum Pass. Karakorum Brangsa Aktagh (2nd Camp) This camp is about one mile to tbe north of Aktagh 1st on the Shdhidula route. Khufelong Camp Kashmir-jilga Camp Kirghiz Camp (near Kirghiz jaugal.) Tupa Diwan Camp 62A. 62 79 78 77 76A 76B 25th April 1874 9th May 1874 15th June 1874 8th May 1874 Nain Sing 5> 5> 13th June 1874 Capt. Trotter Nain Sing Capt. Trotter 12th June 1874 17th Oct. 1873 Capt. Trotter Nain Sing 18th Oct. 1873 20th Oct. 1873 Nain Sing Nain Sing Sextant No. Theodolite Sextant No. Theodolite Jupiter a Virgiuis (Spica) a Ursae Minoris (Polaris) /3 Ursse Minoris Jupiter ... a Ursae Minoris (Polaris) Sun (Centre) a Virginis {Spica) ^ Virginis ... ... a Librae jS Ursse Minoris Jupiter a Virginis (Spica) a Ursae Minoris (Polaris) a Viginis (Spica) a Librae /3 Ursae Minoris /3 Librae Theodolite Sextant No. 7 7 Sextant No. 7 7 7 Sextant No. 7 7 7 7 Sun (Centre) Saturn a Piscis Australis (Fomal- liaut) a Ursae Minoris (Polaris) 13 Orionis (Bigel) a Orionis a Canis Majoris (Sirius). Saturn a Piscis Australis (Fomah Jiaut) a Ursae Minoris (Polaris) a Canis Majoris (Sirius)... ( 15 ) to YABKAND via Karakoram and Kugiar. — (Continued.) Elements 1 Double altitudes zenith distances rected for index s or cor- md USED IN COM- PUTATION OF EEFEACTION, Deduced Latitudes. Remaeks. level errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith. north. • o / '/ Indies. Degrees. / '/ / // / // D. A. 117 32 20 15-6 30 35 32 „ 87 54 40 15-6 30 35 32 50 „ 68 24 30 15-6 30 35 33 15 101 47 30 15-6 30 35 33 22 117 58 30 161 25 35 33 14 68 23 50 161 25 35 32 56 35 33 4 Z. D. 12 18 56 15-6 45 35 38 8 „ 46 7 33 15-6 45 35 37 42 35 34 33 15-6 45 35 37 43 51 8 37 15-6 45 35 37 33 „ 39 2 17 156 45 35 37 36 D. A. 117 48 10 17-4 15 35 37 55 87 45 10 17-4 15 35 37 41 68 32 40 17-4 15 35 37 16 35 37 42 Z. D. 46 30 21 169 31 36 35 * 51 30 46 16 9 31 36 15 38 40 5 16 9 31 35 59 44 44 54 45 169 31 36 9 36 11 Z. D. 12 59 3 17-4 52 36 8 34 36 8 34 D. A. 64 51 40 17-5 30 36 16 42 47 1 30 17-5 30 36 13 2 75 15 17-5 30 36 14 55 36 14 54 D. A. 90 35 50 17-7 20 36 21 59 122 2 30 17-7 20 36 22 10 74 12 20 17-7 20 36 22 19 36 22 9 D. A. 64 38 181 31 86 24 31 46 40 30 18-1 31 36 23 35 75 36 40 18-1 31 36 25 45 74 9 30 18-1 31 36 23 43 36 24 24 ( 16 ) Abstract of Observations for Latitude on road from LEJS. Place of Observation. a a 05 Astronomical date. Observer. Instrument observed with. Object observed. Kirghiz Jangul Camp ... Sasak bulak Camp Teshek Tash Camp Daba Camp Mazar Ehoja Chiklik ... Ak-Masjid ... Fusar Village 76 78A 73A 73 65 65A 63 63A 10th June 1874 17th April 1874 13th April 1874 5th June 1874 12th April 1874 31st May 1874 8th April 1874 9th 6th April 1874 7th Capt. Trotter. Nain Sing: Nain Sing Capt.Trotter. Nain Sing Capt.Trotter. Nain Sing Nain Sing Theodolite Sextant No. 7 7 7 7 7 7 Sextant No. 7 7 7 7 Theodolite Sextant No. 7 7 7 7 Theodolite Sextant No. 7 » 7 „ 7 „ 7 7 Sextant No. 7 „ 7 7 a Virginis (Spica) a Librae /3 UrssB Minoris /3 Librae a Hydrse a Leonis {Regulus) Jupiter a Virginis (Spica) a Ursae Minoris (^Polaris) /3 Ursae Minoris a Hydrae Jupiter a Ursje Minoris (^Polaris) (3 Ursae Minoris a Coronae a Serpeutis Sun (Upper Limb) a Leonis (^Regulus) Jupiter a Ursae Minoris (^Polaris) a Librae a Bootis (Arcturus) /3 Ursae Minoris /3 Librae a Leonis {Regulus) Jupiter Sun (Upper Limb) a Hydrae Jupiter Sun (Upper Limb) Jupiter ( 17 ) to YARKAN D vi^ Karakoram and Kugiar. — (Continued.) Elements Double altitudes or zenith USED IN COM- PUTATION OF Deduced Latitudes. distances corrected for REFRACTION. Remarks. index and level errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith. north. o- ; '/ Iiichei. Degrees. / '/ / tt / '/ Z. D. 46 55 35 18-2 26 36 25 54 „ 51 56 12 18-2 26 36 25 46 38 14 10 18-2 26 36 25 37 45 20 23 18-2 26 36 25 40 36 25 44 D. A. 90 52 30 183 18 36 27 28 132 14 50 183 18 36 27 47 115 16 10 18-3 18 36 27 7 86 5 40 18-3 18 36 27 30 70 13 10 18-3 18 36 27 25 103 36 20 18-3 18 36 27 41 36 27 30 D. A. 90 19 20 19-9 24 36 44 6 114 24 50 199 24 36 44 26 70 46 50 19-9 24 36 44 11 104 10 19-9 24 36 44 27 36 44 18 Z.D. 9 38 13 20-2 55 36 46 37 „ 29 56 53 20-2 55 36 46 30 36 46 34 D. A. 124 8 20 20-9 39 36 50 50 131 30 20 20-9 30 36 50 3 114 7 30 209 30 36 50 52 70 59 20 209 30 36 50 24 36 50 32 Z.D. 52 33 15 21-8 52 37 2 54 17 12 42 218 52 37 3 11 37 37 2 21-8 52 37 2 40 45 57 18 21-8 52 37 2 49 37 2 54 D. A. 130 54 20 21-3 33 37 8 4 113 15 21-3 33 87 7 38 121 20 50 213 40 37 8 17 89 31 21-3 33 37 8 18 113 17 30 21-3 33 37 8 50 37 8 13 D.A. 118 41 10 22-7 47 37 20 40 119 24 22-7 47 37 21 51 112 46 40 22-7 44 37 19 20 37 20 37 ( 18 ) Abstract of Observations for Latitude on road from LER pi s ,0 a Place of Obseevation. G S3 Cm 02 Astronomical date. Observer. Instrument observed with. Object observed. Kugiar Village 69 29th May 1874 •5 Capt. Trotter .. Theodolite 11 .. f] Ursse Majoris jS Ursffi Minoris 2nd April 1874 Nain Sing ... Sextant No. 7 Jupiter 3rd jj ■>■> 7 Sun (Upper Limb) ' „ j> 7 a Leonis {Regulus) 4th 5J 7 7 7 Sun (Upper Limb) a Hydrie a Leonis {Regulus) 5th >5 J5 7 7 Sun (Upper Limb) a Leonis {tiegidus) Tolarik Village 68 27th May 1874 Capt. Trotter .. Theodolite /3 Ursae Minoris n j> » . •• /3 Librae ( 19 ) to YARKAND via Karahoram and Kiigiar. — (Concluded.) Double altitudes or zenith distances corrected for index and level errors. Elements, used in com- putation of eepka.cticn. Baro- meter. Ther. Faht. Deduced Latitudes. By stars north By sun or stars of zenith, south of zenith. Final latitudes north. Rem.veks. Z. D. D. A. Z. D. 12 .32 43 37 15 29 112 10 40 116 15 40 130 23 40 117 1 20 88 59 40 130 23 10 117 47 1.30 22 40 37 12 41 46 21 40 lnche». Begreeg. 23-4 63 23 4 63 232 40 23-2 45 232 40 23-2 45 232 40 23-2 40 23-2 45 23 2 40 23-6 70 23-6 70 87 23 43 37 24 11 37 26 59 37 24 11 37 25 1 37 23 25 37 25 6 37 24 2 37 23 40 37 25 4 37 23 55 37 27 13 37 24 14 37 27 6 ( 20 ) Abstract of Observations for Latitude on road from LES Place of Observation. C3 Angche Chortan, R. bank 108A. 28th July 1874 of Nischu River, 29th Sumzi Ling Camp Chumik Lhakrao Camp,.. Tashliak Khiol, lake bank of. Arash Camp, on right bank of Kiria river. Ghubolik Camp, bank of Ulokshahi Khiol Lake. Polu Village Astronomical date. Observer. 108 107 Kiria Bazar 106 103 102 101 99 26th July 1874 27th 23rd July 1874 24th » » 22ad July 1874 j> » 14th July 1874 15th >» >j 12th July 1874 18th 2nd July 1874 4th 7th 8th „ 9th 18th June 1874 26th „ 27th „ 28th „ 29th Kishen Sing. Kishen Sing. Kishen Sing.. >> »> >> Kishen Sing., » Kishen Sing., 3) » Kishen Sing. Kishen Sing. Kishen Sing... Instrument observed with. Sextant No. 8. Object observed. a Scorpii (Antares) /3 Ceti a Aquilae (Alfair) a Piscis Australis {Fomal- Jiaut) /3 Ceti a Scorpii (Antares) a Aquilse (AUair) Saturn a Piscis Australis (^Fomal- haut) Saturn a Piscis Australis {Fomal- haut) a Scorpii {Antares) a Piscis Australis {Fomal- haut) a Scorpii {Antares) a Aquilae {AUair) Saturn a Scorpii {Antares) a Aquilae {Altair) Saturn a Aquilae {Altair) a Scorpii {Antares) a Aquilae (Altair) B Ursse Minoris a Aquilae {Altair) Saturn /3 Ursse Minoris a Aquilae {Altair) Saturn ( 21 ) to YARKAND yiS, NOH, POLV, and KnOTAN. Elements Double altitudes or zenith distance corrected for index and level USED IN" COM- PUTATION OF EEFBACTION. Deduced Latitudes. Remaeko. errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith. north. « O / II Inches. Degrees. O / II / '/ O 1 II D. A. 60 20 40 17*6 42 33 41 23 „ -75 15 50 17-6 42 33 42 24 33 41 54 D. A. 129 2 10 171 42 34 1 26 51 26 171 42 34 1 1 74 37 50 171 42 34 1 26 34 1 18 D. A. 5S 58 10 16-3 41 34 22 38 „ 128 18 163 41 34 23 31 75 5 10 163 41 34 21 6 „ 50 42 50 163 41 34 22 37 34 22 28 D. A. 74 36 16-4 40 34 38 19 1 50 9 164 40 34 39 29 34 38 54 D. A. 56 44 160 38 35 29 49 • 48 28 40 168 38 35 29 49 56 43 30 168 38 35 30 3 35 29 54 4 D. A. 125 41 40 16-8 40 35 41 40 72 55 20 17-3 40 35 40 10 35 40 55 D. A. 55 20 40 21-8 65 36 11 42 „ 124 39 40 23-0 65 36 12 43 72 16 50 23 65 36 10 13 „ 124 39 30 22-1 65 36 12 49 55 20 50 22-0 65 36 11 37 55 21 21-9 65 36 11 32 „ 124 39 20 21-9 65 36 12 53 36 .11 56 D. A. 104 24 30 25-0 69 36 51 57 36 51 26 „ 123 22 25-0 69 36 51 35 71 11 40 25-1 69 36 50 24 „ 104 24 50 250 69 36 52 8 „ 123 22 25-2 69 36 51 35 71 9 10 25-2 69 36 49 37 ( 22 ) Abstract of Observations for Latitude on road from LJSS CO S s Place of Obsebvation. o a

') 9) * * * 55 /3 Ursse Minoris 12th 5> ... 55 a A quite (Altair) 13th 55 j8 Ursse Minoris 14th JJ 55 •■• a Aquilse (Altair) KHOTANCITY. Obser- 93 19th May 1874 Kishen Sing... J5 a Virginis (Spica) vations taken in Shamal jy JJ o 55 • ■ • a Ursse Minoris {Polaris). Bagh in nearly the same 5> » 13 Ursse Minoris latitude as the centre 20th 55 •• • a Virginis (Spica) of the city. 5J )J 21st J5 55 55 a Ursse Minoris {Polaris). jS Ursse Minoris a Scorpii {Antares) » 55 55 • • • a Virginis {Spica) » >> 55 a Ursse Minoris {Polaris). 9' ') 55 ■ . * /3 Ursse Minoris 26th 55 ' ■ • a Ursse Minoris {Polaris). j> '> 55 ••• j3 Ursse Minoris 27th 5) a Aquilse {Altair) 28th 55 • • . a Scorpii {Antares) » " 55 a Aquilse (Altair) jS Ursse Minoris 29th 55 a Aquilse (Altair) 30th 55 ... a Virginis (Spica) » 5? 55 a Ursse Minoris (Polaris). )J 9i jS Ursse Minoris 3rd June 1874 55 a Virginis (Spica) ' 4th 55 a Aquilse (Altair) » 55 a Virginis (Spica) » 9) 55 a Ursse Minoris (Polaris). /3 Ursse Minoris 5th 55 a Aquilse (Altair) Karakash Bazar 94 23rd May 1874 Kishen Sing... 55 a Virginis (Spica) 24th J5 55 55 a Virginis (Spica) 55 55 55 55 a Ursse Minoris (Polaris). » 55 55 55 j8 Ursse Minoris Guma Bazar 90 12th May 1874 Kishen Sing... 55 Jupiter ... ( 23 ) to YAMKANB via NOS, POLV, and Xfi'Or^iV.— (Concluded.) Elements Double altitudes or zenith distances corrected USED IN COM- PUTATION OE Deduced Latitudes. for index and level EEEEACTION. Remarks. errors. Baro- Ther. By stars north By sun or stars Final latitudes me-ter. Faht. of zenith. south of zenith north. 1 II Inches. Degrees. O 1 II O 1 II o ; )i D. A. 104 40 22-9 66 36 40 5 „ 123 45 50 22-9 66 36 39 37 36 39 51 D. A. 85 1 20 25-4 69 36 59 49 „ 104 41 40 25-4 69 37 30 „ 123 3 50 25-4 69 37 37 „ 104 41 50 25-4 69 37 35 \ „ 123 4 25-4 69 37 32 37 26 D. A. 84 47 252 70 37 6 58 71 35 25-2 70 37 8 20 „ 104 56 40 25-2 70 37 7 55 84 46 50 25-2 70 37 7 3 71 34 40 252 70 37 8 10 „ 104 56 50 252 70 37 8 „ 53 29 40 252 70 37 7 29 84 47 10 252 70 37 6 53 „ 71 34 40 252 70 37 8 10 „ 104 56 40 25-2 70 37 7 55 71 34 50 25-5 70 37 8 16 „ 104 56 30 255 70 37 7 51 „ 122 50 50 25-5 70 37 7 4 53 29 10 255 70 37 7 43 „ 122 50 40 25-5 70 37 7 9 „ 104 56 30 25-5 70 37 7 51 ' „ 122 50 40 255 70 37 7 10 84 46 40 255 70 37 7 8 71 34 10 25-5 70 37 7 58 „ 104 56 20 25-5 70 37 7 48 84 46 40 25-5 70 37 7 9 „ 122 50 20 255 70 37 7 22 84 46 30 255 70 37 7 14 71 34 10 25-5 70 37 7 59 „ 104 56 30 25-5 70 37 7 55 „ 122 50 30 25-5 70 37 7 17 37 7 36 D. A. 84 29 25-7 70 37 15 59 84 28 30 25-7 70 37 16 14 71 54 25-7 70 37 17 49 „ 105 15 25-7 70 37 17 5 37 16 47 D. A. 113 52 30 25-5 69 37 37 31 37 37 31 ( 24 ) Abstract of Observations for Latitude on road from 03 0) S Place op Obsekvation. i3 o c Astronomical date. Observer. Instrument observed with. Object observed. Kok Robat ... 109 20th May 1874 Capt. Trotter ... Theodolite ... Jupiter j> )) j> 99 • • ■ /3 Leonis » >j » 99 a Virginis (Spica) Kizil Village 110 30th Nov, 1873 Capt. Trotter ... Theodolite ... a Ursae Minoris (Polaris)^ » >j » J J » JJ JJ* Yangi Hissar Town 111 1st Dec. 1873 Capt. Trotter ... Theodolite ... a Ursae Minoris (Polaris)* 18th Mar. 1874 >» JJ ... a Leonis {Regulus) Yapehan Village 112 3rd Doe. 1873 Capt. Trotter .. Theodolite ... a Ursae Minoris (Polaris)* j> » j> JJ y Pegasi {Algenih)* KASHGHAR-Yaugi- 113 4th Dec. 1873 Capt. Trotter .. Theodolite ... a Ursae Minoris (Polaris)* Shahr. In Embassy 19th »» . • JJ ... JJ j> JJ Quarters just outside 24th » • * JJ •• . # JJ JJ JJ the gate of the Yangi- » j> jj . . JJ ... a Tauri (Aldeharan) Shahr. » J) »> JJ jS Orionis (Rigel) » j» » " • JJ ... £ Orionis 27th „ j> JJ • . • j8 Orionis (Rigel) )) 99 >> ** JJ ... £ Orionis » J> 5> JJ ... a Ursae Minoris (PoZam)* 29th Jan. 1874 » • • JJ ... S Orionis » >> )> JJ £ Orionis " )) » • . JJ ... a Orionis* 3rd Feb. 1874 5> . • JJ ... a Canis Minoris (Proeyo/i)* 'J j> JJ JJ • • . a Ursae Minoris (Polaris)* 13th „ JJ JJ ••. a Canis Majoris* " » JJ JJ a Ursae Minoris (Polaris)* » » JJ JJ JJ JJ JJ ••• * »J 5» » JJ JJ JJ >j ( 25 ) rABEANB to KASSGEAB. Elements Double altitudes or zenith distances corrected for USED IN COM- PUTATION OF EEPEACTION. Deduced Latitudes. • Bemabes. inde !x and level errors. Baro- Ther. By stars north By sun or stars Pinal latitudes meter. Faht. of zenith. south of zenith. north. 1 II Inche». Degree*. 1 II 1 '/ 1 II Z. D. 33 53 20 253 71 38 26 11 99 28 9 29 25-3 71 38 26 2 33 48 56 25 25-3 71 38 26 2 38 26 5 m M. 8. Z. D. 50 11 57(3) 26-4 20 38 39 26(3) Mean time= 6 19 48 j> 50 7 33^ ^ 26-3 20 38 39 20(2) 38 39 23 „ =6 44 5 z. p. 49 48 24(3) 2G-3 37 38 55 58(3) Mean time= 6 56 27 >> 26 20 59 25-2 33 38 56 17 38 56 8 Z. D. 49 32 48(3) 25-5 38 39 13 32 Mean time= 6 36 15 s> 24 44 53(2) 25-5 38 39 13 30(2) 39 13 31 „ =7 38 48 Z. D. 49 34 40 256 24 39 24 16(3) Mean time=ll 7 4 j> 49 38 39(4) 25-6 14 39 24 19(4) „ =10 18 28 jj 49 13 25 255 17 39 24 5 9} 23 8 40 99 5> 39 24 23 19 47 44 30 99 JJ 39 24 36 J 9 40 40 44 99 >> 39 24 31 }9 47 44 19 99 n 39 24 24 40 40 34 9y » 39 24 21 49 55 38(3) 99 20 39 24 22(3) Mean time=10 51 8 39 47 55 99 17 39 25 40 41 14 99 » 39 24 57 32 1 31(4) 25-7 19 39 24 26(4) Mean time= 9 35 47 33 52 43(4) 25-5 14 39 24 19(4) „ =7 25 33 yy 50 51 16(4) S) 5J 39 24 13(4) » =11 1 45 55 56 32(6) 25-5 20 39 24 32(6) „ =5 53 42 3? 50 44 10(2) 99 J) 39 24 14(2) „ =10 2 5 50 47 47(2) 59 9) 39 24 26(2) » =10 12 48 » 50 49 18(2) 3? J> 39 24 33(2) 39 24 26 » =10 17 34 ( 26 ) Abstract of Observations for Latitude on road from Place of Obseevation. Osten Artysh (Besak Vil- lage). Chung Tirik Village Chakmak Fort Chakmak * Station, three miles north of fort. Torgat Bela ("Ayak Soghon &^ J Kyr Bulak iQ Ui Bulak Astronomical date. ^ fFaizabdd ^^ I Maralbashi Tangitar Tughamati... Kizil-boia orShamba Bazar Khanarik or Khauarik Shamba Bazar. 115 116 118 118A 10th Jan. 1874 9th Jan. 1874 8th Jan. 1874 3rd Jan. 1874 119 7th Jan. 1874 126 Observer. Capt. Trotter. Capt. Trotter. Capt. Trotter. Capt. Trotter. Instrument observed with. Object observed. Theodolite Theodolite Theodolite Theodolite I Orionis a Ursse Minoris {Polaris) Sun (U. L.) 127 129 132 137 139 140 140A. MOB. 20th Feb. 1874 | ;> » 27th Feb 1874 25th Feb 1874 5> j> J> » 2nd March 1874 » » }> >> 11th Jan. 1874 Capt. Trotter... Theodolite ... a Ursse Minoris {Polaris) Observations for Latitude on road Capt. Trotter. 17th „ I 17fch Feb. 1874 18th Feb. 1874 Capt. Trotter... Capt. Trotter... Capt. Trotter . . . » » ... Capt. Biddulph Theodolite Theodolite Theodolite » Theodolite Six-inch Sextant. Sun (L. L.) £ Orionis a Orionis a UrsEB Minoris {Polarisy /3 Ononis {Bigel) Z Orionis e Orionis h Ononis £ Orionis a Orionis Observations for Latitude taken in 5> )) » it 3rd Jan 4th 1874 » >» » 5th Jan 1874 6th 9> j> » Capt. Trotter... Capt. Trotter. Kishen Sing Kishen Sing Theodolite Theodolite Sextant No. 44 „ 44 Sextant No. 44 „ 44 „ 44 „ 44 a Canis Minoris {Vrocyon) (3 Geminorum {Pollux) ... /3 Orionis {Riff el) S Orionis £ Orionis a Orionis a Canis Majoris {Sirius)... a Canis Minoris {Procyon) jS Orionis {Bigel) a Orionis a Canis Majoris {Sirius)..- a Canis Minoris {Procyon) ( 27 ) KASEGSAB to CSADTBKUL. Elements Double altitudes or zenith distances corrected for index and level USED IN COM- PUTATION Of EEFHACTION. Deduced Latitudes. Remabe:s. errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. southof zenith. north. O t II Inchei. Degree». 1 II II 1 II Z. D. 39 59 34 25-0 20 39 36 50 39 36 50 Z. D. 48 50 36 23 2 39 47 39 47 Z. D. 62 3 11 21-5 32 40 5 9 40 5 9 Z. D. 62 41 14 215 10 40 8 28 40 8 28 Z. D. 48 13 51 19-4 —10 40 23 53 40 23 53 from KASEGSAB to AKSU. Z. D. 41 ,, 32 17 5 35 56 Z. D. 50 7 31(3) Z. D. 48 46 1 40 49 22 41 42 41 Z. D. 39 52 31 40 45 51 „ 32 6 15 Z. D. 61 50 7 „ 60 47 12 255 23 255 23 252 29 23-8 25 23-8 25 2;i-8 25 260 42 26-0 42 260 42 26 2 28 26-3 29 40 6 7(3) neighbourhood of KASMGSAB. Z. D. 34 ,, 11 23 22 36 53 Z. D. 48 22 40 24 56 41 18 8 D. A. 116 1 33 „ 68 11 23 „ 112 21 23 D. A. 84 48 33 „ 116 16 13 „ 68 25 53 „ 112 35 3 23-9 32 23 9 32 23-7 30 237 30 237 30 25 20 25-0 20 250 20 26 20 26 20 260 20 260 20 40 47 39 59 25 40 25 57 40 26 24 40 26 20 39 29 34 39 29 30 39 29 42 39 46 1 39 46 46 39 56 41 39 56 51 40 1 54 40 1 57 40 1 46 39 22 44 39 22 59 39 22 44 39 15 46 39 15 24 39 15 44 39 15 53 40 6 40 6 7 40 26 14 39 29 35 39 46 24 39 56 46 40 1 52 39 22 49 39 15 42 H. M. S. Mean time = 9 21 21 ( 28 ) Abstract of Observations for Latitude on road from YANG I to KILAK CO ;-> ID 1 Place of Obseetation. S3 05 O c 0) C3 Astronomical date. Observer. Instrument observed with. Object observed. Aktala Camp 149 22nd March 1874 Capt. Trotter Theodolite ... a Hydrse a Leonis^(SeyMZMs) Kasha-su Camp 151 15th May 1874 Capt. Trotter Theodolite ... a Virgin is (Spica) ^ Virginis • 5J >5 5> » 7/ Ursse Majoris a Bootis (Arciurus) Tarbashi Camp 156 27th March 1874 Capt. Trotter Theodolite ... a Leonis (Begulus) TASHKURGHaN, Camp near the Fort. 160 31st March 1874 Capt. Trotter Theodolite ... 5> a Leonis {Begulus)* n ?) j> a Ursae Minoris {Polaris)* a UrsaB Majoris* » » » » 3> >J » Kogachak Camp 163 3rd April 1874 Capt. Trotter Theodolite ... a Hydrse » » » J, a Leonis Aktash Camp 183 5th May 1874 Capt. Trotter Theodolite ... /3 Corvi Shash Tipa Camp 181 2nd May 1874. Capt. Trotter Theodolite ... a Ursse Majoris 2 Leouis /3 Corvi Parair-kul, Camp on N. edge of Oi-kul or lake of Little Pamir. 165 5th April 1874 7> 7' Capt. Totter Theodolite ... >> 79 ... t Ursse Majoris a Hydrse a Leonis » J> 5> » a Ursse Majoris Daraz Diwan Camp 167 7th April 1874 Capt. Trotter Theodolite ... jj ••• t Ursse Majoris a Hydrse d Ursse Majoris Mazar Tipa Camp 178 30th April 1874 Capt. Trotter Theodolite ... y Leonis . ( 29 ) PANJAH CWAERANJ vik TASEKVMGSAN and return journey to YARKAND. Elements Double altitudes or zenith distances corrected for index and level errors. USED IN COM- PUTATION OF EEFEACTION. Deduced Latitudes. Remaeks. Baro- meter. Ther. Faht. By stars north of zenith. By sun or stars south of zenith. Final latitudes north. O > '/ Inches. Degrees. O 1 II O / '/ 1 II Z. D. 46 35 44 25 53 42 22-2 22-2 29 29 38 29 38 38 28 57 38 29 18 • Z. D. 48 41 45 „ 38 8 44 11 44 50 18 21 49 19-7 19-7 197 19-7 26 26 26 26 38 11 34 ■ 38 12 16 38 12 8 38 12 3 38 12 Z. D. 25 30 55 190 15 38 6 8 38 6 8 Z. D. 25 12 7(3) 25 11 37 53 11 31(3) 24 40 49 ^^^ 24 38 49 20-2 20-2 20-2 202 202 23 23 23 23 23 37 46 30(3) 37 46 55(4) 37 46 48 37 46 59(3) 37 46 50 37 46 49 m M. s. Mean time = 9 23 37 = 9 39 20 = 10 17 48 Z. D. 45 43 37 25 1 46 18-2 18-2 9 9 37 37 23 37 36 58 37 37 11 f Z. D. 60 16 18 18-6 26 37 35 13 37 35 13 Z. D. 24 54 9 16 18 32 „ 60 13 2 260 26-0 260 18 18 18 37 31 36 37 31 54 37 31 32 37 31 39 Z. D. 11 17 55 45 21 48 24 38 58 „ 25 11 49 18-0 180 18-0 18-0 4 4 4 4 37 14 14 37 15 33 37 14 10 37 13 49 37 14 27 Z. D. 11 31 49 45 6 30 15 14 59 19-6 19-6 19-6 27 27 27 37 20 37 16 37 1 37 9 Z. D. 16 59 59 179 20 37 28 53 37 28 53 ( 30 ) Abstract of Ohser'oations f( 'yrLat. onroadfrom YANGIHISSAB to KILASFANJAH Place of Obseetation. CO a a o o z 03 Astronomical date. Observer. Instrument observed with. Object observed. Yol Mazar Camp 176 28th April 1874 Capt. Trotter... Theodolite ... a Leonis Kilah Panjah 174 18th April 1874 Capt. Trotter... Theodolite ... a Ursse Minoris (Polaris) (WAKHAN) 22ud » ... a Virginis (Spica) j» j> 3> Z Virginis ... j> jj „ ri Ursse Majoris a Librae 1 3> 5> » ... a Libr89 ( 31 ) (WAKEAN) \\k TASB.KVB.GSAN and return journey to r^S^^JVD.— (Concld.) Elements. Double altitudes or zenith distances corrected for index and level USED IN COM- PUTATION or EEFRACTION. Deduced Latitudes. Remabks. errors. Baro- Ther. By stars north By sun or stars Final latitudes meter. Faht. of zenith. south of zenith. north. O / II Inehes, Degrees. f II O / '/ / '/ Z. D. 24 42 54 19-0 32 37 18 7 37 18 7 Z. D. 54 19 43 21-3 25 37 14 47 30 3 21-3 25 37 28 36 57 2 21-3 25 37 21 ' 12 56 8 21-3 25 37 8 52 30 39 21-3 25 37 23 S. M. S. 52 33 35^2) 213 25 37 25(2) 37 18 Mean time = 12 51 4 GEOGRAPHICAL APPENDIX SECTION B. LONGITUDES. 9 ( 34 ) The method of observation employed in the determination of absolute longi- tudes was that of lunar zenith distances, as being best adapted to the largest instru- ment carried with the expedition, viz., a six-inch transit theodolite, with micro- meter eye-piece. This method of observation has not hitherto occupied a prominent position in English astronomical works, and as the results at Kashghar cannot but be considered satisfactory, I have thought advisable to enter somewhat at length into the subject and to give an example of the computation of a single night's observations there, drawn up on a form specially prepared fromChauvenet's formulcB by J. B. N. Hennessey, Esq., of the Great Trigonometrical Survey. The subject is gone into somewhat fully in an article furnished by Colonel Walker, R.E., to Mints to Travellers, a publication of the Royal Geographical Society (3rd Edition, December 1871), to which the reader is referred. The instrument employed at Kashghar is furnished with two micrometers, each moving a separate wire, the eye-piece being so arranged that the microme- ter wires may be placed parallel either to the fixed vertical or to the fixed horizontal wire of the diaphragm, according as transits or zenith distances are required to be observed. The distance between the micrometer and centre wires is adjustable at pleasure, and may be set according to the rate of motion of the celestial body observed. A complete observation of the moon, on one face of the instrument, consists in noting the chronometer times of passage of the moon's limb across each of the wires in succession and the corresponding reading of the vertical ver- niers ; a complete pair of observations on both faces gives, altogether, six time» and four readings of the vertical arc. The readings of the ends of the bubble of the level attached to the telescope, object and eye ends being alternately directed towards the object observed, give a correction to be applied to the mean of the readings of the vertical arc which gives a final zenith distance corresponding to the mean of the six chronometer times. In the example I have given it took me just three quarters of an hour to observe ten complete pairs of zenith distances as before described. A quarter of an hour may be allowed for the observation of three pairs of zenith distances to a star for time, prior to the observations to moon, and an equal time for similar observations after. To complete the observations in the time above men- tioned, however, the observer must be thoroughly familiar with his instrument, must have a good recorder, and have his lamps and apparatus in perfect order. The weak point of the system is that it is only applicable at certain times when the moon is favorably situated for observation ; still, however, even in this respect it contrasts favorably with all other methods, excepting that of " lunar distances," for determining longitudes. I give some rules which have been laid down on this subject by Colonel Walker in the JSints to Travellers, modified by subsequent experience : they may I hope be of use to future explorers. " Take pairs of observations of zenith distance on a star for the determin- ation of the local time and chronometer error, then take other pairs of ob- servations of zenith distance on the moon ; in each instance adopt the mean of the chronometer times as that of the ' complete observation ' of zenith distance. Both moon and star should be as nearly easterly or westerly as possible, and not very near (say within 10° of) the horizon. The operations should commence and close with star observations, in order that the chronometer ( 35 ) rate may be duly ascertained and allowed for. The effect of instrumental errors will be materially reduced when the stars and the moon are on the same side of the meridian and at nearly the same zenith distance; if time permits, observations should be taken both east and west of the meridian, and both before and after full moon. In north latitudes, when the moon is going from south to north in declination on any day, she is most favorably situated for observ- ing when west of the meridian ; if moving in declination from north to south, she should be observed east of the meridian. The best time for observation is when the direction of the proper motion of the moon is towards the zenith of the observer. The sidereal time when this occurs may be readily found, graphically, by drawing on a chart of the heavens a tangent to the moon's orbit, at some point near the mean position of the moon on the day of observing, and producing it to cut the declination circle passing through the observer's zenith ; then the hour circle passing through the point of intersec- tion gives the sidereal time of observation. Por practical purposes it will suffice to drop a perpendicular from the point indicating the moon's mean position on to the ecliptic, and drawing through that point a line at right angles with the perpendicular, and prolonging it to cut the declination circle. It will be found that the most favorable times occur when the moon is on the observer's prime vertical, and the least favorable when she is on the meridian. Whenever possible a few observations should be taken daily on several days rather than a large number on a single day." An examination of the results of the observations now published shows, at a glance, that those at Kashghar are both much more complete and satisfactory in every way than those taken at Yarkand and elsewhere. This is easily accounted for by several reasons : — My stay at Yarkand was limited to twenty days in all, many of which were cloudy and unfavorable for observing ; whereas I was at Kashghar on and off for more than two months, during which time I was enabled to select the most favorable days for observing ; I was at Yarkand during the early portion of our stay in the country, and not knowing what opportunities I should have, if any, for further observations, there or elsewhere, I observed the moon when- ever I could get an opportunity quite irrespective of its position being favour- able or otherwise. The observations were taken in a small court-yard, where the paved flooring^ gave anything but a stable footing to the instrument and caused great difficulty with the levels. The noise in the small court of people moving about during the operation was, it may well be imagined, highly detri- mental to such delicate work as observations for longitude, particularly where a pocket chronometer had to be used. At Kashghar, on the other hand, the court-yard was much larger and quieter and the ground more stable, and altogether the surroundings were very much more favorable. These circumstances, combined with the results obtained from the computa- tions, have induced me to employ the longitude of Kashghar as the origin for all my positions in Turkestan. I have merely employed the other observed longitudes as checks upon the general accuracy of the positions of those points as determined by other methods, for which vide the details on the construction of the map which are given in the body of the report. (Sd.) Henet Trotter, Capt., E.E. ( 36 Observations for Time, and resulting Chronometer corrections employed in determining Local Mean Time for the calculation of Longitude from Lunar Zenith Distances. Place of Obser- vation. Astronomi- cal date. Object observed. O Elements employed in com- putation of ke- fbaction. £ .2 ... . t-. Mean of observ- ed Z. Ds cor- rected for dis- levelment. Mean of Chronome- ter Times. Computed correc- tion to Chrono- meter Time to find correspond- ing Mean Time. Baro. Ther. 1873. In. (Fahrt.) o 1 II K. M. 8. JI.M. S. KASHGHAR (Yangi-Shahr.) 6th Dec. ... 7th „ ... » » 27th „ ... 28th „ ... 1874. • y Geminorum ... /3 „ i3 ,, a Leonis a Androm. a Arietis a » E. j> j> J? w. J) 25-7 25-8 5-5 25-6 24 15 20 18 3 3 3 3 3 2 2 53 8 29 41 15 19 47 21 5 66 7 53 42 46 7 50 11 53 52 26 18 10 8 32 11 43 27-3 11 7 57-6 12 40 2*6 9 21 311 11 39 11-7 11 50 45-6 —0 22 220 —0 22 21-6 —0 22 192 —0 22 19-3 —0 22 19-2 —0 22 20-7 —0 22 201 29th Jan.... 31st „ ... /3 Geininorum ... a Leonis a „ E. 5> 25-*5 31 16 15 3 3 4 47 13 25 49 40 49 49 34 17 7 40 17-9 10 38 33-6 6 56 31-0 —0 22 220 —0 22 22-4 +3 12 25-5 1873. , YARKAND ... 8th Nov. ... 9th „ ... 1874. /3 Tauri /3 Geminorum ... » » E. >> » 26-25 26'25 28 23 3 3 3 3 46 4 26 48 57 41 39 3 6 22 15 22 10 45 45-7 12 50 170 13 36 52-9 15 3 56-5 —0 18 25-7 —0 18 26-2 —0 18 23-2 —0 18 22-6 TASHKUR- GHAN. 31st March. a B 1 i s (^Arc- turus) E. 20-2 23 3 58 51 15 9 10 99 +0 2 34-4 KILA PAN- JAH 24th April /5 Geminorum ... a Leonis (fiej'MZMs) W. • 21-4 41 3 4 58 26 41 55 26 9 10 18 29-2 11 45 13-2 —0 10 5-2 —0 10 4*9 ( 37 ) Observations of Lunar Zenith Distances and resulting determination of Longitude, Place of Obsee- YATION. The moon. Astronomical date. Upper or Lower Limb. ^%9 a « > E 13 «4-l ai o o '/3 t. oj § Resulting Longitude. «u S Value from each pair of observa- tions. Mean of each day's observations. a .2 KASHGHAR(Yangi- Shahr). The station of obser- vation was in the centre of the Em- bassy Buildings just outside of and to the north of the YangiShahr or New City. 6th Dec. 1873 7th E. E. L. L. 27th 28th 29th Jan. 1874. W. w. E. U. o t II 48 30 3 47 48 47 46 40 26 46 3 40 43 59 14 43 16 48 42 25 11 41 45 4 49 1 13 48 7 59 47 2 24 46 23 50 45 7 4 44 27 30 43 33 28 42 51 23 53 13 22 H.M. S. 10 27 19-3 30 590 37 4-4 40 19-9 51 244 65 9-3 59 45 2 11 3 18-2 11 24 3-1 28 46-8 34 37-3 38 2-3 44 52-1 48 23-3 53 12-9 54 2 16 57 16 20 58 11 59 4 2 59 42 29 46 56 38 47 43 14 48 33 30 49 43 50 17 58 51 17 43 1 51 53 54 52 45 57 53 23 12 54 25 40 42 31 24 41 49 33 41 1 42 40 14 34 39 13 27 56 580 9 47 18-0 52 17-1 10 11 349 15 52-1 22 2-4 25 44-6 10 25 34-8 30 10-9 1 35 5'8 41 48-7 1 45 10-3 50 52-7 54 18-5 59 13-6 11 2 44-2 1 8 34-2 6 33 49 5 37 32-2 1 41 460 45 571 51 22-2 H.M. S. 5 4 9 4 25 4 6 4 15 3 51 4 16 4 4 25 5 3 59 4 10 3 56 4 19 4 20 4 21 4 7 4 5 4 14 4 11 4 23 4 17 4 32 4 24 4 42 4 34 4 28 4 41 4 40 4 36 4 42 4 36 4 23 4 34 4 41 4 36 4 48 4 41 4 44 IT. M. 8. 4 11 H.M. 3 45 4 10 4 20 4 45 4 40 4 36 4 47 5 15 3 25 10 ( 38 ) Observations of Lunar Zenith Distances and resulting determination of Long. — (contd*) Astronomical date. The Moon. Mean of each pair of observed Z. Ds conected fordislevelment. 6 c . o w to 2 a Resulting Longitude. « CO Place of Obsee- VATION. ^ .1 o o •r; o Value from each pair of observa- tions. Mean of each day's observations. «' ° s o (u g t< c o Cu-43 rt <5 / II n.M. 8. H. M. S. KASHGHAR (Yangi- 29th Jan. E. u. 38 36 19 54 39-8 4 51 Shahr) — concluded. 1874. 37 53 53 37 12 17 58 25-6 7 2 7-9 4 59 4 54 31st Jan. E. u. 42 21 6 8 31 25-7 5 4 44 R.M. 1874, 41 39 35 12-4 4 24 40 2 49 43 487 4 51 5 4 39 5 30 38 29 9 52 154 4 29 37 54 22 55 23 6 4 34 36 36 42 9 2 251 4 50 H. M. 8. Arithmetical Mean of Long itude from six days' observations, 5 4 27-2 which is the value finally adopted. or O 1 II 76 6 47-5 YARKAND 8th Nov. E. L. 54 59 57 11 10 35-0 5 8 30 K. M. S. S.M. Station of observation 1873. 53 48 30 16 54-7 8 30 in the Embassy- 53 7 19 20 33 2 8 35 5 8 51 2 40 Quarters in the cen- 52 21 38 24 351 8 43 tre of the Yangi- 51 34 9 28 44 9 9 13 Shahr or New City. 48 19 50 47 38 27 46 26 52 45 30 28 45 50 8 49 28-3 55 44-5 12 41-2 8 58 9 6 9 8 9 \ Ditto » }> W. U. 57 55 21 58 32 13 20 22 51-3 26 7-9 5 9 40 10 59 15 55 30 1-8 10 7 5 9 56 11 50 ( 60 10 58 34 58-4 9 50 60 47 38 38 15-6 9 55 61 29 37 42 22 9 52 62 16 38 46 160 10 4 63 29 22 52 50-8 9 59 ( 39 ) Observations of Lunar Zenith Distances and resulting determination of Long. — (concld.) '— ^A '-' , _^ The Moon". o Resulting \ Longitude. c3 > r^ -^ -t3 s ^ a> w Astronomical date. Mean of each of observec Ds correcte dislevelmen ^ a o II H. M. S. H. M. S. R . M. S. H.M. YARKAND 9th Nov. E. L. 38 17 7 13 36 28-9 5 8 52 Station of observation 1873. 37 25 58 40 581 9 11 in the Embassy 36 35 26 45 25-9 8 55 5 9 1 5 15 Quarters in the cen- 35 39 54 50 19-5 9 7 tre of the Yangi- 31 57 45 14 10 4 3 9 8 Shahr or New City. 31 6 19 30 6 53 29 11 7 14 41-2 20 29 25 5-8 9 2 8 50 9 1 Arithmetical mean of longitude from three days' observations* 5 o 77 9 16 or 1 II 19 Final longitude adopted for Yarkand, vide body of Report 77 15 55 TASHKtJRGHAN... 31st March E. U. 61 44 7 7 27 42-3 5 2 16 H. M. S. Station of observation 1874. 60 37 23 33 51-1 2 4 about 300 yards to 59 59 7 37 24-3 1 44 5 1 36 8 15 the east of the Fort. 59 16 52 41 20-9 1 26 or 58 4 33 48 7-2 1 7 o 1 II 57 17 40 52 32-7 1 1 75 24 Which gives the Astronomical longitude from one night's observa- tions. o f '/ KILA PANJAH ... Final longitude adopted for Tashkurghan, vide body of Report ... 75 19 1 Station of observation 24th April W. L. 44 53 10 23 21-7 4 50 55 m J/. S. about 300 yards to 1874. 45 13 15 31 0-5 51 17 south of principal 46 15 41 34 53-5 51 7 4 51 2 13 FortofKilaPanjah. 47 3 46 38 59-7 51 2 or 50 52 43 58 28-8 50 43 O / '/ 51 40 43 11 2 32 8 50 57 72 45 30 52 24 10 6 13-8 51 8 53 3 42 9 34-9 51 9 55 9 16 20 141 50 56 Which gives the Astronomical longitude from one night's observa- tions. f / '/ Final longitud e adopt BdforK ila Panjah, t iWeGeographi eal Chapter. 72 45 29 * Observations were also taken at Yarkand on three other nights, when the moon was so unfavorably situated that these have not been employed. t Observations were made on another night at Kila Panjah, but it appeared from the resulting time computations that the chronometer emnloved had been going irregularly. ( 40 ) SPECIMEN COPY OF COMPUTATION OF ONE DAY'S Computation of Longitude fr At Kashghar (Yangi-^liaJir) Station^ on West of Meridian. Moon < Lower Limb observed. \ Lat. N. = = 39 24 32 Assumed Eef. No. No. of observation (Mean of F. L. and F. R.) 1 2 8 (1) Chronometer Time of observation 10 47 55-2 10 52 31-3 10 57 26-2 (2) „ Correction — 22 20-4 22 204 22 204 (3) Local Mean Time (Ast. D.) = 28 days + 10 25 348 10 30 10-9 10 35 58 (4) Approx. Gr. Mean Time = (3) + i^i = 28 days + 5 20 .34-8 5 25 10-9 5 30 5-8 (5) ([ 's observed Zenith Distance = C i6 56 38 47 43 14 48 33 30 (6) Refraction (for B and 'Y) = r ... = 4- 57 59 1 (7) ([ 's Semi-diameter at (4) from N. A. = /S* — 16 2 16 2 16 2 (8) From Table I AS ... — 11 11 11 (9) (5) + (6) + (7) + (8) = ^2 46 41 22 47 28 48 18 17 (10) ([ 's Horizont&l Parallax at (4) from N. A. = tt" ■f 58 45 58 44 58 44 (11) From Table II A w" + 5 5 5 (12) Log. TTi " = log. (tt + A 7r)" ( 3-54777 3-54765 3-54765 (13) Log. sin ^2 T-86192 T-86740 T-87314 (14) (12) + (13) =log. (TTisin^s)" ( 3-40969 3-41505 3-42079 (15) TTi sin ^2 42 49 43 20 43 55 (16) (9) - (15) = ^2-'risin^2 = (see (27) ) I -8880 T-8880 T-8880 (44) Log. sin t (see (31) ) T8239 1-8331 TS426 (45) Log. cosec i^i (see (16) ) 01432 01377 01319 ( 41 ) OBSERVATIONS FOR LONGITUDE AT KASHGHAR. om Lunar Zenith Distances. 2Sth December 1873 fCwil Date, p.m. ) Long. E. = jti = — 76 15 Barometer = B = = 5A. 5m. in time Thermometer = T = 25-6 Inches. 18^ (Fahrenheit.) 4 5 6 7 8 9 10 11 4 91 11 7 30-7 11 13 13-1 11 16 38-9 11 21 340 11 25 4-6 11 30 54-6 22 20-4 22 20-4 22 20-4 22 20-4 22 20-4 22 204 22 20-4 10 41 48-7 10 45 10-3 10 50 52-7 10 54 18-5 10 59 13-6 11 2 44-2 11 8 34-2 5 36 487 5 40 10-3 5 45 52-7 5 49 ]8 5 5 54 13-6 5 57 44-2 6 3 34-2 49 43 50 17 58 51 17 43 51 53 54 52 45 57 53 23 12 54 25 40 1 3 1 4 1 5 1 8 1 10 1 12 1 14 16 2 16 2 16 2 16 2 16 2 16 2 16 2 11 11 10 10 10 10 10 49 27 50 50 2 49 51 2 36 51 38 50 52 30 55 53 8 12 54 10 42 58 44 58 44 58 44 58 44 58 44 58 44 58 44 5 5 5 5 5 5 5 354765 3-54765 3-54765 3-54765, 3-54765 3 54765 3-54765 T-88081 T-88455 T-89077 T89443 r-89956 T 90313 T-90894 3-42846 3 43220 3-43842 3-44208 3-44721 345078 3-45659 44 42 45 5 45 44 46 7 46 40 47 3 47 41 48 43 8 49 17 44 50 12 16 52 50 52 43 51 44 15 52 21 9 53 23 1 12 1 5 12 1 55 3 18 12 4 7 12 5 19 12 6 10 12 7 36 15 15 15 15 15 15 15 12 1 20 12 2 10 12 3 33 12 4 22 12 5 34 12 6 25 12 7 51 27 23 12 27 22 22 27 20 59 27 20 10 27 18 58 27 18 7 27 16 41 76 6 20 76 40 6 77 37 51 78 12 53 79 • 3 13 79 39 16 80 39 42 21 19 56 21 55 22 22 55 53 23 32 33 24 25 17 25 3 2 26 6 20 38 3 10 38 20 3 38 48 56 39 6 27 39 31 37 39 49 38 40 19 51 10 39 58 10 57 41 11 27 57 11 46 17 12 12 39 12 31 31 13 3 10 T-7898535 r7925646 T-7971396 T 7998761 T-8U37582 T-8065019 T-8110386 T-2673721 T -2790905 T -2983804 T-3096455 T-3253301 T-3362001 T3538172 01120256 0-] 120256 0-1120256 0-1120256 01120256 01120256 1120256 0-0096314 00096539 00096912 00097132 00097457 0-0097687 0-0098076 T-1788826 T-1933346 7*2172368 12312604 T-2508596 T-2644963 T2866890 1-5894413 T5966673 T-6086184 T6156302 T-6254298 T-6322482 T6433445 45 43 41 46 32 24 47 55 6 48 44 55 49 56 11 50 46 55 52 11 38 18 28 5-8 18 28 5'8 18 28 5-8 18 28 5-8 18 28 5-8 18 28 58 18 28 58 10 41 48-7 10 45 10-3 10 50 52-7 10 54 18-5 10 59 13 6 11 2 44-2 11 8 34-2 553 55-9 56-8 57-4 58-2 58-8 59-7 5 10 49-8 5 14 120 5 19 553 5 23 217 5 28 17-6 5 31 48-8 5 37 39.7 3 2 54-7 3 6 9-6 3 11 40-4 3 14 59-7 3 19 447 3 23 7-7 3 28 46-5 2 7 551 2 8 2-4 2 8 149 2 8 22 2 8 32-9 2 8 41-1 2 8 53-2 5 37 15-4 5 40 38-6 5 46 26 6 5 49 442 5 54 47-7 5 58 35-9 6 4 10-2 5 4 33-3 5 4 31-7 5 4 261 5 4 343 5 4 259 5 4 8-3 5 4 24-0 26-7 28-3 339 25-7 341 517 360 2156 2156 2-156 2156 2-157 2157 2157 14-603 14 598 14-593 14-588 14-583 14-578 14-572 r-8880 1-8880 T-8880 1-8880 1-8880 1-8880 1-8880 T-8549 T-8608 T-8705 T-8761 T8838 T-8892 T-8977 0-1241 0-1203 0-1140 0-1102 01050 01014 00955 11 ( 42 ) SPECIMEN COPY OF COMPUTATION OF ONE DAY'S Computation of Longitude fr At KasJighar CYangi-ShahrJ Station, on ( West of Meridian. Moon < Lower Limb observed. Lat. N. = for middle observation, and adopt this value as constant for all the other observations. Do. Att ) Do. TT and A(Sf for No. 1 and No. 8, and interpolate for Nos. 2 to 6 with change in Gr. Mean Time for argument. Note.— /8^ and A /Sf have the same sign and are both =±: when HEB^limb of ([ is observed; A5 is + W. lower in N. Latitude ; i is =±: if ([ is —of Meridian ; \ is always + ; /3 is i±r when ([ is moving in Declination from ill S. to N. N. to S. : sign of a = sign of i3 x sign of ;^. ( 43 ) OBSERVATIONS FOR LONGITUDE AT YiA.^Yi(^B.KR.— (Continued.) om Lunar Zenith Distances. 28^A December 1873 (Civil Date, p.m. J Long. E. = Zi = 76 15 5A. 5?w. in time Barometer = B = 25*6 Inches. Thermometer = T = 18° (Fahrenheit.) 4 5 6 7 8 9 10 T-8670 T-8691 T-8725 T8743 1-8768 T-8786 T-8812 00366 00412 00488 00528 0-0586 0-0627 0689 T.9904 T-9903 T-9903 T-9903 T-9903 T-9902 T-9902 03336 0-3336 0-3336 0-3336 0-3339 0-3339 0-3339 11761 1-1761 11761 1'1761 1-1761 1-1761 1-1761 1-5367 1-5412 1-5488 1-5528 1-5589 1-5629 1-5691 11644 1-1643 11641 1-1640 1-1638 1-1637 1-1635 T'6277 T-6231 T-6153 T-6112 T-6049 T-6008 T5944 1-424 1'420 1-412 1-409 1-403 1-399 1-393 19s. 20s. 24s. 18s. 24s. 37s 26s. 4m. 41s. 4w. 40s. 4m. 36s. 4w. 42s. 4?w. 36s. 4wi. 23s. 4m. 34s. Mean resulting longitude from observations on 28th December 1873— 5A. 4m. 36s. or 76° 9' 0" Table used to facilitate the computation. Table I for A >Sf Table II for A tt f Iable III for A 2 : = ^(1-/) 1^ g Horizontal semi- diameter. <6 a Equatorial Parallax. D JM -♦a •'* 2 / 14' 0" 17' 0" 53' 61' a TT o II II o 1/ '/ 53' 61' 10 12-7 18-8 18-6 0-0 0-3 0-0 0-4 1-^ 12-5 10 II // o 20 120 17-7 20 1-2 1-4 0-0 0-0 •00 30 11-0 16-3 30 2-7 31 5 1-8 2-1 5 -00 40 „9-7 14-4 40 4-4 51 10 37 4-2 10 •02 50 8-2 12-1 50 6-2 7-2 15 55 63 15 •03 60 6-4 9-5 60 80 9-2 20 7-2 83 20 •06 70 4-4 6-5 70 9-4 10-8 25 8-9 10-3 25 •09 80 2-3 3-4 80 103 11-9 30 10-6 12-1 30 •13 90 0-1 0-2 90 10-6 122 35 121 139 1 40 45 50 136 149 16-2 15-6 172 186 Example. ^ =30°-5' TT =56'-7, a = 7° From Tables D= ll"-5 - -/^= - 1-2 A 2 =10 GEOGRAPHICAL APPENDIX, SECTION C. HEIGHTS. 12 ( 46 ) ABSTRACT OF OBSERV Observations on road from LEH to YAM '■2 5 At Station r2 ai Reading of ^^ Place of Observation. Date. Observer. barometer or T 'rt boiling point S « , thermometer a corrected for 3 ^ index error. Inches or Degrees. 1 Chimray village 13th Sept. 1873 ... Capt. Trotter... 19-340(2) 2 Zingral Camp 14th » 16-776(4) 3 Chang La (or Sakti Pass) 15 th " . •• 15635(6) 4 Tsultak village J) » jj 16-680(2) 5 Taukse village 16th & 17th Sept. 1873 j> 18-650(8) 6 Chakr Talao Camp 18th Sept. 1873 JJ 18-022(2) 7 Lukong village (on Pangong Lake) 19th & 20th Sept. 1873 JJ ... 17-851(7) 8 Chagra Camp 21st Sept. 1873 JJ 17-217(8) 9 Lankar La (or Marsimik Pass) 22ud JJ 15-135(2) 10 Rimdi Camp j> ;> j> 15-727(4) 11 Pamzal Camp 23rd JJ 17-388(4) 12 Gogra Camp 24th » 16-864(4) 13 o f Shummal Lung pa or Bhao 26th Capt. Biddulph 15-897 14 15 "s'S . j Changlung Burma Pass ... r^ c c J ^isehu (Camp near) ^ *h3 -a j Lingzi Thung plain (south side of) ... 27th JJ JJ 14-596 ' 14-912 16 28th JJ 15-534 17 J^S i Lingzi Thung plain (camp on) ^ l^Siimna Camp east of Kiziljilga j> j> JJ 15-560 18 30th JJ 15-729 19 ^ f Kotajilga Camp 26th Capt. Trotter 16-149(1) 20 ^S'i ( Pangtung Camp 28th JJ 15-725(1).. 21 ^'^ Pangtung or Chang Lung Pass 5> }> JJ 14-805(1) 22 -| «3 -J Sumzum Lung pa Camp 29th JJ 15-714(4) 23 cW > J Dehra Compass Camp 30th J, 15-309(1) 24 ^^S Compass Wala's Pass ... 1st Oct. 1873 j> ... ' A 15-18(1) 25 -4J 2'3 Karakash River near Compass La ... 55 9J ■•• JJ 15-625(1) 26 Shinglung, or Dunglung Camp 2nd & 3rd Oct. 1873... J, 15-844(8) K Ditto >5 JJ ,j 181-60(2) 27 Kiziljilga Camp 1st, 2nd JJ 16-098(2) 28 Chungtash Camp 8th JJ ... 16-786 Note. — The figures in column (5) when given in inches are the corrected readings of a mercurial mountain barometer unless the letter A. is The numbers in brackets following the figures in column (5) indicate the number of sets of observations, the correctecTmean of which has been employed ( 47 ) ATIONS FOR HEIGHT. KAND via Chang chenmo and Shahidula. OP Obseevation. At Base Station LEH.* > o ■«4-( o ^ o . O ^ O 1 OJ ^ 111 emperature air (Fahren heit). Corrected reading of irature rcury enheit rature Fahrei eit). Remaeks. i fa barometer. empe me Fahr g =3 .3 13 H "^ H H ^ Eh C3 Degrees. Degrees. Inches. Degrees. Degrees. Feet. 63 61 19-580 60 56 11,890 57 51 19 580 60 56 15,780 43 31-3 19-617 583 54 17.590 By Captain Biddulph, 17,395 feet. 51 38 19617 58 54 15,950 51 48 19'617 58 54 12,900 61 58 19-617 58 54 13,890 56 55 19-617 58 54 14,130 52-3 48-8 19-617 58-3 54 15,090 45 25 19-617 58.3 54 18,420 By Captain Biddulph, 18,530 feet. 41 42 19617 58 54 17,500 58 54 19-569 56 52 14,790 47-3 45-5 19-569 56-3 54 15,570 . 48-2 19-424 57-6 17,020 " 32 19478 56-5 19,280 ... 35-3 19-396 57-8 18,630 • 333 19522 56-3 17,680 ... 29-0 19522 56-3 17,610 ... 22-0 19443 54-2 17,150 ' 47 44 19569 56 52 16,730 24 22 19-569 56 52 17,250 33 26 19569 56 52 18,910 32 28 19-569 56 52 17,330 21 15 19-569 56-3 49-3 17,890 ... > 23 19 537 49 40-6 18,160 33 80 19-572 49 40-6 17,440 25 25 18 19-572 19-537 49 49 40-6 406 17,030 17,030 1 Mean height = 17,030 feet. 26-5 19-508 43-1 16,590 ... 37-8 19-659 ... 43-5 15,590 attached, in which case an aneroi^ barometer has been used ; when given in degrees the figures are the corrected means of hypsometrical readings, in determining the height. ♦ The height of the observatory at Leh is taken as 11,538 feet above sea level. ( 48 ) Observations on Toad from LEH to YAR •1 At Station -C5 , Reading of Place of Observation. Date. Observer. barometer or •S-^ boiling point thermometer F^^ a corrected for ;3 index error. • ' Indies or Degrees. 29 ^ f Karakash River, Captain Biddulph's Camp 9th Oct. 1873 Capt. Biddulph 16-898 30 Ditto ditto 10th >j 17-233 31 Ditto ditto 11th » ... 17-450 32 ^'i Ditto ditto 12th >j 17-656 33 §.> Ditto ditto (Sora) ... 13th „ » 17-796 34 13 P5^ Ditto ditto 14th » 18036 35 . 1-^ Ditto ditto 15th » 18-376 36 i^ Ditto ditto 16th „ >j 18-491 37 X Fotash Camp .. • »•« 18th Capt. Trotter 18-890(2) 38 Gulbashem ... 17th Capt. Biddulph 18-804 38 ^ ^ Ditto ... 17th & 18th Oct. 1873 Capt. Trotter 19-057(2) 39 fShorjilga Camp ... 10th Oct. 1873 Capt. Trotter 16322(3) 39 Ditto ... ... 1 j> » ••• >> ••• 182-75(2) 40 ^5S Top of hill above Camp ... )5 » j> A 15-35(1) 41 -4^ 1^ Karatagh Pass ... ... ... 11th >j 180-80(2) 42 Karatagh -Lake ... ?> 5' >} .5. 18215(2) 43 Aktagh 1st ... ... 13th „ » 16-571(2) 44 -S<: Chibra Hill ... 15th >> ... A 15-36(1) 45 0) Suget Pass ... • .. 16th » 15-399(1) 46 Suget Hill ... .. • » » IJ A 15-36(1) 47 ^Suget Camp ... 17th J> 18-575(i> 48 Shahidula ... 21st » 19-477(1) 49 Sirki Angar ... 22nd „ » 18-290(1) 50 Sanju Pass (or Grim Pass) ••• ••• 23rd 5> 16-106(1) 51 Tarn village ••• ••• 25th J> 21-700(1) 52 Sanju village ... 28th » 24010(2) 53 Oi Tughrak village 2nd Nov. 1873 >> 24-425(1) 54 Boira village ... , 4th » ••• 24-856(1) 56 Posgiam village •• • ••• 7th 17 25-931(1) 55 Karghalik Town ... 6th }j ... 25-711(3) 57 YARKAND, Yangishahr ... Dec. 1873 to March 1874.* 25-992 * * The mean height of the barometer, derived &om four months' observations at Yarkand, has here ( 49 ) KAND via Cliangchenmo and Shaliidula. — (Concluded.) OP Observation. At Base Station LEH. Resulting O, o , CM o o , ^ rature eury euheit) Corrected readiug g ;h r-l S 2 S 3 ^ • ■*=> ^ IS height above mean sea level. Remaeks. >- o .S d =« 2 'o 2 ^ ^ ^ S o-S^ Inches or Degrees. 58 Sasser La (Pass) 8th October 1873 ... Nain Sing ., 15-419(1) 59 Sasser Pulu Camp 8th » 17-009W 60 Khumdan ... 9tU j> ... 16983(1) 61 Daolatbeguldi Camp 12th „ j> ... 16057(1) 62 Karakoram Brangsa 13th J) ... 15-855(1) 63 Ak Masjid ... 9th April 1874 jj 21-636(5) 64 Tupa or Akoram Pass ... 10th >> 20-392(1) 65 Mazar Khoja Camp 11th j» ... 21-325(2) 66 Yaugi Diwan Pass 16th J, 16-672(1) 55 Karghalik Town 6th Novr. 1873 ... Captain Trotter 25-711(3) 55 Karghalik do. 27th May 1874 )) 203-65 25-286 68 Yolaregh ... 25th >j 200-90 23-901 69 Kugiar Village 29th j> 200-20 23-559 64 Tupa or Akoram Pass ... 1st June 1874 j> ••• 193-19(3) 70 Tizuaf River, Camp on 2nd }> 196-47 21-799 71 Skatlich Camp 2nd » 196-82 21959 65 Mazar Khoja Camp 3rd j> 195 32 21-277 73 Duba Camp 3rd » 193-67 20-547 74 Ueh Ughaz or Chiragsaldi 8th j> 185-72 17-321 66 Yangi Diwan Pass 8th » 183-10(2) 75 Kulunaldi on Yarkand River 9th » 188-62 18-445 76 Kirghiz jangal Camp ... 10th » ••• 187-92 18168 77 Kashmir Jilga Camp ... 11th » 186-92 17-779 78 Khufelong Camp 12th » 185-97 17-415 79 Aktagh, 2nd Camp 13th >> 184-67(2) 80 Wahabjilga Camp 14th » 183-12 16-362 62 Karakoram Brangsa 15th »> ••• 181-72 15-867 81 Karakoram Pass 16th ?> ... 179-32(4) 61 Daolatbeguldi Camp 17th „ 181-92(2) ( 51 ) KAND via KaraJcoram and Kiigiar. or Obseetation. At Base Station LEH. Resulting height above mean sea level. B Temperature of mercury (Fahrenheit). St- Corrected reading of barometer. <4H o cs o a S- !- CD Oi o, si; a- a ^ Cm O , a> a EMABKS. Degrees. Degrees. Inches. Degrees. Degrees. Feet. 28 21 28 15 5 18 15 15 12 15 19-655 19-655 19-655 19 655 19-655 48-5 48-5 48-5 48-5 48-5 44 44 44 44 44 17,840 a 15,240 15,290 16,700 h 17,030 c a Mean height h Mean height c Do. = 17,820 feet. = 16,790 do. = 17,180 do. 47 37 30 23 46 9 45-3 35 30-5 18 48 69 19-589 19589 19589 19575 19666 19-689 43«8S 43-8 438 46 40-1 60 439 439 43-9 461 36-4 59-2 8,870 10,450 d 9,250 e 15,690/ 4,370 4,510 c^Mean height e Do. / Do. [ Do. = 10,465 do. = 9,355 do. = 16,000 do. = 4,440 do. ... 60 19-689 60 69-2 6,150 ... 78 19-689 60 59-2 6,450 ... 66 56 19587 19750 631 631 60.0 60-0 10,480 d 8,800 ... 70 19-750 63-1 60 8,550 , ... 62 19 750 631 60-0 9,460 e ... 60 19-750 631 600 10,440 ... 43 19-614 62-9 60-1 14,940 ... 49 35 19-451 19-614 629 62 9 60-1 60-1 16,310/ 13,210 ... 35 19-614 62 9 60-1 13,620 ... 50 19-614 62-9 601 14,250 ... 49 19614 629 601 14,810 ... 40 42 19451 19614 629 62-9 601 601 15,330 16,490 ... 40 19-628 61-4 57 9 17,330 c ... 40 22 19-469 19-628 61-4 61-4 57-9 57-9 18,550 16,880 h ( 52 ) Observation on road from LER to YAR o ' At Station 203-28(1) 93 KHOTAN City 18th „ „ ... j> 203-24(5) 93 Ditto ... 19th „ „ ... 5J 203.03(3) 93 Ditto ... 31st „ „ ... J> 203-16(3) 94 Karakash town 23rd „ „ ... „ 203-78(1) 95 Borezen Yotkao village... 29th „ „ „ 203-40(2) 93 KHOTAN City 7th June „ ,, 203-50(3) 96 Yurungkash town 8th jj » ... « 203-30(3) 97 Dol Langar village 9th „ „ ... „ ... 203-00(1) 98 Cbira village 11th „ „ ... » 203-38(3) 98 Ditto ... 13th „ „ ... » 203-45(3) 99 Keria Town 18th „ „ ... ,, ... 202-57(3) 100 Sorghak Khiang Shahi Bazaar 22nd „ „ .. . „ 198-42(3) 99 Keria Town 29th „ „ ... „ ... 202-92(3) 101 Polu village 8th July „ ... „ ... 196-33(3) 102 Ghubolik Camp, bank of Ulok Shahi Kul 12th and 13th 1874 July „ 182-1(2) 103 Arash Camp, bank of Keria Eiver ... 15th July 1874 • • • ,, ... 183-92(3) 104 Keria River at Bas Kul 16th „ „ '♦. „ ... 182-25(1) 105 YeshQ Kul (Lake) 18th and 19th 1874 July „ ... 183-58(2) 106 Tashliak Kul (bank of) 22nd July 1874 „ 182-67(4) 107 Chumik Lhakmo Camp... 23rd and 24th 1874 July » ... 182-63(2) 108 Sumzi Ling Camp 26th July 1874 „ 184-50(1) ( 55 ) LEH Y\k Khotan, Polu and Noh. OF Obseevation. o A t § .-s ® R ® g oj O 5 S 2 Degrees. H Degrees. 753 72-5 700 726 76-6 80-2 820 84-9 77-8 67 3 680 75-8 81-2 67-0 77-2 90-8 70-0 450 59-7 470 54-3 490 41-3 73-0 At Base Station LEH. Corrected reading of barometer. Inches. 19-509 19-4G5 19465 19-496 19496 19-530 19481 19502 19586 19 451 19-388 19 451 19451 19-469 19-413 19-413 19-477 19477 19-533 19-454 19-494 19463 19-463 19-428 O) J3 Jj S oj a Sag > o CO '^ ll ;5 d P4 ^ m B -♦a *^ C O ta d (O o d d-^ O) ?3 rd Degrees. 514 63-6 53-6 51-9 51-9 600 62 4 62-5 600 601 64-8 60-1 601 579 666 666 65-4 65 4 68-5 707 71-1 658 65 8 681 Resulting height above mean sea level. Feet. 4,340 4,290 4,430 4,500 a 4,590 a 4,480 a 4,010 4,240 4,380 a 4,370 4,420 4,260 ) 4,180 \ 4,830 i 7,060 4,320 J 8,430 16,960 16,020 16,880 16,160 16,620 16,600 15,570 Bemabks. a Mean height = 4,490 feet. Mean height = 4,220 feet. h Mean height = 4,575 feet. ( 56 ) Observations on road from a Place of Observation. Date. At Station "2 Observer. of baro- r boiling thermo- Drrected X error. »4 U W)0 «^ ^ .S ^^ -p s^ fi g rrt !U >i <1> .S s ^ Inches or Degrees. 109 Kok Robat ... 28th Nov. 1873 Capt. Trotter 205-64(2)1 110 Kizil village . •• 29th „ „ n 205-64(2) 111 Yangi Hissar town 30th Nov., 1st & 2nd Dec. 1873. n 204-71(6) 112 Yapchan village ... 4th Dec. 1873 n 204-64(2) 112 Ditto ... 17th March 1874 ... 1873. n ••• 204-33(2) r 11th Dec, 3-30 p.m. » 25-968 14th „ 9 A.M. J 25-026 17th „ 9 „ > 25-764 113 KASHGHAB (Yangi Shahr} - 18th „ 3 P.M. , 25-680 21st „ 3 „ n ••• 25-971 t 25th „ 3 „ 21st Jan., noon 8th Feb. 3 p.m. > > > 25-754 25-816 25-576 Note. — The values given above as baroa letrical readings at Kash I ghar are actually the readings KASIIGHAR (Yangi Shahr) ... Dec. 1873 to March 1874. Capt. Trotter Corrected mean reading. ^ A. 25-880 Note. — This mean reading of 25 880 inches is obtained from the reduction of Observations on road from KASH Artysh River, bed of Besak village (Osten Artysh^ Chungterek, Kirghiz village Balghun Bashi Camp Ditto ... Chakraak Fort Turgat Bela Camp Ditto Ditto Turgat Pass 31st Dec. 1873 31st „ and 1st Jan. 1874 1st and 2nd Jan, 3rd Jan. ... 7th „ ... 3rd „ ... 5th „ ... 6th „ ... 6th „ ... Capt. Trotter A. 2507(1) 203 •33(1) 19913(1) 194-98(1) A. 21-13(1) 19618(1) 191-83(1) ;l92-03(2) 188-83(2) ( 57 ) YABKAND to KASSGSAM. At Base Station LEH or OF Obseevation. o YARKUND. be'"' o rt <<-> I o s Corrected R EM ABKS. lb. reading •+3 C8 be c S =3 s « - ID /— N of barometer. S o 2i 2^ ^ '::; -w 2 s « c K c K O) s ^ C OJ 1 S S^ 3 *- s H H H H C! Decrees. Degrees. Inches. Degrees. Degrees. ^ee^. 47 19-720 40 825 3,830 38 19720 40 32-5 3,910 Mean height = 4,030 feet. 34 19-660 39 36 4,320 See Station No. 110, page 21. 42 19547 36-5 4,140 Mean height = 4,210 feet. 22 19 353 273 4,280 Base Station Yarkand 3,923 feet abov e sea level. 33 26-100 43 39 4,056 ^ 26 26173 24 20 4,068 23 25195 23 20 4,074 47-5 25-837 52 50 4,088 ! Mean value deduced from Yarkand " 4,060 feet. 32 26 089 34 32 4,041 33 25-950 36 33 4,123 33 25-923 30 31 4,032 1 39 25-648 41 40 3,997 J corresponding to the corrected mean of several boiling point observations. Base Station Leh 11,538 feet above sea level. 33-8 19-446 24-6 32-2 4,027 several hundreds of observations. From Leh 4,027 From Yarkand ... 4,060 Final value ... 4,043 feet. GMAB to CSADYBKVL. 26000 25-976 25 951 26071 26-199 19-449 19-449 19449 19-449 23 28 30 -8 10 -5 20 12 22 22 21 21 20 19 4,860 20 5,290 19-5 7,000 19 9,180 7 9,230 J 19 16-6 8,830 16-6 11,150 ^ 11,030 \ 16-6 166 12,760 Mean height = 5*160 feet. Mean height = 9*205 feet. Mean height = 11090 feet. 15 ( 58 ) Observations on road from KASHGSAR .2 Place of Observation Date. Observer, At Station c8 Is ^•?; ^, ® r: s Reading of meter or b{ point tht meter corr for index ( Inches Sf Degrees. 121 Bibi Miriam village 14th Feb. 1874 Capt. Trotter. A 25-84(2) 121 Ditto 22nd & 24th Feb. 1874 }] A 25.52(4) 122 Artysh Altyn village ... 15th & 16th » A 25-79(8) 122 "Ditto Ditto J, A 25-88 123 Besh Kerira village 26th Feb. 1874 >} A 25-57(1) 124 Kalti Ailak village 24th & 25th Feb. 1874 ,, A 25-77(2) 124 Ditto 1st March 1874 )) 204-72(2) 125 Bash Sogon Camp 19th & 20th Jan. 1874 » 200-63(2) 125 Ditto 19th & 20th Feb. 1874 )j A 23-84(2) 126 Ayak Sogon Camp 21st Feb. 1874 )5 A 24-97(1) 126 Ditto Ditto ,, 202-94 127 Kyr Bulak Camp 28th Feb. 1874 )> ... 202-48 127 Ditto Ditto JJ 202-48 128 Jai Tupa Camp 22nd Feb. 1874 )> A 25-14(1) 129 Ui Bulak ... 23rd }, 200-12 129 Ditto 26th „ 199-54(2) 130 Jigda Camp 23rd )J 202-88(4) 130 Ditto ... 27th » ••• 202-94(2) 131 Belowti Pass 24th » 191-28(2) 131 Ditto ... ... ... \^, 24th >J 191-28 The height of Kasbghar is taken ( 59 ) to AKSV via JJsh Turfan. OF Obseevation. At Base Station LEH oe KASHGHAR.* above el. Temperature of mercury (Fahrenheit). Temperature of air (Fahren- heit;. Corrected reading of barometer. Temperature of mercury (Fahrenheit). Temperature of air (Fahren- heit). 1^ > _c a, 9 s (X) P5 Remarks. Degrees. Degrees. Inches. Degrees. Degrees. Feet. 37 31 25-88 25-96 34 32 4,070 4,470 s Mean height = 4,270 feet. 38 33 54 2591 19-386(LEH) 25-67 29 20-2 35 4,150 4,050 4,130 \ = 4,100 „ 45 20 25-79 25-913 41 32 39-4 4,050 3,950 \ = 4,000 „ 8 29 19-460 (LEH) 26-00 21-5 31 6,490 6,290 \ > = 6,390 „ 22 19 2593 26 037 40 31 40-2 5,010 5,040 = 5,025 20 20 19-436(LEH) 25-92 27-2 34 5,380 5,290 ) 5, = 5,335 „ 24 26-128 38 38-2 4,910 16 16 26 05 25-811 44 40 42-8 6,680 6,620 \ = 6,650 „ 26 25 19454(LEH) 25-986 42 22-2 40-7 5,190 5,000 \ = 5,095 „ 10 10 19-446(LEH) 25-88 22-9 38 11,430 11,280 \ = 11,355 „ aa 4^(M3 feet above sea level. ( 60 ) Observations on road from KASHGHAB 1 Place of Observation. Date. Observer. At Station 1! baro- oiling ermo- •ected error. 'o '^ -G ^^ M •S s-s s.s a T3 *3 a 4J " 3 as « -g o fc- ;2; ^ a a.S^ Inches & Degrees. 132 Faizabad town 1st Jan. 1874 Capt. Biddulph A 20-04 133 Yangi Awat village 2nd „ >> „ 2610 134 Kashmir village 3rd „ » „ 26-37 135 Tojha Sulukh village ... 4th „ » „ 26-45 136 Shujeh village 6th „ » „ 26-60 137 Maralbashi town 8th „ » •• • „ 26-56(3) 138 Charwagh village 15th „ ,, „ 26-57 Points in neighbourhood 139 Tangitar Kurghan 17th & 18th Feb. Capt. Trotter A 24-31(3) 139 Ditto 18th Feb. • • • » 20204(2) 140 Tughamati 18th & 19th Feb. „ A 24-05(3) 140 Ditto ... 19th Feb. » 201-54(2) ( 61 ) to AKSTJ via MaralbasJii. OE Obsebvation. At Base Station KASHaHAR. Resulting (M o , Cm O ^ . rature cury enheit) 60 c3 .-H Corrected readings of rature •cury enheit) perature (Fahren- heit). height above mean sea level. Bemabes. ^2^ O-i"^'^ barometer. ^il sS^ o , a 2j2 H ^ H H H Degrees. Degrees. Inches. Degrees. Degrees. Feet. 28-5 A 26-00 28-5 3,990 As these results mostly depend upon 285 28 3,930 single readings of an Aneroid Baro- 22 22 3,670 meter, they can only be looked upon 24 „ 24 3,590 as approximate. 26-3 26 3,440 27 27 3,480 < 24 24 3,470 of KASRGSAB. ... 32 27 29 4 25-88 26-29 26-02 26-14 42 33-6 33 39-7 33 33 5,670 1 5,790 J 6,090 ) 5,860 ) Mean height = 5,730 feet. Mean height = 5,975 feet. 16 ( 62 ) Observations on road from YANGISISSAB c3 Place of Observation. Date. Observer. At Station )f baro- boiling •mome- ted for rror. a ^ Inches Sf Degrees. 148 Ighizyar village 22nd March 1874 ... Capt. Trotter .. 201-38(2) 149 Aktala Camp J) 5> „ 198-19(2) 150 Sasak Taka Camp 23rd „ A 20-92(2) 152 Kaskasu Pass 25th „ 188-60(1) 153 Chehil Gumbaz Camp ... 5> 5> ••• „ 193 17(2) 154 Turat Pass ... 26th „ 188-07(2) 155 Past Eobat Camp » j> „ 195-02(2) 156 Tarbashi Camp 27th „ 191-24(2) 157 Chichiklik Plain (pass leading into) 28th „ 18567(2) 158 Balghun Camp » » » 192-72(2) 159 Chushman village 29th ,j ... 19357(2) 160 TASHKURGHAN, fort and town ... 31st „ „ 193-27(2) , 1st & 2nd April 1874 „ 193-62(4) :!61 Kanshubar Camp 2nd „ ... >» 188-64(4) 162 Neza Tash Diwan (Pass) 3rd „ ... ,, 185-18(2) 163 Kogachak Camp )> » » 18900(4) 164 Unkul Camp 4th & 5th „ ... ,j ... 188-62(4) 165 Oikul, Kaz-kul or lake of Little Pamir, north side 6th „ ... „ 188-17(2) 166 Langar Camp 6th & 7th „ ... „ 189-39(4) 167 Daraz Diwan Camp 7th & 8th „ ... j> 192-62(5) 168 Sarhadd village 8th „ ... ,, ... 19267(2) 168 Sarbadd village 9th „ ... ,, 192 07(2) 169 Baroghil Pass Capt. Biddulph 170 Patuch village 9th & 10th „ ... Capt. Trotter 192-59(4) 171 Yur village 10th & 11th „ ... ,, 193-19(4) 172 Babatangi (Patur) village 11th & 12th „ ... „ 194-00(3) 173 Zung village 12th & 13th „ ... „ ... 195-65(4) 174 Kila Paiijah (Wakhan) 13th, 14th,&25th„ ... ,, 195-67(6) 174A Langarkish village ... ... 27th „ ... J, ••. 195-17(2) 175 Yumkhana or Jangalik Camp J> 5> ••• „ 191-47(2) Ditto ditto 28th „ ... ,, ... A 19-65(2) 176 Yol Mazar Camp 28th & 29th „ ... » 189-98(4) 177 Bilaor Bas Camp 29th „ ... „ 188-65(2) ( 63 ) to FAN J AH ( WAKSAN). or Observation. ~;^ o Degrees. 0) fi -*J rS -»J H Degrees. 24 30 36 26 14 16 25 17 20 25 10 31 32 34 25 4 19 8 18 29 31 31 28 30 30 87 37 43 40 24 29 54 At Base Station YAEKAND. Corrected readings of barometer. Inches, 25-736 25-736 25-753 19-386 19-386 19-3S6 19-386 CS O fl i. i. 1^ fa Degrees 48 48 46 s pi 03 .T5 fa 2 s-5 Degrees. 48-5 48-5 43-8 33-7 33 7 337 337 At Base Station LEU. Resulting heights above mean sea level. Feet. 5,580 a 7,350 h 9,430 G 12,850 d 10,310 13,130 9,280 h 19-395 375 337 11,370 e 19-395 37-5 33-7 14,180 19-395 375 33 7 10,540 19395 375 33-7 10,100 19-380 37-5 337 10,230/ 19-504 39-3 38-4 10,160/ 19*504 39-3 38-4 12,980 19-486 39-3 384 14,930^ 19-487 38-4 12,740 19-504 393 38-4 12,970 19-486 39-3 38.4 13,200 19-504 393 38-4 12,530 19-547 41-6 41-2 10,780 19-562 43-8 43 9 10,800 19-589 438 439 11,150 Approxi mate. 12,000 19 589 43 8 43-9 10,850 19589 43-8 43-9 10,510 19-589 43-8 43 9 10,060 19589 43-8 43-9 9,110 19-562 43-8 439 9,090 19-589 47-5 46 9 9,350 19 589 47-5 469 11,470 19-589 47-5 46-9 11,410 19-589 475 469 12,320 19-589 475 46 9 13,120 Bemabes. a Mean height ; h Ditto c Ditto d Ditto 5,600 feet. 7,345 „ 9,455 „ 12,930 „ h Ditto = 9,370 e Ditto = 11,515 „ J,U^\J „ /Ditto = 10,230 y Ditto = 14,915 „ Ditto = 10,975 „ Ditto = 11,440 „ ( 64 ) Observations on road from YANGISISSAH At Station 6 ^i S S3 -5 S'=*=< :- Place of Observation. Date. Observer. •s-^ . Si Sh fc,, ^ <1^ « ^ (-, « fee -a S O) c , ^ i; S«1 a ;3 S^ 8^ 52; ^ a g,-2"^ Indies ^ Degrees. 178 Mazar Tupa Camp 30th April 1874 ... Capt. Trotter 187-48(6) 179 Victoria Lake (or Lake of Great Pamir) 1st & 2nd May 1874 JJ 187-03(4) 180 Watershed on Great Pamir 2nd » JJ 18052a) 181 Shash Tupa Camp 2nd & 3rd » JJ 187-42(2) 182 Dahn-i-Isligh Camp 3rd » JJ 188-32(1) 183 Aktash Camp (on Aksu River) 5th & 6th jj JJ 189-42(2) 184 Tagharma Plain 10th j> ... ,, A 20-42(1) 185 Neza Tash Diwan 6th » JJ ... 185-52(2) 160 TASHKURGHAN, fort and town ... 10th „ JJ 193-42(2) 186 Balghun (Darschatt River) j> >j JJ 190-02(2) 187 Kok Mainak Pass 12th >f ••• ji ... 184-17(2) 156 Tarbashi 13th j> ••• JJ ... 191-02(2) 155 Past Robat Camp 14th }) „ 194-87(2) 152 Kaskasu Pass 15th » ••• JJ ... 188-70(2) 151 Kaskasu Camp » » ,j 192-22(2) 150 Sasak Taka Camp 16th ' 5> JJ ... 19482(2) 149 Aktala Camp 17th JJ ... JJ ... 198-47(2) 148 Ighizyar village 18th }> JJ 201-42(2) 110 Kizil village ... ... 19th JJ » 203-94(4) ( 65 ) to PANJAH rJ^AKSANJ.—(Conelnied.) OF Obseetation. At Base Station LEH. Resulting "o Q o .-? & H e^ H ^ H Degrees. Degrees. Inches. Degrees. Degrees. Feet. -•* 32 19.556 47 5 46.9 13,760 ... 23 19524 49-7 49-5 13,950 ... 46 19-524 49-7 495 14,320 '.. 33 19-524 49-7 49 5 13,760 ... 24 19 561 49 7 49 5 13,220 *•« 29 19-524 49 7 49-5 12,600 ... 35 19-548 51-3 51-4 10,310 ... 42 19524 49 7 49-5 14,900^ ... 51 19-508 52-8 51-9 10,270/ ... 30 19548 513 51-4 12,240 ... 40 19-548 51-3 51-4 15,670 ... 35 19-548 513 514 ll,660e ... 36 19548 51-3 514 9,460A ... 55 19-495 52-8 519 13,010i ... 30 19537 52-8 519 106,60 30 19537 52-8 519 9,480c ... 50 19-495 530 51-9 7,3406 , ... 60 19-495 53-0 519 5,610a ... 65 19495 530 519 4,150 37 GEOGRAPHICAL APPENDIX. SECTION D. ALPHABETICAL LIST of LATITUDES, LONGITUDES, & HEIGHTS. 18 ( 68 ) Alphabetical List of Latitudes, Eeference number. Name of Place. Latitude. Longitude. Height. 1 II o ; II Feet. 64 Akkoram (or Tupa) Pass 10,465 63 Ak Masjid Camp 37 8 13 8,870 43 Aktagh, 1st Camp 35 59 3 15,960 79 Aktagh, 2nd Camp 36 11 78 3 20 15,330 149 Aktala Camp 38 29 18 7,345 183 Aktasb Camp on Aksu River 37 35 13 74 53 U,(c) 12,600 108a Angche Chortan 33 41 54 103 Arash Camp, bank of Keria river ... 35 29 54 16,020 115 Artysh Osten (Besak village) 39 36 50 5,160 114 Artysb River (bed of) . . . 4,860 122 Artysh (Altyn) 4,100 126 Ayak Sogon Camp 40 6 6 76 40 32rc; 5,025 172 Babatangi (Pater) village • 10,060 186 Balgbun (Darscbatt River) 12,240 158 Balghun Camp 10,540 117 Balgbun Bashi 9,205 626 Balti Polu or Karakoram Polu 35 24 26 ...... 169 BarogbU Pass 12,000 125 Bash Sogon Camp 6,390 131 Belowti Pass ^ 40 40 20 77 50 11,355 115 Besak vUlage (Osten Artysh) 5,160 123 Besh Kerim village 4,130 13 Bhao or Shummal Lung Pa 34 29 6 17,020 121 Bibi Miriam Khan's village 4,270 177 Bilaor Bas Camp 13,120 54 Bohira village 37 37 19 5,340 95 Borezen Yotkan village ... 4,240 83 Bruchse Camp 35 4 51 15,920 8 Chagra Camp 34 4 59 15,090 118 Chakmak Fort 40 5 9 8,830 118a Chakmak, three miles north of Port 40 8 28 6 Chakr Talao Camp 13,890 3 Changla or Sakti Pass ... 34 4 49 17,590 14 Changlung Burma Pass 19,280 15 Do. Niscbu Camp 34 35 3 18,630 15a Do. or Pangtung Pass 18,910 85 Do. spur (top of) 15,310 86 & 59& Do. village 35 65 43 10,760 138 Charwagh village 3,470 43a Cbibra Camp 36 7 12 44 Do. Hill 17,910 , 153 Chehil Gumbaz Camp ... 10,310 157 Chicbiklik plain (pass leading into) 14,480 65a Chiklik Camp 37 2 54 1 Cbimray village 11,890 98 Chira village 37 oi 26 4,220 107 Chumik Lhakmo Camp 34 22 28 16,600 28 Chung Tash 35 36 56 15,590 116 Chungtirik (Kirghiz village) 39 47 7,000 159 Chushman village 10,100 24 Compass Wala's Pass ... 18,160 (c) denotes that the longitude has been ( 69 ) Longitudes, and Seights. Eeference number. Name of Place. Latitude. Longitude. Height. / // O / tl Feet. 182 Dahn-i-Isligh Camp 13,220 30a Daktod Karpo Sumdo ... 35 45 48 167 Daraz Diwan Camp 37 9 73 46 7(0) 10,780 61 Daulat Beguldi Camp ... 35 22 16 10,790 23 Delira Compass Camp ... 17,890 89 Digar La Pass 17,930 82 Dipsang Col 18,450 97 Dol Langar village 4,420 73 Duba Camp 36 46 34 10,440 26 Dunglung (or Shinglung) 35 9 51 17,030 32 Dungnagu Camp 35 51 34 132 Faizabad town 39 29 35 76 46 10(c) 3,990 37 Fotasb Camp 12,520 41a Do. Pass 85 66 31 63a Fusar village 37 20 37 165 Gazkul (see Oikul) 13,200 102 Ghubolik Camp 35 40 55 16,960 60a Giapchan Kizil 35 17 54 486 Giazgia Camp 36 44 59 12 Gogra Camp 34 21 4 15,570 38 Gulbashem Camp 36 17 28 12,385 90 Guma village 37 37 31 4,340 148 Igbiz Yar village Ishkashm ;.. 5,600 9,500 (Approximate.) 128 Jai Tupa Camp 4,910 130 Jigda Camp 5,095 35 Jung Chidmo Camp 36 10 32 124 Kalti Ailak village 4,000 161 Kanshubar Camp 12,980 25 Karakash River near Compass La ... 17,440 94 Do. town 37 16 47 4,010 29 Do. river, Captain Biddulph's Camp 15,540 30 Do. do. 14,980 31 Do. do. 14,620 32 Do. do. 14,160 33 Do. Sora do. 14,000 34 Do. do. 13,670 35 Do. do. 13,120 36 Do. do. 12,910 62 Karakoram Brangsa 35 37 42 17,180 48a Do. Camp 36 37 30 81 Do. Pass 18,550 62a Do. Nain Sing's Camp near Pass 35 33 4 42 Karatagb Lake 16,890 41 Do. Pass 35 42 54 17,710 55 Kargbalik town 37 53 15 77 27 4,440 113 KASHGHAR— (Yangi-shahr) 39 24 26 76 6 47 4,043 deduced chronometrically from Kashgbar. ( 70 ) Alphahefical List of Reference number. Name of Place. Latitude. Longitude. Height. o 1 It / n Feet. 134 Kashmir village 3,670 77 Kashmir Jilga Camp 36 14 54 14,250 151 Kaskasu Camp 38 12 10,960 152 Kaskasu Pass 12,930 104 Keria River at Bas Khiol 16,880 99 Keria Town 36 51 26 4,575 140& Khanarik or Do Shamba Bazar 39 15 42 61a Khewaz Langar 37 2 47 93 KHOTAN (City centre of) 37 7 36 79 59 4,490 78 Khufelong Camp 36 8 34 14,810 60 Khumdan Camp 35 8 1 15,290 52a Khushtagh village 37 21 5 174 Kila Panj ah ( Wakhan) ... 37 18 72 45 29^^; 9,090 76 Kirghiz j angal Camp ... 36 25 44 13,620 76a Kirghiz Camp near Kirghiz j angal ... 36 22 9 110a Kizil Boia or Shamba Bazar 39 22 49 27 Kizil Jilga Camp 35 20 42 16,590 110 KizU village 38 39 23 4,030 163 Kogachak Camp 37 37 11 74 55 i9rc; 12,740 187 Kok Mainak Pass 15,670 109 KokRobat 38 26 5 3,830 19 Kotajilga Camp 34 29 25 16,730 69 Kugiar village 37 24 14 6,450 75 Kulu Naldi (on Yarkand River) 13,210 127 Kyr Bulak Camp 40 6 7 76 52 26rc; 5,335 166 Langar Camp 12,630 174a Langarkish village 9,350 9 Lankar La (or Marsimik La) Leh Observatory 18,420 11,538 17 Lingzi Thung Plain (Camp on) 34 47 54 17,610 16 Lingzi Thung Plain (south side of) ... 17,680 7 Lukong village (on Pangong Lake) ... 34 6 14,130 137 Maralbashi Town 39 46 24 (78 11 20 Approximate.) 3,480 9 Marsimik La (or Lankar La) 18,420 65 Mazar Khoja Camp 36 50 32 9,355 178 Mazar Tupa Camp 37 28 53 73 34 41 ^c; 13,760 91 Muji village 4,290 84 Murghi Camp 15,190 185 Neza Tash Diwan 14,915 15 Nischu (Camp near) ... 18,630 165 Oi Kul or Lake of Little Pamir (Camp on north side) 37 14 27 74 19 m(e) 13,200 53 Oi Tugrak village 37 30 20 5,760 115 Osten Artysh (Besak village) Pamir Great, Lake of, see Victoria Lake 39 36 50 5,290 13,950 165 Pamir Little, Lake of, see Oi Kul ... 37 14 27 13,200 11 Pamzal Camp 14,790 59a & 87 Panamik village 37 47 10 10,840 20 Pangtung Camp 17,250 21 Pangtung or Chung Lung Pass 18,910 174 Panjah Kila (Wakhan) ... 37 18 72 45 29(c; 9,090 157 Pass to Chichiklik Plain >..**• 14,480 155 Past Robat Camp 9,370 ( 71 ) Latitudes^ ^c. — (Continued.) • Reference number. Name of Place. Latitude. Longitude. Height. O / II O 1 II Feet. 170 Patuch village 10,850 101 Patdr village 36 ii" 56 8,430 56 Posgiam village 38 11 64 4,210 10 Eimdi Camp 17,500 60 Sanju (or Grim Pass) ... 16.760 52 Sanju village 37 11 17 6,070 168 Sarhadd village 10,976 78a Sasak Bulak 36 27 30 150 Sasak Taka Camp 9,455 58 Sasser La Pass 17,820 69 Sasser Polu Camp 36 2 43 15,24^ 48 ShahiduUa (old fort and town) 36 24 57 11,780 181 Shash Tupa Camp 37 31 39 74 15 2Z(c) 13,760 26 Shinglung (or Dunglung Camp) 17,030 39 Shorjilga Camp 35 4i 2 16,490 40 Shorjilga (top of hiU) ... 18,050 136 Shujeh village 3,440 13 Shummal Lungpa or Bhao ^ 17,020 88 Shyok Kiver at junction with Nubra Eiver 10,760 49 Sirki Angar 13,340 71 Skatlich Camp 8,550 33 Sora Camp ... ..." 36 i 9 14,000 100 Sorghak Khiang Shahi 36 39 51 7,060 47 Suget Camp 36 18 45 12,970 46 Sugethill ... 17,990 46 Suget Pass 36 9 53 17,610 108 Sumji Ling Camp 34 1 18 15,570 18 Sumna Camp east of Kizil Jilga 35 16 39 17,150 22 Sumzumlung Pa 34 41 10 17,330 Tagharma or Muztagh Peak 38 35 15 75 22 47 25,350 184 Tagharma Plain 10,310 51 Tarn village 36 52 4 8,790 139 Tangitar Kurghan 39 56 46 6,730 6 Tanks 12,900 156 Tarbashi Camp 38 6 8 11,515 160 TASHKURGHAN Fort and Town 37 46 49 75 19 1(c) 10,230 106 Tasbliak Kul (bank of Lake) 34 38 54 16,620 73a Teshektash 36 44 18 70 Tiznaf River (Camp on) 8,800 135 Tojba Sulukh village 3,690 4 Tsultak village 16,950 140 Tughamati 40 1 52 5,975 173 Tung village 9,110 64 Tupa or Akkorum Diwan 10,465 766 Tupa Diwan Camp 36 24 24 154 Turat Pass 13,130 119 Turgat Bela Camp 40 23 53 11,090 120 Turgat Pass 12,760 59c? Tutialak Camp 35 17 74 Uch TJghaz or Chiraghsaldi 14,940 129 Ui Bulak 40 26 14 77 35 47 (e; 6,660 164 Unkul Camp 12,970 19 ( 72 ) Alphabetical List of Eeference number. Name of Place. Latitude. Longitude. Height. 179 Victoria Tiake, or Lake of Great Pamir (West end) o 37 27 1 It 73 40 38 Feet. 13,950 80 180 Wahabjilga Camp Wamar Port (Junction of Murghabi and Panja River) Water-sbed on Great Pamir 16,490 (Appe.) 7,500 14,320 133 66 111 112 Yangi Awat village Yangi Diwan Pass Yangi Hissar Town ... ... ... Yapcban village 38 39 "56" 8 13 31 76 12" 55 3,930 16,000 4,320 4,210 ( 73 ) Latitudes, 8fC. — (Concluded.) Reference number. Name of Place. Latitude. Longitude. Height. 67 105 68 176 175 171 96 2 92 YARKANP (Yangi-shahr) Yeshil Kul (Lake) Yolarik... Yol Mazar Camp Yumkhana or Jangalik Camp Yur village ... ... ... Yurungkash town „. Zingral Camp Zawa Kurghan ... ... ... / It 38 25 1 37 "27" 6 37 18 7 ; II 11 15 65 73 "5" 49(c) Feet. 3,923 16,160 6,150 12,320 11,440 10,610 4,370 15,780 4,430 GEOGRAPHICAL APPENDIX. SECTION E. MAGNETIC OBSEI^VA-TIOISrS. 20 ( 76 ) MAGNETIC Abstract of results of observations taken by Captain S. Trotter^ R.E., Date of observation. Approximate. Station of observation. North Latitude. Longitude east of Greenwich. Height above sea level. 1873. o / / Feet. j "LEH 1st and 3rd September 34 10 77 37 11,540 ChagraCamp... 21st September 34 5 78 30 15,090 m Chung Tash Camp 8th October 35 37 78 40 15,590 Sanju village 1st November 37 11 78 31 6,070 13 Oi Toghrak village 3rd „ 37 30 78 3 5,760 Karghalik town 6th 37 53 77 41 4,370 c8 YARKUND (Yangi-shahr) 27th 38 25 77 16 3,923 8 Yangi Hissar town 1st December 38 56 76 13 4,320 Yapchan village 3rd „ 39 14 76 7 4,210 rK A SHGHAE (Yangi-shahr) 4th and 19th Decem- 39 24 76 7 4,043 ber and 13th Febru- ary 1874. Ui Bulak. (On road to Ush Turfan and Aksu) 25th February 40 26 77 36 6,650 i- 'Ighizyar 18th May... 38 40 76 12 5,600 TASHKURGHAN Aktash 31st March 5th May ... 37 37 47 35 75 29 74 54 10,230 12,600 o Yol Mazar Camp 28th April 37 18 73 6 12,320 ^Panjah (Wakhan) 18th „ ... 37 72 45 9,090 The observations for Magnetic Dip were taken with Dip Circle No. 2 by Barrow (belonging to the Great " Admiralty Manual of Scientific Enquiry. The rules laid down therein were rigorously adhered to. The observations for declination were taken with the 6-inch Transit Theodolite (by Troughton and Simms). parisons at Dehra and at Leh, both before the start and after the return of the Mission, proved that no sensible ( 77 ) OBSERVATIONS. for Magnetic Inclination (Dip) and Declination (Variation) 1873-74. Magnetic Dip north. Magnetic Variation east. Eemabes. O f 47 21-5 / 3 43 From observations to Sun. (Two sets.) (Mean of two sets.) * 47 22-7 3 56 From observations to Sun and to Polaris. 3 51 From observations to Polaris. 1-5 «©*;; o6' — ■ — • ^7"? frHrH 4 32 Ditto ditto. 00 CD ^■:^p^ 4 32 Ditto ditto. 03 4 53 Ditto ditto. t ; :«o : - - 02 aj 00 63 8-0 4 58 Ditto ditto. 2"^ J (Mean of two sets.) '43 r^ "a 2 fl " -d 4 57 Ditto ditto. 1=2 :g^ : :g 4 55 Ditto ditto. 54 31-7 5 1 Ditto ditto and to Sun. (Pour , E.E., too len the dip ;n July am on July 31 on Septem 6 40 From observations to Sun. ham, 7_wh ition its i tion 420 From observations to /3 TJrsce Minoris. (Two sets.) q'S.S & • g 52 3-3 4 34 From observation to Sun ICun ctober the d( chlagi e dip the de 4 24 Ditto to Polaris ditto. SOr^cc^^ 1.2 i^ a i 4 12 Ditto Ditto ditto, p EH 4 16 From observation to do. and to /3 Ursee Mi * 'noris. Trigonometrical Survey Department). The method of observation was that recommended by General Sabine in the The needle attached to the instrument was re-magnetized prior to the departure of the expedition from India, and coin- displacement had taken place in the position of its magnetic pole. GEOGRAPHICAL APPENDIX. SECTION r. METEOROLOGICAL OBSERVATIONS 21 ( 80 ) Meteorological Ohservations recorded by the Great Trigonometrical Mercurial Baro- Reading of Tempeeatuee of .^ie. Date. meter No. 720 corrected for Temperature of aneroid barometer mercury. Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches, Degrees. Incite s. Degrees. Degrees. Nov. 12th, 1873, 9 A.M. ... 26-112 46 26 46 42 38-5 Noon . . . •107 55 46 52 47-5 3 P.M. ... •077 50 43 48-5 44-6 6 „ ... •007 4-65 46 45 41 9 „ ... •007 44-5 47 42 39-5 „ 13th „ 9 a.m. ... •132 44 51 42 38-5 Noon . . . •127 55 49 45 39 3 p.m. ... •077 49-5 41i 48 44 6 „ ... •087 45 46 45 41 9 „ ... •087 42 46 41 38 „ 14th „ 9 A.M. . . . •147 44 52 42 39 Noon . . . •127 52 48 48 45 3 P.M. ... •087 47-5 45 46 42 6 „ ... •092 46 46 44 39-5 9 „ ... •125 42 50 40-8 36-5 „ 15th „ 9 a.m. ... •157 48-5 54 41-0 370 Noon . . . •167 58-0 53 48 41 3 p.m. ... •142 51-5 52 48-5 40 6 „ ... •197 49-5 56 43 36 9 „ ... •237 47-5 61 38 32 „ 16th „ 9 a.m. ... •282 47-5 68 36 33 Noon . . . •277 600 m 45-3 38-8 3 p.m. ... •227 55-8 62 50-5 43 6 „ ... •207 48 63 46 41-3 9 „ ... •237 43-5 m 42 37 „ 17th „ 9 a.m. ... •187 44-5 61 38 34 Noon , . . •157 46-5 56 42 37-5 3 p.m. ... •107 48 52 45-5 39 6 „ ... •117 45 51 m 35 9 „ ... •037 41-5 54 38 33 „ 18th „ 9 a.m. ... •187 43 59 39 36 Noon , . . •197 55-5 56 47 40-3 3 pm. ... •157 51-5 55 48-5 41 6 „ ... •137 47 56 42-5 37 9 „ ... •207 43 58 39 34 „ 19th „ 9 a.m. ... •207 44 63 35-5 34 Noon . . . •192 57 59 46-5 41 3 p.m. ... •162 63-5 55 49-5 42 6 „ ... •137 47-0 59 45 39 9 „ ... •187 45-0 60 40-5 36 „ 20th „ Noon ... •217 55'5 61 47-5 42 3 p.m. ... •187 51-0 57 46 39 6 „ ... •197 45-0 59 45 39 9 „ ... '197 44-0 60 41 37 „ 21st „ 9 A.M. ... •237 45-0 65 38 32-5 Noon . . . •227 47-5 63 47 42-5 3 P.M. ... •197 52-0 61 51 44 6 „ ... •197 48-0 62 42-5 38 9 „ ... •217 430 64 39 34 „ 22nd „ 9 a.m. ... •157 44-0 58 41 36 Noon . . . •127 56-5 55 60 45 ( 81 ) Survey Fandits at YABKAND during the winter of 1873-74. Minimum in shade. Maximum in shade. Direction of Eemaek s. wind. During preceding 24 hours. Degrees. Degrees. At Yarkand the thermometers were placed in a court- yard — in the open — against a wall and at a height of five W. feet above the ground. The wall fiiced north, and the sun's W. rays never fell on or near it. w. The aneroid barometer was also suspended against the N. same wall. N. The mercurial barometer was placed in complete shade, W. in a porch which opened towards the north. On the 20th, 21st, and 22nd of every month continuous hourly observations were taken extending over a period of N.W. W. 48 hours, i.e., from noon on the 20th up to noon on the E. 22nd of each month. These observations having been S.W. reduced to diagrams, vide body of Keport, it has not been W. thought necessary to reproduce them here. s. The thermometers were all graduated on Fahrenheit's N. scale. S. E. E. N.E. S. E. N. S. S. N. E. W. N. N. W. E. E. S. E. W. N. E. S. E. S. E. S. w. s. s. N. s. N. N. ( 82 ) Meteorological Ohservations recorded by the Great Trigonometrical Survey Mercurial Baro- Eeading of Tempeeatuee of Aie. Date. meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. Nov. 22nd, 1873, 3 P.M. ... 26-087 63-0 26-50 52 46 6 „ ... •087 52-5 •50 46 41 9 •107 49-0 •53 42 38 „ 23rd „ 9 a.m. ... •127 47-0 •55 40 37 Noon . . . •117 60-0 •51 53 46 3 p.m. ... •047 67-0 •46 67 49 6 „ ... •037 60-0 •45 48 42 9 „ ... •047 47-0 •49 44 40 „ 24th „ 9 a.m. ... •047 43-0 •50 42-3 38 Noon . . . •037 58-0 •45 62 46 3 P.M. ... •002 63-0 •38 53-5 46 6 25-962 51-0 •37 44-5 39 9 „ ... •952 45-0 •38 4fd 36 „ 25th „ 9 a.m. ... •922 43-0 •35 34 31 Noon . . . •932 55 •31 46-5 41 3 P.M. ... •882 52 •30 51 43 6 „ ... •902 49 •30 45 40 9 •942 45 •34 39 34-5 „ 26th „ 9 a.m. ... •902 45-5 •32 39 36 Noon . . . •922 58 •31 45 41 3 P.M. ... •952 64 •32 49 44 6 „ ... •952 50-5 •37 44 40-5 9 „ ... •972 45 •40 37-5 34 „ 27th „ 9 A.M. ... 26-122 45-5 •52 36 34 Noon . . . •102 57 •50 49 43-3 3 P.M. ... •092 53 •49 62 44 „ 28th „ Noon ... •422 43 •04 40 33 3 P.M. ... •427 48 •02 45 37 6 „ ... •442 42 •06 36 32 9 „ ... •452 37 •09 31-5 27-5 „ 29th „ 9 a.m. ... •402 28 •03 26 23 Noon •322 35 25-96 33 29 3 P.M. ... •272 37 •87 34 31 6 „ ... •222 35-5 •85 28 24 9 „ ... •212 35 •84 25 22 „ 30th „ 9 a.m. ... •102 27-5 •75 24 24 Noon . . . •052 37-5 •65 36 33 3 P.M. ... •002 39-0 •63 36 32 6 „ ... 25-997 37 •62 30 27 9 „ ... 26-042 33 •66 26 25 ( 83 ) Fandits at YABKAND during the winter of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of Remaees. wind. During preceding 24 hours. Degrees. Degrees. E. W. S. N. N. E. S. w. E. W. s. N. N. W. w. s. N. N. w. For the month of November 1873. E. 9 A.M. Noon. 3 p.m. 6 p.m. 9 p.m. N. N. Mean of Mercurial W. barometer No. 720 W. corrected to 32°... 26143 26-149 26115 26109 26126 33-5 42-5 N.E. N.E. Mean of Dry Bulb N. N. E. Thermometer ... 37-4 45-9 48-0 42-2 381 iis's '44'" Mean of "V^et Bulb S. Thermometer ... 34*2 40*6 416 37*4 34-1 E. W. S. 14 34-8 s.w. w. E. N. 22 ( 84 ) Meteorological Observations recorded by the Great Trigonometrical Mercurial Baro- Eeading of Tempeeatube of Aie. Date. meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- index error. Degrees. corrected. Dry Bulb. Wet Bulb. Inches. Inches. Degrees. Degrees. Dec. 1st, 1873, 9 A.M. ... 26022 26 25-66 22 21 Noon •052 34-5 •68 34 31 3 P.M. •042 39-5 -63 37 33 6 „ •042 37 •68 30 29 9 „ •082 33 •71 25 24 „ 2nd „ 9 A.M. •152 27 •80 25 23 Noon •102 38 •72 36 32 3 P.M. •092 39 -69 36-5 34 6 „ •052 36 -68 31 30 9 „ •027 33 •66 25-3 25 „ 3rd „ 9 A.M. •042 26 •67 23 22-2 Noon •042 37 •67 34 30 3 P.M. •002 44 •64 38 35 6 „ •022 39-8 •62 33 30 9 „ •022 57 •65 29 26 „ 4tll „ 9 A.M. •052 28 •68 24-5 22-3 Noon •027 37-5 •67 35 32 3 P.M. •042 42-8 •61 37-5 36 6 „ •022 38 •62 30 27 9 „ •032 34 -64 24 23 „ 5th . „ 9 A.M. •022 26 •63 23 22 Noon •65 33 30 3 P.M. "■•027 '"42 •63 38 36 6 „ •082 38 •69 29-5 28-3 9 „ •102 33 •71 245 24 „ 6th „ 9 A.M. •202 26-5 •85 24 23" Noon •83 33 30 3 P.M. •162 ""m •78 36-5 34 6 „ •177 37 •80 29 28-5 9 „ •167 32 •98 25 24 „ 7th" „ 9 A.M. 192 ■ 26-3 -83 24 24 Noon •222 39 •86 35 32 3 P.M. •212 44 •83 38 34 6 •242 40 •87 32 29 9 „ '242 35-5 •88 27 24-2 „ 8th „ 9 A.M. •312 26-5 •95 24 23 Noon •307 38-5 •94 36 32 3 P.M. •302 44 •91 40-3 35 6 „ •322 345 •90 34 31 9 „ •282 27 •92 30 27 „ 9th „ 9 a.m. •182 •27 •83 25 23 Noon •152 36 -78 34 30-8 3 P.M. •112 40 •70 38 34 6 „ •062 375 •68 32-5 29 9 „ •052 35 •67 29 27 „ 10th „ 9 a.m. •002 24 •62 21 20 Noon 25-992 33 -61 29 26-5 3 p.m. •962 395 •57 37 32 6 •977 360 -60 28 26^5 9 „ 26^012 325 •63 25-5 250 „ nth „ 9 a.m. •122 26 •77 24 23 Noon •132 38 -79 32 29 ( 85 ) Survey Fandits at YABKANB during the winter of 1873-74. — (Continued.) MlNIMtTM IN SHADE. Maximum in shade. Direction of Re MABKS. wind. During preceding 24 hours. Degrees. Degrees. 13-5 38-8 S.W. w. E. • N. N. 15-5 38-5 E. S. E. S. w. 15-8 39-8 N. N.W. S. N. E. 16 38 N. W. S.W. N. N. 12 37-5 W. N. S.E. n.e". 13 41 N. N.W. N. N.W. S. 12-5 35 w. E. N. S.W. w. 11-5 37 E.S. N.E. E. N. N. 12 38 S.E. W.N. N.E. W. E. 10 42-6 S. E. W. N. W. 12 34 S.W. S.E. . ( 86 ) Meteorological Observations recorded by the Great Trigonometrical Survey Mercurial Baro- Reading ?of Tempeeatxjbe of Aie. Date. meter No. 720 corrected for Temperature of aneroid bar mercury. Solomon' ometer s un- index error. correct ed. es. Dry Bulb. Wet Bulb. Inches. Degrees. Inch Degrees. Degrees. Dec. 11th, 1873,3 p.m. ... 26-134 43 25 64 39-3 35 6 „ •142 38 78 31-5 28-6 9 „ •162 32-5 80 26 24 „ 12th „ 9 a.m. •202 28 -85 27 25 Noon . . . •232 37 86 37 34-3 3 p.m. ... •202 38-8 82 37 33 6 „ ... •202 36 83 33 31 9 » ... •222 33 •84 27 26 „ 13th „ 9 A.M. •192 24 •84 22 21-5 Noon . . . •182 35 •82 32 39-5 3 P.M. ... •162 40 79 36 33-6 6 „ •162 37 78 30-5 29-5 9 „ •172 32-5 81 27-0 25-0 „ 14th „ 9 a.m. •162 24-5 81 20 19-6 Noon . . . •152 35-0 78 32 32 3 p.m. ... •102 39-0 72 36 32 6 „ ... •132 35-5 73 27 26 9 „ •102 32-0 72 25 22-5 „ 15th „ 9 a.m. ... •082 21 71 16-5 160 Noon ... •092 33 75 30 27 3 p.m. ... •062 39 68 26 33 6 „ •052 35 67 26 25 9 „ ... •042 30 66 22 21-3 „ 16th jj t7 A*A1* , . , •022 25 65 21 20 Noon . . . 25^997 36 61 31-5 29-5 3 p.m. ... •952 40-5 58 35 33 6 „ ... •972 35 59 30 29 9 „ •952 32 60 23 22 „ 17th „ 9 a.m. •902 23 54 20 19 Noon ... •877 30-5 53 31 29 3 p.m. ... •857 33-5 69 31-3 28-5 6 „ •862 33-0 59 29 28 9 „ ... •862 32-0 58 28 26 „ 18th „ 9 a.m. •892 330 51 30 28 Noon ... •902 49-3 62 47 38 3 p.m. ... •892 52 51 50 41-5 6 „ ... •902 44 52 33 31 9 „ •927 37 57 33 32 „ 19th „ 9 a.m. 26-077 20-5 72 17-5 16 Noon . . . •112 32 74 30 27 3 P.M. ... •102 36 72 34 31 6 „ ■112 32 7Q 25 24 9 „ ... -122 26-5 77 21 19 „ 20th „ 9 a.m. -202 23-5 85 22 21 Noon ... -192 35 84 33 30 3 p.m. ... •162 37 80 36 31 6 „ ... •152 32 80 28 25 9 „ ... •177 31 82 24 22 „ 21st „ 9 a.m. •132 21 79 17-6 16 Noon . . . •132 30-5 78 28 25-6 y 3 p.m. -102 34-5 75 32 29 6 „ -102 31 72 25 23 ( 87 ) Pandits at YARKAND during the winter of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of BeM AR ES. wind. During preceding 24 hours. Degrees. Degrees. N. N. N. 18 43 S. N.W. N.E. N. S. 10 41 N. N. S.E. S. E. 12 34 N. N.W. N. W. , S.E. 10 33 N.W. W. E. W s. 10 36-5 w. E. E. N. w. 9-8 36 N. s.w. N. S. s. 19 35-5 S.E. N. W. E. S. ioi 49 N.B. N. N.B. W. s. 9-5 si-s N. N. W. E. •6 14 N. N.E. S.E. N. 23 ( 88 ) Meteorological Ohsenatwns recorded by the Oreat Trigonometrical Survey Mercurial Baro- Eeading of 1 Tempebatuee of Aie. ] meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- Date. index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. Dec. 21st, 1873, 9 p.m. ... 26092 27 25-75 21 20 „ 22nd „ 9 a.m. ... •067 20-5 •75 17-5 16-0 Noon •042 31-5 •67 30 27 3 p.m. ... •012 35 •65 32 29 6 „ 25^977 31 •61 26 24 9 „ '952 28 •59 21 19-9 „ 23rd „ 9 a.m. ... •822 19-5 •48 19-5 lb-5 Noon •772 30 •40 29 26 3 p.m. ... •712 33 •35 31 28 6 „ •702 31 •35 22 20 9 „ •752 30 •40 26-5 24 „ 24ih „ 9 a.m. ... •942 36 •57 345 29-6 Noon •932 48 •65 47 47 3 p.m. ... •952 47 •66 47 40 6 „ •952 36 •69 33 32 9 „ 26-002 30-5 •62 26-5 25-5 „ 25tli „ 9 a.m. ... •042 19-0 •66 17 16 Noon •027 32 •64 31 28 3 p.m. ... 25-967 36 •69 33 31 6 •962 31-5 •68 26 24 9 „ •972 28 ■61 24 23 „ 26th „ 9 a.m. ... •922 20 •66 17 16-6 Noon •952 34 •67 32 28 3 p.m. ... •922 35 •56 323 29-3 6 „ •952 32 -59 26 24 9 „ •982 26-5 •65 21 19 „ 27tli „ 9 a.m. ... 26-017 24 •65 24 21-8 Noon •002 32 •63 31-5 28-6 3 P.M. ... 25-952 36 •58 34 32 6 „ •922 32 •66 27 26 9 „ •942 27 •59 20 19 „ 28th „ 9 a.m. ... 26oy7 16 •75 12 11 Noon •127 29-5 •77 28 243 3 p.m. ... •107 33 •75 30 27 6 „ •102 29-5 •73 24 22 9 „ •097 24 •72 18 15-5 ... 29th „ 9 A.M. ... •022 18-5 •67 14 13 5, , ^v w** ,1 Noon •007 31-5 •65 29 27 3 p.m. ... •012 36 •65 33 31 6 „ •027 30 •70 26 24-5 9 „ •077 27 •75 21 20 „ 30th „ 9 a.m. ... •192 10 •86 9 8 " -»-r JN oon •182 23 •85 21 19 3 p.m. ... •142 28 •80 26 23 6 „ •122 25 •77 18 17 9 „ •102 21 •74 13 12 „ Slst „ 9 a.m. ... •052 10 •70 14 14 Noon •007 24-5 •66 22 21 3 p.m. ... 25-952 28 -61 24 22 6 „ •952 26 •61 19 . 18 9 „ ... •962 22 •62 16 14 ( 89 ) Tandits at YARKAND during the winter of 1873-74. — (Continued.) MiNIMUU IN SHADE. Maximum in shade. During preceding 24 hours. Degrees. '8-9 10 15 10-5 11-8 2-5 Degrees. 29-'8 33 29-5 47 46-5 33-3 34 28 30 24 Direction of wind. S.W. W. S. N. N.- W. S.E. E. S.W. N. N. W. W. E. S.W. N. W. N.W. E. E. N. W. W. S.E. W. N. S.W. N. E. W. E N.W. W. S.E. N. E. E. N.E. N. E. E. W. N.W. N. N. N.W. W. N. S.E. W. N. BSMABKS. For the month of December 1873. 9 P.M. Noon. 3 P.M. 6 p.m. 9 p.m Mean of Mercurial Barometer No. 720 ... 26075 26'067 26042 26047 26-060 Mean of Dry Bulb Thermometer... 210 323 353 283 24'1 Mean of Wet Bulb Thermometer ... 197 29*5 32-1 26-4 23-3 Mean of Minimum Thermometer Mean of Maximum Thermometer in shade ... 10-8 86-3 ( 90 ) Meteorological Observations recorded by the Great Trigonometrical Mercurial Baro- Reading of Tempeeatuee of Aie. Date. meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. Jan. 1st, 1874, 9 a.m. ... 25-952 10 25-61 10-3 9-3 Noon ... •952 24-8 •61 25 23 3 p.m. ... •927 28 •57 27 25 6 „ ... •922 26 •55 21 20 9 „ ... •932 23 •58 17-5 16 „ 2nd „ 9 a.m. ... •927 12 •60 13 12 Noon ... •932 21-5 •61 22 20 3 p.m. ... •927 26-5 ■60 24-5 22 6 „ ... •922 23 •60 19 18^5 9 „ ... •942 20-5 •63 16 15 „ 3rd „ 9 a.m. ... 26-052 11 •72 12-0 12 Noon ... •052 22-5 •72 22 19-5 3 P.M. ... •042 28-5 •70 26 23 6 „ ... •052 24-5 •72 18-5 17 9 „ ... •072 21 •74 15-5 14 „ 4th „ 9 a.m. ... 26-062 11 25-73 11 10 Noon. ... •052 25 •71 23 21 3 p.m. ... •047 28 •71 26 23 6 „ ... •072 25 -73 21 19 9 „ ... •082 21 •76 15 14 „ 5th „ 9 a.m. ... •117 10 •80 11 10 Noon ... •107 25 •78 23 21 3 p.m. ... •072 27-5 •72 25 23 6 „ ... •062 25 •74 20 19 9 •072 21 •76 16 15 6th „ 9 a.m. ... •067 10 •83 8 7 Noon ... •152 23-5 •82 22 19 3 p.m. ... •127 27 •78 24 21 6 „ ... •152 25 •80 20 18 9 „ ... •152 22 •80 17 16 7th „ 9 a.m. ... •192 11 •87 13 12 Noon ... •192 21 •86 215 19-5 3 p.m. ... •167 25 •81 24 21 6 „ ... •172 23 •84 20 19 9 » ... •197 22 •86 19 18 „ 8th „ 9 A.M. ... •152 18-8 ■81 17 16 Noon ... •117 25-8 •78 25 22-5 -^ 3 p.m. ... •077 27 •73 25-5 23 6 „ ... •082 23-3 •74 19 18 9 •062 20 •73 16 15 „ 9th „ 9 A.M. ... .067 10 •73 10 9 Noon ... •052 25 •70 23 21 3 p.m. ... 25-992 28 •66 26 23 6 „ ... 26-007 25 •68 21 19 9 •012 21 •69 18 16 „ 10th „ 9 a.m. ... •027 11-5 •70 10 9 Noon ... •022 26-5 •67 26 23 3 p.m. ... 25-972 31 •63 29 27 6 „ ... •972 27 •63 24 23 9 „ ... •952 24 •61 21 19 „ 11th „ 9 a.m. ... 26-202 13 •87 13 12 Noon ... •302 25 •95 23-5 20 ( 91 ) Survey Pandits at YARKAND during the winter q/^ 1873-74!. — (Continued). Minimum in shade. Maximum in shade. Direction of Remarks. wind. During preceding 24 hours. Degrees. Degrees. 23 N. S.E. W. N. S. 7-5 19 W. s. s. N.W. W. 6 29-5 s.w. N. S.E. W. N. 2 23 N. N.W. N. N. W. 1-8 24 N. N.E. E. N. S.E. 1-3 25 N.E. E. W. W. 3 22-5 N.W. W. S.W. N.E. N.W. 9 26 S.W. E. S.E. E. N. 1 29-6 S.W. N.E. E. E. W. 3 23 S. N. W. w. N. 4 31-5 W. N.E. 24i ( 92 ) Meteorological Observations recorded hy the Great Trigonometrical Survey Mercurial Baro- Reading of Tempebatuee of Aib. Date. meter No. 720 corrected for Temperature of mercury. aneroid barome- ter Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. ■ Degrees. Inches. Degrees. Degrees. Jan. 11th 1874 3 p.m. .. 26-352 28-3 26-01 26 23 6 „ .. •382 24 •05 19 18 9 „ .. •412 21 •08 14 13 „ 12th „ 9 A.M. .. •402 17-5 •05 15 14 JN"oon . . •342 26 •01 26 21 3 P.M. .. •302 28-5 25-94 26 23 6 „ .. •262 23-5 •93 20 18 9 „ .. •252 20 •90 14 13 „ 13th „ 9 A.M. .. •192 10 •85 10 8 Noon . . •152 25 •80 23-5 20 3 P.M. .. •102 28 •76 27 23 6 „ .. •092 24-5 •75 20 17 9 „ .. •072 19-8 •73 15 14 „ 14th „ 9 A.M. . . •052 10 •72 10 9 Noon .. •042 25 •70 24 21 3 P.M. .. •022 29 -67 27 24 6 „ .. •032 25 •70 21 19 9 „ .. •042 20-3 •71 17 14-5 „ 15th „ 9 A.M. .. •112 11 •80 11 10 Noon . . •107 25 •77 24 21 3 P.M. .. •102 29 •74 27 23-5 • 6 „ .. •122 25 •78 20 18 9 „ .. •152 21 •80 16 15 „ 16th 99 t7 A.M. •« •207 11 •88 10 9 Noon . . •222 27 '89 26 22 3 P.M. .. •202 30 •87 28 25 6 „ .. •242 26-5 •90 22 20 9 „ .. •257 21 •92 17 15 „ 17th „ 9 a.m. .. 26-312 12-5 26-01 13 11 Noon . . •327 27 25-98 26 22- 3 p.m. .. •302 31 -95 29 26 6 „ .. •302 27 •95 25 23 9 „ .. •292 24 •94 20 18 „ 18th „ 9 a.m. ,. -207 11 -88 13 12 Noon .. •202 27-5 •87 28 24 3 p.m. .. •107 32 •73 30 27 6 „ .. •107 28 •77 24 22 9 „ .. •102 24 •74 20 18 „ 19th „ 9 a.m. .. •077 12 •73 13 11 Noon . . •072 27 •72 27 23 3 p.m. .. •052 31 •69 31 28 6 „ .. •057 28 •70 23 20 9 „ .. •082 23 •72 18 16 „ 20th „ 9 a.m. ,. •062 18-5 •71 12 10 Noon . . '052 31 -70 30 26 3 P.M. .. '052 33 •71 33 31 6 „ .. •052 30 •73 31 28 9 „ .. •027 27-5 -68 22 20 „ 2l8t „ 9 a.m. .. 25'932 22 -58 21 18-5 Noon . . •927 30 •56 30-5 27-5 3 p.m. .. •862 32 -51 31 28-5 6 „ .. •867 30 •64 28 26 ( 93 ) JPandits at YARKAND during the lointer of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of Eemabes. wind. During preceding 24 hours. Degrees. Degrees. N.W. W. S. 3 24 S.E. W. E. W. E. 1 23 E. S.E. N.E. N.W. N. 0-6 23-5 S.W. N.E. S.E. N. N. 1 25-5 W. N.W. W. s. N. 1 25 IS. s. E. E. S. 1 25 N.W. S.E. N.E. E. ^ N. 2-5 27 S.W. w. S.E. E. N. 4 30 E. N.W, S. E. N.E. 7-5 33 S.W. s. E. S. w. 10-8 34-8 s. N. N.E. W. ( 94 ) Meteorological Observations recorded hy the Great Trigonometrical Mercurial Bai'o- Eeading of Tempeeattjee of Aie. Date. meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. Jan. 21st, 1874, 9 P.M. ... •892 29 •56 26-5 22-5 „ 22nd „ 9 a.m. ... •902 25 •55 24 21 Noon . . . •912 31 •56 32 28 3 p.m. ... •912 34 •56 33-5 27 6 „ ... •942 30 •60 26 23 9 „ ... •947 27 •64 22-5 20-5 „ 23rd „ 9 a.m. ... •952 23 •62 22 19 Neon ... •947 31-5 •58 32 27-5 3 P.M. ... •917 33 •57 33 30 6 „ ... •952 31-8 •62 28 26 9 „ ... •972 28-5 •67 25 24 „ 24th „ 9 a.m. ... 26-062 20-5 •72 19 27-3 Noon ... •062 33-5 •72 32-5 29 3 p.m. ... •052 37-5 •69 36 34-5 6 „ ... •057 340 •73 29 28 9 „ ... •072 30 •75 26 23 „ 25th „ 9 a.m. ... •092 21 •75 20-5 19-5 Noon . . . •042 35 •67 34-5 30-3 3 p.m. ... 25-982 38 •62 37 34-5 6 •972 34 •62 30 29 9 „ ... •972 32 •61 29 25 „ 26th „ 9 a.m. ... •992 28 •62 27 25 Noon . . . •977 39 •62 40 36 3 p.m. ... •972 40 •60 38 33-5 6 „ ... 26^002 36 •65 31 29 9 „ ... •007 30-5 •68 26 24 „ 27th „ 9 a.m. ... •022 22 •69 21 19 Noon ... •042 32 •69 33 29 3 p.m. ... •017 38 •66 37 34 6 „ ... •022 33-5 •69 30 28-5 9 „ ... -032 29 •70 24-5 23 „ 28th „ 9 a.m. ... 26-132 21 •80 19 17 Noon ... •132 34 •80 33 31 3 P.M. ... •112 37-5 •75 35 30 6 „ ... •122 33-5 •79 28 25 9 „ ... •172 32 •84 29-5 25-3 „ 29th „ 9 a.m. ... •142 28 •80 29 26 Noon ... •142 35 •81 36 32 3 p.m. ... •092 38 •73 36 31 6 „ ... •102 33 •77 29 26 9 „ ... •097 30 •76 24 22 „ 30th „ 9 a.m. ... 26-062 22-5 25-78 21 19 Noon . . . •052 35-5 •68 35 31 3 P.M. ... •002 38 •64 37 31 6 „ ... •007 345 •65 31 26 9 „ •012 31 .66 27 24 „ 31st „ 9 A.M. . . . •032 22 •69 21 19 Noon , . . •022 34-8 •67 33-5 29 3 p.m. ... 25-997 38-5 • ^63 36-8 33-5 6 „ ... 26-002 34 •65 31 27 9 „ ... •007 32 •67 29 26 ( 95 ) Survey Fandits at YARKAND dtiring the icinter of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of Eemaeks. wind. During preceding 24 hours. Degrees. Degrees. N.W. 19 34 S.W. N.W. N. W. N.E. 17 38 S.W. S. E. N.E. N. 9 89 N.W. E. E. N.W. N. 9 33-8 N.W. N.E. S.W. E. N. 14 . 34-5 S.W. E. S.E. S. w. 11 43-5 w. N. N. E. N.E. N.W. For the month of January 1874 "V-s 42 9 A.M. Noon. 3 P.M. 6 P.M. 9 p.m. N.E. Mean of Mercurial 26-093 26-087 26-060 26-068 26-074 N. Barometer No. 720 S. E corrected to 32°. 17 34 W. Mean of Dry Bulb 15-5 278 300 242 204 E. S W Thermojneter. N. Mean of Wet Bulb 14-0 24-5 26*8 22-2 IS'o N.E. Thermometer. ii 41 S.W. N.E. Mean of minimum thermometer . . . 6°2 6°'2 N.E. S. E. Mean of maximum thermometer in shade ... 29°7 29°'7 8 34-5 S.W. S.E. S.E. N. E. 25 ( 96 ) Meteorological Observations recorded hy the Great Trigonometrical Survey Mercuria Baro- Reading of Tempeeatuee of Aie. Date ^^^^"^ -^ correcte 0.720 d for Temperature of mercury. Aneroid Barome- ter Solomon's index ( ;rror. uncorrected. Dry bulb. Wet bulb. Inch es. Degrees. Inches, Degrees. Degrees. Feb. 1st, 1874, 9 a.m. "... 25' 982 26 25-63 25-5 23 Noon ... 977 35 •62 36 32 3 P.M. ... 907 38 •56 37 32-5 6 „ ... 912 34 •58 30 26 9 „ ... 917 31 •59 26 23 „ 2nd „ 9 A.M. ... 902 20 •58 19 17 Noon ... 897 36 •56 35 31-5 3 P.M. ... 872 42 •52 41 35 6 „ ... 892 38 •53 34 32 9 „ ... 912 33-5 •59 28 25 „ 3rd „ 9 a.m. ... 982 24 •65 24 21 Noon ... 977 36 •62 36 32 3 P.M. ... 952 42 •68 41 35 6 „ ... •942 38-5 •57 31 28 9 „ ... 962 33 •62 29 26 „ 4th „ 9 a.m. ... •962 25 •62 24 21 Noon ... 962 37 •62 37 32-5 3 P.M. ... 927 42-5 •56 41 36 6 „ ... 927 39 •58 32 28 9 „ ... 947 33-3 •60 27 25 „ 5th „ 9 A.M. ... •952 26-5 •61 25-5 225 Noon ... 952 37-5 •60 37 31-5 3 p.m. ... •922 42 •55 42 36 6 „ ... •927 39 •58 32 28 9 „ ... •952 32-5 •61 26 23 „ 6th „ 9 a.m. ... •932 26-5 •62 26 23 Noon ... •927 36 •60 36 31 3 p.m. ... •902 42 •65 41 36 6 „ ... •902 37-6 •65 33 29 9 „ ... •907 33 •68 29 26 „ 7th „ 9 a.m. ... •907 25-5 •68 24 22 Noon ... •922 38 •69 38 32 3 p.m. ... •892 42-5 •62 41 34-5 6 „ ... •892 38 •62 32 28 9 „ ... •897 34 •66 29 26 „ 8th „ 9 a.m. ... •952 24 •50 23 20-5 Noon ... •792 36 •42 36 31 3 P.M. ... •677 41 •31 40 36 6 „ ... •702 36 •36 33 32 9 „ ... •722 34 •37 30 26 „ 9th „ 9 a.m. ... •732 31 •41 31-5 28 Noon ... •712 38 •36 38^ 33 3 p.m. ... •712 41 •37 4ld 34 6 „ ... •752 38 •40 35 31 9 „ ... •757 35 •41 33 30 „ 10th „ 9 a.m. ... •897 31-8 •65 31 28 Noon ... •927 40-3 •57 41 36-3 3 p.m. ... •922 39 •57 38-5 34-5 6 „ ... 947 35^5 •63 33 31 9 •952 32 •64 31 29 „ nth „ 9 a.m. ... •937 31^5 •59 33 31-5 Noon ... 902 37 •56 38 35-3 ( 97 ) Pandits at YABKAND during the month of 1873-74. — (Continued). MiNIMITM IN SHADE. Maximum in shade. Direction of Bemabes. wind. During preceding 24 hours. Degrees. Degrees. 14-3 36 S. s.w. E. N.E. N. 8-5 42 N.W. W. E. W. N. 12-8 38-5 N. w. . s. w. E. 13 39- N.W. N.E. W. E. \ N. 16-5 40 S.W. S.W. w. w. N. 12-8 38-5 w. N.E. E. N. S. 11 43 N.W. N. N.E. E. N. 9-8 45 N.E. W. N.E. N. N. 16-5 49 S.W. N.E. W. N. W. 24-5 49-5 N.W. E. E. W. W. 22 48 N.E. Clouds a ad snow. N. ( 98 ) Meteorological Observations recorded hy the Great Trigonometrical Mercurial Baro- Eeadic g of Tempeeatuee of Aie. Date. meter No. 720 corrected for Temperature of Aneroid B mercury. Solomo urometer a's un- index error. corret 5ted. Dry bulb. Wet bulb. Inches. Degrees. Inc) les. Degrees. Degrees. Feh. 11th, 1874, 3 p.m. ... 25 -872 37-5 2 5-52 37 35 6 „ ... •892 34 •55 32 30 9 „ ... •902 30 •57 28 27 „ 12th „ 9 am. ... •942 30 •61 29-5 27 Noon ... •937 37 •59 35 32 3 p.m. ... •902 39^5 •54 39-3 34 6 „ ... •902 36 -57 30 27 9 „ ... 912 31 •60 27 25 „ 13th „ 9 a.m. ... 9-92 28 •66 ■26-5 25 Noon ... 26-002 36 •67 35 31 3 P.M. ... 25-977 45-5 •66 38 34 6 „ ... •952 37-5 •65 80-5 28 9 „ ... •972 31-3 •67 26-3 25 „ 14th „ 9 a.m. ... 26-027 29 •68 27-5 25-5 Noon ... •002 38 •65 37 33 3 P.M. ... 25-952 41 •58 39 36 6 „ ... ■912 38 •56 33 28 9 „ ... •952 34 •63 31 26-5 ,, 15th „ 9 a.m. ... 26-052 34 •71 34 32 Noon ... •027 41 •69 41-3 35 3 P.M. ... •022 43 •67 41 35 6 „ ... •022 40 •67 36 31 9 „ ... •012 35 •67 31 27 „ 16th „ 9 a.m. ... •027 26 •71 25 22 Noon ... •042 38-5 •66 37 32 3 p.m. ... •022 43 •60 41 35 6 „ ... 25-952 40 •59 33 28 9 „ ... •052 34 •59 30 26 „ 17th „ 9 a.m. ... •942 34-5 •57 34 31 Noon ... •927 41-5 •65 42-5 37 3 p.m. ... •912 45-5 •54 46 38-5 6 „ ... •932 40-5 •60 35 30 9 „ ... •972 38 •65 38 28-5 „ 18th „ 9 a.m. ... 26-102 36-3 •79 36-^ 31 Noon ... •102 ' 44-5 •70 44 34 3 p.m. ... •092 48 -69 48 38 6 „ ... •082 42 •73 39 31 9 „ ... •102 37 •78 31-5 26 „ 19th „ 9 a.m. ... •172 26 -83 25-5 23-5 Noon ... ■162 34 •76 35-5 32-3 3 P.M. ... .•127 38-5 •78 38 ,33 6 „ ... •127 36 •78 34 29 9 „ ... •132 33-5 •79 32 28 „ 20th „ 9 a.m. ... •092 33-5 •74 34 30-5 Noon ... •072 40 •72 42 36 3 P.M. ... •017 43 •63 44-3 38 6 „ ... 25-997 39 •64 35 30 9 „ ... 26-002 35 •65 30 26 „ 21st „ 9 a.m. ... •042 31-5 •68 32 29 Noon ... •052 41 •70 42 35 3 p.m. ... •017 43-5 •65 43 36 6 •032 41 •69 38 31 ( 99 ) Survey Pandits at YARKAND during the winter 0/ 1873-74. — (Continued). Minimum in shade. Maximum in shade. Direction of Remabes. wind. During preceding 24 hours. Degrees. Degrees. S. E. S. 19-8 48 N.-W. N. N.-E. S. S.-W. 11 41 N. s. E. N. S. 17-5 36 N.-W. S.-W. N.-E. S. S. 22 37 N.-W. N.-E. S. S.-E. N. 16 47 N. E. W. E. E. 22 40 W. S.-W. S.-E. N. S. 21 50-5 w. N. S.-E. E. E. 18 62 N.-W. W. S.-W. S.-E. s. - 24-5 41-6 N.-E. N.-E. S.-E. S. E. 20 50 W. S.-B. E. W. 26 ( 100 ) Meteorological Observations recorded by the Great Trigonometrical Mercurial Baro- Eeading of Tempeeatuee of Aifi. Date. meter No. 720 corrected for Temperature of aneroid barometer mercury. Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inche. s. Degrees. Degrees. Feb. 2l8t, 1874,9 p.m. ... 26-062 37 25- 74 35 27 22nd „ 9 a.m. ... •147 34-5 82 35 31-5 Noon ... ■152 39 82 405 35-5 3 p.m. ... •107 38 72 38^o 35 6 „ ... •02 36 75 35 32 9 „ ... •112 33-5 77 32 30 „ 23rd „ 9 a.m. ... •162 33-5 80 33 31 Noon ... •142 37 77 375 34 3 p.m. ... •092 40-3 75 40 32 6 „ ... •082 37-3 73 335 29 9 „ ... •077 325 74 30 27 24th „ 9 p.m. ... •052 305 72 305 28 Noon ... •002 38 •65 37 34 3 p.m. ... 25932 40 •57 41 34 6 „ ... •912 385 56 36 30-5 9 „ ... •902 36 55 33 28-5 „ 25th „ 9 a.m. ... •852 325 52 31-6 28 Noon . . . •812 40 •47 39 34 3 p.m. ... •757 42 •39 42 35 6 „ ... •752 39 •39 36 29 9 „ ... •752 35 •39 31 26-5. 26th „ 9 a.m. ... •802 368 •46 36 32-5 Noon..., •802 48 •43 47 39 3 p.m. ... •772 51 •37 50-5 40-5 6 „ ... •802 45-5 •44 44-5 36 9 „ ... •877 4^ •53 360 31 27th „ 9a.m. ... 26^002 368 •66 36 31-5 Noon ... •002 44 •64 44 38 3 p.m. ... 25952 47 •62 46 37 6 „ ... •972 44 •63 41 33 9 „ ... 26-002 398 •65 365 31-5 „ 28th „ 9 a.m. ... •002 38 •67 38 33 Noon ... •017 46 •66 46 375 3 p.m. ... 25-992 48 •62 49 39-5 6 „ ... 26-002 45 •68 40 34 9 „ ... •052 40 •70 37-5 31 ( 101 ) Survey Pandits at YARKAND during the winter of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of Remarks. wind. During preceding 24 hours. Degrees. Degrees. W. 26-8 47-8 E. N.W. s.w. E. W. ' 27 43 N.E. N.E. S.E. W. w. 2r8 45 S.E. w. i7-5 '480 N.W. W. N.E. W. W. For the month of February 1874. 9 a.m. Noon. 3 p.m. 6 p.m. 9 p.m. s. Mean of Mercurial barometer No. 720. 25-980 25-971 25-935 25-936 25-962 E. '2b-'8 44-'5 N. S. S. Mean of dry bulb S.W. Thermometer ... 297 38-9 41-6 34-5 30°-7 s. Mean of wet bulb w. Thermometer ... 26-8 33-8 35-5 30-0 27°-0 i9'8 60-5. S.E. Mean of Minimum Thermometer . . . 18°'2 w. Mean of Maximum Thermometer in shade 44°" 1 w. w. N.E. 22 45 W. N. S.E. - W. E. ( 102 ) Meteorological Observations recorded by the Great Trigonometrical Survey 1 j Mercurial Baro- Blading of Tempeeatuee op Aie. Date. meter No. 720 • corrected for Temperature of mercury. aneroid barome- ter Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. March 1st, 1874, 9 a.m. .. 25-997 365 •65 36-8 32-3 Noon •992 455 •61 455 40 3 P.M. .. •882 50^5 •50 605 41 6 „ .. . , -802 45-8 •43 41 34 9 „ .. •802 41 •43 365 305 „ 2nd „ 9 A.M. . . •662 38 •32 39 33 Noon •612 48^3 •25 483 41-3 3 p.m. .. •562 525 •18 54 45 6 „ .. •552 49 •16 42 35-5 9 „ .. •542 44 •15 39 33 „ 3rd „ 9 A.M. .. •582 39 •23 40 35 Noon . . •597 48^5 •24 50^3 44 3 p.m. .. •582 50 •22 50 40-8 6 „ .. 25-652 49 2528 47 39 9 „ .. •682 45 •33 42 36 „ 4th „ 9 A.M. . . •832 45 •48 45 37 Noon •827 50 •46 51 41 3 p.m. .. •802 49^8 •43 49^3 40-8 6 „ .. •842 465 •48 43 37-5 9 „ .. •807 43 •45 40 35 „ 5th „ 9 A.M. . . •822 40 •46 39^8 356 Noon . . •797 49 •44 50 42 3 p.m. .. •767 51-8 •36 52 42 6 „ .. •777 49 •45 47 395 9 „ .. •782 45 •46 43 38 6th „ 9 a.m. .. •832 m-5 •48 40 36 Noon •852 48-8 •49 50 42 3 p.m. .. •792 51 •42 51-3 413 6 „ .. •767 48^8 •40 46 38 9 „ .. •752 45 •36 43 36 „ 7th „ 9 A.M. . . •792 40^3 •44 41-5 32 Noon -752 47-3 •36 50 42 3 P.M. .. •652 50 •28 52 42-3 6 „ .. •612 45^8 •22 43 37 9 „ .. •607 42 •25 406 36 8th „ 9 A.M. .. •597 40 •23 36 33 Noon . . •577 42 •20 43 37 3 p.m. .. •552 413 •18 413 36-8 6 „ .. •602 40-5 •23 38 34 9 „ .. •607 383 •25 366 335 9th „ 9 a.m. .. •697 38 •35 39 335 Noon -692 41 •33 41 37 3 p.m. .. •677 42^8 •32 43 375 6 „ .. •727 41 •36 385 34 9 „ .. •772 36 •41 33 30 „ 10th „ 9 a.m. .. •877 365 •52 37 313 Noon . . •892 45^8 •53 46 396 3 p.m. .. •892 475 •53 473 37 6 „ .. •932 425 •58 403 35 9 „ .. 26^002 375 •64 37-3 30-5 „ 11th „ 9 a.m. .. 137 345 ■77 34 30-3 Noon •162 41 •82 43 « ( 103 ) Pandits at YARKAND during the winter of 1873-74.— (Continued.) Minimum in shade. Maximum in shade. i Direction of wind. Rem AH K S. During preceding 24 hours. -. .- Degrees. Degrees. 19-8 49 N.W. N.E. S. N. N. 22-3 65-5 S.W. S.E. E. S. S.W. 24-8 56-8 w. w. S.E. w. S.E. 34-8 57-5 N.E. N.E. N.E. W. E. 32 57-5 W. S.E. N. N.E W. 37 55-5 W. N.E. N.E. E. S.E. 33-0 55- N.E S.E. E. S.W. N.W. 36 58-5 W. E. N.E. N.E. N.E. 29 44-3 W. N. W. N.E. S. One inch of snow at night. 21 49 N.E. S.E. W. W. W. 29 52 W. E. 27 ( 104 ) Meteorological Observations recorded by the Great Trigonometrical Mercmial Baro- Reading ?of Tempeeatuee of Aib. Date. meter No. 720 corrected for Temperature of aneroid bar mercury. Solomon' ometer 3 un- index error. correct ed. Dry Bulb. Wet Bulb. Inches. Degrees. Inche s. Degrees. Degrees. March 11th, 1874, 3 p.m. ... 26142 45 25 76 45 38-5 6 „ ... •152 40-5 81 37-5 33 9 „ ... •162 39 83 36 33 „ 12th „ 9 a.m. ... •147 36 82 35-8 33 Noon . . . •152 38-5 83 39 37 3 P.M. ... •112 37-3 77 37-8 35 6 „ ... •107 35-3 78 34 35 9 „ ... •102 34-5 77 33 30-5 „ 13th „ 9 A.M. ... •102 35 73 36 32 Noon . . . •072 41-8 71 44 140 3 P.M. ... •057 43-5 67 43-5 38-5 6 „ ... •062 41 69 37 34-5 9 „ ... •092 39 75 37 33-0 ,, 14th „ 9 a.m. ... •122 38-8 78 39 35 Noon ... •112 43-5 76 45-3 40 3 p.m. ... •082 47 71 49 42 6 „ ... •102 42-5 73 40-5 35-3 9 „ ... •117 39 76 36 33 M 15th „ 9 a.m. ... •202 39 86 41 37 Noon . . . •197 47 85 47-5 40 3 p.m. ... •177 48-8 82 47 40-3 6 „ ... •202 41-8 84 38 343 9 „ ... •207 39 85 36 33 „ 16th „ 9 A.M. ... •247 37 89 38 32 Noon . . . •212 42 85 43 36-3 3 p.m. ... •152 43-3 74 43 36 6 „ ... •152 41-5 74 37 31 9 „ ... •122 36-3 73 34 30 „ 17th „ 9 A.M. . . . •032 37 67 37-5 33-5 Noon . . . •002 42-5 64 46 39-3 3 p.m. ... 25-927 47 53 49 40 6 „ ... •902 43 51 39 33 9 „ ... •907 39-8 52 37 31 „ 18th „ 9 A.M. ... •862 37-5 52 39 34-5 Noon . . . •852 46-5 48 48 40 3 p.m. ... •807 39-3 42 51 41 6 „ ... •812 46-5 44 43 36 9 „ ... •817 42-5 45 38-5 33 „ 19th „ 9 A.M. ... •897 39 53 39-8 35-8 Noon . . . •892 50 52 51 41-5 3 p.m. ... •872 52-5 49 54 44 6 „ ... •887 50 52 45 38 9 „ ... •892 44-5 53 38-5 34 » 20th „ 9 a.m. ... 26-052 39-5 69 41 36 Noon . . . •052 50-8 69 53-3 42 3 p.m. ... •032 54-8 65 55 43 6 „ ... •047 51-5 68 47 38 9 „ ... •057 44-8 70 43 37 „ 2l8t „ 9 a.m. ... 25-982 43 64 44 37 Noon . . . •882 52 58 53-5 41-5 3 p.m. ... •817 56 47 56-3 44 6 „ ... •762 49 40 47 37-5 ( 105 ) Survey Pandits at YAEKAND during the winter of 1873-74. — (Continued.) Minimum in shade. Maximum in shade. Direction of EeM ABKS. wind. During preceding 24 hours. Degrees. Degrees. S.E. N. N. - 29 52 N. N. S.W. A fall of half an inch of snow at night. E. S.E. 28 43o E. S.W. W. N. E. 29 51-5 N.W. W. N.E. N. W. Cloudy and slight snow. 29 56-3 W. S.E. E. W. E. 29-5 54-8 E. S.E. W. N.W. N. 24-8 49 N.E. S.E. N. N. N.E. 35-6 550 E. N. E. N. N. 4 35-6 50-5 W. N.E. N.E. E. E. 27-5 57-3 E. N.E. N.W. S.W. s. 27 N.E. E. N.W. S. ( 106 ) Meteorological Observations recorded by the Great Trigonometrical Survey Mercurial Baro- Reading of Tempeeatuee of AlE. Date. meter No. 720 corrected for Temperature of mercury. aneroid barometer Solomon's un- index error. corrected. Dry Bulb. Wet Bulb. Inches. Degrees. Inches. Degrees. Degrees. March 21st, 1874,9 p.m. .. 25-752 46 25^41 43 36 „ 22nd „ 9 a.m... •727 42 •34 42 37 Noon . . •752 46-5 •38 47-5 40 3 p.m. .. •742 53 •35 63 43 6 „ .. ■747 50 •35 46 39 9 „ .. •757 44-5 •38 41 35 „ 23rd „ 9 a.m.... •762 43-8 •41 44^5 39^8 Noon . . •762 50 •38 53 445 3 p.m. ... •742 525 •35 44^5 46 6 „ ... •747 49 •37 45 39 9 „ ... •752 45^8 •41 41 36 „ 24th „ 9 A.M. . . . •822 49 •48 50 44 Noon . . . •817 56^5 •46 60^5 60 3 P.M. ... •797 583 •47 60 49 6 „ ... •802 53^8 •49 50 43 9 „ ... •802 48 •49 43 375 „ 25th „ 9 A.M. ... •902 46 •53 .47 42 Noon . . . •872 57^8 •50 60^3 50^5 3 p.m. ... •852 60 •45 61 50 6 „ ... •852 57 •47 54 45 ( 107 ) Pandits at YARKAND during the winter of 1873-74. — (Concluded.) Minimum in shade. Maximum in shade. ^~"" Direction of wind. E E M A R K S : During preceding 24 hours. Degrees Degrees. 34-5 S. S.E. W. N.W. N.W. For the month of March 1874, W. 9 A.M. Noon. 3 P.M. 6 p.m. 9 p.m. 35-5 W. W. Mean of Mercurial 25-908 25-894 25-858 25-862 25-662 N.W. Barometer No. 720 N.W. W corrected to 32°. "aio'-s W. Mean of Dry Bulb 401 484 496 42-7 371 ■ N.E. Thermometer. W. N. Mean of Wet Bulb 361 41-1 413 36-4 33-8 E. Thermometer. 31 W. s.w. Mean of Minimum Thermometer . . . 29°'8 w. Mean of Maximum Thermometer in shade . . . 52°- 5 w. 28 ( 108 ) Meteorological Observations at EASROMAU during Aneroid Barometer Tempeeatitee Maximum Minimum Minimum No. v., by Aneroid Barometer OF AIE. IN IN IN open. Date. Troughton and Simms, corrected for by Dixey, correct- ed for index error. SHADE. SHADE. Dry bulb. Wet bulb. index error. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees. Degrees. Degrees. Dec. 12th, 1873,9 A.M. 2607 2609 44^0, 15-8 Noon •02 -06 "345 3 p.m. 25-95 •02 34-5 6 „ 2602 •07 „ 13th „ 9 a.m. •00 •06 38 53 Noon 25-98 •04 33 3 p.m. •92 25-99 37 * . • 6 „ •98 26-02 „ 14th „ 9 A.M. 26-00 •05 26 '38-5 ios 6 p.m. 25^93 25-98 „ 15th „ 9 A.M. •89 •95 20 375 "7 Noon •86 •92 32 3 p.m. •80 •88 36 6 „ •87 ■90 „ 16th „ 9 a.m. •82 •88 19-5 37 6-5 Noon •78 •83 36-0 3 p.m. •74 -78 36-0 6 „ ■77 •80 „ 17tn „ 9 a.m. •70 •76 23 38 ii Noon •67 •70 33 ... 3 p.m. •66 •69 33 6 „ •69 •72 „ 18th „ 9 A.M. •67 •74 23 35 ri-5 Noon ■ -67 •72 45-5 3 p.m. -65 •69 475 6 ■„ •69 •73 „ 19th „ 9 A.M. •92 •99 60 30-5 "1 Noon •92 -99 3 p.m. •89 -95 6 „ •89 •95 „ 20th „ 9 A.M. •99 26-04 20 11 Noon •99 •05 36 3 P.M. •95 •01 32 ^ 6 „ •97 -03 26^5 „ 2l8t „ 9 AM. •93 •00 28 "'7-5 Noon •92 25-98 35 3 p.m. •89 •94 32 6 „ •90 •95 27 9 „ •94 •99 24 ... „ 22nd „ 9 a.m. •88 •94 255 29^5 8-6 Noon •84 •89 35 3 p.m. •80 -84 33 6 „ •78 •82 26 „ 23rd „ 9 P.M. •63 •67 22 29 "8-5 Noon •58 •60 31 3 p.m. •50 •54 28^5 6 „ •62 •54 25^6 „ 24th „ 9 A.M. •72 -74 38 40 is Noon •73 -76 44 3 pm. •74 •76 38 6 „ .80 -82 28^5 „ 25th „ 9 A.M. •87 -90 27 35 ii-5 3 p.m. •78 •81 33 6 „ •77 •80 26^5 ( 109 ) the winter of 1873-74 by Captain H. Trotter, B.E., and his Native Assistants. Boiling point Boiling point Boiling point Boiling point . Thermometer Thermometer Thermometer Thermometer Direction No. 17970, No. 17972, No. 17974, No. 17975, of ESMABKS. corrected corrected corrected for corrected for wind. for index error. for index error. index error. index error. Degrees. Degrees. \ 205-20 205-20 W. W. At Kashghar the aneroid barometers and the thermometers (with the exception of 205-07 205-07 s. w. the minimum in open) were placed in a covered porch, opening into a large 205-17 205-15 w. E. court-yard and facing the east ; they were completely protected from the direct rays 205-07 205-05 N.E. N. w. of the sun. '206-12 205-10 '205-07 "205-07 The minimum in open was placed on the E. surface of a table, raised three feet above 205-02 205-00 205-02 206-02 W. W. the ground, and placed near the centj:e of the large court-yard. 204-87 204-85 264-77 204-77 s.w. The thermometers were all graduAted (ffi 204-87 204-85 204-77 204-77 S.E. w. Fahrenheit's scale. 204-72 204-70 204-62 204-62 w. E. W. E. S. E. S.W. s. S.E. '204-62 204-60 '2b4'-57 204-57 '204-62 '204-50 '204-62 204-52 204-54 '204'-52 '204-42 204-42 - '204-42 '204-40 '204-37 204-37 W. S.W. W. S.W. S.W. 206-22 205-20 E. s. w. E. W. <■ '20607 "205-05 "2b5'-02 '265-02 '205-12 '205-10 '204-97 '264-97 --- S.E. 206-02 205-00 204-92 26492 E. E. S.W. 204-97 204-95 204-92 264-92 S.E. W. 204-72 204-70 204-67 264-57 w. S.E. 204-39 204-40 204-27 204-27 w. S.E. 204-17 204-16 S. N. S. S.W. N.E. N. 204-92 204-90 204-79 264-82 W. • 204-72 204-73 204-62 204-62 S.E. S. ( no ) Meteorological Observations at KASSGHAR during the winter Aneroid Barometer Tempeeatxjee Maximum Minimum Minimum No. v., by Aneroid Barometer OF AIE. IN IN IN OPEN.* Datb. Troughton and Simms, corrected for by Dixey, corrected for index error. SHADE. SHADE. Dry bulb. Wet bulb. index error. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees. Degrees. Degrees. Dec. 26th, 1873, 9 a.m. 25-74 2577 23 29 7 Noon •75 •78 33 3 p.m. •72 •74 33 6 „ •78 •80 26 „ 27th „ 9 a.m. •83 •86 22 30 "ib5 Noon •78 •82 34-5 3 p.m. •73 •76 33 6 „ •72 •76 32 „ 28th „ 9 a.m. •92 •97 23 30 25 6 p.m. •88 •94 26 „ 29th „ 9 a.m. •82 •88 24 29 ©•0 3 p.m. •80 •85 30-5 6 „ •88 •91 24-6 „ 30th „ 9 a.m. 2693 26^08 24 2"8^6 "'4-5 Noon •97 •02 29 3 p.m. •93 25^98 27 ... ... „ 3l8t „ 9 a.m. 225 315 9-5 Noon 30 3 P.M. 26 6 „ 22 ( 111 ) of 1873-74 by Captain S. Trotter, R.E., and his Native Assistants. — (Continued.) Boiling point Thermometer No. 17970, corrected for index error. Boiling point Thermometer No. 17972, corrected for index error. E oiling point Thermometer No. 17974, corrected for index error. Boiling point Thermometer No. 17975, corrected for index error. Direction of wind. Bemabks . Degrees. 204-72 264-64 Degrees. 204-67 204-62 204-59 264-57 204-62 '204-57 W. E. S.W. S.E. W. s. S.W. s. w. s. N. S.E. E. N. N.W. S.E. W. S.W. s. E. For the month of December 1873. Mean of Aneroid Baro- ( meter No. V., by 1 9 a.m. Troughton and Simms-{ Noon. 3p.m:. 6 p.m. corrected for index 25'87 error. L 25-83 25-80 25-82 Mean of Aneroid Baro- \ ^ a.m. Noon. 3 p.m. 6 p.m. ed for index error. (_ 25"89 25-88 25-84 25-86 Mean of Dry Bulb TherJ^^"' Noon. 3 P.M. 6 p.m. mometer. ) 241 347 33-5 26-4 Mean of Maximum Thermometer in shade ... 35-0 Mean of Minimum in open ... ... 8-9 29 ( 112 ) Meteorological Observations at KASS.GHAR during the Aneroid Barometer Temperatxiee Maximum Minimum Minimum No. v., by Troughton and Simms, corrected for Aneroid Barometer OF AIE. ^^ IN IN Date. by Dixey corrected for index error. SHADE. SHADE. OPEN. index error. Dry bulb. Wet bulb. During preceding 24 hours. Inches, Inches. Degrees. Degrees. Degrees. Degrees. Degrees. Jan. Ist, 1874, 9 a.m. 155 315 125 „ 8th „ 9 A.M. 18 305 7^5 Noon 26^5 3 p.m. 245 ... 6 „ 20 „ 9th „ 9 A.M. 20 28^5 "lO „ 10th „ 9 A.M. 17-5 29 125 Noon 26 3 P.M. 25 6 „ 20 „ 11th „ 9 A.M. 19 '28^5 "7^5 Noon 24-5 3 p.m. 24 6 „ 20-5 „ 12th „ 9 A.M. ' 26-35 ' 2634 16^5 25 " " 6 Noon •31 •30 28-5 3 p.m. •25 •22 26^0 „ 13th „ 9 a.m. •14 •13 18 '245 '" 5 Noon •08 •08 28 3 p.m. •06 •02 25^5 6 „ •06 •02 21 „ 14th „ 9 A.M. •04 •00 20^5 285 "'7^5 Noon •01 2597 29 3 p.m. 2597 •93 265 6 „ 26-01 •97 19 „ 15th „ 9 A.M. •08 2607 20 29 "7^5 Noon •05 •03 275 3 p.m. •03 •02 26^5 6 „ •06 •05 195 „ 16th „ 9 a.m. •17 •17 23 28^5 "" 6 6 p.m. •20 •20 21 „ 17th „ 9 a.m. •28 •29 25 '3b^5 ""6^5 —55 Noon •26 •27 28 „ 18th „ 9 a.m. •17 •18 235 "315 "lO + 1^6 Noon •10 •12 315 3 p.m. •04 •05 29 6 „ •06 •06 „ 19th „ 9 A.M. "00 •02 225 295 ■■'12 "+2 „ 20th „ 9 A.M, 2599 •00 26-5 28 145 + 5 Noon •97, 25-97 36-5 3 P.M. •96 -96 33 6 » •96 •96 „ 2l8t „ 9 A.M. •87 •87 24 36 "'20 + 10 Noon •82 •82 33 3 p.m. •77 •77 32 ... 6 „ •81 •81 27^7 „ 22nd „ 9 A.M. •83 •84 26^5 '39^5 '225 '""17 Noon •83 •84 ' 36 ... 3 p.m. •85 ■86 34 ... 6 „ ■86 •87 29 „ 23rd „ 9 A.M. •87 •88 235 "3'6 195 „ 24th „ 9 A.M. 26^01 2601 33 35 215 "' 16 Noon •00 •00 375 6 „ •00 •00 ( 113 ) winter q/" 1874-75, by Captain S. Trotter, H.B., and his Native Assistants. Boiling point Boiling point Boiling point Boiling point Thermometer Thermometer Thermometer Thermometer Direction No. 17450, No. 17972, No. 17974, No. 17975, of Eemabes. corrected corrected corrected for corrected for wind. for index error. for index error. index error. index error. Degrees. Degrees. s.w. N.E. S. N. N.W S. N. S. S.W. w. N. N.E. W. s. s. w, E. N.W. S. s. s. N.E. S.W. S.E. S. S.W. S.W. S.E. S.W. S.W. w. S.E. N.E. W. S.W. s. S.E. S.W. W. s. S.W. N. w. The break in the observations from Ist to 11th January caused by Captain Trotter's absence on the expedition to Chadyr-Kul. 204-57 204-55 204-57 204-62 S.W. S.E. W. S.W. E. N.E. S. N. W. S.W. w. , ( 114 ) Meteorological Observations at KASSGSAJB, during the winter of 1874-75, Aneroid Barometer No. v., hy Troughton and Simms, corrected for Aneroid Barometer Tempeeatitee Maximum IN Minimum IN Minimum IN Date. by Dixey corrected for index OF AIE. SHADE. SHADE. OPEN. index error. error. Dry Bulb. •Wet Bulb. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees.' Degrees. Degrees. Jan. 25th, 1874, 9 a.m. 2603 26-03 28 39 14-5 —5 Noon 25-97 25-97 38-5 6 p.m. •92 •92 „ 26th „ 9 i.M. •93 ■93 28 "4^-5 22 13 Noon •90 •90 355 3 p.m. •89 •89 35 6 „ •92 •92 26 35 19-5 ■"7-2 „ 27th „ 9 A.M. •99 •99 26-0 35 19-6 7-2 Noon •96 •96 34 ... 3 p.m. •92 •93 35 6 „ •93 •94 27 „ 28th „ 9 A.M. 26-07 26-08 28-5 37-5 ""20 "'8-5 3 p.m. •04 -05 37 6 „ •08 •08 „ 29th „ 9 A.M. •08 •10 32-5 "42 22-5 14-5 Noon •02 •06 42 ... 3 p.m. 25^98 •02 38 6 „ •98 -02 ... „ 30th „ 9 A.M. -96 •00 22 "■"44 "l7-5 "'6-2 Noon •91 25-96 34-5 ... 3 p.m. •90 •91 34-5 6 „ •91 •92 » . . ... . . • „ 31st „ 9 A.M. •95 •98 29 '37-5 17-6 6 Noon •91 •94 39 ■ . ■ ... 6 P.M. •89 •92 ( 115 ) by Captain S. Trotter, M.E., and his Native Assistants. — (Continued.) Boiling point Thermometer No. 17970, cor- rected values. Boiling point Thermometer No. 17972, corrected values. Boiling point Thermometer No. 17974, corrected values. Boiling point Thermometer No. 17975, corrected values. Direction of wind. Eehabes. Degrees. Degrees. W. s.w. N. N.E. N.E. S.E. S. N.W. W. s. w. w. s. S.E. E. W. S. N.E. S.E. W. S.W. w. s. N. For the month of January/ 1874. Mean of Ane- roid Barome- ter No. V. by Troughton & Simms, cor- rected for in- dex error. 9a.m. Noon. 3p.m. 6p.k. 2604 26-01 25-97 25-98 Mean of Ane- roid Barome- ter by Dixey corrected for index error. 9 A.M. Noon. 3 p.m. 6 p.m. 26-04 26-01 25-97 25-98 Mean of Dry 235 32-4 30-3 22-5 Bulb thermo- meter. Mean of minimum thermometer 13°'6 in shade. Mean of maximum thermometer 33°*0 in shade. Mean of minimum in open ... 7°' 6 30 ( 116 ) Meteorological Observations at KASHGHAB during the winter Date. Aneroid barometer No. V. by Troughton and Simms corrected for Aneroid barometer by Dixey corrected for index - Tempeeatuee or AIE. Maximum IN SHADE. Minimum IN SHADE. Minimum IN OPEN. index error. error. Dry Bulb. Wet Bulb. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees. Degrees. Degrees. Feb. 1st, 1874 9 a.m. 25-88 25-86 29 40-5 22 135 Noon •84 •82 38 3 P.M. •82 •79 35 6 P.M. •82 •78 „ 2nd „ 9 A.M. Noon 3 P.M. 6 P.M. •82 •79 •77 •80 •78 •75 •73 •77 32-5 38 38-5 38-5 i3-5 ""' 4 „ 3rd „ 9 a.m. •89 •86 27-6 41-5 '19-5 *"' 8 Noon 3 p.m. 6 p.m. •88 •84 •86 •85 •80 •82 4^ 37 ... „ 4th „ 9 a.m. Noon 3 p.m. 6 P.M. •90 •86 •83 •84 •86 •83 ■79 •80 '"'35 40-5 39 ""42 19-7 """ 9 „ 5th „ 9 A.M. Noon 3 p.m. 6 p.m. •86 •85 •82 •85 •82 •81 •78 •81 "31 41 40 "42 '"'23 '"15 „ 6th „ 9 a.m. Noon 3 P.M. 6 P.M. •84 •79 •76 •80 •83 •78 •73 •77 32-5 43-5 40-5 32 "4i-5 '20-5 "ll .. 7th „ 9 A.M. Noon 3 P.M. 6 P.M. •84 •83 •76 •77 ■81 •80 •75 •76 37 45 40 31 '47-5 19-7 ""'6-6 „ 8th „ 9 A.M. Noon 3p.m. •72 •67 •68 •73 •66 •58 29-5 £8 39 '44-5 "20 "10 „ 9th „ 9 A.M. Noon 3 p.m. 6 p.m. •61 •60 •58 •60 •69 •58 •57 •69 33-5 39 39 34 40-5 28-5 19-5 „ 10th „ 9 A.M. Noon 3 P.M. •81 •83 •81 •81 •83 •81 39-5 37 39 41-5 '"'29 "24 „ 11th „ 9 A.M. Noon 3 P.M. 6 P.M •80 . •79 •75 •80 •82 •79 •76 •81 31-5 38 40 31 "43-5 28-5 24-5 „ 12th „ 9 A.M Noon 3 P.M 6 P.M •79 •78 •76 •76 •82 •79 •77 •77 335 38-5 36 30 '""46 "24 15-5 „ 13th „ 9 A.M Noon 3 P.M 6 P.M •83 •85 •83 •86 •86 •88 •84 •87 29-5 36-5 35-5 31 39-5 125 ""' 2 „ 14th „ 9 A.M Noon 3 P.M •90 •93 •89 •82 30 39-5 38 38-5 2i5 12-5 ( 117 ) of 1874-75 hy Captain K, Trotter, B. jE'.— (Continued.) Boiling point thermometer No. 17970 cor- rected values. Boiling point thermometer No. 17972 corrected values. Degrees. Degrees. Boiling point thermometer No. 17974. Boiling point thej-mometer No. 17975. Direction of wind. W. N.E. S.E. S.E. W. s.w. w. N.W. N.W. N.E. N. S. w. N.E. W. N. W. s. E. S. S.W. S.E. S.W. s. N.E. S.E. S.E. S.E. S.W. E. E. S.W. s. S.E. S. N.E. E. E. N.W. N.E. N.E. N. W. N.W. W. N.W. N.W. N.E. N. N.W. W. N.W. W. Re U ABES. The break in the readings of aneroid No.V. caused by Captain Trotter's absence io the Artysh Districts. ( 118 ) Meteorological Observations at KASJELGJSAJR, during the winter of • Aneroid barometer Tempeeatuee Maximum Minimum Minimum No. v., by Aneroid barometer OF AIE. IN IN IN Date. Troughton and Simms corrected for by Dixey corrected for index error. SHADE. SHADE. OPEN. Dry Bulb. Wet Bulb. index error. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees. Degrees. Degrees, Peb. 16th, 1874 9 a.m. 25-94 31 42-8 25 12-5 Noon •90 42 28 3 p.m. •88 41 30 „ 16th „ 9 a.m. Noon 3 p.m. •95 •91 •87 37 41 39-5 33 33 33-5 44 23 17 „ 17th „ 9 a.m. Noon 3 p.m. •80 •79 •77 34 42 42-5 32 33 34 42 27 15-5 „ 18th „ 9 a.m. Noon 3 p.m. •99 •96 •94 41 45 40 32-5 35-6 35 44 23-5 21-5 „ 19th „ 9 A.M. Noon 3 p.m. 2610 •08 •03 32-5 33 31-5 29 32 32 36-5 20 15 „ 20th „ 9 A.M. Noon 3 p.m. 25-98 •93 •87 41 43 39 31 32 32 43 22 19 „ 21st „ 9 A.M. Noon 3 p.m. •96 •92 •91 32-5 42 40 29 32 32 44 22-5 14 „ 22nd „ 9 a.m. Noon 3 p.m. 26-08 •04 •00 31 34-5 32 34 32 32 42 30 12 » 23rd „ 9 a.m. Noon •06 •04 32-5 40 31 32-6 36 29 14 « 24th „ 9 A.M. Noon 3 p.m. 25-93 •86 •84 39 39-6 36 31 32-8 32 42 20 20-5 „ 25th „ 9 a.m. Noon 3 p.m. •74 •69 •64 45 40 32 33 32 32 45 26 21 » 26th „ 9 a.m. Noon 3 p.m. •69 •68 •67 44 48 40 33 35 36 45 28 22 „ 27th „ 9 a.m. Noon 3 p.m. •90 •91 •89 42-5 41 40 35 36 36 48 31 19 « 28th „ 9 a.m. Noon 3 p.m. •94 •94 •87 41 49 45-8 32 36-2 35-5 48-5 28 22 ( 119 ) 1873-74, by Captain El. Trotter^ R.E., and his Assistants. — (Continued.) Boiling point Boiling point tliermometer thermometer Boiling point Boiling point Direction of wind. No. 17970 No. 17972 thermometer thermometer Remarks. corrected corrected No. 17974. No. 17975. values. values. Degrees. Degrees. E. E.S. W. w. N.E. N. N.E. W.S. N.E. S.W. N.W. S.W. S.W. E. S.W. s. s. S.E. s. E. S.W. N.E. N.E. E. N. N. N W Slight fall of snow. For the month of February 1874. Mean of Aneroid 9 a.m. Noon. 3 p.m. 6 p.m. NS Barometer No. N.S. V.byTroughton N and Simms ... 25-83 25-80 25-78 25-79 N.E. Mean of Aneroid Barometer by E Dixey 25-87 25-84 25-81 25-78 - N Mean of Dry Bulb N E Thermometer... 34-8 40-4 38-4 31-5 g Mean of Wet Bulb E Thermometer... 31*9 ,32-8 33-2 g Mean of Minimum Thermometer in N* shade ... ... ... 23°-5 N Mean of Maximum Thermometer in S. shade ... ... ... 42°'5 Mean of Minimum in open . . . 15°"0 31 ( 120 ) Meteorohgieal Observations at KASHGSAB during the winter of 1874-75, Aneroid harometer Aneroid barometer Tempeeatuee Maximum IN SHADE. Minimum TV Minimum IN No. V. hy by Carpenter, OF AlE. SHADE. OPEN. Date. Troughton and Simms corrected lor Westley, and Dixey corrected for index Dry Bulb. Wet Bulb. index error. error. During preceding 24 hours. Inches. Inches. Degrees. Degrees. Degrees. Degrees. Degrees. Mar. 1st, 1874, 9 A.M. 25-91 49 39 49 27 29 Noon •86 51 38 3 p.m. •84 ~ 47 36 „ 2nd „ 9 A.M. Noon 3 p.m. •55 •51 •44 45^5 53-6 51 39 40 41 51-5 28 27 „ 3rd „ 9 A.M. Noon 3 P.M. •50 •49 •60 40 42 44 36 38 39 55 32 2r5 „ 4th „ 9 A.M. Noon 3 P.M. •74 •71 •69 43 48-5 49-5 40 39 41 47 36 24 „ 5th „ 9 A.M. '■'25-79 •79 43 39 50-5 37 24 Noon •74 •73 50 40 ... 3 P.M. •72 •70 48-3 41 „ 6th „ 9 A.M. •82 •80 38-5 36 52 31 20 Noon •82 •78 49 41 3 P.M. •73 •73 45-5 40 ... » 7th „ 9 A.M. •72 •73 41-2 38 54-5 37-5 3 P.M. •60 •58 45 39 „ 8th „ 9 A.M. •55 ■54 34-5 33 48-5 34-0 28^0 Noon •53 •52 37^5 36 ... ... 3 P.M. •51 •50 40-5 36 6 P.M. •55 •54 38 36 ... „ 9th „ 9 a.m. •63 •63 35 32-5 42 33-5 27-5 Noon •63 •62 40 35 3 p.m. •62 •61 44 37 6 p.m. •67 •66 37-5 34 „ 10th „ 9 a.m. •80 •80 43-5 36-5 44-5 27-5 22-0 Noon •82 '83 47 36 6 p.m. •87 •88 39-5 36 „ 11th „ 9 a.m. 26-08 26-09 44 39 48-5 26-5 17-0 „ 12th „ 9 A.M. •12 •15 32-5 30 50 29 25 Noon •12 •15 39 35 3 p.m. •08 •10 40 36 6 P.M. •07 •08 37-5 36 „ 13th „ 9 a.m. •03 •05 39 35-5 47 30-5 26 Noon 25-98 •00 44 38 ... 3 p.m. •97 25-99 m 38 6 P.M. •97 •99 36-5 36 „ 14th „ 9 a.m. 26-09 2611 37-5 36 45 33^5 27-5 Noon •07 •08 46 40 3 p.m. •02 •04 41 39-5 6p.m •06 •08 39 37-5 „ 15th „ 9 A.M. •16 •19 34-5 32 47 3i5 27-5 Noon •15 •19 40-5 35 6 p.m. '14 •17 375 34-5 . „ 16th „ 9 A.M. •20 •24 34 30 44 315 27 Noon •17 •19 41 34 3 p.m. •11 •12 44 34 6 p.m. •07 •08 38-5 35 ... ( 121 ) by Captain H. Trotter^ R.E., and his Native Assistants. — fConcluded.J Boiling point Boiling point thermometer thermometer Boiling point Boiling point No. 17970 No. 17972 thermometer thermometer wind. Remabks. corrected corrected No. 17974. No. 17975. values. values. Degrees. Degrees. • S. N. S. For the month of March 1874. N. Barometer No. 9 a.m. Noon. 3 p.m. 6 p.m. s N.E. andSimms ... 25-92 25-90 25-82 25-93 N. Mean of Aneroid N. Barometer by S. Dixey and Car- N.E. penterand N. Westley ... 25*80 25-78 25-72 S. Mean of Diy Bulb S. Thermometer... 39-7 44-9 44-6 38-0 N. Mean of Wet Bulb N. Thermometer... 35-7 37-5 38-3 35-6 S.E. Mean of minimum thermometer ... 31°-6 S. Mean of maximum do. in shade 48°-5 N. NW. S. Mean of minimum rod in open ... 24°- 9 N.E. , N.E. S.E. S. Snow falling. N.E. Snow at night. N.E. N. N. , * ■ N.W. N.W. N.W. s.w. , N. N. N. N.W. Heavy fall of snow during night of about N.W. five inches. N.W. N.W. N.W. N. N. N.W. 205-25 205-'25 N.W. N.E. N.E. N.E. N.E. N.E. N.E. N.E. Fall of enow during night. GEOGRAPHICAL APPENDIX. SECTION G. ROUTES. 32 GEOGRAPHICAL APPENDIX. SECTION G.— ROUTES. INDEX. Paet I. — Routes traversed hy members and employes of the Mission. ^Stl"*^ From. To. Remarks. Authoritj. I. Yarkand ... ... Leh (Ladakh) ... Vi& Sanju and Kara- Dr. Bellew. korum Pass. II. Ditto ... ... Leh ... ... Via Kugiar and Kara- Ditto. korum Pass. III. Leh ... ... Shahidula (No. 12 of Fi4 Changchenmo ... Captain Biddulph. Route I.) IIIo. Gogra ... ... Ditto ... ... Variation on No. III. „ Trotter. IV. Yarkand ... ... Kashghar ... „ „ V. Kashghar ... Maralbashi ... „ Biddulph. VI. Ditto ... ... ChadyrKul ... „ Trotter. VII. Ditto ... ...• Belowti Pass ... On road to Ush Tur- „ „ fan. VIII. Yangi Hissar (No. 3 Kila Panjah (Wakhan) Via Tashkurghan and „ Biddulph. of Route IV.) Little Pamir. IX. Kila Panjah ... Aktash (No. 11 of Via Great Pamir. Route VIII.) X. Tashkurghan (No. 9 of Yarkand ... ... Vid Charling River ... „ Trotter (from Pundit Route VIII.) Kishen Sing). XI. Kila Panjah ... Kila Wamur (Roshan) „ Trotter (from Abdul Subhan). XII. Yarkand... ... Aksu ... ... Fid Maralbashi ... „ Chapman (from Bhao Sing). XIII. Aksu ... ... Kuchar ... ... „ ditto. XIV. Karghalik (No. 2 of Tankse (No. 5 of Vid Khotan, Keria, and „ Trotter (from Kishen Route I.) Route III.) Polu, vide Sing). JAN a* Yarkand ... ... Khotan ... ... Vid Karghalik ... „ Chapman (from Ram- chand). Part II. — Routes in Turkestan derived from verbal information supplied by natives. gpyj-g From. To. Remarks. Authority. XV. Kuchar (Route XIII.) Karashahr ... Compiled by Captain Trotter. XVI. Kuchar ... ... Turfan ... ... Ditto ditto. XVII. Turfan ... ... Kamul, Khamil, or Ditto Dr. Bellew. Hami. XVIIo. Turfan ... ... Orumchi or Orumtsi . . . Ditto ditto. XVIII. tJrumtsi or Orumchi .. . Manas ... ... Ditto ditto. XIX. Manas ... ... Yulduz ... ... Ditto ditto. XX. Yulduz ... ... Ghulja, or Kuldja, or Ditto ditto. Hi. XXI. Kurla (No. 8 of Route Lob ... ... Ditto ditto. XV.) XXII. Karashahr (Route XV.) Yulduz Valley ... Ditto ditto. XXIII. Khotan (Route XIV.)... Aksu (Route XII.) ... Ditto Captain Chapman XXIV. Khotan ... ... Polu (No. 14 of Route Ditto ditto. XIV.) * This itinerary includes some notes on Khotan, compiled by Captain Chapman. — H. T. Paet II. — Routes in Turkestan derived from verbal information Bemarks. Number of Route. From. To, XXV. Khotan ... Charchand XXVI. Kila Wamur (Roshan), Route XI. Kila Khumb (Darwaz) XXVII. Bar Panjah(Shighnan), No. 10 of Route XI. Kashghar (Route IV.) XXVIII. Bughrumal Pamir (No. 4 of Route XXVII.) Khokand... XXIX. Yur (near Kila Pan j ah Route VIII.) Chitral ... XXX. Tashkurghan (No. 9 of Route VIII.) Hanza (Kanjud) XXXI. Sarhadd (No. 14 of Route VIII.) Kanjud ... XXXII. Yarkand ... Shahidula Via Kilik Pass supplied hy natives. Authority. Compiled by Captain Trotter. ' Ditto ditto (from Abdul Subhan.) Ditto ditto. Ditto ditto. Ditto Captain Biddulph. Ditto Dr. Bellew. Ditto ditto. Ditto Captain Trotter, GEOGRAPHICAL APPENDIX. SECTION G. ROUTES TRAVEESED BY MEMBERS AND EMPLOYES OP THE MISSION. ROUTE I. Yarkand to Leh via Sanju and Karakoram Pass (Dr. Bellew, October AND November 1873). 1. Yarkand to Posgam (height* 4,210 feet), 17 miles. — Across a cultivated plain covered thickly with farmsteads and traversed by numerous irrigation streams. Trees along the water-courses, and orchards round the farmsteads. Cultivation interrupted by meadows and marshes. At three miles from Posgam cross the Zarafshan river, which flows in two channels separated by patches of tamarisk jangal. Ford across a firm pebbly bed between low sandy banks four to five hundred yards apart, Posgam is a market town of about 600 houses. Galled also Charshamba Bazar. 2. Karghalik (height 4,370 feet), 24 miles. — Cultivated plain, farmsteads and fields, with marshes and jangal patches between. At eight miles cross Tiznaf river. I'ord firm and pebbly between low sandy banks 80 to 100 yards apart. At five miles on pass through Yakshamba Bazar, 300 houses. Then across thin cultivation between patches of saline encrustation, marsh, and waste to Karghalik, 1,000 houses. A market town with widespread farmsteads. Many trees and many water-courses. 3. Bora (height 5,340 feet), 25 miles. — Soon pass beyond cultivation across a stony desert waste six miles to Besharik, or " five streams,^' a populous settlement of farms on the water-courses in a wide hollow running from west to east. Then cross an arid and wide waste of coarse gravel to another hollow, deeper and narrower. In this is the settlement of Bora, 30 to 40 homesteads on the course of the stream from which the cultivation is irrigated. Trees in plenty. 4. Oi-toghrak (height 5,760 feet), 12 miles. — Across an arid desert of undulating surface, coarse gravel, and wind blown ridges of sand, very scanty herbal vegetation, to a deep and winding drainage gully in which, on course of its stream, is the Oi-toghrak settle- ment of 15 to 20 farmsteads. Trees few. 5. Khushtagh, 19 miles. — Across an arid, undulating desert waste of sandy gravel for 18 miles ; then cross a wide boulder strewn hollow with thin tamarisk jangal, and pass through a belt of tall reeds to Khushtagh settlement in a wide hollow. Farmsteads for some miles along the course of the Kilian stream which flows eastward to Giima. 6. Sanju (height 6,070 feet), 25 miles. — Cross arid strip of desert as before, eight miles to a dry ravine in which are four or five farmsteads watered from springs ; this is Langar, and here is a roadside rest-house and tank of water under the shade of tall poplars. From this up a steep bank and across a ridgy desert as before for 1 6 miles to the Sanju valley down a steep sandy slope. The road to Giima branches off north-east on this desert. Sanju is a populous settlement along the course of a river which flows towards Khotan, and is forded on a rough boulder bed. Farmsteads, orchards, and fields here extend in unbroken succession for eight or ten miles along the river. 7. Kiwaz, 13^ miles. — Pass through Sanju settlement five or six miles to high banks of gravel and red sand ; then up a narrowing valley along the Sirikia river, which runs in three or four streams and is crossed twice en route on a boulder bottom, to Kiwaz ; six or eight scattered huts on a limited flat amongst hills. * The heights throughout these routes are supplied by Captain Trotter. 33 ( 128 ) 8. Tarn (height 8,790 feet), 16^ miles. — Up course of Sirigkya river, through a gradually narrowing valley which winds between high and bare hills of schistose slate. River crossed repeatedly en route on a stony bed. Banks fringed with bushes and patches of pasture grass. At 1 1 miles pass the Chuchu glen to the left ; a narrow defile which conducts over the Chuchu ridge to Shahidula, and is taken as an alternative route when the river is unfordable during the summer floods. At Tarn two houses on a small flat leave habitation behind. 9. Gachga, 10 miles. — Up narrow winding valley, and cross river repeatedly as in last stage. No cultivation or habitation. Kirghiz camps in glens and hollows in the vicinity. 10. Kichik Karakoram, 14 miles. — Up by a rapid rise through a widening and branching defile to foot of Sanju Dawan, a sharp ridge of mica slate 16,650 feet high. Then up a steep zigzag, through a narrow and rough gap, and down another on opposite side into a very narrow, deep, rough gorge descending to Kichik Karakoram ; a narrow strip of turf on a trickling stream between lofty vertical cliffs. No fuel nor pasture. 11. Pillataghach, 11^ miles. — Descend narrow, winding, dismal gorge over masses of landslip rock, down course of rivulet for four miles. Then enter valley of Karakash river at Mirza Ababakar camp ground at a cluster of graves on the river bank. Then follow up stream six miles by a very rough road, fording river twice en route to Pillataghach camp ground on a limited flat of brushwood and pasture on the river bank. Valley very narrow ; hills high and bare. 12. Shahidula Khoja (height 11,780 feet), 1 5 miles. — Up stream by rough road in winding valley, with brushwood and forage along river course ; their patches interrupted by projecting moraine banks. At four miles cross Kilyan stream from the right to Korghan, a solitary mud castle at foot of a rock abutting on the river bank. At five miles on cross Toghra stream from the right, then cross Karakash river three times en route to the Fort of Shahidula Khoja, garrison 30 men; frontier post of Kashghar at the junction of the Kizil jangal glen with Karakash valley. Fuel and forage here, and Kirghiz camps around. 13. Sugat (height 12,970 feet), 8 miles. — Up course of Karakash river four miles. Then up course of the Sugat river to the right four miles, and, crossing several times, camp on turfy flat on right bank. Hills on left bank steep down to the river ; on right bank rolling away in wide slopes to high mountains ; everywhere bare schistose slate, and trap. Vegetation confined to river course. 14. Chibra, 21 miles. — Rise out of river channel and pass across wide slopes of hill to a narrow defile coming down from the left. Then up its course between bare banks of shale through a tortuous channel to foot of Sugat Pass, 17,600 feet high. Ascend by a steep path, and follow a gradual slope six miles down to camp ground at Chibra. No vege- tation here. Water very scanty. Snow on Pass from September to April as on Sanju Pass. 15. Aktagh (height 15,590 feet), 10 miles. — Over an elevated, arid, stony plateau, perfectly desert, by a path skirting banks of shale to the right. Breathing oppressed on this march. At ten miles turn slowly to right and slope down to Aktagh camping ground on a patch of turf in the wide, shallow, shingly bed of its stream. The whole region a bleak, desolate, and inhospitable waste. From this down stream is the Yangi Diwan and Kugid,r route to Yarkand. 16. Brangsa Karakoram, 28 miles. — Up a wide, shallow, shingly drainage bed gradually ascend- ing between low banks of shale that roll away in wide sweeps to the mountain tops. Vegetation most scanty in herbal tufts. A few antelopes met with. At half-way pass camp ground of Wahabjilga, where the Aktagh stream flows through a cutting in slate rocks. Then continue over the drainage bed to the Brangsa camp ground at the entrance of a narrow defile. No fuel and no forage in all this region. This Brangsa is also called Balti Brangsa. ( 129 ) 17. Daulatbeg- Uldi (height 16^880 feet), 22 miles. — Through a narrow gorge up course of a little torrent for a mile ; then enter a wide gully branching off amongst the hiUs. Soil soft and spongy, slate detritus. Rise gradually to foot of Karakoram Pass, 18,550 feet high, then up a short ascent and down a steep descent over soft clay to a hill slope along the course of a rivulet, and cross it several times en, route to camp. The pass is half-way on this march. Breathing affected by the elevation on this wide plateau. Surface bare gravel and clay. Prom Daulatbeg there are two routes towards Ladakh. One by Kiimdan, the other by Dipsang, and both meet at the Shayok River opposite Brangsa Saser. The first is only practicable in winter, and is traversed in three stages, viz. — (1.) Across an undulating ravine cut plateau to Gyapshan on the upper course of the Shayok, 15 miles. (2.) Down the bed of the stream in and out of the water repeatedly, and through a narrow straight where the river bed is very nearly blocked by a vast glacier which has slid down across it, on to a bank of loose pebbles and shingle at the foot of a lofty vertical cliff like a wall. This is Kumdan, nine miles. (3.) Brangsa Saser. Down the river course, and through another very narrow and winding straight between a great glacier and the opposite cliffs and then down a wide river channel to camp. The passage of the straight is done on the ice or through the stream where it is broken. A difficult road under any conditions. The second and usual route is the following in continuation from Daulatbeg Uldi. 18. Kizil Langar, 20 miles. — Over the Daulatbeg plateau, across a shallow stream in a wide deep gully with muddy soft bottom in which cattle stick, and rise up to the Dipsang plain ; wide undulating plateau from which the world around subsides, the highest hill tops only peering above the horizon. Soil soft and spongy, gravel and clay mixed, and, where water logged, boggy. No vegetation. Approximate altitude 17,800 feet. Breathing distressed. Prom this descend a steep and stony gully into a very narrow, tortuous gorge between high cliffs of red clay ; and travelling along in and down its torrent half a mile, enter a wider river bed of rolled pebbles over which the stream flows in a net-work of channels. Rocks roll from the hill tops on either side into the channel. Camp at Kizil Langar, where this channel joins a wider one from the north-west. There is no fuel or forage in all this region. 19. Murghi (height 15,190 feet), 16 miles. — Route down a net-work of shallow streams on a loose pebbly bottom, crossing them continually. At four miles pass Borsa camping ground on a gravelly talus shelving to the stream bed. At a mile beyond quit river, and pass over projecting bluffs, ,and again meet it as a raging torrent rolling over great boulders in a tight, winding gorge, and crossing from side to side by narrow fords camp at Murghi on turfy ground, where a gully from the west joins. Road very narrow and difficult, and risk from stone avalanches. 20. Brangsa Saser, 10 miles. — Up the dry, stony gully to the west. At two miles pass Chungtash, " Great Rock," camp ground at a great erratic boulder on a turfy flat. Then descend rapidly into a deep, dark gully and follow down its winding course till it opens into the Shayok River ; pass up its stream a mile and ford opposite Brangsa. 21. TutyaUk, 15 miles. — Up a rough gully and across a glacier at its water-shed for two or three miles. Then up and down by an extremely difficult path between the side of a vast glacier and the opposite hills, a narrow pass full of angular rocks and snow drifts, and in summer purling with torrents on all sides. At half-way pass Sartang camp ground, an open space menaced by half a dozen glaciers around. Beyond, pass along a widening valley over stretches of turf fringing the stream and sloping up the hill sides, and at a glacier projecting from a valley to the west descend into the bed of the stream flowing from it, and camp on a gravelly flat close under the glacier. Fuel scanty ; pasture in plenty here. An extremely difficult march. 22. Changlung (height 10,760 feet), 11 miles. — Down left bank of river amidst granite rocks for three miles. Then cross river by a wood bridge, and pass along a steep hill slope of loose gravel and sand above the river course and rise quickly up to the Lamsa crest at eight miles on. From this descent to the secondary ridge of Karawal Dawan, ( 130 ) and look down on the Nubra valley, the first green spot and inhabited country since leaving Sanju. Drop down to it by a very steep zigzag path and camp at Cbanglang, a small cluster of eight or ten Tatar huts with fields around. 23. Panamik, 11 miles. — Down the valley over two long strips of gravel talus cut by the deep boulder bed of the Tutyalak River, where it joins the Nubra stream, and is crossed by a timber bridge. Then along patches of turf and brushwood jangal of buckthorn, tamarisk, myricaria, and rose to the cultivation and village of Panamik — to comfort and supplies. 24. Tagar, 13 miles. — Down the left bank of the river as in last stage. Midway cross a rocky ridge abutting on the stream, with the populous village of Chirasa on the opposite bank. 25. Sati, 15 miles. — Down the river course, as in last stage passing villages and cultivation, to its junction with the Shayok River. Then up the right bank of the latter to Sati passing villages and cultivation with patches of brushwood and pasture between on the way. From Sati there are two routes to Leh. First, the direct route by the Khardung Pass. Second, the river route by the Digar Pass. The first is in three stages, viz. — (1.) Cross the Shayok by ford or boat according to the season, and pass up the narrow defile of Rong, crossing its torrent several times, four miles ; then rise up to a high cultivated plateau, and at three miles more camp at Khardung village. (2.) Polii, 15 miles. Up the course of a mountain torrent, cross a tributary from the right, and pass over moraine banks to an upland turfy slope. Continue up its winding and narrowing course to the foot of the Pass. Then pass a pool and glacier, and rise over latter by very steep ascent to the crest of the Khardung Pass, nearly 18,000 feet high, and descend by a very stony, steep zigzag to Polii camp ground on a turfy flat, cut by a rivulet coming down from a glacier at the head of a glen to the right. (3.) Leh, 7 miles. Down a winding gully, and over moraine banks, the road gradually improving to the cultivation of Leh, and then to the town itself. This is a very difiicult route. The other continues down the river from Sati. 26. Digar (height 13,080 feet), 17 miles. — Cross river, and then up its left bank for 12 miles. Then rise out of river bed up to a high flat talus of bare clay and gravel. Cross it and pass round a hill spur, and ascend to fields and houses of Digar in an amphitheatre of granite hills. 27. Poll! Digar, 14 miles. — Up a rising moorland amongst granite boulders and across peat beds and bogs for five miles to Polu camp ground on a spur where the ascent increases. Then up a long stony slope covered with snow patches at end of June, and rise suddenly to crest of Digar Pass, 17,900 feet high. Pass through a narrow gap, and drop by a very steep and rough path to the other side ; follow a winding, turfy glen and camp at Polii huts near a thin rivulet. Some pasture here ; but no fuel. Pass very difficult. 28. Leh (height 11,538 feet) or Ladakh, 10 miles. — Down the glen, across its stream to cultivation and homesteads of Sabu, and then up the valley to Leh. ROUTE II. Yarkand to Ladakh via Ktjgiar. (Authority, Dr. Bellew, June 1874.) 1. Yarkand to Yangichik, 12 miles. — Across a populous and cultivated plain well stocked with trees, mostly willow, poplar, mulberry, alsaguns, and orchard trees. At five miles cross Zilchak stream by rustic bridge, and at six miles on ford the Zarafshan or Yarkand river, and camp another mile on at the Yangichik Settlement. 2. Yakshamba Bazar, 18 miles. — Over cultivated plain with farmsteads, meadows, and marshes. At 13 miles pass through Posgam, and on to camp over freely irrigated tract of cultivation. ( 131 ) 3. Karghalik, 16 miles. — At six miles cross Tiznaf river. Country as on last stage. 4. Beshterek, 20 miles. — At three miles out quit cultivation^ and cross a wide gravelly waste of arid desert, strewed with boulders and coursed from west to east by sandy ridges. Pass through a gap in these to Beshterek or '^Five poplars/^ a cluster of 8 or 10 huts. 5. Yolarik, 12 miles. — Pass out of Beshterek gully on to a wide wind swept desert of coarse sand traversed by gravelly ridges. Camp in settlement of Yolarik, a long stretch of farmsteads on the course of a small river. 6. Kugiar, 13 miles. Cross a wide, shallow, pebbly water-run; pass over a high ridge of loose sand on summit of which is the half buried shrine of Sichcaulue Mazar ; and descend to the Kugiar gully. Follow up its course seven miles past farmsteads to camp in the centre of the settlement. 7. Ak Masjid (height 8,870 feet), 24 miles. — Continue up the gully, and cross its stream to hamlet of Fusar, six miles. Here leave cultivation and habitation behind, and enter hills up a narrow winding gully to camp ground on banks of a stream running down an open glade. Hills of shale. Vegetation scanty. 8. Chiklik, 11 miles. Up a narrow winding gully by a very steep rise between hills of loose dust, six miles, to the top of the Tupa Dawan or " Dust Pass." Then descend by a steep, dusty path down a widening gorge to the bed of the Tiznaf river, and camp on a grassy flat, under an overshading bank of rock on its right bank, near a clump of wiUow and poplar trees. 9. Khoja Mazar (height 9,250 feet), 14 miles. — Up the bed of the river, crossing it girth deep 24 times en route on a rough boulder bottom (June), in a deep winding defile, and camp on a turfy slope on its left bank. Brushwood and forage in plenty. In winter the road is over the frozen river ; in summer through it, and dangerous from sudden floods. Road difficult. 10. Duba, 6 miles. — As last stage. Camp on turfy flat at angle of junction of two torrents. Banks fringed with willow and poplar forest. Pakhpo camps in the vicinity. 11. Gurunj Kaldi, 9 miles. — Up the main stream as before, crossing two tributaries from the right, and camp on turfy slope amidst boggy springs. Hills of schist and granitic trap, and perfectly bare. Marmots here. 12. Chiragh Saldi, 11 miles. — Up stream as before through a gradually widening valley. At eight miles pass ruins of Kirghiz Tam, a former outpost of the Chinese rulers, and beyond it cross a projecting spur into the wider bed of the river, which in June is covered with a deep layer of snow over which the road passes. Camp in a patch of brushwood at junction of a tributary from the right. 13. Kulanaldi, 11 miles. — Up a winding and narrowing gully by easy ascent to the top of Yangi Dawan, 15,800 feet high, three miles. Then down an easy descent for two miles to where a guUy joins from the left. Beyond this down an extremely difficult, narrow, tortuous, and deep gorge which is blocked till June by a glacier that melts away in the next month. The passage over it very difficult down to a wider and less steep channel, of loose shingle between steep banks of moraine rubble, which opens into that of the Yarkand river. Cross the river and camp in tamarisk jangal on opposite shore. River channel half a mile wide. 14. Kukat Aghzi, 15 miles. — Up course of Yarkand river through extensive patches of tama- risk and myricaria crossing the river girth deep five or six times en route on a shingly and sandy bottom, and camp in tamarisk jangal. Channel wide with high hills drain- ing to it on each side. 15. Kashmir Jilga, 25 miles. — Up stream as before. At three miles pass ruins of an outpost fort called Nazar Beg Kurghan, at entrance to a glen on the left which leads in two stages to Shahidula by Kirghiz jangal. Beyond this through an alternately widening»r 34 ( 132 ) and narrowing valley to a long stretch of brushwood also called Kirghiz Jangal ; and through this to camp. 16. Khufelung, 11 miles. — Up stream four miles^ then rise up to a shelving slope of slate and shale on right hand ; follow it seven miles and drop into junction point of a stream from the right ; cross its pebbly wide bed and camp on the left bank of Yarkand river in tamarisk jangal. 17. Aktagh, 20 miles. — Up course of river leaving vegetation, and passing over snow fields filling its channel (June) to Aktagh. From this onwards the route is the same as that by Sanju. ROUTE III. Leh to Shahidula (No. 12 or Koute I) by the Changchenmo Eotjte. Authority Captain Biddulph, September and October 1873. 1. Tikshe, 10 miles. — Along Indus valley, road good. The village of Tikshe contains about 600 inhabitants. 2. Chimray (height 11,890 feet), 15 miles. — Along Indus valley for 10 miles, road indifferent in places. This Indus fordable in September after first six miles, turning up valley to north for five miles of pathways through cultivation to Chimray, village of about 500 inhabitants with monastery. Bad camping ground. 3. Zingral (height 15,780 feet), 8 miles. — Up valley about three miles till it forks. Up valley to eastward for 1 1 miles to village of Sakti ; beyond this the ascent gets steeper to Zingral ; no village ; good camping ground. At Zingral the valley forks to the Chang-la and Kay-la Passes, the latter saves about six miles, but is more difficult for loaded animals. 4. Tsultak (height 15,950 feet), 8 miles. — Up most northerly of the two valleys an easy but stony ascent of two miles to top of the Chang-la Pass, 17,600 feet. A very gradual descent of four miles, then turning abruptly to the east to Isultah, a small lake ; no village ; good camping ground. Though the Pass is not formidable either in height or steepness, it must always prove difficult to loaded animals on account of the badness of the road which is a mere track winding through rocks and boulders. 5. Tankse (height 12,900 feet), 14 miles. — Down valley for 6^ miles easy road, cross shoulder of hill into valley with stream running from south-east pass Durgu a small village, continue up valley to large village of Tankse, supplies of all sorts procur- able, the residence of headman of the district. Behind the village is the valley coming in from the Kay-la. 6. Chakar-talab, 14 miles. — Valley above Tankse narrows to a gorge for six miles, then turns to the south and opens out, two miles further is Muglib, very small village ; for three miles the valley is a grassy swamp, then narrows for two miles of gentle ascent among rocky boulders. At Chakar-talab is a small shallow pond, sometimes dry in summer ; coarse grass on further side of it. 7. Lukong (height 14,130 feet), 7i miles. — Five miles up valley to north-west end of Pangong lake, water salt, 2 miles due north from end of the lake to Lukong small patch of cultivation with stream running into lake. 8. Chagra (height 15,090 feet), 8 miles. — 1| miles above Lukong, valley forks up one to north-east-summer pasture ground of Tartars, one or two stone huts, grass plentiful, and fish in the stream. 9. Rimdi (height 17,500 feet), 13 miles. — A short steep ascent out of valley half a mile due )t east into broad valley running east and west. Continue for 5^ miles very slight ( 133 ) ascent to Lunkur, stone hut, uninhabited, a little water, then steeper ascent, but not difficult to top of Lunkur-la or Marsimik Pass, 18,400 feet. Gradual descent down valley turning* due north, at 3^ miles joins valley from west. Rimdi camping ground at junction ; fuel scarce ; water and grass plentiful. 10. Pamzal (height 14,790 feet), 13 miles. — Down valley to east; stony and narrow track for two miles along face of steep bill, valley then bends to north and road improves slightly. At nine miles, bed of stream narrows to stony gorg^e for a few paces, then opens out into quarter mile breadth. Very stony, brushwood plentiful, strike Changchenmo stream running east and west. Camping ground to west of junction. Fuel abundant ; grass plentiful, half mile further down valley. 11. Gogra (height 15,570 feet), 12^ miles.— Up Changchenmo valley into Kugrang valley, north north-west road good ; fuel plentiful ; grass scarce. 12. Shummal Lungpa (height 17,030 feet), 12 miles. — Cross valley, and up Chunglung valley to north-east stream runs in narrow gorge. At 4| miles narrow steep descent and ascent across gorge coming from north. At six miles hot springs in river bed, valley bends round to north road, winds in narrow track on hill side, several steep ascents and descents. Three miles above hot springs is large ravine leading east, up which is road over Changlung Yokma Pass on to Lingzi Thung Plain ; one mile beyond is Shummal Lungpa ravine, running east ; first half mile narrow and stony, then opens out ; camping ground 1^ mile from entrance; water and fuel plentiful; grass very scarce. 13. Camp near Nischu (height 18,630 feet), 14i miles.— Up valley about 3^ miles to fork, up ravine to eastward at head of which appears a practicable pass. At half a mile take up ravine north by west up steepish ascent across Changlung Burma Pass, 19,300 feet high. Descend low hill into broad shallow valley due east, down valley, which bends to north, and camp near black jagged hill. No grass or fuel ; march throughout good for laden animals. 14. Camp Lingzi Thung (height 17,680 feet), 16^ miles. — Down main valley which makes a great sweep round to north-east, and at 6| miles opens out into Shumshul Plain by Kala Pahar. Due north across plain for six miles cross low ridge with 200 feet rise and 700 feet descent on north side into Lingzi Thung Plain, due north for five miles and camp in water-course; fuel and water to be got by digging; no grass. From low ridge above mentioned, rocky peak at head of Kizil Jilga ravine bears 349°. 15. Jungle Camp, 17 miles. — Across plain for 9 miles, straight for rocky peak, across low ridges for 8 miles, and camp by small pond. No grass or fuel, but the latter can be collected on north side of plain where it is plentiful. 16. Camp Sumna (height 17,150 feet), 21^ miles. — Among low hills for 3 miles into broad valley running north in which is plenty of water ; keep up valley northwards for 2 miles towards smooth round hill, and turn up broad valley running in from west for 11 miles to red rock, and cross the Kizil Diwan (height 17,290 feet) at foot of it into Kizil Jilga ravine. Water, grass, and fuel obtainable 3 miles down, and more plenti^ ful still further on. 17. Kizil Jilga (height 16,360 feet), 9 miles. — Down valley to Karakash river flowing between two huge red rocks, camping ground under southern one. Grass and fuel plentiful. 18. Khushk Maidan, 17 miles. — Down Karakash valley, at 5 miles water disappears in the ground. None to be found for 11 miles, where are numerous springs. Camp on south side of valley. Fuel abundant ; grass scai'ce. Road excellent all the way. 19. Chung Tash (height 15,740 feet), 7 miles. — Down valley, which narrows. Huge rock on right bank. No fuel or grass. Road good. 20. Camp Sumnal (height 15,540 feet), 13 miles. — Down valley, which at 3i miles bends round to north, and valley leading to Aktagh comes in west. The Karakash then flows ( 134 ) in narrow gorge and at 6 miles from Chung Tash are hot springs on right bank. A little fuel, but no grass. One mile above hot springs valley opens for a mile then closes again. Road in parts stony and bad. River has to be crossed frequently ; small patches of fuel in side nullahs. Good camping ground at bend of river to east, where large nullah from west joins. Fuel and grass abundant. 21. Camp Tak Marpo (height 15,000 feet), 11 miles. — Valley opens out for 3 miles. Zinchin on right bank. Fuel and grass. Valley then narrows ; road encumbered by huge boulders and masses of rubbish ; very difficult for laden animals ; camp under yellow rock on left bank. Fuel and grass at intervals all the way. 22. Camp Polong Karpo (height 14,600 feet), 21 miles. — Valley opens out and travelling good. At 8 miles is broad valley on left with abundant fuel, after which fuel is to be found all along in main valley; grass very scarce. At 16 miles valley narrows and turns to north, fuel becomes more plentiful. At Polong Karpo is a huge rock in bed of valley on left bank ; good camping ground ; grass abundant. 23. Camp Sora (height 14,000 feet). — At 5 miles river takes sharp bend to north-west into broad valley at foot of Kuen Luen. For 2 miles on either side of the bend is no fuel or grass. Ground at Sora covered with natural salt pans. Good camping ground. Fuel and grass abundant. 24. Jungle Camp, 17 miles. — Camp at mouth of small ravine. Opposite mouth of Karajilga ravine. Fuel and grass abundant. 25. Gulbashem, (height 12,390 feet), 421 miles. — Road down valley good; and grass and fuel abundant everywhere, except for 2 miles above Gulbashem. 12 miles above Gulbashem the river is much increased by springs. Gulbashem is a favourite Kirghiz camping ground. 26. Balakchi, 10 miles. — Grass and fuel. 27. Shahidula (height 11,780 feet), 13 miles. — Small deserted fort on left bank of the Kara- kash. Grass and fuel abundant. At 6 miles strike road from Sugat Pass ; road good. Though parts of the road are practicable for guns and wheeled carriage, it is on the whole only available for camels or horses. ROUTE Ilia. Variation on No. III. (Captain Trotter, Sejotemher and October 1873^. From Gogra (Station 11 of Route III) to Shahidula (No. 27 of Route III). 1. Kotajilga (height 16,730 feet), 8 miles. — Road up stream the whole way good, but somewhat difficult for loaded ponies, as there are several steep ascents and descents in crossing tributary streams, which in the autumn contain only a few inches of water. Pass ravine on right leading to Nischu as per Captain Biddulph's route. At the camp, grass, water, and wood procurable. 2. Pangtong (height 17,250 feet), 7^ miles. — Steady and gentle ascent through a broad stony ravine for four miles, then somewhat steeper. Camping ground covered with snow, but grass and an inferior fuel said to be plentiful. 3. Sumzungling (height 17,310 feet), 15 miles. — Steady and not very steep ascent to the Changlung Pass (18,910 feet). The road then passes over a high table-land for about a mile, after which it enters a ravine along which it passes for 94 miles of execrable road, crossing the stream in numerous places before reaching the camping ground at the junc- tion of three nuUas. Water and a little grass on a neighbouring hill, but no fuel, one of the worst marches on the whole road, as the number of recently dead animals that strewed the road too surely testified. ( 135 ) . 4. Dehra Kompas (height 17^890 feet), 19 miles. — Road runs nearly due north up a gentle ascent for about 5 miles, road fair, then for several miles of good road across the west edge of the Ling-zi-thang plains; crosses several easy open ravines draining eastwards descends into and crosses a branch of the Karakash river and camp at foot of a low pass ; very little water to be obtained by digging ; and no grass or wood. 5. Shinglung or Dunglung (height 17,030 feet), 18 miles. — Across pass, and down a ravine for 5i miles into Karakash river, where plenty of boortsee and water, but no grass. Road follows river, which after 3 miles turns up sharp in a northerly direction, road good, but stony ; boortsee abundant. 6. Kiziljilga (height 16,360 feet), 14^^ miles. — Bad stony road down bed of Karakash River for the first mile, then between about two or three miles of ice bed have to be traversed, the bed extending right across the ravine, here about i mile in breadth ; road very slippery and difficult for laden animals. Near camp passage of Karakash difficult (in October) owing to admixture of ice and water. Fuel (boortsee), grass and water, within reach of camp down Karakash River. 7. Chungtash or Chung Tash (height 15,740 feet), 23^ miles. — Road down Karakash River generally good but stony and bad in the latter portion. Camp badly situated, as there is neither wood nor grass both of which might have been had at Khushk Maidan, a few miles further back. Camp under a big rock near where the bed of the Karakash is very much narrowed by precipitous hills coming down near the river bank. 8. Shorjilga, 14 miles. — Road for two miles down Karakash, which takes another sudden curve to the north-east, the road goes up a tributary stream containing nearly as much water as the Karakash itself. Road bad for two or three miles owing to the number of times the frozen stream has to be crossed and recrossed, it then passes over a toler- ably level plain up to a gorge at the mouth of which is Shoorjilga. In October there was no water there and camp had to be pitched half mile up the gorge at a place where the river water disappears into the ground; not a stick of wood or blade of grass. 9. Kdr^tagh (height 16,890 feet), 9 miles. — Up ravine for several miles, snow and ice nearly the whole way and road bad; short but sharp descent from Karatagh Pass (17,710 feet) into large flat open plain, covered with several inches of snow. Lake frozen over, but water obtained by making hole in ice ; plenty of boortsee, but no grass visible. 10. Aktagh. (height 15,590 feet), 22^ miles. — Road the whole way good, over a level plain, which was entirely covered by snow. About half way at east foot of low double-topped hill is a place called Tamba^ sometimes used as a camping ground. Ak-tagh is at the junction of the Karakoram and Changchenmo Routes. 11. Chibra 10 miles. — Road crosses stream and goes up ravine, steady ascent with fair but stony road. No grass or wood at camp. 12. Suget (height 12,970 feet), 18 miles.— Up ravine to top of Suget Pass (17,600 feet), 4 miles. Steady ascent and road good. Descent steep through the snow down zigzag, then straight down ravine for eight miles due north. Road stony, but descent gentle. Road then turns eastward and soon leaves the nullah, which has a very rapid fall; an alternative road goes right down the nullah in which there is plenty of wood and grass, abundance of both at camp. 13. Shahidula (height 1,780 feet), 8 miles. — The road descends to Karakash River (two miles) and follows the river to camp, crossing it twice en route. Passage somewhat difficult. Plenty of wood and grass a few miles up the Karakash River. Shahidula is the same as Station No. 12 of Route I. * On this march we lost our road and had perforce to halt at Tamba, where we found boortsee ; melted snow served for water, which is not procurable here in summer. 35 ( 136 ) ROUTE IV. TARKAND TO KASHGHAR (CAPTAIN TROTTER), NOVEMBER 1873. 1. Kok Robat or " Blue Hostelry'^ (height 3,830 feet), 22 miles. — Pass through cultivated tracts and at 4^ miles cross the Opo or Arpi canal (from the Yarkand river) by a good bridge. A little distance beyond is the village of Kdrakoram, after which the road passes through grassy swamps, followed by a desert waste. Occasional small villages are passed before reaching Kok Robat, a scattered village of about 200 houses. 2. Kizil (height 3,910 feet) , 26 miles. — Pass over sandy desert, without habitations or cultivation, to Ak Rob^t or " White Hostelry,^' where are two small wells whose surface water is 98 feet below the ground level; another 13^ miles over flat waste to Kizil, a large village of about 500 houses. The dry bed of the Kizil stream is passed, coming from the Kizil Tagh on the west. The country is irrigated by small canals taken higher up from the Kizil River which in summer contains a fair supply of water. 3. Yangi Hissar or "New Fort'' (height 4,320 feet), 28| miles. Over a flat plain irrigated by small water-courses from the Kinkol River, the scattered villages of Chemalung, Kudok, Kosh-gombaz, Tuplok and Kalpin are passed. Low sand hills on right before reaching the Yangi Hissar River, which has its chief source in springs a few miles south-west of the town of the same name, which lies two half miles beyond it. Yangi Hissar is a large town with strong fort, and contains about 4,000 houses. The Yangi Hissar River^ is crossed by a two-spanned timber bridge, about 60 feet long. A low ridge of hills separates it from the town. The fort is about 600 yards to the north of the town. 4. Yapchan (height 4,210 feet), 23^ miles. — Pass at four miles the Zaikash stream fed from springs on the west, further on cross two branches of the Sailik canal from Kusank. After passing the villages of Khanka Sorgoluk, and Tuglok cross by a wooden bridge a large canal from the Kusan River, and then cross the main stream of the Kusan itself by a wooden bridge. The road follows the lower bridge of the stream for about a mile, and is much cut up by branch canals. Soil generally sandy. 5. Yangi-shahr or new city of Kashghar (14^ miles). — Road passes through cultivated coun- try and crosses several streams and canals, the largest of which are the Tazgun or Khanarik or Yamunyar and the Karasu (chiefly fed from springs) . Pass the villages of Tazgun, Turmalak, and Kasr Robat. The old city of Kashghar lies about five miles from the Yangi-shahr in a north-west direction. ROUTE V. Kashghar to Maralbashi (authority, Captain Biddulph) January 1874. 1. Sang, 19 miles. Through cultivation; at six miles village of Barin; 7^ miles Arowah, jimction of road from city of Kashghar ; nine miles village of Yandumba, from where goes cart road to Kizil Boia to east ; cross rivers Yamunyar and Chokanak flowing into Kizil, both bridges swift streams. Sang large village. 2. Faizabad (height 4,000 feet), 16 miles. — Large village, through cultivation; at two miles cross small river ; no bridge. 3. Yangi-awat, 11 miles. — Small village, cultivation almost ceases from Faizabad; at seven miles small village of Shagiat. 4. Kashmir, 28 miles. — Through bush jangal and plain ; at 20 miles cross river Kizil ; bridge used in summer ; ice bears carts, &c. in winter. Kashmir a small post-house, no village. 5. Togha Sulukh, 13 miles. — Through tree jangal and white grass ; a small post-house. * Which I have known called by no lees than nine different names. — H. T, ( 137 ) 6, Shugeh, 18 miles. — A small post-house; road all the way through tree jangal. 7. Maralbashi (height W^C ), 14 miles. — Small village and fort; road through tree jangal and high grass. Maralbashi is at the junction of Kashghar road, with road from Yarkand to Aksu. Carts travel freely all the way. ROUTE VI. Kashghar to Chadyrkul, Captain Trotter. Kashghar (Yangi-shahr) to — 1. Besak, Upper Artysh (height 5,290 feet), 26 miles. — Road leaves on left at 5^ miles the old city of Kashghar, and then crosses the Tuman (Kashghar) river by a good bridge ; passes through numerous gardens into an open stony plain, and then through a gap in a low range of hills, beyond which lies the district of Osten Artysh, consisting of numer- ous scattered townships. 2. Chung Terek, or "Big poplar tree, (height 7,000 feet), 20 miles. — Road passes over level plain and then up the gently sloping Toyanda valley. Road good but stony ; pass en route the old Chinese outpost of Teshektash and the small village of Tupa ; a small Kirghiz village at camp. 3. Chakmak Forts (height 8,830 feet), 20 miles. — The road continues up the Toyanda valley and passes through precipitous hills rising some 3,000 feet above the bed of the stream, which has to be frequently crossed, both on this and the last march. At 12 miles pass the *^Past Kurghan" or Lower Fort, garrisoned by a detachment from the larger forts at Chakmak further on. A few Kirghiz tents en route, but no other habitations. From Chakmak a road goes across the hills to the east to the Terekty Forts. 4. Balghun Bashi (height 10,540 feet), 10 miles. — The road continues up stream, and at about eight miles passes the Suok outpost at the junction of two streams from the Suok and Turgat passes. The former is two days' journey in a north-west direction by a very difficult road ; two miles above the outpost is the camp, where plenty of fire- wood, though but little or no grass. 5. Turgat Bela (height 11,090 feet), 15 miles. — Road always up stream at first through precipitous hills, which open out somewhat as the camp is approached. Occasional Kirghiz tents ; plenty of grass near camp ; but fuel very scarce. 6. To Chadyrkul (Lake), 20 miles. — 13 miles to the crest of the Turgat Pass, a gentle ascent right up to the foot of the pass, from which to the crest (12,760 feet) there is an ascent of about 400 feet in a distance of a mile. From the pass the road passes along a spur for about three miles, and then rapidly descends to the lake, which lies about 1,500 feet lower than the pass. Plenty of grass, but little or no fuel obtainable. The road from Kashghar to Chadyrkul is good throughout, and could with very little labour be made available for light carts and field guns. In January a little snow lay by the roadside to the north of the pass, but none whatever on the south side. The route just described is on the main caravan route between Kashghar and Almati (Fort Vernoye), for details of which maps can be consulted, ROUTE VII. Kashghar to Belowti Pass, (Captain Trotter) February 1874. (On road to Ush Turf an.) * Kashghar (Yangi-Shahr) to — 1, Altyn Artysh (height 4,100 feet), 22 miles. — Over plain for a great part of the way; road good ; pass Kashghar River by wooden bridge (in cold seq^son) , and subsequently ( 138 ) cross two smaller streams and canals, and traverse a low ridge before entering the Artysh valley, which contains numerous villages aggregating perhaps 2,000 houses. 2. Kalti Ailak (height 3,950 feet), 22 miles. — Good road over level plain. 3. Kyr Bulak (height 5,340 feet), 33 miles. — Good road passes over plain and then through a range of low hills up the Sogon stream, on which is a military outpost, then crosses a low pass and emerges into a large plain. Camping ground is occasionally used by Kirghiz. 4. Jai Tupa (height 4,910 feet), 20 miles. — An occasional camping gi'ound of Kirghiz ; water scarce ; situated near the centre of an extensive forest of stunted poplar trees ; good road passing over level plain ; plenty of grass and fuel. 5. Ui Bulak (height 6,650 feet), 27 miles. — Road good but stony; crosses a low spur into the Ui BuMk valley, where plenty of fire-wood and a moderate amount of grass ; water scarce, but plenty of snow. 6. Tigarek, 17 miles. — Road up stream and then over a spur from the main range, stony, but otherwise not difficult. Tigarek is in a large grassy plain (no water) surrounded by hills ; is sometimes used by Kirghiz as a winter pasture ground, when snow is used as a substitute for water. 7. Belowti Pass (height 11,360 feet), 9 miles. — Road crosses some low spurs and then passes up a ravine ; a steady ascent of four miles followed by a sharp pull of two miles up to the pass ; no high peaks near the crest, but undulating grassy ridges. The road descends on the north side of the pass in the valley of the Koksh^l river. The Belowti Pass was the furthest point reached by Captain Trotter, and from it the marches to Ush Turfan were said to be as follows : — 8. Ak-chi, 10 miles, on Kokshal River. 9. Kuyok Tokai, 22 miles, down river. 10. SafrBai. 11. Karawal, 22 miles. 12. Ush Turfan, 16 miles. Total distance, Kashghar to Uch Turfan, 242 miles ; from Safr Bai (No. 10) a road is said to go to Karakul (near the old Issigh-Kul Fort) by four difficult marches, crossing the Bedul and the Zanku passes, and the head-waters of the Naryn River. From Karakul roads go to Kuldja and to Vernoye (Almati). ROUTE VIII. YANGI HISSAB TO KILA PANJAH, WAKHAN, BY THE LITTLE PAMIR (CAPTAIN BIDDULPH) march and APRIL 1874. 1. Ighizyar (height 5,600 feet), 19 miles. — Large village. Road nearly due south, through sand-hills and cultivation, crossing two small streams, over soda plain for 5 miles, cross stream by village, and over stony plain for 7 miles ; practicable for wheeled traffic. 2. Aktala (height 7,345 feet), 17 miles. — Kirghiz camp, south-west four miles over plain to Aktala valley, then up valley 13 miles. At six miles pass through ruined Chinese fort closing the valley. Fuel, grass, and water abundant ; road good. 3. Sasak Taka (height 9,455 feet), 13 miles. Out of main valley, into side valley to south, first few miles narrow and stony, then opens out and travelling improves. Wood and grass abundant. Kirghiz camp at Sasak Taka. 4. Kaskasu (height 10,980 feet), 14 miles. — Kirghiz camp up valley; travelling good; grass plentiful, but fuel scarce. Several Kirghiz camps in the valley. 5. Chehil Gumbaz (height 10,310 feet), 8^ miles. — Kirghiz camp ; half mile from Kaskasu road turns up small valley to south, and at another mile is foot of Kaskasu Pass, first 200 ( 139 ) yards steep, then for three miles winding tbrong-h undulating grassy hills round head of valley to top of pass 13,000 feet, road good half mile along narrow ridge then steep zigzag descent of 1,000 feet into Charling valley to Chehil Gumbaz at junction of two valleys. Grass and water plentiful ; fuel scarce. 6. Past Robat (height 9,370 feet), 9 miles. — Road up valley to west for 2^ miles to foot of Torat Pass; 1^ mile of ascent; not difficult for laden animals to top of pass, 13,400 feet; long steep descent into narrow valley ; road stony and bad; into broad valley, to south to Past Robat ; Kirghiz camp at junction of two valleys ; grass, fuel, and water abundant. 7. Tarbashi (height 11,515 feet), 8 miles. — Kirghiz four miles up valley to fork. The valley to the right leads to Tashkurghan by the Yambulak Pass, only used in summer. Up valley to left for two miles along narrow gorge, among rocks and boulders, the stream having to be crossed and recrossed more than a dozen times. Many hot springs in the defile, steep ascent of 400 feet into upper valley, when road again becomes good. This route can only be used when the stream is low in winter and early spring. 8. Shindi, 17 miles. — Up the valley for eight miles of gentle ascent to the Chichiklik plain, about 1^ mile in diameter ; to south-west is the Kok Moinak Pass, used in spring and summer, by which Sirikol is reached in 1^ marches ; to north is the route by Yambulak Pass before mentioned. Bending to south a long descent, steep and stony, brings one to Shindi, where are a few Sarikoli huts. 9. Tashkurghan (height 10,270 feet), 19 miles. — Down valley for four miles to Yarkand River. From here a road up valley to east leads over the Shindi Dawan to Tarbashi. Continues up Yarkand River to foot of Tashkurghan. Road good all the way. Grass and fuel plentiful. 10. Kanshabiir (height 12,980 feet), 17 miles. — Due west from fort into Shingan valley. The first three miles of narrow defile strewn with boulders, very difficult travelling, after which valley opens out and road gradually improves. Fuel and grass plentiful. 11. Aktash (height 12,600 feet), 18 miles. — Up valley to right, and after a mile up fork to left to foot of Nezatash Pass 15,000 feet. Three miles of gentle ascent, last 300 yards to top steep. Descend into valley running north-west along this for eight miles, over low spur into broad Aktash valley running south. Grass and fuel plentiful. 12. Ghaz Kul, or lake of Little Pamir (height 13,200 feet), 46 miles. — Down Aktash valley to south into Little Pamir due west, travelling excellent the whole way. Grass and fuel plentiful everywhere. Camp by lake. At six miles from the lake is Kabr-i-Bozai, deserted Kirghiz huts, opposite which is road leading to Kunjud by Tagdung-bash Pamir, over Kujroi Pass. 13. Langar (height 12,530 feet), 25 miles. — Continue due west; at 10 miles Pamir narrows into rocky valley, and travelling becomes difficult ; road winding along face of hill with many small ascents and descents. At Langar deserted village ; fuel and grass plentiful. Opposite Langar is road leading to Kunjoot by Bykurra Pass, closed three months in the year. This is the road generally used between Wakhan and Kunjua. 14. Sarhadd (height 11,150 feet), 24 miles. — Down main valley, travelling bad; road much encumbered with boulders, and there are two steep ascents and three steep descents, very trying to animals. The stream has to be crossed many times, and the road in consequence of melting snow becomes impassable after 1st May. From Sarhadd road leads to south to Yassin and Chitral by Baroghil, Darkot, and Ishkaman Passes. Sarhadd is the frontier village of Wakhan. 15. Kila Panja, 55 miles. — Down valley to foot, road good for laden animals villages at intervals the whole way ; grass and fuel plentiful. By the village of Yur, 18 miles from Sarhadd, is a summer road into Mastuj. At Vost, 38 miles from Sarhadd, is a road leading into Chitral said to be only practicable to men on foot. 36 ( 140 ) ROUTE IX. KiLA Panjah (Eoute VIII) TO Aktash by the Great Pamir (Captain Trotter), April 1874. Kila Panjah (Wakhan) to — 1. Langarkish village (height 9,350 feet), 6 miles. — Road lies along the banks of the main Panjah River, and then up the northern branch ; pass on right bank the villages of Zang and Hissar. All supplies for the journey across the Great Pamir have to be taken from Langarkish, which is the highest village on the north branch of the Panjah river. The valley is bounded by lofty and precipitous mountains. 2. Yumkhana or Jangalik (height 11,440 feet), 18 miles. — The road follows the right bank of the river, above which it rises in many places to a height of 1,000 feet ; cross the Ab-i-zer-i Zamin (River), four miles beyond which is the camp. Plenty of grass and fire-wood. 3. Yol Mazar (height 12,320 feet), 13 miles. — Road still along right bank of stream, at four miles cross the Ab-i-Matz (river) up which passes a summer road to Shighnan, two miles further is Boharak, the commencement of the Great Pamir. Plenty of grass and boortsee. Road good. 4. Bilaor Bas (height 13,120 feet). — March along right bank of river through a grassy plain bounded on both sides by undulating hills. 5. Mazar Tiipa (height 13,760 feet), 20 miles. — Road up gentle ascent the whole way, and on the right bank of the stream. 6. Sarikol (head of the lake), (height 13,950 feet), 16^ miles. — Road up gentle grassy slope to Victoria Lake, passing along its northern edge. The lake is ten miles long and nowhere more than two miles in breadth. Camp at the east end of the lake ; whole ground under snow, but very fine pasturage in summer. Prom Sarikol a road leads across to Langar, at the west end of the Little Pamir, crossing the Warram Pass. 7. Shash Tupa (height 13,760 feet), 19^ miles. — Cross a low water-shed a few miles from camp and then enter a valley, the stream down which flows into the Ak-su river ; very gentle descent through broad open valley to camp. Roads traverse the Pamir here in all directions. 8. Dahn-i-Isligh (mouth of the Isligh River), height 13,220 feet, 18 miles. — Gentle descent through open valley, pass several broad open ravines. This camp may be considered the termination of the Great Pamir. Plenty of grass and fuel. 9. Ak-tash (height 12,600 feet), 37 miles. — Road follows the Isligh River until it joins the Ak-su, both of these streams partially frozen, but ice breaking up making journey somewhat difficult. The Isligh River passes through precipitous mountains ; after entering the Ak-su valley turn south to Ak-tash, which is the same as station 1 1 of Route VIU. ROUTE X. Tashkurghan to Yarkanp, via Charling Eiver (Captain Trotter, from KisHEN Sing). Tashkurghan (No. 9 of Route VIII) to— 3. Chehil Gumbaz (No. 5 of Route VIII), 56 miles. 4, Tashkerim, 19 miles. — Road passes down Charling River ; Kirghiz encampments ; wood and grass ; pass on road Alumbitte Kurghan. From Tashkerim a foot-path leads across the hills to Kinkol camp. ( 141 ) 5. Khaizak-vil, 14^ miles. — The road continues down Charling River. Pass on left bank at eig-ht miles the small village of Bagh (30 houses), also the villages of Kiok-tash, Mirgul (25 houses), and Yoya (15 houses). 6. Arpalik, 21 miles. — Road leaves Charling River and passes up a tributary stream to small village of Yamunarik. Thence goes over two low passes, the Kara Diwan (at 11^ miles) and the Kizil Diwan. 7. Kizil-tagh, 15^ miles. — Road good for three miles to Tangitar, where the river (Kizil) passes for five miles along a very narrow ravine, very difficult to traverse in the after- noon owing to floods caused by the melting of the snow on the hills above. 8. Yak-arik, 23^ miles. — Good road over the " Shaitan kum'' or " DeviFs sand." 9. Yarkand, 20 miles. Road passes over a well cultivated and thickly inhabited plain. Total distance, Tashkurghan to Yarkand, 188 miles. ROUTE XI. Route from KilaTanj ah (WAKE AN) to Kila Wdmar (ROSIIAN) along the river Panjah. Authority Captain Trotter {from Abdul Suhhan.) Names of places. Kila Panjah to Khan- dut. Pigish ... Shikharbi Paresban... Patur Yaghduru (Dovam) or 2nd. Barsbar... Nawabad Country or district. Wakban Ditto Ditto Ditto Sbikashim Gliaran sban.) Ditto (Badak- p* « ffl to s o 16 6 20 17 17 14 Kemarks. A village witb about 30 bouses and mud fort. Road stony, passes through village of Parg at 2\ miles, Pakui at 5i miles, and ruins of Ishmurgh at 9 miles. From Pakui to Khandut is a dense forest of stunted poplar trees. A village of 30 houses. Road good, and along left bank of river ; much cultivation. Supplies plentiful. A scattered village of 30 bouses. At 13 miles is the large village of Argund. Road bad and stony and along left bank of river. Supplies plentiful. A frontier village between Wakhan and Ishkashim, subject to Mir of Wakban, consisting of 25 houses, villages passed on road are Verg at 7 miles, Sad Ishtrag at 10 miles, and Kazideh at Hi miles. From Sad Ishtrag a road strikes off southward to Chitral, the capital of Kashkaro. Four miles beyond Kazideh the river turns abruptly northwards, and is dangerous here for horses, as it passes over a narrow ledge of rock overhanging the river. Road stony. Supplies plentiful. A village of 15 bouses. Valley in this march wider and river broader. At 6 miles is the large village of Ishka- shim of 60 bouses. A road joins here from Faizabad. At 8 miles village of Yaghduru (Aval) or first. Road good, and through rich cultivation. Supplies plen- tiful. A small village belonging to the Gharan district situated on the right bank of river Panjah. River forded 3 miles above village of Sar-i-Shakh, left bank avoided being dangerous for horses. Road stony and bad. Supplies plentiful. A deserted village on the left bank, river recrossed at ford near deserted village of Kazideh at a mile and half from Barsbar. Another road from Badakshan to Gharan {via Aghirda Pass) and to Shakh Darrah valley, crosses at the same ford. At 5 miles is the deserted village of Zich, on the opposite bank of river is a ruby mine. No supplies here. Cultivation scanty. Grass and wood plentiful. ( 142 ) ROUTE XI. Route from Kila Panjah (WAKHAN) to Kila Wamar (ROSE AN) along the river Panj ah. Authority Captain Trotter (from Abdul SubhanJ — concluded. Names of places. Country or district. ii Remabks. 8 9 10 11 12 Darmarakht Viar ... Kila Bar Panjah ... Sacharb Kila Wamar Shighnan Ditto Ditto Ditto Eoshan Total distance ... Hi 14^ 5i 9 22 Camp on the left bank. Darmarakht is on the opposite bank from whence supplies are procured by means of a wooden bridge thrown across the river, which is about 150 yards wide here. From Nawabad at 3 miles is the large village of Shekh Beg in ruins, and further on the road runs through a tunnel called Kuguz Parin, or " hole in the rock." Road bad and stony. The Kuguz Parin in the boundary between Gharan and Shighnan. A scattered village of about 40 houses. At two miles from Darmarakht a large tributary of the Panjah river called Arakht is crossed by a strong wooden bridge. At 9i miles the road ascends and traverses the Mithinz and Tarseb passes. Road bad and stony. Supplies plentiful. A large to-wn on the left bank, the capital of Shighnan. Adjacent to it stands a stone fort on the margin of the river. At Dasht-i-Khust, the river Suchan falls into the Panjah. At SJ miles is the small village of Deh-i-Mur- ghan. Road good. Supplies plentiful. A village of 20 houses on the right bank. Sacharb is reached by crossing the river either by ferry at Kila Bar Panjah, or by ford at the village of Disbar at 3 miles lower down. Yumj village is at 4i miles from Sacharb. Road good. Supplies plentiful. A large town, the capital of Roshan, situated about IJ miles above the junction of the Murghabi river with the Panjah. Fort and town on the high bank of the Panjah. At 13 miles is the small village of Past Khuf. At 16 miles stands the Darband tower, built on a rock. This is the boundary between Shignan and Roshan. 161i ROUTE XII. Yakkand to Akstj. Authority, Captain Chapman. Bhan Sing. FROM YARKAND TO AKStJ AND ONWARDS. 1. Yarkand, TereJc Langar, 7 tash,* Ek Shamba and Char Shamba Bazaars en route ; cultivation and gardens to within one mile of the halting ground. * Note. — A tash is the ordinary uuit of measurement of distances in Eastern Turkistan, and on many of the principal roads tasJi-boards have been errected ^similar to the wooden sign posts, still existing in some parts of England. They were put up between Khotan and Kashghar shortly after the accession of the present Ruler, but the Yarkand road the following measurements were made by Kishen Sing Pandit : — Number cf Average number paces. ' of paces per tash. From 1st to 5th Tash Post ... ... ... 36,350 9,112 „ 5th to 8th „ ... ... ... 27,880 92,93 „ 8th to nth „ ... ... ... 26,800 8,933 Mean value of each tash ... 9,113 or almost exactly 4| English statute miles. Theoretically the tash is equal to 12,000 paces of a riding camel, and it is by means of this measure that the distances are said to have been laid out between Yarkand and Kashghar.— [H. T.] ( 143 ) Terek Langar, a villag-e of about 40 houses, with two musjids, in every house a room for the accommodation of travellers. 2. To Lai Lik, 7 fas/t, through desert and low jungle; at 4 tash a deserted Langar of the time of the Chinese ; at Lailik, 22 houses, the inhabitants support them- selves by the entertainment of travellers ; supplies, &c. •3. To Menui, 4 task, through a jungle of high trees. The road within one tash of Lailik (on the Yarkand side) approaches the Yarkand river ; it is touched three times by the road during this stage. At Menut 16 houses in all, accommodation for travellers ; supplies, water, &c, 4. To Alaigur, 5 tash, through a jungle of high trees, the river is encountered twice en route. At Alaigur 23 houses, accommodation in each ; supplies and water ; the river is nowhere crossed, but the road follows its bank, 5. To AJcsdk Mardl, 3 tash, through jungle as during previous stage, the road approach- ing the river once en route ; 14 houses at the halting stage, accommodation in each, and supplies, &c. 6. To Shamdl, 5 tash. Here is only an old rest-house, but about a mile to the east on the bank of the river is a cluster of some six houses, whence the traveller may get some supplies ; the entire route through jungle. 1 rom this point the river runs wide of the road. 7. To Mardlbdshi, 3 tash. High jungle encountered en route, but in patches, there being here and there strips of sand and bog, the only water being obtained from springs. A fort, and considerable place. Vide Captain Biddulph^s report. 8. To Charwdgh, 3 tash. The spurs of a range of hills stretching out from the Tianshan are to the north of the road which runs through jungle with cultivation here and there, the drainage from this point is into the Kashghar River from which canals are cut to Maralbashi and onwards ; there are about 40 houses in Charwagh, each having accommodation for travellers. 9. To Tumshuk, 4 tash. Half way a hill is to the north of the road under which the road immediately runs; on the top of this hill is a Mazar and also one at the base. Captain Biddulph gives this hill the name of Pir Shereh Kuddam Mur- taza AH Tagh, and thinks the rock must be basaltic. Beyond this hill is a fort, and at the distance of about a tash is a ruined city at the base, and on the slope of a second hill (this is one of the buried cities) the houses are of earth and not of stone. One tash further to Tumshuk, through a low jungle, a place of 32 houses, accommodation for travellers. A canal from the Kashghar river is crossed at 2 miles from Tumshuk by a bridge ; this is narrow and com- paratively deep, being only some 10 yards across, it runs away east. 10. To Chadyrkul, 2>\ tash, through a jungle of high trees; 15 houses. 11. To Yaka Kuduk, 4 tash, through a jungle of high trees; water from a well about 30 feet deep ; there is a slight drainage from this point ; southwards towards the Yarkand river about 12 houses in Yarkudut, where travellers are accommo- dated. At this point the road divides into two, the shorter and more direct road going by — 12. Yazdah, 5 tash. High trees during half the march when these change to low jungle. No water en route, spring near Yazdah itself. About eight houses in Yazdah. 18. To Childn,"^ tash. Low jungle and sandy desert ; no vi^X^x en route ; 22 houses in Chilan ; two large trees at this place which are conspicuous ; two tanks at Chilan supplied by springs. The longer one by — 12 8nget, 4 tash. A deserted Langar, but no water ; no one halts here ; if a halt is intended, water must be carried. 37 ( 144 ) 13 Childn, 6 task. A low jungle, but no water on the surface ; but it may readily be obtained by digging. This is the easterly of the two roads. This road is closed after the winter season when the thaw sets in and occasionally when the springs swell and bring an extra amount of water. 14. To Choi Kuduk, 4 task. Through desert without water. At Choi Kuduk water obtained from a well, but the water is brackish. There is a large serai here with a musjid. Here is a low range of hills on the north-west, close to which the road runs, and behind which is the bed of the Kashghar Daria. 15. To Sai Arik Langar, 4 task. Through a desert, with sand and small stones. At Sarek Lang^r there are two serais, and a post for the examination of passports ; gome 30 or 40 houses with cultivation, &c. ; water by a canal from the Kashghar Daria. 16. To Kumush or Kumbdsk, 3 task. The Kashghar Daria is crossed at one mile from Sai Arik Langar. After the crossing there is a group of hamlets known as Aykol, beyond this eastwards at about one and half tash is a considerable sheet of water ; in the district, which takes its name of Aykol from this, are some 2,000 houses. Cultivation may be said to extend from Sai Arik Langar as far as Aksu ; there is a bazaar held at Kum. In Aykol are two serais and a considerable bazaar (Thursday) ; the country is cut up by canals from the Kashghar river. The Kum district stretches towards Ush Turfan and eastwards ; it is said to contain 8,000 houses. Kum Bazaar, which is the head-quarters of this district, is off the road some 2^ miles. At the stage of Kumbash there is merely a Langar. 17. To Aksu, 4 tash. After leaving Kumbash, about two tash, the Ush Turfan river is crossed ; it runs in three principal channels, one of which is crossed by a ferry during the cold season ; the Kashghar Daria was crossed in four separate channels at some distance from each other, and all bridged. After leaving the Ush Turfan river the road rises to a plateau along the skirts of which it passes. It drops suddenly upon Aksii. A small stream from the north passes to the west of Aksu at about one mile distance and falls into the Aksu Daria about three miles south of the town. Total distance 73^ tash or 75^ tash from Yarkand to Aksii. At Terek Langar, the first march from Yarkand, the natives are Dulans, a tribe presum- ably of Kalmak origin, having a distinct dialect of Turki and many customs different from those in force elsewhere; they extend as far as G^i'^w, the 13th march ; they remain distinct from the natives of the Aksii and Yarkand districts. Kokshal is the name usually given, not only to the Ush Turfan river, but to all the streams in the Aksii district, on account of the rice grown in the fields which they fertilize. Kokshal (rice producing). Kok really means blue, all green things springing from the ground are called Kok. From Aksu to Ush Turfan — 1. To Sayik, 4 tash, through cultivation at the base of the hills and in the valley. 2. Achtdgh, 6 tash. 3. Ush Turfan, 4 tash. Two serais in Ush Turfan, the last two places are in the Ush Turfan district, which is a separate command; it contains 8,000 houses, and is a highly productive district ; flocks and cattle abound. The water of Aksu is from springs, there is only one tank in the Yangi-shahr. There is a very large sale of horses in Aksu ; the Dadkhwah taxes the sale of horses, taking 12 puis on each transaction. On market day 600 loads of Indian-corn and wheat, 1 tanga per charak ; 300 loads of rice, 2 tangas the charak. There are 200 dyers in Aksu. 1. Khotan Serai, 50 rooms. 2. Kashguree Serai, 60 rooms. 3. Sheik Beg Serai. 4. Mullah Saduk Serai. ( 145 ) SERAIS IN AKSU. 9, 5. 6. Dhung Serai. Khona Serai. 7. Nar Kurgan Bai Serai. 8. Yarkandee Serai. Andijani Serai. the walls. Outside the walls. 13. Eesah Kor. 14. Arjak. 15. Abdullah Bai. 16. Shah Mahomed Niaz 17. Laid Sheik. 18. Yaniis Bai. 19. Tiidi Bai. 20. Miisa Akhiind. 21. Mahomed Tej Bai. 22. Abrahim. 23. Shamsh Akhund. 24. Toonganee Serai. There are three other serais within the walls. 1. The Custom House. 2. Charee Hakim Serai. 3. Kirghiz ditto. 4. Aid Darogah Serai. 5. Mahomed Tokhta Bai. 6. Badshahi Serai. 7. Hajf Serai. 8. Kush Najuk Serai. 9. Imam Khwaja Serai. 10. Shukutlik Serai. 11. Abdullah Beg Serai. 12. Hikmut Baki Serai. There are altogether 33 serais outside the walls. Inside the walls there are 84 musjids, and in the Yangi-shahr of Aksu 4 musjids. Inside the walls 800 shops; every house is a shop almost. Outside the walls 500 shops; 35 Coppersmiths; 33 Butchers; 22 Ashkhanas; 19 Bakers. Outside the walls ; 45 Coppersmiths ; 40 Bakers ; 34 Ashpaz ; 50 Shoemakers. The greater part of the population are in the suburb outside the walls. In the entire district of Aksu there are 30,000 houses. The principal merchants resident are — 1. Shumsh Tar Ahhoond. — This man has agents who travel to Turf an, tTrumtsi, and Hi; he is said to be worth 500 Yamboos. 2. Ahmed Shah Bai.' — He trades with Almati, Kashghar, Turfan, 0rumtsi; property valued at 400 Yamboos. 3. Jait JDarogah. — Trades with all the cities before mentioned and possesses property valued at 500 Yamboos. 4. Kassim Bai, Andijani. — Property valued at 1,000 Yamboos : this is the principal trader. The tanab in Aksu district is measured in the square of 12 Kulaj (the length covered by the arms at sketch), a tanab is calculated to take 2 charaks of grain. Five tanabs make a putmun. The tax on crops is fixed by valuation for cotton. ROUTE No. XIII. From Aksu to Ktjchar. Authority, Captain Chapman. From Aksu to Kuchdr. The tash on this road are marked on regular Tash posts. 1. To Jamgu, 4 tash, through cultivation and past frequent habitations; watered by small canals. At Jamgu two serais and a small bazaar ; about 50 houses. 2. To Kara-Yalghdn, 3 tash. Kara (black) Yalghan (tamarisk), a low shrub jungle with willows, &c., tamarisk ; water from springs about half way, where are a few shepherds' huts. ( 146 ) 3. To Yagk-Arik, 7 task. (Oil-canal) Through desert and low hills ; at 3 tash there is a small place of shelter for cattle known as a Dhung ; at the 4th tash there is a similar shelter^ water procurable at both these places, streams from springs ; at the 6th tash is a newly- constructed Karawul and some few houses with cultivation. About two miles to the south-east of this Karawul is a copper mine. The road here passes through a spur of the main range which is, however, very low at this point, canals. About 50 houses in Yagh Arik, a tank, gardens, &c. Yagl' Arik is in the district of Bdi. 4. To Bai, 7 tash. For 1^ tash through highly cultivated district with gardens, houses, &c., to the bank of a river flowing south ; the bed of the stream nearly half a mile across and very stony, the stream is rapid even in winter and is divided into three channels; the cultivation continues from the opposite bank all the way to Bai. This is a long stage and can be broken without difficulty at the hamlets en route. Bai is a walled town with three gates, and has its own Governor. There is a regular urda, four serais, musjids, &c., and three large tanks in the town; there are 62 principal shops inside the walls; there are two Madrassas and two schools. Bazaar is held on Friday after mid-day. There is a garrison of 200 soldiers, 4 Yuzbashis, and 20 Panja Bashis. There are six serais outside the walls of the town. It is estimated that there are 4,000 houses in the entire district of Bai. Mahomedan population of the same type as at Aksu. About 8 tash to the north of Bai are hot springs to which miraculous cures are attri- buted, the springs having been, it is said, called into existence by Hazrat Alii after a fight with infidels. There is a Ziarat at this place, and it is a place of pilgrimage ; the road to it is a very difficult one. 5. ^o Sairdm, 4i tash. The Bai district is left at about two miles from the walls; the entire road is through hamlets and cultivation, a considerable stream running * Sh h-Y R* through a shallow* bed immediately after leaving Bai. Sair^m is larger than Bai, but there are no walls round it. The Hakim is under the orders of the Governor of Bai. There are 16 serais, 11 musjids, 4 tanks, and 75 principal shops, the population of the district is approximately the same as that of Bai. Bazaar is held on Thursday. Intermediate between Bai and Sairdm and to the north about one stage, iron of a superior quality is obtained : this is only worked for local purposes. Grain is exceedingly cheap, and rice is grown, but in smaller quantity than in the Aksu district. 6. To Toffa Dhung. A single stage house, where provisions can be obtained ; water pro- cured from a distance, 5 tash. The road on leaving Sairam very soon passes strips of cultivated ground and through a tamarisk jungle, 2 tash to Kizzil. Kizzil lies in a sandy strip ; a stream is here encountered flowing south ; there are willow trees and a few houses grouped on either bank ; the bed of the stream is 40 yards across ; from Kizzil to Toga Dhung a stony desert ; to the south there are small sand hills, and the road here takes a more northerly direction. 7. To Kuchdr, 6 tash. About two miles from Toga Dhung across a low ridge on the top of which is a serai. This gives a better halting place than the last stage, but water is obtained at the serai with great difficulty and must be paid for, the road turns southwards immediately after crossing this low ridge. There is no cultivation to within about two miles of Kuchdr, but about 2 tash from the ridge is a Karawul in a gorge where the rocks appear to have been subject to volcanic action and are of a very dark color on a high hill to the north-east. At this point is a ruined city, the people commonly call this " Takht-i-Touran," the outskirts of these ruins are actually on the road ; the ( 147 ) hill is o£ bare rock and, as before stated, presumably volcanic, but the ruins are of earth of a deep yellow color quite unlike anything on the hill itself ; there are besides a larg-e number of eaves, excavated for residence ; from below a high wall is visible, which is said to be the wall of an old palace. The city is said to have existed previous to the first Chinese occupation ; the current story is that the city was consumed by fire sent down from heaven owing to thp refusal of its ruler to adopt the Mahomedan faith, the blackened appearance of the rocks having given rise to this tale. From the Karawul to Kuchar proper is 3 task. Kwchdr is a walled city of a circular form with four gates two of which have been lately closed. The garrison of Kuchar is as follows : — There are two Panjsads, 20 Yuzbashis, 50 Panja Bashis, and about 600 soldiers; there are two schools and three Madrassas. The present Dadkhwah is Mahomed Tokhta Beg. There are 205 principal shops inside the walls, 100 of which are always open, the remainder being closed except on bazar days. Four serais inside the wall; the city wall is surrounded by a ditch, which is kept full of water; there are 140 shops outside the wall, 15 serais outside. The Tungani have a separate quarter; they have 45 shops and have 9 serais ; corn is ground by mills in which horses are worked ; these are kept by Tunganis ; the suburbs of the city are large in proportion to the rest of the town, there being only some 400 houses inside the walls, and 1,300 houses outside. The population of the district is said to be considerable, there being, according to calculated accounts, 22,000 houses in the district. Alum and salammoniac are brought from Kuchar, and Pushum of a superior quality ; it is considered the best obtainable. Rice is grown in small quantities, but this is produced in large quantities at Shah Yart, the south of Kuchar, some 8 tash distant. About 16 tash to north of Kuchar a large idol is said to exist, which is cut out of the rock; it is reported to be from 40 to 50 feet in height, it has 10 heads and 20 hands, and it is carved with the tongue hanging outside the mouth ; the mountain behind this idol is exceedingly difficult of ascent ; rumour says that it is resorted to by game of all kinds, but that the animals, owing to the protection of the idol, cannot be killed by the huntsman. A mountain lake of considerable size is said to exist in this neigh- bourhood, the drainage of which falls into the Yulduz and makes its way to Karashahr. The idol referred to above is said to grow thin during the daytime, but to increase in size during the night. Salammoniac is obtained in large quantities from the neighbourhood of a volcano, which is at a distance of eight tash from Kuchar ; this sells in Kuchar at 3 tangas the jing. The people of Kuchar declare that a description of rat circulates freely in the flames of this volcano without being injured ; it goes by the name of Salamander. Surrundoo (alum) is also obtained in this neighbourhood, and sells for a half tanga the jing. The farming of the salammoniac and alum is let out for 4 kurus yearly. There are copper mines between Yagh Arikand Bai. There is no monopoly with reference to the mining for copper ore ; there are regular miners who can be hired by any one who chooses to try for copper, the agreement with them being that they are to receive one-half of what is extracted. The copper is found in a low range of hills, and at a depth of from 30 to 36 feet ; there is a smelting furnace on the spot, which is under Government supervision ; the charcoal and the wages of the smelters are paid for by the finders of the ore, and from the copper extracted one-seventh goes to the Government. It is usually calculated that the ore yields from one-ninth to one-eighth of pure copper which sells in Bai for 3^ tangas the jing. 38 ( 148 ) Route XIV. Route from Karghalik via KHOTAN to Tankse. Authority Captain Trotter from Pundit Kishen Singh. "a * J2i 9 10 Names of stages. Karghalik to Yakin Langar. Chulak Langar Guma (height 4,340 feet). Moji village (height 4,290 feet). Pialma ... Jawa or Zawa Kurghan (height 4,430 feet). Khotan City (Ilchi Shahr) (height 4,490 feet). Yurung Kash (lieight 4,370 feet) . Dol Langar (height 4,420 feet). Chira (height 4,220 feet). Country or district. Karghalik Guma Ditto Ditto Ditto Khotan Ditto Ditto Ditto Ditto s ^ 13 20 23 24 35^ 25 20 3i 13 35 Kemarzs. A small village of four or five houses only, country well cultivated. At four miles is the village of Besharik, and at eight miles Lob village. Cart road all the way. Yakin is a halting-place for traders. Water, fuel, and supplies plentiful. A small village of 10 houses. At four miles is Khush Langar, and is the boundary between Karghalik and Guma. Country up to Khush Langar is cultivated, the rest of the journey is over a sandy desert, and no water except in a reservoir at Dabzokum brought daily from Chulak Langar. Supplies plentiful. A small town and district with about 1,000 houses and a bazar of 300 shops. A Dadkwali or Governor and 200 sepoys are posted here. Two Langars or rest-houses are built on the road, viz., Silak Langar at 9^ miles, and Hajif Langar at 14 miles. The Kilian river is crossed near Guma. Road runs over a sandy plain the whole way. A large village with a bazar. Road through a level culti- vated country. At three miles the dry bed of a branch of the Kilian river, about 200 paces broad : is crossed. Supf Khajam village is 9 miles, Cholo village of 50 houses 10 miles ; Mukhila Langar at 11 miles ; and the large village of Chuda at 19 miles. Road over a sandy plain. Fuel and pasture plentiful. A large village and bazar. At 2\ miles is Kosha Langar; at 10^ miles Kondla Langar; at 14 miles Jhanguia, a large place with a fort. The road from Sanju to Khotan joins at Jhanguia. The entire journey is over a sandy plain without habitation between Jhanguia and Pialma. Water, fuel, and pasture plentiful. Road over sand hills all the way. Water scarce, to be had only at Ak Robat at 15 miles, from a deep pucka well ; again at Imam Salar's tomb and at Jawa Kurghan, a large village and bazar. Supplies plentiful. Road over a cultivated country thickly inhabited through- out. The Karakash river, about half a mile wide with several channels, is crossed at 14 miles. Khotan is a large town, where a Governor and several hundred sepoys are posted. Numerous canals from the Karakash river intersect a large area of country around Khotan. At a distance of 15 miles north-west is the large com- mercial town of Karakash. A large place of 500 shops. At 2i miles the river Yurung Kash, (the same size as the Karakash) in two channels is crossed The road for several weeks is flooded in hot weather. Jade and gold are found up the stream. Road good, and rich cultivation all round. A large village of 150 houses. Excellent road, thickly inhabited, and rich cultivation all through the journey. At lOi miles is Lob village and bazar of 50 shops. A town on the banks of a small stream with a bazar of 150 shops. Road as far as Ak Langar. The first 6 miles over a sandy plain covered with jungle. Elman Bazar 10 miles ; Beshtoghrak Langar at 15 miles ; Aisma Langar at 26 miles ; Yakin Langar at 30 miles : all these places have rest-houses for travellers, with water and supplies. No cultivation except at Chira. ( 149 ) ROUTE XIV. Route from Karghalik via KHOTAN to Tankse. — (Continued.) 0) CJ ^ bo 11 12 13 14 15 16 17 18 19 20 21 22 Names of stages. Earakia Langar Kiria Bazar (height 4,580 feet). Country or district. Khotan Ditto s 25 2n Toghrak Langar Polu village (height 8,430 feet). Khiakde Camt) GhubolikCamp (height 16,960 feet). Aksu Camp Arash Camp (height 16,020 feet). Kiria Daria-i-bash Kul (height 16,880 feet). Camp ... Nikong Chumik Daknak Camp Ditto Ditto Ditto Ditto Ditto Ditto Ditto Rudok Ditto Ditto 23 32 17 25i 19i 12 15i 22 m 25 Remaeks. A small village of 10 houses. Road over an open country. At \0\ miles is the village and bazar of Giilukma ; at 1Q\ miles Domakd village ; from thence the country is covered with high forest. Supplies plentiful. A large town and bazar of 600 shops ; at 5 miles is Ya Langar ; at 9i miles Siasgol ; at 14J miles Yaka Langar ; at 22 miles Phundra village of about 100 houses. From Yaka Langar to Kiria country thickly populated and extensive cultivation. Felt is manufactured at Kiria. A branch road goes from Kiria to Sorghak gold fields as follows i—lst March, Toghrak (height 5,760 feet), 15 miles, a village of 50 houses, road over sand hills. No habitation except at Oi Toghrak. Kiria river 500 paces wide (with several channels), crossed near Kiria. 2nd March, 15 miles, a rest-house with scanty cultivationr ound it. Road open and over sand hills. Zrd March, Sorghak (height 7,060 feet), 34 miles, a place famed for its gold fields : these fields are worked all round the year by men from Kiria, who with their wives and families sleep in temporary huts. One-fifth of the produce is paid as a tax to the ruler of Kashghar, who also purchases the remaining produce at a fixed rate. A small village of 5 houses. Road through cultivation at 15 miles, the rest-house of Bughuz, to east of which, at 200 yards, is the Kiria canal here called Toghrak Ustang. A village of 50 houses with scanty cultivation. Road runs along side the river over a plain for 8 miles, to where the river issues from a mountain gorge, up which the road passes to within 2 miles of Polu, when the river turns off to south-east. Road good. Road, stony and bad, runs along the valley of the Khurap or Polu river to Khiakde. A little fuel and grass. Gold dust is found in the stream here. Camp near Ulugh Shahi Kul. A lake with sulphur mines in its neighbourhood. For 6 miles from Khiakde the road runs along the Polu stream through a narrow gorge between hills called Tangitar, then ascends the Ghubolik. At Diwan Pass, difficult for laden ponies. A gradual descent from thence for 14 miles to Ghubolik. Road bad at the pass. Grass, fuel, and water plentiful. Camp on grassy plain between two small streams. At 12 miles a small pass crossed. Road good. Fuel and grass plentiful. Camp on northern bank of the Kiria river. At 8 miles the Kizil Diwan pass is crossed. Road good, but stony at the pass. Grass and fuel plentiful. Camp on small lake, the source of the Kiria river. Road stony and bad, slightly ascending to the lake. Grass scarce, and fuel plentiful. Camp near a small stream. Grass and wood plentiful. At 16i miles a pass is crossed forming boundary between Kashghar and Thibet, Camp on an extensive plain, with grass and fuel at hand. A mile from camp a pass is crossed. At 10 miles road runs along the west bank of the Yeshil Kul lake (height 16,160 feet). Grass and wood plentiful. Water scarce. Road good the whole way. Camp on an extensive plain. Grass and wood plentiful. Road along the banks of a dry water-course. ( 150 ) ROUTE XIV. Route from Karghalik via KHOTAN to TanJcse. — (Concluded.) o . S3 S Names of stages. Country or district. .S (D • 11 Remarks. 23 Tashliak Kul (height 16,620 feet). Rudok 28J Camp on the northern edge of a fresh water lake. Two small passes are crossed, one at 10 miles and the other at 25 miles. Road good. Grass and wood plenti- ful. Camp at the base of low hills. At 9^ miles the road runs 24 Chumik Lakmo (height Ditto 27^ 16,600 feet). along the eastern edge of the Mangchaka or Mangtza lake. At 13 miles ascends a low range of hills. At 19 miles passes a small lake. Road good. Water from a neighbouring spring. Wood and grass plentiful. 25 Lugrang Camp Ditto 19 Camp between a range of hills. A pass crossed at about a couple of miles from Chumik Lakmo, from thence the road to Lugrang along bed of a partially dry stream. Water scarce. Wood and grass plentiful. 26 Sumzi Ling* (height 15,570 feet). Ditto 17 Camp on left bank of Rudok river. At 4 miles the Kiangla (pass) is crossed, from thence along the Rudok stream to Sumzi Ling. Road stony. Fuel and grass plentiful. Camp on right bank of Naichu river. Road for 8 miles 27 Angche Chiortan ... Ditto 31 good, at 10 miles ascend a pass, from thence along the river to camp. Wood and grass plentiful. 28 Noh village Ditto 10 A small village of 25 houses on the Naichu river. Sparse cultivation. Road good and along the stream. Rudok lies about one long day's journey (by a circuitous route) to the south. 39 Pal Ditto 20 Camp on the upper or east end of the Pangong Lake called here Chomo Ngala Ring Cho. Road in a westerly direction and along the banks of the Lake but very stony. Water good. Wood and grass plentiful. 30 Dal ... Ditto 21 Camp on the Chomo Ngala Ring lake. Road stony, along the edge of the lake. Water good. Fuel and grass plentiful. 31 Aot Ditto 12 Camp on south side of Pangong. Road crosses the lake by a shallow ford near encampment. About 2 or 3 miles north-west is the ruined fort of Kharnak. Water fuel and grass plentiful. 32 Thakang Ladakh 29 Camp on south side of lake near a mountain stream which falls into it. Scarcity of fresh water. Road stony and along banks of the lake. Wood and grass plentiful. 33 Shushul Ditto Si A village of 30 or 40 houses, with sparse cultivation. Road good, and supplies plentiful. 34 Lung Barma Ditto 31 Good camping ground in the Lungchu valley. At 4i miles from Shushul cross Gongta-la-pass, from thence along river bank. Road stony. Fuel and grass plentiful. 35 Tankse (Station V of Eoute III) (height 12,900 feet). Ditto From Tankse to Leh, vide Route III. Total distance ... 24 742 A village of 50 or 60 houses, with some cultivation. Road stony and along Lungchu stream. Fuel and grass plentiful. * The Pandit's thermometers were broken here. ( 151 ) ROUTE No. XlVa. Yarkand to Khotan. Authority Captain Chapman (compiled prom Notes by Ramchand). FROM yarkand TO KHOTAN, BY KARGHALIK. From Karghalik to Eg-un, two tash, across a ravine and for four miles across desert, and then entering cultivated land which continues for two miles, after which there is desert close up to Egun, beyond Egun desert, at three miles a Langar (Gombaz) with tank and two old tombs ; eight miles "beyond this through a bare desert to a place where a tank (containing about 20 mussuks of water) is prepared and covered in. Water is brought to this daily (one donkey load) from a distance of ten miles, under the orders of the Hakim of Giima. Beyond this 10 miles through desert to Chulak (Langar), where there is a tank. This is the usual halting stage, and there is a post here for the examination of passports. Erom Chulak, 13 miles, to Serik Langar through desert. These Langars are all com- fortably fitted up for travellers. Four miles beyond this is another Lacgar, where are trees and water ; at this point the road divides, one going direct to Khotan and the other to the Guma Bazaar ; six miles to Giima by taking the direct road, and not going to Giima about four miles are saved, but all Rahdaris (passports) have to be shewn to the Hakim of Giima. Giima is a considerable place ; its district comprises that of Sanju. It extends from Egun to Pialma in the Khotan direction, and southward as far as the Sanju pass. Giiraa, for three miles through cultivated ground, then through desert for 13 miles to Mocha. Here is a comfortable rest-house prepared by a Mullah where travellers are entertained. A road diverges at this point to Sanju, which is distant 15 miles. From Mocha (or Moji) for about 14 miles through desert to Zungoe or Jhanguia ; here is a small bazaar, and the place is surrounded by a wall built in the time of Habbibulla ; cultivation beyond this for some two miles, where there is a Langar, after which there is desert for 12 or 15 miles. Here there is a Langar, but no tank, only a well of extreme depth from which water is drawn. Five miles to Pialma. Here there is a small bazaar. From Pialma 15 miles to Ak-Robat. Here is a new rest-house prepared by the Amir's orders where travellers are entertained ; there are two wells in the court-yard, which is 100 feet square : these wells are deeper than the one at the last Langar. Here there is a very high pole upon which a bell is hung, in order that during storms of sand travellers may be directed to the rest-house ; during the night it is cus- tomary to place a light on the top of this pole for the same purpose. There is a regular establish- ment for the care of the serai. From Ak-Robat five miles to the Mazar of Imam Maho- med Shah. This stands in the middle of a desert ; a large number of pigeons are kept by the Shaik custodian, for which a regular allowance of grain is made ; the road is through a heavy sandy desert. There is a high pole at this point with a bell on it similar to the one at Ak- Robat. Through desert for five miles to Jawa : this is a small village of 20 houses, also sur- rounded by a wall in the time of Habbibulla ; there is a post for the inspection of passes ai\d for search for gold and jade carried out of the Khotan District without permission. This is the point where the Atalik halted before the capture of Khotan, and from whence he enticed Habbibulla into his camp. To the east of this village is a considerable stream flowing from the Sanju Diwan, which is crossed by a wooden bridge built after the fashion of Kashmir bridges. This is about 20 yards across. On the other side is a regular rest-house for the Hakim of superior construction. From this point cultivation and habitations extend uninterruptedly to Khotan at a distance of 25 miles. Ten miles from Jawa the Karakash is crossed, where the breadth of the bed of the stream is about 400 yards. On the bank of the Karakash there is also a rest-house built to accom- modate travellers detained by the rising of the river. 39 ( 152 ) Seven miles short of Khotan is Do-Shamba Bazaar : this is a small village. Marches. Stages. 1. Posgam. 2. Karghalik. 8. Egun. 4. Chulak. 5. Guma. 6. Mocha. 7. Zungoe or Jhanguia. 8. Pialma.^ 9. Ak-Robat, a short stage owing to heavy sand. 10. Beyond Jawa, to the banks of Karakash.' 11. Khotan (Ilchi). A road starting between Pialma and Ak-Bobat makes up the Karakash valley to the Suget pass and the Karakorum. To the east of Khotan and flowing close to the Yangi-shahr, 500 yards outside the wall, is the Yiiriing Kash River; the bed of this stream is 600 yards broad, and in the dry season it flows in two channels; the road is often closed in this direction. Niaz Beg attempted the construction of a bridge, which was carried away by the stream. From Khotan up the valley of the Yurung Kash to Ladak ; this was the road taken by Jumma Khan ; it is said to be very diflicult. From Khotan (Ilchi) by 6 marches in a southerly direction crossing the Yurung Kash at (Ilchi) — 3 tash to Sumpula. 3 Hasha. 3 Gunjutagh. 4 Nura. 4 Imam Mazar. 5 PoM. (These six marches are through cultivated lands by small vil- f lages; no river encountered. Fifteen marches from this point by a comparatively easy road to Changthang where the road between Leh and Lhassa is joined. Grass and wood dming the entire route. Changthang is from 12 to 15 marches distance from Leh. Water is procured by digging. This is a summer route ; it was pursued by Nujjuf Sbah, Envoy of the Maharaja of Kashmir, in 1864, with Nika, Mogul, who is now in Khotan. The province of Khotan is divided into the following districts ; — Ilchi (Khotan proper). Karakash (a Beg). Keria (a Beg). .Chira (a Beg). Yurung Kash united with Ilchi. Nia united with Chira. The city of Khotan is of an irregular form, the circumference being approximately a mile. The Fort and Urda of the Hakim are outside. The old Chinese wall round the town has fallen into ruin, but a wall now stands encircling the town and a large portion of the district at some considerable distance from the place itself : this is said to be some 20 miles in circuit. The gate through this on the Yarkand road is 1^ miles distant from the town. Khotan is the great manufacturing city in the Amir's territory. The province yields very little cotton and very little grain, these being imported from Kashghar and from Guma ( 153 ) and Karg-halik. Copper comes from Aksii and is worked into vessels, in great favor throughout the country. Rice is obtained from the same place. Khotan is the great silk-producing province. Its gold mines and supply of jade are sources of vt^ealth, the population, however, remain poor owing in a great measure to their indolence, work being taken up only as there is immediate necessity for the supply of daily wants. The immorality of the women of Khotan is proverbial, and the excess of women over men leads to much licentiousness. The inhabitants are chiefly artizans as distinguished from the cultivators of the other provinces of the Amir's kingdom. The resources of this province may be best arrived at by re^aewing them separately. ISorghak. Kappa. Chuggulaka. Charchand. Kard,tagh. There are said to be altogether 22 places in which this mineral is found, but the above are those which are regularly worked. 3,000 people are employed at Sorghak; the mine at this place is said to be 400 feet deep. 4,000 people are employed at Kappa; the mine is said to be 100 feet deep. At the other places there are no more than 40 or 50 workers. The Sorghak gold is obtained in small beads and is of a red color. At Kappa, large nuggets are obtained, but the gold is of a light color and mixed with sand. Gold is also obtained in the Yuriing Kash sands after the flood of the hot season has passed. The only tools used iire a pick and shovel; no sieve is in use, but the soil is dug out in blocks and disintegrated by the heat of the sun. The sieve is used in extracting gold from the sand of the Yiirung Kash river. On this last there is no tax paid. The working of the mines is thus conducted. The workers are the poor of the country who sell the gold they obtain to established buyers, who keep a supply of utensils of food, &c., to meet the requirements of the workers. From these gold purchasers one-fifth of the yield is at once confiscated as the property of the Amir, who retains the right of purchasing any further quantity he may require at 120 tangas per ser (the market value being 138 tangas the ser). The whole of the gold obtained is indeed supposed to be purchased for the Amir, but a large amount finds its way surreptitiously into the market. On the road between Keria and Khotan there is a regular searching house where men are stripped if they are supposed to have concealed gold about them, women are examined and are then made to jump over a ditch, in order that any gold they have concealed may fall out. No large guard is kept at the mines, but a small detachment of soldiers watch the proceedings of the buyers. The officials, however, even to the Beg of Keria, who is in charge, are said to be open to bribes and to study their own advantage. The punishment for secreting gold is very light. The probable yearly yield of gold in the Khotan District is said to approach 7,000 sers, of which 5,000 sers,^ reach the jAmir and about 2,000 are smuggled into the market. The sale of gold is winked at, though disallowed. If a merchant is discovered to have obtained gold, no more than one-fifth will be confiscated, and the remainder is purchased at the fixed price. Gold is readily bought up by merchants from India and Andijan. A profit of one-eighth may be realized by conveying Khotan gold to either country. Silver is also found in the province, but the yield was found' insufficient to pay the working expenses and the mine opened has been abandoned. ( 154 ) Jade is obtained near the bed of the Yiirung Kash. There are two principal mines, one at a distance of 15 miles and the other at 25 miles from Ilchi. It is also procured from the bed of the river. The tax on the working is one-fifth part paid into the treasury, and a tax of 1 in 40 from all traders who carry jade to other places ; besides this all pieces that are of superior size and quality are boug-ht up at a price fixed by the Dadkhwah of Khotan. The old skilled carvers of jade have almost disappeared since the Chinese have been ejected. The mines and the working of the jade were closed until two years ago when Chinese traders began to reach Alm^ti. The Amir, however, allowed the market of Almati to be flooded the first year and so much of the jade carried there from Khotan was inferior that it was not purchased, but returned to Khotan in this year. The value of large pieces of this mineral may be judged by the following fact : — Quite lately some five men obtained a large block of a good description w^eighing some 40 jings ; it was taken before the Dadkhwah, who purchased it as it was, in an uncut state, for 12yamboos, the market price fixed upon it was however 60 yamboos ; there was however no single merchant rich enough to purchase. The existence of gold and jade is necessarily demoralising to the papulation ; the number of workers in jade fluctuates, but the supply is in- no wise exhaifeted. Coal is said to be obtained in the Kuen Luen, but it has not been used since the Chinese were in authority; it was then brought from a considerable distance to Khotan. 8ilk. — From the earliest time Khotan has been celebrated for its silk manufactures. Sericul- ture in Khotan is the same as already noted in Yangi Hissar, this is purely a domestic business : there is however a regular sale for cocoons in the market, the purchasers are regular traders who sell ag-ain to the reelers who purchase as they are in want of cocoons. There are poor people who dispose of the spun silk in the markets, which is chiefly bought by Andijani merchants ; there are however a great number of silk weavers (these color their own silks) . Silk cloth made at Khotan is not exported, but spun raw silk goes in large quantities to Andijan. It is also found profitable to send the '^ waste'^ to the Almati market, where it is purchased for Russian paper manufactories (it is a rumor that Russian notes are made from silk " waste'') . Reeling does not go on during the winter season. The white, black, and red and a fruitless mulberry are all known in Khotan, but the worms are fed only on the leaves from the fruitless tree and from that yielding a black fruit. The produce of silk in its various stages, from the tending of the worm to the final ope- ration, affords occupation to the bulk of the population in the Khotan province. There are two kinds of silk, the white and the yellow, the latter being most esteemed : this is known as " Taiful," and is chiefly bought up for Andijan ; the former is called " Kalawur," each of these are classed in two classes. As a rule the silk is reeled off on a single chirka, but lately an Andijani has set up a wheel working 16 reels at one time. Carpets. — Khotan carpets are celebrated for the excellence of their manufacture and for the variety of their patterns ; they are made at three places in the Khotan province, more particularly Sumpula, Yui'ung Kash (Char Shamba Bazaar), and at another village on the Keria road about three tash from the city (Se-Shamba Bazaar). Carpets are made of silk and of wool, gold thread is also sometimes worked into the silk carpets. The wool used in the carpets is chiefly obtained from the hill districts through the Kirghiz ; it is spun off and sold in the weekly bazaars. The dyeing is carried out by the carpet-makers. The mordants used are — alum for dyes of yellow, brown, and red, and their various shades. Grape juice for blues, and for mixed colors. If green, the wool is first dyed yellow then put into an indigo solution. If purple, it is first dyed red and then put into an indigo solution. The dyes are indigo, madder, tookmuck (a seed), 'bukum.' ( 155 ) The price of labor is exceedingly cheap in the Khotan District, and the carpet makers are hired as required by those who are rich enough to purchase the materials and set up the frames. There are two kinds of frames, standard, which are placed perpendicularly, and long flat frames near the ground ; the latter are usually required for very large carpets, but the standard frames allow of better work being done. The size of the carpets ranges from the small saddle carpet on which one man works at a time to carpets 3| yards wide, upon which 10 men are employed at one time. The pattern is given out by head of the party whom they term Aksakal. The patterns do not exist on paper, but are passed on from master to pupil and so remain from generation to generation. There are some 200 masters who are known for their carpet work. It is to be noticed that in comparison with other parts of the country sheep are very abundant in the Khotan province, and that the wool is Khotan lamb skins of white color, form the linings of hats and posteens. Men employed in carpet making under the Government receive 20 phools daily each man, and are not supplied with food ; in ordinary houses, the daily wages is 10 phools and food for the day. The copper vessels made in Khotan are superior to all others prepared in the Amir^s terri- tory ; they are worked in a varietj'^ of patterns. There are about 30 shops at Khotan (Ilchi) where these are made. The copper, as previously stated, is obtained from Aksu. Vessels for tea and for water are those principally made, the iron vessels from Russia being preferred as cooking utensils. Patterns of various kinds are cut upon the Khotan copper work in very neat fashion. Iron. — Iron work, stirrup irons, spoons, knife handles, &c., made of iron are inlaid with copper wire, which is usually of English manufacture and is obtained from Hindoostan ; there are some 12 masters in this work at Khotan ; the pattern is cut out in the iron of the stirrup, spoon or whatever article it is intended to ornament, and the wire is run into the pattern and the whole heated in the fire. Rides. — Cow hides are largely exported from Khotan to Yarkand with sheep skins and goat. This is due rather to the existence of a large number of cattle and sheep in the Khotan province than to superiority in the manner of preparation. Skins of animals obtained in the Kuen Luen are also exported for the lining of posteens. Sandal-wood and tea (brick) are obtained from the buried cities. The sandal is worked into beads, and the tea is sold in the market. The buried cities proper are said to be at a distance of many marches to the east of Khotan. A discovery of buried ruins has, however, lately been made quite close to the city of Khotan (Ilchi) at a distance of about four miles to the north-west. A cultivator working in his field was watering his crop ; on the water suddenly disappearing into a hole and continuing to be absorbed, he dug up- the place where the water disappeared and obtained a golden ornament said to have been a cow • Lately excavations have been ordered by the Dadkhwah and more gold has bee^ found; the diggers are paid for any gold they may excavate at 110 tangas the ser. I^^ the beginning of April 1874 a gold ornament of about eight sers weight was found by a man who had gone out in search of charcoal — this was in the shape of a small vase and had a chain attached to it. Rumour declared it to be neck ornament of the Great Afrasiab, and the finder was declared to have hit upon the spot where Af rasiab's treasure was buried ; he was paid for the ornament at the rate of 100 tangas the ser, and a party was at once organized to search the neighbourhood. At present no fresh discovery has been made, and I cannot in any way fix the locality, but it is at no great distance from Khotan itself. Peices. Cotton cloth, Kdm, per than ( 7 yards X \ yard) „ Tolma, „ ( 7 „ „ „ ) ... „ Chakman „ (22 „ X ^ „ ) ... Kummerband (10 girra.s X 2| yards) 40 Tangas. Phools 1 30 2 25 4 1 Tang as. Phools ... 6 ... 4 ... 3 ... 20 ... 17 ( 156 ) Mixed cloths of cotton and silk. Tangas. Phools. Gazina (10 yards X 9 girras) per than ... ... ... ... 6 „ ' (10 „ „ „ ) „ No. 2 ... ... ... 4 Mushroo(7 „ X i yard ) „ ... ... ... ... 7 ,. ( 7 „ „ „ ) „ ... ... ... ... 5 This is largely exported and is in general use in the province ; it formerly obtained nearly double its pressent price, but the introduction of a large quantity of red chintz (Gulanar) from Russia has swamped the market. Mushroo is both dyed and stamped. Silks. Dariaye (7 yards x \ yard) per strip ,, (7 „ „ „ ) „ No. 2 »> (7 „ „ „ ) »> » 3 Silk Chakman (13 yards X \ yard) (13 „ ,. „ ) No. 2 This is chiefly sold for the making up of chogas, the coloring of the strips is invariably made in pairs ; the size of the than is however arranged so as to make one choga. This silk is largely used for chogas throughout the country, but is not carried across the frontier. Tangas. Phools. Shiaye (7 yards X 5 yard), Nos. 1 and 2 ... ... ... ... 15 Dorooya(7 „ „ „ ) „ „ ... ... ... 10 These descriptions are largely imported from Andijan, the manufactured articles of Khotan being coarser than those from Andijan. Numdahs for spreading on floor — Tangas. Phools. Banging from the prayer Numdah ... ... ... ... 2 25 To large Numdahs at ... ... ... ... ... 20 They are made largely at Yurung Kash. Tangas. Phools. Numdahs for packing bales of merchandize of a dark color ... ... 1 10 Rope is made from the bark of the mulberry and also from hemp, which is, however, very little grown in this province, each rope ... ... 2 ' Rope is also made of wool, per rope ... ... ... ... 2 There is no large sale in the Khotan market for foreign goods, nor are there established merchants in Khotan who trade across the frontier ; the result is that goods are not so easily disposed of as in the Yarkand market, and there is more difficulty in obtaining by barter articles required for a return consignment. Shrines in the Province of Khotan to which pilgrimages are made are — Imam Akbar, Mazar, in the Karakash valley. „ Asgar „ „ „ " K ^^ ■ "\ ^^^ Bakr. [ Yurung Kash valley. „ lehran at Chira. „ Nasrudin . . . ] „ Khw^mudin . . . f * n ^ ht i.^ \. " rz 1 T- }- All at Mayartagh. „ Zahurudm ... I •' ° „ Mayaudin ... j „ Azail ... Kerai. „ Saydlik ... Nia. Limgi Khanem, the gift of Imam Jafir Sadik, at Nia. Imam Jafir Sadik, in the desert beyond Nia, the principal place of pilgrimage from Khotan. Imam Musa Kasim, at Khotdn. ( 157 ) Trees known in the Khotan district are — / Tarek ... ...1 iKaraTarek Safeda, poplar, six kinds J Kuppuk Tarek ... J Hangi Tarek •9 (3 O / Malja Tarek 1 V^Tagh Tarek Stages through desert 7. Booksam (or Bash Bonksem). 8. Kolu. 9. Badlik Kotagh {or Bedelik Kudok). 10. Nurs-shakum. 11. Balfuz-nakum. 12. Khal. 13. Darialoe. 14. Mahtung. Here cultivation is encountered. 15. Karatal {Kharatal). 16. Besh-turkimirum {Besk-arik). 17. Aksii. These are stages for donkeys, the chief trade being carried on with donkeys — copper, rice, iron, gold silk, and country cotton cloths going from Khotan. A trader with horses will accomplish the journey in 10 stages. ' , Noi£. — This route would appear to be reliable. Names in italics are foui^ by me in Klaproth's Map. — H. T. ROUTE XXIV. Feom Khotan to Folu (direct). Authoeity Captain Chapman, from Native information. Khotan to — 1. 2. 3. 4. 5. 6. Sampula Hasha Ganju-tagh Miira ... Imam Mazar Polu ... or ... 3 tash. "^ ... 3 tash. ... 3 tash. A large town. ... 4 tash. Chehar Im^m, 4 tash. ,..5 tash. J These six marches are through cultivated lands with small villages. No rivers encoun- tered. Note. — This route agrees very well with another procured by me frc )m a different source. — H. T. EOUTE XXV. Khotan to Charchand and Kurla. Authority Captain Trotter, compiled from various sources of native information. Khotan to — 1. Dol 2. 3. 4. 5. 6. 7. 8. 16i 35 25 27i 15 Miles. y Vide Route XIV. Chira Karakia . . . Kiria Ui Toghrak Yessulghun. Auras. Naia. At a day's journey from Naia in a southerly direction lie the Sorghak gold fields. The river, from Sorghak flows by Naia and passes in a northerly direction to Maz^r Imam Jafr Sadik, a favourite place of pilgrimage, two days' journey north of Naia. From Naia to Charchand lie two roads across the desert, the northern road passes the camps of — Kara Buran. Yantagh Kuduk. Kok-muran. Yang-arik. Charchand. 9. Baliklik. 14. 10. Yer-tunguz. 15. 11. Haidil-shah Kuduk. 16. 12. Andhira. 17. 13. Shiodang. 18. ( 163 ) At all these camps wells have been dug. On the alternative southern route the camping grounds are as follows : — 9. Subed (from here a road goes to Sorghak). 10. Apalik. 14. Tokpai. 11. Shrine of Bibi Tujilik. 15. Hassan Gunj. 12. Moljia. 16. Achian. 13. The Kapa gold fields. 17. Charchand. The marches from Naia probably average between 20 and 25 miles in length. Charchand is a place of some importance, and is said to be used as a penal settlement ; a large river is said to flow through it coming from Thibet and ultimately finding its way to Lob, The geographical position of Charchand is not fixed with any degree of certainty but it is probably about equidistant from Keria (Route XIV.) and Kurla (Route XV.), to which latter place a road passes via Tartang and Chaktuk. Between Charchand and the Lob District are said to be oases where wandering tribes of Sokpos (Kalmaks) wander about with their flocks and herds. Near Charchand are the Khadlak gold fields, where 100 diggers are employed. Road down Panjah or Amu river the whole way. Valley very narrow and precipitous, and not practicable for laden horses. ROUTE XXVI. KiLA Wamur (Roshan) to Kila Khumb (Darwaz), authority Captain Trotter through Abdul Subhan, from native information. Kila W^mur, chief town of Roshan. 1. Wazmid. 2. Amern. 3. Kila Chamarj. 4. Wadkhod. 5. Jarf. 6. Kila Khumb. An alternative road goes in a northerly direction by which Kila Khumb may be reached in three days from Wamur. ROUTE XXVII. From Bar Panjah (Shighnan) to Kashghar, authority Captain Trotter through Abdul Subhan, from native information. From Bar Punjab — 1. Ghund village. 7. Murghabi. 2. Ishtim or Wiar. 8. Rang-kul. 3. Charsim Fort. 9. Moji Chakr Arghin (Kirghiz) 4. Bugrumal Pamir.'^ From here a road 10. Bulghar Pass (very high). diverges to Khokand via Bartang. 11. Tashbalig. 5. Sasik-kul (2 lakes and Pamir) . 12. Kashghar. 6. Kara-su. ROUTE XXVIII. From Bugrumal Pamir (No. 4 of Route XXVII) to Khokand. 6.) 1. Marjanai. 2. Sirich Fort (the capital of Bartang) . 3. Kara Bulak (on Murghabi river). 4. Takhta Korum. 5. Altun Mazar, cross the Mazar Pass. 11. 7. j Two marches in the Alai. 9. Osh Kurghan by Draot (Deraout). 10. Marjilan. Khokand. Note. — These routes are very meagre, but have been used in conjunction with other sources of inform- ation in the compilation of my preliminary map. From Wamur there is a pathway up the Murghabi Eiver to Sirich Fort, but so difficult that travellers nearly always go by the Ghund river in preference. — H. T. * Probably the western prolongation of the Alichur Pamir. 42 ( 164 ) ROUTE XXIX. Feom Yuk (see No. 15 op Route VIII) to Chitral, authority Captain Biddulph from Native information. Yur to — These ai-e marches for a man on foot. 1. 2. 3. 4. 5. Suneen. Over pass.* Kusht. Topkhana. Gazan (inhabited). 11. < 6. Manzagram. 7. Mastuch. 8. Booni. 9. Risht. 10. Ragh. ^hitral. ROUTE XXX. Tashkurghan, No. 9 of Route VIII, or Sarigh-Kul to Hanza in Kanjud (Dr. Bellew). 1. Davdar, 4 tash. Across valley and amongst hills to camp ground. The first stage from 'Tashkurghan. No habitation. 2. .Tilghar, 3 tash. Easy road amongst hills. Desert. No fuel or forage. 3. Ghajakbay,t 4^ tash. The same. 4. Rang or Zastol, 4 tash. Up a narrow gorge and over a glacier to 5. Rangal, 3 tash. Down a defile along a torrent. Road rough. Hills bare. 6. Talictay;, 4 tash. The same as last stage. 7. Lupgal^ 5 tash. Continue down the valley. Mountains high and bare, 8. Udmurkish, 4 tash. Desert country amongst hills. 9. Misgar. ^ 10. Sas. / These are all the names of camp grounds. Each a day^s journey from 11. Khybar. I the other in vallies amongst hills. Streams from all sides, and 12. Passu. ( scanty brushwood. Country very difficult and bare. Under snow 13. Garnit. \ for half the year. 14. Syaban. J 15. Muhammadabad. First village from Tashkurgh^n. Fields and orchards on river bank. 16. Kanjiid, 3 tash. 1_,000 houses and fort. Capital of Hanza on a large river. Fields and fruit trees in terraces on hill slopes. ROUTE XXXI. Sarhadd, No. 14 of Eotjte VIII, Wakhaj^, to Kanjtjd (Dr, Bellew). 1. Shawar. In a glen. Fuel, water, and grass. 2. Langar.J On border of Pamir Khurd. Grass, fuel, and water. 3. Khaldarchit. In a glen of the Pamir hills. Ditto. No trees. 4. Luptuk. A deep narrow defile in the mountains. Grass and water. 5. Irshal. Over a high mountain and a glacier down to 6. Ast^n. A long march down a defile along a river which flows all the way to Kanjiid. 7. Ispinj. A short march down course of the same river which is unfordable and only crossed on the ice in winter. 8. Reshit. Short march down the river. 9. Kirmin. Ditto. 10. Gircha or GoorM. Twenty scattered houses and terraced fields. 11. Khybar. 6 houses. Ditto. 12. Passu or Basoo. 20 houses. Ditto. * Closed for several months in the year, t Probably Kila Ujadbai.— H. T. ( 165 ) 13. Sissiini or Sasoni. 10 houses, scattered houses and terraced fields. A very short stag-e to 14. Gholki. 30 houses. Ditto. 15. Gulmik or Gulmit. 100 houses on right bank of river. Leave river here and cross Durband Kotal, not high nor difficult in two stages to 17. Kanjud. 1,000 houses and a fort. Capital of the country, on a river which flows to Gilgit. These routes^ XXX. and XXXI., prohahly meet about Gircha. The accounts given are incon- sistent, but as very little is known of the Hanza country, every contribution to a knovdedge of it is valuable. — H. T. Yarkand to Shahidula via Kilik (Captain Trotter). 1. Yarkand to 2. Karghahk ^ Route I. 3. Beshterek. 40 or 50 houses ; 4 tash. 4. Balerak Kurba. 200 houses, 2 tash. 5. Akchik. 4 or 5 houses, 2 tash. 6. Takmk or Chakmk Camp, 4 tash. Kirghiz camp. Plenty of cattle. 7. Tupa Diwan, 4 tash. Pass. Good road. 8. Azghan or Kilik. Plenty of wood and grass. 9. Kilik Diwan. Higher than Tupa Diwan. 10. Larcha Ya Tuba, 3 tash. Good road. Plenty of grass and wood. 11. Gor Jilga, 4 tash. 12. Mazar Khoja, 2 tash. Large stream. 13. Shahidula. (Routes I. to III.) This road is said to be shorter and better than any other road between Yarkand and Shahidula ; grass and wood are to be found at every stage. Shahidula can be reached by a horseman in five days from Yarkand. The man who supplied this route has tried all the roads from Yarkand and reports the road via the Kilik Pass to be much the best for foot passengers. ^* ^" book is due on th , *'*'^'« K<''^"»&V,h'^^ 0)476B Yl ^318836 1^ '^'■SitJM^ m mirnm^^, )M.\^ ^•,^^' •:^'r;V