The

Geographical Journal.

No. 4. OCTOBER, 1909. VoL. XXXIV.

A NEW BRITISH ANTARCTIC EXPEDITION.

Evkky one interested in the exploration of the South Polar region will be glad to learn that (’aptain R. F. Scott is now prepared to complete the work so well begun by himself, and continued by ^Ir. Shackleton, in the h’oss ({uadrant of the Antarctic continent. So far as the attainment of the South Pole is concerned, it is well known that only about 100 miles remain to be covered before that point is reached. It is true that this spot may not exhibit any features of exceptional scientific interest, and the Royal Geographical Society could hardly advocate an expedition with the South Pole as its sole objective. Captain Scott, however, has proved his capacity as a lea'ler, and he is certain not to adopt any narrow view of his duties. Moreover, it is useless to deny that there is a widespread feeling that this feat should be accomplished by an expedition sailing from this country. As citizens we all share in this sentiment, and we may, moreover, as geographers, legitimately take full use of this opportunity of obtaining public assistance for Antarctic exploration.

Captain Scott estimates that at least £40,000 will be required to eejuip his expedition. Part of this has been promised. Captain Scott has already begun his preparations, and if he is to be at his base of operations in December, 1910, as he should be, ho will have to leave England not later than August next. It is to be hoped, therefore, that those among the FelloAvs of the Society who arc disposed to lend sub¬ stantial support to this new British Expedition, will do so without delay. Subscriptions should bo sent to Messrs. Cocks, Biddulph & Co., Charing Cross, S.W.

The President has written the following letter to Captain Scott : —

“ Koyal (ieugraiiliical Society, 1, Siivilc Row,

“ ISurlingtoii (iatdens, Luiiduu, W., Scjit. 17.

“My DKAii Scott, — Many thanks for your letter in which you tell mo about your proposed Antarctic expedition. You may certainly count

No. IV. — OcTOBEK, 1909.] 2 c

362

A NEW BRITISH ANTARCTIC EXPEDITION.

on me and on the officers of this Society for every help it is in our ]»ower to give ; and, after 1 have laid the matter before our Council, I am certain I shall be authorized to inform you of their warm approval of your renewed efforts. In the meantime, please accept my best wishes.

“ Yours sincerely,

“Lionaud Dauwin.”

The objects of the expedition are —

( 1 ) To reach the South Polo.

(2) The further scientific exploration of the Koss sea area.

To achieve these objects, it is projwsed to establish two bases, one in

McMurdo sound, and one in King Edward Yll. Land. In which of these bases the main force of the expedition is established depends entirely on circumstances.

Should a favourable sj)ot bo found in King Edward VII. Land, it may bo that the greater force, with provision for an extended southern effort, is landed there, and to the lesser force is given the continuance of the im]>ortant station-observations to be made in McMurdo sound, and the working out of further geographical and geological details in that region. On the other hand, should the wintering spot in King Edward VII. Land prove unsuitable for southern exploration, or its latitude greatly increase the distance to be traversed to the pole, the programme will be to establish a small party in that region capable of conducting the

THE LOHIT-BRAHMAPUTRA.

363

Btatiuii obHervatiuna aud the scientific exploration of the region, whilst the main force is landed in McMurdo sound.

In this case it is hoped that the geographical work of the main (Mirty will not he confined to the main southern effort. If, as is anticipated, a large quantity of provisions can be carried to the foot of the great glacier, a sufficient portion, together witli the remaining I)onies (which cannot be taken up the glacier), can 1)C utili/>cd by a small party to prosecute the further exploration of the mountainous land in a south-easterly direction, and determine finally whether it has connec¬ tion with King Edward Vll. Land, or whether it continues in the direction of Graham Land. Such work should bo of great geographical importance, and, together with the exploration of King Edward VI 1. Land, should throw great light on the nature and extent of the Great Harrier. The general outline of the scientific work of such an expedition is too well known to need repetition, and the advantage of comparative observations in two stations needs no comment. It may be well, how¬ ever, to lay stress on the advantages which such sciences as magnetism and meteorology will gain by the duplication of observations in known places as being the only means by which secular change and seasonal variation can be determined. Attention might be drawn also to another consideration of great importance. The collaboration between those who have made observations on recent exj)editious, and those who have dealt with such observations, should lead to the solution of many elusive problems. For instance, the examination of geological results may point to the exploration of critical localities whence data of the highest import¬ ance may be gleaned. Again in meteorology, it seems not too much to exjKJCt that consultation will lead to devices whereby some accurate con¬ ception of the precipitation and evaporation in various regions may be ascertained. Yet again, the extent to which the study of ice-physics can be carried should now be more surely known, and suggestions as to the prosecution of such study should be more clearly defined. For the better elucidation of these and kindred matters, it is hoped that it will be possible to consult the best expert advice, aud secure the formulation of the most detailed and practical instructions. In conclusion, it may be urged that such a programme as has been briefly outlined cannot fail to produce scientific results of high importance, as well as servo a patriotic end, if a proper attention is paid to the details of organization.

THE LOHIT-BRAHMAPUTRA BETWEEN ASSAM AND SOUTH¬ EASTERN TIBET, NOVEMBER, 1907, TO JANUARY, 1908.*

By NOEL WILLIAMSON.

Tiiii river shown on the mai)S as the Hrahmaputra enters the north-eastern corner of Assam through the Mishmi hills. To the Assamese it is known

364

THE LOHlT-BRAHMAPUTEA BETWEEN

as the Lobit ; Tibetans call it the Zajnil Chu, and the Mishmi name is Tellu. It is the eastern branch of the Tsangpo or Brahmaputra, and after meeting the Dibong, another branch from the north, the two join the Tsangpo, here known as the Dihong, some 15 miles to the west of Sadiya.

Attempts to trace the Lobit have been made at various times — Wilcox and Griffith in the early part of last century ; then Kowlatt, and, in 1869, Cooper. Of these Wilcox reached the farthest point, the Ghalum river. In 1854, the French missionaries Krick and Bouiy^ penetrated as far as the Zayul valley, in which Kima lies, but were murdered by Mishmis the same year. In 1882, the great traveller A-K carried his explorations down the Zayul Chu as far as Sama, a few miles below Ilima, and prior to this certain influential Khamtis had on three different occasions traced the river from Assam up to Kima. In 1885 an attempt was made to prove that the Zayul eventually found its way, not into the Brahma¬ putra, but into the Irawadi. The following year the ground was cut from IhjIow this theory when N(^edham and Molesworth followed the river up to within a short distance of Bima. None of these travellers gave us a reliable map of the region, and it remained unvisited for the next twenty- one years.

The first two years after my arrival at iSadiya in 1905 were occupied in making myself acquainted with the frontier tribes generally. Anxious to see something of the Mishmis and make myself acquainted with their country, it was my intention to penetrate into the hills as far as possible with local transport, and, if 02)portuuity occurred, to map in the course of the Lohit as far as the limits of British territory. That 1 should have been in a position to carry on survey work was duo to the great interest taken in the matter by Sir Lancelot Hare, k.c.s.1., Lieutenant-Governor of Eastern Bengal and Assam, and to the valuable assistance given me by Mr. John Eccles, m.a., of the Survey of India.

The names of the trilies inhabiting the hills surrounding the Assam valley as shown on our maps are those by which they were known to the Assamese when we occupied the country-. These names are still in general use, but they are neither recognized nor understood by the races to whom they refer. Thus the term “ Mishmi ” is used for the tribes living in the hills between 95° 30' and 97° longitudes along the north¬ eastern frontier of Assam, and the peojile through whose country the first l)art of the present journey was made are commonly known as Digaru Mishmis, presumably because they live somewhere near a river called the Digaru. They sj)eak of themselves us Taroan.* The latter half of the journey was amongst a tribe calling themselves Miju (M’ju).

I left Sadiya by Ixjat on November 28, 1907 ; and on December 7 a Taroan chief, with several men of his village, apjreared at my camp at Samkha in rej)ly to a message that I wanted him to help me with j»ortors to take me as far as his village, Tashalun. lie arrived with a largo * Pronounced Ta-ro-au, the /• being nusul.

ASSAM AND SOFTTI-EASTERN TIBET.

365

following, but it consisted mostly of small children. So it was late before wo got away, since to readjust the loads to my diminutive porters’ capacity took some time and not a little tact. On December 9, Tasbaliin was reached after a weary three days’ trek up and down the dry stony beds of streams, the only paths in these parts. For the ^lishinis, who come down to visit Sadiya only during the w-inter months, the beds of rocky streams, wdiich are' dry at that lime of the year, form cfm- venient paths. But for the white man wearing nailed shoes it is a tiring game. The country from 4 miles north of Sadiya and from Sunpura to the foot of the hills is covered with dense forest without a sign of human habitation. Hero and there, burietl in the growth of centmies, are to lie found traces of a kingdom of prehistoric times, fiait, in bis ‘ History of .Assam,’ gives legends of these ruins dating back to the time of Krishna, but tells us that nothing is really known alsmt Sadij-a prior to the beginning of the thirteenth century, when the Chutiya dynasty held sw’ay. The Chutij’as sacrificed human beings as a part of their religio*, and this method of propitiating the household spirit is still in vogue amongst the Bangpangs living in the hills between Assam and the Hukawng valley.

Tiishalun was found to consist of one house surrounded by a number of granaries hidden in dense jungle. In these parts a few houses, or even one, make a village. The explanation is that (piite a largo com¬ munity collects under the one roof and really represents the inhabitants of several houses of the ordinary hill type, a household consisting of wives, brothers, brothers’ wives, sons, and their wives, etc., etc.

On the 10th I was ready for a forward move, but the elders of the place arrived to say that the porters were all crying and unable to carry loads that day owing to the terrible hardships they had suffered on the road from Samkha ! Of course this report -was nonsense, and only a pre¬ lude to a suggestion of higher wages. Having satisfied them on this point with a promise of one rupee per porter as far as the Tidding river, we arranged to start the next day without fail. The rest of the 10th w'as occupied in climbing to the crest of the range overlooking the plains, and from a height of 47u0 feet a magnificent view was obtained of the Lohit where it debouches on the plains and divides up into a multitude of channels. The water looked a deep blue, and distance rounded off the edges of the numerous islands, giving the whole the appearance of a toy archipelago. Turning in the opposite direction, I got my first peep into the country of the Taroan — a rugged mass of hills, the nearer peaks here and there covered with snow, and at the back of all a great wall of snow-clad ranges. On my return to Tashalun that evening, the village elders came to tea and stayed until it was time for rum.

On December 11 we got off safely. Alsnit half my porters consiste*! of women, who are more used to heavy loads than the men. They all carried a fair 40-lb. load, but I found them verj' slow. They were a

1

3G6

THE LOHIT-BRAHMAPUTRA BETWEEN

cheery, merrj’ crew, however, and I liad not the heart to be very angry when only a portion of my tent arrived in camp that night.

^ly party coneisted of my cook — a Iluddhist Magh from Chittagong — and a chaprasi, an Angami Naga. I had to confine myself to these two owing to diCBculties in obtaining more porters. The cook, however, was not long before he had enlisted one of the local gentry to assist him in the “ kitchen,” Chowna Gohain, a Khamti chief with great influence in these parts, and a son of Chowsam Gohain, who assisted Cooper in ’69, accompanied me, and with him came two of his Ehamtis. In addition to the transport difficulty, there was that of food. Rice, I was told, was not available in large quantities, and the fewer mouths to fill the l>ettor. Cur march on the 11th took us over the first range facing the plains at an altitude of 4000 feet, and we camped that night at the Aharo stream (4200 feet), looking down into the valley of the Tidding. Next day we descended to a village called Salumgum (2300 feet), where I was told I must stoj* the night, as fresh porters would have to be arranged for for the next stage, though I had hoped to have got as far«ns the Tidding river that day. However, there was nothing else for it, and as it commenced to rain later in the day, it was perhaps just as well. The people evinced the greatest curiosity as to the object and direction of my journey, and were firmly convinced that 1 was making for Rima. As news of any event out of the common travels fast in these hills, the anthoi ities at Rima must have heard about my approach to their Iwrder long before I reached it, and had they intended to offer objections to my crossing into their territory, I should have seen or heard something to that effect when at my farthest camp — Sati. But the Tibetans made no sign, and I am pretty sure my journey might have extended to Rima without difficulty or objection. This may not be the case again (unless we find China in possession), as possibly in another few years the effects of the Lhasa expedition will have worn off. But as the orders against British subjects entering Tibet are strict, 1 had to turn back short of Tibetan territory. The 13th found me scrambling down the remainder of the range to the Tidding river, which was crossed (altitude 1200 feet) a mile above its junction with the Ijohit. The crossing was made by a bamboo trestle-bridge, which also served as a fish-trap. The streams running into the Lohit are of no depth in the dry winter months, but the current is too strong to admit of their being fordeil. So a temporary bridge is run across on bamboo supports, held in place by iMjulders, the footway consisting of a single bamboo with a very shaky rail on one side. Below the footway, in the stream at the }>oint where the current is strongest, several cone-shaped baskets are fixed to the supports, and all fish swept in are verj' soon drowned. In addition to this temporary bridge every village has its rojsj bridge, many of them 300 feet in length for use when the rivers are swollen. They are of the single-rope typo, and all goes well as far as the bottom of the sag, but from there onwards

ASSAM AND SOUTH-EASTERN TIBET. 307

hard work with hands and feet is requiretl to haul one’s self up the other side. One variety consists of several canes laid together, with a hoop of cane as a runner, through which the traveller passes his shoulders, and, with his legs crossed over the canes, shoots off liead first. Tlio other is constructed of a fibrous creeper twisted into a cable, which is both soft and smooth. On this is a wooden runner roundeil to fit the cable, and to it several ropes are attached. The passenger ties himself with these

FIRST VIEW OF UPPER LOIIIT.

ropes below the runner in a sitting position, and away he goes down to the centre, from where he pulls himself up the other incline.

On Decemlier 14 we climbed a small spur and got into the valley of the upj)er Lohit, at which I got my first peep from a height of several hundred feet. From here on we practically had no climbing. Occasionally we had a slight rise of a couple of hundred feet or so to get over a spur, but beyond that I was surprised to find how easy the country was. In places we came on long flat shelves along and aliove the river, which only wanted the undergrowth cleared away to become

..Ok

THE LOniT.ERAnMA?rTRA BETWEEN

3r,8

an excellent level road. Rut oh 1 that undergrowth ! It was a mass of thick reed, with a tunnel through it which the people called a path . On December 14 I reached the Tellua river, and on the 15th the path took us close to the edge of the Lohit and past a charming sandy bay called Narra. A rope bridge over the Lohit was near at hand, and the offer of a rupee soon obtained a volunteer to give an exhibition of how a crossing was made. This took him three and a quarter minutes. We then turned up the left bank of the L^m, and after a short climb came out on a level piece of country surrounded by hills, where was Kupa (2000 feet), a village of three houses. The headman Kuranu on first acquaintance seemed a bit sulky, but wo soon became excellent friends, and next morning ho had arranged for porters to tike mo on to Sameling.

From Kupa the path returned to the Lohit, and we crossed the Delli river at its junction with the Lohit by the ordinar}’ typo of rickety trestle bridge. A short distance brought us to a flat piece of ground below Sameling (1050 feet), where we decided to camp. That night we found that we had selected the spot to which the village methan were accustomed to return of an evening to sleep, or rather where they did not sleep ! I wish they had, instead of wandering round during the night poking their noses into everj’thing. Here Vichy, my chaprasi, who had started fever, became much worse; and on the 17th, as no porters had appeared by 0.30, we halted to give him a chance of recovery. This halt gave an opportunity of climbing out of the valley to look at the surrounding country, of which no view was obtainable near the river, except of an occasional peak, the hills on each side Wing so steep that they shut out every thing. A friendly Taroah took mo up a hill called Rirakhu,on which had stood Kaisha’s village, destroyed by Fiden in 1855 for the murders the previous year of the French missionaries, Krick and Boury. The site has never been reoccupied. Rirakhu, having lately been cultivated, was fairly clear of trees, and a glorious view was obtained. As one plodded up the steep path the headwaters of the Delli first came into view, and then snow and more snow and more snow. I looke<l forward to a grand panorama at the top, but at an altitude of 5200 feet the mist swept up from the valleys and all was blotted out. After a time it cleared sufficiently to show me the hills and give mo a view of the country in the neighbourhood, but nothing of the great ranges stretching from the north-west round to the north-east. On return to camp that evening I found Vichy worse. The local nobleman who had consented to help the cook in return for his food plus pay had com plained in the morning of a stomach ache, the result, probably, of a too liberal diet at my expense. I had dosed him with strong ginger, but evening found him still in pain, and he refused to take more. He expected one dose to effect a magic cure, and if it did not he considered the medicine was not worth taking a second time. An evil spirit had

ASSAM AND SOUTn-EASTERN TIBET.

360

got hold of him, he confided to me, and the only sure remedy was to pro¬ pitiate it hy sacrificing a white fowl. To do so with full ceremonial three whole days apart from the world were necessary, and therefore ho had to tender his resignation, lie came to see me a few weeks ago in Sadiya, and was much hurt at my mildly suggesting that “ the evil one ” had l)een a surfeit of pork. The custom of shutting one’s self off from contact with one’s fellows is common aH along this frontier, on such occasions as a birth, a death, sickness, or some other unusual occurrence. It is called doing “genna,” and takes place whenever the slightest ground can be found for it. There is a story of a certain well-known frontier officer being refused admittance to a house where usually he was an honoured guest. On asking the reason, he was told that the household was doing “ genua,” as the family bitch had just ]>roduced pups. On the 18th I again halted for Vichy.

The next day, as he was no better,

I left him l)ehind with money and medicine in the house of a head¬ man, who promised to look afttT him, and so lost one of my only- two servants. ( >n the 19th, porters liaving arrived, we marched to I’angum, crossing the Du river, which is smaller than the Delli, but too broad to l)e forded in com¬ fort. The Du is the l)oundary l)etween the Taroan .and Miju tribes, and Pangum is the first of the Miju villages. It consisted of some seven houses, the ruler l)eing Dagre.sson, a headman of great influence in these parts, a sturdy old gentleman with a pleasant face.

On the road between Sameling and I’angum I p.assed a cane bridge over the liohit, where I found a large party of Mijus waiting to cross, as all the cane hoops had been taken to the other side by previous travellers. < )n the 20th and 2l8t 1 remained at I’angum, while Dagresson made arrangements to get me ]>orters who would go through to the end of my journey and return here with me. I arranged to give each man Rs.lo for the trip. Through porters are a great advan¬ tage, as the delay in collecting others daily at each stage is thus avoided.

HALLl RIVER FROM SOCTII BANK OF I.OHIT.

370

THE lohit-hrahmaputra between

One day at Pangmn was occupied in climbing a hill near the village, from which I was able to fix my position. No view, however, was obtained, as the forest was thick and the hills round me too high. The altitude of the village was found by ll.P. to be 2131 feet, and that of the winter level of Ix)hit ImjIow the village, 1756 feet.

At Pangum I found a couple of Tibetans, traders from Rima, barter¬ ing Chinese opium for Mishmi teeta (Coptis teeta), from the roots of which a decoction is made valued in Tibet and India as a tonic and febrifuge. The opium is soft stuft' from which the moisture had not l)een properly extracted, unlike the hard Indian article, which is much preferred by these tribes. All villages I passed through had })atches of poppy, but only in very small areas. The cultivation of the jmppy gives the owner much trouble. The field has to l»e fenced in and con¬ stantly watched to guard against the depredations of deer and village cattle, who are particularly partial to the plant. The drug is commonly used by the hill people, Dagresson being an exception to the rule. He was a great trader, and he told me one day, “ No, I don’t take opium, because it is an article with which I trade. Were I to eat opium, 1 should be eating up my profits.”

The want of small coin gave me some trouble during my journey, as I had not brought enough change. A porter for an ordinary march received 8 annas pay, and when one had to give a rupee between two of them it caused a lot of explanation and worry, not only to one’s self, but also to the porters. There is little money in these hills, and a Mishmi remarked to me one day that the only way to divide a ruiiee l)etwecn two ^leople was to cut it in half.

Dagresson, shortly after my arrival at his village, produced a couple of letters addressed to “ Mr. Nicholl, Rima.” Mr. Nicholl was a traveller who was exjiected to arrive at Sadiya from Rima in 1904, and these letters had been sent up by my predecessor at Sadiya through Dagresson for delivery. Mr. Nicholl had never got as far as Rima, and the letters had been carefully preserved for three years. AV'hile at Pangum, Dagresson was particularly anxious that I should go as far as Rima, but on my telling him that it was impossible, we agreed that I should make my way to Walung, a village he described as belonging to himself, though inhabited by Tibetans and close to the Rima border. Tibetans, he told me, resorteil in large numbers to the headwaters of the Du river to trade, and owing to the isolated position of the Rima province, I am inclined to think that that portion of Tibet relies to a great extent on Assam for imports through the hill jieople, who are never likely to give us trouble, lest w'e close our marts to them.

We left Pangum on the 22nd, and reached Tila (2600 feet), following the Lobit. As Dagresson had not been able to get the full number of porters, it was a question of leaving behind the theodolite and plane- table or my tent. In the end the tent was left with one of the many

ASSAM AND SOUTH-EASTERN TIBET.

371

n

I

Mrs. DagroBsons, who was selected custodian as she was the prond possessor of a padlock on one of her granaries. The mention of the theodolite reminds me that on firet entering the hills 1 explained that it was utilized for keeping my watch correct to time. Some of these people liad a glimmering what a watch was, one of them describing it to his fellows as a device which told noon-time even in the height of the rains when the sun could not he seen Cor days. This explanation allayed any superstitious fear which may have existed in tlieir minds, and after that I was frequently aske<l to set iip the “ time maker.”

At Tila I tot'k up my quarters for the night in a Miju house. Things

. 'I )
X	li
	?

MIJCS AND TIBETAN TRADER AT SATI.

were not made pleasanter hy the cook having to use the fireplace along¬ side my bed as a kitchen. Culinary operations always cause great interest amongst savages, and consequently the smell of Miju combined with the smoke from the kitchen fire was a little overpowering.

The Ilalli river (2075 feet) was crossed on the 23rd near its junction with the Lohit, and we then suddenly entered a country covered with pines. From here onwards the rapids of the Lohit l>ecame more frequent and resembled small falls. Tho upper Lohit is nowhere navigable. In fact, it is possible only with great difficulty to get small “dug-outs” as far as the Brahmakund. The night of tho 23rd wo s|>ent in the forest near the Sa stream (2580 feet), and again out in the ojien on the 24th at the Ma Ti (stream), as there were no villages at

372

THE LOniT-BRAHMAPUTRA BETWEEN

hand (2580 feet). At the Ma Ti a crowd of people came from Chang-gn, on tho other side of the Lohit, to see me, l)ringing presents of fowls and eggs. Chang-gn is the village from which (’ooper turned hack to Assam in 1870. As it was Christmas Eve and cleanliness is next to godliness, I had to cast conventionality to the winds and l)athc in front of my bonhre surrounded by an admiring circle. After leaving tho llalli river no large streams like the Delli and the Du are met with on the right bank. From here the largo streams which feed the Lohit flow in from tho left bank.

On Christmas Day we reached Warning (28.50 feet ), a village with one house, which was found empty on arrival except for an idiot child. The owners were considerably surprised when they came in later to find us in occupation and making ourselves quite at home. The road ran in many places over tho boulders in the Lohit bed, and tho going was extremely bad. Tho boulders were enormous, and a slip down iKJtween them would h.avo me^int a broken log. Halfway to Warning w'e got a view up-stream of a long stretch of the Lohit, with the rjhaliim flowing into it in nearly tho same line. From a distance they looked one river; a closer vieiv, however, showed the Ghalum to Ihj less than hiilf the size of the I^ohit. At the junction of tho two rivers, which wi* passed just liefore reaching Warning, the Lohit makes a Ix'nd of close on a right angle and narrows from 100 to 00 yards. Tho river is then a constant succession of rapids and small falls. Ih'cemlier 26 found ns making for Sati, tho path as usual running over level tiers ivell .alxive the river, varied by occasional drops to those awful Ixmlders. 'J’he scenery was verj’ grand and wild— great steep spurs studded ivith pines falling to the Lohit on cither bank, and a foaming roaring m.ass of water cutting through the centre of the picture. Sati was not reached as expected on the 26th, as a halt for tho night was made at Dagresson’s suggestion in the foro.st near tho Rlang Ti ('.‘1020 feet;. Sati could have Isjen reached that day, but as an influential Mij'i lived there, etiquette demanded that my arrival should bo announced with due ceremony ; a sudden appearance might have clouded tho political horizon! So Dagresson went on, and I followed ‘him next morning, arriving at Sati in half an hour. After all he need not have been so particular, as we found the local king, Maiyuonson, aw’ay on a trading trip ; one of his wives, however, welcomed mo and took mo in. f'howna, my Khamti friend, shareil with mo tho front room, in which as usual there were two firepbaces. He used to light his fire and have a cheerful blaze ; I didn't ! The result was that his warm side of tho room was .alw.ays crowded, and my cold end left severely .alone, for which I was very thankful at times.

f)n arrival at Sati, Dagresson again urgeil my going on to Kima, saying there was mtthing to prevent my doing so. Much as I should have liked to, it w.as under the present orders impossible. Besides,

ASSAM AND SOUTH-EASTERN TIBET.

3T3

except fur purely seutiinentul rcusouH, there wan nut much to he gained.

I had learnt what I wanted tu — the attitude of the hill {tcople and the nature of their country — and no had to refuse the tempter Dagiesson. It had been my intention to go as far as Walung, but Dagresson had to Uilmit that though he had great influence there, still the people paid revenue to Tibet ; so that village was also barred to me. I could sec the old gentleman was very disappointed at my decision. However, ho packed off his son Tungno to Walung shortly after our arrival at Siati, with orders to kill a cow of his there, and come back sharp with Ijcef and rice with which he wished to feast the party. Tungno returned next day, and the beef was excellent ; not that I had anything to complain of in the way of scarcity of meat, us these people go in largely for capons, and a fine bird can bo easily got for one or two rupees. Walung is one march from Sati, and it is two more marches on to Kima — the last being a short one. With Tungno came three Tibetans from Walung, who presented me with some eggs ami a fowl. They were wild-looking men and very dirty, more so than the Mijus. The authori¬ tative manner in which Dagre.<son talked to them struck me particularly, and I fancy the Kima })eoplu look with a ccrtiiin amount of awe on the inhabitants of these hills. I remained at .Sati during December 28 and 2'.i, and fixed my position with theotlolite and plane-table. Dagres- sou, who was doing the honours of his country, seemed to think a return journey to Pangum by the same route a waste of time. “ If you won't go to Kima, and have only come to see the Miju country, see us much of it as you can. C'ro.'s the Lohit here and travel back to Pangum along the other bank,” was his suggestion, which suited me in every way.

While at Sati 1 climl>cd a i>cak to the west of the village, and at an elevation of (jOoo feet had a fine view to the north-east and cast, only sj)oiled by a ha/.e through which one couhl see no great distance. The great snowy range Pj the east, commencing north of the Khamti Long, stretches away in about long. 97’ to the north, and curves round towards the Lohit and Kima, being in jdaces 15 to 2<) miles distant, and forming the divide between the Lohit and the Ira wadi. A Miju of Sati, who came up the hill with me, explained that due east on the other side of this range the Kunung river to«jk its rise, and, running south, entered a country “ where Sahibs live.” Ily this he meant ISurma, and was referring to the N’Mai Kha. “Kunung” is the name by which a tribe occupying the hills to the north-east of the Khamti Long is known. Another tribe, known to the Khamtis as Kinung, has its habitat between the Kunung country and China. The Khamtis report the Kunungs to bj decent, <juiet folk, but speak of the Kinungs as a warlike, turbulent people who dress like Tibetans. At Sati I met another Tilx-taii trader from Kima. Ho seemed to think I was on my way there, but appeared (juitc indifferent. The iKjsition of Sati was fouud to be lat. 28’ 1' 3" and long. 90° 54' 2 ’.

I

374

THE LOHIT-BRAHMAPUTRA BETWEEN

On December ‘J'J I commenced my return journey, crosaing on a bamboo raft a deep pool of the Lohit where there was little current. The river here was about 60 yards wide. At my crossing-point at low-water level a B.P. observation for the height of the Lohit gave 3186 feet. A-K gives the elevation of the Lohit at llima as 4650 feet, a drop to Sati of about 1500 feet. This is a good deal for the distance, but, as 1 have already said, the Lohit here descends very rapidly. Another B.P. observation taken at Shirong gave 2688 feet, a fall of 500 feet in 9 miles. The difference between flood and low-water level is 35 to 40 feet ; on the Dihohg (Tsanpo) the difference is 55 feet some 10 miles before it enters Assam.

Our camp on the 30th was amongst charming surroundings. A flat piece of ground covered with dry bracken in a pine forest about 80 feet above the Lohit, and a clear hill stream alongside ; a clean smell in the air, a great bonfire of pine logs — one sighed at the thought of a return to civilization and all its worries — and round me a crowd of Mijus, very savage and very dirty, but withal merry and willing, watching me eat my dinner with the greatest interest. The Miju and other so-calletl Mishmi tribes compare in manners most favourably with the Abors. It would be impossible with the Abor to allow him near one at meals. Out would go his paw to seize anything new to him, and he would bo annoyed if one objected. On the 30th, after leaving Sati, I asked Dagresson to join Chowma and myself at lunch by the wayside, the piece de resistance being a capon presented me by my late hostess, Maiyuonson’s wife. Dagresson declined the capon, but asked for some bread, as he was hungry. It appeared that Maiyuonson was his father- in-law, in whose house he could touch no meat ; nor could he after he had left the house eat any meat which had come from it. My bread had no connection with Maiyuonson, and so he could eat that. A curious custom, and calculated, I should imagine, to discourage a pro¬ longed visit by a Miju to his father-in-law 1

On December 31 we passed through Kraw’, a village belonging to the Lamat clan of Mijus. The Lamats live on the left bank and up the Ghalum, which was crossed by a fishing-weir near its mouth. We halted close by on a level plain of short grass under the Miju village of Shirong. This is about the farthest point reached by Wilcox in 1826, owing to the refusal of the Lamats to let him proceed past the Ghalum. After dark I took observations for latitude with the help of Chowna, who held the lamp. It was bitterly cold, and poor Chowna must have been glad when it was over. He was not very enthusiastic on star-gazing in a wintry wind. The Shirong men were all away at Sadiya buying cattle, but soon after our arrival the ladies of the place appeared with presents of fowls and eggs. It is here that the Lohit makes its great bend from north to west, the latitude being 27“ 53' 5" and longitude 96“ 53' 7", the altitude, as 1 have already said, being 2688 feet. On

ASSAM AND SOUTH-EASTERN TIBET.

375

New Year’s Day we passed along above the Lohit, over a very fair path to the Ija Ti (2400 feet), a river about 30 yards broad. Crossing it, we struck up a steep hillside, and, after rising 1300 feet, found ourselves on a big open plateau between the hills and the Lohit, about three- quarters of a mile broad and 4 miles long, covered with short grass. On it were dotted here and there a number of houses, each surrounded by their own granaries and bamboo clumps. All w'as so different from what one had been accustomed to. Instead of being close to the Lohit surrounded by dense forest, we were up high on rolling downs well away from the river. The plateau generally, as well as the village, is

THE LOHIT AFTER IT MAKES ITS GREAT TURN FROM SOUTH TO WEST,

called Chang-gu, the headquarters of the clan called Malu by the Taroan and Peri by the Mijus. It was from here Cooper turned back in the seventies, 'I'he story told me at Chang-gu was that the Tibetans, hearing of Cooper’s intended visit to Bima, sent w'ord to the Lamats and Malus not to let him through. The Tibetans gave out that if anything hapjiened to Cooiier, these clans would bo held responsible, Tliis frightened the Malus in case of any misfortune befalling Cooper, and they objected to his going forward. So he had to turn back. It is (juite possible the Tibetans did honestly think that harm might come to him amongst the Mijus, and that they themselves might be held to blame. There is little doubt, I think, that the Bima people did have.

1

376

THE LOHIT-BRAHMAPUTRA BETWEEN

autl probably have to this day, a dread of the wild Miju. Take the case of A-K. The Tibetans had no idea who ho was, and appear to have acted solely with a friendly object w’hen they warned him about the perils of the route between Uiina and Assam. The head Tibetan respon¬ sible for Cooper’s return sent him a present, I was told, of a Tibetan mastiff and a sword, w'ith the message that as ho had now seen Tibetan specimens of both, there was no necessity for him to proceed as far as Kima itself. Chang-gu (3700 feet) is marked on maps as Prim, which name nobody recognized. It may have been a mistake for Pori, the Miju name of the clan.

At Chang-gu I saw an animal much like a highland bull. It had Iteen purchased at Rima, and had a thick, shaggy black coat, long slender horns curving forw'ard, then turning back, with sharp points, a tail thick and short, with tail-hairs about 18 inches long; height at the withers (which were not high) 12.1 hands. The Mijus called the animal chitla, the Tibetan name lieing dzu. It is a different species from the ordinary Tibetan cattle called mun-tsu by the Miju.s, and lor by the Tibetans. The price paid for the animal had been three Assam silk cloths, or about 30 rupees. These cloths and musk-pods arc the articles principally taken by the Mijus to Rima for barter, 10 rupees a tolah being calculated as the value of the latter.

In Chang-gu I came across a couple of Tibetan traders. They came one day to the house I was putting up in, solemnly sat down close to where I was writing, and after gazing on me for half an hour, got up and w’alked away without a word. I trust I met with their approval ! I remained at Chang-gu on January 2 and 3, 1908, and climbed a hill (5300 feet) to the south-east. The Chang-gu plateau lay at my feet to the north-west, and to the south I looked down a sheer drop of nearly 4000 feet into the La Ti. The haze was very thick that day, and I could only get occasional glimpses of the great snow-barrier between the Lohit valley and the Khamti Long to the south and south-east.

The Mijus bury their dead in a suitable spot near the house. A child is buried at once, but the corpse of an adult is kept in the house from two to four days. 'I'he length of time intervening between death and burial depends on the wealth of the deceased — the wealthier the man, the longer the period. The reason given me was, “ A rich man has a bigger house and more property in it than one poorer. It is a greater wrench for the spirit of the rich man to part for ever from all his belongings, and consequently we let his body remain in the house us long as possible to please the dead.” A month or so after burial, when the deceased’s relations have had time to prepare for it, a feast is given to the village. The grave is then opened, and the remains are burnt, the ashes being left untouched.

My stay at Chang-gu came to an end on the 4th, when I continued my homeward journey, reaching the Cam stream that day. Eu route

ASSAM AND SOrTII-EASTERN TIHET

we mot a party of Mijiis, carrying groat loads of pork. They proved to ho a bridegroom and his friends ofi' to Chiing-gn to fetch the bride, and were taking the wherewithal to feast her relations. Dagresson failed to arrive that night in camp, and his son Tnngno reported that papa was sleeping peacefully on the road, having been overcome by the many stirrup-cnps he had drunk with the local kings of the f'hang-gii plateau. He arrived next morning, looking bedraggled and bleary -eyed, but a drink of hot tea made him look more himself. Tea I foun<l very much appreciated by all these people, and I am not sure that they ilo not prefer it to rum. The tea, however, they get from the shops on the

TAROAS.

frontier is vile stuff and expensive. On my showing the Til)etans at Pangum some of my tea, I was told that only the wealthy classes at Rima drank it in that form.

The cold at our camp of the dth was very bitter Next morning at dawn the thermometer registered 20” Fahr., and it must have been a good deal lower during the night. The whole of this valley in the winter has a damp cold temperature, and this is especially the case on the left bank, where, owing to its low declination in I)eceml>er and January, the sun is shut out from many places by the height of the hills to the south. All along the banks pine trees lie rotting. My suggestion

No. IV.— OCTOBKR, lOOO.] 2 11

i

i i

378

THE EOUlT.BRAIIMArrTRA BETWEEN

to float down logs to Sailiya, whore there was a market for them, met with all sorts of objections as to its impossibility. The hill savage is very wanting in initiative, and the idea will not catch on until some outsider leads the way. On January 6 I recrossod the Lohit to Pangum, the bamboo raft being worked by one of the Tibetans whom I had seen there in December.

The spectators at meals solve<l a matter one day which had l>een puzzling them for some time. “ What is that brown stuff the sahib breaks up and drinks with hot water!” One brilliant mind at last declared it to be lanimj (opium).* Cadbury’s chcKjolate accused of l)eing a narcotic 1

While at Pangum I measured Dagresson’s house, which was 252 feet long and 18 feet broad. It contained twelve rooms, three having two fireplaces, the rest only one. In addition, there was the front common room with the usual two fires. Numerous doors down one side of the house, and on the other a narrow verandah with exits on to it from each room. The house is raised off the ground about 3 feet, and is made of bamboo with timber sup})orts and a thatched roof. This house is of the ordinary type, but of course few are as largo. On my asking Dagresson how many people lived with him, he replied that he really did not know, but would count them up for my benefit. So a long piece of stick was selected, a place on the ground in front of mo swept clean ; down squatted Dagresson, and after many false starts the census began. The sons living with him headed the list, and after he had muttered each one’s name a piece was broken off the stick and laid on the ground : these totalled fourteen. The same process for his daughters, a row of nine representing them, and six more for his wives. He had had many others, he explained quite cheerily, but six only remained to him then. A row of thirteen pieces gave the number of his male slaves, and ten the female ones. Hero the stick gave out, and a fresh one had to bo called for. His sons’ wives came to thirteen— a total with himself of si.xty-six. But nothing had been said about his grandchildren, and on my mentioning them the old man groaned with weariness. However, more space was swept clean, and he proceeded to tackle the sum. He made out there w'ero thirteen living with him, but thought there were more whom he could not remember. We had to stop here, as the unaccustomed mental exertion was telling on the old gentleman. How¬ ever, he had accounted for seventy-nine persons in his owm house. Leaving out the twenty-three slaves, and fifteen for his wives and daughters-in-law, there were forty ]>eople under his own roof directly desceuded from himself. Also ho had other sons with their offspring living in se^wirate houses. Of course this case is an exceptional one, but generally tribes on this frontier are verj' prolific ; even among those who

corruption of Ifeon/, the Assamese word for opium.

ASSAM AND SOUTH-EASTEr.N TIEET.

1

;i79

aro monogaraoiiH it is (initc common for a woman to l)oar lior lius1>aml ton children. Tho infant mortality from exposure, h<jwever, is very great — 70 per cent, at the least ; and it is an example of the survival of tho fittest, from tho survivors springing a hardy race of mountaineers.

On January 9 I left I’angum, after an aifoctionate farewell from l)agres80n,Tungno his son, and all their wives, lie came in to see me at Sadiya in March, 190H, bringing with him a Tibetan from liima, but, 1 regret to say, died shortly after his return home, llis influence rendered the portion of my journey east of his village nearly free of all transport worries, usually one’s greatest difficulty, and his place will Ikj hard to fill. Tungno his son has succeeded, but unfortunately ho has taken to opium, and can never be the man his father was. This reminds me that a report has come down in tho last month that tho sale at Rima of Chinese opium to tho Mijus has lately Isjen put a stop to. I hope this is true, as the habit is doing untold harm in this little corner of the globe.

Cn January 12 we reached the Tidding river. Instead of crossing we followed it up as far as I’ariling. From here wo moved to Teronlung (1970 feet), avillage on the right bank, and on tho 14th crossed tho outer range at an elevation of 5922 feet (B.P. ), following tho path by which the hill people bring up cattle from Assam. A w’ell-graded bridle¬ path could l>e constructed over this range without difficulty. That night we camped at the Tiju river (1550 feet) amongst the low foothills, and next day our party was back in the plains.

In appeariinco and dress there is nothing to distinguish the Miju from the Taroan. Each speak a dialect which is understood by tho other. They intermarry, and both tribes are polygamous, the only limit to the number of wives Ijeing the length of tho purse. Each tribe is divided into clans, which aro exogamous, and marriage is hotw’een adults. Though living on tho borders of Tibet, no trace of Buddhism is found among them. Their religion is animistic, and consists in the propitiation of the various spirits to whom sickness, failure of crops, and suchlike calamities aro attributed. The propitiation takes the form usually of tho sacrifice of a fowl or a pig, a small portion being set aside for the spirit, the rest going down the throats of the offerer and his family.

Travelling amongst the numerous tribes which occupy the hills to the east and north of Assam, one is struck by the strange* similarity which is sometimes met with between people divided from one another by great distances, and l>etween whom there never can have been any communication in their present locations. The Miju and Taroan tie their hair on the top of their heads like some of the Lushei clans. Among tho “ crop-haired ” Mishmis we find the men and women cutting their hair very much like the Angami men, and, more curious still, tho women wearing the siime broad black cane garter below the knee w’hich one meets with amongst the Angami men. The earring of the Taroan

2 I) 2

I

380

THE LOHIT-BRAHMAPFTRA BETWEEN

and Miju women is identical with the large ring of thin brass wire which the Eastern Angami women pass through their ears. T.ut strangest of all is the likeness between the Abors and the Ao, Lhota, Serna, and Trans-Dikhu tribes of Nagas, separated from one another by the whole breadth of the Assam valley and tho Biahmaputra. The Abors male and female cut their hair in the round fashion peculiar to these Nagas, and the curious tattoo-markings on the legs and faces of the Abor, Ao, and Trans-Dikhu women are extraordinarily similar, though wo find no trace of such marking amongst the tribes occupying the 200 miles of intervening hills. I doubt if many corners of the globe can com|mre with the region between the Brahmaputra and Ira wadi systems in respect of the numl>er of tribes speaking dialects so totallj' different that no two tribes can understand one another, and yet they spring (except the K basis and Khamtis) from tho same stock. Their languages are classified by Grierson as Tilieto-Burman, and a comparison of common words, as^^re, trnod, etc., and of the numerals of such widely separate<l people as Abors, Aos, Lusheis, and Eastern 'I'ibetans, shows a marked resemblance between the various languages.

As already said, in appearance, dress, etc., there is practically no difference between the Taroan and Miju tribes, and, unless otherwise sjtccified, any description applies to the two tribes equally. Tho men let their hair grow long, and tie it in a knot on the top of tho head. They wear a small apron in front, a sleeveless coat of a very dark blue, usually ornamented with rod thread, oiien in front, and reaching well down the thigh ; a cloth 3 feet wide and 5 feet in length, worn during tlic day like a Highlander’s plaid, at night serving the purjiose of a blanket ; slung over one shoulder, a largo bag, usually of bearskin, hangs in front to tho waist, and over tho shoulder a strap supjiorts a long knife. Sometimes a piece of coloured cloth twisted round tho hair knot completes the attire. Ornaments consist of long cylindrical silver earrings with bell-mouthed ends ; some quarter-rupee pieces sown on a leather strap as a necklace, and a silver charm l)ox of Tilsjtan manufacture, studded with turquoise, on the chest. The women wear a short little sleeveless jacket covering the breasts, of the same colour as tho men’s coats ; a short under-cloth as a jHstticoat, and a longer one reaching to below the knee, which is used as a covering at night. They tie Iheir hair behind, and above the forehead wear a thin band of silver fastened under the hair at tho back. They wear earrings like those of the men, and some in addition pass a largo ring of thin brass or silver through the top of the ear, which is held up in place by a cord across the head. Long strings of ghvss l>eads, amber in colour, depend from tho shoulders; a number of thin silver rings hang round tho neck with a silver charm box like that of the men. All of them, men, women, and children, are groat smokers, using home-grown sun-dried hdtacco in pijMis. The pipes, al)out 10 to 12 inches long, are of various

ASSAM AND SOUTH-EASTERN TIBET.

381

kiudu, some with silver buwls and mouthpieces, others plain bamboo roots.

Those people usually build their villages low down close to the rivers. The want of water prevents their occupying the cooler heights, and as they live at peace with one another, there is no necessity for a strategical site on the top of a hill. They may occasionally have a light, but they are not raiders or head hunters like the Naga tribes on this frontier. Even in war heads are not taken. All that tlm victor does, I am told, is to cut off the hair of the slain, which is buried in front of the victor’s house. Living at peace as they do, they are constantly on the move, trading between Tibet and Assam. Agriculture and hunting are their other means of livelihood. They are well to do.

BOCLOERS os THE LOUIT RIVER.

and their villages are plentifully stocked with fowls and pigs. Of cattle I saw few, though they buy quite a number from Tibet and Assam, but quickly slaughter them at festivals. There is no permanent cultivation, but a fresh patch of the hillside is cleared and utilized for a couple of years, after which it is allowed to rest for eight or nine years, w’hen it is again cleared, all growth being burnt and worked into the ground. Rice is not largely grown. The staple crop is maize, which they grind into a coarse flour or boil whole. They make from fermented grain a litpjor of whieh they drink large quantifies, but 1 saw little drunkenness.

Their weapons consist of a single-edged knife about 18 inches long.

3H2

THE EOIIIT-BIIAHMAPUTRA BETWEEN

and a spear. They own crossbows, but the ordinary bow with whicii they move about is a light one of the usual pattern.

Wandering in the wilds, one is apt to wonder what will be the future of the country — whether in another twcntj' j'cars it will bo in the same uncivilized state as it now is, or whether sufficient reason will arise for opening it up. These hills, it must be remembered, are peculiarly situated. They separate two countries, Assam and Tilnjt, from one another ; the distance from the plains of Assam to the border of Tibet is less than 50 miles in a straight line and 110 by road, the whole route lying through a country occupied by a quiet, peaceable people, who at present look to us as the paramount power, and from whom we may expect nothing but obedienefc ; a jKJople amongst whom no trace of Tibetan influence is found. With the awakening of China, how long these conditions will continue is a subject that I cannot touch on here. An effort to take advantage of the present situation should, I think, be w'orthy of consideration from a commercial point of view. To connect India with the borders of south-east Tibet by a good mule-track as a beginning would be easy, could be edrried out at no great cost, and should attract trade. The attention of the Calcutta Chaml>er of Commerce was drawn to a trade route from .Assam to Tibet over forty years ago by Cooper, who went up to investigate. He met with failure, but in those days the hill triltes were unfriendly, which made all the difference. The country itself presents no difficulties ; it is, in fact, a strikingly easy one for a mountainous tract. The highest altitude is met with when crossing the outer (or first) range, and even here we have to face only' an actual rise of 4600 feet, after which no high altitudes olisiruct the way. The banks of the river would api>ear siKJcially formed for a road ; large flat tiers running parallel to the Ijoliit, with easily surmounted spurs extending to the river itself, rising gradually from 1200 feet at the Tidding, to 3100 at Sati, an ascent of It'OO feet in 70 miles. It is a natural highway into Tibet, and only requires the hand of man to render it easy and expeditious.

At present trade is infinitesimal. The imports which pass up to Tibet from Assam through Miju traders amount to little, and of Til)etan exports there are none. Hut would these conditions continue if an eiisy and fairly expeditious route existed? 1 very much doubt it. At present south-eastern Tibet, or the Rung, as the country is known, has no industries, because she has no incentive for the development of her resources. She is cut oft’ from convenient marts on all sides. Thousands of maunds of wo(d are wasted annually simply l^ecause there is no market, and that not only’ wool of the ordinary quality’, but also of the costly variety’ called bashm from which shawls are made. Were communiciitions improved along the natural outlet and the lino of least resistance, viz. the Lohit valley, facilities for exjmrt would be brought within the reach of all. < hice the Tibetan learned that every’ pound of

3

ASSAM AND SOUTH-EASTERN TIBET. ;I8:3

wool had a luarkotablo value in Assam, and that Assam could bo reached quickly, comfortably, and safely, and that there he could purchase tea, clothiug, etc., in return for his wool, commercial interchanges would bo assured, and both countries would benefit to a considerable extent. Trade intercourse just now is impossible, as Tibet is a forbidden land to the trader. Hut a good bridle path from the limit of British territory to Sadiya, a place in close proximity t® the terminus of the L)ibru-Sadiya railway, would attract the Tiljctau to trade with us.

Events have been taking place of late which are likely to increase interest in this section of the Lohit valley. 1 refer to the Sinification of Tibet, and if reports in the public press be true, it is only a matter of months, not years, before the Bong, instead of forming a part of Tibet, will l)ecomo a Chinese province. Assam will then 1)e separated from China by only 50 miles (in an air line) of mountainous eountry, which can be traversed with ease.

ITow this is likely to affect aft'airs it is difficult to say. The Chinese colonist may under orders prevent all intercourse l>etween the “ foreign devils ” and the Tibetans. (Jr if left to his own devices, we may find him keenly alive to the advantages accruing from easy communications with Assam. Sir Thomas Iloldich, in ‘Til»et, the Mysterious,’ draws a picture of the Tsangpo valley with a railway and Gyala Sindong with a hotel. Why not, then, a railway running up the Lohit towards Scchuan ? A railroad as far as Bima is practicable, but from there the country is difficult. The TilaLaand two other high passes have to be surmounted, and the Salween, Mekong, and Yangtse to be crossed — at first sight a formidable array. But then we must remember that the altitudes of these passes do not represent their heights above the surrounding country. 'I'he altitude of the country itself is 5000 feet or more, which substantially reduces the height to be encountered. Again, the rivers are not the enormous rivers wo know them to bo further south, but at the points crosseil are confined within narrow limits with a rock forma¬ tion suitable for bridge foundations. But however costly, were there facilities for quick communication Ijctween India and Western China, the possibilities of commercial expansion would appear to bo lioundless. Given a railway, every ton of our exports for Sechuan w<mhl Ix) captured for this route instead of being carried a long sea voyage from Calcutta, only then to commence the difficult journey up the Yangtse. With such improved communications, the resources of Sechuan, one of the wealthiest provinces of China, would develop enormously ; with an easy and expeditious route, there is no reason why the Chinese coolie should not seek for employment on the tea gardens of Assam, and so possibly solve some of the present labour difficulties. The pros[)cct of forging the link connecting India with China may l>e visionary, but, again to quote Sir 'rhomas Iloldich, “ it is not more visionary than twenty -five years ago was that of a modern hotel at the Victoria falls of tlie Zambezi.”

1

( 3H4 )

THE MONGOLIA-SZE>CHUAN EXPEDITION OF THE IMPERIAL RUSSIAN GEOGRAPHICAL SOCIETY.

By Captain P. E. KOZLOFF.

In tlio summer of 1907, tlio Imperial Kusaian Geograpliical Society, with the approval aud assistance of Ilis Majesty the CVar, organized an expedition to Central Asia, entrusting it to me from its commencement. My com2)anions were the geologist A. A. Chernof, the topograjjher Staff-Captain Napalkof, and S. S. Chetyrkin, botanist and entomologist, besides ten men to jirepare specimens, take meteorological observations, etc., and to act as a guard.

The object of the expedition was to explore parts of Mongolia not yet visited by Europeans, to take soundings of the alpine lake Eoko- nor, and to investigate the region of the upper course of the Huang-ho, where it winds among the mountains.

The expedition started excellently etpiipiied from St. I’etorsburg, and on November 22 left Moscow. Though the year was drawing to a close, my compatriots flocked to the station to see me off — relations, friends, acquaintances, comrades of the Ekateriuoslav Guanls, and our journey commenced to the strains of the Ekaterinoslav March. The railway journey of ten days and a couple of days’ ride in a Siberian carriage soon passed, and wo found ourselves on the borders of the emjure at Kiakhta-Uvazhda, and wore kindly welcomed by old acquaint¬ ances.

lleyond Kiakhta begins Mongolia, at first a steitpe country, but mountainous beyond, with ranges more or less massive, rich in animal and vegetable life. Then comes Urga, the Mongolian Lhasa, beyond whicli the character of the country rapidly changes, the surface becomes more even, the vegetation is poor, the population scanty, especially south of the mountains constituting the eastern prolongation of the Mongolian, or Gobi, Altai. There is the actual desert of Gobi, hero taking the form of even slojxis of sand and stone, there moulded into smootli or rocky hills, on the toj) of which at even the setting sun casts 2>icturcsquo gleams. In fine, southern Mongolian presents almost everywhere a sea of arid sand, but in many parts occur meridional sand- dunes, not infrequently rising to a height of 100 feet or more.

The native inhabitants of Mongolia, the Mongols, also vary with the j)hy8ical conditions and the predominance of their own chiefs or of the Chinese. In Northern Mongolia the traveller meets with nomads still priding themselves on their traditional boldness and courage, their many-coloured garments, their lively hacks, their ornaments, their dexterous horsemanship, tlieir cunning ; these Mongols carry one’s thoughts buck to the time long gone by, when they had a history of their own. The maintenance of the independent condition of the

KOKO NOR

to ilhiKtratt* n papf*r CAPT P K KOZLOFF

Scak- 1 1 OOtl (KH) or 1 hitli • 15 78 StaJ Ms

*kje saruLttnt

^Ktttsit I.

■A}uikulti]

Ser.chim MV»

SrT€mHnfy,r«lirt lak^ strrtjehinq aZon^ th^ bank

THE MONGOLIA-SZE-CHUA\ EXPEDITION. 3H5

Mongols is promoted to a great degree Ly the Bogdo-gegen of Urga and the ruling authorities of the tribes, the Mongolian princes, whom the Chinese treat courteously, striving to attach them to the court of the Bogdykhan (Emperor of China). The Mongols of the central parts of the country are markedly inferior to their northern neighbours.

but, on the other hand, are often superior to the inhabitants of the south, who are becoming ever more and more assimilated to the Chinese, both outwardly and inwardly. Among these people the national life is forgotten, the racial feeling is disappearing. On the whole, judged from our standpoint, the Mongol passes his life in a miserable, un¬ enviable environment, and his mental horizon is limited, though he is

386

THE MONGOLIA-SZE-CHUAN EXPEDITION OF THE

difitiuguishcd from his nomadic neighbours by having attaincil a com¬ paratively higher intellectual development ; has his own literature and printed laws, learns to road and write Tibetan, studies religious questions — has, in fact, some slight conception of culture; but, after all, the nomad, according to Eurof>eau ideas, represents a low, effete stage of social development.

In crossing Mongolia in various directions, and observing and study¬ ing different sides of its simple but original character, I was far from expecting to find a dead, forgotten city, such as the expeilition actually came across in the lower basin of the river Etsin-gol, which rises on the northern slopes of the mighty Nan-shan. The lost, deserted city Ehara- khoto, or llaishen-khoto,* now stands 8 to 10 miles east of the most easterly branch of the Etsin-gol. The “ Black City,” or rather its ruins, were well known to the Etsin-gol Torguts, who migrated thither from Kobuk-saire-Dzungaria about four centuries ago, when the banks of the Etsin-gol were still clothed with virgin forest, impenetrable in parts, which the Torguts burned down during the first three years to make room for their encampments and pasture. According to the present Torguts, their ancestors found the ruins just as they are now — that is, a town of the Chinese type, with high clay walls facing the cardinal jwints, placed on a rather high abrupt terrace, which was at one time washed by the Etsin-gol flowing past to the north-east. The remains of the channel can be traced the whole distance from its head at Botok-deerek to the salt sandy basin Khodon-khoshu, lying in a line with the existing basins Gashun-nor and Sogo-nor. At some distance before reaching Khara-khoto we were much interested in the subnrgans, or tombs, standing singly or in groups of two to five along the road which leads to this historical monument, and overwhelmed with sand from the neighbouring desert. Two miles further the city itself came into view, its principal suhunjan rising above the north-western angle of the fortress, from among a number of smaller neighbours erected by the wall and in a line with the wall outside the fort. The explorer approaching Khara-khoto from the west notices a small building with a dome-8hai>ed roof standing at some distance from the south¬ western corner of the fortress, and somewhat resembling a Mussulman house of j)rayer. The walls of the town are covered with sand, in some places so deeply that it is possible to walk up the slope and enter the fortress. A few minutes more, and we passed into the dead city by the western gate, placed directly opposite the other remaining gateway in the eastern wall. Here we found a quadrangular space,! whereon were scattered high and low, broad and narrow, ruins of buildings with

KImni-khoto an<l Baisbcn-kliuto are the Mongolian names signifying Black City and Fortress City.

t A side of the siiuare measures about a iiuarter of a mile.

IMl’ERIAI, RUSSIAN GEOORArHICAL SOCIETY.

387

rubhibh of all kinds at their feet, including a heap of fragments of clay and china pottery. Here and there stood tombs. Wo at once divined the interest with which our labours would be rewarded in the investi¬ gation and excavation of all that now lay around us.

Our camp was pitched in the middle of the fortress beside the ruins of a large two-storied building, to which adjoined, on the south side, a temple ruined to its foundations. ( >wing to the absence of water on the spot, wo wore obliged to carry with us all our vessels filled with water, that we might remain as long as possiljlo at the ruins. Ilefore an hour had passed, the interior of the fortress had come to life again. Men moved about, totds were at w’ork, dust rose into the aii'. About the bivouac was the desert bird, the chough-thrush {Podoctx HttKlirsum ), chiittering loudly as it perched on the twigs of the saxaul ; it was answered softly by the sweet singer of the desert, the whinchat (^Saxicvlo), Interesting occupation made the time pass quickly and imperceptibly. The didl grey and geucially windy day soon gave place to a calm clear night, in which the ruins seemed stem and gloomy. Being tired wo soon fell a.sleep, but some of us w’ere disturbed by the unpleasant voice of the owl (^Athene) screeching ominously from the top of the principal mhurijan.

During the few days spent at the ruins of Khara-khoto, the exiHjdi- tion gathered all kinds of articles — books, letters, documents, coins and paper money (assignats of the Min dynasty), female ornaments, some household utensils, ordinary articles of trade, objects of the Buddhist cult, etc. The quantity collected was enough to fill ton chests of 36 lbs. each, which have been forwarded to the Geographical Society and the Academy of Sciences. On the return home of the expedition we shall be able to learn positively when the “ Dead City ” existed and who were its inhabitants.* From the slight legendary lore handed down from generation to generation, the Torguts of Etsin-gid who accompanied us to the ruins communicated to us information of no little interest. The native tradition runs as follows ; The last ruler of Khara-khoto, the Itatyr f Khara-tsian-tsiun, relying on his invincible army, designed to hike possession of the throne of China, and consequently the Chinese authorities were obliged to send a considerable force against him. A series of battles t<x)k place between the imperial forces and those of Khara-tsian-tsiun to the east of Khara-khoto, aliout the present northern boundary of Ala-shan in the Shar-tsa mountains, and were unfavourable to the latter. Having got the upper hand, the imirerial army forced its

* ('a]>tain Kozloflf has. since writiuj; tlic alwvc, receive 1 a letter from V. (Jregorief, vice-Prcsideiit of the Kussiun Geographical Society, iuforniing him that the manuscripts ami other articles forwarded to St. retersburg show that Khara-khoto must he identical with llsi-hsia, the capital of a Tangiit kimrdom which nourished from the eleventh to the fourteenth century. — En.

t .\ Mongol word lueaidug a valiant man, a hero.

388 THE MONGOLIA-SZE-CHUAN EXPEDITION OF THE

enemy to retire, and finally to shut himself up iji his last refuge, Khara- baishen, which was blockaded on all sides. Whether the siege was prolonged is not known, but, at any rate, the fortress was not taken at once. The imperial army, being unable to take Khara-khoto by assault, detenuined to deprive the beleaguered city of water, with which object they diverted the river Etsin-gol, which, as mentioned above, at that time washed the wall of the town, to the left, that is westwards, blocking up the original channel with bags of sand. And to this day there are to bo seen remains of this dam, in which the Torguts not long ago found remnants of bags.

The besieged then began to dig a well in the north-west corner of the fortress, and sank it down to 80 citokan — a chokun is more than 11^ feet — and still did not find water. Then the hatyr resolved to meet his enemy in a last general engagement, but, in case of defeat, he buried in the well, before it was filled in, all his treasures, which, it is said, filled no fewer than eighty wagons or carts, each holding 600 to 000 lbs. weight ; and this consisted of silver only, and did not include other valuables. He then killed his two wives, and sou and daughter, that the enemy might not abuse them. These preparations completed, he ordered a breach • to be made in the northern wall, near the place where his wealth was hidden, and through it rushed on the enemy at the head of his troops. In this decisive action he perished, and also his army, hitherto considered invincible. The imiK-rial army, as usual, destroyed the captured city utterly, but did not discover the hidden treasure, which, it is said, lies there still, though the Chinese of the nearest towns and the local Mongols have trietl more than once to get possession of it. They attribute their failure wholly to a sihjII worked by Khara-tsian-tsiun himself. The natives are more convinced of the efficacy of this jwtent sjhjII since the last time they searched, when instead of treasure they found two large snakes brightly glistening with red and green scales.

After its work at the ruins of Khara-khoto, the expedition followed the historic road to the residence of the Alasha-tsin-vau, the town L)yn-iuau-in, and then made somewhat extended excursions into the Ala-shan range and the adjoining valley of the Iloang-ho before pro¬ ceeding to the Nan-shau and further south. In Dyn-iuan-in we estab¬ lished a depot and a meteorological station of the second class, which was in working order from the beginning of May, 19< >8.

Here ended the first j)eriod of the expedition’s activity, and the second lay before it — the Koko-nor period.

On October 1 5, 1 908, the expedition came to the Ilui-dui oasis, where it was accommodated in the Chinese temple, Viu-tsy-miao, Iwautifully

* TUia can still be seen.

1

IMPERIAL RUSSIAN GEOGRAPHICAL SOCIETY.

sitnatod on tho heights lx)unding the oasis on the sonth. Three months had passed since we left Alasha-yamyn, and in this interval wo had visited the burning desert of south Ala-shan, the eastern Nan-shan, and tho high alpine lake Koko-nor — had, in fact, accomplished the second part of our programme. We happened to be among the Ala-shan sands in tho very hottest season, in tho Nan-shan at tho time of tho development of vegetation and animad life, and at Koko-nor under the l)est weather conditions, comparatively speaking.

Tho heart of Koko-nor, its mysterious island Kuisu, was visited. This event made a groat impression, not only on tho Kok(j-nor natives, but also on the Chinese of Sinin — so great that both conceived it possible that we had subdued the dragon of tho lake ; that the water retired before us, and we marched along the bottom, caught a golden fish, etc.

LANDSCAPE IN CKNTnAD^MONGOI.IA :“MONr.OL ENCAMPMENT.

At any rate, it is eertiiin that bfith Chinese and Tanguts began to treat us with greater respect and attention.

Uninterrupti-d rain for the space of a week delayed the departure of tho main caravan from Dyn-iuan-in till July 18.* The noise and bustle of tho town gave place to tho deep silence of tho desert. The air, purified by the rain, had become clear, and it was possible to trace on the one side tho outline of tho Ala-shan range, on the other a sandy plain stretching far away to tho western horizon. Tho route of the expedition crossed the desert at its narrowest part, from l)yn-iuan-in to Sa-yan-tsin, in a south-westerly direction, leaving on the east a wedge of sandy waste that extends to tho undulating heights on the loft bank of the Yellow river. Our caravan consisted exclusively of camels, to tho number of thirty, which travelle<l easily over the cracked sandy ground. At our feet lizards scurried about, liectles crawled everywhere, flies swarmed, buzzing as usual ; butterflies were less

* The eastern detachment, under Captain Nap:iIkof, left the town on .Tuly II ; the western, under the geologint M. Chernof, on July I t.

390

TITK MOXGOLIA-SZE-CHUAX EXPEDITIOX OF THE

common. A great scarcity of l>irtls was remarked, ami still scarcer were the mammals of the desert ; occasionally the profile of a swift¬ footed antelope appeared for a moment on the summit of a hilly ridge or sandy mound. Oppressively monotonous and wearisome is the desert in summer. At night, indeed, one feels pretty comfortable, but as soon as the rays of the summer sun appear the heat becomes unbearable, crushing all one’s energy. Even the famous ships of the desert Itecome heavy and stupid in the very hottest season. Again and again you look at your watch, straining your eyes to catch sight in tlio distance of the green spot that is to afford a resting-place for the caravan l)e.'ido its well. In order to cross the desert with greater ease, especially where there were long stretches without water, we usually left the wells after dinner at two or three o’clock, and travelled till dusk, then unloaded the caravan beside the road, and, after taking some refreshment, imme¬ diately lay down to sleep. As soon as the morning dawned the caravan was ready to start, and tried to reach as s<x)n as possible another halting-place with water.

The first half of the sands we traversed under a burning sun, far from feeling the full “ charm ” of the desert. The sand was heated almost up to loS" Fahr., and burned our feet through the thin soles of our shoes. The poor dogs suffered more severely than ourselves, in spite of the great care and precautions of Sergeant-Major Ivanof, who every half-hour, took down a trough from the baggage, and sparingly watered them. The dogs know all about it, and at the proper time ran, without Ireing summoned, to the front and fawned ujKUi Ivanof.

The sand-dunes (barkham') were piled one on the other, higher and higher. The camels strain at the leading-reins, and breathe heavily as they ascend to the top of a dune and then descend to its base. The broad hoof of the camel falls softly on the sandy surface ; the curious sound it omits is hardlj' heard, being over[towered by the breathings of the troop. On ascending a high sand-dune wo have always the same view — sand, sand, sand. Our mouths are parched ; the dryness of the desert air is extreme. The expedition rested quite twenty-four hours at Shirigin-gol, where in one of the small lakes, or rather jMX)la, wo bathed several times, vainly seeking coolness.

AVe were fortunate in crossing the second half of the desert, the Tengori sands, when the sky was overcast. Dark storm-clouds swept up from the north, rain fell, and the air became fresh and pleasant. In such weather the sands wore no difficulty. The camels left barkhan after l)arkhan l>ehind, and at length passed the sacred obo, erected, according to tradition, by the Iluddhist prelate llangcn-bogdo, who opened this road. In memory of the great pilgrim and his oj)eniug up of the road, the appropriate name of Tengeri, t.c. Celestial, was given tf> this sandy waste. Beyond the sands the road soon passes through hills of argillaceous sbme, and then comes to a mountiiin ridge, a northern

IMPERIAL RUSSIAN G?:()GRAPHICAL SOCIETY. 391

offshoot of the eastern Nan-shaii, crossing the scarcely pcrceptihle remains of the Great Wall at the small decayed town, Sa-yan-tsin, the residence of the Chinese custom-officer. There, for the first time after leaving Dyn-iuan-in, we found good water. A dense agricultural popu¬ lation extends thence right up to the little town Shara-khoto, near the pass of the same name, where the basin of the Koko-nor begins — a country well suited for nomadic life.

Every day we advanced along the eastern Nan-shan the air iMKiame more transparent, the sky bluer, the vegetation more luxuriant and greener. Behind us, to the north, clouds of dust rising a1)Ove the plain reminded us of the desert. From the Chagiyn steppe, from the town

DYN-ICAN-IH (SORTII-WEST ANGLE OF FORTRESS); CARAVAN ON THE MARCH.

Snn-shangen,* we ctuld already admire mighty mountain chains, espe¬ cially to the south-west, where the impetuous mountain stream, the Tetung, hides itself, which was visited more than once by my never-to- be-forgotten instructor, N. ]\I. Prjevalsky. ^lore to the east, along our road, l)cyond the town Pin-fan, rocky chains sloped down, dividing into secondary ridges, which were covered from top to bottom by the tilled fields of Chinese. Not long ago, conjparatively, the land there was waste, a memento of tho terrible Dungan insurrection ; to-day it is almost all occupied ; there is absolutely no free, uncultivated spot (hat is suitable for tillage.

• In this town we stayed in a .small Ituddhist mnnastery, presided over by the amiable ijegen Nian-;;okik, who bad erected a very fine temple at his own exiK*nse. We soon l)ceame intiinuto with this priest, for he anil I had many friends in common — senior lamas of Nan-shan monasteries. At partiiio we exchanged prosimts, tho priest giving me an excellent metal figure of Tsagan-durkhe, the tVhite goddess.

392

THE MONGOLTA-SZE-CHUAN EXPEDITTOX OF THE

The first stream of importance on our route was the Churmyn-gol, the second the Tetung, and the third and last the Sinin-ho ( with the little town Lo-vachen) with its tributaries. Along the valley of the last we tnmed sharply to the west, and, gradually ascending, passed beyond the region of cultivation. Wo crossed the Churmyn, which divides into arms, without much difficulty by a ford at the town Pin-fan, where we found good camping-ground close to the wall. On the larger and more impetuous Tetung there is a ferry-boat, unfortunately with a rather high deck, rendering it troublesome to get the animals on Ixiard and off again. The river Sinin-ho, along which runs a much-frequented trade route, Lan-cbou-fu to Sinin, was spanned by some hunch-backed bridges. The valleys of all these three rivers, like the mountains that confine them, run from north-west to south-east, and bear the same character; their water helps to fill the main drainage artery, the Iloang-ho.

This eastern section of the Nan-shan has a different character to that of the more western districts, where not only the central, but also the lateral, ranges exhibit crystalline outcrops in the form of sharp summits, peaks or great cliffs rising one above the other in picturesque, enchanting grandeur. Here the traveller can, with comparative ease, reach the summits of the passes with a caravan of camels, or in a Chinese cart drawn by horses or mules, for loess predominates everywhere, overlying for the most part Han-hai deposits, Ilan-hai red sandstones, and limestones or siliceous schists occur only in small quantities. The carriage-road is deeply sunk in the layers of loess, and the surface is cut up by trenches very troublesome to travellers, and in some placc.s impossible to avoid.

The towns and inhabitated places in general on the route of the main caravan do not differ essentially from other towns of Western China well known from the re|)orts of former travellers. Here, too, the towns are enclosed in walls of stone or clay, and here are collected officials, traders, soldiers, artisans, etc. Here also the Government buildings and the bazaars are full of natives during the day. It may safely be asserted that Chinese towns have from times long past been constructed, and are still constructed, on one jilan ; having thoroughly examined one or two inhabitetl places, one can form a correct idea of the others. The only variations are the extent of the town quarters and the strength and solidity of the walls.

At the time of our journey in the Nan-shan, at the end of July, the grain harvest had commenced in the valleys and on the lower and middle l»elts of the adjacent mountain slopes. The fields were alive with natives, both men and women taking part in the work. Broad straw hats protected them from the scorching rays of the sun. The villages, as with us at home, were deserted by old and young.

Towards the middle of August the expedition came to Sinin, and

IMPERIAL RUSSIAN GEOGRAPIIirAL SOCIETY.

39.‘{

jiitchoil its camp on the eastern side in the little hamlet Tsav-dia-tsai. The animals were hard up for pasturage ; wo had to put them on dry fixlder and deprive them of their usual liberty of movement.

Sinin * is a large provincial capital with the residenee of the Chinese dignitary who exercises authority, not only over the nomads of Koko- nor, hut also over those of extreme north-eastern Tibet. To get all my business in Sinin done as satisfactorily and quickly as possible, I, with two interj)reter8, moved into the city, where I found Icxlging with the Chinese trading firm Tsian-tai-mao, known to me during my former Tibetan journey. In the course of the three days I sj)ent in Sinin, I was able to settle all questions definitely. The Chinese oBBeials received

TOWS OF SISIS, TAKKS FROM THK TEMPI.K, WHICH STANDS ON THE SOVTH-SOUTH- WEBT OF THE TOWN.

me very pleasantly, and professed themselves ready to assist me in my projected visit to Koko-nor, though they all with one voice declared, “ You know that the times are changed ; that many Tanguts and Tilietans are armed with quick-firing rifles; that these savages collect more frcriuently than formerly into roblier bands, attack caravans, and mercilessly pillage them. Pray do not go to Koko-nor. The Tanguts will not leave you in peace, and unpleasantness will arise which will give us all no end of trouble.” The Tsin-tsai went still further. In the course of a long conversation, he heard from me that we intended

* Founded, according to Chinese nnn.Tls, 22‘2.'i years ago, eonsi'qnently in the year 317 B.C.

No. IV. — OcroRKR, 1909.] 2 k

394

THE MONGOLIA-SZE-CHUAX EXPEDITION OF THE

to navigate Koko-nor, and for this purpose were provided with a folding canvas boat. The Chinese official almost jumped off his chair, and, raising his voice, declared that this was folly, fur the water of Eoko-nor was peculiar, not only stones, but even wood sinking in it. “Try it yourself, and you will see. Besides,” continued the Tsin-tsai, “ at the present time there is no one on the island Khaisin-shan ; * the pilgrims flock to it only in winter.” For my part, I thanked the honourable official for his kind attention, but could not refrain from telling him that I was now more determined than ever to visit Koko-nor, to convince myself of the truth of this singular property of the water of drawing dry wood to the bottom. Thereupon the Tsin-tsai said, “ Very well, I will do everything to facilitate your journey to Eoko-nor. I will write to the tribal chiefs ; I will provide you with an escort ; but I beg you to hand me a paper, stating that you are going in spite of my warning and polite remonstrances, and that you will be personally responsible for any troublesome consequences that may ensue.” Being familiar with written declarations of this kind, I this time also complied with the demand of the obstinate official, who kept his word in supplying me with an interpreter knowing the Chinese, Mongolian, and Tibetan languages, and with four mounted troopers, who soon made friends with my grenadiers and cossaoks.

As regards the information given by the Sinin officials about the greater audacity and activity of the Eoko-nor and other Tibetan robbers, and of their possessing quick-firing rifles, this we ourselves had oppor¬ tunities of learning, having noticed large or small parties of Nan-shan Tanguts bearing across their shoulders German, more frequently Japanese, and occasionally liussian, rifles. Seeing the wild energetic nomads sweeping past on their mettlesome horses, wo reflected that in no distant future these free sons of the steppes of Eoko-nor and Tibet would become formidable to the Chinese Empire. This the Chinese of Sinin recognize, and they recognize their own present helplessness and the absolute necessity of training and arming their troojis before long after the European model.

My affairs in Sinin being settled, I sent off the caravan by the direct road to Dongor, turning myself, with the Cossack Badmazhanof, south¬ wards, with the object of visiting the groat Amdos monastery, a day’s journey from Sinin. Gumbum is hidden among high hills of clay and loess, immediately adjoining a very lofty alpine range, the southern foot of which is washetl by the rapid waters of the Yellow river. The Gumbum monastery was founded five hundred years ago by Bogdo- gegen, who afterwarils made a pilgrimage in Tibet, to Lhasa, where ho fixed his permanent residence. The monastery he founded came

* That is, “New mountain of the B<“a;” so the Chinese name the island Kuisii. The latter word is Mongolian, and signifies “ navel.”

IMPERIAL RUSSIAN OEOORAPHirAL SOCIETY.

under the control of the Acha-gegon, supposed to bo now in his fifth regeneration.

In tlie twelve temples are said to reside sixty-three gegenit, presiding over a confraternity of fully two thousand men. There are four principal temples, which were saved from destruction by the Dungans by the forces of the monastery — young fanatical lamas, who fought splendidly, gun in hand, against the daring foe. The solid ancient temples glitter outwardly with gorgeous gilded roofs, and within are abundantly

THE TSIN-T8AI OP 8ISIN.

decorated with historical images of the best Mongolian, Til)etan, and even Indian workmanship. Especially magnificent and revered is the temple called the Golden Tomb. Before this temple, the worshippers, prostrating themselves and moving their hands and feet from time to time, have worn in the boarded floor of the porch large holes, in which they can easily place the toes and hands when prostrating themselves at full length. The large cathedral temple is also preserved in good condition, and is capable of holding five thousand worshippers. On

2 E 2

390

THE MONGOLIA-SZE-OHUAN EXPEDITIOX OF THE

entering the jwrch I caught sight of seven stout whips liung on the wall. These whips, as I was infornietl hy the local inhabitants and the older lamas, only keep up tho long-established rule of the monastery for novices, Kich and beautiful also is the temple, standing l)eside eight white tombs,* according to tradition on tho spot where is concealed in the earth the placenta of tho child which afterwards became tho famous reformer of lluddhism, Tson-khava.

In Gumbum are carefully preserved relics, such as tho raiment of a Dalai-lama, Rangon-ergen, Tson-kava ; also the carts of the Dalai-lama .and Rangen-orgen, the hat of tho reformer, and tho saddle of a Rogdykhan with the dragon quivering on tho bow.

All tho time I was in Gumbum, tho head of this monastery, tho Acha-gegen, was absent on an extended tour to tho East — Poking and Japan, .\coording to information obtained from tho inhabitants of the neighbourhood, tho Acha-gegen sides with the Ja^mneso. Tho latter call at this monastery from time to time, and on particular occ.a8ions, as, for instance, during tho presence of tho Dalai-lama, stay there for a long time — a year or more — under tho pretence of lc.arning tho Til)ctan language.

I left Gumbum early in tho morning of August 18, and in the even¬ ing of tho same day reached Donger, where an ishind between tho arms of the river on the southern side of tho town was whitened by the tents of the expedition. Tho caravan had arrived only tho day before mj’ return from Gumbum. As the governor of tho town had received timely notice of our exjiedition, wo received nothing but kindness and attention from him. Tho memlKjrs of tho ex|>odition could wander freely through the town, enter at will tho stores and shops, and felt themselves gene- riilly at homo. In this market town the streets were crowded with Koko-nor Tanguts. The young ones compelle«l notice by their bright dress and the singular decorati<m of their backs, consisting of ribl*ons in twos or throes richly trimmc<l wHh coins, shells, turquoise, etc. Still more animation was lent to tho streets by the Tangut horsemen career¬ ing along fully armed. Tho proud and haughty steppe-dwellers fear nothing and no one, but rather, on the contrary, inspire tho Chiiu'sc inhabitants with fear and humiliation.

Rain again delayo<l our departure for some days till August 24, when at length we continued our march, (hi the third day, from tho summit of the Shara-khoto pass, we joyfully greeted tho soft blue surface

• Tlic tiiintis of eight ge<jeiis, put to death here hy order of n Chinese prince notorious for )iis eruelty and severity in the exereisc of jiidieittl fiinctiuns. This prinee was despatched to Giindmni Ity the Itogdykhan to reduce tlic rels-IIious Tanguts. The prince charged the eight gifgiim with lK‘ing the instigators of tlie n'voit, and siiid to them : “ You that arc Itorii again know all tilings, nut only the past and the presi'iit, hiit also the future. Tell me when yon will die.” “ To-morrow.” they replied. *• No, you are wning ; it will l»c to-day ; ” and he ordertsl their heads to he eiit oft' at once.

IMI’KIliAL UUSSiAX UKoGRAlMIIt'AL SUCIKTY.

CiUMBUM MOSASTEUY FUOM TUE SOVTU-EAST.

US to ttduiirc his stately flight, particularly graiul when, without moving his wings, ho soars or glides along the mountain.

Tlic next day, August 27, the expedition reached the shore of the aljiine liasin, and Koko-nor smiled on us with still greater charm. Itefore us spread its azure surface slightly agitated by gentle undula¬ tions, which from the first day lulled us to sleep with their monotonous murmur. In calm clear Aveather Koko-nor was simply enchanting. This imposing sheet of water seemed to us all more like a sea tlian a lake. Its really grand dimensions,* its surface stretching beyond the horizon, the colour and saltness of the water, the depth, the high waves.

3i»7

of tile Koko-nor, stretching away to the west. To the south mountains were crowded together, among which the rocky summit of Ser-chim was conspicuous, silvered over with fresh-ialleii snow. This stands near Koko-nor, and farther off, in the same southerly direction, a still grander range lifts itself proudly to the sky, reflecting a dull whiteness from its thick mantle of snow. A solemn calm reigned in the marvellously transparent atmosphere ; the sun made its heat perceptible ; in the clear blue vault of heaven the feathered denizens of the kingdom of the air hovered here and there — the Himalayan griffon, the black vulture, and the bearded lanunergeier ; the last seblom approached near enough for

* Koko-nor lias a circuit of about ‘i'.W uiilcs.

398

THK MONGOLIA-SZE-CHUAN EXPEDITION OF THE

ami at tliucs the heavy surf, conveyed the idea of a Bca rather than a lake. We from the firat to the last day of our atay there ajxtke of it aa Biioh. The bathing in Koko-nor waa excellent; we bathed aeveral times a day. Floating easily on the water, we could swim to a con¬ siderable distance, and then giving oui'selvea up to the {K)wer of the waves, reach the shore again. The transparency of the water was so great that the sandy bottom and the fishes swimming over it could be plainly seen at a great depth.

liut if Koko-nor was lK3autifiil by day, at eve, esi>ecially in fine weather, it was truly bewitching, as it was on the day of the arrival of the expedition on its bank. The sun, having finished its daily journey, was sinking to its rest, its rays lighting up the mountains and spread¬ ing over the visible horizon and being reflected from the bosom of the water. A transparent feathery cloud, like golden lacc, moved slowly southwards. Yonder on the mountains |>erfect stillness prevailed, and Koko-nor became calm; it no longer blustered and beat heavily on the bank, but gently whispered to it.

On the following day our camp woke up when the dawn had just apjieared in the east. »Scr-chim still slept veiled in eumulo-stratus clouds that retiched down to the shore. From the lake geese cackled, snipe screamed and whistled in various tones, and the white-tailed fish-eagle uttered its call ; in the meadows the lark awoke and began to rise aloft, singing as it went. The caravan was in motion along a soft beaten track. The rays of the orb of day played on »Sor-chim ; the clouds, wanderers of the sky, rose upwards and melted away. After an hour the first sharp gust of wind rushed upon us, quickly followed by a second and third. Koko-nor knit its brows.

AVe followed the southern shore, meeting occasionally Tangut caravans of yaks. The nomads were moving from one side of the valley to the other, from the western to the eastern, where lay rich {•astures still untouched. Finally, the island Kuisu, the centre of attraction to us, showed itself. From the mcadowland sloping to the shore, where our caravan was now travelling, wo could see a still greater expanse of blue and azure water than before. Kuisu, like a gigantic war-ship, rose from the dark-blue waves and allured us by its mysteriousness.

It may bo remarked that Koko-nor is drying up year by year, its level is sinking, and the line of its banks is shrinking. Native observers, e8}>ecially former inhabitants of the district — Mongols — assert that Kuisu is increasing in size; that long ago, as the old men say, this island was scarcely noticeable, whereas now it appears quite as a mound.

On August 29 the expedition jutched its camp at Urto, the nearest point to Kuisu on the southern shore, where it remained about throe weeks, enjoying the full and unrestrained life of the nomads 3 miles

away from us. To the north, half a mile distant, lay Koko-nor ; to the south, 4 miles off, stretched the mountains, the western prolongation of Ser-chim. From the mountains flowed a stream, traversing a valley slightly sloping to the lake and entering a secondary or partially enclosed basin along the shore, mile in circumference. On a third raised Ijeach by the same stream, continuous with the mainland plain, we had our camp, whence we enjoyed an extensive view on all sides. The swift clear brook enlivened our bivouac, winding round it on three sides. About the brook, and still more on the neighbouring lake, geese, ducks, and widgeons were constantly flying. On the sandy shore

IMl’ERIAL RUSSIAN GEOGRAPHICAL SOCIETY.

399

THK TOWN OF OONGEU FBOM THE SOUTH-WEST.

of the lake itself sat cormorants and gulls in small and large flocks and solitary flsh-eagles.

On arriving at Koko-nor we took out our folding boat, and, having put it carefully together, commenced making trial trips to a distance of 2 miles and more from the shore.* We boated in all kinds of weather, when Koko-nor was stormy and calm. On the whole, the result of these trials was satisfactory. The boat swam on the water like a cork, obeyed the helm, and was fairly easily steered in any direction. It had, however, its bad qualities. On the first day the rudder and the fastenings began to get loose, the fastenings parted,

• The depth of Koko-nor is almost a fathom at the shore, and it gradually increases with the distance : at two-thirds of a mile it sinks to 50 foet, and ut 3 or 4 miles to 05 to 80 foot.

400

TUE MUXG(IL1A-SZE-1’HUAN EXPEDITION <)E THE

and, the Hides of the Imat o})euing, high waveH fell coiiHtantly into it. Thorough repairs were neoessary before we could have Hufficieut contidence in our craft to atten)[it the sail to Kuisu. A wooden strip laid along the sidcH strengtheueil the whole boat aud kept the fastenings firmly in place.

Having tested the boat, w’e had to think of provisions, adjust sound¬ ing and other instruments, and decide who were to take part in the navigation. As a starting-point, a place was chosen about 4 miles to the west of our camp and 2^ miles from the vbo on the bank, which serves as a landmark for the pilgrims who cross to Kuisu in wdnter on the ice.

Amidst these interesting occupations the first week of our stay at Koko-nor glided past imperceptibly. At this time Chemof returned from his excursion, having covered 500 miles. Crossing the Ala-shan desert from Dyn-iuan-in to Lian-chou-fu, he penetrated into the Nan- shan in the direction of the town Hatunga, a region hitherto unvisited, and discovered some glaciers occupying ravines on the northern slojre of the highest chain. Then he descended to the valley of the Donger-ho, and travelled by the caravan road to rejoin the exjredition. The first days after his return he was engaged in setting in order his observations and collections. We, as before, were occupied with Koko-nor, daily observing the condition of the w'ater, which was not the same at all times and parts. Thus, for instance, the south-eastern bay, calm and quiet, reflected the lovely azure shades of the sky when the northern part of the lake was agitated with waves of astonishing height, while dark high rollers tipjied with foam swept up from the north-west. Koko-nor quickly changes from calm to rough, but takes a long time to settle down after a violent storm. Calm or agitated, Koko-nor is always exceetlingly beautiful. I sat for hours on its shore, or wralked at a disUince up and down from our encampment, never tired of gazing at the lx>undless expanse of water, nor wearied of the monotonous splash of its breakers, which reminded me of the southern shore of the ( 'rimeu.

I decided to take the first tuni in the voyage to Kuisu. All was ready, aud in the evening of September y, I and the orderly I’oliutof moved to the landing-place. Koko-nor wiis peaceful ; the sun set in a clear transparent sky ; the barometer was high. Scarcely had the red of sunset vanished from the west when the moon rose from the east, illuminating all the visible surface of the lake. I was vexed that I had not made ready to sail for Kuisu a day sooner, that I might have taken advantage of the good weather. What would the morrow bo?

Having had enough of wandering on the shore aud enjoying the beauties of Nature, 1 turnetl to the tent to sleep. The sea also slumbtired, and not a sound disturbed the absolute stillness; but at two o’clock in the morning I was roused by heavy waves beating violently against the shore. At dawn, when we hail intended to start, the storm was

il

II

IMI'EUIAL lirsSlA.N OBOURAI’UK'Al. SIM:IETY. 4nl 4*

tfc

8till Uluru furiuuu. At midday it bu^'uu tu abutu, aud wu mudu 8uvunil '1

attumpta to get away from tho bank, but the tenth or twulftli wave i

forced 118 liauk again. ^

lieturning to the camp, I projiOBed to my comrades f’liernof and 1

Cliutyrkin, that they sliould go to tho starting-point aud patiently wait . ■

for smoother water. They were more fortunate than I. Going to tho ||

starling-place on September 11, they put oft’ from the shore the next

day at one o’clock, with tho intention of proceeding perhaps halfway, ^

if the lake did not settle down. However, as they advanced tho weather

imjiroved ; tho wind dropped, and tho waves liecame lower, liejoicing 4

in this circumstance, my companions put forth greater strength, with '

SOCTHEKS FLASK OF THE AI.PISE CHEST I.YISCi BETWEEN SIXIN AXU UCI-1»UI : CHINESE TEMPLE NEAU VILLAGE OF CUAX-HC.

more hope of accomplishing their purpose. Five hours of heavy toil at tho oars ])asscd, and half the distance lietwcen tho shore aud the island lay behind, though the island still appeared of no great size. Another tedious hour passed, and the island seemed as far off as ever. The desire to reach Euisu at all costs inspired them with fresh energy. Meanwhile twilight spread over the water quite unnoticed, and tho wind rose, bringing with it a heavy swell. Kvery minute it Itecame darker and darker, and tho island vanished from sight. The high waves threw jets of cold water into tho boat. The travellers had only their own strength to depend on, but their reserve of strength was Itcing gradually exhausted. Working at the oars, they encouraged one

THE MONGOLIA-SZECHDAN EXPEDITION OF THE

4(i2

anutlier with the idea that the island was near. Four trying hours I)assed, and suddenly, like a dark monster, Kuisu rose before them. Our travellers fortunately chanced on a smooth beach, where the boat could be put in ; rocks stood to right and left. One can imagine with what delight my comrades landed at Kuisu. At the same time the clouds Ijecame thinner, the moon showed itself, and Kuisu stood revealed in all its beauty. Weariness, hunger, and cold (from wet clothes) were for the time forgotten.

My companions were on the water seven and a quarter hours altogether, never leaving the oars, during which they rowed 18 miles. The boat, being a third full of water, was too heavy for the travellers to drag ashore till the ordinary lading was taken out. Then they drew it on to the beach, and, turning it over, made a shelter fur themselves. Having refreshed themselves with brandy, eggs, and tea, they tried to sleep, but their wet clothes made them too uncomfortable. They felt great chilliness, a premonition of fever, so they set off to wander over the island in the direction of the obo and shrine, the situation of which was very well known to them from numerous observations from the shore with a field-glass or in astronomical work. On the way to the fibo they fell in with a horse, which, neighing with alarm, quickly cantered away. This discovery was a welcome indication of the presence of i)eople on the island. After seeing the obo and temple, they returned to the boat, but soon started off again, this time in the opjKHiite direction, along the bank, where they came across a human habitation — a cave, surrounded by an enclosure, from which a flock of goats and sheep gazed with curiosity at the travellers. Fearing lest they should frighten some one, the two companions went on, and soon found a small empty cave, with a store of fuel (dung). There they made a fire, warmed and dried themselves, and at two o’clock in the morning fell asleep. '

On leaving the cave in the morning, September 13, Chernof and Chetyrkin saw a thin whiff of smoke rising from a cave-dwelling about one hundred and fifty paces off. They went to the boat, and having got the kodak and presents which had been brought in case they were wanted, they set off to visit the unseen hermit. On approaching the cave, they heard the voice of the monk offering up hb morning prayer, “ Oiumani padme-hum,” I'.e. O Treasure of the Lotus.* To give notice of their presence, Chetyrkin coughed loudly, when the invisible anchor¬ ite l^egan to pmy more fervently with raised voice, trembled, and began to break down. AVhen the travellers entered the cave he was terrified, made a long face, shook, opened wide his eyes, and putting his fingers to his throat, repeated volubly, “Ter-zanda? ter-zanda? ” ».c. What is to be done 1 what is to be done ? The travellers tried as much as possible

* Tlie lotuH, aeoordin!; to Indian mytkulo^'y, is the thmne of the Creator, and is also a symbol of the world.

IMPKUIAL RUSSIAN GEOGRAIMIK’AL SOCIETY.

403

to pacify the anchorite ; but ho was a long time in recovering, and pro¬ bably would not have come to his senses if the l)oat had not l)een injinted out to him, at sight of which he smiled, brightened up, and concluded that ho had to do with ordinary men, and not such beings as his imagination had eonjured up. The meeting with the other two anchor¬ ites dwelling on the island, Tangut lamas * like the first, passetl off (piietly, as they wore already informed .of (he presence of Ku-ssians on Kuisu, and their companion was i)resent.

I!

li

	1

OASIS OF UCl-nUI, TASKS FllOM TUB TKMFLK V1S-T8Y-MIAO, I.OOKISO NOltTU-WEST : TllK llOASO-UO IS TllK liACKGUOUSD.

My Colleagues sjxmt four days on the island, exploring it and making acquaintance with the life of the monks. Kuisu lies almost in the centre of Koko-nor, a little nearer the southern shore, and stretches from west-north-west to east-south-east. It is composed of granite and gneiss containing biotite, exjiosed along the coast and the higher parts

* One of the lamas eamc from T.uhraii, another from Gumbum ; of the third we have no information.

•104

THE M(»NU0LIA-SZE-C1IUAN EXPEDITION OF THE

of the iishtuJ, but iu other places covered with soil of sand aud clay, which sup{x>rts herbaceous vegetation. The relative height of the crest of Kuisu is 200 feet; the shore-line scarcely exceeds 3^ miles, the length of its axis is about 1 j mile, and its maximum breadth, about the middle, is a third of a mile. The island is esiiecially rocky along the southern aud western shores ; elsewhere it descends to the lake in long gentle sloi)e8. 'I’he vegetation is identical with that of the land along the southern shore of the lake. There are no springs or streams; the anchorites and their cattle satisfy their needs with rain-water collected iu hollows made for the jiurpose.

Hesides the temple and obo, erected on the highest part of the island, Kuisu is adorned with tombs, cave oratories, minor oIkw, etc. In the temple aud caves arc some interesting old idols. The chief occupations of the anchorites are the regular i»erformance of religious rites, copying out prayers, and making numerous Uatm — clay images of goils. They also attend to their flocks of sheep and goats, which uumljer in the aggregate ninety head. Each monk lives iu a separate cave, with a fold fur his own flock. The animals graze together, but on returning from the pasture part into three groups, each going to its own home. The horse is common property. The monks store up abundant supplies of curds, butter, aud fuel. They live exclusively on dairy products, with which they kindly regaled my comrades daily.

Eight foxes live on the island in the com|>any of the monks and their flocks; there are probably also alpine hares and other smaller rodents. As for birds, the chief sjiecies were swimmers and waders — geese, ducks, widgeons, cormorants, and gulls, usually sitting on the rocky coasts or on the saudflats. There were alwaj s fish * near the shore basking in the sun. When the cormorants came down to the water these shoals made otf, and the birds pursued them for a long distance. There, too, fish-eagles fished, perched here and there on the clifls ; a solitary specimen of the black-eared kite {Milvug me!an<it!$), probably a passing visitor, was noticed circling above the caves or flying from one end of the island to the other. Repeated inquiries from the natives about the existence of fur-bearing animals in Koko-nor, as well as jiersonal observations, led to no positive results beyond the information obtaiueil from the Tao-tai (Governor) of Sinin that he had seen skins of animals caught in the Koko-nor waters answering to the description tirst given by V. A. Obruchef.

In winter there is a change in the monotonous existence of the anchorites, for then the pilgrims cross to the island on the ice. How¬ ever, the ice cannot be traversed every year, Ixjing not compact enough, and intersecteil by numerous cracks. From the summit of Kuisu a fine

* It is interesting to notice th»t there lire no large fisli iu Koko-nor; the finest bpeciuicns were not more than 2S inches in length, or live or six |>oiinds iu weight.

rMPERIAL RUSSIAN OEOORArHICAL SOCIETY.

40.5

view is obtaine<l on all sitles. A small rocky islet towards the south¬ west looked very pretty, and a caix; or peninsula on the west stood out clearly. My comrades could not see the northern shore at all, whereas the southern shore was (piite distinct.

On Septeinl)er 17, after hourly harometrieal and other meteorological observations, my colleagues loft Kuisn and its hospitable anchorites, who invited their “first summer guests” and “ first foreign visitors” to wait till winter and return to the shore over the ice more safely than in “ that cockleshell,” they said, iwinting at the l)oat and shaking their heads.

On the return voyage four soundings were taken. The first, close to Kuisn, showed the maximum depth of the southern part of the liasin,* n.amely, 120 feet; the second and third, aliout halfw.ay across, showed the same depth of 1 1 5 feet ; and the last, 3 or 4 miles from the shore of Koko-nor, .agreed with the soundings taken while the boat w.as Injing tested, marking 80 feet, or a little more. The surface temperature of the water w.aa 58° Fahr. The bottom, over a belt extending from 3 to miles from the shore, was sandy, and further out muddy.

While f'hcmof and Chetyrkin were rowing to Kuisn, I made an excursion of 00 miles in the southern Kokrenor mountains. On my return to camp, Ivanofl’t joyfully greeted me with the news, “ Yesterday rScptemlwr 13) a rocket went up from Kuisu.” This informed mo that our “cockleshell” (9 feet long, 2 to 3 broad, and weighing 72 lbs.) h.ad reached the island with its crew. My comrades and 1 had arranged that from the time they loft the shore a watch should be kept at the head camp, at iline o’clock in the evening, for the appearance of two rockets. The first rocket was to bo tired on the day of their landing at Kuisu or the fidlowing day (as, indeed, hapi>eno<l) ; the second on the eve of their dejuirture from the island. “ The rocket flashed like a sm.all snake of fire,” repeated my constant attendant Ivanof. “ Thank G(h1 ! ” I .said ; “ now wo must watch for our friends’ second signal.” On Soj)teml>er 14, 1.5, and 16 I stood at the .astronomical pillar, sur¬ rounded by my fellow-travellers, the Sinin Chinamen and the Mongol drivers of Ala-shan. On the last evening, punctually at nine o’clock, a streak of light flashed for a moment ,acn)ss the d.arkness of the night. Immediately the silence was broken by joyous voices, and the Chinamen and Mongols, who had not seen the first rocket, were especially noisy. Next day, about nine o’clock in the evening, when Koko-nor was b(‘ginning to hash the shore, and our hopes were getting faint, T heard the voice of my comrades as they neared the tent. t)ur delight was

• Tliat is, 8iipp<Hiing Kuisu to lie on tlio line separatiu" the nortliern and sontlieni parts of the hasiii.

t Sergeaut-iuajor of the eseort of tli«‘ exis-dilinu, wlio has slinrol with me many hardships ami privations since the fourth ami last journey of my reorettisl master. X. M. Prjevalsky,

406

THE MONGOLIA-SZE-CHUAN EXPEDITION OF THE

unbounded. Questions and cross-questions followed, narrative and repetition, which formed the base of the description given above of the voyage to Euisu.

On the return of my colleagues, the camp was moved lU miles to the east, to the base of a peninsula or cape called Chon-shakalur (great penfold), which ends in a sandbank jutting out into the lake. From this saudspit Chetyrkin and the orderly Poliutof made two l»oat excursions, one 4;^ miles to the north, and the other south-eastwards, returning by the shore, a distance of 10 miles. The former was in the open water of the lake, the latter in the south-cjistern bay partly shut otf by the promontory.* On the second excursion, about midway, an accident happened. The iron cross-ties of the boat broke, in con¬ sequence of which the edges parted, the fastenings came undone, and the l)oat suddenly rolled to one side. Struggling with a considerable swell, the Iwatmen could scarcely manage the l)oat, and hastened to run ashore.

With this our work at Koko-nor oame to an end. I considered the second task of the expedition to be accomplished as far as lay in our power, and made preparations for our return to Sinin, which we reached by the former route on Septeml)er 28. In Sinin wo had the pleasure of meeting our companion N. Y. Napalkof, who had safely returned from an excursion in the eastern part of the Kan-su ])rovince. The route of Captain Napalkofs detachment included Djn-iuan, Tsin-an, and Sinin, forming a triangle the sides of which were connected with places explored by our predecessors. My colleague visited a corner of Kan-su hitherto quite unknown, and put down for the first tim9on the map the names of nine new towns.

Tliis time the Sinin authorities assigned to the expedition an excellent house specially reserved for distinguished visitors, standing in the centre

♦ Two soundings tetween the camp and the sandspit (IJ mile) gave a depth of .'to feet ; bottom muddy.

8ept< mb«T 21, soundings to the north. Temi>eratiirp at 11.45 a.m., of the air, 44'''C Fahr. ; of the water, 41t°"l ; —

Depth at 1 vc'rst (two-thirds of a mile) from the spit, 2S.i feet ; laittom sand.

., 2 versts „ 47A sand and mud.

r,7i

.50.4

07

82.4

8.5

mud.

mud and weeils. mud.

September 22, soundings to the south-east of the spit. Temix'raturc at 9.110 a.m., of the air, 4:5®'7 ; of the water, 45°'5 : —

Depth at 2 versts from the spit. 74 feet ; l>ot tom mud.

.. 4 „ ., 89

0 954

‘.•II

III

IMPERIAL RUSSIAN GEOGRAPHICAL SOCIETY.

407

of the town. They met us, as we say, with outstretched arms. They were already informed by the interpreter and troopers, who reached Sinin the day before us, of all that we had done at Koko-nor, and how well the Koko-nor natives behaved to us ; and, more than all, that wc were all safe and sound, and none of us lost. The Tsin-tsai and other Sinin officials, when they saw us, could talk of nothing else but our navigation of Koko-nor and visit to the island, and all examined our boat with great interest, l)oth when put together and taken to pieces, and also my colleagues, whose hands still boro thick callosities. At last the Tsin-tsai remarked, “ You Russians have navigated Koko-nor for the first time, are the first to give me information about its depth, and are the first foreigners who have visited Kuisu or Khaisin-shan.

1 shall certainly report all this to Pekin.” Not to endanger our excellent relations and weaken the goo<l impression produced by every interview with the Sinin authorities, I thought it l)etter not to refer to the former conviction of the Tsin-tsai, that not only stones, but also wootl, sank in Koko-nor. The truth was self-evident.

In Sinin the members of the exjiedition had the pleasure of making acquaintance with the English family of H. French Ridley, who had lived there as a missionary for more than fifteen years. In the company of these amiable people wo passed many pleasant leisure hours, reminded of the culture of Europe. French Ridley was also interested in our navigation of Koko-nor, and whenever an opportunity cKscurred led the conversation to this subject. Being an old inhabitant of Sinin, the friendly missionary gave me much valuable information about the dis¬ trict, and effectively assisted in the enlargement of our ethnographical collection by obtaining articles from the inhabitants of the place.

In Sinin, too, thanks to the general improvement of postal manage¬ ment in China, we received our letters fairly regularly, and could also dispatch our correspondence to our distant home. One always receives pleasant nows and another unpleasant. So it was now. Chemof heard sad news from home, and was compelled to make an early start for Russia. On his way across Ala-shan, Central and Northern Mongolia, he would make geological and geographical investigations.

Taking advantage of Chemofs departure, I sent off to the depot of the expedition at Alasha-yamyn seven baggage animals, chiefly laden with collections. For our further journey we exchanged camels for mules, which have a particular facility for travelling in mountains, but, on the other hand, require careful attention and plenty of grain fodder. A good mule carries, without much difficulty, a load of 220 to 260 lbs. equally balanced on both sides of the animal. As we did not try to curtail our baggage, the number of animals remained the same, namely, twenty-three, which, with eleven riding-mules, formed the caravan, and extended along a considerable stretch of road. The expedition left Sinin in two sections, starting at different times. The camel caravan.

408

TnE MONGOUA-SZE-CHUAN EXPEDITION.

under Chemof, left on October 0. The main caravan set out on October 12, and arrived on the 15th at the oasis Hui-dui, having crossed on the way a lofty alpine chain by the pass La-chi-lin, more than 1:1,000 feet above sea-level. This journey was enlivened by the passing of natives, travelling in one direction or the other. At this time, besides trading caravans and caravans laden with pears from Hui-dni to 1)0 sold at Sinin, the whole roatl was full of smart Tanguts repairing in motley companies to the monastery of Gumbum for a festival.

The oasis Ilui-dui, which the expedition is to leave one day in the early part of January of the present year, is spread out over a wide part of the valley, on the right bank of the Hoang-ho. Two tributaries, which supply the agriculturists of Hui-dui with water and are fed by the southern mountains, divide the oasis into three parts — a western, central, and eastern. In the middle part, near the main river, is situated the town; on the southern border, the Chinese temple Viu-tsy-miao. On the borders of the other parts stand Buddhist monasteries, of which the most noteworthy is the monastery Goml)a-sy, hidden among the wooils of the south-eastern outskirts of Hui-dui. The native population of Ilui-dui, numbering al)Out 7o0(j to 8000 souls, consists of Chinese collected in groups in the interior of the oasis ; Tanguts, only a fifth or sixth as numerous as the principal inhabitants, dwell on the outskirts near the mountains. Both peoples live on friendly terms, and often intermarry.

In autumn and winter the oasis is bare, and is of a dirty, dark, dull colour. It is very different in spring and summer, when a rich vegeta¬ tion displays itself, and Ilui-dui presents a harmonious emerald airpet, strikingly contrasting with the general yellow monotony of loess landscapes.

In conclusion, I wish to say a few words on the results obtained by the cxjiedition during the second (Koko-nor) period. The chief caravan and two detachments traversed and surveyed IfioO miles of country, connecting its observations with three new astronomically fixe<l points, besides old ones. On all journeys meteorological readings were systema¬ tically taken, prominent points were determined hypsometrically, and cfillections were made — geological, zoological, and botanical, the last, however, only with the main caravan. These collections, the zoological in all its sections, constitute a full record of forms. In the ethnographical collection objects of the Buddhist cult are conspicuous, such as idols of metal or painted, gau or amulets, articles indispensable in the performance of religious rites, Tibetan books. This collection owes much to the valuable information and advice kindly given me at starting by our graduate S. O. Oldenburg. The expedition has also acquired a fair number of siiecimens of Chinese painting and photographic plates of views and types of the country piissed through.

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909: PRELIMINARY REPORT.

By C. SEOTTSBERG, D.Sc., Leader.

VI. OvKItLANU FROM LaOO NaHUKLHUAI’I TO PuNTA AuKNAS.

As I have remarked on several occiRsions, our exj)edition has profiteil very much from the very valuable assistance oflTered by the < 'hileau Government. We were not only able to carry out our investigations in less time than ever expected, but also to enlarge our working-field in a considerable degree. As, in the beginning of October, 1908, wo were ready to return south with the awakening spring, one single excursion of our original plan was left, viz. to the Beagle channel. To spend all the spring and summer there would have been a mistake, and therefore we had to enlarge our work a little, taking care to put our new work in strict relation with w'hat had been done before. During the last half-year we had often discussed the matter, and resolved, if ]>ossible, to return to I’unta Arenas overland. Finally, wo were able to arrange the economical side of the matter satisfactorily, and arrived at Puerto Montt in order to pack and send off the collections made during the winter. After this we travelled over the < 'ordillora to Lago Nahuelhuapi in the Argentine Republic, where we t<x)k up our headquarters in the little village of Bariloche. Hero we bought horses, gear, and provisions, and also found time to make a couple of excursions.

tin October 23 our caravan stood ready to start. It counted four membei's — Messrs. Halle, Quensel, and myself, and our old assistant from the expeditions in the south, Pagels, who again had joined us. We had the smallest number of horses ever used in Patagonia for a trip of any extent— ten and a mare, the inevitable “yegua madrina;” and I very much doubt that anybody in Bariloche expected us ever to reach Punta Arenas with those horses, especially as they were not even in very good condition. Each of us disposed of two animals, and the remaining three carried the cargo. It is clear that only the most necessary things could be brought along : some instruments, a Win¬ chester, a kodak 9x1- with films, small bottles for plants and insects, etc., and sleeping-bags, a tent for two persons, and provisions for the first month, during which wo could reckon on the hospitality of the farmers, allowing us to completely omit some articles.

Our plan was to follow the Cordillera, passing close to the great lakes, and using the passes between the mesetas and the main range. The region wo thus intended to explore is known only from the work of Prof. Steffen and the members of the Argentine-Chilean Boundary ( 'ommission, well known to all English readers. It resulted in a very fine map that we used all the way down. But the geology and botany No. IV.— OCTOBKR, 1909.] 2 F

410

THE SWEDISH MAOEI.LAXIC EXl'EDITloN,

were little kuuwu. Not a few jteople have i>a8»etl along the luouiitaiiis, but they have kept outaide on the painpa, where travelling ia caay. Wo rode to Norquinco, paaacd Kio t'hubut near Mayten, and proceeded to Lelej, where we had aeveral excuraions to make. Here wo aaid good¬ bye to civilization in the form of jKjet and telegraph, and atarted on November 3. Through the Nahuelimu paas wo came down in the “ Valle IG de octubre,” a fertile >’alloy of the aubaiidine ty|»c, deacribed by Steffen in this Juumal some years ago. Hero wo made an excuraion of three days through the forest region to the ( 'hilean frontier.

From hero we went right across south, passing close weat of Lago Itoaario, and came down in the valley of Kio Carronleiifu, generally cidlcd “ Valle < ’orcovado.” Over large swauqta w’O then found our way eastward to the valley of Rio Tecka, where we etopi»ed several days. The vegetation was now fairly well develoiied, and gave me a lot ol work.

From Tecka river we again set course south and passed the I’ico valley, from which we hml to cross a 4U0il-feet-high meseta to come down into the Frias or Cisnes valley, where we stayed a couple of days to explore westward. A road is projeeted here across to the I’acific. In order to reach the next large depression, where the Aysen river system is laid out, we had to pass outside a very stony meseta and to cross the Senguerr river, one of the largest Patagonian rivers, at the foot of the mountains. We soon crossed the Chilean frontier and halted at Nire- huau, a place belonging to the Aysen ( 'ompany. We had now left the pampa, and found the country thickly wooded everywhere.

From this place a road is made across to the Coyaike river, and thence along to the main estancia, situateil between the high mountains. \N'e stayed a week there in order to give our horses a goo<l rest and to follow the road ourselves down to the I’acific ocean. It is a most interesting road, as it offers a complete section through the mountains and a splendid opportunity to follow the gradual development of the Pumilio- forest into a rich rain-forest. It luis lieen a tremendous work to make this road. For long stretches it is laid with thin sticks, as the ground is extremely swampy ; it is im])ossible to pa.ss at the side of the road, the ground being covered with fallen trunks and over¬ grown with thick Iximboo (^Chmqnea quilu). lJuring most of the time we had torrential niiu. The distance to the Pacific is 50 miles.

On December 3 we left Aysen. IJefore us we had a long stretch of uninhabited country, and consetpicutly we had to carry more provisions, making the cargo rather heavy for our two jiaek-horscs. In Tecka we had secured two more animads, making it possible for us to change also the pack-horses, which had jjrovetl almost necessary. The provisions were intended to last a month, and I dare say that we only carried the most necessary things, and were prepared to suffer some privations. Rice, some Hour, tea, and cocoa were the most important articles, but

THE SWEDISH MAGELLANIC EXPEDITIDN, 1907-190').

411

our iirinoipul fooil was thuiuoat of guaiiaeu and deer (the hueimil), which in this region are rather identiful. Mostly wo had to do without bread — that wo regarded as a luxury.

With some difliculty wo {Kissed tho swaiujiy valloy of Kio Mayo, and thus shnsl at tho foot of a 450<*-feet-high meseta, which we had to cross, as wo knew no {lass between it and tho main range. It took us three days of hard work. Tho snow had recently molted some little brooks could iio {lassed on snow-bridges. The ground looked very firm, being covered with round stones just like an old-fashioned pavement, but the soil underneath was so saturated with moisture that it gave way everywhere under tho hoofs of our horses, and we had to wado in tho thick clay for hours and hours. 'I'ho ditliculties wore highly increased by tho sources of Rio C'halia, extended like a fan over tho entire {dateau and deejdy cut down, forming narrow canadons, filled with an almost im]>enetrublu brusliwoiKl. Tho second day was worse than tho first : all the little firm j»atches of heath were comjiletely undermined by that small rodent, the tuco-tuco, and the descent into tho Koslowsky valley, climbing down in the ravines of a very steep 2.j0<>-foet-high barranca, was rather adventurous. Rut down wo ctimc, and I got rich remunera¬ tion in the tlora of the valloy. After a day’s rest we crossed the valley and sought the next pass over the mountains. Hero we could leave tho meseta on our left, and had a comfortablo march over a 40DO- feet-high jiass down to tho vast dei)ression of Lago Ruenos Aires, the greatest Patagonian lake, double tlie size of Lake Constance. Wo followed the Kio Fenix, which comes down from tho j)a6s, and crossed it near its outlet in the lake. Round this lake, I, for tho first time, mot with the remarkable and undescribed vegetation mentioned below.

Following the south side of the lake, we crossed the canadon of Rio Chilcas, and cut across towards Kio do los Antiguos, a large river Mowing to the lake. Arrived at the edge of the plateau, we found at our feet a several hundred yards high barranca. Venturing to descend, we slid down in the loose sand till we, to our surprise, found ourselves at the top of an almost vertical, barren sandstone wall, where all further advance seemed impossible. With great risk we climbed down into a crevice, but nearly lost one of our {Kick-horses, that tumbled down about l<tU feet— another inch, and it would have been lost. The cargo was lowered down with ro{)cs, and, cutting steps with a knife in the sandstone, we managed to get the animal over into a ravine, where we could drag it down. We soon crossed the river, and arrived at tho {Kirullel river Yeinemeni, which we had to follow south u{) to tho {lass. The following days were a constant chain of difficulties. Wo tried to advance along the river, but every few hundreil yards tho rushing waters had cut close to the foot of the Uirranca, and only with tho utmost care could wo lead the horses across these dangerous {lassages, where they hardly found any foothold. Now and then they broke out

412

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909.

mad with fright, turuud uut iu the rivur ur climbed the barranca, and had to be fetched Uick, often with a lot of trouble. At last the road was shut by a vertical wall, and we now climbed about 1000 feet, till we were well above the barrancas, and continued on this height. It certainly was a mistake ever to try the bottom of the valley, but as it came off without disaster, it was only another experience useful enough for the future. At Hio Zeballos, a tributary to liio Yeinemeni, we made a day’s halt to give the horses a rest before crossing the pass and to make an ascent of the mountains to the west of the river. Above 3200 feet the country is thickly wooded. Wo found the snow-lino at about 5500 feet.

The pass gave us some trouble with its “ floating soil,” where wo had to be very careful. The height >vas about 6000 feet. We followed the Kio Gio down till wo found some fuel uear the Chilean frontier*. Some 2600 feet below us we now had the lakes I’ueyrredon and Posadas. The terrain north of them is exceedingly broken and intersected by cracks and canadons, and along the shore we found the most beautiful traces from the great Ice Age. The two lakes are separated by a narrow isthmus. Lago I’osadas lies a yard higher alxrve the sea-level than Lago Pueyrredou, and a short river joins them. Without adventure we passed the river where it flows out in the lake, a high west wdnd jrressing heavy breakers against the current and terrifying the horses, which certainly never had seen anything like it Ixjfore, and had to bo forced out in the foaming swell. Wading across the sand-dunes on the isthmus, wg met the mountain range on the south side, where we had to climb from little more than 300 to 3500 feet in a few hours’ time. On that height we slowly found our way to the valley of a small river, Pio Tarde, which we thence had to follow up to the next pass. At the last forest patch wo halted. New geological formations craved our attention, and the horses badly wanted a rest before crossing the pass. In large snow’drifts and floating soil w'e crossed at 6800 feet, the greatest height reached with the horses. Between remarkable pillars of basalt we descended into the valley of h’io Belgrano down on to thp 2300 to 2500 feet high i>lateau of Lago Belgrano, where we arrived on December 23. On the way through the canadou w’e killed three deer for Christmas — we had been without meat a couple of days, and felt somewhat hungry. A vegetarian has not a very happy time in Patagonia.

We resolved to make a longer stay at the lake. Mr. Halle intended a geological survey to the east, and the rest of us prepared for a boat trip, iu order to jxjnetrate as far w'est as possible. The boundary commission had left a small canvas boat on the south shore of the {)euiusula. Wind, sun, and rain had treated it rather badly ; but still we were able to use it, and set out on the 2Gth. It was leaking suflicieutly to keep one of us occupied. Too much wind and sea

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909,

413

made progrcBS very slow at first, but wo waited and sot out again. We found Lago Azara situated about 30 feet IksIow Lago llelgrano, that empties through a cascade into the former. We carried Imt and luggage the short stretch between the two lakes, and then staited again to roach the west arm of the Lago Azara. The entrance is very shallow, and almost shut by innumeraVde trunks carried out from the forests to the west. We now had reached the forest region. I had expected the forest to change going westward, but only found solitary specimens of the evergreen beech (Nolhofagus hetuloideg) on two places. From the end of the west arm we made an ascent of Cerro Aspero up to the glaciers. From this place wo had a magnificent view of a very picturesque part of the Cordillera; to the west the undulating, forest- clad hills, with numerous lakes, rising to the high crests and the inland ice; to the north the imposing 12,000-fect-high mount of San Lorenzo, differing in geological structure from the surrounding peaks, of which many reach a height of more than 3200 feet. Returning through Lago Relgranoon the 30th, we passed north of the peninsuba. Clenerallj’ the Andine lakes, as far as known, are deep ; but here wo found this part of Lago Belgrano so shallow that wo hardly were able to pass with our little boat. This is caused by the Rio Lacteo depositing enormous masses of sand or mud.

On Janmary 1, 1009, wo continued our march, cro8se<l the rivers Relgrano and Lista, and reached the valley of Rio Nires in two days. Thence wo passed into the valley’ of Rio Tuco-Tuco, whose name is sufficient to frighten anybody. The entire valley’ is one extensive and dangerous bog, and tbo firm margins are completely undermined by’ the tuco-tuco. Wo wore happy to pass without adventure, cut straight through a dense forest, and slid down the steep barranca of Rio Carls m, where it unites with the Rio Mayer. This river receives the water from the Lakes Nansen, Azara, and Belgrano, makes a sharp turn, and flows to Lago San Martin. Wo wore now separated from that lake by’ a high pass in the valley l)etwecn the Meseta and the main range. After a survey round these rivers, wo starte*! to ascend the steep and inaccessible canadon of Rio Carbon towards the pass. As an illustration, I mention that we had to cross the river twenty-seven times l)cforo reaching the pass, at a height of, more or less, 5000 feet. It was ct>mplotely covered by extensive snow- and ice-fields. The last stretch wo rotlo in the river itself between high walls of perpetual snow. On the south side wo met with the floating soil, the most treacherous ground one can imagine, sliding down like a thick porridge, and where horses easily got caught as well as people. After a struggle of thirteen hours, during which we had walked perhaps half the stretch, we reached the first bushes and grass in the darkness. A very important geological discovery was made during these day’s, and Mr. TTalle later returned to the place.

Through a narrow canadon wo reached an esiancia on the shore of

1

414

THE SWEniSTI MAOET.T.ANIC EXPEHITION, 1007-190!).

Lago Sail Martin on January 7. The owner, Mr. S. Frank, waa an old ac(iuaintanc<.‘, and wc had agreed with him to make a lioat excursion together in the lake with its curious inlets. Some years ago the last lioundary commission had brought a large, fine Berthon boat there. It was left on the beach and never launched. We found it very imudi in want of repair, and started to work without delay, replaced the broken wooden parts with new, mended the canvas, and paintetl it all over, and got it tight.

We had to wait several days liefore we could think of starting. A constant west wind blows for days and weeks, making it impossible to get out of the narrow east arm, which is blocked by the deposits of Kio Fbsiles. We wont to sleep on the beach the 15th, and started the next day in a fresh head wind. There happened to bo sufficient water to let us pass the entrance of the bninch, and we steered for the north shore. The party consisted of (^uonsel, Frank, myself, Pagels and two men from the farm, who had some knowledge of life on water. Bad luck followed us. We kept close under land, looking for sKolter l»ohind the protruding points, struggling with a head wind and a heavy sea ; but, thanks to the iierfect sea-boat, we got out of it without other inconvenience than so.akod clothes. Day after day wont with rain and stonn. The difficulties were increased by the fiict that one is always unable to land with a canvas Imat on other places than a sheltered sand beach. For long stretches the cliff runs perixjndicular into the water.

After eight days’ hard work we had enterotl the western of the two north arms. The vegetation had changed. Bound the entrance of the c!ist north arm wo came into the forest l)elt; and, arrived at the lirst- mentionod branch, we found the mixed forest of evergreen and deciduous¬ leaved trees. Wo had a very strong desire to reach the lM)ttom of the branch, where the great lake has its outlet through Itio Paseua to the Pacific, but the scarce provisions and the almost insufferable head wind made us give up this jdan when we were alsnit two days’ journey from the goal. Instead we turnotl south and sailed down to the great glacier at the entrance of the south arm, where wo ha<l to navigate l)etween icebergs measuring up to .‘u) feet above the water. Wo returned south of the Isla Central, and expected to l)e able to snil Imck — wdiat we really longed for, rather tired of eight days’ hard pulling. But now, at last, we got calm w-eather, and oven easterly wind, and thus were forced to continue pulling all the way. The distance covered was l.'iO miles, and in spite of all ImuI luck wo were satisfied with the results. We arrived at the farm on the 26th, and continued our march two days later. Mr. Hallo had l»een very successful in his .stratigraphi(^al and palicontological w'ork during our absence. (Joing west of Laguna Tar, wo reache<l Bio Shehuen and crossed the Meseta, where we, from a height of only :U)00 feet, got a glimi»sc of both

TIIK SWEDISH MAOELI.AXir EXPEDITION, 1907-190!). 415

Lago Viedina and Lago Argontiuo. In the canadon of l!io (Jangrojo, a tributary to tlio former lake, we found water, fuel, ami grass, the necessary articles for Patagonian travelling. Thence we came down to the sandy and barren shores of Lago Viedma, and followed IJio de la Leona, which carries the water of that lake down to Ijago Argentino This part of the Cordillera is the most sterile wo have seen. There was no grass for the horses, and we had to watch them closely at night. There were guanaco, and even horse-trficks along the steep barranca, making it possible, but not more, to pass for some miles; but then we climbed up in a ravine, and continued the march at some dishincc from the river. The landscape, where wo now found ourselves, was very remarkable — as far as one could see, only thousands of hills, intersected by deep ravines, not a drop of water or a blade of glass. The ground was full of hidden, funnel-shaped hollows, where only a narrow crack indicated the danger. It reminds one of the “ loess” in China, though of course the geological origin has nothing in common with that formation. Close to Lago Argentino wo found a small farm, and grass for our hungry animals.

On February 2 wo crossed llio Santa Cm/, and made for the llaguales mountains. Quensel and I had some (disorvations to make in the forest region further west, and wo met the other party south of the 4000-feet-high pass at Itio Centinela. In a dry summer like this had lieen, this pass is gfXKl enough ; extensive and often impassable swamps cover the ground on the south side. Wo reached ContriTas, and were back to civilization. Thousands and thousands of sheep belonging to the big “ Comjiahia Fsplotadora de Tierra del Fuego ” graze here, and in tlie main estancia, called “Cerro Castillo,” wo got our first corre¬ spondence since October. From there Quensel wont to Ultima Esperanza (Last Hope inlet), famous for the (Iryptotherium cave, and made a lioat excursion to the channels west of that place, discovering two new inlets, several miles in length. I myself, accompanied by Pagels, lotle in to the farm of Mr. Ferrier, at the foot of Patagonia's most l»eautiful mountain, the 10,000-feet-high Cerro Payne, where Quensel had made very important geological studies in the previous summer. Wanting to make some phytogoographical observ’ations, we rode farther west through a thickly wooded country, but had to abandon the horses and continue on foot, till w’o reached the edge of the inland ice and the sources of the river wo had followed.

On my way back from Ferrier, I passed Ultima Esperanza and joined Quensel there, and some days later wo met llalle, who had occupied himself with some interesting layers with plant fossils in the vicinity of Cerro Contreras ; and together we started for Punta Arenas, where we arrived on February 25, looking back upon a distance of 15t>0 miles, larger excursions included. The straight lino from Lago Nahuelhuapi is only 85D miles, more or less.

416

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909.

Phytogeographieal Bemarls. — It is evident that I had to confine my work to general studies on plant geography and physiognomy. Most of the region had not been visited by any botanist before. I can divide my studies into three parts ; (1) Observations on the changes the flora undergoes from north to south ; (2) fixing the limits l)etween the different forest zones and between those and the pampa ; (3) a more detailed study on the little-known alpine vegetation.

(1) Our way led through the boundary region between forest and steppe, with numerous excursions to both sides. I found the flora of the latter rich in species, but not undergoing very great changes with a different latitude. It seems convenient to establish the following natural formations : the grass steppe ; the Mulinum steppe (iff.

an Umbellifer) ; the heath, which has many features of a desert, and as far south as liago Argentine is characterized by the presence of various Cactaceae; the brush wootl round the rivers, formed mostly hy Nothofagtis antarctica ; and the water vegetation. A local but rather remarkable formation is the bush steppe round the great lakes, formed by species of Lj'thrum, Escallonia, and Verbena, etc. Rich in species at Lago Buenos Aires, it gradually gets poorer, and in the south hardly more than the Escallonias and BerJmris efr, mtrrophylln, the most common bush of the pampa, are left. It often was the only fuel available, and a very spiny one indeed.

Concerning the geographical distribution of the pampa plants, the greater part seem to l>e found all along the Cordillera ; in consequence hereof, it is difficult to divide the pampa in latitudinal zones.

I also could follow the longitudinal distribution of the forest. Down to the Corcovado valley we find, to a height of 240D feet, more or less, patches of Libocedrut chtlengis, at the very edge of the pampa. All along the mountains stretches a belt of Nothofagw pumiUo, in the northern part growing from a height of 2.500 to 3500 feet and upwards, gradually appearing on a lower level. As we remember, it reaches the sea at the Straits of Magellan. I also had many oj»portunities to complete my studies of the rain forest.

(2) My conclusions are based upon our excursions across the Cordillera to I^ago Nahuclhuapi, in the valleys 10 de Octubre, Frias, and Aysen, and in the lakes Belgrano and San Martin ; finally, in the region of Ultima Esperanza. The general rule is that no rain forest is found east of the main range. The low passes and the northern valleys, where rivers cut through to the Pacific, make an exception ; here a typical forest of evergreens extends as far as to the east of the Cordillera’s highest ridges. After a narrow but often well-marked mixed zone comes a deciduous-leaved wood of Nothofagm* pumiUo, and east of this a margin zone of N. antarctica, mostly round the rivers.

I have studied the steppe from 300 feet above sea-level up to the snow limit, which from about G500 feet in the north of Patagonia

1

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909.

417

sinks down to al)Out 3800 feet in the south. With exception of the alpine region, which starts at a height of 25<>0 to ."»0O0 feet, the vegetation is very uniform at different levels above the sea.

(3) Of great interest was the study of the alpine flora and the vertical limits of forest. N. jrumilio seems to form the alpine dw.arf forest everywhere on the east slopes of the Cordillera ; on the west side we find N. nnUtrctiea above the evergreens. The alpine vegetation is rich in species, and contains many remarkable typos, and, l)oing little known, the survey ought to give good results. I can mention here that a lot of species, described lieforo from one single place and regarded as verj' rare, have l>een found to characterize the wh<de visited region. The xeromorphic construction of Patagonian alpine plants is veiy striking. The shajM) of compact, hard cushions is developed to a very high degree, having representatives in the most different natur.al orders. Some of these types were probably unknown l)eforo.

As a hint to travellers, 1 will add that I did not carry other plant- press than two pieces of cardboard, and no other paper than the old newspapers I was able to pick up in the farms under way. Still, the collections are quite as big as if I had b<‘en working under the most normal conditions.

We did not make any other zordogical collections than of insects and spiders.

Geolthjleal liemarlit. — Mr. P. I). Quensel, B.sc., communicates as follows : —

“Our long ride along the eastern slopes of the Andes gave me a splendid opportunity to study the geological construction of the moun- biin range, and thereby to conqileto my eailier researches, partly fionj Ultima Esperanza and Tierra del Fuego, and partly from the west coast. Naturally on an expedition stich as we were now undertaking, and with the small equipment wo had at our disposal, it w'as out of question to make detailed observations everywhere on the route. In most cases I had to 1)6 contented with a survey of the principal geological features of the country. We had planned our route along a geological Isjundary- line, so to say. Most of the way south from Aysen, wo passed exactly where the slightly disturbed strata of the mesetas meet the older much- change<l sedimentary or volcanic deposits of the Andes. The meseta deposits, which have a vertical extent of lietween .‘5000 and 4<>00 feet, are cut through by the great pre-glaeial valleys, which now contain the subandine lakes and the large rivers, which either flow eastwar*! to the Atlantic, and in this cjise follow the original course of the v.alloy, or westward through the transandine valleys. In the latter case it is without exception glacial deposits which form the water-divide, and have caused the rivers to seek their way through the mountains, at the same time in many inst.ances causing the formation of the large lakes. These same lakes were an obstacle that forced us aw.ay from our direct

418

THE SWEDISH MAOEELANTC EXPEDITION, 1007 1009.

march-line out on the pampa, hut only to nsturn immediately to the eon tact-lino between moseta and the (lordillera, where the erosion has had a favourable j)oint fur its attack, and has everywhere formed available pa.ssc8.

“ On our way from Puerto Montt to Nahuelhnapi, I gaineil a complete section through the Cordillera of those latitudes, aud again could state the regular construction of the Patagonian Andes, wherein they so strongly dissemble the more northern parts of the same range. Of interest is the very great pro})ortion of eruptive rocks, which here as well as farther south jmrtake in the formation of the chain ; the sedi¬ mentary ones are scarce, and mostly to lie found on the eastern side. The younger volcanic rocks play the same part as forther south ; this question I have already dealt with in an earlier paper.

“ From Nahuelhnapi to Aysen, the oast Cordillera is almost entirely composed of porphyries and jiorphyric tuffs of varying appearance and geological age; only in few places the older, highly metamorphosed, underlying sedimentary rocks are visible. At Aysen the al»ove-men- tioned road gave another section. Here the eruptives, C8j)ecially granites, play a still greater part than farther north. For over 4o miles the road traverses one granitic eruptive zone. To the n.aked eye these granites show no sign of dynamic destruction. One can here cla.ssify the Cordillera rather ixs an eruptive than a dynamic or foldetl range of mountains. South of Aysen, we come into the region of the mesetas or great plateau-mountains. They are only slightly disturbed, having just l>eeu caught in the most recent period of folding, and generally dip only some ItC to 20° t«i the east. They consist of comi>ara- tively loose sandstones and conglomerates, covero«l by recent alluvial dejwsits of sand aud gravel. The surface is perfectly smooth, excx3pt where the rivers, always running west to east, have cut their deep cauadons. South of the lake lluenos Aires, the mesetas are crowned by a cover of basalt and traversed in all directions by volcanic dykes. These eruptions are of recent geological age, and belong to the bas.altic eruptions of the panqia formation, dating from I’lioceno to j)08t-f}lacial age. Farther west these formations discordant overlie the elder jKjrphyric tuffs of the central Cordillera, which disappear 8<mth of I.Akc llclgrano. On a 1>oat excursion to the alpine Lake Azara I again found some representatives of the granitic lakkolites I had previously oxploro<l iu South Patagonia, where they are so characteristic for the eastern Cordillera. A good example is the 1 2,000-feet-high Mount San Lorenzo, north-west of Lake Belgrano ; the highest peaks west of the lakes Vieslma and Argentine l>elong to the same tyj>o. Hero and there one can still trace the cover of slates as a dark cap on the top of these mountains. South of Lago llelgrano the sedimeutarj* dejM^sits reach farther to the west, and consetpiently are more distuil>e<l. In Lago San Martin we unide another loat excursion to the west, and reached

THE SWEDISH MAOEIJ.ANIC EXl’EDITTOX, 1!»07 1009.

419

the groat glacier, which reaches tl»c water south-west of Isla t’enlral. AH round the lake I hud found highly distur1>od slates, htit the moraine of the glacier only carried andesitic hlocks, showing that more to the west the volcanic rocks again play the same important part in the central Cordillera. After my arrival at Ultima Esperan/a, I finally made another Isiat excursion in order to complete mj’ observations of an earlier date in those parts.”

Mr. T. C. llalle, n..sc., gives the following communication of his re.searches : —

“Dtiring the journey from Bariloehe to I’unta .Arenas, we had goorl opportunity to study the geology of the transitional zone iKstween the Cordillera and the Pampa. The older nsiks which here appear at the surface are comparatively little known, and suhseiiucntly foiined the most interesting object for our researches. Much time was also si>cnt in studying the plant-hearing layers, fo.s8il plants having l>een describi'd hitherto only from a few places south of Lago Argontino.

“ In the great Tertiary tuff and sandstone formation, which oxtemds over a large area south from T.ngo Nahuelhuapi, fossil plants were found in several localities. Near Bariloehe the impressions occur in great folde<l beds of a grey tuff. Consequently the formation of the mountain range was not yet finished at the time when the tuft’-lieds were deposited. Other localities for fossil plants arc Cerro Colorado, alstut one day^’s ride south from Bariloehe, and Estancia Eelej, some days further south. Many of the siieeimens are well presen’e<l, and will make it jKwsible to determinate the age of the formation, and al.s«» of the last upraising of the Cordillera in this region.

“At Rio Ayson, where wo had opimrtunity to make an excrursion through the Cordillera and down to the Pacific coast, the stnicture of the mountain range is quite different. Even close to the eruptive zone the range is built up of slates, sandstone, and tuff, lying in horizontiil Is'ds and not showing any signs of folding or regional metamorphasis. In some places the slates cfuitain fossils, by’ the presence of ammonites and l>elemnites proving not to bo younger than Cretaceous, and eon- sfspicntly tdder than the raising of the mountain range as known further north or south.

“ South from Rio Aysen we crossed the high mesetas round the upjier parts of the rivers Chalia and (Suenguel. These mesetas wore found to consist of light-c«dourtHl and very’ soft sandskines and tuffs. Also here fossil plants were found, which by comparison with those from the tuff series further north will give some information on the relation l)etween these two formations. South of Lago Buenos Airi'S, in the region explored by Hatcher, collections of fossils were maile at several hsalities.

“ The geology of the large meseta north of Lago San ^lartin proved to bo of great interest. Especially the older layers are here very well dis¬ played in the valleys f)f the rivers Carls'm and Fdsiles. The undermost

420

THE SWEDISH MAGELLANIC EXPEDITION, 1907-1909.

layers in the meseta consist of dark folded slates containing a compara¬ tively rich marine fauna. This series probably is identical with the ‘ Mayer River-beds ’ of Hatcher, which were referred by that same author to the Upper Jurassic, though the determination remained somewhat doubtful, the fossils being scarce and badly preserved. The black slates are covered by mighty beds of a yellow sandstone, which to the west also are folded and dislocated, but further cast apjiear undi8turl)ed, only dip¬ ping slightly to the east.

“ In the bottom of the sandstone series I found in one place a fossil flora, consisting of a cycad, a Balera, and several conifera and ferns, bnt no dicotyledons. 'Phis flora is of great interest, as nothing of the kind is known before from Patagonia.

“In the region south from Ijago Argentine, where R. Ilauthal has made detaileil studies of the geology during a long time, piy attention was confined more especially to the plant-bearing layers, and collections were made in several places.”

VII, The Region of the Beagle Channel.

The last link in our chain of excursions was to l»o a visit to the Beagle channel and its neighlwnrhood. 'i’his region has lieen visited plenty of times by scientific expeditions since the discfivery by H.M.S. Brngle; but in spite of this fact, our researches in the. Aniarclic gave good results, and when that ves-sel left Tekenika bay in Novemlter, 1902, it carried a precious load of various materials. 'I'hese, however, were all lost with the ship, and the intentions of our little enterprise were principally to secure new collections and also to solve some fjues- tions aroused by the first-mentioned expe<lition.

With orders from the Chilean Government, which never failed to render us the most valuable assistance, the new chief of the naval station in Punta Arenas, Rear-Admiral F. Valenzuela, put a small steamer, the Porrt'titr, at our disi^osal, and we left that place on ]March 7. At night¬ fall we anchored in Puerto Barrow, and the next day steamed through the Brecknock passage, entering the western extremity of the Beagle channel, when our first station was I’uerto Fortuna. After having visited the unknown Ventisqueros sound, Bahia Ronianche, and the Darwin glacier, we anchored in Lapataia on the 7th. From here Mr. Quensel and I pulled up the river, and thence along the Lake Acigaini to its north-west end, from where we made an excursion on fixit in order to ascertain that the river w’hose sources we diseovore<l in March, 19(i8, south of Sierra Valdivieso, and name<l Rio Rojas, is the same that empties in the lake. Thus here is a pass from Ijake Cami to the Beagle channel. Later, I got to know that the Indians who lived hero l>cforo probably used this pass to come from Almirantazgodown to the channel.

Wo left on the 9th, and passeil the Mtirray Narrows l)etween the islands Navarin and TIoste, visited the small mission station in Douglas

A SlXTEKNTll-CENTrRY MAP OF THE BRITISH ISF.ES.

421

bay, where the lust roiunants of the Yagliaii trilio are eollecteil under the Kuperinteudence of Mr. Williams, who moved his place from Tekeiiika two yeai‘8 ago, and anehored in Allen (iardiner bay. Here Mr. Ualle went ashore and stayed for some days. The Porvenir went to Orange l)ay, where the great Freueh expedition had its lieadcjuarters in 1882- 8.>. The commcmorativu pyramid was found to be intact still. On the 12th we visited the Wollaston islands, and returned at night to I’aek- saddle bay, where also Paek.saddle islaifd was visited.

In Tckenika we picked up Hallo, and went straight to Usliuaia, where we stayed one and a half days for botanical excursions. From there we went to (jiable island for geological investigations, and linally to llarberton and Slogget lay. Paying a second visit to llarbertou, we arrived at Ushuaia the 18th, leaving the following morning.

On our return voyage we visited Puerto Edwards, and landed in Puerto Arenas tm March 21.

In our original plan we had prepared for a longer stay in the Beagle ehanncl, esjiecially for anthropoinetrieal work amongst the Ona Indians. But already, when we came to Buenos Aires in Octol>er, ltt<.'7, wc learnt that the well-known ethnographer. Prof. Lehmann-Nietschc, of Ea Plata, was working with the same subject, and in Ilarbe-rtou Mr. W. Bridges informed mo that an American had visited him and obbiined his valuable observations on the life of those Indians. In conse(iuence, wo resolved to give up this part of our programme.

The expedition returned to Buenos Aires March 31, and left fur Sweden on May 23.

A SIXTEENTH-CENTURY MAP OF THE BRITISH ISLES.

The first Dumber of the current volume of (Jlobm (vol. 92; contains the reproduc¬ tion, with short description by Mr. W. Keinhard, of an early manuscript map of the British Isles, now in the British Museum, which does not apjiear to have met with much notice on the part of writers on British toi(Ograpl;y. The original, which is clearly drawn on vellum and coloured, measures 2 feet 1 inch by 1 foot (5 inches. We arc informed that it was acquired by Sir Robert Cotton early in the seven¬ teenth century, and came to the museum in George the Third’s time.* In the catalogue of manuscript maiis in the British Museum it is a.ssigued to the time of Queen Elizabeth, though ilr. Reinhard is inelintHl to jJace it somewhat earlier. The grounds for this opinion do not seem entirely satisfactory, the priucijial being the inferior accuracy of the outlines of Scotland as compared with those in the map by George Lily (1540), frequently reprinted in Italy during the course of the sixteenth century, and stated by Gough to have been “ the first exact map of this island.” Should his view, however, be correct, the map will ixrssess the

* It would thus se< m to have not been included in the Cotton Library as originally placed in the museum in IT.^S. This may explain the fact that it is not mentioned by Gough in his essay on British ma]>e in the second edition of his ‘ British Topography ’ (1780), though other maps in the Cotton Library are referred to.

422

A SIXTEEXTII-CKNTDKY MA!* OF THE HRITISH ISLES.

interest of being the earliest in which the outlines of Scotland are shown with an ajiproxiniation to the truth ; this result being ])ossibly due, as Mr. lleiuhard suggests, to the circumnavigation of Scotland by King James V'. in 1540.

The map shows other ditferences from Lily's map (or, at least, from the later versions of it), though there is some general resemblance. Ireland, in itarticular, is shown as running almost due north iuid south (as in Nowell's map of the same cen¬ tury reproduced by the Ordnance Survey), instead of being inclined to the meridian as it really is, and as Lily drew it. As regards England, there is little, if any, inferiority, and the amount of detail shown is at first sight remarkable, as is the general accuracy of the information. Besides the chief towns, many of the more imjiortant villages are marked, including some which have remained little more than villages to the present day. We still finil Orwell (placed, however, south of the river on which Ipswich is marked, while Harwich is also niis|)laceel to the north), and Dunwich, though there is no trace of Ravenspur in Yorkshire. Just north of the Humber is Sisterkirk, re'pre'senting the two villages of Witherusea anil Owthorni', the latter of which has now disapiieared. Among minor |ioints of detail, we tind the isles of Axholme in Lincolnshire and Oxncy in Kint (both shown as entirely surrounded by rivers). Broinholme, which figures prominently on the Norfolk coast, reiiresents Broomholm Priory, to be looked for in vain on a moihrn general map of England. Many of the small islands round the coast are markeil, including several which aie, now at least, almost or quite joined to the mainland ( I'hanet, Shei)i«y, Mersea, Foulness, etc.).

In regard to the English section, however, it must be remembcreii that long before the sixteenth century fairly detailed maiis of that kingdom were in exist¬ ence, as is shown by the manuscript map once in the jiosscssion of Gough, and now preserved with his jiapers in the Bodleian at Oxford, which is supiiosed to date from about 1:>U0. Mr. Reinhard mentions this, but makes no attempt at a coin- ]>arison. In several jKiiuts the Cotton map shows traces of the influence of the earlier map or of intermediate examples now h'st, as, c.y., in the general shape of Wales and in the jiobition of the islands round its coasts. A majority, perhajis, of the names and features are common to Iroth, though the later map contains a good many additions and variations. Thus, though both Portland and Selsea ate marked as islands in both ma{s, in the Cotton map the firmer bears the name Corfe.* The most curious jioint of couni‘Ction between the two maj s is as follows : A row of five islands shown in the Cotton map off the coast of Lancashire and Cumberland is at first sight a puzzle, but the explanation is sujiplied by the earlier map, where the names of the rivers of this (lart of England apjtear, in small njuarcs, off their resfiective mouihs. The squares have evidently been mistaken by a cojiyist for islands, and accoidingly in the Cotton map we find that the names given to the islands are in reality (in somewhat corrupted form) those of the rivers Lune, Kent, Leven, Duddon, and Esk.

►Scotland, Wales, and Ireland are naturally shown more roughly than England, and, as Mr. Reinhard says, the data for these may have bi‘en taken from the writings of Bjcthius, Polydor Virgil, and Giraldus Cambreusis. As an example of a name due to the second of these, he quotes the forms employed for Argyll — Argatelia or Argadia — though as these occur in other literature of the time (i.y. in the list of place-names ascribed to the above-mt ntioned George Lily, in Paulus Jovius’ Description of Britain), this alone would not prove a direct borrowing from Polydor Virgil. In Wales (Wallis) Snowdon is prominently marked (as “Snowdone hillis”), and in Scotland the Grampian range is eiiually in evidence.

* In Mercator's luaj) of 15<i4 Corfe is also shown as an island, but in its proper place, while Portland appears in addition.

THE SURVEY AND MAPPING OF NEW AREAS.

423

The mill' may be placetl with lliat of Lily iis an exaiiiple of the greiitest advance in the mapping of Great IJritain l>efore the middle of the t-ixteeuth century,* though falling short of the maps printcil by Mercator and Ortelius, one or two decades later, from originals sent over from England (by Humphrey Lhwyd, in the case of the Ortelius map). The more detailed maiw of Saxton belong, of course, to a new ejioch.

THE SURVEY AND MAPPING OF NEW AREAS, f

By Colonel Sir DUNCAN JOHNSTON, K.C.M.G., C.B., R.E., P.R.G.S., F.G.S.

Aktek referring to the principal geographical achievements of the past years, Sir Duncan Johnston proceeded to say—

The best modem explorers are not now content with exploration or even with a rough route traverse and an occasional observation for latitude ; they either them¬ selves make careful reconnais-sanee surveys of the country adjoining their route or they are accomiianied by trained surveyors, who m;Jie such surveys. Again, every year the area surveyed on correct scientific principles is extended. The intere.-ting address of my prclecessor. Major Hills, will have told you what is being done in this way in the llritish Crown colonies. In the llritish self-governing colonies and in the colonies and deiiendencies of other ixnvers the area of regular survey is Ifing continually extended, and in more remote regions surveys are It-ing carried out by Iktundary Commissions or for railways or other purposes. Along with the increasing appreciation of the value of geography which luis taken place of late years, there has Ix-en an increasing recognition of the need for regular surveys, and it is jirobable that the next generation will find that not only is no ceusideralile area of the Earth’s surface unexplored, but that the area not yet surveyed at least ge igt.iphically, or for which a regular survey has not been projectitl, is getting limited.

I proi»ose in the rest of my address to de.al with the regular survey and ma|)piug of new areas, and to discuss various '[uestions connected therewith; if I am right in Ixdieving that large areas will Ik; regularly surveyed in the near future, such questions merit careful consideration. I shall state on these jioints the practice of some of the great national surveys, Ix-cause their exiierience seems the best guide for future work ; but I recognize that methods suitable for rich and jiopulous countries, such as Germany, Fr.ance, or Great Britain, may be too costly for many countries and provinces whose survey has still to be made, and mention will be m.ade of less exiKUsive methtds which are likely to lx; much in demand in future.

It would be diflicult to say anything new on the subject I propose to deal with, and I lay no claim to do so, still les.s do 1 wish to dogmatize as to the best methods. When I express opinions I shtill also state the practice of .some of the principal surveys of the world, and my hearers, having weighed the matter, can accept my opinions or not according to their judgment. In either case my object will have been attaine«l if careful consideration is given to the is'ints raised.

^laps may l>e roughly divided into three classes : —

(1) Geographic d mai>s — i.e. those on very small scales.

* Another map which bears traces of the use of eoumiun material is that of Sebastian Munster, inserteil both in the Bu.sel rtoleinys of l.olO and later years, and in his ‘ t ‘osiiiography,’ though this is a far rougher production.

t I’resiileutial Address to the tieogniphieul Section of the Britisli .\.ssooiation, Winnipeg, August, 19o'J.

THE SURVEY AND MAIM'ING OF NEW ARP:AS.

4L'4

(2) Tuitographical maps. Tlie dividing line between ttiese and geographical

mai« is not very clearly defined. For the purpose of this address ma{)S between the scales of 4 miles to the inch and » j JO0 ^ considered as topographical.

(3) Cadastral mai« — i.e. majis on large scales mainly for property purposes.

As the lime at my disposal will not admit of my discussing all three classes of mai>s, and as I have on a previous occasion read a paper to this Association on ‘ Cadastral Surveying/ I prui>ose to limit my remarks to to^)ographical surveys and ma|>s.

In most of the older countries to]K>graphical surveys have originally been made to meet military needs/ and as a rule they are carried out under military supervision. In order that they may be useful in case of war such surveys must have been made before war breaks out. The use, however, of topographical maps is not limited to military pur|K)ses ; ou the contrary, they have invariably proved of great value for civil requirements. In one resj)ect they are more useful for civil than military pur¬ poses, as a state of war occurs rarely, and hence while the ma^is are only occasion¬ ally used in counectiou with war, they are constantly used in connection with civil administration and with public and ]irivate business of all kinds. The toiographical maps of the Ordnance Surve}’, prepared originally solely for military requirements, have proved extremely useful for civil puriwses. Directly or indirectly, all the numerous ma]>s prejiared by the trade in Great Britain for civil use are based ou them. I believe the cxjHirience of all other countries is similar to that of the Ordnance Survey. In most countries in which land is of any value, a cadastral survey for laud transfer purixjscs is needed, as well as a tojiographical survey. In some cases, indeed, the ueetl for a property survey has first made itself felt ; thus in the Transvaal and in the Cajie Colony, neither of which yet has a topographical survey, there has for many years been a Government Survey Deptartment for making ])rop)erty surveys. The question arises whether there should be two sepKrrate surveys, one for topographical and one for cadastral map)s, or whether there should be only one survey, the topograpihical map>8 being pirepwred by reducing the aidastral survey. Incidentally the further question arises whether, if two sepiarate surveys are made, they should be under one head.

In most countries — the Ordnance Survey of the United Kingdom being an excep)tion — not only are entirely separate surveys made for these two classes of maps, but these surveys are generally uuder dill'ereut departments. In some cases the cadastral surveys arc isolated farm surveys, showing little detail excepA propierty boundaries. Such surveys would, of course, not answer as a basis for topographical map>s. In other cases, however, the cadastral s\irveys show all necessary detail cxcep>t ground forms, which can be adiled by a septaratc survey. The only cadastral survey, so far as 1 know', which shows ground forms is the Ordnance Survey, whoso U-iiich map® are contoured.

A difliculty in the way of utilizing the cadastral survey for the smaller-scalc maprs arises from the fact that a cadastral survey is, from its large size, much slower than a topxigrapihical survey. It is often found advisable to take up the survey of the former somewhat irregularly, while it is imptortaut for the p>rop)or ptrogress of the latter that it should be taken up regularly and methodically. 'The Ordnance Survey 1-inch map) has, since 1824, not had a separate survey of its own, but has been basol on the cadastral survey. Ordnance Survey exp)erience has shown that the delays in comp)letiug the topiographical map), due to this course, have been much greater than one would have exp)ected, and that there are grave disadvantages in having the scale of survey very much larger than that of the finished map. These objections do not ap)p)ly, or can be overcome, if the cadastral survey of any locality is com¬ pleted before the top)ograpbical map is taken up, This is a condition not likely to

THE SURVEY AND MAPPING OF NFAV AREAS.

425

be often fulfilletl in the case of future topographical surveys. I advocate therefore that, following the general practice, there should be entirely separate topographical and cadastral surveys. I should advocate this even where it is es.senlial to keep the exjjenso as low as possible. More economy would probably result from the adoption of a fairly small scale for the topographical map, from curtailing the small detail to be shown on it, and from showing on the cailastral maps only such detail as is needed for i)roperty purposes, than would result from making one survey do for both classes of maps.

On the other hand, I consider that, even when separate surveys are made for the two classes of maps, it is advantageous that both should Ire made under the same head. The more usual course is, however, to have the two surveys independent, and in some cases local circumstances may make the course I .advocate in.advisalrle.

Trianguhition. — The first preliminary to any survey should be a t i, angulation. It is the most satisfactory course, and the best economy in the long run, to carry out with the greatest accuracy possible the primary triangulation on which the survey is basorl. Such a tri.angulaticn will remain good for a very long i>eriu<l. For example, the primary triangulation of the Ordnance Survey was commenced in 1791; while some doubts have been expressed whether it is accurate enough to comhine with other more recent work for the purpose of investig.ating the figure of the Earth, no one has questioned that even the earliest part of this triangiilation is amply .accurate enough for map-making purposes.

On the other hand, I do not advocate carrying out a primary triangul.ation until .arr.angements have been made for basing a survey on it. In South Africa an excel¬ lent and very accurate primary triangul.ation htis been c.arrie<l otit. This triangula¬ tion was undertaken largely, no doubt, for scientific purposes. While answering its purpose in that resjiect, it has so far had no surveys of .any great extent base<l on it. An accurate triangulation is now a much quicker and less expensive operation than it used to be. The introduction of Invar tapes and wires has largely exi)editcd and simplified the accurate measurement of base lines, while the improvements effected in theodolites enable equal or greater accuracy to be obtained with the comparatively small and handy instruments now made than could be got formerly with large .an<i cumbrous instruments, such as the 3(>-inch theodolites, with which most of the primary triangulation of Great Britain and Ireland was carried out. Unless obser- v.ations are rendered difficult by numerous buildings, by trees, or by a hazy or smoky atmosphere, a good primary triangulation should not now be very expensive. It is usual to base on the primary triangulation a minor triangulation of several orders, the object being to have an accurate framework of trigonometrical prunts on which to kase the suivey. If it is important to keep the expense low, the trigonometrical points may bo rather far apart, intermediate points being fixeil by plane-table; but it should be remembered that it is the truest economy to make the best triangula¬ tion which funds admit t f. In forests or in wooded and rather flat country, where triangulation would bo very expensive, linos of traverse made with every possible accuracy, and starting and closing on trigonometrical points, may be UBe«l instead of minor triangulation.

Detail Survey. — Provided the detail survey is kased on triangulation, it may be made by any recognizeil method. Plane-t.abling is now almost universally resorted to, and is probably as cheap and convenient as any other methotl. The vertical heights of the trigonometrical points will have been fixetl by vertical angles with leferenco to some datum. The height of intermediate points can bo flxal by clino¬ meter lines, especially down spurs and valleys, and even by aneroid, and from these heights the contour-lines c.an be sketched in. Altitudes can be more accurately fixed by spirit-levelling, but this is an expensive method not likely to be much used No. IV. — OcTOBKit, 1909.1 2 0

THE SURVEY AND MAPPING OF NEW' AREAS.

42r,

in the case of topographical surveys. It is possible that in exceptional cases photo¬ graphic surveying maybe resorted to with advantage, and undoubtedly photographic methods sometimes enable work to be done which would not otherwise be feasible. The photographic method suggested by Captain F. V. Thoraiison, B.K.,i8an advance on previous methods. In Canada, I understand that a good deal of photographic surveying has been done, and presumably the conditions in Canada have been found suitable for this method. It has been little used elsewhere.

Sralt of Map. — The next point for consideration is the scale on which the map is to be published, and it is an important one. Speaking generally, the cost increases with the scale, and cost is therefore one of the main determining considerations. The physical and artificial character of the country, the amount of detail it may be decided to show on the map, the method adopted for representing hills and other detail, and the method of reproduction to be used, all affect the question. Clearness and legibility are among the first essentials of a good map, and it is desirable that the scale should be such that all detail it may be decided to show on the map can l>e inserted without overcrowding ; or conversely, if the scale is fixed, the amount of detail and method of showing it should be such as to avoid the common fault of overcrowding the map. In populous countries such as Belgium, France, and Germany, where buildings, roads, railways, &c., are numerous, a larger scale is, emteris desirable, than in less jK>pulous countries. All important detail,

such as roads, railways, canals, forests, woods, &c., should appear on the map, as should the more important names, but it is a matter for consideration how far minor detail, such as orchards, marshes, rough pasture, state of cultivation, etc., should be inserted on the map, and to what extent the less important names should be omitted. In billy country hachures and contours, es])ecially if in black, tend to obscure the detail and names, and the smaller the scale the greater this tendency.

Methods of reproduction will be dealt with later, but I may here say that more detail and names can be shown clearly on a given scale if the map is engraved on copper than if reproduced in any other way. The scales adopted by different countries vary very much — I give below the scales adopted by some of the principal surveys.

i-sJoe scale— Switzerland (the more populous parts), Prussia, Baden, Saxony, Bavaria, and Wiirtemberg (these German mai^s, although called ma}>s of {xHiition, are practically topographical).

4oioD — Belgium and Denmark.

soioo — France (the new topographical map), Algeria, Tunis, Holland,

Japan, Spain, Switzerland (the less (wpulous parts).

rtiioo United States (the more populous parts).

scale (1 inch to a mile) — Great Britain and Ireland, and Canada.

Tsioff Austrian Empire.

Boion ^^® — ^^® France.

lOcAsoo ®®‘^1® — *^® German Empire, Italy, Norway, Portugal, Swollen, and Switzerland (Dufour atlas).

i j.i'ooo ®®*^® — *^^® United States (the less populous parts).

1!!«\500 ®®*^® — Russia.

isjfooo seal®— the United States (barren districts).

The introduction of cycles, motors, and other rapid means of locomotion has led to a demand for a scale which will show a considerable tract of country on a sheet of moderate size. If the standard map is already on rather a large scale, this demand is best met by publishing a reduction of the standard map. This course is followed by Great Britain and Ireland and by Canada, whose 1-inch map is reduced to and published on the J-inch scale; but if only one scale is used a

i

THE SURVEY AND MAPPlNCf OF NEW AREAS. 427

compromise must be arrived at which will meet the reasonable requirements of rapid locomotion, as well as the other essentials of a topographical map.

If I may venture an opinion in a matter in which practice varies so much, it is that for countries using Dritish measures in which, owing to dense jwpulation, the detail is close, the 1-inch scale is a very gotni one,-and that for more

ojKjn iiarts the i-inch scale may with advantage be adopted. For countries using metrical measures, I should advocate and i-j.-.ViOu resjiectively. These scales

do not differ largely from those adopted by most of the principal countries, the majority of whom use scales between and for fairly close countries.

Where it is important to keep the cost down, I should advocate a half-inch to the mile or a xa.’i'ooa scale. All except the most closely populated country can l)e shown clearly on such scales, provided the maps do not show too many names or too much small detail. The United States have scales of •'‘"d

3(?ui5o* general closeness of detail in any area determining which of those tliree scales is adopted. This arrangement is a good one, and would be st'll better if the areas published on the djiod scale were also reiluced to an*l pnblishc*! on the i-Uddd scale, and if the whole country were publishe^l on the scale.

The principle here advocated of having each scale as far as possible complete for the whole country has been carried out by Great Britain, where the whole country, except some uncultivated areas, is published on the 25-inch scale, and the

whole country on the C-inch, the 1-inch, the i-inch, the J-inch, and other smaller scales.

Scale of Field Survey. — It is usual to make the field survey for small-scale maps on a larger scale than that on which the map is to be published with a view to securing greater accuracy of detail, but this should not be overdone. If the field survey is on too large a scale it entails needless expense, also when the surveyor is w’orking on too large a scale he is apt not to realize the cfifect of reduction on his survey, and is likely to survey so much detail as to overcrowd the map, thus increasing the cost of the work and injuring the map. When the map is reproiluced by photographic methods the fair drawing is usually on a larger scale than the finished map, so as to get finer results on reduction ; but in this case also, for somewhat similar reasons to those stated above, there are limits to the amount of reduction which can be made with>advantage. In these respects the practice of different countries varies considerably. In Austria the field survey is on the .js,joo scale; this is reduced to and drawn on the floHea scale, and this drawing is reproduced by heliogravure on the - scale. In France the field survey is on the mJeo or scale. The survey is reduced to and drawn on

the scale. In Algeria and Tunis, both field survey and drawing are the

lOoijo scale. In all cases the French maps are now reproducetl by heliogravure on the joJ^oo ^^® 4 00(50 drawings. In Germany the field survey is on the

•■.-.ooo scjile. This is reduced to the o" which scale the maps are engraved

on copper. In Great Britain the I-inch map is based on the 25-inch and 6-inch survey. These were reduced, and a fair drawing was made on the 2-inch scale in a manner suitable for reduction to the 1-inch scale — i.e. the detail, lettering, etc., were drawn so that when reduced to the 1-inch scale they should be in proper proportion. This drawing was reduced and printed by heliozincography on the 1-inch scale, and from these prints was engraved on copper. In America the field surveys are on the scales of uo^tjo* (^(^d f^® »5*j.\oo» ^^® 12.5^045*

and the ®®*'‘l® niaps respectively. The drawings, on the same scale as the

field survey, are reiluced by photography and engraved on copper.

I considi-r that the best results are obtained when the field survey is made on tlouble the scale of the finished map; that if reproduction is ti> be by engraving,

2 11 2

I

f

428

THE SrRVET AND MAPPING OF NEW AREAS.

the fair drawing should he on the same scale as the finished map ; that if, on the other hand, reproduction is to be by photographic methods, the fair drawing should be on the same scale as the survey, f.e. doable that of the finished map. The reduction I advocate should conduce to accuracy of detail, and, if reproduced photographically, to fineness of detail, while it is not so great that the surveyor and draughtsman should be unable to realize the effect of redaction.

DftaiK — The need of considering the amount of detail, etc., to bo shown is not always sufficiently realized. The way in which detail is to be represented also needs consideration, as on small-scale maps much detail has to be represented conventionally. Railways have to be shown conventionally, and should be so marked that they catch the eye without being too heavy. Roads also should 1m> clearly marked. Where different classes of roads exist they should be distinctively shown, main roads being more prominent than others. It is important to know what roads are fit for fast wheeled traffic in all weathers, and which are fit only for slow traffic. The exact classification of roads must depend on the conditions obtaining in the country, llie most elaborate classification is that shown on the French maps, and next that shown on the maps of Great Britain. Provided that important distinctions are represented, the simpler the classification the better. Forests, woods, marshes, and in some cases pasture, rough pasture, orchards, vineyards, gardens, etc., are shown by conventional signs. While forests, woods, and marshes should certainly be distinguished on the maps, I incline to the opinion that the state of cultivation is better omitted, and that the less small detail shown the better. Such small detail increases the cost and often overcrowds the map. The German ] shows much small detail, and although the maps arc

l)eautifully and delicately engrav&l on copper, the detail is rather crowded on some sheets. The French Carte Vicinale is, in my opinion, rather crowded with names.

The most difficult question, and that on which opinions differ most, is the method of representing ground forms. Methods which answer well on steep ground are less satisfactory on gentle slopes, and vice versa, and each method is open to some objection. Ground forms may be indicated by contours ; hill-shading in stipple, vertical haebures, horizontal hachures, the layer system, or by a combination of some of these. Groimd forms are representeil by contours on the » ,j^(jg-scale maps of the German States, the Swiss Siegfried Atl.as, the maps of the United States, the 1-inch map of Canada, the ^^J^^-scale map of Denmark, and the mai^s of Japan. Where the slopes are steep the contours give almost the effect of hill-shading. Some of these maps give a very good representation of the ground, the l>est l)eing those in which the contours are in colour. Hill features are shown by stipple shading on the French Carte Vicinale and the Ordnance Survey 4-mile map. In mountainous country stipple shading gives a good pictorial representation of the ground, but it fails in flatter country, and it is often difficult to tell from it which way slopes run. The Swiss Dufour Atlas (150*505 scale) is a good example of vertical hachuring, as are some of the German 1 oi^ooo-scale maps. Vertical hachures are also used on the Austrian and Swedish ma])s, and in conjunction with contours on the maps of several other countries. Vertical hachures, when well executed, give an artistic and graphic representation of the hills. In the Swiss and British maps the pictorial effect is enhanced by assuming a light from the left-hand top corner. In steep ground, especially when the hachures are in black, these are apt to oljscure detail and names. I think hachures are better when printed in colour, hut many will disagree with me on this ix)int. Horizontal hachuring, while having some advantages, is less effective and is little used.

The system generally known as the layer system has been used in Great Britain by the well-known Scotch firm of J. Bartholomew & Co., has recently been

1

THE SURVEY AND MAPPING OF NEW AREAS. 429

adopted by the Ordnance Survey for ita ^-inch mapsi, and is used in the .^-inch tiiaps of Canada. It consists in indicating by various shades of colour the area lying between certain contours ; thus one shade may be given to all ground below the oO-feet contour, another shade to ground between the 50 and 100 feet contour, and so on. This system gives a general indication of ground form, and enables the . contour-lines to be followed more easily. Its shades of colour enable the eye to ])ick out mure easily all land lying at about the same level. It is most effective in ground with a small range of veitical height, as the vertical depth of layers can then be small and the distinction in colour between successive layers marked. In hilly ground the depth of the layer must be increased, which means that many ground features are ignored on the map, or the number of layers on the map must be large, in which case the distinction in shade between successive layers will be less marked. This method is popular in Great Britain, and enables those who are not versed in reading contours and hachures to realize something of the nature of the ground forms.

A combination of these methods has been used as follows: France on her sti^inr^le maps shows ground forms by contour-lines and stipple shading. This gives a very fair representation of the ground, but where the contours are very close together the effect of the coloured contours on the stipple is nut pleasant. Nor docs the stipple always look well when it falls on colour. The German coloured ij;,^^,ig-acale map, the Italian British 1-inch show both contours

and vertical hachures. The Norwegian ,Q(^^^^-&cale map shows the features by contours, vertical hachures, and shading. The new British 1-inch scale map has both contours, layers, and stipple shading.

Opinions differ so much on this subject, and there is so much to be said fur and against each method, that I will conBue myself to the opinion that contours reasonably close together should form the principal feature of any method of representing ground forms; that contours by themselves give a very fair repre¬ sentation of the ground ; that vertical hachures, if printed so as not to obscure the detail and names, or stipple shading whtn there is not too much colour on the maps, increase the pictorial effect and are useful additions to contours ; that ground forms should preferably be in colour ; and that where hachures or stipple are used as well as contours, both should be in the same colour. The German coloured ,(y„V,ua-8cale map (brown hachures and contours), the British 1-inch scale copper¬ plate printed map (brown hachures and black contours), the British 1-inch coloured map (brown hachures and red contours), and the French -scale (grey stipple and brown contours), all give a good rt presentation of the ground, and there are other maps which might be name 1 almost, if not quite, as good.

Vertical Interval of Contours. — The vertical interval between contours should de{)end partly on the scale, partly on the steepness of the ground. Practice varies considerably in this matter. The j j^^-scale maps of Switzerland and of Germany, except Prussia, are contoured at 10-metre intervals. The ,7-scale maps of ('ranee are contoured at 10-metre intervals. The -.(j^jj-scale maps of Japan and St)ain are contoured at 20-metre intervals. On the Swiss scale contours are 30 metres apart. On the United States scale the contour interval varies

from 20 to 100 feet. On the British 1-inch map there are contours at 50 feet, at every 100 feet up to 1000 feet, and thence at 250-feet intervals. On the Canadian 1-inch and .1-inch maps the contour interval is only 25 feet, but the sheets published have been in ground with only moderate elevations. On the German ®cale

the contour interval is 50 metres.

I consider that if the contours are printed in colour the vertical interval may with advantage be such that on steep ground the contours are reasonably close together, every fourth or fifth contour being printed heavier so as to be more easily

430

THE SURVEY AND MAITING OF NEW AREAS.

if-

folluw«xi. If the contours are in black they cannot with advantage be so close. It is, in my opinion, best if the contour interval is uniform all over a country. Failing this, it seems des'u^ble that it should l)e uniform over considerable areas, and at least throughout a sheet ; but this view is not universally held. 1 do not like the varying interval adopted by the Ordnance Survey. The contours on the Ordnance Survey maps are surveyed with great accuracy and at great exitensc. Fur topographical ma^js much cheaper and more rapid methods will suffice.

Cartography — I have, with a view to clearness, kept the question of the method of reproduction sejiarate, but it has a bearing on some of the i)oints already considered. Thus the fine engraving of the German ™^l' enables an amount of

small detail and ornament to bs shown on that map which could not have been clearly shown if any other method of reproduction had been used. The older maps were generally engraved on copper, or sometimes on stone, and printed in black and white. Subsequently photographic methods, such as the photogravure of the Austrian and the more recent French maps, were used, and colour

printing is now largely resorted to. In some cases the colour plates are prepared by engraving on copper, stone, or zinc. The maps of the United States and Switzerland are engraved on copper. In other cases — for instance, the 1-inch Ordnance Survey — colour-plates are prepared on stone by transfers and offsets from the engraved copper plate. In other cases — e.g. the France — the colour-pUtes

are prepared by photographic methods. For clearness, delicacy of outline, and artistic eflect nothing equals engraving on copper. It forms also the best basis fur colour-printing. Unfortunately, it is very slow and costly. Engraving on stone is quicker and less expensive than copjwr engraving. It is inferior in delicacy to the latter, but some of the best stone engraving is very good. Photographic methods are the most rapid and the cheapest, and with care give very fair results. As good examples I may quote the . .5^,^-scaIe maps of Austria, prepared by heliogravure, and the 6-iuch maps of the Ordnance Survey, prepared by heliozincograpby, both black and white maj^s. Of colour-printed maps I may instance the new map of France prepared by heliogravure, and the ^-inch Ordnance Survey map hitherto prepared by photo-etching, although I understand that in future the outline will be engraved on copi)er. When rapid reproduction and moderate cost are desired, I do not hesitate to recommend photographic methods, which, although not so good as engraving, give, when carefully executed, reasonably good results.

Opinions differ as to the extent to which colour should be used, the modern tendency being to use it very freely. I can hardly be accused of prejudice against colour, as during my tenure of office at the Ordnance Survey colour-printing was largely developed, but I think it is often overdone. 1 consider that a moderate amount of colour is a great improvement to a map. Ground forms, however indicated, can, in my opinion, be better shown by colour than by black ; it is advantageous also to distinguish water by colour, to give prominence to main roads by colouring them, and to colour woods and forests, but 1 do not advocate going much beyond this. It is difficult to choose colours which are suitable, distinctive, and harmonious, and the more numerous the colours used the greater the difficulty of doing so.

Colour-printing introduces possible sources of error. Colour inaja are based on a drawing on which all detail to appear on the map is shown. A jdate is prejared for each colour on which there should be only such detail as shall be printed in its particular colour. In preparing this plate there is a risk that detail which should appear may be omitted, or that detail be inserted which should be on another plate, or that the detail may be slightly out of ix)sition. Again, owing to change of teiiqieiature and to the varying amount of moisture in the air, {)aper contracts or expands, llegistration can rarely be mathematically correct, and with every care

I

THE SURVEY AND MAl*riN(} OF NEW AREAS.

4:il

may uometimes be apprc<.iably out. While with Ciire errors such as I have indicated can be minimized so as not appreciably to a£fect the map, it is difficult to ensure that they should be altogether absent.

To recapitulate my views, I advocate for a topographical map a scale between rtzioe tZsVsa to a mile), according to circumstances. The sc.ale of

survey to be double that of the finished map; ground forms to be shown by contours