Heinrich Handel-Mazzetti: [A Botanical Pioneer In South West China] - chapter 25


[p.109:] Chapter 25. To the Upper Irrawaddy

The Lisu — a dugout canoe as ferry — climbs in die Tjiontson-lumba — discovery of Taiwania — slow progress —bridge construction —over the Chiangsbel —asleep descent —a hazardous ropebridge —among the Dmng in die Irrawaddy gorge — I fall sick and return over five passes — distant vistas — Gomba-la and Tsukue — subtropical flora on the Nu Jiang (Salween)

There were two routes from Gongshan to the Drung Jiang, neither yet trodden by a European. For the outward journey I chose the first, which did not actually pass through Gongshan, for the return the second. Five of my men from the Lancang Jiang (Mekong) remained with me, and the other ten porters were Tibetans and Nu tribesmen from the Bahan district The guide whom I ultimately found was a somewhat suspect character of sinister appearance with the euphonious and meaningful name of Kru (donkey); he had fled to the Drung Jiang because he owed money to the missionary, and had only recently ventured back after hearing that the latter had been transferred. To make the journey with fewer porters would have been impracticable since the men had to carry enough food to last them for a fortnight; I had already left the paraffin lamp, the camp table and chair and other nonessential items at Cizhong. Pere Ouvrard kindly undertook to record barometer and thermometer readings as a baseline for calculating altitudes on my journey.

After setting off on the morning of 26th June, I sent Lu back from the Alulaka ridge with the horses, for ro one could possibly have ridden down the 1200 m descent to the river itself, and there was no question of taking them any further. He had the task of changing the paper between which I had pressed the plants I had already gathered, and of looking after the larger collections which I hoped to send back in the next two or three days. A few Lisu families from the Lancang Jiang and the lower Nu Jiang had recently settled on the ridge. They were the best hunters in the region; they would track a deer or a bear all day until they caught up with it, and in their own country they shot their poisoned arrows from the bushes at caravans and at any stranger who ventured there without a safe conduct. They were spreading resolutely northwards along the Nu Jiang and seemed likely to displace, exterminate or swallow up the feeble Nu. The village chief and his henchmen had recently absconded, having murdered and robbed four Tibetans on the path to Londjre the year before. Their houses consisted of a single low-roofed room, the floor covered with springy bamboo mats. I bartered salt in exchange for potatoes from them.

Flowering in the remnants of forest along the path was the bizarre Arisaema speciosum, its spadix drawn out into a thread a metre long dangling over the leaves; creeping over rotting stumps was the yellow Lysionotus sulphureus (a new species). Tjionra [note # 121: Kingdon Ward calls it Cho-la. (The Land of the Blue Poppy, p. 189).] on the east and Tjiontson on the west side of the Nu Jiang (Salween) were united by a ferry. Two dugout canoes, each carrying four men and four loads and propelled by two paddlers, took my caravan across. They were so narrow that one could sit with an elbow on each gunwale (Fig.38). The river was nearly at its highest level; its dark brown waters swirled round the boat and made it rock alarmingly. Not long before one of the boats had been swept away by the flood waters. Everything was green. The sandbanks were colonised by grasses spreading by stolons, and the rocks at the banks were covered as far as the highwater mark by pale green liverworts (Marchantia cuneiloba) and maidenhair fem (Adiantum capillus-veneris). One of the houses in Tjiontson offered good accommodation. Some built of stone with timber framing and others of wood alone with a gently sloping gabled roof, the Nu houses of this village were quite imposing.

Debouching at Tjiontson was a major lateral valley from the west-southwest; its Tibetan name was Tjiontson-lumba. My route to the Jiu Jiang passed along it First we climbed 700 m steeply up the arid hillside to a Nu village named Xuelamen-kou; numerous tribesmen had already moved up there for the summer. It was situated on a flat shoulder, one of the few level places in the whole valley. Here I bought a few small musical instruments of a kind common to all the native tribes, among them three carved bamboo plectra in neat bamboo cases, also carved and painted. My servant indignantly showed me an opium poppy he had discovered there.

"Well, so it is, but there's only one", said I dismissively.

"But there are three seed pods on it!" he exclaimed.

It was in fact the Chinese official in Gongshan who had induced the natives to cultivate opium for his own use. But when an opium commissioner came along he of course left them in the lurch and allowed them to be punished. We crossed a side valley where wild walnut trees grew in large numbers and stopped for the night in an empty house. On the opposite side was a stratum of marble which extended along the entire valley-side, in some places forming tall cliffs and elsewhere a narrow vertical edge. Wild bees nested in holes in the marble. Beeswax obtained from them was one of the few taxes which the Nu had to pay to the Naxi tusi in Yezhi. Up to this point the path had been quite tolerable. On the next day, however, we rounded a corner to find that it consisted of steps barely half the width of a man's foot leading along the side of a precipitous slope. To traverse it called for a good head for heights, complete surefootedness and extreme caution, especially as continuous rain had made the clay soil slippery and it was covered with bracken and other tall vegetation. Though carrying loads of 40 kg, the natives walked along barefoot [p.110:] with complete confidence, and my servant found it best to follow their example.

After several steep descents we reached the valley bottom, which contained mixed woodland drenched with moisture and similar to the forest in the Doyon-lumba. On the slopes there were numerous fallen tree trunks which we had to clamber over, a task made easier by notched poles placed where the branches joined the trunk or fitted into small manmade recesses. In many places we had to climb over steep rock shelves, most of them dripping with water, and we had to traverse old rockfalls, crossing boulders as big as houses consisting of extremely coarse-grained granite by balancing on sharp-angled edges or wriggling across a fallen tree trunk with a yawning black space on either side. In some spots there were ladders made from notched logs set almost vertically, but the footholds were too small to give a safe grip for climbing boots. A slip would have meant a long drop and a hard landing.

Twice I came across small trees, quite young and in appearance like the juvenile forms of a scale-leaved juniper, but having three lines on their needles, which were keeled and diverged from the stem in their upper halves only. In these respects they resembled Cryptomeria, and I was inclined to assign them to that genus. Growing nearby was a small conifer with dark pendent branches and fruits at first glance like those of a cypress; it seemed so unusual that I thought it worth struggling up through brambles and boulders to take a photograph, and I sent Lao Li up the tree. To my disappointment he brought me twigs of a spruce Picea complanata (now P.brachytyla) widely grown in Europe, bearing the well known Adelges galls. Before long, however, I saw some enormous fully grown trees undoubtedly belonging to the juvenile form of the first species [note # 122: It was Taiwania cryptomerioides (Handel-Mazzetti, Symbolae Sinicae, Vol.7, 717).]. In their habit of growth they were reminiscent of the American sequoias. Their pale grey trunks, well over 6 m in girth, had smooth bark, though on older trees there was some longitudinal fissuring. They soared arrow-straight into the sky, certainly reaching heights of 70 m, and were comparable in size to the largest pines of the region, though their lower branches spread out horizontally and were so large that any one of them could have stood comparison with a fair-sized tree. In some instances the trunk forked high above the ground and the separate branches towered up like a candelabrum into the mist which hung above these woods nearly all day, saturating the deep humus and the bulky moss cushions on the forest floor. However, I was still unable to identify the trees; they were so tall that I could not reach or even see their flowers or fruits, and no tree climber could have scaled a smooth trunk of such dimensions; nor were there any climbable trees nearby from which the necessary specimens might have been secured.

I pitched camp for the night among thickets of bamboo. Next day we were held up by a stream swollen by incessant rain. As there were no bridges I looked for a spot where the stream divided, waded knee-deep through the first part and crossed the second by balancing on a tree trunk, the usual expedient for bridging small watercourses. The porters were not so lucky: at another spot they tried to find a better way of crossing, but one of them flopped into the water with the blankets belonging to the rest of the men and another managed to slip down, bruising and lacerating his shin quite severely. As I was able to disinfect it at once, the wound healed quite well. Towards evening we reached another stream in spate. This one was so swollen and turbulent that any attempt to cross it was out of the question. We had to go back, pitch camp at 3000 m and wait However, next day the rain continued relentlessly and the water did not subside at all. The main stream was a raging torrent, surging round whole groups of-trees and tumbling down in roaring cascades. Only in-a few stretches did the path come close to the edge of the deep vertical-sided trench which the stream had cut for itself; standing on the edge, one glimpsed the water between bamboos and moss-hung branches, but the view was obscured by swirling clouds of spray.

At this altitude the broad-leaved trees thinned out somewhat — among them were Euonymus ting-ens, its white flowers veined with purple, Malus ombrophila (a new species), Rhododendron sinonuttallii and Styrax shweliensis — and conifers began to appear in the woods, notably the new Tsuga intermedia, together with larches and firs, though the pines dwindled and disappeared. Lilium giganteum flourished in large numbers in the patches of tall herbaceous vegetation, and besides its bulbs we found another useful addition to our diet: the young sprouts of a tall bamboo with stout culms, which was common here and at similar spots in the forests of the Nu Jiang. In Kunming they fetched high prices as a delicacy. They tasted delicious and made a welcome change to my diet, as even my cook, inept as he was, could not spoil them.

Nearly all the time we were soaked to the skin; if the rain ceased for a while the vegetation — two and a half metres tall and dripping wet — drenched us from every side. To avoid catching cold I had to strip to the skin every evening as soon as the tent had been pitched and have my clothes and boots dried in front of the fire. This maltreatment reduced my boots — which were now for the most part of Yunnanese fabrication — to an indescribable state. I should have preferred to throw away the whole lot if replacements had been available, but I could not yet resign myself to going barefoot The plant papers had to be dried sheet by sheet in front of the fire, beneath the tattered pieces of cloth which the Tibetans spread out between the trees for shelter, and all too often the rain fell on them from above while the fire scorched them from below. For kindling the men used splinters of resinous pine which they carried up from the valley with their baggage. They were wonderfully well adapted to the hardships of life in their mountains and the rigours of their weather; they could make themselves comfortable anywhere and they never grumbled. Even when the rain was pouring down they sang songs and called out jokes [p.111:] from one shelter to the next They soon picked up my servant's favourite expression, a word which was constantly on his lips instead of the usual indescribably filthy Chinese oaths. It sounded like "lungtent-io",[note # : Perhaps (allowing for dialect pronunciation and errors of transcription) "longtan huxue" meaning "dragon's pool and tiger's cave" — a dangerous place. A Chinese would have found the Nu Jriang area unfamiliar and threatening (S.G.Haw)] but I was never able to find out what it meant The/ repeated it from dawn till dusk, and it echoed back and forth between them, and he very soon stopped using it However, they were sometimes careless in handling unfamiliar European objects, and one of the porters continually leant on my pith helmet which in cloudy weather was placed on top of his load, and soon squashed it into a shapeless mass.

The valley had the U-shaped profile typical of glaciated terrain; its floor was covered by vegetation and its sides were almost vertical, in some places extending upwards in a staircase of rock slabs — mostly dark in hue, but here and there sparkling with water from rain or melting snow — and culminating on the south side in Shatsakon, a summit apparently 5000 m in altitude. We had to wait until 2 July before the stream, which splashed down the valley side in splendid waterfalls, was low enough to cross. Gleidovia ruborum, a short-stalked, branching plant of the broomrape family previously found only in the northwest Himalaya, peeped out from beneath a rotting tree trunk, forming clusters which looked like birds' nests. Otherwise the number of new plants was not yet sufficient to be worth sending back a load. Towards noon we had to cross the main stream. An uprooted spruce lengthened with a sapling bridged it almost completely, and we continued up the south side of the valley. Magnolia tsarongensis, its long narrow leaves covered with brown felt on the undersides, formed what amounted almost to a tangle-wool. Its broad-cupped snow-white flowers were opening among the rhododendrons of the forest floor, some of which were quite sizeable little trees, in particular R ixeuticum, whose leaves, also covered with tawny felt beneath, gave large stretches of the forest a brownish tint. Beneath them grew the tall Arisaema vetmcosum with warty hairy steins and leaf stalks. A lateral stream gushing out through a rock cleft proved too deep to wade. A young fir was felled, split in half and laid across at a narrow spot; my nope was stretched out to make a handrail, and in a quarter of an hour we had all reached the further bank There were leeches here, and as we had no chance of choosing our footholds and handholds some of the men were bitten. Yet again the stream was impassable, so I called it a day, pitched the tent on a boggy patch, collected the liverworts growing in profusion on the boulders and tried to work out how long it would take the men, working with their small cleavers, to fell a fir big enough to reach across the stream. The tree which they cut down that day fell on to a trunk lying in the water, broke into pieces and was lost Next morning, however, after an hour's work a second fir was felled and made a splerdid bridge. Using my rope as a handrail once more, we were soon across, and having struggled over fallen tree trunks and through thick vegetation, in which I found the tall Cathcartia smithiana (a new species) and Corydalis pterygopetala, we found that the stream was now so calm and shallow that we could easily have waded it at that spot on the day before. Further on there was a delightful waterfall, past which we climbed steeply, using a rivulet in the cane-brake as our path. Beneath a boulder lay the body of a man — not yet completely reduced to a skeleton — who had frozen to death here in the winter, his pack beside him and his dog lying at his head. Every year a few people die from cold and exposure in these mountains, and even more bleed to death after stepping on the razor-sharp end of a bamboo culm, chopped off obliquely near the ground.

Soon we came to the first of the high alpine plants in a grass-grown avalanche channel: there were primulas (P. dickieana and Omphalogramma elegans), orchids (Pogonia yunnanensis and Pleione scopuloruni), edelweiss (Leontopodium jacotianum), the tiny pink Utricularia salwinensis, Vaccinium modestum with solitary pendent flowers on long stalks, gentians and others; they made a pleasant change. In this stretch the south side of the valley was less steep, and its lateral valleys entered it between rounded forest-covered spurs, descending at a gentle slope and evidently beginning a long way off. At its last or more correctly first fork we crossed the northern tributary of the main stream and camped on the tongue of land between them — the last camp on this side of the divide, reached after eight days' travel averaging barely over 6 km a day. There was not a patch of level ground anywhere, and the tent looked lopsided and askew, although the men had hacked away the earth on the uphill side and built up the downhill side with stones and brushwood. There were some thick roots which defied their efforts and made the campbed tilt even further.

On 4th July it was still raining as I at last went up along the southern tributary towards the pass. The little valley was a veritable garden of flowers. Nomocharis aperta, Doronicum altaicum and many other species already known to science grew in colourful throngs above the fresh green of the grasses and sedges on the banks of the stream, which is some stretches offered the only practicable route. Primula agleniana, its large bell-shaped pale crimson flowers emitting a delightful perfume of vanilla (Fig.41), flourished in profusion beneath the tangle-wood, which here consisted of a cherry, the new Cerasus mugus (Prunus mugus), together with Sorbus poteriifolia, Rhododendron saluenense and the yellow R. chryseum. The sky was gloomy and overcast; clouds drifted between the mica-schist towers of the crest which still separated me from the goal I had dreamt of so long, and hung above the snowfields, along which we ascended through gulleys and conies. Flowering at the end of the snow were the first-comers of spring at that altitude: the new Corydalis polyphylla, Caltha scaposa, some splendid louseworts and Omphalogramma minus (apparently a dwarf form of O. souliei). On compact turf I found the small Primula valentiniana (also a [p.112:] new species) with bell-shaped flowers of deepest purple-red, together with Rhododendron campylogy-num, a dwarf creeping shrublet with solitary long-stalked pendent flowers of a colour which can only be described as black. New for China was another discovery — Diapensia himalaica, growing on boulders in small cushions with flowers only 8 mm in diameter. Though the Chiangshel pass was only 4075 m high it was noticeably above the treeline and on its western side there were other finds to keep me busy, such as the dainty Meconopsis lyrata and the dwarf shrub Diplarche multiffora. The porters had gone ahead, and in the mist my collectors and I wandered from the track and started to go straight downhill. Fortunately the absence of any trace of a path soon aroused my suspicions and vigorous blasts on my whistle brought the guide back to us; otherwise we would have strayed into trackless primeval forests. In fact the path curved sharply to the right and led further uphill on to a crest largely covered with dwarf cherries and Viburnum cordifolium var. hypsophilum, though it soon descended again to the uppermost firs. Between them were patches of turf and creeping among the grasses was a rhododendron — probably R. repens once more; sitting there on the ground, its flowers seemed quite extraordinary: they looked almost like purple gentians. We went on for some distance, continually clambering up and down between piles of granite rubble but not losing or gaining much height, and finally pitched camp beside a pool at the western extremity of the crest. Next morning we began the craziest part of the whole journey — a dead straight descent for more than 1600 m down a natural stairway of rock steps, roots, tree stumps and mud holes through a jungle of bamboo. Except in one short stretch where it ran obliquely, the track led straight down the mountainside. In many places we had to face inwards and though there was usually a curved root or a bamboo culm to cling to there was sometimes no handhold at all. In some spots we gazed almost vertically down from the top of a rock face, through the gaps between the roots and boulders on which we had to place our feet, on to the tree-tops below. From 3400 m downwards there was a larch forest with only a few firs, and lower down mixed forest similar to that in the Nu Jiang. The tree-trunks were covered with epiphytes; besides the shrubs we had previously seen I found Gaultheria nummularioides forming large woody clumps on trunks and boulders, and several orchids, including the new Coelogyne taronensis with small numbers of large green flowers marked with rust-coloured spots which made them look as if they had been singed. All these were growing high up in the branches among the lichens and golden-green mosses (notably Plagiochila species and Pseudospiridentopsis horrida) which formed thick mats over every piece of wood and dangled down in long strands even from the leaves themselves. At last we came out into a small clearing and saw that we had nearly reached the valley bottom. However, it was only a lateral valley coming down from the north and curving round to the west. Its name was Naiwanglong. At the edge of the clearing was an unusual pine, the Indian Pinus excelsa, a huge tree with exceptionally slender needles hanging down as if in mourning and narrow cones over 20 cm long, resembling those of the Weymouth pine. Its rounded, bluish green crowns stood out conspicuously torn the surrounding forest We clambered down another steep descent into the gorge itself. Two large :ems grew there, looking oddly out of place. One was Gleichenia glauca with arching fronds several metres in length, branching and forking several times before terminating in finely pinnate arborizations. The other, Dipteria conjugata, was erect and looked like an enormously enlarged Ginkgo leaf with its Made deeply cleft in the midline, and tiny heaps 1-up sporangia thinly scattered on its back. Both species would normally be regarded as tropical rather than subtropical.

Rope bridge at Naiwanglong [Handel-Mazzetti's sketch]

The porters had hurried on ahead, and suddenly I found myself standing among them on a rock platform above the raging torrent. If anyone says that he does not know the meaning of fear I should like to despatch him to that very spot to learn what it is, under the same conditions as those which I encountered. It was a gloomy ravine, filled with spray from the waterfall which crashed down it Dangling straight across the water were three bamboo ropes, each no thicker than a man's finger, some distance above one another. We had to rely on them to get across. There was no means of telling whether they were rotten, and for extra safety I added my rope to them. The natives launched themselves across without an instant's hesitation; as evening was drawing near they used the opportunity, the mission caretaker at their head, to extort a few rupees for expediting the task. My men, however, gazed apprehensively at the ropes and at me. Not until the weight of the wata (the wooden slider) was applied to them did the ropes come together into a reasonably compact strand or cable. As my own rope was [p.113:] somewhat shorter than the others it was impossible to sxure it without tying a knot and all it did was to impede us on starting. Resting one foot on the stump of i small tree, one pushed off from the bank and slid across on the wata as far as the halfway point, but from there on one had to haul oneself up hand ove' hand. However, the men pulled me up faster than I could manage and my hands sustained a few graies. The trees to which the bridge was fastened had to be supported by men pulling on short ropes, otherwise the weight would have uprooted them and the entire bridge would have collapsed into the water. The scene will always remain imprinted on my memory and I was sorry not to be able to photograph it; however, the ravine was far too dark for a snapshot and the movements of the branches and the ropes were too lively for a time exposure; in any case the view was obscured by twigs and foliage in the foreground. At a drier season the scene might havs been less daunting and the river might have beei easily waded, but such considerations did nothing to alter the situation as we faced it on that day. As we had only one wata — and that only by lucty chance — the work proceeded very slowly, and conmunication was hindered by the roaring of the torrent, which drowned our voices although the banks were only 25 metres apart.

After a short but extremely slippery and exposed scranble I pitched the tent on a small spur projecting from the mountainside. I spent the next morning there, putting the collections in order, and that afternoon proceeded along the stream almost as far as ihe Drung Jiang. On the mountainside there were patches of bracken, and once again I found birches (Betula cylindrostachya) and Pinus excelsa, their twigs and needles festooned with orchids (Bulbophy-Uusrfi) which twined among them like a tropical Tilandsia. Unfortunately nearly all of them, like the maiy other orchids which grew there, had already finished flowering. The only species still unfaded was Galeola lindleyana, an orchid devoid of leaves anc chlorophyll. Its eggyolk yellow flowers, never mote than half open, projected above the decaying undergrowth, through which rambled Rubus chrysebotrys, a new species of bramble with golden berdes. Not until we had dug up the orchid did it disclose its full height of 2Yi m, not to mention its creeping stolons a metre long. Beside the river there was an ancient rockfall where the incessant rise and fall of the water had precluded any accumulation of smtll debris, and there was nothing except a little sard between the granite boulders, some of them as large as houses. We squeezed along pitchblack passages between and beneath them. Scrambling up them was Rhaphidophora deeursiva, an aroid with thick creeping stems and dark green leathery leaves cleft into irregular pinnate divisions.

When I arrived, the iceaxe in my hand, at the first Drung house in the little hamlet of Nicheluang the people were at first so timid that they hid from me, but the mission servant, who had previously made friends with them, managed to calm their fears. Was there anything for sale?

"They've just barred the chicken coops with crisscross sticks", said my cook, showing me what he meant, "but otherwise they're all right. They haven't any weapons or even any clothes on."

That night was marked by the onset of an illness which seriously hindered my work. Whether it was due to the water from the forest streams which I had drunk unboiled, the Japanese canned meat acting directly on the intestine, or simply a chill from exposure to rain and wind, the diarrhoea which I had already experienced now became unbearable. Furthermore, the house was so densely populated with fleas that a good night's rest was out of the question, and lack of sleep added to my feelings of weakness. The three kilometre stage to the next village upstream, Shuche, was almost a day's march, so severe were the ascents and descents. There is probably nowhere else in the world where the tracks which link neighbouring villages are such miserable breakneck scrambles. As Burrard says in the notes to his new map of TibeC of all the parts of the Irrawaddy district surveyed by his men, it was here on the Drung Jiang that the difficulties of the route were at their worst From a comer 275 m above Nicheluang there was a good view of the valley. Its walls were even steeper and less dissected than those of the Nu Jiang valley. As its channel was more or less straight, one could see a long distance upstream to the north west and downstream to the south. The river itself was about half the size of the Nu Jiang, with numerous rapids and some minor bends round small projecting spurs. According to the Prince d'Orleans, who travelled to it from Yuragan, it makes a major bend to the west at Turong 30 km to the south. However, as my view down the valley extended for a distance which, at my estimate, must have been almost 50 km, I felt obliged to ask whether — contrary to all appearances — a major west-east channel might not really constitute the river valley. However, this has been categorically denied and was in fact subsequently disproved by a better view from the mountain. According to the map drawn by Roux (d'Orleans' topographer), the Prince's further route passed close to that channel and when I was drafting my survey I formed the impression that the error which undoubtedly exists in his map arose from the circumstance that he never got a view of the course of the river as a whole and sketched the bend as too small. Remembering that Burrard's map of Tibet, which is notoriously inexact and, where it covers the adjacent parts of Yunnan to the east of the boundary, is extremely sketchy and full of errors, shows the bend as extending very much more broadly, it seems to me highly probable that the river actually makes a complete reversal, in the form of a huge S-shaped loop. The terrain being what it is, such an assumption is by no means absurd, and on the relevant sheet of the latest Indian survey (not for sale) there are indications of just such a bend [note # 124: In fact the Irrawaddy — this part is now known as the Drung Jiang — does not continue in a straight line (or as far as Handel-Mazzetti believed. After running south for some 20 km it curves to the west and crosses the Burmese frontier (see Operational Navigation Chart [ONC] H-10).] The mountain range on the opposite side reached an average [p.114:] altitude of 4000 m, and an impressive peak further upstream, between the fifth and sixth lateral valleys, was considerably higher, though even that was devoid of glaciers. The whole range consisted of dark igneous rock. Below a crest across the valley a detached portion of blue sky peeped through an enormous hole in the rock.

We now came to a deep ravine where the men who had been sent ahead had bridged the tributary stream with branches and bamboos. Here I hoped to learn something of the plants growing down at river level at 1650 m. Gleichenia glauca hung down over the rocks and there was a Brassaiopsis (probably B. hookeri) with inflorescences at least a metre in length, but once again it was out of reach. Then we climbed up almost vertical ladders made from notched logs for a distance of some 50 m to the village of Shuche, which consisted of two groups of houses, the upper group at an altitude of 2025 m. There was no longer any trace of the woods of Yunnan pine and of oak which exist on the Nu Jiang or of the maquis of sclerophyllous shrubs; this valley was much wetter than the Nu Jiang. Nor was there any sign of the palm woods which Genestier had seen further down the valley. Yet I found trees and shrubs belonging to the subtropical flora of the Sikkim Himalaya, among them Bucklandia popuJnea, Neillia thyrsiflora, Pentapyxis stipulate and the twining Trichosanthes palmata, all of which were new to the flora of China. As defined by the frontier line, though it was totally unnatural and had not yet been substantiated by any treaty, this part of the Irrawaddy valley belonged to China, and the people paid taxes in kind to the tusi of Yezhi. They told me that beyond the mountains there was English territory, and there was a good track. From time to time a minor official, who though Chinese was subordinate to the Yezhi tusi, was sent here to collect taxes, but the year before he had arrived too late; the Tibetans from Tsarong, who were alleged to carry their raids for booty and slaves much further down the valley, had appropriated them. Burrard depicts the entire valley including Ridong, a Tibetan settlement much further upstream, as part of Burma.

I had my bed set up under the projecting roof of a house in the upper part of Shuche, on a platform just over a metre in width, and decided to take a rest day. The men had carelessly neglected to bring enough provisions and on the previous day had sent Kru to another village to buy meal. The delay was not unwelcome, for besides the diarrhoea I was troubled by soreness in the thigh muscles after the long descent. Kru came back that evening without having bought any corn; that had probably not been his intention at all — he simply wanted to visit some old friends and give the others a rest day. I therefore consulted the village headman, who sold us millet and a pig — at a price! Two men dressed only in loin cloths of loosely woven fabric rolled up into a narrow strip pounded the grain with a wooden pestle made from a piece of tree trunk in a hollow tree-stump set up on the "veranda" of the house; while engaged in this strenuous labour they chattered continuously, repeating over and over again two scraps of Chinese which they had just picked up from my men. The Drung people were at a far lower level of civilisation even than the Nu. They had dark hair hanging thickly over their foreheads. By no means all of them were of short stature, and they displayed a wide range of facial shapes, from high and narrow to low and round. The women wore somewhat more clothing than the men and dressed their hair in exactly the same style. Their faces were tattooed with blue squares covering the area between the eye and the corner of the mouth, and also on the forehead and chin. Their speech was allegedly a dialect of the Nu language, and not greatly different from it [note # 125: There are great differences between different dialects of the Nu language; its exact affinities are still unclear. But the Nu dialect of the Gongshan region is close to the Drung language. The two are almost mutually intelligible. Many Nu can also speak the Lisu language (S.Haw).] Their houses were small, built of wood on low piles and covered with a gable roof thatched with reeds. The single room had a small door at one end opening on to a narrow platform hi front, and an even smaller window at the other end. Ox and yak skulls were hung up round the doors, although the tribesmen did not themselves keep cattle of any kind, and in the interior were numerous skulls of smaller animals killed by their hunters. Maize, buckwheat, beans, yams, millet, a little wheat, and tobacco were their crops. They tilled their fields with a wooden implement like an axe but without any blade or point of iron; it was simply a forked branch. They hunted with crossbows and arrows like all the indigenous peoples of the region. The weather now cleared, and I was able to measure out a baseline, survey the features and take photographs. As is clear from Burrard's reports, the terrain had previously been surveyed as far as a village named Naktai some distance further north. However, if his map represents the entire outcome of these labours, it seems fair to say that my efforts were by no means superfluous.

On 9th July I set off along a somewhat better track on the side of the next lateral valley where the men had built me a bridge beforehand, and then northwards up the valley itself. In the mixed forests there were three new woody plants, Ficus Glicauda with a long drip-tip, Pentapterygium interdiction, an epiphyte related to our bilberry, and Hydrangea taronensis, a low growing subshrub with blue flowers. On the steep slope below the track there were pure stands of sombre Tsuga. We pitched camp by the stream and next morning we had to wade il immediately after starting; but for this we might have stayed dry all day. A steep ascent up to the right brought us to the pass, which had the typical Nu name of Pangblanglong. From there I had a good view south westwards beyond the Drung Jiang to ai range which displayed broad rounded ridges above the headstreams of a fair-sized lateral valley. Then, joining it to the north, on a bearing of 228 , I saw dark spires with little glaciers between them. I was granted only a brief glimpse of these, just long enough to take bearings on the district as a whole, before they were again hidden in cloud. Beyond [p.115:] Pangblanglong the route ran down a side valley of the Naiwanglong, deep down in which the remnants of avalanches were still lying. We had to stay high up and reached the crest itself once more near a nodal point between the valley in which we had ascended, the Naiwanglong valley with its numerous branches, and another valley which ran northwards and then bent sharply westwards towards the Irrawa-ddy. The structure of this part of the range is extremely confusing, and erosion by the abundant rainfall has obscured its original simplicity. The river Naiwanglong has cut into the main range, which runs almost exactly northwest to southeast, and it was not entirely clear why the whole group now appears to be made up of several chains running due north and south, terminating 8 km northeast of Pangblanglong in a transverse block which forms part of the watershed between the Nu Jiang and the Irrawaddy.

It was a stiff climb up to the ridge, and no less steep on the west side beneath its crest up to a third col, which shared the name Bushao with the one before it We then proceeded on flatter ground across a broad cirque situated at the foot of a chain of rock towers, becoming taller as they continued northwards to the watershed. We marched steadily at an altitude of 4100 m above the firwoods, which, grievously ftjirowed by avalanche tracks, covered the upper slopes of the valleys. The white-flowered Potentilla brachystemon (Sibbaldia perpusilloides), a most peculiar plant, probably the smallest of its genus, formed low mats on the boulders, as did Leontopodi-um muscoides, an edelweiss with very narrow stellate bracts. Draba granitica was slender and had small flowers; Primula genesteriana, like all these a new species, is related to our birds-eye primrose but was several times smaller, and the eggyolk yellow Draba involucrata grew in masses on the scree slopes. At the point where the transverse ridge joined the jagged crest we crossed the latter by a pass named Shualuo, the fourth of that day's march, and then went along the top of the crest towards the Gomba-la. Late that evening we pitched the tent on a small patch of level ground above the valley which ran northwards to the Nu Jiang. Fearing that he would not be able to find us, I had to send some of the men back in the dark with my lantern to look for one of the porters who, feeling unwell, had sat down beside the track. Up there, above the treeline, we were able to cover greater distances in each day's march than on the outward journey. The track and the weather were also better.

Next morning was wonderful. Mount Gomba-la, an enormous dome with a small glacier in a dark ravine, towered above us into the deep blue sky. Crowned by serrated peaks, the chain to which it belonged plunged down in vertical precipices on our right into the valley running northwards and debouching into the Nu Jiang in an inaccessible gorge near the Tibetan frontier. To the left there was another rugged rock peak, but straight ahead, above the intersections of the valley, was a line of gigantic snow peaks gleaming in the sunshine far beyond the Nu Jiang — the summits of the Lancang Jiang-Nu Jiang divide to the north of the Doker-la. They were

mantled in fresh snow extending far down their lower slopes, and silhouetted against it were dark fir trees looking like tiny dots. The peaks themselves, incredibly steep pyramids and towers, each one of them a colossus, towered up to heights certainly over 6000 m [note #126: Handel-Mazzetti's map shows Kakerbo (ca. 6000 m) to the east of the main line of the Mekong-Salween divide. The higher peak further north is not shown, but according to ONC H-10 it reaches 6810 m. The main peak is called Meili Xueshan in Chinese, and its height is given on Chinese maps as 6740 m (SGH).] Kakerbo projects towards the Lancang Jiang and was not visible from there, but I consider that it is only an outlying peak of these huge mountains, the highest I have ever seen. This interpretation is supported by Gebauer's observations and photographs taken from the mountains near Deqen (Atendse). A foreground to this superb alpine scene was provided by Pegaeophytan sinense, a plant with solitary bluish white flowers growing here and there in patches of sandy mud and among flat pieces of slate in spring-fed bogs (Fig.43). It is one of the largest of its family (Gw/atee) and its clumps, sprouting from carrot-shaped roots, covered considerable stretches. By stopping down the diaphragm to a barely visible pinpoint, I took a photograph which showed the plant and the distant peaks, both in sharp focus. The fifth pass, Tsukue, at 4175 m the highest in this part of the route, led over the Gomba-la crest just to the east of its junction with the main ridge on the same side of the Naiwanglong that we had skirted- on the day before, and then turned southwards into a side valley leading down to the Nu Jiang. From the top of the pass one can trace several narrow vertical bands of light grey marble which run exactly north westwards through the bedrock. They bear witness to the original structure of the mountain, though the present line of the highest crests does not correspond to it at all. In front of us, below the treeline, was a lake bordered on the right by a green meadow, and to our left the screes of the Gomba-la, intersected by strips of turf and bands of rock, plunged steeply down from the windgap below its twin summits. The gap would probably have been within reach and the high level circuit would certainly have been uncommonly rewarding, but I could only have tackled it if I had been in good health; furthermore, supplies were running short and my boots were worn out.

I therefore felt obliged to drop down to the lake, and in its meadow found some plants of the daisy family not yet in flower, including the stemless Jurinea salwinensis (a new species), Lactuca amoena, similar but with blue flowers, and Leontopodium himalayanum. We went on into the main valley which led northeast for some 18 km to the Nu Jiang. Down to the right we saw the gleaming water of some small bog pools among the forest on a flat shoulder. They were arranged in a pattern which I often encountered and which reminded me of the sinter basins at Bede, although they were certainly not formed in the same way. The track led to the left over a spur on the valleyside; from the top it offered [p.116:] brief glimpses through the trees of the tangle of summits in the direction of the Doker-la, without giving a satisfactory panorama. There were some more scrambles, and though they were not very long, in my weakened state I found them quite exhausting. After crossing rock slabs scoured by avalanches, the path finally reached the valley bottom at a point where the entire valleyside on the left, formed from slabs of dark granite, was completely denuded. Elsewhere the slabs were clothed with trees, but here their cover consisted only of shrubs pliant enough to survive the unceasing onslaught of snow and ice which sweeps down every winter from the glaciers gleaming high above us on the Gomba-la. That evening there was no place large enough for the tent, and I had to set up my camp bed under the shelter of a huge boulder in the forest However, the rain soon began again, and it offered scant protection against the water which ran down its face and the overhanging roots and dripped on to the bed. In the forest I noticed a bright red toadstool. Because of the difficulty of preserving such fungi I did not usually trouble to collect them, but on this occasion I told my collector to fetch it, remarking that it would not have any roots. However, he found that it actually had a large and troublesome root, and on looking' more closely I recognised it as Balanophora involu-crata, a parasitic flowering plant, which I later encountered quite frequently in similar places in the forests of the Nu Jiang. The valley curved gradually to the right, remaining narrow and steep walled. Again and again we thought its end was in sight, but what we saw was only the crests of the lateral slopes, while new bends came into view further ahead. The puzzling conifer which I had seen on the ascent from Tjiontson reappeared here, at the same altitude as before. Once again the trees were gigantic and inaccessible, and I could not make out any detail at their tops. Evening was coming on and we had no time for any serious attempt to climb them. But I had decided on my plan. I was not going to depart without identifiable material from those conifers. As the hign alpine flora on the Nu Jiang-Irrawaddy divide was still in its vernal condition, I should in any case have to send my plant collectors back there in late summer, and then they would have time to get some cones for me. That evening I reached Niualo — as its name indicates, a Lisu village, and indeed the most northerly of them — and enjoyed a hospitable reception.

From there it was not much further to the Nu Jiang. As we descended we entered subtropical rain forest just below 2200 m. One of its commonest trees was Sloanea fotrestii, which resembles our beech in its foliage and the bristly spines on its fruit. Another huge tree was Schima khasiana. Hanging from the tips of the highest twigs were the long racemes of Dendrobium devonianum with deep yellow flowers 8 cm in diameter. Once again there was shrub vegetation with Saurania napaulensis and the tall Rhamnus henryi with soft foliage. Then a stretch on bare rock brought us to the slopes of the Nu Jiang valley itself. After a short descent through arid terrain with a few pines we reached Xiqitong, a scattered village 3 km north of Gongshan. The Nu Jiang flows down from Wuli in the northeast, breaking through a bed of crystalline limestone. A short distance further on, in the Gongshan gorge, it turns to the east, but very soon corrects this deviation and resumes its general trend from NNE to SSW. This hard rock formed a steep precipice everywhere, and in the lower gorge it rose up into huge pillars 600 m high, one of which had necessitated the building of a high wall or causeway in the river itself to carry the path. When the river rises, however, it floods this stretch and the only route to Gongshan is high up, over the ridge. Although there were still a few xerophytes on the sandbank deposited along the river below Xiqitong, among them Schefflera delavayi, here with a dense layer of brownish felt, the gorge itself was filled with luxuriant subtropical vegetation. Enormous lianas, the new Mucuna coriocarpa with thick trunks and seedpods 50 cm long, climbed high up into the Sloanea trees, the soft Rhaphidophora peepla moulded its juicy green stems closely against the treetrunks, bearing flower spikes sheathed in white high up among its elliptical leaves. Agapetes lacei had produced its strange swollen woody stem bases, and among them flourished the unmistakably tropical Asplenium nidus, a fem which forms large nests of tongue-shaped fronds 70 cm in length. Among essentially xerophilic plants were numerous small epiphytic orchids, though unfortunately they had all finished flowering, and the palm Trachycarpus martiana, on the rock faces across the river, where it was almost inaccessible, it had grown into trees with substantial trunks. Gongshan is situated on a flat talus fan which forces the Nu Jiang to make a smooth curve round the eastern side of the mountain. I took a shortcut across it below the main village, as I was in haste and felt no need to call on the resident official; at the time I did not know that he had previously gone mad and died from overindulgence in opium and schnapps, which he drank in quantities of eight to ten rice bowlfuls daily. I stopped for the night at Dara, a village up on the hillside, populated mainly by Tibetans. I received maize cobs as a gift from my hosts, but my porters did not have much success in their search for provisions. When I was mustering them next morning my servant suggested that I should blow my whistle, because there was so much shouting in the village. Before I could stop him he seized the whistle, which was hanging round my neck, and put it into his mouth — yet another example of Chinese disregard for hygiene! At Tjionra, where I crossed to the east bank, there was a soldier who had just arrived from Deqen. I was afraid that the authorities might be looking for me, but it transpired that he had been sent there because of the death of the official, and that he was in fact a good friend of my men from the Lijiang district My horse had been sent there to meet me on my return. After that long march I was so delighted to see it that I could gladly have hugged it, shaggy and filthy as it was. On 14th July I was back at Bahan, and was at last able to concentrate on restoring my inside to health.

[chapter 26:]