Heinrich Handel-Mazzetti: [A Botanical Pioneer In South West China] - chapter 16


Chapter 16. The Land of Muli

The Chesha pass —the monastery town of Muli —a distrustful lama —interior of the temple —prayer wheels —up Mount Saganai in the rain — tales of bandits — the Li Qu valley — three days' march at over 4000m — ascent of Gonshiga (4750m) — Xifan people along die Shuiluo He — die Yangzi bend

The friendly relations between the lama and the official were purely superficial, and it now suited the lama to help me in my efforts to reach Muli. In the meanwhile he had sent ahead a messenger with a letter hi Tibetan, and he said mat the official would not be able to stop me.

"He is really just a child, and if he still raises objection) I shall simply shout at him, and then he will certiinly give way", said I, to the lama's approval.

Ths inhabitants were by no means friendly, and the mail who had stayed in the town complained that he was unable to purchase anything; wherever he went, as soon as the people saw him coming with his sack they slammed their doors. Not until the lama had provided a man to accompany my servant was he able to buy the necessary food supplies. On the day before I planned to depart the official sent a soldier with an invitation to dinner for the next evening.

"What does he think he's playing at?" Lao Li said to him, "that's not the way we do things. When a European says he's departing tomorrow, that's just what he will do, and you can forget about your invitation to dinner. What I want is a letter for Zhongdian. The lama is already fixing things for Muli."

This produced the desired effect, and a soldier duly turned up on the morning of 23rd July. The abbot still insisted on being photographed in various costumes, but when the proposed subject did not appeal to me I simply left the cover of the slide-holder in place. However, I took pictures of his eighty year old father dressed as the living Buddha and of a mendicant lama with his yellow hat and praying kit, also a prayer wheel and a drum covered with human skin. Then I departed with my caravan, which had been augmented by one of the abbot's henchmen, a man who spoke Chinese, Tibetan, Naxi, Xifan [note # 83: Xifan = Pumi (SGH)], and Liidi, the last being totally different dialects of the Naxi language [note # 84: The Naxi language is related to the language of the Yi, and belongs to the Tibeto-Burman section of the Sino-Tibetan group. It has two major dialects (SGH).].

The route led northwards across Naxi territory, through green valleys, over low ridges and between countless dolines, many of them concealed in sombre woods of Pinus tabulaeformis interspersed with holly-leaved oak scrub. Once beyond the Yongning basin it continued straight through into the province of Sichuan as far as Wujiao, the first Pumi village, where I set up my campbed under the eaves of a house, as the interior rooms were too malodorous for me. It was not until long afterwards that I realised that this revolting stench, which I often encountered in Tibetan houses, simply came from their meat-smoking operations and was really quite wholesome.

Next morning along came an old woman with bitten off (or torn off) oat stalks in one hand and a pile of horse dung in the other. On the basis of these corpora delicti she asserted that my horses had been trespassing in her oats, and claimed damages. However, as they had not been outside at all, we finally threw her out

The next day's march took us at first slightiy downhill along the lateral valley in which Wujiao is located, a valley which runs eastwards to the Yongning river. Then we went uphill to another tributary, its banks bordered with old trees of sea buckthorn covered with moss (Didymodon corticola — a new species). We continued up the hillside through magnificent forests, in some parts consisting of firs and numerous yews with an undergrowth of bamboo, and elsewhere of huge oaks, finally emerging on to the jagged mountain which bounds the valley of the Litang river. In many places the track was a deeply but trench, used by riders at their peril. Tree trunks had fallen across it and once, having just ridden beneath one, ducking as low as I could and yet scraping my back on it, as I started to straighten up, I saw just in time that there was a second trunk beyond the first and only managed to dodge it by a hairs-breadth. If my forehead had struck against it, I should certainly have been swept senseless from the saddle. I pitched camp below the pass at 3950m at the foot of rugged limestone peaks full of caves and grottoes, beside a little waterfall and amid fir woods and Potentilla bushes. The Chesha pass lay 150m higher, between short mountain chains running north and south. The whole mountain consisted of such chains, but erosion attacking it from north and south has converted it into a range which for the most part runs east and west The descent began through a high alpine valley where I gathered some botanical treasures: the tall purple Saussurea longifolia, the large Aster yunnanensis, Codonopsis subscaposa with white beUflowers marked with brownish-purple and Campanula aristata with tiny blue flowers. In a gloomy forest of firs and rhododendrons I found a new species of primula (P. muliensis [note # 85: Now Primula boreio-calliantha. Handel-Mazzetti's tepid description of this "truly magnificent Nivalid" — he omits even to mention the flower colour (lilac, purple, mauve or violet) — is in marked contrast to Kingdon Ward's ecstatic depiction in "The Romance of Plant Hunting", 1924, p. 119-120 (under the name P. coryana). See also K. Ward, Gard. Chron. Ser. 3 Ixxxii, fig. 35 (1927). It has never been successfully cultivated.]) . Its leaves were white beneath and its flowers, over 4 cm in diameter — among the largest of its genus — were spread horizontally to catch the light.

We turned left over a whaleback ridge and plunged steeply down through pine woods to the Muli valley. Clearings in the forest gave glimpses to the north. From afar Muli looked like a heap of pale stones flung down on the steep valley side. The Litang river ran past to its right in a deeply incised [p.74:] channel. Rugged crags towered to heights of over 2000m above it on both sides of the valley. Far upstream its course was delineated by intersecting lateral spurs, which compelled it to make seemingly minor zigzags. On that day, however, all the peaks were shrouded in low cloud. We camped for one more night down in the valley among aromatic Artemisia — one of the plants which made up the garrigue growing on the arid valley slopes. Below it was the lush subtropical bush of the humid river gorge. Next morning three hours' march brought us to the little town. The abbot's man had announced my impending arrival, and the headman of the temple was expecting me, a plump lama with a piercing gaze. His skin was smeared with rancid butter and he wore a red cowl which left his right shoulder and arm bare. The actual chief — I find myself slipping into the jargon used by Lao Li, to whom anyone with another man under him was a "chef — was temporarily absent, having set out with two hundred men in pursuit of three hundred of his yaks which bandits from Gongling had driven into the mountains. I was accommodated in a house just outside the town. It had been built by a Chinese trader, but his business had not lasted very long. He had been murdered on the road to Kangding and his dismembered corpse had been despatched to his widow.

Muli had been built in a quadrangular plan but was without a perimeter wall [note # 86: wing to the deteriorating conditions in China a town wall has since been built (see illustration by Rock in the National Geographic Magazine XLVII, p. 466). Muli must be the only town to have been fortified in this way in the twentieth century (Handel-Mazzetti 's footnote).]. Situated at 2800m, beneath forests of huge evergreen oaks overtopped by steep rock faces, it offered long vistas deep into the green valley of the Litang river. Its solid stone houses were roofed with shingles, and their small windows had frames with an expansion beneath and a little cornice above to keep the rain off. Two large temples, one at the upper and the other at the lower boundary of the town, served the needs of worship. However, the whole town was really a monastery, for it housed seven or eight hundred lamas, and the few lay inhabitants in their grey penitential garments seemed to blend into the background. I was surprised to find that the lamas, like all of them here, were red, and not, as one might expect from Davies' book, yellow. According to Rock, they belonged to the yellow sect of the reformed Tibetan church, but their only distinguishing mark was their yellow headdress; usually, however, they went bareheaded. As for women, there were only a few in the small farms round about; they came in to sell eggs, vegetables or wood or barter them for tea. Travellers in the mountains had to carry a stock of tea to offer instead of currency when making small purchases, for away from the main trade routes the natives would not accept copper coins. In Muli the Sichuan coins worth 40 cents — often called "rupees" [note # 87: "The Chinese rupee is a curious coin struck in Szechuan for Tibetan trade. The Tibetans of central Tibet trade mostly with India, and small traders take a sum of about 3000 rupees (£200) to Calcutta, where they buy Indian goods for sale in Tibet. Three thousand silver rupees are about one man's load, and, apart from the expense of carrying the money down, there is the danger of robbers. Consequently, the trader prefers to enter the British Post Office at Gyantse and send a money order to himself at Calcutta, which he does at a cost of one per cent. The BiMsb-.lndis.n Post Office 3) Gyantse would not accept any but Indian currency; Chinese and Tibetan currency, theretore, depreciated in value in comparison with Indian. The Chinese sought to overcome this by corning a Chinese rupee which would be of value equal to the Indian rupee; but, of course, though similar in size, weight and almost so in pattern, it was still ol no use at the Gyantse Post Office and fell in value to twelve annas (three quarters of a rupee), much to the annoyance ol the Chinese. I found that on this eastern border the Chinese rupee, not being influenced by the British Post Office, had a higher value than the Indian coin." P.M. Bailey, China-Tibet-Assam, London, 1945. Bailey made the journey in 1911.] by Europeans — were in circulation. There were only a few Chinese living there — traders from Shanxi — and they had a miserable existence: they were not permiitted to smoke their pipes except in the privacy of their houses, nor to open their mouths too wide, nor to adorn their dwellings with the customary greetings printed on red paper — not that these did anything to enhance the appearance of the houses. The only place where anything of the sort was to be iseen was above the gate of the great lamasery itself. Here there was a dedication in large Chinese lettering from the dujun of Chengdu (the provincial capital of Sichuan), who had granted it to the native prince (the "tusi") in recognition of his peaceable conduct.

As evening came on the little town prepjared for divine service. It was heralded by sonorous Iblasts on a horn, and the streets were immediately swarming with people. Red figures hurried towards the: temple, and from it came murmured chanting, mingled with the sounds .of gongs and drums. Choral singing, sometimes remarkably pleasing, was also to be heard, and after the monks had dispersed to their dwellings the noisy gabble of their prayers continued late into the night It sounded like an endlessly repeated "katanatanoten noten", but what the words actually were I was unable to ascertain. Not until 9 pm did quiet gradually descend. The Pumi llanguage spoken in Muli had certain striking peculiarities, notably the sharp rasping "r" and some (cadences resembling the second half of the well known song "Dearest father, send me cash!" I gave the Iheadman of the temple a watch; though it cost only $2!^ my servant told him that its value was $15. With an astonished cry of "a-i" he held it to his ear and listened to its ticking. Then he brought mie an old alarm clock which no longer went and asked whether I could repair it It was soldered into its case and there was no way of opening it, so there was nothing I could do, and my gift, simply because it was the only timepiece in the place, was ipso facto of greater value.

On the third day of my stay I called on Ihim with the purpose of viewing and if possible photographing the temples. First of all I took a picture of an enormous prayer wheel in its own little house, with an old man turning it, but then the door was immediately shut. The lower temples were locked up, and the keyholder, so I was told, was a long way off, but the upper temple was open. We climbed up to it, but found it closed. On trying the door the large lock at its foot gave way of its own accord; it was evidently [p.75:] defective. The man who was escorting me said that we ought not to open the door without the lama's help, and I myself felt somewhat uneasy, since all the rest of the men had remained below. There were, so it seemed, plenty of keys, but the headman of the temple could not bring himself to permit photography within the sanctuaries. Other foreign visitors had stayed only one night or even less, he was reported as saying, but I had stayed much longer. At this I told my men to speak politely to him and to ask whether the watch was still going and whether he had remembered to wind it This had the desired effect, and that afternoon I was permitted to visit the temples, though without my camera. First of all, however, I was entertained to a dish of tsamba, a Tibetan food made by mixing tea with butter and adding various kinds of flour and, in this instance, sugar. Then I had to purchase some lengths of silk as offerings, and incense candles, before we were allowed to proceed.

Black curtains edged with white concealed the huge doors of the three-storey building and lent it an air of mystery. On feast days enormous incense burners sent columns of smoke towards the heavens. Before entering the sanctuary we shuffled through a pile of bracken to clean our shoes. The Tibetans are totally different from the Chinese in their attitude towaxls their temples. No one in Muli would have dreamt of holding feasts or revels in front of the gods or of using the temples for play or merrymaking. Hie chambers were separated from one another by bills with wooden pillars, and their floors were strewn with red cushions and rugs to permit worshippers to kneel or prostrate themselves in comfort. Massive articles of copper and silver bore witness to great wealth. Figures made of butter on flat tablets of clay were being offered as sacrifices. Kneeling in front of each of the images of the principal deities were two lamas, reciting their rosaries in husky tones, My men paid their respects, but Lao Li, being a freethinker, was unmoved. The finest of all was a temple adorned with silk (Fig.28). There was a row of columns stretching up to the ceiling, all of them hung with layer upon layer of silk in various colours; Jie wooden pillars were swathed with similar fabric, and hanging down from all the galleries were silk-embroidered banners, side by side or overlapping. The colour scheme was not garish, but soft and festive, a light blue predominating. The religious books, each volume between wooden covers, were ranged round the whole room, set out in bookcases on rows of shelves. In a narrow hall at one side of the temple was long row of prayer cylinders with the well known inscription "Om maid padme hum". From time to time they were set in motion, each revolution counting as a prayer. Close by was a group of prayer flags, narrow strips of fabric fastened to tall poles in order of length, each having the words "Om mam padme hum" printed upon it Everywhere in the neighbourhood were swarms of obps, low pyramids of stone slabs with prayers chiselled upon them, each pile topped by a carved pole.

It was still raining hard and looked as if it would never stop, so I decided not to put off my trip to the mountains any longer. Next day, taking the tent and essential baggage, I set off towards the mountain to the west I had been told that it should be possible to travel two days' journey into the range without risk from bandits, and I was attracted by the prospect of making a photogrammetric survey of the country towards Gongling and up the Litang river, which should be feasible if the weather was kind. When Lao Li heard that the route I had chosen was the one leading to Zhongdian he said he thought the lama was trying to get rid of me. Before we had gone far, however, we came to a path which branched off to the right and led steeply uphill. When we reached 3900m the guide declared that we had already completed the first day's march and that there was no water further on. I accordingly pitched camp and spent the rest of the day collecting plants.

Next morning (30th July) it was still raining but I went further up. I had eyes for nothing except the wonderful plants of the limestone screes, among them Cremanthodium smithianum, Cyananthus formosus, Eriophyton wallichianum, Aconitum tatsienense, Corydalis calcicola, Hemilophia pulchel-/a, Dipoma iberideum, the little Scrophularia chasm-ophila with large greenish-yellow flowers, the new Pedicularis lophocentra, growing in mats beset with long-stalked red flowers, and Valeriana trichostoma. Up on a saddle known as Santante there was an unpleasantly cold wind, but otherwise the weather was quite agreeable, though unremittingly wet I climbed a peak called Saganai (4525m), clothed to its very top with patches of turf, rhododendrons and willow scrub. Photogrammetric surveying was out of the question and I was able to take only one plant photograph, but I made some interesting discoveries including a small blue Corydalis (C trilobipetala, a new species) and the well known northern Saxifraga ttagellaris in a form with exceptionally large flowers. At the tree line, which on the south side of the peak was constituted, quite exceptionally, by Quencus aquifolioides scrub, I once more found some huge spikes of rhubarb, growing above a wall of rock slabs dripping with moisture.

On die return journey next day, acting surreptitiously to avoid detection, I pocketed some papers printed with figures of gods and spirits which were tied to trees and bamboo thickets near the camp. Lower down I also appropriated a few small, neatly chiselled stone slabs from an obo, and some little grooved balls, moulded from clay, which had been placed togedier with incense candles in rock crevices to propitiate the mountain spirits. On my return I found the young lamas at their lessons; half of them were sitting at the foot of a wall while the others stood in front of them, addressing the pupils and clapping their hands unceasingly.

I then discussed with the lama the route to Zhongdian via Eya, and soon afterwards Lao Li came to me, once more pale and trembling.

"Sir", he said, "don't go through Eya. A few days ago it was raided by sixty bandits from Gong-ling. They stole a hundred pack animals and killed some people".

Banditry was rife at that time, since the force of four thousand soldiers stationed in Gongling and Xiangcheng, having been paid with paper money — something quite unfamiliar to them — had deserted and occupied Kangding and Ya'an. The major caravan route leading from Heqing between Muli [p.76:] traders who told me that they had been obliged to return from Bawolong, travelling all night along secret paths, and in view of what they said I felt it would be unwise to make a journey from Muli to the north. However, I was in no position to judge the truth of the stories I was told about the bandits. Chinese servants are inclined to fabricate such tales if they are afraid or too lazy to go somewhere, especially in districts inhabited by "Manzi". They use this term or "Yiren" to denote all non-Chinese peoples, and indeed we Europeans were commonly called "Yangguizi" (foreign devils), until the authorities were compelled to forbid the practice. Even at that time children still used to call out "foreign devil" as I passed, and they certainly had not invented the insult themselves. Some time afterwards Kok said to me that according to information which he had received I could have gone to Gongling after all — the situation was not really so bad, even there. I had by now gained the temple headman's trust He actually invited me to take pictures, and I succeeded in photographing the interior of the main silk temple, though even then there was a young lama who hurriedly tried to lock the door when he saw me approaching. Considering that the last white man to visit the temple — Jack [note # 88: R. Logan Jack, a Scottish geologist, had been engaged by the Chinese government as an adviser on mining and was stationed at Chengdu. Owing to the outbreak of the Boxer rebellion, he and his companions were obliged to leave China. Setting out on 27th June 1900, they travelled south east through Ya'an and Guabi,reaching Yongsheng on 30th August. They went on through Lijiang, up the Yangzi, across to the Lancang Jiang and as far as Weixi, where they turned back as the route to Burma was impracticable. Then they travelled south through Baoshan and Tengcnong, reaching Burma on 20th October. Muli is shown on his map (as "Mili Gen-chen"), but his route passed 60 km to the east. (R. Logan Jack, The Back Blocks of China, Edward Arnold, London, 1904). Though H.-M. was mistaken in implying that Jack had visited Muli, his memory was not entirely at fault. Major H.R. Davies spent a night there on 19th March 1900, but did not enter the temple. ("Yunnan", Cambridge, 1909, p. 242-243). Davies says that he was not the first European to visit Muli: Amundsen had been there in 1899 (Geographical Journal, June and November 1900) and wrote: "I believe M. Bonin also visited Muli in 1897".]— was not permitted to cross the threshold, this was a gratifying achievement

A climb down the arid hillside to the bridge over the Litang river — the Xiao Jin He in Chinese [note # 89: The Litang River below Muli is often called Xiao Jin He [Lesser Gold River] (SGH).]— was less rewarding than my journey hither, although my route along the side of the lateral valley hi which the town is situated was not at a substantially higher altitude. Low growing pale blue Gesnerads were flowering (Didissandra bullata and the taller Rhabdothamnopsis sinensis); others included the new Primula barybotrys with spikes of dark red flowers, Sedum engleri, similar to our own S. maximum, and Vitex yunmnensis, and the new Quercus cocetferoides was in fruit. The rock in the lower part of the valley consisted of clay-slate, compressed and flattened; it was said to be rich in gold, and in Muli I was shown articles carved from talc which had been quarried here. The bridge was a substantial structure built from wooden beams cantilevered out from each bank, arranged hi such a way that each layer projected some distance further out than the layer beneath, until finally their ends came close enough for them to be bridged by a platform high above the water. This type of construction, completed by a handrail supported by a tall arch in the middle, is characteristic of medium-sized Tibetan bridges.

Having soldered up a box full of dried herbarium specimens and entrusted it to a trader for forwarding to Lijiang, I left Muli on 3rd August with an addition to my caravan, a schnapps-sodden crony of the lama. Since the soldiers had not brought with them the promised letter of recommendation for Zhong-dian, I sent them back to Yongning with orders to ask the official if he would kindly make good this omission at once, but in reality it was a matter of complete indifference to me. As the weather was better I told the caravan men to pitch my camp on die Zhongdian road at the first stream beyond the branching of the track, and once more climbed the same mountain as far as the Santante col. Unfortunately I was again unable to photograph dlistant vistas, but I returned with some useful additions to my botanical collection and some good pictures of plants. The next day's march led from our hiillside down into a beautiful woodland valley where the people, most conveniently for me, had just completed the reconstruction of a bridge washed away by the river. The luxuriance of the mixed woodland — one striking feature was the large-leaved Populus szechuanica growing near the stream — was matched by the lushness of the meadows which fillijd its clearings. The mules from a small caravan were grazing in the meadows and the vegetation — consisting almost entirely of herbaceous plants with hardly any grasses — was so tall that one saw nothiing of them except their backs emerging here and there. The route continued south-south-west along the side of the wooded valley. As far as the Deke pass, which formed the watershed between the Litang river and the Shuiluo He (the river which runs into the Yangzi at the northern apex of its great loop), there were still a few scattered trees, although the pass was at an altitude of 4350m; between them, however, there was a flourishing high alpine flora. The slopes were dotted with silvery clumps of Anaphalis yunnanensis, a subshrub anchored to the clay-slate talus by its branching rootstocks. In moist spots where fine grit was exposed I found Ljgotis micrantha, a new species, deeply buried in the gravel, its leathery leaves pressed against the ground and its spikes of violet flowers lying on the muddy shingle. Another find was Crepis hookeriatia, its short stem expanded into a round swelling completely buried in the earth, with a rosette of numerous pinnatifid leaves pressed against the ground and in the middle a broad disc made up of countless narrow yellow flowerheads at ground level. Siimilar to it was Lactuca souliei, a small plant with blue flowers. Saussurea steUa had the same haibit of growth with linear grasslike leaves and flourished in the yak pastures, which covered large areas chiefly on the southern side of the pass. Cyananthus macrocalyx with yellow flowers also grew in them, together with Anemone rupestris and Tamicetum delavayi, only 1.5 to 5 cm in height, and here and there among them a small hairmoss (Polytrichum) was putting up its little "flowers".

[p.77:] We pitched camp just beyond the pass and next morning I visited it again, for I realised that its riches were not exhausted and the weather was now more favourable for photography. The track ascended the ridge, following the crest which led southwards between the Yongning valley landscape and the Shuiluo He river towards Waha and Alo, continuing at altitudes well over 4000m. It was a good path, the very one which Davies had taken in March 1900, but in his map he depicts the course of the crest incorrectly; he travelled along it in a snowstorm and had obviously lost his bearings. The flowers were superb. The fragrant yellow Cremanthodium cyclaminanthum (a new species) grew in small patches of turf; hi muddy spring-fed bogs I found the beautiful Ligular-ia paradoxa (also new), a comparatively small species with deeply cleft palmate leaves and greenish flowers, and Swertia elate, another new species, with blue flowers. Growing at the edges of the thickets was Haplosphaera phaea, representing a new genus of Umbelliferae, with single umbels of purplish-brown flowers, and Corydalis radicans, also new.

Scattered along the windswept ridge were groups of firs, patches of tanglewood and little thickets of rhododendrons — both shrubs and small trees — all growing on a variety of geological formations, mainly craggy limestone and soft igneous rocks, and presenting gorgeous pictures of unspoilt natural beauty. To the west, below the crest, among sombre firs and tree azaleas, were some little mountain tarns, as deep and clear as the lakes of the Bohemian forest Looking into their green waters one could see massive brown tree trunks rotting at the bottom; every stone between them was clearly visible even from a distance. Growing in the soft mossy soil were some superb flowers, notably Cremanthodium campanulatum, Doronicum stenoglossum and, once more, Primula muliensis. There were some splendid and informative vistas, especially from a little summit projecting towards the east I climbed it so as to photograph the range which I had seen when crossing the Chesha pass. From this side its numerous peaks seemed compressed and foreshortened. Beyond the broad green vale of Yongning I could see all the mountainous tract which extended to the south and southwest of Yanyuan. On the maps it was still a blank, but it was far too distant for any attempt to fill it in. During a somewhat belated midday rest I climbed a little summit to the right of the track in the hope of seeing the snow-capped Korikaling range, but I was disappointed. To the northwest there was indeed a chain consisting of dark rock, exposed to view up to an altitude where it was still sprinkled with numerous snow patches. Above this, however, it was completely hidden by cloud. Among the slate rocks of the peak I found Sedum pleurogynanthum, a new species with white flowers resembling those of a gentian. The little summit was situated at a nodal point; the main crest slopsd down to a saddle beyond which it gradually rose again into a broad forested mountain named Shagu, but the track led westwards along a branch ridge which terminated in Gonshiga [note # 90: This peak, about 30 km NNW of Yongning, is called Kan-ju-shan on Rock's map. Handel-Mazzetti's altitude of 4750m is remarkably close to the figure of 15,450 ft = 4770m given on ONCH-10.], a mountain which projected towards the Shuiluo He gorge. It was one of the highest mountains in the district, and at its foot I pitched what proved to be the highest camp in all my travels at 4250m.

Next morning (6th August) I set off to climb Gonshiga. The mountain sought to hide itself behind a hedge of rhododendrons, but I was able to see that one of its three peaks was the highest and I made it my objective. I soon wormed my way between the gnarled trunks of the tree rhododendrons and reached a watercourse which flowed down through a little ravine to the camp site. Up here the stream was bordered by an abundant growth of roseroot (Sedum purpureoviride) and Potentilla bushes, their stems and branches thickly clustered with cushions of moss, notably a new species of the family Pottiace-ae, a brownish moss with terminal fruits and small thin tightly packed, spore capsules (Didymodon handelu). Also growing beside the stream, generally as isolated plants, was Saussurea obvallata, its succulent stems tipped with densely compacted flower-heads and enveloped by large pale leaves, resembling those of Cirsium oleraceum though more closely apposed. I climbed obliquely up scree slopes of calcareous clay-slate, where I again found Arenar-ia'kansuensis, Hemilophia pulchella in a form with large yellow flowers, then Saxifraga muliensis, a new species with spreading runners, the narrow-leaved Primula rigida and the new Saussurea kato-chaetoides with broad, stiffly toothed leaves and a sessile flower head. The uppermost summit was a marble crag at 4750m; it was the highest point that I ever attained. Growing in its clefts and fissures was Saussurea leucoma, covered with white wool, and our own Pseudostereodon pulcherrimus forming golden-green cushions. The other plants ceased together with the clay-slate less than 50m below, yet up to that level there were still patches of turf comprising no fewer than fifty species of flowering plants, including the shrubby Potentilla fhiticosa and Lonicera hispida, while Rhododendron adenogynum climbed up to 4600m.

Although the sun broke through at intervals, distant views were totally hidden; I therefore turned back and devoted my attention once more to the mosses, which flourished in great profusion of species in moist shady spots beneath boulders and on bare patches left by earth slides. Some were our own alpine species such as Riccia sorocarpa, Distichium capillaceum, Encalypta ciliata and Campylopus schimped Then there were new species (Desmatod-on setschwanicus and Tayloria pygmaea). Growing on the rocks were Grimmia subconferta and G. micropyxis together with our own Gymnomitrium revolutum, two little yellow saxifrages (Saxifraga Oiaristulata and S. elatinoides) resembling S. montana, both of them new to science, and lastly Aconitum pulchellum, also new. On this occasion, possibly because of the denser air of summer, I had no [p.78:] symptoms of mountain sickness. I returned along the stream to the camp, ate my midday meal while the men packed up the tent, and then began the descent. The main track led down north-westwards through a short and very steep rocky valley to Weisha on the Shuiluo He river, but here I chose a detour to avoid bandit territory, a route which suited me since it led through stretches, admittedly not very extensive, of still unsurveyed terrain, while not diverting my steps away from anything else of interest.

Continuing south-westwards along the narrow crest we were drenched by a cloudburst, but as we went down the track leading off to the left we came to a shoulder from which we had a fine view of the forest-filled valley through which we now had to travel. In conjunction with a photograph from the lower end it was enough to fill in the map. Situated in small clearings on the hillsides were some Pumi villages, each consisting of a few stone houses. I stopped at the highest of them, Ryi (3250m), where my lodging was an open verandah and I was unable to dry my clothes properly. I sat up late that night putting my rich haul of plants into the press. Next day the narrow path led through the gorge and then along the opposite side of the valley — part of the slopes of Mount Shagu — up into each lateral ravine and then down again. The rain had washed it away in places and our guide had to send men ahead to repair it; our progress was slow. The valley was administered by Muli and hence belonged to Sichuan, not as the maps show, to Yunnan [note # : On modern maps the valley of the Shuiluo He is entirely within Sichuan (SGH).].

Battened upon by the lamas, the people were poverty-stricken. Men and women alike were clad in miserable rags. Nearly all the men wore pigtails and most of them had a knife and a snuffbox hanging from their belts. The women adorned their hair with chains of red and blue stones. Most of the children were completely naked, and all of them were streaked with dirt Their houses were small and exceedingly dirty, and I thought it wiser to sleep in the outer room — and on the second evening in my tent —in preference to accepting their friendly invitations. My guide certainly understood how to enlist their speedy cooperation. On one occasion he shouted a command to a youth standing on the outskirts of a village, and when the young man stared at us open-mouthed, he immediately threw a stone at hiim. In another village he made all the young men of military age turn out and devote their labour to the peaceful cause of science.

At the mouth of the valley the track descended into the arid subtropical zone belonging to the outer part of the Shuiluo He valley. The scenery here was similar to that on the Yalong. The ridge between the river and its western tributary, the Londa He, was made up of vertical strata of phyllite. As we went on the villages became more frequent; Dsengo, inhabited by Naxi people, looked like a fortress situated on a knoll projecting from the ridge into the river valley. The people gathered in front of the villages to stare at us; .die lower we went, the less clothing they troubled to wear. In this dusty oven I nryself began to wilt, and gazed longingly at the sombre fir woods now far above us. But this was not the time to let such longings prevail. The subtropical flora was certainly of no less interest than the plants which I might expect to see up there, and its study was perhaps of greater value in understanding the botanical history of the country, and its past climatic fluctuations and plant migrations. Here I encountered a new species, Lysimachia reflexiloba, with white flowers, the tips of the petals bent backwards; also Primula bathangensis, a tall plant with raceme :s of yellow flowers, and then Scilla chinensis, Chlotvph-ytum flaccidum and Cnoton caudatiformis in ripe fruit I had to wait half a day in Yumi while a bridge washed away by the water was being rebuilt; in fact, when I reached the spot I had to begin by telling the men how to lay the planks. As we progressed the valley scenery grew even more spleindid. The path climbed up the mountainside again, i overhung by steep rock faces consisting of limestone overlying sandstone and phyllite dipping southesist at about 30. These rocks rested unconformably on an older stratum of phyllite which was folded almost vertically. On the opposite side of the valley its wall also became steeper instead of flatter, and ahead of us we saw the sharp ridges and spurs round I the northern side of the great loop in the Jinsha Jiang river, which I finally reached on llth August at a village called Sanjiangkou.

[chapter 17:]