Full text of "A naturalist in western China, with vasculum, camera, and gun; being some account of eleven years' travel, exploration, and observation in the more remote parts of the Flowery kingdom"

See other formats

at 4 4 '
1 SDOLTNED SLL Pa He TE LEST TREY LOS IT OTL RTA LE NE OS RET I ED PEARSE aE GLE REE NRRL BOE VL

eae tae tec Pee gear Oe Pm ert Ad ek IEA ed eres ee 8 See Bei et as ie ie rsa oe 8 De i is oe

FS ES et LA ORI EE es a tate

A NATURALIST
IN WESTERN CHINA

THE AUTHOR

A NATURALIST IN
WESTERN CHINA

WITH VASCULUM, CAMERA, AND GUN

BEING SOME ACCOUNT OF ELEVEN YEARS’ TRAVEL,
EXPLORATION, AND OBSERVATION IN THE MORE
REMOTE PARTS OF THE FLOWERY KINGDOM

ERNEST HENRY WILSON, V.M.H.

WITH AN INTRODUCTION BY

CHARLES SPRAGUE SARGENT, LL.D.

WITH ONE HUNDRED AND ONE
FULL-PAGE ILLUSTRATIONS AND A MAP

VOL. I

LIBRARY
NEW YORK
BOTANICAL

GARDEN

METHUEN & CO. LTD.
36 ESSEX STREET W.C.
LONDON

QK355

Ae
\A/

e¥ B|
v1

First Published in 1913

PREFACE

give a general account of Western China, more
especially of its natural history and of the manners
and customs of the non-Chinese peoples inhabiting
the Chino-Thibetan borderland. The attempt is based
as broadly as possible, and it is earnestly hoped that
the information will be of interest to many sorts and
conditions of people.
» My travels in Western China began early in 1899,
_and had for their object the collecting of botanical
_ specimens and the introducing of new plants into the
gardens of Europe and North America. I have made
_ four separate expeditions, covering in all nearly eleven
years, and the nature of my work made it necessary
for me to eschew the beaten tracks of the Flowery
_ Kingdom.
The opportunity to travel and study the natural
, history of China I owe to the business enterprise of the
‘house of Veitch, the famous nurserymen of Chelsea,
-to whom I was recommended by Sir William T.
| Thiselton-Dyer, then Director of Kew Gardens, at the
instigation of Mr. W. Watson, the present Curator
, of that establishment. My first two expeditions were
in the interest of Messrs. Veitch ; the last two in that
of the Arnold Arboretum of ‘Harvard University.
The results of these four trips are well known in
~ the horticultural and botanical circles of Europe
and North America.
In my wanderings in China I have been singularly
“fortunate. The Chinese treated me always with
-kindly courtesy and respect. I was in interior China
un? b

[« the following pages I have endeavoured to

B8 19

LIBRAR:
NEW YOR:
BOTANICA!

GARDEN

vii A NATURALIST IN WESTERN CHINA

during the Boxer outbreak and the Russo-Japanese
War, and visited places shortly before or after anti-
foreign riots, but never experienced any incivility
meriting thename. I engaged and trained as collectors
a number of Chinese peasants, who served me faithfully
throughout my journeys, and we parted with genuine
regrets. At the commencement of my travels in
China, Mr. Augustine Henry, now Professor of Forestry
at Dublin, imparted to me much sound advice which
I did my utmost to follow. To this gentleman and
to the devoted services of my Chinese collectors must
be largely attributed the results of my work in China.

It is exceedingly pleasant to recall the kindly
acts and hospitality of the many people I have been
privileged to meet during my wanderings. Exigencies
of space forbid the mention of names but do not affect
my sincere appreciation. But for meeting them
one’s life would have been very much the poorer
and lonelier. To my friend, W. J. Tutcher of Hong-
Kong, this book in part owes its inception, and to
another friend, J. Hutson Edgar, I am indebted for
much information concerning the peculiar customs of
the Thibetans and other non-Chinese races.

In the preparation of this work, I have received
much encouragement from Professor Charles S.
Sargent, who has also contributed an introduction of
the greatest value. To my friend Herman Spooner
I am indebted for invaluable criticisms of the manu-
script. To Walter R. Zappey, my associate on the
third expedition, I owe much for assistance in details
concerning the colours and measurements of the game-
birds and mammals.

Two or three of the chapters I first published in
the Gardeners’ Chronicle during 1905-6, and that on
insect white-wax in the Chemist and Druggist, 1906,
but these have been remodelled to suit present require-
ments, and amended and corrected in accordance with
increased knowledge.

With six exceptions, the illustrations are from

PREFACE 1X

photographs taken by myself with a whole plate
Sanderson camera for the Arnold Arboretum, and
permission to use them I owe to Professor Sargent.
The photographs were developed and printed by Mr.
E. J. Wallis of Kew, who obtained from the negatives
the best possible results. For the illustration of
Budorcas tibetanus, I am indebted to Mr. Samuel
Henshaw, Director of the Museum of Comparative
Zoology at Harvard College.

It is quite impossible to record the full extent of
one’s obligations, since much information is un-
consciously absorbed through contact with many
people and extensive reading. I should, however,
be lacking in filial respect did I not record my sense
of indebtedness to the Alma Mater who gave me both
inspiration and opportunity—the Royal Gardens, Kew.

ERNEST H. WILSON

THE ARNOLD ARBORETUM
HARVARD UNIVERSITY
July 1913

CHAP.

TT.
ITI.

Vil.

VIII.

XI.
XII.

aaiil.

CONTENTS

INTRODUCTION BY PROFESSOR CHARLES S. SARGENT 2

WESTERN CHINA: MOUNTAIN RANGES AND RIVER
SYSTEMS . E a F 3 ; ‘i

WESTERN HuPEH : GENERAL TOPOGRAPHY AND GEOLOGY
METHODS OF TRAVEL: ROADS AND ACCOMMODATION :

In Quest oF FLowers: A JoURNEY IN NORTH-
WESTERN HUPEH : . : . :

ForEST AND CRAG: ACROSS THE HUPEH-SZECHUAN
FRONTIER A i - Z .

TuE RED BASIN oF SZECHUAN: ITs GEOLOGY, MINERAL,
AND AGRICULTURAL WEALTH . s A -

EASTERN SZECHUAN: NARRATIVE OF A JOURNEY FROM
TANING HsIEN TO TUNGHSIANG HSIEN ‘

THE ANCIENT KINGDOM oF PA: NARRATIVE OF A
JOURNEY FROM TUNGHSIANG HsIEN TO PAONING FU .

THE CHENGTU PLAIN: ‘‘ THE GARDEN OF WESTERN
CHINA ’’ . U z J ‘ d .

NorTH- WESTERN SZECHUAN: NARRATIVE OF A CROSS-
MOUNTAIN JOURNEY TO SUNGPAN TING : :

SuNGPAN TING: THE LAND OF THE SIFAN s 5

THE CHINO-THIBETAN BORDERLAND: ‘‘ THE MARCHES
OF THE MANTzU ”’ ; ° A ‘ ;

THE CHIARUNG TRIBES: THEIR History, MANNERS,
AND CUSTOMS . ; i - : :

104

116

141

149

160

xi A NATURALIST IN WESTERN CHINA

CHAP.
XIV. Across THE CHINO-THIBETAN BoORDERLAND: KUAN

HsIEN TO RoMI CHANGO; THE FLORA OF THE PAN-
LAN SHAN . ‘ * . :

XV. AcROSS THE CHINO-THIBETAN BORDERLAND: Romi
CHANGO TO TACHIENLU ; THE FoRESTS OF THE TA-P’AO
SHAN. - 5 : - °

XVI. TACHIENLU, THE GATE OF THIBET: THE KINGDOM OF
CHIALA, ITS PEOPLE, THEIR MANNERS AND CUSTOMS

XVII. SAcRED OMEI SHAN: ITs TEMPLES AND ITS FLORA .

XVIII. THRouGH THE LAOLIN (WILDERNESS): NARRATIVE OF
A JOURNEY FROM KIATING To MALIE, vIA Wa-
WU SHAN : . 5 . ‘

XIX. Wa SHAN AND ITs FLORA 3 F ,

PAGE

170

191

205

219

228

245

LIST OF ILLUSTRATIONS

THE AUTHOR. : : : : . Frontispiece
FACING PAGE

A PEAK OF THE BARRIER RANGE (TA-P’AO SHAN), Civ¢a 21,000 FT.
THE YANGTSZE RIVER AT ICHANG , F ; Z
A GENERAL VIEW IN NORTH-WESTERN HUPEH A ;
A VIEW NEAR HSINGSHAN HSIEN WITH TERRACED FIELDS ‘
SAN-YU-TUNG GLEN ; CLIFFS OF CARBONIFERIOUS LIMESTONE .
PRIMULA SINENSIS ; P ‘ : ; .
OuR CHINESE HousE-BoAaT ‘ - : -
OurR CARAVAN . : : ‘ X
HOSTEL AT CHE-TSZE-KOU, PINUS ARMANDI BEHIND s ;
HIGHWAY LEAD THROUGH A NATURAL TUNNEL . - :
WAN-TIAO SHAN, 8100 FT. i 5 is 3 a
In SAN-YU-TUNG GLEN, CUPRESSUS FUNEBRIS IN FOREGROUND .
THE CHINESE PRIVET (LIGUSTRUM SINENSE), 20 FT. TALL :

CERCIDIPHYLLUM JAPONICUM, VAR. SINENSE, 80 FT. TALL, GIRTH
7 BY. - i - : ‘ j ;

THE LarGE-LEAVED POPLAR (POPULUS LASIOCARPA), 50 FT. TALL,
GIRTH 5 FT. ; ‘ : , , :

THE MARKET VILLAGE OF TAN-CHIA-TIEN 3 : -
A TyPiIcaL VIEW IN THE RED BaAsIN ‘ P - :
A Soap-TREE (SAPINDUS MUKOROSSI), 80 FT. TALL, GIRTHI2 FT. .
MAUSOLEUM WITH ORNATE MURAL SCULPTURING ‘ :

TANING HstIEN: NorTH GATE . : 3 2 :

xili

Z
6
8

X1V A NATURALIST IN WESTERN CHINA

FACING PAGE

THE CHINESE PISTACHIO (PISTACIA CHINENSIS), 60 FT. TALL,

GIRTH 25 ET. Ps = 3 . : -
A SANDSTONE BRIDGE WITH CYPRESS AND BAMBOOS e
THE MARKET VILLAGE OF TAI-LU CH’ANG 2 :
THE REDBEAN TREE (ORMOSIA HOSIEI), 60 FT. TALL, GIRTH 20 ET.
NAN-CHING KUAN, OPPOSITE PAONING Fu 3 : :

THE DIvIDED WATERS AND BRIDGE (AN-LAN CHIAO), 250 YARDS

LONG 4 ; ; . ‘ ; Y
CHANNEL CUT THROUGH THE Li-TIU SHAN By LI-PING .
VIEW IN THE MANCHU SECTION OF CHENGTU CITY : 3
THE VILLAGE TEMPLE, KUNG-CHING CH’ANG ‘
A Memoria ARCH ° : ;
VIEW FROM HOSTEL, LAO-TANG-FANG . :

. .

HSUEH-PO-TING, WITH SNOWS; BED OF STREAM ENCRUSTED
WITH LIME DEPOSITS - . :

LooxinG East-SouTH-East FROM HsuEH SHAN Pass, RUINED
ForT IN FOREGROUND : ; 5

THE CITY OF SUNGPAN TING . E
Two SIFAN z : i ‘ é “

ToMB OF MAN MURDERED BY BANDITS ON HSUEH SHAN Pass 5

THE HicHway To TACHIENLU AND BEYOND, HERE BLASTED
FROM SOLID Rock @ :

A PRICKLY PEAR (OPUNTIA DILLENII) NATURALIZED .
THE TACHIENLU RIVER ; ¢
THE WOKJE VILLAGE OF TaA-wWEI : ‘ ‘
IDoLs IN A BONPA TEMPLE ‘ x ;
A BamsBoo SUSPENSION BRIDGE, 70 YARDS LONG ‘ ‘
THE VILLAGE OF HSUAN-KOU . : :

PRIMULA VEITCHII ‘ F 2

THE GI-LUNG LAMASERY ¢ ‘

82
86
90
94

100

106
106
110
110
114

122

132

142

LIST OF ILLUSTRATIONS

FACING PAGE

THe Town oF Romi CHANGO .

Forests WEST OF CHANGO, A POPLAR TREE IN FOREGROUND

RHEUM ALEXANDRE AND OTHER ALPINE FLOWERS

. .

THE YA-CHIA-K’AN, PEAKS 21,000 FT. .

. . .

TACHIENLU ; SITE OF FORMER TOWN IN FOREGROUND

MANI STONES . 2

. . . . .

THE LAMASERIES JUST OUTSIDE TACHIENLU

THE CHINESE BANYAN (FICUS INFECTORIA), 70 FT. TALL, GIRTH
47 FT. ‘ : ; ; :

HYDRANGEA XANTHONEURA, VAR. WILSONII, 15 FT. TALL

TEMPLE ON THE CiTyY WALL, KiaTING Fu

VILLAGE OF PING-LING-SHIH, MT. Wa-wuw IN DISTANCE

VIEW FROM TEMPLE ON SUMMIT OF MT. WA-WU, CLIFFS CLOTHED
WITH SILVER Fir (ABIES DELAVAYI)

BAMBOO JUNGLE AND SILVER FIR (ABIES DELAVAYI) .

WA SHAN, II,200 FT.

. . . . . .

192

194

242

246

INTRODUCTION

HE botanical explorations carried on in China in recent years

make it possible to compare the forest flora of eastern con-

tinental Asia north of about lat. 22° 30’ with that of eastern
North America north of the Rio Grande. In these explorations Mr.
Wilson has played an important part, and more than any other
traveller has shown us the remarkable richness of the flora of
western central China and the distribution and value of many of the
most important Chinese trees. A comparison of the flora of eastern
continental Asia with that of eastern North America made at this
time cannot be entirely conclusive, for although much has been
done to make known the Chinese flora much is still left undone ; and
there are still vast regions of the Celestial Empire into which no
botanist has as yet penetrated, and these may be expected to yield
new harvests of still unknown plants.

It is not surprising that the forest flora of China is richer in
genera than that of eastern North America, for although the area
of the two regions under consideration is not very dissimilar there
is a great difference in their topography. In eastern North America
only a few mountain peaks reach an altitude of 6000 ft., and these
are wooded to the summit. In China mountain ranges are more
numerous, with peaks which sometimes rise far above the upper
limits of vegetation, and on some of these mountain ranges the
timber line is at least twice as high as the highest land in eastern
America. The connection of the great mass of mountains of
south-western China with the Himalayas which must be considered
their western prolongation, and the great tropical region which
extends uninterrupted by any large body of water southward from
south-western China, will account for the presence in the Chinese
flora of many Himalayan and tropical forms which have no counter-
part in eastern North America. On the other hand, the flora of
eastern North America has drawn from the large and arid plateau
of Mexico many genera of Cactacez, the Agaves, Yuccas, Dasylirion,
and other genera which have no representativesin China. While the

xvii

xviii A NATURALIST IN WESTERN CHINA

larger mountain systems, the greater height of land, and its more pro-
lific neighbours can account for a larger number of genera in eastern
Asia than in eastern North America, it is not possible to find an
explanation for the greater number of species there of widely
distributed genera like Acer, Picea, Prunus, Sorbus, and Berberis,
which are more numerous in China than in any other part of the
world, or for the absence from eastern Asia of larger numbers of
species in genera like Crateegus and Amelanchier.

In eastern continental Asia there is nothing to compare with
the great maritime pine belt which extends from southern Virginia
to eastern Texas, and is one of the remarkable features of the flora
of eastern North America ; and the great forests of Pinus Strobus L.,
which once extended from northern New England and eastern
Canada to northern Minnesota, are but poorly replaced in north-
eastern Asia by trees of Pinus koratensis S. and Z., scattered over a
comparatively restricted area in eastern Siberia and Korea. The
Black Oaks, with their lustrous leaves and biennial fructification,
which are so abundant and conspicuous, except in the extreme north,
all over eastern North America, are wanting in eastern Asia ; while
the Bamboos, the most widely distributed and the most generally
useful of all the forest plants of China, are represented in North
America by two small and unimportant species of Arundinaria
confined to the swamps and river bottoms of the southern states.

As a rule, to which, of course, there are a few exceptions, the
trees of eastern North America are larger and more valuable than
related Chinese species ; but of Chinese shrubs it can be said gener-
ally that they produce more beautiful flowers than the shrubs of
eastern North America, although to this statement there are also
some exceptions. A more detailed examination of the principal
groups of forest plants in the two regions will show the similarities
and the differences of the forest flora of the two regions.

CYCADACE#.—Four species of Cycas are found in southern
China, and in Florida the family appears in two species of Zamia.

CONIFERZ.—This family is represented in China by fourteen,
and in eastern North America by nine, genera. In eastern North
America there are only two genera which are not also represented
in China, Taxodium, which is replaced there by the nearly related
Glyptostrobus and Chamecyparis, represented, however, in Japan
by two important trees. Libocedrus, Cupressus, Cunninghamia,
Pseudolarix, Keteleeria, and Fokienia have no eastern American
representative. In Pinus, eastern North America, with fifteen
species, has the advantage of eastern continental Asia, in which

———

INTRODUCTION xix

only eight species occur; and in eastern America Pine trees are
individually larger, more numerous, and more generally distributed
than in China. In Picea, China, with its twentyspecies, has a decided
advantage over eastern America, where only three species occur,
and in Abies, China, with its nine species, is richer than eastern North
America, where there are only two species, and of these one is found
only on the highest peaks of the southern Appalachian Mountains.
Of Tsuga there are two species in eastern North America and two
species in China, but Tswga canadensis Carr. is a larger tree and
much more widely distributed than any of the Chinese species.
Larix, on the other hand, 1s better represented in eastern continental
Asia, where it is widely distributed with several species from eastern
Siberia to the mountains of Western China, where it sometimes forms
large forests, while in eastern America there is a single species con-
fined to the north-eastern part of the continent and is a small tree
which southward is found onlyinswamps. Juniperus is represented
in China by six species, and in North America by five species ; but
none of the Chinese Junipers are as large or as widely distributed as
Juniperus virginiana L. ; and none of them produce so valuable wood
as that species and J. barbadensis L. Thuya is represented in each
region with a single species of about equal importance. In eastern
America Taxodium is a large and valuable timber tree widely
distributed in the South Atlantic and Gulf regions, while its Asiatic
representative, Glyptostrobus, is a small tree confined to the banks
of a few streams in south-eastern tropical China.

TAXACE#.—In this family the advantage is all with China,
with Taxus, Torreya, Cephalotaxus, Gingko, and Podocarpus,
while in eastern America it appears with only a single species each
of Taxus and Torreya, small trees found only in a few small isolated
groves in western Florida.

GNETACE. — Represented in continental Asia by Ephedra
and Gnetum, this family does not appear in eastern North America.

PANDANACEZ#.—One species of Freycinetia and two species of
Pandanus represent this Old World family in southern China.

PatmM#.—About the same number of species of Palms are re-
ported from the two regions, fourteen species in seven genera in
China, and sixteen species in eight genera in eastern North America.
The species in the two regions belong to different genera, with the
exception of Cocos nucifera L., which is found on all the tropical
shores. In eastern North America Palms extend farther north than
in China, and some of the dwarf species cover in the southern
United States great areas of dry sandy land with almost impene-

xx A NATURALIST IN WESTERN CHINA

trable thickets, as dwarf Bamboos make travel on some of the
mountain slopes of Western China almost impossible.

LirtacE#.—Of this family Heterosmilax is found only in
eastern Asia, but Smilax occurs in the two regions, in each of which
it is about equally represented by a number of species ; Yucca and
Dasylirion are American.

ARACE&.—Plants of this family are sometimes woody in southern
China, where species of Pothos and Rhaphidophora are large
climbers, but in eastern North America all the members of this
family are herbaceous.

PIPERACE#.—Piper, with woody species in southern China, is
the only genus of this family in the two regions.

CHLORANTHACE.—Of this small tropical family Chloranthus, in
China, is the only representative in the two regions,

SALICACE#.—In the number of species of Populus in the two
floras there is no great difference. None of the Asiatic species
grow to a larger size, perhaps, than the American Cottonwood
(Populus deltoidea Marsh.), but with this exception the Asiatic
species are larger and more valuable trees than the American
species, notably the Manchurian Populus Maximowiczii A. Henry,
and the north China Populus tomentosa Carr., which are among the
largest and most beautiful Poplar trees of the world. In Salix
there is probably no great difference in the number of species in the
two regions, although there is still much to be learned of the alpine
species of Western China. In eastern North America Willows are
mostly shrubby, only three or four species attaining to the dignity
of small trees, while in eastern Asia there are probably ten or twelve
arborescent species, and some of these are trees of considerable size.

JUGLANDACE#.—In this family the advantage is with eastern
continental Asia, with four genera against two in eastern North
America, where there is no representative of Pterocarya, Engel-
hardtia, or Platycarya. Juglans is common to the two regions, but
Carya is not known in China; and the presence in eastern North
America of this genus in many widely distributed species, valued
as timber trees and for the nuts produced by some of the species,
economically, at least, makes up for the absence of the three genera
which occur in China and not in eastern North America. Juglans, in
eastern America with three species, and continental Asia, with four
species, is of nearly equal importance in the two regions. None
of the Asiatic species, however, compare in size with the American
Juglans nigra L., but Juglans regia L., whose original home is now
believed to be on the mountains of northern and Western China,

INTRODUCTION xxi

yields the most valuable nuts and the most valuable timber pro-
duced by any species of the genus.

LEITNERIACE, a family of a single species of Leitneria, is
North American.

MyricacE#.—Comptonia is confined to eastern North America,
but Myrica, with a small number of species, occurs in the two regions,
with one species, Myrica Gale L., common to them both. In
North America there is no species which at all resembles Myrica
rubra Lour. with its edible fruits.

BETULACE2.—In eastern North America Carpinus appears
as a small widely distributed tree, but in continental Asia seven
or eight species of the two sections, into which this genus has been
separated (Eucarpinus and Distigocarpus), are common. Ostrya is
represented in each region by a single species, the eastern American
Ostrya being much more generally distributed and more abundant
than the Chinese species, which appears to be confined to the moun-
tain forests of Hupeh and Szechuan. The monotypic Ostryopsis is
confined to Mongolia and China. Of Alnus there are five species
in eastern North America, four of these being shrubs and one a
small tree; but in eastern continental Asia there are at least six
or seven species of this genus, and one of these, Alnus cremastogyne
Burk., is a large tree sometimes 100 ft. high, shading the banks of
many streams in Western China with groves of splendid specimens.
Betula forms a considerable part of the forests of eastern Siberia,
and is common on many of the mountain ranges of China, especially
those in western Szechuan, where it reaches altitudes of 10,000 ft.
In eastern Asia, however, there is no species which, like Betula
nigra L. of eastern North America, can thrive on the banks of
streams in the nearly tropical heat of regions like Florida, Louisiana,
and Texas, where this tree grows to its largest size. The number
of species in the two regions is not very different, and as timber
trees the Birches of one region are probably as valuable as those
of the other. It is doubtful, however, if any eastern Asiatic Birch
tree ever grows to the size sometimes attained by Betula lutea
Michx., of the forests of north-eastern North America.

FaGACE#.—In eastern North America there is a single species
of Fagus ranging from eastern Canada to Florida and Texas, and
one of the largest and most common trees of all this region. In
eastern continental Asia Fagus does not extend into the north,
and appears to be confined to the mountain forests of the central
western provinces, in which three species are now known; these
are smaller and less important trees than the American Beech.

xxii A NATURALIST IN WESTERN CHINA

Castanea is more important in the number of species in eastern
continental Asia than in North America, but it is doubtful if any
Asiatic species is anywhere as common or forms such a large part
of the forest as the American Castanea dentata Borkh. forms on
some Appalachian slopes; and in height and girth of trunk this
American tree has no Asiatic rival. In eastern North America
there are two other species; of these one is a small tree or shrub
and the other a shrub, both bearing a single nut in the involucre.
In Western China there are also two species with similar fruit, but
one of these, Castanea Vilmoriniana Dode, is a noble tree and
the largest of the eastern Asiatic Chestnuts; the other is a small
shrub, to be compared with the American Castanea nana Muhl.
The Japanese Castanea crenata S. and Z. reaches Korea, and Castanea
mollissima Blume, another small tree, ranges from the neighbour-
hood of Peking to the mountains on the Thibetan border. Cas-
tanopsis, which is related closely to Castanea, has its headquarters
in south-eastern Asia, with several species in southern China,
and one in California, but no representative in eastern North
America. It is possible that the number of species of Quercus is
greater in eastern continental Asia than in eastern North America.
Oak trees, however, are much less widely distributed in the former
region and are not numerous at the north, the Chinese Oaks being
chiefly confined to the southern provinces and usually evergreen.
Some of these evergreen Oaks should be referred to Pasania,
distinguished from the true Oaks by the arrangement of their flowers
in bisexual aments, the pistillate in several-flowered clusters below
the staminate, like the flowers of Castanea and Castanopsis. As
has been already stated, there are no Black Oaks in China, and
no species which are counterparts of the eastern American Chestnut
Oaks. The northern White Oaks are inferior in size to several
of the White Oaks of eastern North America, and it is doubtful
if any of the southern evergreen species equal in size Quercus
virginiana Mill., the Live Oak of the southern United States.
ULMACE#.—No Elm tree of eastern Asia equals the so-called
American Elm (Ulmus americana L.) in size and beauty, but it is
probable that the genus Ulmus has a larger representation of
species in western continental Asia than it has in eastern North
America, although it is still impossible to speak with much know-
ledge of Chinese Elm trees, which are very imperfectly understood.
It is interesting that the section of the genus (Microptelea), which
flowers in the autumn, has representatives in the two regions,
two in eastern America and one in China, the only other species

INTRODUCTION XxIli

of this group growing on the Himalayas. The monotypic Planera
occurs only in eastern North America, and Pteroceltis and Zelkowa
have no American representatives. Celtis is common to the two
regions, but the trees of this genus are larger and appear to be
more generally distributed in eastern North America than in
eastern continental Asia. The tropical genus Trema is represented
in both regions.

MoraAcE&.—Of this family the monotypic Maclura is American,
and Broussonetia, Cudrania, and Artocarpus, which occur in China,
have no American representatives. There are two species of Morus
in each of the two regions, but neither of the two American Mul-
berries compare in value with the Chinese Morus alba L., for the
leaves of this tree and its numerous varieties furnish the best and
chief food of the silkworm in all countries where silk is made. In
Ficus the advantage is with China, both in the number of species
and in the size of individuals, only two species having secured a
foothold in tropical Florida, where they are comparatively small
trees. Its nearness to tropical south-eastern Asia, which is one
of the great centres of distribution of this genus, will account
for the presence of some forty species of Ficus in southern China,
where some of the species grow to a very large size.

PROTEACE.—Helicia, in south-eastern China, is the only genus
represented in the floras of the two regions.

LoRANTHACE#.—In this family Phorodendron is North American;
Arceuthobium is North American and Japanese, and Loranthus and
Viscum are found in China and not in eastern North America.

SANTALACE#.—Pyrularia and the monotypic Darbya have not
been found in China. Henslowia is eastern Asiatic, and Buckleya
has a representative in the floras of the two regions.

OPILIACEZ.—One species of the tropical genus Cansjera, in
south-eastern China, is the only woody member of this family in
our two regions.

OLAcACEz.—Of this tropical family Schoepfia and Ximenia have
reached southern Florida from the West Indies, each with a single
species ; in China it appears only in Schcepfia.

ARISTOLOCHIACE2.—This family is represented in the floras
of the two regions by the genus Aristolochia, with more numerous
woody species in China than in eastern North America.

POLYGONACE#.—No arborescent or shrubby species of this
family is reported in China, but in tropical Florida two species of
Coccolobis occur, and a species of Brunnichia is widely scattered
through the southern United States.

xxiv A NATURALIST IN WESTERN CHINA

NYCTAGINACEH.—Pisonia aculeata L., an inhabitant of tropical
shores in many parts of the world, probably reaches south-eastern
China, as it occurs in Formosa; with other species of this genus
it is common in tropical Florida.

TROCHODENDRACE&.—This family is not represented in the
flora of eastern North America but appears in China in Euptelea.

CERCIDIPHYLLACEZ and EUCOMMIACE2, too, have no American
representatives, but appear in Western China each with a monotypic
genus, Cercidiphyllum and Eucommia.

RANUNCULACE2.—In this family Pzeonia occurs in China but
not in eastern North America, and the monotypic Xanthorrhiza is
Appalachian. Clematis is common to both regions, with a much
larger number of species in eastern Asia than in eastern North
America, where the genus is poorly developed.

LARDIZABALACEZ& is an Asiaticand Chilian family, with Decaisnea,
Stauntonia, Holbcellia, Akebia, Sinofranchetia, and Sargentodoxa
in China.

BERBERIDACE&.—The woody plants of this family are much
more numerous in China than in eastern North America. Mahonia
and the monotypic Nandina do not occur in the latter region, where
there is only one species of Berberis, while in eastern continental
Asia, which must be considered the headquarters of the genus,
some forty species are now recognized.

MENISPERMACEH.—IlThe woody members of this family are
represented in eastern North America by Menispermum and Coc-
culus. These occur also in western continental Asia, where Sino-
menium, Diploclisia, Stephania, Cyclea, Tinospora, Limacia, and
the monotypic Pericampylus are also interesting members of this
family.

MAGNOLIACEH.—Magnolia is represented in the two floras by
about the same number of species. In China, however, species
occur in two groups, one of which produces its flowers before the
appearance of the leaves, and in the other the leaves are nearly fully
grown before the flowers open. To the latter group all of the
American species belong. Some of the American Magnolias are
larger trees than the Chinese species, and no Asiatic Magnolia com-
pares in beauty with Magnolia grandiflora L. of the southern United
States or equals Magnolia macrophylia Michx. in the size of leaves
and flowers. Liriodendron appears in each region with a single
species, but the American representative of this genus is a larger
and much more widely distributed tree. Illictum and Schisandra
appear in the two regions, the former with three species in China

INTRODUCTION XXV

and one in the south-eastern United States, and the latter with
one American and eight Chinese species. Michelia, Kadsura, and
the monotypic Manglietia and Tetracentron are Chinese, the
latter being one of the largest and most interesting of the Chinese
trees.

CALYCANTHACE2.—Calycanthus, with several species, is eastern
North American only, and Meratia (Chimonanthus), with two
species, is Chinese and does not appear in eastern North America.

ANONACEH.—This tropical family reaches eastern North
America with several species of Asimina, its most northern repre-
sentative, and with Anona in tropical Florida. Uvaria, Artabotrys,
Unona, Polyalthia, and Melodorum represent it in south-eastern
tropical Asia.

LauRACE#.—In eastern North America this great, mostly
tropical, family appears only in Persea, Ocotea, Sassafras, Litsea,
Lindera, and Misanteca, but in eastern Asia there are eight genera
of Lauraceze, Cryptocarya, Beilschmiedia, Cinnamomum, Machilus,
Sassafras, Litsea, Lindera, and Cassytha. Of Sassafras there is
a species in each region, but the American species is a much more
widely and generally distributed tree, the Chinese Sassafras being
confined to the mountain slopes of western Hupeh and Szechuan.
Litsea, which appears in eastern North America in one small shrub,
in China is represented by at least a dozen species, among them
several small trees. Lindera, too, is more important in eastern
continental Asia than in eastern North America, where there are
only two shrubby species, while China can boast of nearly ten times
as many species ; some of these are large trees. The greater wealth
of China in plants of this family appears, too, in the important genus
Cinnamomum, including the species which yields the camphor of
commerce, and in Machilus, of which several species are large and
valuable timber trees.

CAPPARIDACE®.—This family appears in the tropics of the two
regions with Capparis and Crateeva. Capparis is common to both,
but Crateva of south-eastern China has no American representative.

NEPENTHACE.—One species of Nepenthes represents, in southern
China, this Old-World family of a single genus.

SAXIFRAGACE#.—This family occurs in each of the two regions,
with Philadelphus, Hydrangea, Decumaria, Itea,and Ribes. Deutzia
(with one species in Mexico), Cardiandra, Schizophragma, Pileos-
tegia, and Dichroa occur in China but not in eastern North America,

which has no woody genus of the family not found also in eastern
continental Asia.

xxvi A NATURALIST IN WESTERN CHINA

PITTOSPORACEZ.—Pittosporum, which reaches southern and
Western China with a few species, is the only genus of this family
in the two regions.

HAMAMELIDACEZ.—This family is more important in the number
of genera in eastern continental Asia than in eastern North America,
where there is one endemic genus, Fothergilla. Hamamelis and
Liquidambar occur in the two regions, and in China the family is
represented also by Distylium, Corylopsis, Fortunearia, Sinowilsonia,
Loropetalum, Sycopsis, Eustigma, Rhodoleia, and Altingia. In each
region the Liquidambar is a large, widely distributed, and valuable
timber tree. The Chinese Hamamelis, like one of the American
species and the species from Japan, flowers in the winter.

PLATANACEH.—Platanus, the only genus of the family, which is
represented in eastern North America bya large, common, and widely
distributed tree, has not reached eastern Asia.

Rosace#.—Of the thirty-four genera of the woody plants of
this family found in the two regions, Neillia, Stephanandra, Sorbaria,
Sibirzea, Exochorda, Cotoneaster, Osteomeles, Cheenomeles, Docynia,
Pyrus, Eriolobus, Pyracantha, Rhaphiolepis, Eriobotrya, Photinia,
Stranvesia, Rhodotypos, Kerria, Prinsepia, Pygzeum, and Maddenia
occurinChinaonly. Three genera, Aronia, the monotypic Neviusia,
and Chrysobalanus are American and not Chinese ; and ten genera
are common in the two regions, Physocarpus, Spirzea, Rosa, Malus,
Sorbus, Amelanchier, Crategus, Rubus, Potentilla, and Prunus. Of
the genera common to the two regions, Physocarpus, with one species
in eastern Siberia, is better represented in eastern North America,
where the genus is widely distributed with several species. On the
other hand, the closely related Spirezea has a few small eastern
American species, but abounds in China, which is the centre of
greatest distribution of this genus. Eastern continental Asia, too,
is greatly superior to eastern North America in species of Rosa, and
in their variety and horticultural value, for China is the home of
Rosa levigata Michx., Rosa bracteata Wendl., Rosa Banksie R. Br.,
Rosa multiflora Thunb., Rosa indica L., the origin of the Tea Roses
of gardens, and of Rosa rugosa Thunb. The number of species of
Malus is probably about the same in the two regions, but it is inter-
esting that those of eastern North America all belong to a group
(Coronariz) which is not represented in eastern Asia, where the
small-fruited species with a deciduous calyx predominate. Sorbus
in eastern North America is represented by two species of the
Aucuparia section, while in eastern Asia there are nearly thirty
species in this group and at least ten species of Aria which does not

INTRODUCTION XXVli

appear at all in the flora of eastern North America. Amelanchier,
which is very widely distributed through eastern North America,
with a number of species, of which two are small trees, has but one
shrubby Chinese species. In Crategus the difference between the
floras of the two regions is even more remarkable. In all of eastern
continental Asia only twelve species have been found ; in eastern
North America are more forms of Crategus than of any other genus
of plants, and probably a thousand species. In Rubus the difference
in the number of species in the two regions is probably not great ;
several of the American species, however, produce more valuable
fruit than any of the Asiatic species. Potentilla fruticosa L. ap-
pears in the two regions with two other related species in eastern
Asia and one in eastern North America. The composition of
Prunus is unlike in the tworegions. Of the true Plums (Prunophora)
there is only a single species in eastern continental Asia (Prunus
salicina Lindl.), confined to southern and Western China, no plum
tree being found anywhere in the north ; in eastern North America
plum trees have a wide distribution from the valley of the St.
Lawrence River to Florida and Texas, a larger number of species
occurring in the Arkansas-Texas region than in any other part
of the world. Padus, on the contrary, is represented in eastern
North America by only four species, while in eastern continental
Asia about seventeen species are recognized. None of these,
however, equal the American P. serotina Ehrh. in size or in
value as a timber tree. Laurocerasus appears in eastern North
America in two species and in eastern China in three species.
Cerasus has but three eastern North American representatives and a
much larger number in eastern continental Asia; and Amygdalus,
Persica, and Armeniaca occur in eastern Asia and not in eastern
North America.

CONNARACEZ.—Rourea in southern China is the only representa-
tive of this family in the two regions.

LEGUMINOS&.—Of the genera of woody piants of this family
the following occur in eastern North America but not in eastern
continental Asia: Lysiloma, Prosopis, Parkinsonia, Cercidium,
Amorpha, Eysenhardtia, Robinia, Coursetia, and Ichthyomethia ;
and the following in eastern continental Asia and not in eastern
North America: Fordia, Ormosia, Millettia, Maackia, Caragana,
Clitoria, Pueraria, Rhynchosia, Dalbergia, Euchresta, Mezoneurum,
Cesalpinia, Pterolobium, Entada, and Albizzia. The following
genera have representatives in the two regions: Pithecolobium,
Acacia, Leucena (probably naturalized in southern China), Mimosa,

xxvill A NATURALIST IN WESTERN CHINA

Cercis, Cassia, Gleditsia, Gymnocladus, Sophora, Cladrastis, Wisteria,
Erythrina, Desmodium, Lespedeza, Dalbergia, and Sesbania.

RutacE#.—In eastern Asia representatives of this family are
certainly more important than those found in eastern North
America, for they include Citrus, Limonia, Atalantia, two genera
of interesting trees, Phellodendron and Evodia, besides Toddalia,
Acronychia, Murraya, Clausena, Orixa, and Skimmia, while in eastern
North America this family is represented only by Helietta, Ptelea,
Amyris, and by Zanthoxylum which occurs also in the other region
which contains the larger number of species.

ZYGOPHYLLACE# is represented in eastern North America by
Guaicum and Porlieria, and in eastern continental Asia by
Zygophyllum.

COCHLOSPERMACE.—A species of Amoreuxia in southern Texas
is the only member of this family in the two regions.

SIMARUBACE# appear in tropical Florida in a species of Sima-
ruba and in Picrasma, and in Texas in Castela, while of this family
China has given to the world one of its valuable trees in Ailanthus,
and is also represented by Picrasma, Brucea, and Harrisonia.

BURSERACE#.—One species of Bursera is eastern North American,
and the species of Canarium in China represent this family in the
two regions.

MELIACE2.—Of the woody plants of this family in the two
regions Swietenia is certainly the most valuable, although it is the
only eastern American representative of the family ; while in China,
representatives of this family are Aglaia, Amoora, Turrea, Cedrela,
and Melia, one species of the last being widely and generally
naturalized in the southern United States.

MALPIGHIACE2.—This family reaches tropical Florida with a
species of Byrsonima and southern Texas with a species of Mal-
pighia : its only genus in China is Hiptage.

POLYGALACEZ.—Of this family only a few Chinese species of
Polygala are frutescent, the species of this genus in eastern North
America being herbaceous.

DICHAPETALACE#.—This small family is represented in the two
regions by a single species of Dichapetalum in south-eastern China.

EUPHORBIACEZ.—Woody plants of this family are more
numerous in China than in eastern North America, where the follow-
ing genera only appear: Andrachne, Drypetes, Croton, Ditaxis,
Ricinella, Bernardia, Gymnanthes, Sebastiana, Stillingia, Hippo-
mane,and Mozinna. These eastern North American representatives
of the family are all small shrubs with the exception of Drypetes,

INTRODUCTION XXIX

Gymnanthes, and Hippomane, which are small trees of tropical
Florida. In eastern continental Asia woody plants of this family
occur in Bridelia, Andrachne, Sauropus, Phyllanthus, Glochidion,
Securinega, Breynia, Bischofia, Aporosa, Daphniphyllum, Anti-
desma, Microdesmis, Aleurites, Croton, Blachia, Claoxylum,
Acalypha, Alchornea, Mallotus, Macaranga, Homonoia, Endosper-
mum, Gelonium, Homolanthus, Erismanthus, Sapium, Sebastiana,
and Exccecaria. Nearly all of these are tropical genera which,
coming from the south, have obtained a foothold in south-eastern
China. Only Andrachne, Croton, and Sebastiana have representa-
tives in the two regions. Aleurites, a genus of trees which produce
the wood-oil of commerce, is probably the most valuable genus of
the family in the two regions.

BuxaAcE#.—One species of Pachysandra occurs in each of the
two regions; the other genera of this family in the two regions,
Buxus and Sarcococca, are Chinese.

CORIARIACEH, a family of a single genus, Coriaria, has a re-
presentative in China but not in America.

EMPETRACEZ.—Empetrum occurs in north-eastern North
America and in north-eastern continental Asia, and the other
genera of this family, Corema and Ceratiola, are eastern North
American, and have no Asiatic representatives.

ANACARDIACEZ.—In this family, Pistacia, Rhus, and Cotinus
are represented in the flora of the two regions. Metopium is
American, and Spondias, Mangifera, and Dracontomelum are
Chinese. In China, Pistacia chinensis Bunge is a large, widely
distributed, and valuable tree, but in the United States Pistacia
mexicana H.B.K. has secured only a precarious foothold on the
northern bank of the Rio Grande in Texas. Of the members of
this family in the two regions Rhus verniciflua Stokes, the Chinese
Lacquer tree, is no doubt the most valuable.

CyRILLACE#.—This exclusively American family is represented
in eastern North America by Cliftonia and Cyrilla.

AQUIFOLIACEH.—Of this family, Ilex is widely distributed in
the two regions, and the monotypic Nemopanthus is east North
American. Ilex usually grows to a larger size in China than in
eastern North America, but is less northern in its range in the
former region where most of the species are evergreen.

CELASTRACEH.—In this family, Celastrus and Evonymus are
common to the two regions. Tripterygium, Perrottetia, Micro-
tropis, and Elzodendron are eastern Asiatic and not American,
and Pachystima, Maytenus, Crossopetalum, Gyminda, Scheefferia,

xxx A NATURALIST IN WESTERN CHINA

and Mortonia are American. Only one species of Celastrus and
three species of Evonymus occur in eastern North America, but in
eastern continental Asia the species of these genera are much more
numerous, and the species of Evonymus are usually larger and more
beautiful plants.

HIPPOCRATEACEZ.—Of this small tropical family there is a
species of Hippocratea in tropical Florida and two or three in
southern China.

STAPHYLEACE&.—Represented in China by Staphylea, Turpinia,
Euscaphis, and Tapiscia, this family is much more important in
eastern continental Asia than in eastern North America, where there
is a single species only of Staphylea.

IcACINACEZ.—Without an eastern North American genus this
family appears in China in Iodes, Mappia, and the monotypic
Hosiea.

ACERACE#.—Eastern continental Asia with its sixty-four
species is far richer in Acer than eastern North America, where only
ten species occur. The American Maples, however, are more
widely distributed, and are larger and more valuable timber trees ;
Dipteronia and Dodonza of this family are Chinese.

HIPPOCASTANACEH.—Of this family, Aésculus appears in two
arborescent species in China, one in the north and one on the
mountains of the west, but in eastern North America, where more
species are segregated than in any other part of the world, four
arborescent and four shrubby species occur in the southern United
States. The monotypic Bretschneidera is Chinese.

SAPINDACEZ.—Of the woody plants of this family found in our
two regions, Urvillea, Serjania, Exothea, Hypelate, Cupania, and
the monotypic Ungnadia are American, and the monotypic genera
Xanthoceras and Delavaya, with Nephelium, Schmidelia, Keelreu-
teria, and Pancovia, are Chinese; Sapindus is common to the two
regions.

SABIACEH.—Without representatives in eastern North America,
this family appears in China in Sabia and Meliosma.

RHAMNACE#.—Of this family, several genera reach tropical
Florida from the West Indies and the dry region of Texas from
Mexico, and the number is larger in eastern North America than in
eastern continental Asia. The exclusively American genera are
Rhamnidium, Reynosia, Ceanothus, Condalia, Karwinskia, Colu-
brina, and Gouania ; and the Asiatic genera are Ventilago, Paliurus,
and the monotypic Hovenia. Sageretia, Zizyphus, Berchemia, and
Rhamnus have representatives in the two floras. Species of

INTRODUCTION XXxXi

Rhamnus, however, are more numerous in eastern continental Asia
than in eastern North America, and Rhamnus davuricus Pall., and
other Chinese species, from which a green dye is made, are more
valuable than any of the American species.

ViTacE&.—Of the Grape family, three genera, Tetrastigma,
Cayratia, and Leea, occur in China and not in eastern America; and
Ampelopsis, Parthenocissus, and Vitis are common to the two
regions ; Cissus reaches tropical Florida but has not been reported
from southern China. Species of Vitis are less numerous in eastern
North America than in eastern continental Asia; and in North
America there is no species which corresponds with the spiny-
stemmed Grape vines of China (Spinovitis).

ELZOCARPACE&.—The forest flora of the two regions is only
represented by the Asiatic Eleocarpus and Sloanea (Echinocarpus)
of this family.

TILIACE#.—Tilia is widely distributed in the two regions with
rather more species in the Asiatic region. In size and in value as
ornamental trees there is not much difference between the American
and Asiatic Lindens. The Asiatic genera Grewia, Corchoropsis, and
Triumfetta do not appear in eastern North America.

MALvACcEz#.—In eastern North America there are woody species
in Pavonia, Hibiscus, and Thespesia, and in western continental
Asia only in Urena, Hibiscus, and Abutilon.

BomMBACE#.—Only the Asiatic Bombax represents this family
in the woody plants of the two regions.

STERCULIACE%.—The large genus Sterculia has several Chinese
representatives, including Sterculia platanifolia L. f., now natural-
ized in several of the southern United States. Of this family these
genera also appear in China: Heritiera, Reevesia, Kleinhovia,
Helicteres, Pterospermum, Abroma, and Buettneria, among which
are several large trees, while in eastern North America are only
Hermannia, Melochis, and Nephropetalum, all small shrubs of arid
Texas.

DILLENIACEZ.—Unrepresented in eastern North America, the
family appears in China in Tetracera, Actinidia, and Clemato-
clethra.

THEACE#.—Much more important in eastern continental Asia
than in eastern North America where only Gordonia and Stewartia
occur, this family has several woody plants in China, including Thea,
Gordonia, Stewartia, Schima, Ternstroemia, Eurya, Hartia, Tutcheria,
and Adinandra. One of the species of Thea, from the leaves of
which tea is made, is the most important member of the family ;

xxxii A NATURALIST IN WESTERN CHINA

and in another (Camellia) are found some of the most valued and
generally cultivated ornamental trees and shrubs.

GUTTIFERZ.—Of this family only Ascyrum, Hypericum, and
Clusia appear in eastern North America, but in eastern continental
Asia, Ascyrum, Hypericum, Cratoxylon, Garcinia, and Calophyllum
represent this family.

TAMARICACE2®.—With Tamarix and Myricaria in eastern con-
tinental Asia this family has no representative in eastern North
America, although one species of Tamarix is occasionally naturalized
in the southern United States.

CisTACE#.—Hudsonia of the Atlantic coast region is the
woody representative of this family in the two regions.

COCHLOSPERMACE2 and K@BERLINIAN#%.—A shrubby species
of Amoreuxia of the former, and a species of Kceberlinia of the
latter, both in Texas, are the only members of these families in the
two regions.

CANELLACE2.—A West Indian species of Canella which has
reached tropical Florida is the only member of this family in the
flora of the two regions.

FLACOURTIACE2.—Without a representative in eastern North
America this family contributes some of its most interesting trees
to the Chinese flora in Xylosma, the monotypic Carrieria, Itoa,
and Idesia, and in Poliothyrsis.

STACHYURACE2, of which Stachyurus is the only genus, is Asiatic.

TURNERACE&.—A shrubby species of Turnera from southern
Texas is the representative of this family in the two regions.

PASSIFLOR2Z.—Although Passiflora appears in several herbaceous
species in the southern states, Passiflora ligulifolia Mast. of southern
China is the only woody species of the family in our two regions.

CARICACEZ.—Carica Papaya L., now naturalized in many of
the warm countries of the world, is possibly a native of southern
Florida.

DaAPHNACE#.—In this family the advantage is with eastern
continental Asia, as Dirca is its only American representative,
while in China there are species of Daphne, Edgeworthia, Wick-
stroemia, and Aquilaria.

EL#AGNACE#&.—In this family Shepherdia is American, Hippo-
phaé Chinese, and Elzagnus is found in the two regions.

LyTHRACEZ.—The monotypic Decodon is the only woody plant
of this family in eastern North America, while in China appear

species of Lagerstroeemia and Woodfordia, and the monotypic
Lawsonia.

INTRODUCTION XXXili

RHIZOPHORACE%.—Rhizophora is common on the shores of
tropical Florida, but the family is more largely represented in
tropical China by Kandalia, Bruguiera, and Carallia.

COMBRETACEZ.—Represented in tropical Florida by Bucida,
Conocarpus, and Laguncularia, this family appears in China in
Combretum, Quisqualis, and Illigera.

MyrTACE£.—Rhodomyrtus, Eugenia, Psidium, and Beckea
of the Myrtle family have reached south-eastern China, while in
tropical Florida occur Eugenia, Anamomis, Chytraculia, and
Syzygium.

MELASTOMACE2.—Of this family woody species of Barthea,
Allomorphia, Blastus, Bredia, Anplectrum, and Memecylon occur in
China, but a species of Tetrazygia is the only woody member of
the family in eastern North America.

ARALIACE#.—An arborescent species of Aralia and a species of
Echinopanax are the only tree and shrub of this family in eastern
North America; in eastern continental Asia the family is more
largely represented by Aralia, Acanthopanax, Fatsia, Nothopanax,
Heptapleurum, Dendropanax, Heteropanax, and Hedera.

CorNACE#.—Of this family Garrya occurs in eastern North
America but not in China, where are found the monotypic Campto-
theca, a large tree, Davidia, Alangium, Helwingia, Torricellia,
Marlea, and Aucuba. Nyssa and Cornus are common to the two
regions. Nyssain America is widely and generally distributed from
New England to Florida and Texas, with several species, of which
two are large trees, but in China only a small tree is now known,
confined to the central provinces. On the other hand, Cornus is
more numerous in species in China than in eastern North America,
six of the species at least being arborescent and one a tree occasion-
ally roo ft. high.

CLETHRACE%.—Of this family there are three species of Clethra
in each of our two regions.

ERICACEZ.—In the number of genera of this family eastern
North America has the advantage of the Asiatic region, with twenty-
three genera in the former and only seventeen in the latter. Such
genera as Bejaria, Leiophyllum, Menziesia, Kalmia, Zenobia,
the monotypic Oxydendrum, Gaylussacia, and Arctostaphylos have
no eastern Asiatic representatives. Enkianthus, Craibiodendron,
and Diplycosia are Chinese, without American representatives,
and the following genera are common to the two regions: Vaccin-
ium, Gaultheria, Chamzdaphne, Loiseleuria, Phyllodoce, Andromeda,
Arctous, and Rhododendron. No single genus except Cratzgus so

xxxiv A NATURALIST IN WESTERN CHINA

well illustrates, perhaps, the differences in the floras of the two
regions as Rhododendron. In eastern North America there are
only six species of true Rhododendron, all confined to the extreme
eastern part of the continent, and, with one exception, of restricted
range, but some one hundred and sixty species have already been
distinguished in eastern continental Asia, where the genus is widely
distributed and where, on the mountains of the western and south-
western provinces, is the greatest segregation of these plants in the
world. On the other hand, only three species of Azalea have been
found in China, while in eastern North America, which is the
region of their greatest development, ten or twelve species are
recognized.

THEOPHRASTACE®.—The only member of this family in the
floras of the two regions is a single species of Jacquinia in tropical
Florida.

MyYRSINACEH.—This family has a much larger representation
in eastern continental Asia than in eastern North America, only one
species each of Ardisia and Rapanea having reached tropical
Florida, while in southern China there are several species of Ardisia,
Rapanea, and Myrsine, and where also Mesa, Embelia, and Aégiceras
occur.

PLUMBAGINACEZ.—With Plumbago in the two regions the
family is also represented in China by Ceratostigma.

SAPOTACE#.—In this family eastern North America has the
advantage with six genera, while only three genera reach southern
China. Of these only Sarcosperma is not represented in the
American flora. The other genera which are found in China,
Sideroxylon and Chrysophyllum, occur in tropical Florida, which
has been reached also by Dipholis and Mimusops, while Bumelia,
which is an American genus, is widely distributed through the
southern United States with several species.

EBENACE#.—Of this family only one genus, Diospyros, is
represented with two species in eastern North America, and eight or
nine species in China. As a fruit tree one of the Chinese species
is much more valuable than the North American species.

STYRACE&.—Of this family Styrax occurs in the two regions.
Halesia, with three species, is eastern North American; and the
monotypic Alniphyllum and Pterostyrax are Chinese.

SYMPLOCACE&.—Symplocos, the only genus of the family,
appears with one species in the southern United States, and is
largely represented in China, where twenty species are distin-
guished.

INTRODUCTION xxv

OLEACE2.—In this family also the advantage is with eastern
continental Asia, with eight genera, while only four are eastern
North American. Fraxinus, Chionanthus, and Osmanthus are
common to the two regions, Adelia is American only, and Fontanesia,
Forsythia, Syringa, Ligustrum, and Jasminum are Chinese and not
American. Fraxinus is widely distributed in each of the two
regions, with probably about the same number of species in each,
but the American species are usually larger and more valuable
timber trees. As an ornamental plant the American Chionanthus
is superior to the Chinese representative of the genus, but China’s
contributions to gardens from this family in Forsythia, Syringa,
Ligustrum, and Jasminum more than make up for the beauty of the
American Chionanthus.

LOGANACE#.—Gelsemium, with one species, and Buddleia, with
a species of southern Texas, are the only woody representatives of
this family in eastern North America. These genera occur in
China with Strychnos, Gertnera, and Gardneria.

APOCYNACE#.—Vallesia, Thevetia, and Trachelospermum are
woody plants of this family in eastern North America. It has a
larger representation in southern China in Plumeria, Melodinus,
Rauwolfia, Alyxia, Alstonia, Parsonsia, Pottsia, Wrightia, Ecdy-
santhera, Anodendron, Trachelospermum, and Scindechites.

ASCLEPIADACE#.—Roulinia, with a species of southern Texas,
is the only woody plant of this family in eastern North America,
but in eastern continental Asia occur woody species of Pentaneura,
Cryptolepis, Periploca, Taxocarpus, Calotropis, Holostemma,
Graphistemma, Metaplexis, Henrya, Gymnema, Marsdenia,
Stephanotis, Pergularia, Dregea, and Hoya.

CONVOLVULACE&.—Ipomecea and Argyreia are Chineserepresenta-
tives of this family, and a woody species of Ipomcea has reached
Florida from the tropics.

BoORRAGINACEH.—Cordia, Bourreria, and Ehretia are the North
American genera of this family, with woody species. Cordia and
Ehretia appear in China in a larger number of species than in
eastern North America, and Tournefortia is Asiatic and not
American.

VERBENACEZ.—Aloysia, Lantana, Citharexylon, Duranta, Calli-
carpa, and Avicennia, represent this family in the southern United
States. In China, Callicarpa, with several species, Premna, Gmelina,
Vitex, Clerodendron, Caryopteris, Sphenodesma, and Avicennia
make a larger representation of the family, only Callicarpa, with one
American species, and Avicennia occurring in the two regions.

xxxvi A NATURALIST IN WESTERN CHINA

LaBIAT&.—A few small shrubs of Salvia in Texas, and the genus
Elsholtzia, with several species in eastern continental Asia, represent
the woody plants of this family in our two regions.

SOLANACEZ.—Species of Lycium and Solanum, which occur in
each of the two regions, are the only woody plants of this family.

SCROPHULARIACEH.—Of this family Leucophyllum, a small
shrub of western Texas, is the only woody plant in eastern North
America, but in eastern continental Asia it is represented by the
important genus Paulownia, of several species of large trees, and by
Brandisia.

BIGNONIACE2.—In this family Campsis and Catalpa are
common to the two regions. The monotypic Chilopsis, and Aniso-
stichus and Crescentia occur only in the southern United States.
Oroxylum, Dolichandrone, Stereospermum, and Radermachera are
Chinese and not American.

GESNERACE®.—/Eschynanthus and Lysinotus,with woodyspecies
in China, are the representatives of this family in the two regions.

MyoporacE&.—A species of Myoporum of southern China is
the representative of this small family in the two regions.

RuBIACE.—Of this family the American woody representatives
are Cephalanthus, the monotypic Pinckneya, Exostemma, Genipa,
Randia, Catesbea, Hamelia, Guettarda, Erithalis, Chiococca,
Strumpfia, Psychotria, Morinda, and Ernodea. Of these Cephal-
anthus, Randia, Guettarda, Psychotria, and Morinda occur also in
China, where also are woody representatives of Adina, Luculia,
Wendlandia, Hedyotis, Mussenda, Adenosacme, Myrioneuron,
Webera, Gardenia, Diplospora, Antirrhcea, Conthium, Ixora,
Damnacanthus, Lasianthus, Pederia, Hamiltonia, Leptodermis,
Serissa, Emmenopterys, Dunnia, Pavetta, and Uncaria.

CAPRIFOLIACE®.—Of the ten genera of woody plants of this
family found in the two regions, Dipelta, Leycesteria, Kolkwitzia,
and Abelia (with a species in Mexico) occur only in China. Of the
other genera, Sambucus, Viburnum, Symphoricarpos, Linnea,
Lonicera, and Diervilla are common to the two regions. Symphori-
carpos is chiefly American, with a single Chinese species, and Vibur-
num, with some seventy species, is richer in China than in eastern
continental America, although the American species grow to a
larger size and are more ornamental. Lonicera is poorly represented
in eastern North America with twelve species, while in eastern
continental Asia more than one hundred species are recognized, the
region of their greatest segregation being on the mountains of the
central and western provinces.

INTRODUCTION XXXVIi

GOODENIACEH.—A Chinese species of Sczevola is the only woody
plant of this family in the two regions.

ComposiTt#.—Iva and Baccharis, of the Atlantic coast region,
are the American shrubby representatives of this family, which
occurs in China in several species of Blumea and in Pertya.

It appears, therefore, that 129 natural families are represented
in the two regions under consideration ; that of these 92 are common
to the two regions, that 12 occur in eastern North America, but not
in eastern continental Asia, and that 25 occur in eastern continental
Asia and not in eastern North America. Of the 692 genera in the
two regions 155 are common to both, while 158 are found in eastern
North America and not in eastern continental Asia, and 379 are
found in eastern continental Asia and not in eastern North America.
Of the tropical genera 76 have reached southern Florida and 89
south-eastern China.

From Mexico the flora of the United States has derived 42
genera of the woody plants of Texas.

It is impossible to form an accurate estimate of the comparative
number of species of woody plants in the two regions at this time,
but including Crategus it is probable that the number is as great
in eastern North America as in eastern continental Asia.

C. S. SARGENT
ARNOLD ARBORETUM

July 1913

A NATURALIST IN WESTERN
CHINA

CPAP TE, L1
WESTERN CHINA

MOUNTAIN RANGES AND RIVER SYSTEMS

\ , Y ESTERN China is separated from Thibet proper by
a series of parallel mountain ranges running almost
due north and south, and divided by narrow valleys.

On some maps the name Yun-ling is applied to the whole
system, with sections marked Hsueh shan, Hung shan, Taliang
shan, and so on. A great many local names, the majority of
them unpronounceable when converted into English, are also
applied to this system, but outside certain maps no one general
name for it exists. Later we shall have much to say about
this region, for the time being it suffices to note the general
trend of the ranges and a few of their important features.

Made up largely of razor-backed ridges, following one
another in quick succession, these ranges are separated by
narrow valleys, or rather ravines. The higher peaks are well
above the snow-line, and for height, savage grandeur, and
wondrous scenery are comparable only to the Himalayan alps
of India. The whole region is practically uncharted and un-
surveyed, and it is the author’s firm conviction that some of the
peaks rival in altitude the greatest of the Himalayan giants.

About lat. 33° N., in the neighbourhood of Sungpan Ting,
a mighty spur is thrown out from these ranges of perpetual
snow, and extends, with a slight southerly dip, due east for
some 10° of longitude, terminating in low hills near Anluh

VOL. I.—I

2 A NATURALIST IN WESTERN CHINA

Hsien, in north-eastern Hupeh. This spur appears on maps
under the general name of Kiu-tiao shan (nine mountain ridges),
Ta-pa ling, or Ta-pa shan. The two latter names have direct
reference only to important peaks of the spur, and the first is
the most appropriate, since it denotes a series of parallel chains
closely packed together. The Kiu-tiao shan forms the boun-
dary between Szechuan and the northern provinces of Kansu
and Shensi, and is the watershed between the middle Yangtsze
and the Han River systems. Attaining its greatest altitude in
the neighbourhood of Chengkou Ting, long. 108° 30’, lat. 32°
15’ N. (approx.), it radiates from this climax buttress-like
spurs in all directions. Those on the south form the boundary
between Szechuan and Hupeh and extend downwards beyond
the Yangtsze River. Subsidiary spurs and others thrust out
from more easterly points of the range, make the whole of
north-western Hupeh exceedingly wild and rugged. In the
middle of the province the Yangtsze River has forced itself
through these spurs, which run at right angles to its course,
and formed the famous Yangtsze Gorges.

Another spur, or rather series of spurs, not so clearly defined
as the preceding, and of less altitude, is thrown out in the
neighbourhood of Tali Fu, long. 100° E., lat. 25° 42’ N.
(approx.), in western Yunnan. It extends across northern
Yunnan, southern Kweichou, and northern Kwangsi, and forms
the boundary between Hunan and Kiangsi on the north and
Kwangtung on the south. In eastern Kiangsi it is deflected
north and north-north-east, finally reaching the sea in the
neighbourhood of Ningpo, long. 121° 35’ E., lat. 29° 50’ N.
(approx.). This mountain system extends across some 21
parallels of longitude, and forms the watershed between the
Yangtsze River on the north and numberless rivers on the
south. Of these the Red River, reaching the sea in the Gulf
of Tongking, and the West River, which enters the sea near
Macao and Hongkong, are the chief.

Innumerable lateral spurs are given off by this system, and
the country is extremely broken, especially in the western
parts, with which we are concerned. The province of Kwei-
chou is one mass of mountains, and the same is true of
southern Hupeh and southern Szechuan. In these three

“, ee ee

/ ‘LA 00012 VOMIO “NVHS OV.d-VL) AONVY UAINAVA AHL AO NVAd V

WESTERN CHINA 3

areas there are subsidiary ranges of considerable altitude
dipping in various directions and connected up with spurs to
form a heterogeneous and complex mountain system. The
outstanding feature of the whole region west of 112th parallel
of longitude is the entire absence of plain or plateau, or any-
thing in the nature of flat, level country, with the solitary
exception of the area forming the Chengtu Plain. Of this we
shall speak in due course. East of the 112th parallel the
Yangtsze River flows through a flat, alluvial plain in which
isolated, or more or less connected, mountain ranges and spurs
occur, but with this region we are not in this work concerned.

The most important region comprised within the mountain
systems above described and west of the 112th parallel is that
termed by Richthofen the “‘ Red Basin of Szechuan.’”’ This
region includes the whole of Szechuan east of the Min (Fu)
River to near the Hupeh boundary. It is a region of vast
agricultural wealth, with a magnificent river system, teeming
with large cities, towns, and villages, and supporting an enor-
mous population. With thesolitary exception of cotton, which
is imported from the coast, it is self-contained, with a surplus
of produce to spare for export. Salt is produced in unlimited
quantities in very many districts; coal, iron, and other
minerals of economic importance abound. In short, the
“Red Basin”’ is one of the richest and fairest regions in the
Chinese Empire.

The whole of Western China, with which this work is con-
cerned, lies within the Yangtsze River basin. According to the
geographical information at present available, the Yangtsze has
its source almost due north of Calcutta, in latitude about 35° N.,
on the south-east edge of the Central Asian steppes. Its exact
length is unknown, but it is estimated to exceed 3000 miles.
From its source it pursues a tortuous course, nearly due south,
through wild and partially unknown country for 1000 miles.
Then suddenly turning eastward it flows right through the
heart of China for some 2000 miles, finally reaching the sea
immediately to the north of Shanghai.

From its mouth to Ichang, rooo miles, it is navigable for
steamers at all seasons of the year, though in winter difficulties
in the way of shoals and sand-bars are encountered. The

4 A NATURALIST IN WESTERN CHINA

greatest difficulty is experienced between Hankow and Ichang,
and this section is operated by a small fleet of shallow draught
steamers specially built for the trade. The regular steamer-
fleet plying between Shanghai and Hankow is also specially
designed for the service and is luxuriously fitted. Ocean-going
steamers of deep draught can ascend as far up as Hankow,
except in low-water season. In summer the river overflows
and invades much of the low-lying country contiguous to its
course, and the chief difficulty in navigation at such times is
to keep to the channel. The difference between summer and
winter level is very considerable and varies to a large extent,
according to the width of the river and the nature of its banks.
At Ichang the river is 1100 yards from bank to bank, and the
average difference between summer and winter levels is about
40 feet ; in the gorges which commence some 5 miles west of
Ichang, the river is narrowed to a third of its usual breadth and
the difference exceeds 100 feet. Above Ichang the river is
obstructed by rapids, rocks, and other impedimenta, and is
navigated by specially built native boats that range up to
80 tons displacement. The difficulties of navigation are
more especially confined to the stretch of the river between
Ichang and Wan Hsien, a distance of about 200 miles. From
Wan Hsien to Pingshan Hsien, some 500 miles farther west,
the navigation becomes easier.

Much has been written on the possibility of opening the
Yangtsze River to merchant steamer traffic from Ichang west-
wards. So long ago as April 1900, two British river gunboats
of shallow draught, small in beam and length, and of a special
design, ascended as far as Chungking, the commercial capital
of Western China, distant above Ichang some 400 miles. Later
these boats ascended as far west as Pingshan Hsien and one of
them succeeded in reaching Mei Chou, a city on the Min (Fu)
River, about 140 miles above its junction with the Yangtsze at
Sui Fu. Since this exploit two larger and more powerful
British gunboats have been built for this work and are now
stationed at Chungking, which has been made a naval base.
France and Germany, following the British lead, have also
gunboats stationed at Chungking. During suitable seasons
these craft move up and down the river, and regularly every

WESTERN CHINA 5

year one or more visit Pingshan Hsien and Kiating Fu, the
latter city being about 100 miles north of Sui Fu on the Min
River.

The advent of the gunboats had been anticipated early in
1898 by a small launch called the Leechuan, commanded by
Captain C. Plant and owned by the late Mr. Archibald Little,
the pioneer foreign merchant of these regions. The experi-
mental test made by this launch took practical shape in 1900
when a commercial steamer named the Pioneer, captained by
Plant and operated by a British syndicate, in which Mr. Little
figured, was placed on this service. She madea trip prior to the
Boxer outbreak, after which she was chartered by the British
Government and was finally purchased for naval purposes.

On 27th December 1900, a German merchant ship, the
Suihsiang, specially designed and built for the purpose, left
Ichang for Chungking, but was wrecked and totally lost below
the Tungling Rapid only some 40 miles above Ichang.

Early in 1910 the task was again taken in hand; this time a
powerfully constructed tug named the Shutuwng, towing along-
side a flat for passengers and cargo, wasemployed. This outfit,
owned by a Chinese syndicate, was commanded by the same
Captain C. Plant. The venture proved successful, and four-
teen round trips were made during the year. It is fitting that
the man who pioneered the whole business should succeed in
demonstrating the practicability of merchant steam navigation
on the Upper Yangtsze. The work, however, is dangerous,
exceedingly difficult, and, moreover, costly, and unless some
improvements are made in the river-bed, it will be some time
before any considerable fleet of steamers ply on these waters.

Above Pingshan Hsien navigation is only practicable for
small native craft in certain short interrupted sections. The
river flows for the most part through gorges or between steep
mountains, and its course is frequently broken by dangerous
rapids and cataracts that produce a seething, foaming swirl in
which nothing canlive. Inthe autumn of 1911 an adventurous
French naval officer made an extraordinary journey down the
Upper Yangtsze to Sui Fu in native boats specially built for the
purpose. An account of this journey should prove exciting
reading.

6 A NATURALIST IN WESTERN CHINA

It has been mentioned that the most difficult stretch of
the Middle Yangtsze was that between Ichang and Wan Hsien.
This is the region of the world-famous Yangtsze Gorges. Five
in number, these gorges extend from the immediate west of
Ichang to Kuichou Fu, a distance of about 150 miles. Through-
out this stretch the river flows between perpendicular walls of
rock, is narrowed to a third or less of its usual width, and
becomes in consequence very deep. Soundings taken by the
British gunboats in their ascent in 1g00 gave 634 fathoms of
water in two places, and this when the water at Ichang was
rather less than 6 feet above zero mark! The cliffs, composed
largely of hard limestone, are 500 to 2000 feet or more high,
and commonly 500 to 1000 feet or more sheer. The scenery
hereabouts is savagely grand and awe-inspiring.

Foreign maps without exception give the name of Yangtsze
kiang (variously spelt) to this magnificent river. So far I have
never met a Chinese to whom this name is intelligible. I have
read that the name denotes “‘ Son of the Ocean,”’ and is applied
to the section between Wuhu and the sea. This may be so, I
have no knowledge on the point. Many local names are given
to stretches of this river, but from Sui Fu, in western Szechuan,
to its mouth it is universally spoken of by Chinese as the
Ta kiang (Great River), occasionally it is rendered Chang kiang
(Long River), or simply Kiang, meaning The River. West of
Sui Fu it is called the Kinsha Ho (River of Golden Sands); the
Chinese do not consider this the main stream, but regard the
Min (Fu) River asthe principal. Theyrecognizethat the Kinsha
has the larger volume, but it is navigable only for some 40 miles
and then loses itself in wild and barbarous regions. The Min,
on the other hand, is navigable to Chengtu, some 200 miles
above Sui Fu, and is therefore to the utilitarian mind of the
Chinese of much greater importance. From near Batang
northward the Kinsha Ho is known by the Thibetan name of
Drechu, and finally near its source it goes under the Tangut
name of Murussu.

In ascending the Yangtsze from Ichang to Chungking the
observant traveller is struck by the insignificant character of
the tributary streams. Apparently the only one of importance
joins the main stream at Fu Chou on the right bank. This

ONVHOI LV AHATA AZSLONVA AHL

WESTERN CHINA ~

stream, the Kien kiang, rises in western Kweichou and flows
through the heart of this province. It is navigable from its
mouth to Szenan Fu, and beyond, for specially constructed
native boats. Apart from this river there is no tributary of
seeming importance until Chungking is reached, yet nearly
every town and village of note stands at the junction of some
small stream with the Yangtsze. Here and there men will be
found hauling small, stout-bottomed boats over the stones at
the mouths of these small rivers. That the main stream is
joined by many tributary streams a glance at the map proves.
In western Hupeh the country is wildly mountainous, and the
streams are torrents, pure and simple. In eastern Szechuan
the country is much less wild and the streams of a different
character, and why they appear unnavigable is, on the surface,
not obvious.

In rgto I journeyed overland by a little-known route from
Ichang to Chengtu. Entering Szechuan a little to the north of
Taning Hsien, I travelled due west to Paoning Fu, and from
thence south-west by the mainroad. On this journey I crossed
all the principal streams which join the Yangtsze on its left
bank east of the Min River. The surprising thing observed
was the fact that they were one and all navigable for boats of
varying sizes for long distances. On inquiry, I found that
navigation ceased on most of them some 2 to 5 miles before
their union with the Yangtsze. The Kuichou Fu, Yunyang, and
Kai Hsien Rivers may be cited as examples, affording evidence
of this state of things.

Near the embouchure of tributary streams the Yangtsze is
generally narrowed and the water gorged by boulders and
detritus choking the mouths of these lateral waterways ;
rapids and races frequently occur at these points. The accepted
view is that enormous quantities of debris are brought down by
these tributaries and deposited at their mouths. This theory
is all very well when applied to mountain torrents, but most of
the streams under discussion pursue a comparatively placid
course with easy currents for some 50 miles or more before
reaching the Yangtsze. Their volume and force of current is
insufficient even in summer floods to carry down the enormous
quantities of detritus which choke up their mouths. My

8 A NATURALIST IN WESTERN CHINA

personal observations put the responsibility on the main stream
itself. During the summer floods the Yangtsze brings down
* vast quantities of mud and detritus, which it deposits wherever
opportunity offers. Flowing as the Yangtsze does, more or less
between steep banks, the mouths of tributary streams afford
the most favourable places for the deposition of this debris.
The volume of the main stream is enormously greater, and its
current so much stronger, than that of the tributaries that it
simply thrusts them back and silts up their mouths. The
small quantity of debris brought down by the tributary streams
would also be deposited hereabouts owing to the slacking of
the flow consequent upon the damming of their debouchure.

At Chungking the Yangtsze is joined on its left bank by the
Kialing or so-called “‘ Little River.’’ A glance at a map shows
that this river is made up of three streams which unite near
Ho Chou. The Kialing River and its tributaries drain a fan-
shaped area, in extent more than half of the entire Red Basin
situated north of the Yangtsze. Their importance is due to
their being navigable for such extreme distances. The most
easterly branch is navigable, for small craft, to some 40 miles
north of Tunghsiang Hsien; the next branch is navigable to
Tungchiang Hsien ; and the next to north of Pa Chou. The
central (Paoning) river is navigable for fairly large boats to
Kuangyuan Hsien, and skiffs laden with medicines and other
nativeproducts descend to this town from Pikou, inthe province
of Kansu. The most westerly branch is navigable to Pai-shih-
pu, a few miles north of Chungpa, and one of its western tri-
butaries taps the north-east corner of the Chengtu Plain.

The Kialing River system is thus the greatest collecting and
distributing waterway in Szechuan, and its commercial im-
portance, probably greater than that of the Yangtsze itself and
its tributaries west of Chungking, is not generally understood.
The To kiang, which joins the Yangtsze at Lu Chou, though a
natural stream, owes very much of its volume and importance
to water artificially lead from Kuan Hsien across the northern
part of the Chengtu Plain via Sintu Hsien, and a secondary
branch via Han Chou, which meet together at the great market
town of Chao-chia-to. In summer it is possible to descend in
boats from Han Chou and Sintu Hsien to Lu Chou.

A GENERAL VIEW IN NORTH-WESTERN HUPEH

WESTERN CHINA 9

The Min (Fu) River proper, save at lowest water, is navigable
from Kuan Hsien and Chengtu downwards. The Chengtu
branch is artificially formed by canals led across the plain from
Kuan Hsien, and unites with the Kuan Hsien stream and its
tributaries at Chiangkou. A tributary of the Min, which joins
at Hsinhsin Hsien, is navigable in high water for small boats to
Kiung Chou, a city situated at the extreme south-west corner
of the Chengtu Plain.

The Min (Fu) River rises some 35 miles north of the Sungpan
Ting, near the boundary of north-west Szechuan and the Amdo
region in lat. 33° N. (approx.). Immediately to the south
of Sungpan city it plunges into wild, mountainous country,
flowing through a gorge from which it emerges only a few miles
north of Kuan Hsien, where it becomes navigable for rafts only.

At Kuan Hsien a famous and gigantic irrigation system is
in operation, but of this we shall deal in due place.

The Min is really onlya tributary of the Tung River, which it
joins at Kiating Fu, but since it admits of navigation it is of
more practical importance, and for this cause the Chinese give
it pre-eminence. The Tung River is only navigable for a few
miles above Kiating, though rafts descend from a much higher
point west. Its tributary the Ya, which joins it immediately
west of Kiating, is of greater commercial importance, and a very
considerable raft traffic ascends and descends this stream from
Yachou, which is the centre of the brick-tea industry of western
Szechuan.

The Tung River is really one of the longest rivers in Sze-
chuan, having its source in the north-eastern corner of Thibet,
about lat. 33° 40’ N. It flows through the western frontier
of the tribes country, where it is known as the Tachin Ho
(Great Gold River), and ultimately strikes the highway from
Chengtu to Lhassa at Wassu-kou, a hamlet 18 miles east of
Tachienlu. From this point to its union with the Min at
Kiating it is called the Tung Ho, though around Fulin it goes
by the name of Tatu Ho. Owing to its unnavigability its
commercial importance is small, but this does not excuse the
geographer’s scant appreciation of it in the past, even if it
explains the Chinese view.

Considerably west of Pingshan Hsien the Yangtsze is

se) A NATURALIST IN WESTERN CHINA

joined by the Yalung River, an artery almost equal in volume
to the main stream itself. This river rises in the north-eastern
limits of the Thibetan highlands in the same general country as
the Yangtsze, but to the south-east. It flows more or less due
south throughout the whole course, but the region it traverses
is, 1f anything, less known than that through which the Yang-
tsze flows. In its upper parts it is called the Niachu, since
it flows through the country of the Niarung tribes, and its
map cognomen, “ Yalung,” is probably a transliteration of
the name Niarung.

On its right (south) bank the Yangtsze receives many
streams rising in northern Yunnan and Kweichou, but none
equalling in importance those uniting on the left bank. How-
ever, all are significant factors in the distribution of mer-
chandise in these parts, even though geographically they are of
comparatively small moment. The important thing to be re-
membered in connection with the river-system here mentioned
is this—the Yangtsze River is the main artery of China in
general and Western China in particular, but Szechuan owes
its agricultural wealth and general prosperity principally to
the Kialing and Min Rivers with their network of navigable
contributary streams and canals. The rivers west of Sui Fu
flow through wild, mountainous, sparsely populated zegions,
and their course is so much obstructed that practically no
boats ply on their waters and even ferries are scarce.

CHAPTER II
WESTERN HUPEH

GENERAL TOPOGRAPHY AND GEOLOGY

[Te country comprising western Hupeh, with which

we are concerned, lies west of the 112th parallel of

longitude. The city of Ichang, situated on the
Yangtsze River, just west of this parallel and about 1000
geographical miles from the mouth of the river, is a convenient
starting-point for exploring this region. This important town
is a treaty port, opened to foreign trade in 1877. The popula-
tion is roughly estimated at 30,000. There is also a small
foreign community consisting of a British Consul, Imperial
Maritime Customs’ staff, a few business men, and missionaries
of Roman Catholic and various Protestant denominations.
There is very little local trade, but Ichang being practically the
head of steam navigation on the river, is a most important
transhipping port. Six steamers regularly trade between
Ichang and Hankow, and the thousands of native craft lined
up in tiers attest its importance as an entrepdt of trade. In
the near future it is destined to be a most important junction
on the Hankow-Chengtu railway and already important work
on this enterprise has been commenced there. Ichang is well
known, and every year foreigners visit it in increasing numbers
intent on seeing the famous gorges which lie immediately
beyond. It is easily reached from Shanghai and from Peking.
From Shanghai palatially fitted and specially designed steamers
leave every night for Hankow, 600 miles up the Yangtsze.
From Peking and Hankow a good express train runs to and
fro weekly. Between Hankow and Ichang a regular fleet of
steamers keep up constant communication. Ascending the
river by steamer from Hankow the hilly country commences

12 A NATURALIST IN WESTERN CHINA

about 40 miles below Ichang. At first low, the hills gradually
increase in height, and by the time Ichang is reached one is
fairly among the mountains. In the vicinity of the town the
hills are pyramidal in outline, with prominent cliffs near by ;
north, south, and west of the town the country is much cut up,
_forming an archipelago of peaks 2000 to 4000 feet high,
the peaks themselves being offsets from spurs attaining alti-
tudes of 7000 to gooo feet, situated some days’ distance beyond.
These pyramidal hills around Ichang are very interesting and
never fail to attract the attention of travellers. They are
made up of a substratum of pebbly conglomerate, on which are
reared thin, horizontally deposed strata of marine limestone,
red shale and sandstone, over-capped with sandy clays. The
strata are piled with great regularity, and when erosion is
equal on all sides the characteristic pyramidal shape is pro-
duced and maintained. This formation is general from the
edge of the great plain to Ichang, and occasionally it contains
thin beds of coal. It is of comparatively recent age, dating
back to Permo-Mesozoic times. The dominant fossils it
contains are Cycads, and the youngest rocks probably belong
to the Odlitic series. The cliffs and bold peaks to the north,
south, and west of Ichang are made up principally of Paleozoic
limestones, with a little shale and sandstone, the latter of the
Mesozoic period. The strata are folded in apparent conformity
and are without notable metamorphism. In eastern Szechuan
these rocks extend beneath the Red Basin. The Yangtsze
has forced itself right through them and formed a series of
mighty chasms in which the structure of the various forma-
tions is beautifully exhibited.

In the neighbourhood of Hwangling Miao (30 miles west of
Ichang), and westwards for Io miles to the Tungling Rapid,
granitic gneiss is exposed. These are the oldest rocks in this
region and the only Pre-Cambrian formation known 7m situ in
the Middle Yangtsze. This section of the river is called the
Ta-shih Ho (River of Dregs and Boulders), and well does it
deserve this appellation.

The next oldest rocks of importance are those forming the
cliffs opposite Nanto, in the Niukan Gorge and in the eastern
half of the Wushan Gorge. This is a massive formation 4000 to

SQ THA GHOVUAUAL HLIM NYISH NVHSONISH YVAN MAIA V

Mey Fi i‘ ) ;
wh Ee.
ce ee

WESTERN HUPEH © 13

5000 feet thick, in the major part composed of dark grey or
liver-coloured limestone free from chert and containing both
Cambrian and Ordovician fossils. It is, in fact, a great marine
limestone in all its phases. It weathers into wonderful escarp-
ments, often sheer for 1000 to 2000 feet, with slightly projecting
summits, and is frequently many miles in extent. The cliffs
on the right bank of the river opposite Nanto, which extend
nearly to Hwangling Miao, are typical examples. At one of
the major rapids during the low-water season, known as the
Hsintan, some 45 miles west of Ichang, a bed of shale is beau-
tifully exposed. This bed is some 1800 feet thick, and com-
posed principally of olive-green argillite, with local black shale
and quartzite. It is of the Middle Paleozoic age.

Resting apparently conformably on this series of shale is
a vast deposit of Upper Carboniferous limestone 4000 or more
feet thick. This is the characteristic formation throughout
the Ichang and Mitan Gorges ; it occurs also throughout the
western end of the Wushan Gorge and in the Kui Fu(Wind-box)
Gorge beyond. The prevailing rock is dark grey or blackish
limestone, full of marine fossils and with occasional thin layers
of anthracite coal. This also weathers into wonderful escarp-
ments, but commonly they are boldly rounded with less linear
dimensions. This formation is the most general throughout
western Hupeh, on both sides of the river, though greater on
the north than on the south, where the Cambrian-Ordovician
formation preponderates. Next in succession come the Permo-
Mesozoic beds of red shale and sandstone, with thin layers of
marine limestone and coal, which were described in reference
to Ichang. These beds are characteristic of the country west
of the Mitan Gorge, as far as the entrance to the Wushan Gorge,
principally on the left bank. Coal occurs in this stretch in many
places, more especially around Patung Hsien. Glacial deposits
and signs of glacial action are in evidence in many parts of
western Hupeh, though nowhere on a large scale. The most
accessible of these is on the Yangtsze itself, opposite Nanto, a
hamlet situated on the extreme western end of the Ichang Gorge
and some 20 miles above Ichang city. At this point can be seen
a glacial deposit about 120 feet thick, overlaid by marine lime-
stone of the Cambrian-Ordovician age referred to above. All

14 A NATURALIST IN WESTERN CHINA

the evidences of ice action are well disclosed, and the whole
deposit is most instructively exhibited. Since the deposition of
these various systems great regional disturbances have taken
place and the strata has commonly been bent up from a great
depth. The summits of the very highest peaks in western
Hupeh are usually comprised of Silurian (? Devonian) shales.

None of the useful or precious minerals occurs in quantity in
western Hupeh. Coal is scattered through the entire region,
but is nowhere found in abundance and the quality is in-
different. Iron ore is worked in places and in one or two
localities the quality is good, but usually it is poor. Copper
occurs in two districts (Chienshih and Hsingshan) but is not
worked to any great extent. Salt, so abundant throughout
the Red Basin of Szechuan, does not occur. The sandy clays
and marls are used in brick and tile making, and lime is burnt
in several places and used for building purposes. Both the
clays and the limestone here mentioned belong to the Permo-
Mesozoic beds. The carboniferous limestones are quarried and
used for various construction works.

In the Gorges the main stream is joined by numerous lateral
branches which flow through glens of wondrous beauty. These
streams, winding their way through, usually fill nearly the entire
bed of the glen and are bounded by walls of cliff 300 to 1000
feet sheer. Waterfalls are numerous and wherever it is possible
vegetation is rampant. The tops of the cliffs are worn into
curious and grotesque shapes. Caves abound and in these
stalactites and stalagmites occur. Subterranean springs are
common and many of the small rivers originate from such
sources. They issue forth from some cave, or from the face
of a cliff, or well up through level rocks. The Hsingshan
River is an example of this mode of origin. The Chinese attach
much legendary lore to all these caves and subterranean springs,
and frequently associate fine temples with such spots.

In the vicinity of the Yangtsze the more commanding peaks
and crags are crowned by temples, usually belonging to the
Taouist cult. Commonly these temples cap seemingly inac-
cessible points, and one marvels how the material used in
erecting them was transported thither. Whenever possible a
few trees, usually Xylosma racemosum, var. pubescens (Winter-

SAN-YU-TUNG GLEN; CLIFFS OF CARBONIFEROUS LIMESTONE

WESTERN HUPEH 15

green), Gleditsia, Cypress, Ginkgo, and Pine are planted near
the temples and add much to the beauty of the scene. Such
temples are well built, but unfortunately, since the interior is
usually dark, filthy, and uninviting, a close inspection robs
them of most of their charm. From the distance they look
most picturesque, the style of architecture being in harmony
with the surroundings, and one admires very much the taste and
culture which called them into existence. The preservation of
the ‘‘ Good Luck ” of towns, villages, and communities by the
warding off of evil influences is a matter of great moment in
China, and with this good work the temples are associated.
The pagodas, found all over China, have been erected solely
with this end in view. Geomancy enters very largely into
Taouism and holds a most important place in Chinese thought,
and, in fact, governs many of their actions. As illustrating this
we will take an example at Ichang. Facing the town, on the
right bank of the river opposite, is a pyramidal hill nearly 600
feet high, called by foreigners the ‘‘ Pyramid.” This hill was
supposed 1o exert a baneful influence over the town, and was
held responsible for the town’s poverty in local literati. Not
until a temple was built on an eminence behind the town,
sufficiently high to enable it to overlook the Pyramid, was this
evil influence counteracted, and the Goddess of Good Luck
induced to smile on the town. The very year this temple was
completed a student passed the provincial examinations with
high honours. Was not this the beneficial result of the
building? The temple, called Tungshantzu, is richly endowed
and forms a strikingly conspicuous object from all points of
approach. The logic of “ Fung Shui,” as this cult is called, is
beyond the grasp of the average Occidental brain, but of its
effect on the Chinese mind one is constantly made familiar.

Too wildand savage for extensive agricultural development,
and with a marked absence of useful mineral deposits, western
Hupeh is one of the poorest, most sparsely populated, and
least known parts of China. For these same reasons it is of
particular interest to the botanist, since the vegetation there
has been less molested than is usually found to be the case in
China generally. Even here, it is hardly necessary to say,
every available bit of land either is, or has been, under cultiva-

WESTERN HUPEH 17

APPENDIX

THE FLORA OF ICHANG

Tue Flora of Ichang and the neighbourhood up to 2000 feet
altitude, as included in this note, is essentially of a warm
temperate character, and includes not a few sub-tropical
forms. Nevertheless, we find also a number of cool temperate
plants, and what really obtains is a fusion of these three
floras, with the warm temperate element in the ascendancy.
The following characteristic plants will serve to illustrate the
point: Aleurites Fordii, Liquidambar formosana, Ligustrum
lucidum, Cesalpima sepiaria, Toddalia asiatica, Wisteria
sinensis, Rhododendron indicum, Pyracantha crenulata, Primula
sinensis, Anemone japonica, Aspidistra punctata, Reinwardtia
trigyna, and Woodwardia radicans. The low hills around
Ichang are very barren-looking, being mostly clad with
“spear grass’’ (Heteropogon contortus), with a few shrubs
and herbs here and there, and relieved by small woods of
Pinus Massoniana and Cupressus funebris, with occasional
groves of the common Bamboo, Phyllostachys pubescens.

However, it is not to these low hills that we look for the
floral wealth of Ichang, but to the limestone cliffs of the glens
and gorges. Here the variety is astonishing, a striking feature
being the quantity of well-known flowering shrubs.

The two first shrubs to flower in the early spring are
Daphne genkwa and Coriaria nepalensis. It is a thousand
pities we cannot succeed with the Daphne in England, since it
is such a lovely plant—by far the finest species of the genus.
Here, at Ichang, it grows everywhere, on the bare exposed
hills, amongst conglomerate rock and limestone boulders, on
graves, and amongst the stones which are piled around the
tiny cultivated plats in the gorge, sometimes in partial shade,
but more usually fully exposed to the scorching sun. The
plants are, on the average, about 2 feet in height, and are but
seldom branched. Imagine the annual suckers from a Plum
tree, and you have the appearance of these Daphne plants.

VOL. I.—2

18 A NATURALIST IN WESTERN CHINA

For two-thirds of their height they are so densely clad with
flowers that they look like one large thyrse. The colour is
lilac, often very dark; but a white form is not uncommon.
Its outward resemblance to Lilac leads to its being so called by
the foreign residents at Ichang.

The Coriaria is not so well known and is not nearly so
attractive. Its flowers are polygamous, and the plant when in
fruit israther showy. The Chinese consider its foliage and stem
poisonous to cattle.

Wisteria sinensis is abundant, often scaling high trees, but
the semi-bush form is the more common. Its flowers are borne
in great abundance, and vary much in shade of colour, the
white form being, however, rather rare.

Another well-known shrub which abounds here is Loro-
petalum chinense. On the tops of the cliffs, amongst loose
conglomerate and limestone boulders, it forms a well-nigh im-
penetrable scrub. The bushes are seldom more than 3 feet in
height, very much branched, and when in full flower look like
patches of snow at a distance. Messrs. Veitch show the plant
very well, but there is an enormous gulf between the best grown
pot plants and the plants in a state of nature. In Devon and
Cornwall, if planted in a rockery, it ought to thrive.

Rose bushes abound everywhere, and in April perhaps
afford the greatest show of any one kind of flower. Rosa
levigata and R. microcarpa are more common in fully exposed
places. Rosa multiflora, R. moschata, and R. Banksi@ are
particularly abundant on the cliffs and crags of the glens and
gorges, though by no means confined thereto. The Musk and
Banksian Roses often scale tall trees, and a tree thus festooned
with their branches laden with flowers is a sight to be re-
membered. To walk through a glen in the early morning or
after a slight shower, when the air is laden with the soft delicious
perfume from myriads of Rose flowers, is truly a walk through
an earthly paradise.

In March and April Sophora viciifolia is very fine in the
glens and gorges when it is covered with masses of bluish white
flowers. This plant has a very wide distribution. It iscommon
in Yunnan, and in the warm valleys of rivers bordering Thibet.
The Ichang plant is much less spiny than that of Yunnan and

WESTERN HUPEH 19

western Szechuan. Possibly the latter is really the Indian
S. Moorcroftianum.

Two very common plants on the cliffs in the glens are
Eriobotrya japonica (Loquat) and Meratia precox. Both flower
about Christmas. These are two out of many plants which
formerly were erroneously supposed to be natives of Japan.

Among conglomerate boulders Caryopterts incana is
common, but is not nearly so fine as it is farther west. Pyra-
cantha crenulata and Vitex Negundo are exceedingly common,
and so also is Cesalpinia sepiaria. This thorny shrub is semi-
scandent in habit, and very like the better known C. japonica.
Its handsome foliage and erect thyrsoid racemes of bright
yellow flowers make it a very conspicuous object.

Symplocos crategotdes, with its pretty white flowers and
bright blue fruits, is abundant. This is a useful and charming
shrub, and deserves to be better known. Deuizia Schnetderiana,
Lagerstremia indica, Rhododendron indicum, Jasminum floridum,
Nandina domestica, Ilex cornuta, Viburnum utile, and Buddleta
officinalis are all extremely common shrubs. Of other well or
lesser known shrubs which are common, I may mention—

Abelia chinensis, A. parvifolia, Rhus Cotinus, Buddleia
astatica, Ilex pedunculata, I. corallina, Deutzia discolor, Des-
modium floribundum, Eleagnus pungens, E. glabra, Spirea
chinensis, Eurya japonica, Hypericum chinense, Hydrangea
strigosa, Berchemia lineata, Evonymus alata, Polygala Manesi,
Viburnum brachybotryum, V. propinquum, Thea cuspidata,
Rubus parvifolius,and many other species. Chaenomeles sinensis
with red, and C. cathayensis with white or blush-white flowers,
are commonly cultivated. Lengthy as is the list, I am not
justified in omitting Itea ilicifolia. This Holly-like shrub, with
long, pendent racemes of white flowers, is one of the handsomest
of all the Ichang shrubs. Of fluviatile shrubs, the commonest
are Distylium chinense, Salix vaniegata, Ficus impressa, Rhamnus
davuricus, Adina globiflora, Myricaria germanica, and a curious
Box (Buxus stenophylla). Climbers are very much in evidence,
and include such beautiful plants as Lonicera japonica, Trachelo-
spermum jasminotdes, Pueraria Thunbergiana, Clematis Henryt,
C. Benthamiana, C. Armandt, C. uncinata, Vitis flexuosa, Par-
thenocissus Henryana, P. Thomsonit, and Mucuna sempervirens.

20 A NATURALIST IN WESTERN CHINA

This last is a rather remarkable plant. Two miles above
Ichang, on the right bank, is an enormous specimen, called by
foreigners the “ Big Creeper.’ It covers several hundred
square feet of ground, climbing over several Pine trees and
many Bamboos. The base of the main trunk is almost as thick
as a man’s body ; the flowers are dark chocolate coloured, and
are borne in racemes on the old wood ; the legumes are 2 to 24
feet in length, and contain many large black bean-like seeds.
It flowers in May.

Ichang does not possess a great number of trees, but the
variety is really astonishing. Paulownia Duclouxi and Mela
Azedarach, with their enormous panicles of flowers, are very
striking in the spring. In the autumn, Sapium sebiferum, with
its wonderful autumnal tints, stands alone. In winter the ever-
green Ligustrum lucidum, and Xylosma racemosum, var. pube-
scens, are very conspicuous. The latter nearly always shelters
some wayside shrine. Perhaps the commonest trees are—

Gleditsia sinensis, Rhus javanica, Platycarya strobilacea,
Quercus serrata, Cedrela sinensis, and Pterocarya stenoptera.
The Mistletoe occurs on the last-named tree. Other lesscommon
trees are Sterculia platanifolia, Populus Silvestri, Crategus
hupehensis, Celtis sinensis, Dalbergia hupeana, Acer oblongum,
Cunninghamia lanceolata, Atlanthus glandulosa, Broussonetia
papyrifera, Ulmus parvifolia, Hovenia dulcis, Sapindus muko-
vosst, Salix babylonica, and Sophora japonica. Of this latter a
curious variety occurs in which the leaves and young shoots are
clothed with a dense white velvety indumentum.

As with flowering shrubs, so with herbs, though in a less
degree, Ichang is the home of many favourite garden plants.
One of the commonest and best known is Primula obconica.
This charming herb abounds everywhere, but more especially
in moist, grassy places on the banks of the Yangtsze and in the
glens. Occasionally, under very favourable conditions, in
height, size of flower, and luxuriance of foliage, it approaches
the cultivated form, but more usually it is a dwarf and
almost insignificant weed.

Again, Ichang is the home of the Chinese Primrose, and
the type of the cultivated Chrysanthemum occurs there also.
Other favourites which are common are—

WESTERN HUPEH 21

Corydalis thalictrifolia, Anemone japonica, Sedum sarmen-
tosum, Saxifraga sarmentosa, Iris japonica, Reinwardtia tigyna,
Lycoris aurea, L. radiata, Rehmannia angulata, Hemerocallis
fulva, and H. flava. Other characteristic herbs are: Adeno-
phora polymorpha, Bletia hyacinthina, Asarum maximum,
Ophiorrhiza cantonensis, Viola Patrinit, Delphinium chinense,
Lysimachia Henryt, L. clethroides, Potentilla chinensis, P. discolor,
Fragaria indica, Thalictrum minus, Mazus pulchellus, Verbena
officinalis, Platycodon grandiflorum, and many Composite,
Leguminose, and Umbellifere.

Perhaps Ichang is best known to horticulturalists generally
as the home of the lovely Likium Henryi. This acknowledged
favourite occurs on the limestone and conglomerate rocks, but
isnow bynomeanscommon. Lilium Brownii and its varieties,
chloraster and leucanthum, are fairly common; L. concolor
occurs, but is rare.

Ferns are not rich in species, but Woodwardia radicans,
Osmunda regalis, Picris longifolia, P. serrulata, Nephrodiummolle,
Cheilanthes patula, and Gleichenia linearis are very abundant.
A variety of Adiantum Capillus-Veneris is very common on
stalagmitic limestone in the glen. Pieces of these rocks covered
with Ferns are detached and find their way all over China, being
popularly known as “‘ Ichang Fern-stones.”’

A hasty reference to the common floating plants of the ponds
and ditches around Ichang must bring this note to a close.
Euryale ferox, with its handsome foliage, is very common ;
Nelumbium speciosum is, of course, cultivated. Other common
aquatics are—Limnanthemum mnymphoides, Jussie@a repens,
Salvinia natans, Trapa natans, Azolla filiculoides, Marsilea
quadrifolia, Monochonia vaginalis, Eviocaulon Buergerianum,
and several species of Potamogeton and Utricularia. In late
autumn, when the Azolla changes to a rich crimson tint, the
ponds look very fine. In some rice fields near Ichang Dr.
Henry found a very anomalous plant. It was made the type
of a new genus—Tvafella sinensis, and doubtfully referred to
the natural order Pedalince.

CHAPTER ‘ITd
METHODS OF TRAVEL

RoADS AND ACCOMMODATION

HE advent of steam navigation on the upper-middle
Yangtsze has brought Chungking, the commercial

metropolis of Western China, three weeks nearer the
coast and occidental civilization. This is a very considerable
gain to the would-be traveller in these regions, yet it only post-
pones for a littie time longer the inevitable. Sooner or later
the traveller must dispense with the comforts and luxuriesof
modern occidental methods of travel and adapt himself tothose
more primitive and decidedly less comfortable of the Oriental.
In the regions with which we deal there is nothing in the nature
of wheeled vehicular traffic save only the rude wheel-barrows
in use on the Chengtu Plain. There are no mule caravans, and
scarcely a riding pony is to be found. For overland travel
there is the native sedan-chair and one’s own legs; for river-
travel the native boat. Patience, tact, and abundance of time
are necessary, and the would-be traveller lacking any of these
essentials should seek lands where less primitive methods
obtain. Endowed with the virtues mentioned, and having
unlimited time at his disposal, he may travel anywhere and
everywhere in China in safety, with considerable pleasure and
abundant profit in knowledge. With her industrious toiling
millions, her old, old civilization, her enormous natural wealth
and wondrous scenery, China alternately charms and fascin-
ates, irritates and plunges into despair, all who sojourn long
within her borders. No country, outside Europe and North
America, is of such perennial interest to the world at large as
China. Ever-changing yet ever the same, she is the link
which connects the twentieth century with the dawn of

OUR CARAVAN

METHODS OF TRAVEL 23

civilization, epochs before the Christian era. To travel
leisurely through this vast country is an education which
leaves an indelible impress on all fortunate enough to have
had the experience. The Chinese do not see time from the
Westerner’s view-point, and for the traveller in the interior
parts of China the first, last, and most important thing of all
is to ever bear this in mind.

The majority of travellers still ascend the river above
Ichang in native boats, and it will probably be a long time
before a regular fleet of steamers ply these dangerous waters
and render the native boat obsolete. The journey from Ichang
to Chungking and beyond has been described so often that
the subject is threadbare, and I have no intention of describing
it over again. Volumes have been written on this subject,
and some day perhaps a writer will arise and do full pees
to the theme.

I have made the journey up and down many times, and on
each occasion have been more and more impressed with the
sublime beauty of the Gorges. The scenery in these savage
chasms is all and more than any writer has described it as
being. It must beseen to be fully understood and appreciated.
The more often one travels up and down this stretch of the
river the deeper grows one’s awe and respect for the many
rapids, swift currents, and innumerable difficulties which im-
pede navigation.

The native boats are perfectly fitted for the navigation of
these difficult waters ; they are the outcome of generations of
experience, and the balance-rudder and turret-build have been
used in these craft long before their adoption by Western
nations. The men, too, who earn their livelihood in navigating
these boats, understand their business thoroughly. Much has
been written by hasty travellers on the shortcomings and
incompetence of these men, that is as unwarrantable as it is un-
deserved. These Chinese boatmen are careful, absolutely com-
petent and thorough masters of their craft, and the more one
sees of them and their work the more one’s admiration grows.
Oriental methods are not occidental methods, but they suc-
ceed just the same! When on the boat the Westerner will do
well to adapt himself to Eastern methods; any attempt to

24 A NATURALIST IN WESTERN CHINA

enforce those of the West generally ends in disaster. Many
accidents on the Yangtsze have been caused through the
foreigner, ignorant of local conditions, difficulties and dangers,
forcing the captain of the boat to proceed against his better
judgment. The traveller is advised when engaging a boat to
do so through a responsible Chinese business house, to have an
agreement drawn up setting forth the arrangements desired,
and then to leave the boat-master to carry out his engagement
in hisown way. This is the only way to ensure safety, and on
paper no one would attempt to gainsay it, yet in practice this
is commonly done, but always to the jeopardy of the trans-
gressor.

Since we shall have much to say on the subject of over-
land travel a word or two anent roads seems fitting and
desirable. To the uninitiated this subject may seem trivial,
but to the experienced it is otherwise. Chinese roads make a
lasting impression on all who travel over them, and the vocabu-
lary of the average traveller is not rich enough to thoroughly
relieve the mind in this matter. The roads are of two kinds,
paved and unpaved. I have yet to meet the traveller whose
mind is thoroughly made up as to which of these is worse and
the more difficult to negotiate. A clever writer once wrote :
“An Imperial highway in China is not one which is kept in
order by the Emperor, but rather one which may have to be
put in order for the Emperor.”’! When any important official
takes up duties in a distant part of the empire the local officials
put the roads over which he has to travel in some semblance of
repair. Such work is always hastily done by labour forced and
grudgingly given, and in mountainous districts the first severe
rainstorm destroys considerable portions of it.

It is nobody’s real business to look after the roads, and
nobody does. The land devoted to roadways is com-
mandeered, and in agricultural districts the farmer takes
good care to keep these roads down to a minimum width. It
usually happens that the roadways get narrower and narrower
every year, until the advent of some important official forces
the local authorities into having them repaired and restored
to their original width. Roads in China owe their origin to

1 Arthur H. Smith, Village Life in China, p. 35.

METHODS OF TRAVEL 25

the same causes that obtain elsewhere in the world, namely,
military conquest and commercial interchange between distant
localities.

Throughout the length and breadth of China run imperial
highways, few in number, it is true, but of vast importance,
since they connect the imperial capital with the capitals of the
provinces. These were made for military purposes in early
times, when the Emperors were busy conquering the country
and extending their territories. They are all of great strategical
importance, and were originally paved throughout with huge
blocks of stone. Often, indeed, they were actually blasted and
excavated from solid rock. They vary in width according to
the configuration of the country and the nature of the traffic
they have to carry. In the northern parts overland travel is
commonly done by cart, and the roads are adapted to such
traffic. In the parts with which we are concerned the country
is too wild and rugged for wheels, and the only recognized mode
of travelling is by means of sedan-chair. The imperial roads
were originally made sufficiently wide to enable two chairs to
pass one another freely. Ten to twelve feet is a broad highway
in these parts, and it must be conceded that roads of such width
amply serve their purpose. Unfortunately this width is rarely
maintained for any considerable distance. The grading of
these ancient highways was well done, and the whole work
speaks volumes for the ability and energy of those old-time
engineers. Like much else in China these roads were once
magnificent, but to-day they are far from this. In general
they are sadly neglected. Floods have destroyed them here
and there, often the paving blocks have been stolen for house-
building and other purposes, and gaps of unpaved, muddy
stretches, almost impassable in rainy weather, occur all too
frequently. Sufficient of the original road remains to stir
admiration for the skill and foresight of the engineers, long
since dead, and to set the traveller longing for those halcyon
days of old.

In the prosperous parts of China, highways connect all the
principal cities, town, and villages. These are usually 8 to 10
feet wide, and though originally paved throughout, are now in
a state of more or less disrepair. Nearly all the towns and

26 A NATURALIST IN WESTERN CHINA

villages in Western China are situated on the banks of streams
for the simple reason that the valleys offered lines of least resist-
ance. Even when the streams are not navigable they afford
easier means of access to the interior than the mountains and
forest-clad country. In a general way all the older roads
in China follow the courses of streams as closely as possible,
leaving them only when the nature of the country necessitates
the departure, and watersheds intervene.

Bypaths and narrow tracks permeate the country in every
direction, and abound even in the most sparsely populated
mountainous regions. Some one has very wisely made out that
the exchange of salt was the first commerce engaged in by
mankind at large. Salt is, and long has been, a Government
monopoly in China, consequently the practice of salt-smuggling
has gone on from time immemorial, and the majority of
the mountain-paths were very probably first struck out by
smugglersofsalt. Indeed, many important trade-routes to-day,
in China, presumably originated in this way. The province of
Szechuan is abundantly rich in salt and also in mountain-paths.
From a lengthy study I have come to regard this network of
bypaths as the result of salt traffic, and more especially illicit
traffic. There are to-day many such paths throughout the
Hupeh-Szechuan boundary, used for practically no other traffic
than that of salt, and by these paths salt still reaches certain
districts in defiance of the law. Very useful, if difficult, the
traveller finds these bypaths, for without them it would be
impossible to traverse some of the wildest and most interesting
parts of central and Western China.

When travelling overland in China it is not possible to use
tents, and one has perforce to make use of such accommodation
as the country affords. The Chinese do not understand tents,
and it is unwise to try innovations in a land where the people
are unduly inquisitive. The traveller gets along best when he
avoids publicity as much as possible. On all the main roads
there are inns of sorts, usually very filthy, and in season abound-
ing in mosquitoes, creeping things, and stinks, the latter, in
fact, being always in evidence. On the byways, and more
especially in the mountains, accommodation is hard to find and
is of the meanest description. However, one is usually tired,

CGNIHAd IGNVWUYV SANId ‘NOM AZSL-AHOD LV 'IHLSOH

METHODS OF TRAVEL 27

and any shelter suffices for a night’s halt. In wet weather, or
when held up through flooded torrents or what not, the absence
of proper accommodation is acutely felt. In the wilds of China
one hungers for the dak bungalows of India and Ceylon, or
accommodation on similar lines.

A traveller in China should have with him an outfit, com-
prising bed, bedding, victuals, cooking paraphernalia, and
insect-powder. It sounds rather forbidding on paper, but
labour is cheap, and a little experience enables one to keep the
size of outfit within reasonable limits. The necessary coolies
should always be obtained through a respectable agency and
an agreement made in writing, stating all necessary details.
A head-man, called a “ Fu-tou,”’ should be given charge of
the coolies.

In parts of China where foreigners are well known, it is
possible to dispense with the luxury of a sedan-chair, but it
must be remembered that a sedan-chair is an outward and
visible sign of respectability. Itis the recognized medium of
travel, and, quite apart from its real use, it is a necessity, since
its presence ensures respect. In the out-of-the-way parts of
China, even though it is carried piecemeal, a chair is of greater
service and value to the traveller than a passport. According
to treaty, all foreigners travelling in China must furnish them-
selves with a passport, which must be shown on request. This
is a matter of considerable importance, and should never be
omitted.

Onething more is necessary ere the caravan is fully equipped,
and that is a good cook. Unless the traveller speaks Chinese
he must have a servant able to speak broken English. A good
travelling servant is hard to find, but the last thing the average
traveller should dream of doing is to engage an interpreter. A
good domestic servant will fill this function in so far as it is
necessary.

CHAPTER IV
IN QUEST OF FLOWERS

A JourNEY IN NorTH-WESTERN HUPEH

N 4th June roto I left Ichang for Chengtu, via a new
route through the wilds of north-west Hupeh. With

600 miles of overland travel ahead the caravan had
been fitted up with all the skill at my command, and with
enthusiasm to spur us on I felt that the difficulties would not
prove insurmountable. Nearly all the men had been associated
with me on former journeys of a similar nature.

We took the lesser road by way of San-yu-tung glen for
Hsingshan Hsien, in consequence of the main road being
congested by coolies engaged in blazing a trail for the Hankow-
Szechuan Railway. The caravan consisted of twenty carrying-
coolies, several men for collecting and general work en route,
a chair for the Boy, and another for myself. My own start
was not propitious. I was riding in my chair and had scarcely
cleared the precincts of the foreign settlement when one of
the poles snapped. This occasioned an hour’s delay, but
happening where it did new poles were secured without diffi-
culty. It is never easy to make an early start the first day,
and it is always advisable to count on a short stage. It was
one o'clock when we reached the mouth of the San-yu-tung
glen, 5 miles above Ichang, and overtook the main caravan.
The weather was hot, and we only did another 15 li! to Sha-
lao-che, making 35 li in all. This little hamlet consists of a
few scattered houses, and we availed ourselves of the largest,
which happened to be a wine-distillery, and the smell of stale
brewing was very strong.

The journey up the San-yu-tung glen was very interesting,

1 Ten apes” Seat miles.

IN QUEST OF FLOWERS 29

much of the scenery being rugged and grand. The cliffs of hard
limestone are usually 500 feet or more sheer, and are the home
of Goral and other animals, and also of many cliff-loving plants.
In the crevices and niches the Chinese Primrose (Primula
sinensis) finds its home, but the flowers were past and the
flower-stems all bent towards the cliffs to ensure the seeds being
deposited in the rock crevices. This plant is the parent of
our greenhouse Primulas, and in February and early March
the cliffs present a wonderful picture, being covered with
colonies of plants, one mass of warm mauve-pink flowers.
Wherever the cliffs are not absolutely sheer, vegetation is
rampant. Pine trees (Pinus Massoniana) fringe the summits
and Rosa microcarpa was in full flower, otherwise there was
very little blossom to be seen. Most of the shrubs being spring-
flowering were in young fruit.

There was considerable delay in starting the next morning.
One or two of the coolies gave up, and others had to be found.
The road was vile all day, and it took us 10% hours to cover
45 li. For the first 10 li the road continues to ascend the
glen, which narrows and presents even finer scenery than that
of yesterday. We passed a lovely natural grotto full of
stalagmites inside, and with the dripping external rocks one
mass Of Maiden-hair Fern. These rocks are known throughout
the Lower Yangtsze Valley as “‘ Ichang Fern-stones,” and
command a ready sale.

In the glen Parthenocissus Henryana is abundant ; in the
juvenile stage the leaves of this plant have prominent white
veins, and are very attractive, but in the adult stage this
variegation is lost, and they become very ordinary by com-
parison.

The glen soon became impassable, and we climbed the
cliffs and ultimately overlooked the country generally.
Terraced fields are much in evidence, and every available
inch of country is under cultivation. Wheat, barley, and peas,
all ripe, were the principal crops, and their yellow culms en-
livened the landscape. We saw a small patch or two of the
Opium Poppy hidden away under trees and of very poor
quality. Pear and Plum trees are commonly cultivated
hereabouts, Bamboo groves and Cypress trees abound. Here :

30 A NATURALIST IN WESTERN CHINA

and there we caught an occasional glimpse of the white-tailed
Paradise Fly-catcher (Tchitrea incet). Pheasants were calling,
and likewise the English Cuckoo.

Around Niu Ping (Cow-flat), which was our destination
for the day, much rice is cultivated, and the farmers were
busy transplanting the tiny rice-plants. The whole country
is finely terraced and is backed by limestone cliffs of Cambrian-
Ordovician Age. Near our destination we passed a fine Ginkgo
tree showing curious root-like protuberances on the branches.
In rocky places by the wayside, and especially in the walls of
the terraced fields, Rehmannia angulata abounds. Plants 14
to 2 feet high carry six to a dozen large, rosy-pink, foxglove-
like flowers. The local name is “ Féng-tang Hwa” (Honey-
bee Flower).

“ Cow-flat ’’ is a tiny place of about a dozen houses. Our
quarters were cramped but comfortable, and the people very
nice. There is a road from this hamlet to Nanto, distant
30 li. When I first visited this place in rgor I was an object
of great curiosity from the moment of my arrival to the time
of departure. I have been here several times since and am
now treated as an old-time acquaintance.

It was quite cool during the night, and a blanket was
required. At Ichang the very thought of a blanket was
enough to bring forth perspiration! We left about 6a.m., and
after ascending and descending a series of lateral spurs finally
reached the small river which enters the Yangtsze at Nanto.
After ascending this river for a few miles we commenced a
steep ascent. Now by an easy and then by a heavy grade the
road winds in and out among the mountains, and we did not
reach our halting-place for the night until6.30p.m. The last
coolie arrived an hour later. The length of the whole journey
is supposed to be only 60 li, but we all agreed that it is a good 70.
Whatever the distance, it is certainly a hard day’s travel.

The mountain sides are very steep, with razor-like ridges.
Terraced cultivation is everywhere carried out, rice is cul-
tivated in the bottom-lands and maize on the slopes, with
occasional patches of Irish potato. Where it is too steep, or
for other reasons unsuitable for cultivation, the mountain-sides
are covered with shrubs and trees, chiefly scrub Oak and the

HIGHWAY LEAD THROUGH A NATURAL TUNNEL

IN QUEST OF FLOWERS 31

common Pine. Small trees of Cornus Wilsoniana in full flower
were common here and there. Odd trees of C. Rousa, also, in
full flower, were conspicuous in the outskirts of the woods and
copses. This small tree is exceedingly floriferous. In habit it
is flat-topped with horizontally-spreading branches, and the
flowers borne erect, well above the foliage. The white bracts,
which are so conspicuous, frequently exceed 5 inches in
diameter ; with age they become tinged with pink. The fruit is
large, orange-red, and edible. This Chinese form will probably
prove a better plant under cultivation than the Japanese form
with which gardeners are familiar. The plant loves a sunny,
well-drained situation. But the display of the day was made
by the wild Roses. By the side of streams the Rambler Rose
(Rosa multiflora), with both white and pink flowers, was abund-
ant. In the woods higher up the Musk Rose (R. moschata)
filled the air with its soft fragrance. Here and there occurred
Actinidia chinensis, scaling tall trees and wreathing them
with white and buff-yellow fragrant flowers. In the forenoon
noted Rehmannia angulata, especially common on steep stony
places in full sun.

Our halting-place, Lao-mu-chia, is about 3500 feet altitude,
and consists of about six houses and a tile-factory. Hereabouts
much charcoal is burnt for export to Nanto and down river.
During the day’s journey we met several men laden with bales
of Pear and Crab-apple leaves. These leaves are commonly
used as a substitute for tea, and there is a considerable export
from these parts to Shasi.

On leaving Lao-mu-chia we immediately commenced the
steep ascent of the Hsan-lung shan, and a climb of 1000 feet
brought us to the summit, where there is a small temple ina
ruinous condition. After a precipitous descent of a few
hundred feet the road meanders over and among the tops of
hills, composed of granitic-gneiss, which is rapidly disintegrat-
ing, and ultimately descends to the bed of a torrent and joins
the main road from Ichang to Hsingshan Hsien.

Near the summit of Hsan-lung shan, which is composed
of Cambrian-Ordovician limestones, the Chinese Tulip tree
(Liriodendron chinense) is common in the woods, and so is
Viburnum tomentosum with its sprays of snow-white flowers.

32 A NATURALIST IN WESTERN CHINA

Styrax Hemsleyanum and Amelanchier asiatica, var. sinica, the
June berry, are other trees with white flowers remarkable for
their beauty and abundance of blossom. On the more open
slopes Symplocos crategoides, Lomicera Maackii, var. podocarpa,
Diervilla japonica, and Crategus cuneata made a fine display.
Thin woods of Pinus Massoniana and Sweet Chestnut (Castanea)
also occur ; the Pine trunks are gashed for the ultimate purpose
of producing kindling wood. In open places Rubus corchori-
folius abounds, and its red, raspberry-like fruits with their
delicious vinous flavour were good eating. In the descent
Dipteronia sinensis, a small bushy tree with erect trusses of
small white flowers, occurs, and Actinidia chinensis is common.
The hermaphrodite and male forms of this climber have large
white flowers quickly changing to buff-yellow, and the fragrance
is verypleasing. A form with purely female flowers isunknown.
At the foot of the descent we joined the main road from Ichang |
to Hsingshan Hsien, and following this route we reached
Shui-yueh-tsze, a village of 100 houses, situated in a tiny rice
flat, at five o’clock. The people were very inquisitive, and I
held an impromptu reception until bedtime.

On joining the main road, we saw evidences of the survey
for the Hankow-Szechuan Railway. The proposed route was
marked by bamboo poles, and on the rocks with Arabic
numerals and initials in Roman letters.. The route descends
a stream, just before reaching Shui-yueh-tsze, to Liang-ho-
kou, and then continues down the Hsingshan River to the
Yangtsze, which it connects with at Hsiang-che. Its con-
struction even in this region promises to be a difficult task,
and will call for great ability on the part of the engineers.
Much tunnelling and blasting will be necessary, yet from
Hankow to this point the task is simple compared with that
which lies beyond. The cost will be enormous even in a
land of cheap labour. It is highly improbable that the gentry,
who are so violently opposed to the employment of foreign
capital in this venture, realize the magnitude of the task and
its ultimate cost.

The next day’s journey proved interesting but arduous.
By an undulating path we reached the top of the ridge, which is
known as T’an-shu-ya (Lime tree Pass), from a gigantic Linden

IN QUEST OF FLOWERS 33

which occurs there. This tree (Tilia Henryana) is about 80
feet tall and 27 feet in girth, and though hollow appears to be in
good health. The young leaves are silvery, and the tree, from
its size, is a conspicuous object for miles around.

Descending through a cultivated area we entered a glen
which we followed for 20 li: the scenery in the lower end
is magnificent. Cliffs of hard limestone rear themselves almost
perpendicularly some 2000 feet and more. In the upper part
of the glen Pterocarya hupehensis is common alongside the
burn. An odd tree or two of the rare Pteroceltis Tatarinowit
also occurs here. Throughout the glen Lady Banks’s rose
(Rosa Banksia) is especially abundant. Bushes Io to 20 feet
high and more through them were one mass of fragrant white
flowers. It occurs in thousands and is particularly happy,
growing on rocks and over boulders by the side of streams.
Cesalpinia sepiaria, with erect thyrsoid panicles of fragrant
yellow flowers, is also abundant hereabouts. Growing on the
cliffs, Illictum Henryi, with its dull crimson flowers, is also
worthy of note. On issuing from the glen we struck a shallow,
rock-strewn stream of considerable width, and after ascending
it for a short distance made a very precipitous ascent of a
couple of thousand feet. Crossing over a ridge and a flat area,
a descending road led to Shih-tsao-che, which we reached as
night was closing in. This hamlet consists of about a dozen
houses scattered through a narrow valley.

During the day I collected specimens of thirty different
kinds of woody plants. The striking plants of the afternoon’s
journey were the Amelanchier and Dipelta floribunda, both
masses of flower. Walnut (Juglans regia) and Varnish trees
are abundant above 3000 feet; the sides and tops of the
mountains are clothed with woods of Oak and Pine, particu-
larly the former. We also saw many fine Willow and Ailan-
thus trees. Primula obconica, Lysimachia crispidens, and a
blue-flowered Salvia are abundant up to 2000 feet. Near
the inn a few trees of Catalpa Fargesii occur, but were not yet
in flower. Hereabouts Daphne genkwa is abundant, but it
was scarcely in flower at this altitude.

It rained a little in the early morning and showers fell at

intervals during the day, nevertheless, the weather was good
VOL. I.—3

34 A NATURALIST IN WESTERN CHINA

for travelling, since it was not too hot. Most of the journey
was downhill. Soon after starting in the morning we crossed
one or two low ridges, intercepted by narrow plateaux, and
about noon commenced the descent to Hsingshan Hsien. The
descent is precipitous in parts, but the mountain-sides are
mostly under cultivation. About half-way down coal is
mined, but the quality appears to be indifferent. Lime is
burnt in small quantities and paper-mills occur near Hsing-
shan.

Hsingshan, the only district city in these wilds, may claim to
be one of the smallest and poorest Hsiens (i.e. cities of the fourth
class) in the whole of China. It is situated on the left bank of a.
stream and contains scarcely a hundred houses, most of which
are in a ruinous state. The wall facing the river varies from
4 to 12 feet in height. A road, apparently the main road, runs
along the top of this wall. The east gate is closed by sewage ;
the north gate is so low that one has to bend the head when
passing through! The whole town is dull and lifeless, as far as
business is concerned, but children are plentiful, as they are
everywhere else in China. The town is backed by a steep
mountain, up two sides of which a wall is carried: most of
the mountain-side enclosed within the wall is given over to
terraced fields. The river is broad, with a shingly bottom, and
the water clear andlimpid. Thick-bottomed boats ply between
Hsiang-t’an and Hsiang-che, a village at the head of the
Mitan Gorge, on the Yangtsze. No one stays in Hsingshan,
and we journeyed on to Hsiang-t’an. This name signifies
“ fragrant rapid’’: the waters may perhaps be sweet, but the
village is foul and stinking. We had some little difficulty in
securing lodgings, poor as they were, and an objectionable coolie
had to be evicted before we could settle down for the night.

Flowers were not common during the day. We passed
a magnificent tree of Keteleeria Davidiana, 80 feet tall and 16
feet in girth. This tree shelters some graves, and was probably
planted long ago. In the descent we passed through orchards
of Crategus hupehensis, all in full flower. This Hawthorn is
one of several kinds cultivated in China for their edible fruit.
The interesting Torricellia angulata occurs sparingly, and here
and there large plants of Mucuna sempervirens cover large

IN QUEST OF FLOWERS 35

trees. Catalpa ovata is common on the plateaux and an
interesting small-leaved Poplar occurs around farmhouses, but
is rare.

Hsiang-t’an being in water communication with the Yang-
tsze boasts quite a considerable trade. Medicines are the
principal export. Rifle-stocks, roughly shaped out of Walnut
wood, are exported from this neighbourhood to Hangyang
in increasing quantities annually. They are worth locally 300
cash (about 6d.) each. The village is situated on the left bank
of the river, and possesses an Opium Likin and a Viceroy’s
Bank. Pigs seem more in evidence than human beings, as
judged from the four visits I have paid the place in different
years. Being only 300 feet above Ichang, Hsiang-t’an enjoys
a hot, dry climate.

Leaving Hsiang-t’an we immediately crossed the river by
ferry and ascended a narrow valley, which soon becomes a
ravine and finally a wild, entrancing gorge. At the head of
this gorge we took a small mountain-path which entailed a
severe climb from the river-bed to the tops of the surrounding
mountains. In this ascent the Musk Rose was a wonderful
sight, and Lovopetalum chinense abundant but out of flower.
Once on top of the mountains an undulating path leads to Peh-
yang-tsai, where we found lodgings in a new and fairly clean
farmhouse.

In the gorge I gathered Rehmannia Henryi, a herb less than
I foot tall, with large, white, foxglove-like flowers. Here-
abouts the root-bark of Lady Banks’s rose is collected, and
after being dried is pressed into bales for export to Shasi.
This bark is used for dyeing and strengthening fish-nets, and it
is claimed that it renders the net invisible to fish. In the
valley Kelreuteria bipinnata occurs, but is rare; the flora of
the ravine generally is similar to that of the San-yu-tung glen.

The mountains are clad with Oak (largely scrub), Pinus
Massoniana, and Cypress. A few Keteleeria trees occur and
also Liguidambar formosana. Populus Silvestrii, with its light
grey bark, is a very common tree hereabouts. Wood Oil trees
were a wonderful sight and most abundant. In the ravine
they were in full leaf, and the fruits were swelling, but from 1500
feet to 3000 feet they were leafless and covered with flowers.

36 A NATURALIST IN WESTERN CHINA

By the side of streams at low altitudes the Rambler Rose
(Rosa multiflora) was a pretty sight with its white and pink
blossoms, but the Musk Rose (R. moschata) was the flower of the
day—bushes 6 to 20 feet tall and more in diameter, nothing
but clusters of white fragrant flowers. Growing on some old -
graves I found a sulphur-yellow flowered form of Rosa Bank-
si@; this, I think, must have been planted. Rose bushes are a
special feature in this region and numerically are the common-
est of shrubs. Around our lodgings the Hardy Rubber tree
(Eucommit ulmoides) is cultivated for its bark, which is
valued as a tonic medicine.

Peh-yang-tsai is a scattered hamlet, situated in a narrow
valley, some 2500 feet altitude. Facing our lodgings is a
massive peak called Wan-tiao shan, its face a sheer precipice
of hard limestone, the summit and farther slopes apparently
well forested. The people of this hamlet, like the country
people everywhere in these parts, were extremely nice and
obliging, and it was a real pleasure to be amongst them.

Wan-tiao shan looked too tempting to be passed by without
investigation, so we spent a day, and a very hard day too,
in making its ascent and descent. Leaving our lodgings at
8 a.m., several hours were occupied in rounding the spurs and
surmounting the cultivated and scrub-clad land which subtend
the mountain proper. At 6000 feet we reached Bamboo
scrub, and a path through this led to an area where medicinal
Rhubarb was cultivated, and where the drug “ Tang-shén ”’
was extraordinarily abundant. At 6500 feet we entered the
timber. At the margin of the woods, to the left of the road,
are extensive plantations of the drug ‘‘ Huang-lien.’’ This
interesting plant (Coptis chinensis) is grown under a frame-
work of brushwood reared some 3 to 4 feet above the ground.
The drug is used as a tonic and blood-purifier.

As the path winds the trees are at first small, with plenty
of Bamboo scrub, but this belt is very narrow and speedily
gives place to large trees which extend to within 500 feet of the
summit, where Bamboo scrub again becomes troublesome.
Everywhere above 5000 feet, where the woods are thin and
sunlight penetrates freely, Bamboo scrub is found, rendering
travel excessively arduous and, unless a path is cut, im-

“La 001s ‘NVHS OVIL-NVA

IN QUEST OF FLOWERS 37

possible. In the dense shade of the forest the Bamboo does
not thrive.

The forest, though full of splendid timber, is not rich in
variety. The Chinese Beech (Fagus sinenis) is the commonest
tree. This species always has many trunks, and trees 60 to
70 feet high, with stems 3 to 6 feet in girth, abound. The
interesting Tetracentron sinense is very abundant ; trees 60 to
70 feet by 8 to ro feet girth are plentiful. The leafage of this
tree is very thin and characteristic. Large trees of White
Birch and of several species of Maple occur scattered through
the forest. The smooth-leaved Davidia (D. tnvolucrata, var.
Vilmoriniana) occurs sparingly, and good-sized trees of various
Cherries, Bird Cherries, Mountain Ash, and Wild Pear are

‘common. Rambling over the tops of the largest trees is
Berchemia Givaldiana. Several species of Rhododendron
occur ; one species (R. sutchuenense) forms a tree 30 feet and
more tall and 5 feet in girth. Shrubs in variety abound ; in
the glades Viburnum tomentosum was wreathed in snow-white
flowers. In more open places the Musk Rose is rampant,
and near the summit Rosa sericea is abundant.

The summit forms a sloping, undulating flat, about an
acre in extent, covered with grass and a few shrubs. On the
apex stands a small temple now partly in ruins. A sharp,
rocky ridge extends from the summit, linking the mountain
up with the ranges to the northward. The face on two sides
is a vertical precipice, 2000 feet and more sheer. From the
summit (alt. 7850 feet) we got an extensive view of the sur-
rounding country. Nothing but mountains on every side ;
to the north and north-west these are heaped one beyond
another in quick succession and are separated by narrow
defiles down which torrentsrush and roar. Very difficult looked
the country in front of us, but the call of the unknown was
strong. We descended by the same devious path, indeed, there
is no other, and reached our lodgings as darkness overtook us.
Specimens of some forty odd different plants rewarded the
day’s labour, several of them new and uncommonly interesting.
On the extreme summit Box is a common shrub, and growing
with it I discovered a new species of Lilac (Syringa verrucosa).

The following day we continued our journey northwards.

38 A NATURALIST IN WESTERN CHINA

Just beyond Peh-yang-tsai we passed through copses of
small Oak (Quercus variabilis), where the Jew’s ear Fungus is
cultivated. The culture is as follows: Oak saplings, about
6 inches thick, are cut down, trimmed of their branches, and
cut into staves 8 to 10 feet long. These are allowed to lie on
the ground for several months, where they become infested with
the mycelium of the fungus. They are then stacked slantingly
in scores or thereabouts, and the fructifications of the fungus
develop. These are ear-shaped and gelatinous and are by the
Chinese esteemed a delicacy. I tried them, but did not find
them very palatable, and the experiment resulted in a bad
stomach-ache !

On leaving these plantations the road descends to a ravine
along which it meanders for a mile or two. Many shrubs were
in flower in the ravine, and I gathered amongst other things
specimens of a new genus, allied to Holbcellia, with fragrant
yellow flowers. (I subsequently secured seeds of this plant,
since named Sargentodoxa cuneata, and succeeded in intro-
ducing it into cultivation.) At the head of this ravine a
steep ascent through woods of Oak and Birch leads to a
cultivated area where there are two or three scattered houses
and many Tea bushes. Near one house the Chinese Coffee
tree (Gymnocladus chinensis) occurs; the pods of this tree are
saponaceous and are esteemed for laundry purposes.

From the Tea plantations the road leads through Pine
woods, now by an easy, now by a heavy grade, but always
ascending, and we were all glad when our destination (Hsin-
tientsze) wasreached. Near this place are some fine old woods,
rich in a variety of deciduous trees and shrubs. I noted a
Horse Chestnut (#sculus Wilsonit), two kinds of Beech,
Styrax Hemsleyanum, Meliosma Veitchiorum, the Davidia, and
many different kinds of Maple and Oak—all of them large
trees. In the margins of the woods Viburnum ichangense was
particularly fine, and many Cherry trees, with both pink and
white flowers, common. In moist shady places in the woods
a blue Primrose (P. ovalifolia) carpets the ground for miles.
The yellow-flowered Stylophorum japonicum, an Epimedium,

and various species of Corydalis are abundant in and near the
woods.

IN QUEST OF FLOWERS 39

The hamlet of Hsin-tientsze, alt. 5600 feet, consists of one
rather large house. It is built on a slope a few hundred feet
below the summit of the ridge, and from the front of the house
a wonderful view of the surrounding country is afforded.
Nothing but mountains as far as the eye can range, and
not 20 square yards of level ground in sight! Our quarters,
though cramped, were, all things considered, fairly comfortable,
and were as good as could be expected in these wilds.

The next morning we made an early start in order to cover
the 60 li between Hsin-tientsze and Mao-fu-lien. Immediately
on leaving we traversed an old wood especially rich in species
of Maple. Davidia and Beech are also common, whilst the
interesting Cornus sinensis occurs sparingly as a thin tree
60 feet tall. Pinus Avmandi is present, but Conifers generally
are very scarce in this particular locality.

We meandered around the mountain-sides, by a tortuous
ascending path, until we reached a gap in the ridge and crossing
over made a breakneck descent of a couple of thousand feet.
A new kind of Poplar, having the young foliage bronzy-red, was
common on all sides, and in the descent I gathered Primula
violodora, Rhododendron Augustinii, Acer griseum, and a pink-
flowered Staphylea, the last two both small trees. The most
interesting find, however, was anew Hydrangea (H. Sargentiana),
a shrub 5 to 6 feet tall, with stems densely felted with short
bristly hairs and large, dark green leaves with a velvety lustre—
in foliage alone this species is strikingly handsome.

At the foot of the descent we came upon small woods of
Pinus Henry, a tree averaging 60 feet in height, more or less
pyramidal in shape, with bark usually rough and black, but
sometimes red in the upper parts. The cones vary considerably
in size and are retained on the tree for several years. In the
valley near the Pine woods there is considerable cultivation.
Walnut trees are common and Cunninghamia abounds.

Leaving this valley, a long but fairly easy ascent led to the
top of another ridge, and a precipitous descent brought us to
another narrow valley. These ascents and descents were most
fatiguing and occurred with exasperating frequency every day,
and several times a day at that. Another climb of over 2000
feet and we reached our destination for the day, finding

40 A NATURALIST IN WESTERN CHINA

accommodation in an inn which is also a large medicine depot,
and is owned by a wealthy man from the province of Kiangsi.
This inn is a large, rambling two-storied structure with several
outhouses and a large courtyard. There is not sufficient level
space to accommodate the whole place, and the front part is
supported on posts. It serves as general store for the whole
country-side, and in addition is a veritable museum. Dirt in
every shape and form draped everything, and the stink from
adjacent piggeries was tempered with the odour of various
aromatic herbs. The business instinct of the house is strong, as
I found to my cost when changing some silver and buying a goat.
The rites of ancestor-worship were strictly carried out every
morning and evening, and everything done to ensure continued
and increasing prosperity. The burning of incense and candles
and the performance of mystical genuflexions may assist
business, but a little more attention to cleanliness and sanitation
would make a stronger appeal to the foreigner. At least, such
were my conclusions after a thirty-six hours’ stay in the place.

It rained a good part of the next day, but as we had decided
upon a day’s rest it did not inconvenience us. In the forenoon
I went out for a few hours to investigate the woods around
Mao-fu-lien. Some very large trees of Sassafras (S. tzumu)
occur here—the largest specimen is nearly roo feet tall and 12
feet girth. The Chinese Sassafras has no medicinal value, and
the wood is used for box-making and fuel only. Oak and
Sweet Chestnut are plentiful and form small woods. The
Chestnut (Castanea Vilmoriniana) is a singular species, with
a single ovoid nut inside the spiny fruit ; the flowers have a
peculiarly unpleasant smell. Around the inn are cultivated
many trees of the Hardy Rubber and also Magnolia officinalis.
Walnut and Varnish trees are abundant, and behind the house
is a fine flat-leaved Spruce (Picea pachyclada). The mountain-
tops are clothed with Grass, Brambles, scrub Oak, bushes
of the pink-flowered Rhododendron Mariesii, and the scarlet
R. indicum.

The view from the inn is one of steep ridges and high
mountains, separated by deep, narrow chasms as far as the
eye canrange. It is indeed a fascinating country, but exhaust-
ing to travel over.

IN SAN-YU-TUNG GLEN, CUPRESSUS FUNEBRIS IN FOREGROUND

IN QUEST OF FLOWERS 4I

A fine morning followed yesterday’s rain, the country looked
refreshed, and the air was laden with fragrance from the myriad
flowers on every side. The coolies grumbled loudly over the
extortionate charges at the inn, and several hours elapsed before
they recovered their cheerfulness. The day’s journey com-
menced in a steady ascent to the top of a ridge followed by the
usual precipitous descent. Hereabouts Staphylea holocarpa, a
small, very floriferous tree, with both white and pink flowers,
is very common and most strikingly beautiful. Another in-
teresting plant is Salix Fargesit, a dwari-growing Willow,
having large very dark green leaves. A small torrent marks
the foot of the descent, and from this point on we occupied
several hours in an exhausting climb to the summit of another
ridge, finally crossing over at 7300 feet altitude. In the ascent
a new Spruce, having short square leaves and small cones, was
discovered, and many small trees of Hemlock Spruce were
noted. Near the head of the ridge, on cliffs, Box (Buxus
microphylla) is very common, and a rosy-red flowered Primrose
is abundant in grassy places. A dwarf Bamboo forms dense
thickets on the top of the wind-swept ridge.

The descent quickly leads into copses of Birch, and later into
fine woods composed of mixed deciduous trees and shrubs and
a few conifers. In these woods we spent a profitable time,
collecting in all specimens of some fifty different kinds of woody
plants. We saw one or two large trees of Davidia and many
of Tetracentron. Cherries in variety are plentiful, and were a
wonderful sight—nothing but masses of pink and white. Three
kinds of Rhododendron were collected, and six in all noted.
Maples in variety are very common, but one large tree ot Acer
griseum, with its chestnut-red bark, exfoliating like that of the
River Birch, was the gem of all. Various Pomacee and one or
two species of Lauracee make up a fair percentage of the small
trees. Viburnumsin variety, Honeysuckles, Diervillas, Deutzias,
Philadelphus, and Neiillia sinensis are everywhere abundant.
In rocky, more open places Viburnum rhytidophyllum with its
long, thick wrinkled leaves looked particularly happy, and
in places exposed to the sun a Crab Apple (Malus sp.) with
pink flowers was a sight for the gods. On wet, humus-clad
rocks Pleione Henvyi luxuriates, and herbs in endless variety

42 A NATURALIST IN WESTERN CHINA

crowd every available spot. A fine torrent collects up the
waters of countless smaller streams, and falls down the narrow
ravine, often in a series of waterfalls hundreds of feet high, the
noise of the falling water alone breaking the silence of the forest
depths.

With some difficulty, owing to the timidity of the people,
we obtained lodgings in a peasant’s hut at Wén-tsao, alt.
6150 feet. This tiny hamlet consists of four small houses,
scattered and pitched on the steep mountain-slope. It is
surrounded on all sides by precipitous mountains covered with
forests. Around the houses small patches have been cleared,
and wheat, a little maize, and a few peas and vegetables are
cultivated.

The forests of this region are particularly rich, and in order
to better appreciate them I propose to interpolate here extracts
from my journal of another date :—

“ May 30.—Weéen-tsao. On a precipitous slope facing our
lodgings a score or more Davidia trees occur; they are one
mass of white, and are most conspicuous as the shades of
night close in. Two large trees of Pterostyrax hispidus are
growing amongst these Davidias, and are laden with pendu-
lous chains of creamy-white flowers.”

““ May 31.—Go over and investigate the Davidia trees
and the forests generally. Crossing a narrow neck a wood-
cutter’s circuitous path leads us down to a narrow defile
through a fine shady wood. Ascending a precipice with diffi-
culty, we soon reach the Davidia trees. There are over a
score of them growing on a steep, rocky declivity ; they vary
from 35 to 60 feet in height, and the largest is 6 feet in girth.
Being in a dense wood they are bare of branches for half their
height, but their presence is readily detected by the numerous
white bracts which have fallen and lie strewn over the ground.
The tree starts up from below when felled ; indeed, it naturally
throws up small stems after it gets old. The bark is dark
and scales off in small, irregular flakes. By climbing a large
Tetracentron tree growing on the edge of a cliff, and chopping
off some branches to make a clear space, I manage to take
some snapshots of the upper part of the Davidia tree in full
flower. A difficult task and highly dangerous. Three of us

IN QUEST OF FLOWERS 43

climb the tree to different heights and haul up axe and camera
from one to another by means of arope. The wood of Tetracen-
tron is brittle, and the knowledge of this does not add to one’s
peace of mind when sitting astride a branch about 4 inches
thick with a sheer drop of a couple of hundred feet beneath.
However, all went well, and we drank in the beauties of this
extraordinary tree. The distinctive beauty of Davidia is in
the two snow-white connate bracts which subtend the flower
proper. These are always unequal in size, the larger usually
6 inches long by 3 inches broad, and the smaller 34 inches
by 24 inches; they range up to 8 inches by 4 inches and
5 inches by 3inches. At first greenish, they become pure white
as the flowers mature and change to brown with age. The
flowers and their attendant bracts are pendulous on fairly
long stalks, and when stirred by the slightest breeze they
resemble huge Butterflies hovering amongst the trees. The
bracts are somewhat boat-shaped and flimsy in texture, and the
leaves often hide them considerably, but so freely are they
borne that the tree looks, from a short distance, as if flecked
with snow. On dull days and in the early morning and
evening the bracts are most conspicuous. The fruit super-
ficially resembles a small walnut, but the inner shell is abso-
lutely unbreakable. To my mind Dawvidia involucrata is at
once the most interesting and beautiful of all trees of the
north-temperate flora.

“With the Davidia is a good-sized tree of the Horse
Chestnut (50 feet in height by 4 feet in girth). Higher up
Hornbeam and Tetracentron are common, and Birch, white,
red, and black, luxuriate.

““ Maples are a feature of these woods; all are tall trees, but
of no great thickness. Unfortunately very few are flowering,
and indeed this is true of the forest trees generally this year.

“‘ Perhaps the commonest tree in these forests is the Beech ;
parts being formed entirely of these trees. So light-demanding
are they that they suffer no competitors or even undergrowth.
For the first time it is possible for me to say definitely that
two distinct species of Beech exist in this region. One forms
a tree with a single trunk, the other always has several trunks.
The former species has glabrous, shining green leaves, a large,

44 A NATURALIST IN WESTERN CHINA

dense, much-branched head ; it makes a tree 40 to 50 feet
high with a trunk 5 to ro feet in girth, and, save for its smaller
stature, very strongly resembles the European Beech. The
second species, which is the recognized Chinese Beech, grows
much taller, but never attains the girth of the other. It
generally has six to twelve trunks, averaging 2 to 5 feet in girth,
arising closely together and slanting away from one another
as they grow. The bark is light grey and the leaves sub-
glaucous and hairy below; branches somewhat ascending
but with the young branchlets slender and pendulous. A
local name for the Beech is ‘ Peh Litzu.’ Small plants are
common, but no flowers are to be discovered.!

“Tn the shade of trees, Ribes longeracemosum, var. Wilsonitz,
a remarkable black currant, with racemes 1 to 14 feet long, is
common, whilst Rodgersia esculifolia, with large, erect, thyrsoid
panicles of white flowers, is rampant.

“‘ Five species of Oak—three deciduous and two evergreen
occur. Meliosma Veitchiorum and many species of Pomacee.
and Cherries are common, whilst the Varnish tree is every-
whereabundant. In dense shade various evergreen Barberries
occur,and in open country Neillia sinensis forms dense thickets.

“ Of Conifers, Pinus Armandi and P. Henryi are scattered
over the cliffs; Picea Wilsonit and a flat-leaved Spruce (P.
pachyclada) are rare, whilst the Hemlock Spruce? is fairly
common on the cliffs—neat, dense trees of no great size with
their young leaves just unfolding and old cones abundant.
The White Pine (P. Armandz) is more common higher up on
the mountains ; withits long needles, graceful port, and light
grey bark this tree is strikingly handsome; the cones are
pendulous, borne at the ends of the glabrous branches. The
very resinous wood is used locally for torches, burning with a
clear, bright flame, and gives a good light.”’

1In 1910 I succeeded in introducing young plants of both species into

the Arnold Arboretum from this region.
2 Tsuga chinensis.

CHAPTER V
FOREST AND CRAG

ACROSS THE HUPEH-SZECHUAN FRONTIER

N leaving Wén-tsao a sharp descent for a couple of
() hours brought us to the upper waters of the Hsingshan
River, which we left several days ago. Crossing this
stream by a covered bridge we reached the hamlet of Li-erh-kou.
Around this hamlet trees of the Hardy Rubber (Eucommia) and
Magnolia officinalis (Hou-p’o) are cultivated for their bark. A
steady ascent from Li-erh-kou through occasional woods of Oak
and Birch, interrupted by areas where people were busy plough-
ing the fields andsowingmaize, brought us to the hamlet of Chin-
tien-po, wherewe lunched. Nearthis place is a fine new Meliosma
(M. Beaniana), atree 60 feet high. It was leafless, but one mass
of creamy-white flowers borne in pendulous panicles. Near by
this tree I discovered one small specimen of the ‘‘ Judas tree ”’
(Cercis racemosa). Prior to this discovery I knew of only two
trees some fifteen days’ journey south-west of Ichang. This new
tree was about 25 feet high, with a stem half decayed through
at the base, and a mop-like head. In spite of its partial decay
the tree appeared in vigorous health, and was one mass of
silvery-rose coloured flowers, borne in short racemes. The
leaves of this species are hairy below. Varnish and Walnut
trees occur in abundance, and we met several coolies laden
with cakes of fat, expressed from the fruits of the Varnish
tree (Rhus verniciflua). The double-flowered form of Spirea
pruntfolia is commonly planted on graves, and the bushes were
wreathed in flowers.
Soon after leaving Chin-tien-po we commenced a precipitous
ascent, and after climbing for several miles reached the neck of

a ridge where Viburnum rhytidophyllum laxuriates. From this
45

46 A NATURALIST IN WESTERN CHINA

neck the ascent is more gradual, and but few crops are grown,
as it is nearing the limits of cultivation in these regions. Near
some limestone cliffs are two magnificent trees of Maackia,
each 60 feet tall and 7 feet in girth. The bark of this tree is
smooth, of a light grey colour, and the unfolding leaves are
silvery grey. Here, too, are many small trees of the Bladder
tree (Staphylea holocarpa) and Peach bushes. These were in full
flower, and flitting amongst the flowers and drinking in the
honey were many beautiful little sun-birds (4thopyga dabry1).
Rhododendron indicum was left behind at 5500 feet altitude.

A few hundred yards beyond the limestone cliffs we crossed
over at 7000 feet altitude, into Fang Hsien, and traversed a
narrow moorland valley clothed with grass and bounded by
rounded hills covered with thickets. In this moorland are acres
of Astilbe Davidii and A. grandis, with several Senecios and
other ornamental herbs. The thickets are composed chiefly of
Birch and Willow, with a few Poplar and Silver Fir, and an
occasional flat-leaved Spruce. The vegetation was scarcely in
leaf, and it was evident from the appearance of the ground that
snow had only just melted away. We flushed a Solitary Snipe
and secured a cock pheasant for the larder, but very little
life of any sort was visible in these uplands. At the head of
this moorland valley we entered a narrow defile and, after
skirting the side of a mountain through thickets in which
various Maples and Currants were prominent, reached Hung-
shih-kou. This is a miserable hut of wood in a half-ruinous
condition, kept by a family clothed in rags. It is situated
at an altitude of 6300 feet, by the side of a considerable torrent,
and is walled in by precipitous, well-wooded mountains.

At night some of the coolies slept in a loft above the room I
occupied, and every movement they made caused dust and dirt
to fall over my bed. Ox waking in the morning I found myself
covered with this filth, and nearly choked with the dust into
the bargain. The owner of this hovel is a hunter, and he has
shot the Serow of this region, which is known as “ Ming-tsen
Yang.’ He hadacouple of pairs of horns and a flat skin which
we secured, and, judging from this fragmentary material, the
beast must be larger than any known species of Serow. (In
1907 my associate, Mr. Zappey, made several trips after this

THE CHINESE PRIVET (LIGUSTRUM SINENSE) 20 FT. TALL

FOREST AND CRAG 47

animal, but to no purpose, though he secured a tantalizing
glimpse of just one specimen.)

The name Hung-shih-kou signifies ““ Red stone mouth,” and
has reference to the outcropping of red sandstone which occurs
here and extends to Hsao-lung-tang, 20 li distant, which we
made our halting-place for the next day. Though we had only
20 li to cover we started early, glad to escape from the miserable
lodgings into the woods again. Ascending a stream, through
brushwood thickets composed of Willow, Birch, Spirea, and
Roses, we twice crossed the stream by rotten bridges of roughly
hewn tree-logs before reaching our destination. On the way
we passed several fine trees of Pzcea Wilsonit, beneath which old
graves nestle. The largest trees are about 70 feet tall and
6 feet in girth ; the leaves bright green, and the habit distinctly
stately ; the cones are borne in large clusters, and many still
remained on the trees. Here also are small trees of the White
Pine (Pinus Armand) with cones g inches long. A new Poplar
was discovered in flower, and Veitch’s Viburnum and Spirzea
were common with their young leaves just unfolding.

The handsomest tree in these parts is, however, the Chinese
form of Betula utilis, a Birch with orange-red bark, which on
exfoliating exposes the glaucous waxy bloom of the layer below.
Trees 40 feet high are still pyramidal in habit, much branched,
with slender, ascending branches on which the lenticels are very
prominent. The older trees, as seen on the tops of the moun-
tains, are mop-headed, 60 to 8o feet tall, with a clean trunk for
40 feet more, and are still strikingly handsome though blown
and battered by the wind.

The hamlet of Hsao-lung-tang (Small Dragon-pool), alt.
7400 feet, consists of two dilapidated wooden huts pitched on
opposite sides of a lovely burn, which flows through a narrow
sloping valley lying almost due east and west. This valley is
flanked by steep ridges clad only with grass and scrub. Odd
patches of Birch and Silver Fir attest to forests which have all
been destroyed by fire. From the numerous old graves and
abandoned fields it is evident that formerly more people dwelt
in this valley than do so to-day. Tiny patches of cabbage
and Irish potato occur around the huts ; and also plantations
of Tang-kuei (Angelica polymorpha, var. sinensis), a valued

48 A NATURALIST IN WESTERN CHINA

Chinese medicine. The people declare the valley too cold for
wheat or barley !

On the occasion of my first visit to this place in 1901, I had
to retrace my steps owing to dearth of supplies. Since that
date no white man had visited this region. In the direction in
which we were bound these are the last inhabited houses for
over a hundred li.

I took a photograph of the hostel on my arrival, but what
I should have liked to photograph was the interior. This was
impossible, since, even at midday, a light was necessary to see
into the farthest corners. Dirt and filth in many forms
abounded, and although plenty of timber is to be had for the
felling, the house, through the idleness of its keeper, has been
allowed to fall into a most ruinous state. Of one low story,
the house is bisected into four compartments, and is provided
with no outlet for the smoke or for the ingress of light, save
through the doorway and holes in the roof ; the floor, of course,
is mother earth. Pigs were quartered in one section, into which
our arrival also forced the owners. Cows and goats occupied
a hovel 6 feet from the door, the floor of which was fully a
foot deep in filth. Luckily, the weather continued gloriously
fine, and the miserable surroundings were less evident in con-
sequence. (In passing, I might record the fact that this was
the only occasion on which I enjoyed fine weather in this place.
Twice previously I had been marooned here for days, and either
stayed in bed or shivered by the doorway watching the rain.)

Bee-keeping is one of the principal industries of the peasants
in these wilds, and around this hostel are scores of bee-
hives. The hives are hollowed-out logs of Silver Fir, about
3 feet 6 inches long by 1 foot wide, two pieces of wood are fixed
crosswise in the centre, and opposite these three or four holes
are bored to allow the bees ingress and egress. Rude boxes
often take the place of these logs. The beeswax is not separ-
ated from the honey, the honeycomb being eaten as removed
from the hives. Though the climate is rigorous, the bees are
healthy and strong, and disease is unknown among them.

The morning following our arrival we ascended the Sha-
mu-jen range behind our lodgings. The first 500 feet was
steep going, but afterwards the climb was easy. At about

FOREST AND CRAG 49

8000 feet woods of Silver Fir occur. The trees at first are of
no great size, but their dimensions increased as we ascended.
Most of the larger trees have been felled and converted into
coffins ; the remains of thousands of them are scattered every-
where around. On the decayed trunks of many of these trees
large bushes of Rhododendron are growing, thereby proving
that the trees have lain there these many years past. Some
of the prostrate trunks measured over 150 feet in length and 6
feet in diameter. None of this size is now standing, but plenty
that are over 100 feet tall occur. The upper part of the
ridge is a cliff some 200 feet high, under the lee of which Birch
and Maple are common and wild Rhubarb is also found. We
discovered a more or less easy path up the cliff, and crossed over
at 9700 feet altitude. The highest peak in this range is prob-
ably a couple of hundred feet higher. The summit is of hard
limestone with rare outcroppings of red sandstone. Stunted
wind-swept Silver Fir and various kinds of Currant extend to
thesummit. Rhododendron and a dwarf Juniper (J.squamata)
are also common. The descent was through woods of Birch
and Bamboo to an open, grassy, scrub-clad, sloping moorland,
through which a considerable torrent flows. The Bamboo, so
common hereabouts, is very beautiful, forming clumps 3 to Io
feet through. The culms are 5 to 12 feet tall, golden yellow,
with dark, feathery foliage; the young culms have broad
sheathing bracts protecting the branchlets. Taken all in all,
this is the handsomest Bamboo I have seen.!

In the vicinity of the stream shrubs in great variety
abound ; of these the Willows, Roses, Spirzeas, Philadelphus,
Hydrangeas, odd bushes of Rhododendron Fargesit, and clumps
of Avalia chinensis are the principal features. The Rhododen-
dron referred to is one of the most beautiful, with compact
trusses of white or, more commonly, rosy-red (occasionally
deep red) flowers ; the leaves are small, displaying the trusses
of flowers to great advantage. This species is usually a bush
5 to 8 feet tall, and of about the same dimensions through the
head ; more rarely it is 15 to 20 feet tall. The steep grassy
slopes are almost devoid of trees; the fine pasture land
and the typical moorland character of this narrow valley

1 In toro I successfully introduced it into cultivation.
VOL. I.—4

50 A NATURALIST IN WESTERN CHINA

constitute a region that is very different from others in
central China.

In the afternoon we visited Ta-lung-tang (Large Dragon-
pool), a deep, silent pond about a stone’s throw across, nearly
circular in outline with reedy margins, walled in by steep
grassy mountain slopes. In short, in situation and appearance
the very kind of pool that in any country legends would be
wrapped around, and so in this case many curious stories con-
cerning elfs and demons are centred round this silent pool.
The day was gloriously fine and sunny, but the wind, which
swept through the valley in considerable force, was very cold.
Whether it be due to local conditions or to the altitude I could
not determine, but the tree flora is comparatively poor and of
little interest, and very unlike the belts that occur between
4000 and 6500 feet. The altitude, however, favours coarse
herbs, and these are rampant. Many interesting shrubs also
occur, but with the exception of Silver Fir, Birch and Poplar
trees are rare.

With a prospect of 60 li of unknown road before us we
planned a daylight start, but this scheme did not mature,
as the men had to prepare and cook their morning meal
before starting. The entire absence of food supplies makes
travelling hereabouts extraordinarily difficult. Yesterday
four of the men journeyed back 45 li in order to buy
food-stuffs, and returned only after dark; several of them
were up most of the night grinding maize and preparing
cakes for the march.

On leaving Hsao-lung-tang we ascended the lesser branch
of a stream through a narrow valley flanked by bare grassy
mountains having here and there small patches of Silver Fir
and Birch forest. The road is one steady climb, never steep
but often difficult owing to the Bamboo scrub. The decaying
stumps and stark tree trunks speak eloquently of the magni-
ficent forests which must have formerly existed here until
destroyed by axe and fire. To the botanist and lover of
Nature this vandalism is painful, but presumably it was
necessary for economic reasons. The unwitting cause of it
all has been the Irish potato. But Nature took her revenge
when, twenty-three years ago, the Potato disease devastated

CERCIDIPHYLLUM JAPONICUM VAR. SINENSE, 80 FT. TALL, GIRTH 7 FT.

FOREST AND CRAG 51

the crop and ruined the country-side, causing a general exodus
of all the people. Nature is fast reclaiming the whole region,
but re-afforestation is a slow process.

Nearing the head of the pass we entered large timber—
a fragment of the virgin forest, composed exclusively of
Silver Fir and Birch with a dense undergrowth of Rhodo-
dendron. The last named comprise four species—R. Fargesit,
R. maculiferum, R. sutchuenense, and R. adenopodum, most
of them bushes 10 to 20 feet tall, their flowers making one
blaze of colour. The Silver Fir and Birch trees are of huge
dimensions, but none was fruiting. On emerging from this
patch of forest we entered a rolling moorland covered with
Bamboo scrub which merges gradually into areas clad with
the dwarf Juniper, coarse grasses, and herbs, amongst which
a species of Onion was abundant. This moorland extends
across the rounded saddle of the range and for several miles
down the other side. The crest of the saddle I made 9500 feet
altitude, and from this point we obtained a fine view of the
series of bare, savagely jagged peaks from which the range
(Sheng-néng-chia) takes its name. The highest peaks prob-
ably exceed 11,000 feet altitude, and the lower slopes are
forested, but the country is not attractive. Animal life is
remarkable for its absence, and hardly a bird was to be seen.
The solitude which reigned in this remote, inaccessible region
was broken only by the noise of rushing waters and the low
whining of the wind amongst the tree-tops. In shady places
blocks of ice still remained, and about the head of the pass
the grass was only just beginning to show green. Save for an
alpine Primula and a Dandelion no flowers of any sort were
to be seen.

On crossing the pass we again entered Hsingshan Hsien,
and after wandering across moorland for a few miles a short
steep ascent led us across a lateral spur into Patung Hsien.
From this point a precipitous descent of 2000 feet brought us
to a ruined and deserted hut at a place called Wapéng, the
only accommodation the country-side affords. In the descent
we passed hundreds of curious rock-stacks—bare blocks of
shale standing erect, with acute edges, like gaunt sentinels
guarding the neighbourhood. The mountain-side was formerly

52 A NATURALIST IN WESTERN CHINA

under cultivation, but is now abandoned and covered with
grass and coarse herbs. Around the hut a little Medicinal
Rhubarb and much Tang-shén was growing, telling of former
plantations of these and other medicines. The country on all
sides is very steep and much cut up, but stark decaying tree
trunks, the sole remnants of former forests, mar the beauty
of the landscape on all sides.

We reached Wapéng (alt. 8400 feet) fairly early in the
afternoon, and the men were busy till nightfall collecting fuel
and rigging up a bamboo shelter beneath which to pass the
night. The day had been gloriously fine and the night proved
equally so, with a distinctly frosty nip after sundown. A
roaring fire made things look cheerful, and everybody was in
the best of health and spirits. The sides of the hut were airy
and the wind played about one all night. The roof was
partially wanting and afforded a good view of the starry
heavens above. It was a lonely place, yet one felt peculiarly
happy and glad to be privileged to visit a region so remote from
the world in general.

There was no difficulty in getting the men up next morning,
and we were off just as the sun’s rays broke over the landscape.
Dark mists obscured the view for an hour or so, but as the sun
rose these disappeared and we enjoyed another gloriously fine
day. A steep and precipitous, nay breakneck, descent of a
1000 feet brought us to a narrow well-wooded valley, walled
in by forest-clad mountains. The Silver Fir does not descend
more than 500 feet from Wapéng, below which its place is
taken by Hemlock Spruce. This Spruce is not plentiful, but
giants 100 feet tall by 12 feet in girth occur. The forests as
we descended quickly become of mixed character, and finally
conifers completely disappear. The variety of trees and
shrubs was astonishing, and nearly all the more interesting trees
of western Hupeh were to be found and in quantity. Maples
are particularly abundant, and I gathered specimens of a
dozen species in flower. Four species of Rhododendron occur
scattered, but not in quantity. On rocks in places an
interesting orchid (Plezone Henry) abounds and was one mass
of flowers. The Davidia is fairly common, and the curious
Euptelea Franchetii and Tetracentron sinense are the commonest

FOREST AND CRAG 53

of trees. A feature in these woods was Staphylea holocarpa, a
small tree covered with pendulous trusses of white and rosy-
pink flowers. A Horse Chestnut (4sculus Wuilsonit), the
Chinese Yellow-wood (Cladrastis sinensis), Hemsley’s Styrax
and Plerostyrax hispidus, all of them large trees, were fairly
common ; Cherries, Bird Cherries and many Pomacee abound.
Birch is one of the commonest constituents of these forests ;
in the more open areas Bamboo scrub forms dense thickets,
and high up in the woods Rhododendron maculiferum forms
trees 25 feet tall with a trunk 1 foot in diameter.

Here and there clearings have been made for the cultiva-
tion of the medicine “‘ Huang-lien’”’ (Coptis chinensis). In
one abandoned clearing were hundreds of Lilium tigrinum
luxuriating amongst the grass and tall herbs. In dark shady
places the noble Lilium mirabile is common. This lily has
tubular snow-white flowers spotted with red within, and
glossy green, cordate leaves. An occasional Spruce or Pine
tree occurs, and at the edge of the forests Cunninghamia
appears. Many of the cliffs are clothed with Hemlock Spruce.
Birch is fairly common, but, with the exception of one or two
evergreen species, Oak is very scarce. Hornbeam is not
plentiful, and Magnolias are decidedly rare trees; Ash is
general, and the Linden, represented by three or four species,
abundant. The Laurel family is represented by four species,
all of them deciduous, including a handsome kind with young
foliage of a bronze-red. Honeysuckles are rare, save for the
climbing species Lonicera tragophylla, which has golden-
yellow flowers. Clematis in variety are common, especially
C. montana (white and rosy-red forms) and C. pogonandra
with its top-shaped yellow flowers. Many species of Schis-
andria, all of them a wealth of flowers, Holbellia Fargesii,
and the botanically interesting Sinofranchetia sinensis are the
principal climbers.

The road follows the course of a torrent which rises near
Wapéng and quickly becomes a considerable stream. The
path is narrow, very rocky and difficult to follow, and how
our chairs got through was a puzzle. Both torrent and path
ultimately plunge into a narrow ravine shut in by lofty cliffs,
unclimbable and bare. In places the rocks are of limestone,

54 A NATURALIST IN WESTERN CHINA

but from 5000 feet downwards slate and mud-shales pre-
dominate.

At 4500 feet altitude we reached the edge of the forest
and entered a cultivated area, where there are a few inhabited
houses—the first we had seen for two days. Barley and Irish
potato are the crops. Near the edge of the forest the torrent
flows underground for about a mile. On rocks here Lomicera
pileata abounds as a fluviatile shrub; the curious climber
Hosiea sinensis is common, covering rocks fully exposed to the
sun. In the open country I noted in full flower a fine
specimen of the Chinese Tulip tree (Liviodendron chinense),
70 feet tall and 5 feet in girth.

A precipitous descent, through fields margined with Tea-
bushes, led to the tiny hamlet of Sha-kou-ping, where the torrent
we had followed joins with a very considerable stream flowing
down from the north-east. The united waters plunge at once
into a ravine and finally enter the Yangtsze a few miles
above the city of Patung. Sha-kou-ping is only 2600 feet
above sea-level, and is hemmed in on all sides by lofty
cliffs. The flora is that common to the glens and gorges
around Ichang, and the wealth of flowers was extraordinary.
The Banksian rose is one of the commonest shrubs _ here-
abouts, and was laden with masses of fragrant white flowers.
Opium Poppy was abundant and the whole country-
side was gay with the colour of flowers. Styrax Veitchiorum
occurs here, and trees 12 to 4o feet tall were masses of ivory-
white.

From Sha-kou-ping we toiled slowly up the rocky ravine
down which the main stream rushes. A paper-mill or two
are located here, but houses are few and far between. The
rocks are of slaty shales, often very rotten, and the torrent
is a succession of rapids and cataracts. In spite of the
turbulent nature of its waters it is full of fish, some of them
of good size.

The hamlet of Ma-hsien-ping, our intended destination,
proved to be a miserable place of some half a dozen hovels
all filled with people engaged in collecting tea. We there-
fore journeyed on for another ro li to some farmhouses at
Shui-ting-liangtSze, and arrived just as the sun was setting

FOREST AND CRAG | 55

behind the range. We found accommodation in a large farm-
house, alt. 3900 feet. The day’s journey proved very arduous,
but there was much by way of compensation. The scenery
was sublime and the flora wonderfully rich and varied. In all
I gathered specimens of upwards of fifty new kinds of woody
plants, many of them previously unknown. This region is one
of the richest I have visited, and I subsequently secured a fine
haul of seeds, the great majority of the plants raised from them
being now found growing and thriving in many gardens of
Europe and America. (Later I again traversed this same
region, and owing to heavy rains was over a week in crossing
the country between Hsao-lung-tang and Shui-ting-liangtsze,
a flooded torrent holding us up for three consecutive days.)

It was nearly midnight when all was quiet last night, the
men being loud in their grumbling against taking the road to
Taning Hsien instead of that to Wushan Hsien. The reports
we had heard indicated a bad time ahead for all of us and for
the men in particular, owing to the extreme poverty of the
country-side. I heard them as I lay in bed, but fortunately
no complaints were brought to me.

It was later than usual when we got away in the morning.
After a steep ascent we meandered along the mountain-side,
and ultimately crossed over into Fang Hsien again by a low
pass, alt. 5600 feet. This is the real watershed of the Han
and Yangtsze River systems. The Sheng-néng-chia is a gigan-
tic spur thrust out from the backbone of the chain, and the
streams which take their rise from three sides of this spur
flow down to the Yangtsze. From the watershed we had a
good view of the Sheng-néng-chia peaks bearing E.S.E.,
and of some equally lofty mountains to the east, evidently
in the vicinity of the Yangtsze itself. On both sides of the
watershed is a rather broad cultivated valley bounded by razor-
backed hills clothed with woods of Oak and Pine. Varnish
trees abound on the edges of the fields and Walnut trees are
alsocommon. Farmhouses are scattered over the country-side,
and the crow of the Pheasant, the coo of the Wood Pigeon
and the notes of the Cuckoo were heard on all sides. By the
wayside are many fine trees of Sweet Chestnut and Magnolia,
and one very fine specimen of Corylus chinensis, 120 feet tall

56 A NATURALIST IN WESTERN CHINA

and 12feetingirth. Many medicines arecultivated hereabouts,
more especially Rhubarb and Tang-shén. Populus lasio-
carpa, with huge handsome foliage, is one of the commonest
trees.

After a few miles the cultivated valley ends and we entered
a narrow defile flanked by steep, well-wooded mountains.
Hereabouts the interesting Sinowilsonia Henryt is common,
forming a small, bushy tree with handsome foliage and long,
pendulous racemes of inconspicuous flowers. The most orna-
mental tree, however, is a fine Crab Apple, which was laden
with umbels of pure white fragrant flowers borne on long slender
stalks. Issuing from this defile we entered a small cultivated
flat and found lodgings at the hamlet of Pien-chin, alt. 5200 feet.

The vegetation during the day’s journey was not very
remarkable, though I added sixteen kinds of plants to the
collection. Noteworthy on the rocks and cliffs was Viburnum
rhytidophyllum, with its large flat corymbs of dirty white
flowers, which are not very pleasing to the nostril. In the
defile the mountain-side is rich in shrubs; amongst which various
Rhododendrons were prominent ; Rhododendron indicum was
common and Rosa sericea was just opening its flowers. All day
Oak woods were common; but these never contain much that is
of more than passing interest. In abandoned cultivated areas
a small Poppy, resembling the common Iceland Poppy, with
deep yellow (occasionally orange) flowers was very abundant
and attractive. In shady places the large yellow flowers of
Chelidonium lastocarbum made a fine show, and common on
bare limestone cliffs are Corydalis Wilsonit and C. tomentosa,
both species with yellow flowers and glaucous foliage. Around
our lodgings there was much cultivation, maize, barley, pulse,
and the Irish potato being the principal crops. Several paper-
mills occur by the side of the stream, bamboo pulp being the
raw material from which the paper is made.

On leaving Pien-chin we followed a river to a point where it
is joined by a tributary stream which we crossed and then
ascended the road which skirts its banks. This stream is gentle
for a Hupeh torrent, and for 10 lithe road is of the easiest. The
mountain-sides are covered with shrubs and trees; among which
Cercidiphyllum was conspicuous. Occasional houses and small

THE LARGE-LEAVED POPLAR (POPULUS LASIOCARPA) 50 FT. TALL,
GIRTH 5 FT.

FOREST AND CRAG 57

patches of cultivation occur, but the country generally is very
sparsely peopled. Populus lasiocarpa is abundant, and large
branches are commonly driven in the ground to make fences ;
these branches take root and form groves. A magnificent tree
of Ailanthus Vilmoriniana, 150 feet tall and 20 feet in girth,
was passed, andI was astoundedat the huge sizeof this specimen.
Tangled masses of Actinidia chinensis and various kinds of wild
roses were everywhere abundant, filling the air with soft
fragrance. Leaving this delightful mountain stream we made
a steep ascent of goo feet and then, to our great surprise,
entered a broad level valley. This valley was evidently in
earlier times a mountain lake—to-day the margins arecultivated
and the centre is a marsh. The whole district is known as
Chu-ku-ping or Ta-chu-hu,—the latter name having reference
to its former condition as a lake. A flat area of this character
is unique in these regions, as far as my knowledge goes. Several
roads cross this flat and we took the one for Taning Hsien.
By the wayside strawberries, white and red, luxuriate and were
very good eating. Quite a number of horses and cattle were
grazing in this valley, and the country could support many more.

After meandering some 15 li over the easiest of roads we
made a very steep and fatiguing ascent to alt. 7300 feet, and
crossed over into the province of Szechuan. From the neck of
the divide, looking away E.S.E., we obtained a good view of
the Sheng-néng-chia and the main and subsidiary ranges and
peaks—nothing but mountains on every side save the tiny
valley at our feet which we had just crossed. In the ascent we
passed many shrubs in full flower; particularly striking were
the various kinds of Viburnum, Deutzia, Abelia, and Cornus.
A precipitous descent through a ravine and we reached the
hostel at Hwa-kuo-ling, alt. 6350 feet, where plantations of
Rhubarb were common and several other medicinal plants
cultivated.

The road we were following is called the “‘ Great salt road,”
but we only met four men carrying salt in the day’s march.
Indeed, on the whole journey we encountered practically no
traffic. This wild mountainous country supports only a very
Sparse population and foreign trade has no chance hereabouts.
Our great difficulty was in securing enough food for the men.

58 A NATURALIST IN WESTERN CHINA

At Chu-ku-ping we managed to get one good meal from the
local head-man and bought portions of a wild pig recently
killed. At the hostel nothing was obtainable and the men had
to eke out on the small rations they had with them. Goitre
is common in these regions and nearly every one is affected. It
would seem to be hereditary, since I noticed children in arms
showing unmistakable swellings in the throat.

Boisterous winds and heavy clouds alternating with bright
sunshine marked our first day’s journey in eastern Szechuan.
We were again amongst cliffs of hard limestone and the scenery
strikingly resembles that of the Yangtsze Gorges and contiguous
country. The whole region is too steep for cultivation, and
habitations are few and far between and most dilapidated in
character. The soil is stiff, clayey loam and the few crops we
saw were wheat, Rye (Secale fragile), Irish potato, maize, and
pulse. The cliffs are for the most part well timbered, and
the common trees and shrubs of Hupeh are represented. Pinus
Armandi is very abundant and P. Henry: is also common.
Odd trees of Spruce and Hemlock also occur. A fine speci-
men of Acer griseum, 60 feet tall, 7 feet in girth, with curious
cinnamon-red papery bark was the feature of the day’s march ;
unfortunately, it was badly situated for photographing. Beech,
Yellow-wood and Dipteronia sinensis were common trees en
route.

The road is one long succession of ascents and descents
and most fatiguing. In the afternoon, after a particularly
trying ascent, we wandered for an hour or so through woods of
Oak (chiefly Quercus variabilis and Q. aliena) and Sweet
Chestnut, the latter laden with its white, evil-smelling flowers.
Walnut and Varnish trees are everywhere abundant and
Campanula punctata is a common weed of cultivation. No
foreigner had ever before traversed this region and the people
were very timid, locking up their houses and hiding themselves
from view at our approach. The cliffs in this neighbourhood are
full of caves and many of these are bricked up to form places of
refuge in troublous times. We found lodgings for the night at
Peh-kuo-yiien, alt. 3750 feet, in the house of the head-man of
the hamlet. Food-stuffs were scarce and there was great
difficulty in persuading the people to part. What little we

FOREST AND CRAG 59

did eventually obtain was at famine prices and the grumbling
was loud on all sides.

The following morning we descended by a moderately easy
path to a torrent and then commenced a _ heart-breaking
ascent of some 2600 feet. It was excessively hot and I do not
remember perspiring so much before. A rugged, precipitous,
sparsely populated country is this, and I never wish to see it
again. Limestone regions are magnificent from the scenic
point of view, but for travelling over they are fierce and
arduous beyond words! Our destination was Hsao-pingtsze
and no one knew the distance. Inquiries made as often as
possible always elicited the same reply: “‘ Seven or 8 li from
Peh-kuo-yiien, 7 or 8 li to Hsao-pingtsze.’’ Late in the after- _
noon the distance to go increased to 30 li and did not shorten
until we suddenly sighted the two huts which form the hamlet
of Hsao-pingtsze !

The ascent was largely under cultivation, but the final
stage was through jungle. Lomicera tragophylla is common
and was in full flower, but we saw no good plants. A bush
of Schizophragma integrifolium, one mass of the purest white,
on the cliffs, was conspicuous from afar. But the flora
generally is very ordinary, with Rhododendron discolor and
R. Mariesit common here and there. On reaching the top of
the cliffs we entered a cultivated slope where Walnut and
Varnish trees abound. The district is called Ta-ping-shan
and consists of several scattered farmhouses surrounded by
fields of maize, pulse, barley, and Irish potato. At one of
these farmhouses my followers managed to secure a good meal
and high spirits prevailed in consequence.

On leaving this place we continued to ascend by an easy
path skirting rolling downs. A few scattered houses occur
for a couple of miles but were mostly deserted, and we soon
left all signs of cultivation and habitation behind us. The
downs are treeless and clad mostly with grass with scattered
bushes of Willow, Barberry and Spirea. The depressions
between the hills were masses of blue Forget-me-not. The
whole region would make excellent grazing ground for cattle.
Crossing over at 7950 feet altitude, we descended by an easy
road for a mile or so and passed a couple of huts surrounded

60 A NATURALIST IN WESTERN CHINA

by extensive plantations of Medicinal Rhubarb. Many fine
herbs luxuriate hereabouts, and among them Ivis Wilsonii
with its yellow flowers was conspicuous, covering large areas.
Eventually we reached the edge of a precipice, down which
the road fairly tumbles for 5 li to Hsao-pingtsze. This
hamlet, as the name indicates, is situated on a tiny flat
(probably caused by a landslide) and boasts two miserable,
dilapidated houses. We took up lodgings in the smaller and
presumably less squalid of the two, but there were little to
choose between them in all conscience. On three sides the
hamlet is walled in by steep cliffs and the fourth is the edge
of a precipice itself. It was only some 30 yards from our hut
to the edge of this precipice, and the view from this point is
one of the most extraordinary and wonderful my eyes have
ever beheld. Below me (some 4000 feet the morrow proved)
at an acute angle lay a small village with a considerable river
flowing past it. Beyond this was range upon range of bare,
treeless, sharp-edged ridges, averaging 5000 to 6000 feet in
height, with outstanding higher peaks and grander ranges in
the beyond. The rocks are mainly of limestone, white, grey
and reddish, giving a bizarre appearance to the whole scene.
Never have I looked upon a wilder, more savage and less
inviting region. A storm was brewing and the light rapidly
failing, making it impossible to take a photograph, though
no photograph could have produced a picture that would give
an adequate idea of the savage grandeur of the whole scene.
It was indeed sufficient to awe and terrorize one. Such scenes
sink deep into the memory and the impressive stillness
produces an effect which is felt for long years afterwards.
Soon the angry rain-clouds darkened and blotted out the
whole scene and the next moment a thunderstorm burst over
us. This storm lasted through the night and, the roof of our
hovel being like a sieve, the rain soon converted the mud floor
of the hut into a quagmire. We huddled together and did
what we could to keep dry and warm, but the night proved
long and cheerless.

Soon after daybreak next morning we made our escape
from these wretched quarters, but rain was still falling, and of
the wonderful scene of the preceding evening nothing was

NOLL VIHONVI AO HOWVTTIA LAYAVWN AHL

FOREST AND CRAG 61

visible from the gap but an ocean of clouds. The descent is
most precipitous and for the first 2000 feet we fairly tumbled
down. Afterwards it became more gradual and led over a
steep cultivated slope of red clayeysoil, making walking difficult.
Nowhere is this descent easy, and very glad were we all that
our route was down instead of up this mountain-side. At the
foot of the descent the road leads through a rocky defile to
emerge on the banks of a clear-water river some 60 yards
broad. Across this we were ferried to Tan-chia-tien, the village
we saw from near our lodgings last night. This village consists
of some fifty houses which are huddled together and overhang
the river in front and cling to the cliff behind in an extra-
ordinary manner. From this village a kind of long street
with houses scattered here and there along its length extends
for 2 miles to the village of Chikou, situated at the junction of
this river with another of almost equal size. A mile or so
from Chikou up the secondary stream are the salt wells of
Taning-ching.

The road we struck at Tan-chia-tien is a highway leading
northwards to Shensi and southwards to Kuichou Fu on the
Yangtsze River. Hereabouts and down to Taning Hsien,
12 miles distant, and northwards I know not how far, the
cliffs are sheer to the water’s edge. The road is well graded
and a good 6 feet broad, and has been excavated or blasted
from the solid rock.

From Chikou to Taning Hsien is said to be 30 li with not
a house or hovel between. To cover this we with difficulty
engaged boats, long, narrow, lightly built affairs (Sin-po-tzu),
turned up at prow and stern, with no oars and steered by
long sweeps projected fore and aft. The current was strong
and rapids numerous; aided by a freshet we covered the
whole distance in half an hour. The brief journey was through
one grand chasm, the walls of rock being sheer to the water’s
edge with no space even for a shingle-bank to lodge. These
cliffs are treeless and mostly bare with here and there grassy
patches and clumps of delicate, graceful Bamboo (Avundinaria
mitida). The road zigzags around the cliffs on the right
bank well above high-water mark, and every inch of it has
been blasted from the hard wall of rock. Stone gates and

62 A NATURALIST IN WESTERN CHINA

barriers occur at intervals, but there are no houses. This road
is of such a nature that time and neglect can affect it but little,
but it is now scarcely used except by occasional pedestrians
and salt-carriers when the river is impracticable. I tried hard
to discover when and by whom the road was built but found
no one who could tell me. It is evidently one of the ancient
arteries of China, and probably dates back to the discovery
of the salt-wells. It struck me as being an old military road
and may probably have been built centuries ago when Kuichou
Fu was a place of infinitely greater importance than it is
to-day.

The river I have mentioned, known locally as the Taning
Ho, rises near the borders of Shensi, Hupeh, and Szechuan, and
after flowing nearly due south enters the Yangtsze at Wushan
Hsien. From Chikou boats descend to its mouth, 200 li distant.

Taning Hsien, alt. 750 feet, the most easterly inland town
in Szechuan, is situated on the right bank of the river, here
about 100 yards broad, and sweeping from the gorge in a fine
curve. The town is wedged in on the side of a mountain-slope
up which the city wall ascends for several hundred feet. The
river-front is bounded onone side bythe city wall, and the shops,
houses and ydamens are crowded together near the river. The
upper slopes enclosed within the city wall are given over to
agriculture. The town, comprising about 400 houses, is the
residence of a district magistrate, and boasts a trade in salt
and odds and ends. Formerly it was the centre of a large
opium traffic.

At Taning Hsien the Chinese Banyan (Ficus infectoria), so
abundant and characteristic of the central parts of Szechuan,
puts in an appearance. Near a temple, a few hundred yards
from the north gate of the town, I observed from the boat
what appeared to be a Mantzu cave built in the face of lime-
stone rock. On inquiring I was told of four or five similar
caves in this neighbourhood. Later I may have something
to say about these caves, but it is interesting to be able to
register their presence at the extreme eastern edge of the
province, since heretofore they have been considered a feature
of the more western parts. Physically and geologically
speaking, the country east of the Taning River belongs to

FOREST AND CRAG 63

western Hupeh. Almost immediately west of it the charac-
teristic red sandstone of Szechuan commences.

For twenty-two consecutive days my followers and I had
struggled through the wild, lonely mountain fastnesses of north-
western Hupeh, suffering much from bad roads, worse accom-
modation and scarcity of food supplies. For the first time on
record the journey had been accomplished by a foreigner, and
one and all of my followers were happy in the thought of the
comparative luxury and plenty of the country which was now
before us.

CHAPTER VI
THE RED BASIN OF SZECHUAN

GEOLOGY, MINERAL, AND AGRICULTURAL WEALTH

province of Szechuan, from near the Hupeh boundary
to the valley of the Min River, the predominant rocks are
red clayey sandstones, probably of Jurassic age. These rocks
are of immense thickness and impart a characteristic red colour
to the surface, and for this reason the late Baron Richthofen
gave the term “‘ Red Basin ”’ to the whole region. This basin
is nearly triangular in shape, the city of Kuichou Fu marking
the ‘‘apex.’’ Imaginary lines connecting Kuichou Fu with
Lungan Fu in the north-west, and Kuichou Fu with Pingshan
Hsien keeping a little to the south of the Yangtsze River,
respectively mark the northern and southern “sides.” Another
line from Lungan Fu and thence skirting the valley of the Min
River to Pingshan Hsien marks the “ base’’ of the triangle.
The entire basin is nearly 100,000 square miles in area, and is
surrounded on all sides by lofty mountain ranges, those on
the west rising above the snow-line. In the east the boundary
ranges are composed principally of Upper Carboniferous lime-
stone, as described in Chapter II. The western boundary
ranges are largely made up of shales. The Yangtsze River
traverses the basin from west to east, following a course nearly
parallel with the southern limits of the basin itself. Within
this triangle there is abundant life, industry, prosperity,
wealth, and intercommunication by water. Outside of it on
all sides the contiguous country is sparsely inhabited, little
productive and no river is navigable save the Yangtsze, where

it leaves the basin.
In ancient geological times ae region was doubtless a vast

province of Su the eastern and central parts of the

THE RED BASIN OF SZECHUAN 65

inland lake with a fairly even floor. Since the draining off of
the waters the Yangtsze River and its network of tributary
streams have eroded channels 1500 to 2500 feet deep through
these soft sedimentary rocks, and converted the whole basin
into a thoroughly hillycountry. To-day practically the only level
area is the Plain of Chengtu, some 80 miles long and 65 miles
wide, with an average altitude of 1800 to 2000 feet above sea-
level. The rest of the basin is broken up into a network of
low, rolling or flat-topped mountains averaging about 3000
feet above sea-level, and nowhere exceeding 4000 feet altitude.
The whole of this region is under agriculture, the highest
development of which obtains on the Chengtu Plain, perhaps
the richest area in the whole of China. Anent this particular
part we shall have something to say later.

How great a period of time has elapsed since the disappear-
ance of the waters from this basin is purely conjectural. But
that this triangle has long constituted a well-marked boundary
is evidenced by the fact that remarkably few of the plants
found in the mountains bordering the eastern limits at 2000
feet altitude and upwards are common to the mountains
bordering the western limits. The genera are of course the
same, but the species are usually distinct. The difference
between the floras of the eastern and western border-ranges is
too great for a mere 500 miles of longitude to account for
solely. The same is true of the fauna in so far as the game
birds and animals are concerned, as Chapters XI and XIII,
Vol. II, dealing with these will confirm.

From evidence presented by the flora to-day it appears
doubtful if ever the Red Basin was covered with great forests.
Rather would I suppose that subsequent to the disappearance
of the waters the region bore some resemblance to the “ bad
lands”’ of certain parts of the United States of America.
All this is admittedly pure conjecture. Everywhere to-day,
trees, shrubs, and herbs are common, but the flora, in contra-
distinction to that of the contiguous regions, is relatively poor,
and the species largely common to the entire basin. Further,
the majority of these species are widely spread throughout
the warmer low-level legions in China, some indeed ranging
to the extreme eastern limits of the country. A theory is apt

VOL. IL.—5

66 A NATURALIST IN WESTERN CHINA

to become fascinating, and may easily be carried too far. The
facts above recorded are best left until the geology of China
generally is more accurately known.

Coming down to historical times we learn that the region
previous to the advent of the Chinese was peopled by an
aboriginal population divided into the kingdom of Pa in the
east and the kingdom of Shu in the west. This aboriginal
population has entirely disappeared, but records in the shape
of well-constructed caves having square entrances are found
scattered all over the Red Basin. These caves are especially
abundant around Kiating Fu. A little investigation of these
interesting places has been attempted, and fragments of pottery
and odds and ends discovered. The entrances to these caves
could only be closed from the outside, and from this fact, and
other details, it is probable that they served as the burial-places
of the chiefs and more wealthy among this extinct people,
rather than as dwelling-places or harbours of refuge. Doubtless
they have been subsequently used for these latter purposes,
but that they were designed for tombs seems to best explain
their origin. From Chinese history we learn that as early
as 600 B.c. the kingdom of Pa had relations with the Chinese
kingdoms of Ts’u, which occupied the regions north of the
barrier ranges. Later, Pa princesses married Ts’u kings.
Ts’uwas in time conquered by Ts’in (another Chinese kingdom),
which gradually absorbed Pa, and finally conquered Shu about
315 B.c. A military road was commenced from the neighbour-
hood of modern Hanchung Fu, designed to connect with the
region around modern Chengtu, by Ts’in-shih Hwang about
220 B.C. This road, which enters Szechuan from across the
barrier ranges near Kuangyuan Hsien, is still in existence as
the great highway connecting Chengtu with Hanchung Fu,
Sian Fu, and, ultirnately, Peking itself. For the next fifteen
centuries the history of this region is full of war, rebellion, and
internecine strife. Usurpers established petty dominion over
the country from time to time, only to disappear amongst
awful slaughter and bloodshed. There is scarcely a square mile
of the whole region but what recalls scenes of valour, treachery,
and carnage. In the latter half of the thirteenth century the
famous Tartar, Kublai Khan, carried his arms victoriously

NISVd GH WHHL NI MIA TVOIdAL V

4 3 ’
. ot

*
ing eee ed

THE RED BASIN OF SZECHUAN 67

over nearly the whole of modern China, and formed an Empire
which the succeeding Ming and Manchu dynasties maintained
more or less intact.

Since the time of Kublai Khan many rebellions have
swept over Szechuan, decimating the population and paralysing
industry. The present population is mainly derived from
immigrants (voluntary and otherwise) who settled there during
the early half of the eighteenth century. A census taken in
A.D. 1710 returned only 144,154 souls for the whole province.
To-day the population is estimated at 45,000,000! In spite
of all the long-sustained wars and bloody rebellions, agri-
culture has managed to subsist, and the whole of the Red Basin
is a lasting monument to Chinese genius and industry in matters
agricultural. An abundant water-supply and constant tillage
are necessary to obtain a full crop from these sandy clays
and marls. Fortunately, the whole region is one vast net-
work of streams, all of which drain into the mighty Yangtsze.
The Chinese have taken full advantage of this intricate river
system, and devised manifold methods of irrigation. These
devices, combined with the untiring patient industry of the
people, have converted an incipient ‘‘ bad land”’ into a rich
and fertile region of terraced fields. In no part of China that
I have visited are the people entitled to greater praise for
meritorious agricultural accomplishment than throughout this
Red Basin.

In many parts of this region the river valleys are so steeply
eroded that very little cultivatable bottom-land is formed.
Consequently the rice belt is relegated to slopes and summits
of the low, flattened hills. In limestone regions the bottom-
lands constitute the main rice belt, but in the sandstone
regions the opposite obtains. The climate of the whole region
is mild and genial, and during both winter and summer the land
is cropped. Rice is the great summer crop with maize, millet,
sweet potato, sugar-cane, tobacco, pulse, and various other
crops. The principal winter crops are wheat, rape, peas,
broad beans, cabbage, Irish potato, etc. Formerly opium was
cultivated in enormous quantities as a winter-crop, but this
has lately been almost entirely suppressed. Cotton does not
thrive in the Red Basin, though its culture is attempted in

68 A NATURALIST IN WESTERN CHINA

many districts, notably Yilung Hsien and in Tungchuan Fu.
Cotton is the one commodity that this region has to import;
and nearly all its surplus products go to meet this deficiency.
But, if cotton is very little grown, many kinds of hemp are
produced in quantity, though very little is used for textile
purposes. Silk production is everywhere an industry of import-
ance; and in many districts the staple. Only the very poorest
are without some silk garment, though such is only habitually
worn by the more wealthy. Tea is grown in many districts
both for local consumption and for export. In the more
westerly parts tea for the Thibetan market is a staple product.
Wood Oil and many other valuable economic trees are also
largely cultivated. Fruit is generally grown, including peaches,
apricots, plums, apples, pears, and oranges in variety.
Oranges thrive remarkably well in this red sandstone; and the
extensive orchards are a wonderful sight during the month of
December. Tangerine varieties are most generally cultivated,
and the fruit in season can be purchased at the rate of twelve
hundred or more for two shillings! The tight-skinned varieties
are less frequently grown, and are more expensive. Around
Lu Chou are plantations of Litchi trees. When they came from
their original homes the settlers evidently brought with them
their favourite trees and grains and planted them around their
new homesteads. These introductions; and the favourable
climate, explain the presence of such a vast variety of culti-
vated plants, which is probably greater than that found in any
other province in China.

The steeper and rougher country is covered with small woods
of Oak, Pine; and Cypress, elsewhere trees are confined to the
vicinity of streams, houses, temple-grounds, wayside shrines,
and tombs.

The streams are navigable for extreme distances, and a
perfect network of roads traverse the basin in every direction.
These roads are, on the whole, well built for Chinese roads, but
are not kept in thorough repair any more than those elsewhere
in the land. The streams, however, are well supplied with
ferries, and well-built bridges, substantially constructed of
stone, and kept in good repair, are a feature throughout the
entire region. Large cities, market villages, hamlets, and farm-

‘LA fl HIMID “TIVE “La 08 (GSSOMOMNW SQGNIdVS) ANAL-dVOS V

THE RED BASIN OF SZECHUAN 69

houses dot the land, which everywhere appears prosperous
and its inhabitants contented. Drought occasionally brings
famine, but, on the whole, the Red Basin suffers much less
from this dread calamity than do other and less favoured parts
of the eighteen provinces of China.

The mineral wealth of the Red Basin is not varied, but
enormous brine deposits occur scattered over the whole area,
and are worked at depths varying from almost surface level to
3000 feet. In the eastern parts, Kuichou Fu, Wén-tang-ching,
for example, the rivers have scoured the rocks until the brine-
deposits are practically exposed. In the west, however, as at
Wu-ting-chiao, situated on the left bank of the Min River a few
miles below Kiating Fu, the brine is found at about 500 feet
down. At Tzu-liu-ching, on the left bank of the To River,
where the richest deposits occur, the brine is found at depths
from 1000 to 3000 feet.

Salt is worked in some thirty-nine districts in the Red Basin.
It is everywhere a Government monopoly, and its production
and subsequent distribution are rigorously controlled. The
annual output is estimated at about 300,000 tons. At Tzu-liu-
ching most of the brine is evaporated by inflammable gas; in all
other places the brine is evaporated by coal heat. In boring the
deep wells, it is uncertain whether brine or gas will be struck,
but both are equally valuable. The occurrence of this in-
flammable gas indicates the presence of petroleum beds at
still greater depth.

Coal is found in greater or lesser quantities scattered all
over the Red Basin, and is always found not very far removed
from brine pits. This coal varies from lignite to anthracite.
The average quality is poor, but one or two good seams have
been found, notably at Lung-wang-tung, a few miles north
of Chungking.

Our early description of the Red Basin needs some ampli-
fication to explain the presence of coal and other minerals.
Although the sedimentary sandstones are in a state of undis-
turbed stratification over a great part of this area, yet there is
dissecting this Red Basin a number of linear elevations, in
which the underlying limestone is bent up from a great depth.
This limestone forms in every case an axial core, lined on either

70 A NATURALIST IN WESTERN CHINA

side by highly inclined strata, among which there is ordinarily
noticeable, next to the axis, a double belt of coal-formation,
followed on either side by strata of red sandstone standing on
edge. Baron Richthofen estimates that “ the area of the coal-
bearing ground in Szechuan probably exceeds in size the total
area of every other province of China.’”’ But probably through-
out nine-tenths of this area the coal-measures are buried deep
beneath the superincumbent strata, and with trifling exceptions
can never become available for mining. In the linear elevations,
above referred to, the belts of coal-formations, though narrow,
are of great length. They are most readily accessible in those
places where rivers have cut through and exposed the ends of
the seams. Mining is done by means of horizontal adits
working from an exposed surface inwards. Coal is very
generally obtainable throughout the Red Basin, and is the
ordinary fuel of the entire region.

Iron-ores occur scattered throughout the entire region, but
though in the aggregate the iron-smelting industry is a con-
siderable one, in no one place is iron made on a large scale.

Sulphate of iron (copperas) is found in combination with
coal in one or two districts, notably in Kiangan Hsien. Lime
is common to all the linear elevations mentioned above,
occurring in juxtaposition with coal, and is burnt in kilns in
the usual way.

Gypsum is found and worked in one or two places, notably
Mei Chou and Pengshan Hsien, both districts on the Min River,
between Kiating and Chengtu.

Mineral oil in small quantity occurs in the district of Pengch’i
Hsien, where a native company has made some attempts to
develop the industry, but with unsatisfactory results.

Other less important minerals occur in small quantities.
The precious metals, gold and silver, are not found in the Red
Basin proper but in the mountainous country to the west of
this region, where copper, lead, and zinc ores also occur.

In reference to gold it should, however, be mentioned that
rude placer mining is carried on during the winter months,
throughout the numerous shingle banks exposed in the beds of
the Yangtsze, Kialing, and Min Rivers. On the Yangtsze this
precarious industry is first to be noted some 50 miles below

THE RED BASIN OF SZECHUAN 71

Ichang, but it is not general until the region west of the Gorges
is reached. The industry is carried on by the unemployed
peasantry, and the returns are most insignificant. This gold
is in all probability brought down by the Yangtsze and its
larger tributaries during the summer floods. There is no record
of any gold-bearing quartz having been found 7m situ in the
Red Basin proper. In the mountains bordering its western
and north-western limits, gold quartz is found in greater or
less quantities, and all the principal rivers of this region either
take their rise in, or flow through, these ranges. This fact
explains the presence of small quantities of gold far removed
from the gold-bearing strata.

CHAPTER VII
EASTERN SZECHUAN

NARRATIVE OF A JOURNEY FROM TANING HSIEN TO
TUNGHSIANG HSIEN

HE region described in this chapter was traversed by
Lieut.-Colonel C. C. Manifold and Captain E. W. S.

Mahon when surveying for a possible route for the pro-
posed Hankow-Szechuan Railway in 1903 or 1904, I am not
sure which. There is no record of any other traveller having
crossed this part of eastern Szechuan, though it is very
possible that missionaries may have done so. I do not know
the conclusions arrived at by these surveyors, but the con-
struction of a railway along the route I traversed would be
a difficult and costly undertaking.

The following narrative is compiled from my diary,and may,
perhaps, convey a brief idea of the nature of the country and
the flora found in the more easterly parts of the Red Basin.
As will be gathered, I took ten days to cover the distance,
but I travelled leisurely, and the journey could be accomplished
in six days.

June 28.—Yesterday we spent the day at Taning Hsien,
refitting and preparing for our journey westwards to Chengtu
Fu. Money exchange proved an involved and difficult business.
Ten-cash pieces, both Hupeh and Szechuan currency, are
accepted here at 20 per cent. discount. This means that the
purchasing power of a thousand such cash is only equal to
800 string-cash. Farther west, Hupeh 10-cash pieces are not
current, and the Szechuan r10-cash piece is only accepted for
two days’ journey west of this town. We had therefore to
burden ourselves with string-cash, which added considerably

to the weight of our loads. A thousand cash in r10-cash pieces
72

EASTERN SZECHUAN 73

weighs less than 2 lb.; in string-cash the same equivalent
weighs over 8 Ib.!_ If there is one reform more badly needed
than another in China it most certainly is in the matter of
currency.

Leaving Taning Hsien by way of the west gate we made
a slight ascent and entered a narrow, highly cultivated valley,
flanked on our right by fairly high and on the left by lower
mountains, nearly treeless and sparsely cultivated. The town
of Taning lies in a depression, and the morning mists obscured
the general view. It is a very small place, with much of the
land enclosed within its walls given over to cultivation. An
outer gate, wall, and block-house guards the west gate proper.

Ascending the valley by an easy road which more or less
skirts a fairly large tributary stream of the Taning Ho, we
reached the village of Che-tou-pa before noon. Rice was
abundantly cultivated everywhere, irrigation being effected
by means of large ‘‘ Persian’’ wheels. Much cotton is culti-
vated following wheat, the winter crop. Maize was 5 feet
tall and in full flower. Paliurus orientalis, a thin tree 30 to
50 feet tall, is very common, and was laden with white,
circular, odd-looking fruits. Weeping Willows, Cypress, and
fine specimens of a hairy-leaved, small-fruited Hog Plum
(Spondias) were noteworthy, with Bamboo groves in abundance.

On leaving Che-tou-pa we deserted the main tributary
stream and ascended a small branch. The valley narrows,
and the hills are more wooded, chiefly with Cypress. The road
is easy, though here and there sadly in need of repair. We
journeyed slowly, and eventually crossed over a ridge of low
hills to the hamlet of Lao-shih-che, which we reached at
5 p.m. This tiny place, alt. 1950 feet, and 55 li from Taning
Hsien, consists of half a dozen houses, scattered through a
narrow valley with rice fields on all sides. The people were
very nice, but inquisitive.

We were on the edge of the Red Basin and much of the
soil had the characteristic red colour. Wood Oil trees are
commonly cultivated, but cotton was not in evidence during
the afternoon. In a grove I noted some magnificent trees of
Pistacia chinensis and Sapindus mukorosst. The young shoots
of the former are cooked and eaten, but the round fruits of

74 A NATURALIST IN WESTERN CHINA

the Sapindus are. used as soap. Celtis trees are common,
their smooth, pale-grey bark rendering them conspicuous.
On a ridge we noted many trees of the interesting “‘ Button
tree’ (Adina racemosa). These trees were 30 to 60 feet tall,
2 to 4 feet in girth, and the finest specimens of their kind I have
met with. The Chinese Pine (Pinus Massoniana) is general,
but by far the commonest tree of the day was the Cypress
(Cupressus funebris).

The road proved a pleasant change; instead of wild and
savage scenery, low rounded hills backed by steeper mountains,
all rather treeless, and for the most part cultivated, were the
order of the day. Here and there were a few outstanding
cliffs of limestone with an occasional temple crowning odd
crags. At Taning Hsien we secured a number of new coolies,
and these men described the country passed through in the
afternoon as Laolin (wilderness). This immensely amused my
Ichang men, who recommended these newcomers to try the
Sheng-néng-chia before speaking of “‘ Laolin”’ !

The day was grilling hot, and all were fairly exhausted on
arrival at Lao-shih-che. Whether it was the heat or the after
effects of a day’s holiday I could not determine, but I was
called upon to play “ Doctor”’ to nearly half my followers.
The majority were suffering from stomach troubles, several
from filthy sores. Epsom salts, permanganate of potash, and
iodoform dressings soon improved the majority.

The next day was gloriously fine, but scarcely so hot as
the previous day, or perhaps the slightly increased altitude
made it more bearable. The whole day we travelled nearly
due west through a narrow valley bounded by moderately
high parallel ranges. The road continues easy with occasional
ascents and descents. We were still on the fringe of the
Red Basin, but in the afternoon grey sandy soils were most
in evidence. Rice is cultivated wherever sufficient water is
obtainable, and was scarcely ever out of our view. Maize is
the other principal crop, with various kinds of pulse and the
Irish potato. Thesweet potato is cultivated here and there,
and Wood Oil trees are even more abundant than before.
Much oil is evidently produced in this region, and we noted
many oil-presses during the day. The parallel ranges are

ONIANIdTAOS TVANN ALVNYO HALIM WOATOSNVIN

EASTERN SZECHUAN 78

from 500 to 1000 feet above the valley, sparsely cultivated,
and for the most part well timbered with Cypress (Cupressus
funebris), Pine (Pinus Massoniana), and Oak (Quercus serrata).
Poplar is a common tree, and by the sides of streams Weeping
Willows abound. Shrubs in variety occur the most note-
worthy being Jtea tlicifolia and Torricellia angulata. Nowhere
else have I seen this latter shrub so plentiful; it favours the
sides of streams, ditches, and rocky gullies, forming a densely
leafy bush 8 to 12 feet tall. The fruit when ripe is black, and
is borne in large pendulous cymes. The Itea occurs in rocky
places, and its pendulous tails of greenish-white flowers are
often 18 inches long. The leaves very closely resemble those
of the common Holly, and when not in flower it might easily
be mistaken for that plant.

Houses are scattered along the route, but the population
is sparse. We met a few mule trains, but there was really
very little traffic on the road. We found accommodation for
the night at Hsia-kou, a prettily situated hamlet, alt. 2800
feet, 65 li from Lao-shih-che. Our lodgings were spacious,
but the occupants of the house looked unprepossessing opium
sots.

At To-chia-pa, a small hamlet passed a few miles before
reaching Hsia-kou, a road branches off to the northward and
leads to Chéngkou Ting. It was said to be a hard road to
travel over.

On leaving Hsia-kou we immediately plunged into a ravine
with steep limestone cliffs 300 to 500 feet high; the road
follows the dry bed of a torrent. At the head of this ravine
we made a slight ascent, and wandered across low mountain-
tops for a few miles, then descended and crossed a branch of
the Kuichou Fu River by a covered bridge. Up to this
point Pine and the Chinese Fir (Cunninghamia lanceolata) are
common. At the bridge I photographed the largest tree of
Platycarya strobilacea 1 have seen. This specimen was fully
75 feet tall, with a girth of 6 feet. I had no idea it could attain
such dimensions. A few miles beyond this point we forded
the main branch of the Kuichou Fu River, a broad, shallow,
clear-water stream, and about noon reached the village of
Chiao-yang-tung. Soon afterwards we were overtaken by a

76 A NATURALIST IN WESTERN CHINA

furious thunderstorm, which arose with amazing suddenness.
The fury of the storm spent itself in a torrential downpour of
short duration, but rain fell steadily during the rest of the
day. The rain did not improve the mud road, and our progress
was slow and difficult in consequence. During the whole
afternoon we made a steady ascent, skirting the mountain-sides
through woods of Pine and Oak. Eventually the road enters
a narrow sloping valley, at the head of which we found lodgings
for the night in two houses which constitute the hamlet of
Shan-chia-kou, having travelled 65 li. Around this place the
flora is varied and essentially cool-temperate in character.
Bushes of Mock Orange (Philadelphus) were conspicuous on
all sides with their wealth of pure white flowers. The Hautboy
strawberry is abundant, and around our hostel I gathered in
a few minutes enough of these luscious fine-flavoured white
berries to stew for dinner. The Torricellia was again common.
It ascends up to 3500 feet altitude, and often forms a small
inelegant tree.

We saw very little rice during the day, maize and Irish
potato being the chief crops. There is practically no traffic
on this road; the mule-trains seen yesterday evidently came
down the road from Chéngkou Ting. Population is sparse, and
what there is looked strongly addicted to the opium habit.
So far, however, we had not seen any signs of poppy.

A magnificent day ushered in the new month. The morning
was bearably hot, but the afternoon scorchingly so. A hundred
yards beyond our lodgings we reached the head of a ridge, and
an abrupt descent of a couple of thousand feet or so led to a
narrow valley where much rice, maize, Irish potato, and a
little Hemp (Cannabis) are cultivated. The parallel ranges
flanking this valley are of limestone with outstanding bare
rocks and cliffs, very little cultivated but with good woods of
the common Pine. Here and there in the valley we passed fine
trees of Sassafras, Sweet Chestnut, Sweet Gum (Liquidambar),
Chinese Fir, and Poplar. At the head of the valley we made a
slight ascent to the top of a ridge. Below us, some 2500 feet,
flowed a considerable river walled in by lofty limestone
precipices. It was 10.30 a.m. when we reached the top of
this ridge, and the rest of the day’s march was a more or less

EASTERN SZECHUAN 77

precipitous descent to the river, which we reached at Sha-to-tzu
about 3 p.m. In its early stages the descent is as difficult as
it well could be—over loose Rowley-raglike debris, down and
up steep steps, and over slopes of greasy clay. We crossed
one or two cultivated slopes, but most of the time the road
skirts around the sides of cliffs. At the edge of one precipice,
500 to 1000 feet sheer, the road is carried through a narrow
tunnel some 50 yards long and 34 feet broad at the exit. This
tunnel is partly natural and partly made by blasting the hard
limestone. It was quite dark within the tunnel save for a
faint glimmer of light at the exit. Both chairs and loads were
with difficulty carried through this tunnel. This roadway is
of recent date, and is unique in my experience of Chinese roads.
Rough as it is it saves about 10 li and a very steep ascent
and descent.

From the tunnel-way the road skirts the tops of the cliffs
with many exasperating and wearying ascents and descents.
Finally we descended to a small tributary of the main stream
and, crossing over, reach Sha-to-tzu, a busy market village and,
for the nature of the country, of considerable size. Up the
tributary stream some Io li, iron is mined and smelted, the
quality being described as good. Around Sha-to-tzu, coal
is worked and lime burnt.

The river we had with so much fatiguing travel reached
enters the Yangtsze at Yunyang Hsien, distant 150 li. It is
a clear-water stream of considerable volume, and is navigable
for small boats from just below Sha-to-tzu to within 15 li of its
mouth. Salt and a little peddling traffic was noticeable on the
road; also odd loads of medicines, including Tu-chung, the bark
of Eucommia ulmoides. The salt is a product of Yunyang
Hsien, and is not allowed to enter Taning Hsien. The quality
is said to be superior to that found within the latter district.

The flora of the day’s journey was not particularly interest-
ing, being very much the same as that found in the glens
and gorges around Ichang. A new Stachyurus and Abelia
Engleriana were collected. The Heavenly Bamboo (Nandina
domestica) was particularly abundant in rocky places, its
elegant foliage and large erect trusses of white flowers with
conspicuous yellow anthers making it very attractive. In

78 A NATURALIST IN WESTERN CHINA

autumn and winter the masses of scarlet fruit render it ex-
tremely beautiful. Wood Oil trees were general in rocky places,
and Hypericum chinense, a wealth of rich golden yellow, was
strikingly handsome, nestling on the cliffs everywhere. Quite
a little Ramie (Behmeria nivea) is cultivated, and the people
were busy stripping the fibre-containing bark from the stems.
The leaves, like those of several other plants, are used for
feeding pigs. The stripping and cleaning is all done by hand
labour.

The day’s march was full of interest, but the intense heat
and hard road made the 60 li very trying, and all were glad
when the end of the stage was reached. The scenery was
magnificent, and forcibly reminded us of the glens and ravines
around Ichang. The railway surveyors must have been filled
with despair when they encountered this steep limestone
country !

Sha-to-tzu is only about 700 feet altitude, and, in spite of
the swift-flowing stream which passes its “‘ front door,” was
suffocatingly hot. We managed to find a good inn with
quarters removed from the street and remarkably private in
character. Wehad no difficulty in changing silver here, but
ro-cash pieces are no longer negotiable. String-cash was the
only kind the people would accept.

Just below Sha-to-tzu we crossed the river by a ferry which
is assisted by a convenient rapid, and commenced a steep
ascent. A few hundred feet up we were afforded a good view
of the village we had just left. It contains about a hundred
houses, crowded together on a narrow, fan-shaped slope. A
few temples shaded by large Banyan trees were conspicuous,
and the whole made a decidedly pretty scene. The ascent is
through cultivated fields, groves of Wood Oil trees, and finally
Pine woods. At 3100 feet altitude we crossed a gap, and 200
feet more led to the top of the range. The rest of the day we
followed an undulating, easy road which meanders through
rocky, Pine and Cypress clad mountain-tops, and finally
descends to Ché-kou-tzu, which was our destination for the day.

The country is very pretty ; farmhouses are scattered along
the route, and where possible the land is under cultivation.
Rice was of course the crop where water is obtainable, maize

ALVD HLYON :‘NAISH DNINVI

EASTERN SZECHUAN 79

and Irish potato elsewhere. Tobacco is grown; a little of
this crop has been noted every day since leaving Taning Hsien.
Limekilns were common all day. In one place we saw a
number of men out with guns after Muntjac. They fired
several times, but did not succeed in killing the animal during
the time we watched the sport.

A few li before reaching Ché-kou-tzu we passed an un-
usually large house of much architectural beauty. It was
erected by a rich man named T’ao, who held the purchased
rank of “‘ Hsien.’”” He died some twenty years ago, and the
family has fallen on evil times, thanks to idleness and opium.

The flora was not very interesting. Some fine trees of
Cypress and odd ones of Catalpa Duclouxit were noteworthy.
Pine abounds, and I saw several examples of “ clustered
cones.’’ These cones, a hundred and more crowded together,
were all small, and appeared to have displaced the male
flowers. Ché-kou-tzu, alt. 2050 feet, consists of some forty
houses situate above the mouth of a stony stream and backed
by low mountains, on the top of which is an ancient fort.

On leaving Ché-kou-tzu we immediately entered a pretty
valley, highly cultivated with rice and bounded by low,
rolling hills. A large number of farmhouses and a small
hamlet occur in this small but prosperous valley. Through-
out the whole forenoon we traversed a number of such depres-
sions separated one from another by low ridges, always
ascending slightly with the valleys narrowing until finally they
become mere basins surrounded by rocky limestone mountain-
tops. Crossing a final ridge we entered Kai Hsien at a place
called Shih-ya-tzu, 35 li from Ché-kou-tzu. Up to this point
the scenery is very pretty, the rocky mountain-tops being
clothed with woods of Pine and Cypress. Oak is common,
and in more open places and around habitations we passed fine
trees of Spondias, Pistacia, Paulownia, Tapiscia, and Hovenia
dulcis ; the last covered with masses of white flowers.

The afternoon’s journey was all downhill, ending in a very
precipitous descent to Wén-tang-ching. The road led through
maize plats, odd rice fields, and bare, treeless hilltops, with
no flora of interest. Nearing our halting-place for the day it
was fearfully hot, and the absence of shade was severely felt.

80 A NATURALIST IN WESTERN CHINA

Here and there the hilltops are crowned by old forts built
of dressed stone. These relics (Chaitzu) of turbulent times
abound all over the salt districts and more wealthy regions
of eastern Szechuan. Limekilns, small clay-covered affairs,
were common en voute, and many of the rice fields had been
dressed with slaked lime.

Wén-tang-ching is a town of considerable size, by far the
largest place we had met with since leaving Ichang. It is
built on steep slopes bounding the two sides of a clear-water
stream, and backed by high limestone cliffs. On the south-
west side these cliffs are stark and sun-baked. Large quan-
tities of salt are produced here. The brine pits are situated on
the foreshore and immediate neighbourhood of the stream. The
supply depends on the state of the river, the lower the water the
more brine is obtainable. During summer floods the industry
is suspended. The salt is white, powdery, of moderate quality,
valued at twenty-six cash per 16-ounce catty. It is dis-
tributed throughout the north and west of the Hsien, but
cannot enter the city of Kai Hsien itself. Dust-coal is mined
in the neighbourhood and used for evaporating the brine.

The town consists of about a thousand houses and boasts
several temples and large guild-halls, that belonging to the
Shensi guild being very prominent on account of its large size
and ornate architecture. Two small pagodas protect the luck of
the place, and many Chaitzu crown the surrounding hills. The
inhabitants are not prepossessing, being unusually dirty and
over-curious. Some were not over-civil, and there was a slight
scuffle between my men and some rowdies. Our inn was dark,
suffocatingly hot, and most undesirable in every way. It was
the best we could find, and served its purpose, uncomfortable
as it was. Behind the inn is a huge cave with vast stalactites
and a cool breeze blowing through it. This is the curiosity of
the town, and was pointed out with a great show of pride.

All along the route from Taning Hsien there has been much
argument over the price of food-stuffs. The natives constantly
putting up the price on my men, this led to heated words, but
generally ended in the men getting a fair price. Many of
them had travelled too far.not to know “ the ropes.”

Wén-tang-ching is only 750 feet altitude, and with the

EASTERN SZECHUAN 81

heat from stark surrounding cliffs and hundreds of furnaces
is a regular inferno. Prosperous it may be, but it failed to
appeal to us, and one and all were glad to quit it.

A steep ascent of a few hundred feet and we cleared the
town. After passing through a large graveyard we descended
to an alluvial valley where much sugar-cane, maize, tobacco,
and a little cotton is cultivated. The road is broad, paved
with blocks of hard stone, and traverses the valley to its head
at Ma-chia-kou, 12 li from Wén-tang-ching. Ma-chia-kou is
the coal port for the salt-wells. Coal is carried overland
some 30 li, and at this point put into small boats and conveyed
to the brine-pits. This coal is valued at three cash per catty,
the carriers receiving one cash per catty for carrying it down.
The boats are small, steered by sweeps fore and aft, and can
descend this stream to Kai Hsien, 60 li below Wén-tang-ching,
and from thence to Hsiao Ch’ang on the Yangtsze, 110 li
distant. At Ma-chia-kou the road leaves the main stream,
which flows down from the northward, and after crossing a
neck descends to a broad stony torrent, which it ascends through
uninteresting country, eventually leading through a limestone
ravine. The coal supply is of primary importance to the salt-
wells, consequently the road is kept in good repair. During
the forenoon we met hundreds of coolies and many women
laden with coal. Iron is found in this neighbourhood, and
pigs of this metal were being carried down to the boats.

On leaving the above-mentioned ravine we traversed a
valley of rice fields and reached Yi-chiao-tsao about noon.
Five li above this hamlet we crossed over, and during the rest
of the day’s march descended a narrow valley flanked by steep
Cypress-clad slopes. Sweet potato is abundantly cultivated,
also rice and maize. Houses are frequent, and the people
appear fairly well-to-do.

We found lodgings for the night at Wang-tung-tsao,
alt. 1350 feet, having covered our usual 60 li. The day was
terribly hot, making the journey very fatiguing. The inn is
beautifully situated in a grove of Bamboo and Cypress, but is
poor and abominably stinking. Really, it is a pity that such a
vile house should defile such a charming spot.

The next day was also grilling hot, with no signs of a storm

VOL. I.—6

82 A NATURALIST IN WESTERN CHINA

to cool the air. Descending a few li we struck a rather broad
stream with many red-sandstone boulders in its bed. The
road ascends this stream to its source, and steep ascents and
descents were all too frequent. We lunched at the village of
Kao-chiao, and a more hot, fly-infested, stinking hole, with
people more inquisitive, I have not experienced. Savage,
snarling, yelping dogs abounded, and these, with the other dis-
comforts, did not add relish to the meal. My followers seemed
to share my views of this village, and grumbling and male-
diction were loud on allsides. Our meal did not occupy long,
and we all felt better when clear of this filthy, pestiferous place.

The whole day was spent among sandstone, grey and red,
and we were seldom out of sight of rice fields. Pine abounds,
but the Cypress does not appear to be at home here, and occurs
very sparingly as compared with previous days. Wood Oil
trees are common, but the flora generally is not interesting.
Eleagnus bushes are common and were in ripe fruit. The
stems of this shrub (Shan-yeh-wangtzu, or Yang-ming-nitzu)
are commonly used for making the long stems of tobacco
pipes so frequently seen in this region. The Burdock (Arctium
major) is common in stony places and often cultivated, being
used as medicine under the name of “‘ Yu-pangtzu.”

Three li before reaching our lodgings we crossed a ridge,
and passing through a stone gateway, entered the district of
Tunghsiang Hsien. We found an inn at P’ao-tsze, a small
scattered hamlet, alt. 2650 feet, 65 li from Wang-tung-tsao.
The inn is clean and prettily situated in a little valley bounded
by low red-stone hills all under cultivation. The host is
evidently a man of substance, and amongst other things owns
a reclining chair of novel workmanship, of which he is evidently
very proud.

There was no breeze last night, and I slept badly, partly
owing to the heat and partly to the occupants of the inn talk-
ing in high argumentative tones till past midnight. This is a
common habit of the Chinese and very exasperating to any one
trying to get to sleep.

With only 50 li to do to Nan-pa ch’ang the men were in
high spirits and set out in style. The road proved easy—by
one o’clock we had covered the distance, and had a couple of

‘Ld &@ HIMID “TIVE “1a 09 (SISNHNIHO VIOVISId) OTHOVISId ASHNIHO AHL

EASTERN SZECHUAN 83

long rests into the bargain. On leaving P’ao-tsze we made a
short, steep ascent, and then descended by an easy road lead-
ing over and among sandstone bluffs. Twenty-five li on we
reached the bed of a small stream and followed it to its union
with a large, clear-water stream flowing down from the north-
ward. This stream flows past Nan-pa ch’ang and is navigable
for small boats down to Tunghsiang Hsien and up-stream some
290 li to Tu-li-kou. Near our destination we passed many
coolies carrying down bright anthracitecoal. This comes from
Fu-che-kou, some 50 li away, and the men receive 200 cash
per picul (100 catties) for carrying it down. We also noted
iron in flat slabs, which comes from Tung-che-kou, 25 li distant.

Pine was again the common tree, but Cypress also was
fairly common. The sandstone is evidently more favourable
to the Pine than to the Cypress. We saw two or three trees of
the rare “‘ Hung-tou-shu’”’ (Ovmosia Hosiei). The wood of
this tree is highly valued and so heavy that it sinks in water.
Wood Oil trees continued abundant, and around Nan-pa ch’ang
plantations of Mulberry were being made. Evidently seri-
culture is about to be attempted in this district.

Nan-pa ch’ang, alt. 1550 feet, is a village of considerable
size, and is built on a flat bordering the stream. Formerly it
was one of the most important centres of the opium trade
in Szechuan, and its product was of very superior quality.
The opium trade is now completely stopped, and this place
has suffered tremendously in consequence. It also boasts a
trade in general merchandise, supplying a large area of country
to the northward. But opium was its real source of wealth,
and with the disappearance of the opium traffic all trade has
declined. To the northward a lot of tea is grown and the
leading people of Nan-pa ch’ang are endeavouring to divert
this trade from its present headquarters, Taiping Hsien, to
their own village.

Around Nan-pa ch’ang there are a few Mantzu caves.
Everything was very quiet in the village and we attracted
little or no attention. We saw a couple of uniformed police,
odd street lamps, and other signs of modern ideas. Leaving
this village the next morning at 7 a.m., in four small boats, we
dropped down the beautiful clear-water stream, and reached

84 A NATURALIST IN WESTERN CHINA

Tunghsiang Hsien at 3 o’clock. The distance is 140 li by water,
go li by land. Numerous rapids obstruct the stream, but
since the volume of water is comparatively small they are not
dangerous. The river is bounded by sandstone cliffs, often
steep and covered with Pine, Cypress, and mixed shrubby
vegetation. Chaitzu are common, and here and there we
passed villages. Cultivation is general, and the crops were
beginning to show signs of suffering from drought. Pulse in
variety is abundantly cultivated, together with rice and other
favourite articles of food. Ordinarily the whole region is one
of plenty and prosperity.

It was a pleasant change dropping swiftly down this beauti-
ful river, and we all enjoyed the journey. On reaching Tung-
hsiang Hsien a thunderstorm broke and the rain cooled the air
delightfully.

We entered the city of Tunghsiang, alt. 1400 feet, through
the east gate, and found accommodation in a quiet and
moderately clean inn. The city, though not large, seemed a
busy place. Formerly it boasted a large traffic in opium, and
its general trade was then very considerable. It nestles among
low hills on the right bank of the river, and is faced on the
opposite bank by steeper and higher mountains. Sandstone
cliffs and bluffs abound, and in some respects the whole scene
reminded me of the country around Kiating Fu.

Our inquiries into the matter of currency disclosed the fact
that Szechuan dollars are accepted here, but 10-cash pieces
were still useless. The Roman Catholic and China Inland
Mission have established outstations here. An Irish missionary
belonging to the latter was staying here at the time of my visit,
and I enjoyed for an hour or so the pleasure of his company.
It was pleasant to hear my own tongue spoken again. Not
since leaving Ichang, 35 days before, had I encountered a
single foreigner.

CHAPTER Viti
THE ANCIENT KINGDOM OF PA

NARRATIVE OF A JOURNEY FROM TUNGHSIANG HSIEN
TO PAONING Fu

or Paoning Fu descends the river via Suiting Fu to Ch’u

Hsien, then strikes westward to Chengtu, north-west to
Paoning Fu. I had no fancy for the main route, since, by
going due west from Tunghsiang Hsien to Paoning Fu, we
should explore new ground. My map (War Office, Province of
Ssu-ch’uan, Eastern Sheet) gave no route, but indicated villages,
and it was evident, therefore, that these villages were connected
by a road of some sort. Chiangkou seemed a good place to
start for, so my men were instructed to find a cross-country
road to this town. At first the innkeepers, chair hongs, and
local officials denied all knowledge of any such road, and indeed
of sucha place. But any one who has travelled in China values
such denials at their proper worth and is not discouraged. The
men who had charge of these inquiries were trusted followers of
ten years’ standing, andthough entirely ignorant of the geography
of the region could be relied upon to ferret out a route if such
existed. After about six hours’ investigation, from the magis-
trates Yamén downwards, I was informed that a small
mountain road did exist, but was over hard and difficult
country, affording the poorest of accommodation. This was
sufficient ; they were told to get an itinerary of this route and
engage a few local men as extra carriers. I went to bed
about 10.30 p.m., satisfied that by 6 o’clock next morning
everything would be ready for our cross-country jaunt. In my
travels about China I have been singularly fortunate in never
having any trouble with the Chinese. In the spring of 1900

‘ ROM Tunghsiang Hsien the recognized route to Chengtu

86 A NATURALIST IN WESTERN CHINA

I engaged about a dozen peasants from near Ichang. These
men remained with me and rendered faithful service during
the whole of my peregrinations. After a few months’
training they understood my habits thoroughly and never
involved me in any trouble or difficulty. Once they grasped
what was wanted they could be relied upon to do their part,
thereby adding much to the pleasure and profit of my many
journeys. When we finally parted in February Ig1I, it was
with genuine regret on both sides. Faithful, intelligent,
reliable, cheerful under adverse circumstances, and always
willing to give their best, no men could have rendered better
service.

This cross-country journey from Tunghsiang Hsien to
Paoning Fu, via Chiangkou, promised to be of morethanordinary
interest. There was a novelty about it also, since there was no
record of any foreigner having attempted it before. The route
lay across the old kingdom of Pa (see Chapter VI, p. 66), and I
hoped to find some evidences of this ancient race. Chinese
history is dry, difficult reading, and it is hard to dig out solid
facts. Wars, rebellions, and massacres deluge everything in
blood; the arts of peace are seldom given any prominence.
The Chinese historians have always treated the aboriginal
races with arrogant contumely, rendering it almost impossible
to discover at this late date anything about the arts and life
of these lost peoples. That the modern province of Szechuan
boasted kingdoms and dynasties of its own before the advent of
the early Chinese is historical fact. The first Emperor of the
Ts’in dynasty, Tsin-shih Hwang or Shih Hwang-ti(221-2009B.C.),
incorporated part of the kingdom of Pa with the rest of
his dominions and nominally also that of Shu, whose capital
was near modern Chengtu Fu. The succeeding Han dynasty
(206 B.c. to A.D. 25) made the conquest complete. Since this
time no aboriginal chief has ruled the Red Basin of Szechuan,
though it has been conquered and re-conquered time and again
by usurping Chinese and alien races. During the period a.D.
221-265, the Chinese Empire was divided into three kingdoms,
one of which, under the Emperor Liu-pei, had its capital at
Chengtu. Liu-pei and three of his generals and statesmen are
handed down as popular idols, and everywhere in Szechuan

A SANDSTONE BRIDGE WITH CYPRESS AND BAMBOOS

THE ANCIENT KINDGOM OF PA 87

stories are told of the doughty deeds accomplished by these
heroes of old. With this brief introduction I again take up my
narrative :—

My principal men once more proved equal to the occasion,
and on 8th July everything was arranged for our cross-country
journey. An itinerary had been made out, and the Hsien
provided us with a couple of uniformed soldiers. (He sent six,
but I managed to get them reduced to two.) Heretofore on
this journey we had managed to avoid taking official escort,
although it is the custom to do so in Szechuan. No ordinary
traveller desires this honour, but it is thrust upon him and
cannot easily be avoided. The presence of this escort renders
the officials responsible for the traveller’s safety in accordance
with treaty arrangements. It is necessary to pay these men a
few cash, but often they prove useful in odd ways. Cash is
cheap, and an extra hundred per day for each soldier does
not amount to any considerable sum. The difficulty is in
keeping the escort down totwomen. Four andsix are common
numbers, and if one did not protest continuously an almost
unlimited number of authorized and unauthorized ragamuffins
would attach themselves to one’s caravan. If there is cash to
be made the legitimate escort is often not above farming in a
few extra “‘ hands,” thus securing more money. The escort is
provided with a letter from the official supplying it wherein the
number of men dispatched and their destination is given, so
by examining this it is possible to check any attempt at fraud.
On dismissing these men at their journey’s end it is necessary to
give them acard to carry back to their superior. Their letter is
stamped by the official who provides the new escort, and the
card signifies that their duty has been satisfactorily carried out.
If they return without a card for any reason or other they are
liable to be punished.

Leaving Tunghsiang Hsien by the west gate we followed
the main road to Suiting Fu for a few li, then branched off to
theright. The road is well paved, and we met plenty of traffic.
For the first 20 li the road is practically level, winding in and
out among low hills. It then makes a steep ascent to the top of
some bluffs, where Mien-yueh ch’ang is situated, 30 li from the
city of Tunghsiang. Throughout the rest of the day the road

88 A NATURALIST IN WESTERN CHINA

was easy, leading through and among low hilltops and shallow
valleys intercepted by hills 300 to 500 feet high. Cypress and
Pine are abundant, so also are Pistacia and Albizzia lebbek,
both making large umbrageous trees. Vitex Negundo is the
‘commonest shrub, sometimes attaining to the dignity of a small
tree : it was everywhere covered with masses of lavender-purple
flowers.

The country is highly cultivated. Rice predominates, with
various kinds of beans (especially Lutou, 7.e. green beans) next
in importance—both crops evidently follow after wheat. We
passed odd patches of cotton and very many Plum trees.
The region is well populated, bypaths abound, and it was
no easy matter for us to keep to the right road. At one point
the road bifurcates, one branch leading to Shuang-ho ch’ang,
the other to Shuang-miao ch’ang, our proposed halting-place
for the night. The names of these two villages, when spoken
rapidly, sound much alike, even to Chinese ears. My men got
somewhat confused, and for a time there was danger of the
caravan following two divergent routes.

We passed through the market village of Wang-chia
ch’ang (ch’ang signifies market village), a curious little place,
dominated by a temple in the middle, the roofs of the houses
uniting to form a central covered way, beneath which the road
passes through the village.

Shuang-miao ch’ang was our intended destination for the
day, but being market day the village was filled to overflowing.
A hundred or more people followed us into an inn, and in
a little while there was hardly room to breathe. Many were
obviously under the influence of wine. It was too hot to
tolerate such overcrowding curiosity, so we pushed on a
further 5 li, where we happened on a decent farmhouse, which
we commandeered. The owner being away, his wife was at
first sorely afraid, but in a couple of hours her confidence was
gained and all was well. The men had difficulty in obtaining
food and lodging. The majority went back to the village,
but none complained: they all realized the impossibility of
my remaining the night in such a crowded place.

Our quarters were new and shaded by a grove of Bamboo
and Cypress, but mosquitoes were multitudinous, rendering

THE ANCIENT KINGDOM OF PA 89

life miserable. The place is called Hsin-chia-pa, alt. 1950
feet. We had covered 80 li, through a rich and interest-
ing country. Lady Banks’s rose was particularly abundant,
with stems 2 feet round, festooning trees 40 to 50 feet tall.
Mantzu caves occur sparingly. In several places we passed
cultivated patches of Panicum crus-galli, var. frumentaceum.

We parted excellent friends with our hostess at Hsin-chia-
pa, a trifling present and 400 cash made her extremely happy ;
her thanks were both genuine and profuse. Soon after
starting we made a precipitous ascent of 1000 feet and crossed
what is probably the water-shed of the Suiting and Sanhuei
Rivers. A descent led to the head-waters of a small river,
where is situated the tiny market village of San-che-miao.
Market was in full swing, the one short street with its few
hovels being crowded with people. We passed through
without stopping to satisfy the curiosity of the crowd. At this
village several roads converge, the one we followed continuing
to descend the stream, and leading through a rocky jungle-
clad defile. The cliffs are of red and grey sandstone, steep,
rugged, and crowned with Pine and Cypress. As fluviatile
shrubs Distylium chinense, various Privets (Ligustrum) and
Cornus paucinervis abound. The last-named is a low-grow-
ing shrub with spreading branches, and laden with small
flat corymbs of white flowers it formed a most attractive
bush by the water’s edge. In the jungle-clad slopes through
which the road winds Tea bushes 15 feet and more tall are
common. They looked uncommonly like spontaneous speci-
mens, but were possibly planted long ago, though some of them
have been undoubtedly naturalized. Occasional trees of the
Red Bean (“ Hung-tou”’), Orvmosia Hosiet, occur ; at one time
this was probably a very common tree in this region. Its tim-
ber is most valuable, and the tree has been ruthlessly felled.
There is practically no cultivation in this defile, or room for
any, and not a house for 20 li.

After traversing this wild and interesting ravine for several
hours we made a steep ascent to the top of the cliffs, and on
the way up discovered spontaneous plants of the Tea Rose
(Rosa indica) in fruit. These were the first really wild specimens
I had met with. Once on top of the cliffs we found that

go A NATURALIST IN WESTERN CHINA

the country all around is under cultivation, chiefly rice, with
houses at frequent intervals. After a few li the road descends
to the river again, and crossing by stone steps we reached the
market village of Peh-pai-ho, where we found accommodation
in a large house. This village, alt. 1600 feet, also known as
Peh-pai ch’ang, is a small place with unprepossessing residents.
Our quarters were dark, fairly filthy, and loafers crowded
around until bedtime.

The day’s journey of 60 li was through a sparsely populated
country, which, considering the low altitude, was unusually
wild and jungle-clad. The flora had points of interest, the
finding of Tea bushes and bushes of the Tea Rose in the rocky
defile being particularly noteworthy. On bare sandstone
cliffs large white trumpet-flowered Lilies were common, with
their stems thrust out at nearly right angles to the cliffs.
We met very few people on the road, and most of the women
“we saw had natural feet. In the early morning we passed quite
a lot of Panicum crus-galli, var. frumentaceum, cultivated.

The itinerary my men secured at Tunghsiang Hsien did
not err on the side of accuracy. Constant inquiries were
necessary, but the results were confusing. The river which
flows past Peh-pai ch’ang was said to unite with the Chiangkou
stream at Chiang-ling-che, 70 li distant.

A heavy thunderstorm occurred in the night, accompanied
by a downpour of rain which lasted intermittently into the
early forenoon of the next day. The country needed rain
badly, and the air was cool and fresh in the morning. Peh-pai
ch’ang is a regular warren of dilapidated houses, filthy and
stinking, with a loafing and unduly curious population. A
loin-cloth belonging to one of my chair-bearers was stolen
during the night, and my followers had little that was com-
plimentary to say about the village or its inhabitants.

Following the river down-stream for 5 li, we reached Lei-
kang-k’éng of the maps. This hamlet (pronounced Lei-kang-
t’an, from a fine waterfall on a small river which, flowing
from the north, joins the main stream at this point) consists
of a deserted temple, a few scattered houses, and an old fort
high up on the cliffs. It and Ta-chén-chai, another old
fortress, are the only places marked on the map—both are

ONV.HO ONTIVL 40 AOVITIA LANYUVN AHL

THE ANCIENT KINGDOM OF PA gI

to-day of no importance. The market villages, the real
places of importance, are not shown. Maybe these villages
have sprung up comparatively recently, and the forts, from
long-continued peace, lost their importance. This is the
only feasible explanation which occurs to me. This section
of the country is only known from Chinese maps, and these
were probably compiled during military times long ago.

From Lei-kang-k’éng a steady ascent for 30 li leads to the
top of a ridge where is situated the important market village
of Peh-shan. This place boasts a fine temple and about a
hundred houses. Like all such villages in these parts it con-
sists of one central street, practically closed over by the nearly
uniting eaves of the houses. These market villages are a
striking feature of this part of Szechuan. They are situated
approximately 30 li apart, and nine markets are held monthly
in each. These are arranged in such manner that the three
villages lying nearest to one another hold market on different
days, thus between them practically covering the month.
On market days the country-folk assemble from all sides to
buy and sell. Pedlars and itinerant merchants constantly
journey from market village to market village. Such markets
are of the highest importance in a sparsely populated country,
but the denizens of these villages suffer from too much spare
time. Market days are what they exist for, and on: the other
days are mainly spent in gambling and sloth. This system
of market villages dates away back to the very dawn of
Chinese civilization, and in the region we are concerned
with here, is very little changed from what it was in the
earliest times.

Five li before reaching Peh-shan ch’ang we struck a road
which comes from Suiting Fu, 120 li distant. The country
hereabouts is split up in low mountain ranges, averaging
3000 feet altitude, composed of grey and red sandstones.
The river-valleys are mere ravines clothed with dense jungle,
Pines, and Cypress, with no bottom lands nor cultivation of
any sort. Some 500 feet up the cultivated area begins and
extends to the summit. Terraced rice fields abound, tier
upon tier, intercepted by low bluffs, the tops of all of which
are cultivated. The whole country is very pretty, and in

92 A NATURALIST IN WESTERN CHINA

many respects peculiar, as far as my experience goes. Most of
the women have natural feet, and many were busy weeding
and firming the rice plants.

On leaving Peh-shan ch’ang the road makes a steep descent
to a stream and a correspondingly steep ascent to the top of
the bluffs again, winding round to the crest of a ridge where
is situated the market village of Yuen-fang. This place,
alt. 3100 feet, which was our destination for the day, hav-
ing covered the allotted 60 li, is prettily situated. We found
lodgings in a new and clean house boasting a veranda over-
looking a grove of Pine and Cypress trees. The crowd which
collected was small and though inquisitive kept at a respectful
distance.

The flora proved identical with that of the previous day’s
journey. I again met with sub-spontaneous Tea bushes in
the jungle and also saw a number of the Red Bean trees.
Perhaps the most interesting objects noted during the day
were the tombstones. These are very different from any I
have seen elsewhere. They are of freestone, often highly
sculptured, the workmanship being superior and the effect
both artistic and dignified. One or two old stone mausoleums
were magnificently sculptured. The aboriginal population
of this region were accomplished workers in stone, and their
work may have served as patterns for the Chinese to copy
from. In conception the designs are evidently not pure
Chinese, and I strongly suspect “‘ Mantzu’”’ influence, to use
the Chinese term for the aboriginal population.

At Fu-erh-tang there is a particularly fine family temple,
and near by a Mantzu cave in an isolated piece of rock.
Around many of the mausoleums and family temples ancient
stone pillars (wei-tzu, 7.e. masts) occur. Wayside shrines and
small temples, dedicated to Kwanyin (Goddess of Mercy) and
to the tutelary genii are common, the images being carved in
stone and mostly coloured blue and white. The day’s journey
was more than usually interesting ; somehow one felt instinc-
tively that one was traversing a region closely associated with
man from very ancient times.

Leaving Yuen-fang ch’ang soon after 6 a.m., we traversed
country similar to that of the day before, and reached Pai-

THE ANCIENT KINGDOM OF PA 93

(pronounced P’an)-miao ch’ang at 10 o'clock. Here, contrary
to what my map indicated, I found no river. Replies to
inquiries gave it as 30 li farther on, and so it proved. The
map for this region is hopelessly inaccurate, and it was quite
useless attempting to be guided by it. Pai-miao ch’ang is a
small village built on the top of a ridge and surrounded in
part by woods of Cypress and Pine. Crossing an undulating
area we descended by an easy path, finally reaching the
T’ungchiang River, 10 li above Chiangkou. This river is fully
roo yards broad, with red-coloured water and a sluggish
current. Boats were easily secured and we dropped down-
stream to Chiangkou, which we reached at 3 o’clock, just before
a heavy thunderstorm broke. The day’s journey was said to
be 7o li, the road was easy, with flora and scenery identical
with that of the preceding days.

Chiangkou (alt. 1600 feet) is the second town in size and
importance in the department of Pa Chou. It consists of
about 500 houses, built on the fringe of a promontory between
two rivers, backed by low, steep, well-wooded hills. The
rivers unite at this point and are navigable downwards to
Chungking. The more easterly stream descends from T’ung-
chiang Hsien, the westerly stream from Pa Chou, each town
distant from Chiangkou 180 li. Both streams are navigable
for small boats up-stream to these towns.

A Féng Chou (official next below a Chou in rank) resides
at Chiangkou. From a distance the town looks well-built
and prosperous, but it does not improve on closer inspection.
The position is admirable and undoubtedly the town is of con-
siderable commercial importance, yet we had great difficulty
in exchanging twenty taels of silver. Like other towns we had
passed through, Chiangkou was feeling the suppression of the
opium traffic severely, and until new industries arise to take
the place of the opium trade the resources of all these places
will be crippled.

We found accommodation in a poor but quiet inn, and,
thanks to the thunderstorm, no curious crowd gathered to
annoy us. My principal men spent several hours in finding
out a cross-country road to Yilung Hsien, and eventually
succeeded.

94 A NATURALIST IN WESTERN CHINA

On leaving Chiangkou we ferried across the Pa Chou River
and then made a steep ascent of a few hundred feet. The
rest of the day we meandered along the crest of a range of low
mountains, following an undulating path. In parts the road
was good, in others ankle deep in slippery mud. Thunder-
showers fell at intervals and it was fairly cool.

The country generally is similar to that traversed during
previous days. Tobacco is a rather common crop hereabouts
and we saw a little cotton. Maize is very rare, but rice is
abundantly cultivated. Shrines and small temples continued
common and in good repair. Kwanyin and Tuti are the
common deities, the latter representing an old man and
his wife, constituting the tutelary genii. Dignified, ornately
carved tombstones and mausoleums were everywhere in
evidence.

Our intended destination for the day was Chén-lung
ch’ang, 60 li from Chiangkou, but on reaching there we found
market in full swing, and, to avoid the crowd, we journeyed on
another 6 li. On market days these villages are impossible,
from the foreigner’s point of view. I rode through this village
in my chair, and the crowd which gathered at the upper end
of the place mustered several hundreds. Wine appears to flow
freely on market days and many were under its exciting in-
fluence. Prudence as well as comfort therefore demands that
one avoid all crowds as much as possible when travelling in the
interior regions of China. Women attend these markets in
force and appear to be a power in this part of the Celestial
Empire. Their bearing and manners generally are very free
for Chinese women ; natural, unbound feet are the rule.

Chén-lung ch’ang is clustered on the narrow neck of a
sandstone ridge, and in common with all such villages boasts
a fine village temple. We lodged for the night in a poor way-
side inn at Hei-tou-k’an, alt. 3100 feet.

The next day was cool, with showers at odd times, but of no
consequence. With the exception of one steep descent and
an ascent in the late afternoon, the road was more or less level
all day, traversing the tops of the low mountains. These
sandstone mountains are dissected by innumerable deep,
narrow ravines, clothed with Pine, Cypress, and a dense jungle

‘Lad 02 HLYID “TIVE “LA 09 (IHISOH VISOWYMO) AUNL NVAAGAA WAHL

THE ANCIENT KINGDOM OF PA 95

of miscellaneous shrubs. Unlike limestone country no bottom-
lands are formed, and cultivation is relegated to thehigher parts
of the ranges. Farmhouses are scattered here, there, and
everywhere, but the villages are all situated on the tops of the
mountains, most frequently on the divide of a ridge.

Fourteen li from Hei-tou-k’an we passed through the
village of Tai-lu ch’ang, where market was in progress and
many pigs on sale. Thirty li from this place we passed Ting-
shan ch’ang, a village of considerable size, charmingly situated
on the neck of a ridge, backed by a Chaitzu and a fine cypress
grove. Chaitzu, of which frequent mention has been made,
are a feature of these parts. They are old forts, said to have
been mostly constructed during the great sectarian rebellion
of A.D. 1796-1803. A small official (Hsao-shoa-tang) resides
at Ting-shan ch’ang. In spite of its fine situation this village
was unusually filthy and was dominated with the strong
odours of a wine distillery. The usual crowd of loafers
followed us for some distance on quitting this village.

In the late afternoon we arrived at Lung-peh ch’ang,
alt. 3000 feet, after travelling 74 li. We lodged in a rambling,
dilapidated inn, fairly clean, with rooms removed some little
distance from the street—the village sewer. Market not being
in progress the crowd of inquisitive idlers was relatively small.

The flora was not particularly interesting, but we passed a
number of fine Camphor trees (Cinnamomum Camphora). The
crops, however, were rich and varied. Rice and sweet
potato preponderate, odd patches of cotton were noted
and also others of Indigo (Strobilanthes flaccidifolius). In the
afternoon coolies laden with salt passed us. This salt is pure
white and granular and comes from Nanpu Hsien. From our
lodging Ting-shan ch’ang was visible, 30 li distant and nearly
due east. The map shows a river flowing past this village,
but the only one we could get tidings of was 50 li from that
place.

After a comfortable night’s rest we continued our journey
through country similar to that of foregoing days, but less
well-wooded and more inclined to be arid, with broader
valleys more under cultivation. Our route followed the
boundary between Pa Chou and Yilung Hsien. We passed

96 A NATURALIST IN WESTERN CHINA

through two market villages and stayed for the night in a
farmhouse 1 li before reaching Fu-ling ch’ang, alt. 2800 feet.
We purposely stopped at this place in order to escape market
day at the village, but did not avoid a constant crowd until
after dark, when the doors were closed. We found all these
crowds quiet and orderly enough, but a continuous mass of
faces, with wooden expression, blocking the doorway, obstruct-
ing light and air, is very trying. Immensely useful as these
markets are to the country-side, they have decided drawbacks
from a traveller’s point of view. A good police force is really
more necessary in these villages than in the cities. The more
lawless element fears a Hsien (Magistrate), but has little respect
for a Ti-pao (Village Head-man). Local produce is mostly in
evidence in these markets; a few needles, aniline dyes,
trumpery odds and ends, chiefly of Japanese origin, are about
the only foreign goods met with.

We saw more cotton during the day than we had else-
where observed on this journey, and the crop looked flourish-
ing. Kao-liang (Sorghum vulgare) was a common crop, but
rice and sweet potato again preponderated. The sorghum
and rice were bursting into ear. Wood Oil trees occur, but
are not plentiful, and commercially this crop is unimportant
hereabouts. Mixed with the cotton were odd plants of the
oil-seed yielding Sesamum indicum (‘“ Hsiang-yu’’).

In the late afternoon we traversed country which somewhat
resembled that around Tunghsiang Hsien—on all sides, as
far as the eye could see, nothing but ridge upon ridge of low
sandstone mountains. These ranges average about 3000 feet
in altitude, those to the east and north being higher than
those to the west and south. The map is all wrong for the
region, so I could not definitely place our route. The river
Sheng-to, so boldlyindicated, escaped us, though we should have
crossed it had the map been correct. The market villages
passed were smaller than heretofore, very filthy and stinking,
yet most charmingly situated on the neck of low ridges, and
well shaded with trees. Camphor trees are very common,
and “‘ Pride of India’’ trees (Melia Azedarach) particularly
abundant. The stage said to be 70 li proved very easy, the
weather being dull and cool.

THE ANCIENT KINGDOM OF PA 97

Our stay over at the farmhouse was hardly a success;
we had a full crowd until bedtime, and in spite of fair promises
four of my men who remained in the house with me had
neither dinner nor bedding. As a punishment I paid only
half our usual rate, much to the householder’s chagrin. Fu-
ling ch’ang was quite deserted when we passed through in the
early morning. It occupies the narrow neck of a sandstone
ridge, after the usual manner of these villages. The same is
true of Shih-ya ch’ang, 30 li farther on. Ten li beyond this
latter village we passed a nine-storied pagoda and sighted the
town of Yilung Hsien, to the northwards, about a mile distant
as the crow flies and at equal altitude (2500 feet). Yilung
is a very small town, situated on the mountain-top, backed
by a steep bluff and surrounded by a wall of dressed sandstone.
Two-thirds of the land enclosed within the city wall is given
over to cultivation. We passed to the south-west of the town
by a road which makes a steep descent and ascent and then
meanders along the tops of the mountains until Tu-mén-pu
is reached. The mountains are lower, more flat, the valleys
wider, and the whole country more treeless. Cotton is abun-
dantly cultivated throughout this region, and it is evident
that the district of Yilung produces a very considerable quan-
tity. Rice and sweet potato are the common crops, the
latter thriving on the hot almost soilless rocks. The earth is
drawn into ridges, often leaving bare rock between, and cuttings
are inserted. These cuttings, leafy shoots about 6 inches long,
quickly take root and form plants that produce an abundant
crop. Sorghum is fairly common in places, but maize is very
scarce. Stone monuments were less in evidence, but we passed
a fine O-mi-to Fu stone surmounted by a hideous T’eng-kou.
Six old hats protected this stone from the rain and sun; in
front was a huge mass of ashes and the remains of many
Joss sticks. We were informed that the tutelary genius of
this spot is renowned for his benevolence, and that it was
hoped shortly to erect a shrine over the spot.

We had been unfortunate in the matter of market days all
along, and found another in progress at Tu-mén-pu. Seemingly
having gained nothing by staying the night a little beyond or
before reaching these villages, we experimented and stayed at

VOL. I.—7

98 A NATURALIST IN WESTERN CHINA

one. It wasnotasuccess. A mobrushed our inn and bedlam
reigned for a couple of hours. Eventually it thinned down,
but many of the more insistent and curious remained until
bedtime. There was much noise, but the crowd was friendly
enough; nevertheless, I was glad it proved to be the last
market village of its kind we encountered before reaching
Chengtu.

Tu-mén-pu or ch’ang, alt. 1950 feet, 70 li from Fu-ling
ch’ang, is a large and prosperous village boasting much trade
on market-days. Something of everything in the way of native
produce was on sale, and the narrow street was thronged to
overflowing. Five li before reaching this place our road con-
verged with one leading to Pa Chou city by way of Yilung
Hsien.

I had a poor night’s sleep in consequence of loud talking
being carried on far into the early hours, a woman (as usual)
being the principal offender. This was an emphatic reminder
of the hubbub of the crowd which besieged us on arrival, and
I was really glad to quit Tu-mén-pu. A few li beyond this
village we branched off from the main road, which goes to
Nanpu Hsien. Much salt comes from this township, and
during the last two or three days we had met many carriers
laden with this commodity.

Forty li beyond Tu-mén-pu we passed the poor village of
Shui-kuan-ying, protected by dilapidated gates which denote
its former military character. In years gone by it was a
barrier of some considerable importance. Twenty li farther
on we reached the village of Chin-ya ch’ang, alt. 2150 feet,
which differs from all we had met with heretofore in having a
broad main street fully exposed to the heavens. To our great
joy market was not in progress. We found lodgings in a new
and quiet inn, which proved a welcome change; the people,
too, were courteous and much less inquisitive. The day was
exceptionally hot, and all were glad to reach the end of the
allotted stage of 60 li. Twelve li before reaching Chin-ya
ch’ang we struck a main road leading from Nanpu Hsien,
and following it entered the village through an isolated ornate
gateway. Beyond the village is a bluff of grey sandstone
studded with square-mouthed caves. These caves are crude

a 5

THE ANCIENT KINGDOM OF PA 99

imitations of Mantzu caves, and are of recent origin, and
purely Chinese.

The day’s journey was through less interesting country
than usual. The broad valleys and nearly treeless mountains
are all under cultivation. Cotton was again common in the
forenoon, but much less so afterwards. This crop looked as
flourishing as Chinese cotton usually does. Tobacco is spar-
ingly cultivated. The tobacco leaves are merely sun-dried before
using, and the quality is therefore poor. Sweet potato was
more plentiful than ever; the arid sandstone rocks evidently
suit this crop. Rice was, of course, everywhere abundant,
sorghum common, but maize was very scarce and suffering
from drought. The Irish potato is very little cultivated in
these parts. Around Tu-mén-pu white-wax is produced in
small quantities on the Privet (Ligustrum lucidum), but the
cultivation is slovenly carried out, the trees being dwarf and
ill-cared for. A few Cypress trees were noted, but Paulownia
is a common tree, and Wood Oil trees rather plentiful. A little
silk is raised, but the industry is unimportant hereabouts.
Odd trees of the Banyan (Ficus infectoria) occur near houses
and shrines. We passed a few fine tombs, but the average
headstone is less ornate than those formerly met with.

We experienced a brief but terrific thunderstorm during
the early hours of the morning, and rain continued to fall
slightly when we set out from Chin-ya ch’ang. For 20 li
we followed an abominable road of mud. This was very
greasy, and caused many of us to come “ croppers.”’ Ultim-
ately, we reached a paved road, and, 6 li farther, a tributary
stream of the Kialing River. This tributary is broad, broken
by cataracts and rapids, and quite unnavigable at this point.
It unites with the Kialing River, locally known as the “ Paoning
Ho,” at Ho-che kuan. This is a small riverine port boasting a
remarkably fine shop where coal, lime, and especially Chinese
wine (sam-shu), were on sale. On the paved road we met
several men carrying Bombay cotton yarn—the first example
of foreign goods we had encountered on the whole journey !

At Ho-che kuan the Kialing River is smooth and placid,
and when in flood is fully 400 yards broad. We ferried across
the river to the right bank, and then traversed an alluvial flat

100 A NATURALIST IN WESTERN CHINA

of considerable size, highly cultivated with rice and sorghum,
with here and there a little abutilon hemp. At the head of
this flat, some 10 l from the river, we crossed over some
levelled hillocks into a basin—evidently an old lake bed—
surrounded by bare mountains 200 to 500 feet high. This
depression was a lake of luxuriant padi (rice), with houses here
and there, nestling in clumps of trees. From this basin we
passed through a low, narrow gap between the hills, and came
abruptly to the Paoning River a little below the city itself. We
were ferried across and found lodgings in a large and fairly com-
fortable inn. The flora of the day’s journey was without special
interest, Cypress being the only kind of tree really common.
But shading some graves, opposite Ho-che kuan, occurs the
largest specimen of the “ Pride of India”’ (Melia Azedarach) I
have met with, This tree is 70 feet tall, and ro feet in girth.

Paoning Fu is a city of past rather than of present great-
ness. It is stilla most important administrative centre, but its
real interest lies in its great historic past. From the early days
of Chinese conquest it has been a strategical point of vast
importance. During the Ming dynasty (A.D. 1368-1644) a
generalissimo of forces had a palace here. The terrible rebel,
Chang Hien-tsung (A.D. 1630-46 circa), ravaged the country
roundabout, but spared the city itself. The result is that
many of the official residences and temples date back to
ancient times.

Formerly Paoning was the centre of a lucrative and thriving
silk industry, but this has steadily declined during the last
twenty years, and to-day it is a mere figment in comparison.
Attempts are now being made by the officials to rejuvenate
and foster this industry, which apparently failed more through
lack of business ability and tenacity than anything else. On
the neighbouring hills I was told “ wild silk” is produced, the
“worms ”’ feeding on the leaves of a scrub Oak, ‘‘ Ching-kang ”’
(Quercus serrata).

The city occupies an extensive alluvial flat on the left bank
of the river within an amphitheatre of low, bare, often pyra-
midal, hills, 300 to 600 feet high. Viewed from the opposite
bank there are no outstanding architectural features visible,
save a pavilion, which is practically the only building breaking

Na ONINOVd ALISOddO ‘NVOM DNIHO-NWN

iy | . fa

‘ ‘
yO ee

oe ae py Ce ew
ee We dues ir) ey

;
rs

THE ANCIENT KINGDOM OF PA Io

the monotony of level roofs. The area within the city walls
is largely occupied by yaméns, temples, and residences of the
more wealthy. Business is mostly carried on outside the city
proper, and is confined mainly to one street. Umbrellas were
the most noticeable articles on sale, but the city is famous for
its superior vinegar, great jars of which were on view.

Hedges of the thorny shrub, Citrus trifolata, are a
prominent feature of this city and its suburbs, giving to the
quieter streets a country-lane-like appearance. The water
supply of the city is from wells, which are often very deep.
This water is said to be good, but that supplied to our inn
had a very “earthy” flavour. From what I saw of the
city during a day’s stay there, I received the impression of
its being clean, its people very orderly and courteous, and the
decline in its prosperity most marked. The Paoning Ho is
a shallow river, and opposite the city about 500 yards broad
when in flood. It is navigable for boats of considerable size
downwards to Chungking. Up-stream small boats ascend
to Kuangyuan Hsien. A certain amount of merchandise
descends in small boats from Pikou, in Kansu, to Chaohua
Hsien. These rivers are most important to Paoning Fu, for,
in addition to export trade, the coal and wood used in the city
itself are conveyed over these waterways. On the right bank
facing the city is a ledge of cliff, on which nestle several
temples and pavilions, sheltered by groves of Cypress. In a
gap in this cliff is situated the busy little village of Nan-ching
kuan. Timber is very scarce around Paoning. Cypress wood
is commonly used in house-building; Alder wood (Alnus
cremastogyne) occasionally being employed for window frames,
etc., but its chief use is as fuel. Pine occurs, but, save as fuel,
is worthless. Cunninghamia, that most useful of Chinese
conifers, does not occur in this neighbourhood. The wood
of the Hung-tou tree (Ovmosia Hosiet), so highly esteemed for
carpentry, was formerly fairly common and cheap. To-day,
however, it has to be brought from a distance, and, in con-
sequence, is expensive. Oak and “ Huang-lien”’ (Pistacia
chinensis) are the only other timber trees of note. Paoning is
an important missionary centre, and the seat of a Protestant
bishopric. During my brief visit I had the pleasure of

102 A NATURALIST IN WESTERN CHINA

spending a few hours with the kindly and energetic Bishop
Cassels and certain of his coadjutors, who did all they could
to render my stay pleasant.

Leaving Paoning Fu and following the main road via
Tungchuan Fu, by easy stages I entered the city of Chengtu Fu
nine days later, having occupied fifty-four days on the journey
from Ichang.

The journey from Tunghsiang Hsien to Paoning Fu fully
bore out my expectations. The crowds on market-days were a
decided drawback, but not once was I insulted or called (in my
hearing) uncomplimentary names. The avaricious greed and
cunning of the inhabitants were most marked. They were
constantly putting up the prices of food-stuffs on my followers,
which led to much argument and high words, and several
times I was called upon to settle such disputes. The greed
of the Szechuanese peasant and small shopkeeper is a byword
among the Chinese of other provinces. The term ‘‘ Szechuan
Lao-ssu”’ (“Szechuan Rat’’) is applied derisively to the
whole population by the Chinese from other provinces.
Niggardly and avaricious they undoubtedly are, but they
are great agriculturists, and the question of the ‘ mote
and beam” may well be left open. As mentioned before,
the province is largely peopled by descendants of immigrants,
and these folk almost invariably style themselves men of the
provinces their ancestors came from !

The outstanding features of this ancient part of Szechuan
are :—

1. The elaborate system of market villages situated at
equal distances of 30 li apart, each with its nine market-days
per month, and alternating with the markets of neighbouring
villages. Each village is situated on the mountain-top and
usually on the neck of a divide, with one central more or less
covered street.

2. The rice belt is confined to the mountain slopes and
summits, the valleys being ravines, jungle-clad as a rule, with
little or no cultivatable bottom-lands. The highly cultivated
nature of the region and the presence of cotton in quantity
around Yilung Hsien.

3. The numerous fine mausoleums with remarkably good

THE ANCIENT KINGDOM OF PA 103

sculpturing ; the peculiar, dignified style of headstones and
mural monuments generally. The number of wayside shrines
and deities all in excellent repair.

4. The independent bearing and buxom appearance of the
women, and their evident influence in general market business.
Throughout the whole region natural, unbound feet are the rule.

5. The region is far from being thickly populated, and
cannot be termed wealthy, but apparently it is largely self-
contained and self-sufficient.

6. The intense curiosity of the people due to the fact that
few had ever seen a foreigner before.

CHAPTER IX
THE CHENGTU PLAIN

“THE GARDEN OF WESTERN CHINA”

HE plain of Chengtu is the only large expanse of level
ground in the great province of Szechuan; it is also

one of the richest, most fertile, and thickly populated
areas in the whole of China. Its extreme length from Chiang-
kou in the south to Hsao-shui Ho beyond Mienchu Hsien in
the north, is about 80 miles as the crow flies; its extreme
width from Chao-chia-tu in the east to Kuan Hsien in the
west, about 65 miles, in a straight line. From Kiung Chou in
the extreme south-west to its north-east limits beyond Teyang
Hsien is about 80 miles. The circumferential boundaries are
very irregular, the total area being under 3500 square miles.
Chengtu Fu, the provincial capital, and seventeen other walled
cities, are situated on this plain, together with very many un-
walled towns of large size. Farmhouses dot the plain in
every direction; the total population probably exceeds
6,000,000.

This plain is really part basin part sloping alluvial delta,
having an elevation ranging from about 1500 feet above sea-
level in the south and east to 2300 feet in the north-west and
west. It is bounded to the west and north-west by the steep
descent of a high mountainous region, which at very little
distance from it reaches above the snowline. In the extreme
north-west the snowclad Chiuting shan actually overlooks the
plain. On its other boundaries the sandstone hills of the
Red Basin rise sharply in bluffs 1000 to 1500 feet above the
level of the plain. The high barrier ranges protect the plain
from the cold northerly and westerly winds, but to these
must be ascribed the rapid changes in temperature, the fogs,

104

THE CHENGTU PLAIN 105

raw atmosphere, and the overcast skies so characteristic of
Chengtu Fu.

The plain owes its abundant fertility to a complete and
marvellous system of irrigation, inaugurated some 2100 years
ago by a Chinese official named Li-ping and his son. The
headquarters of this irrigation system is Kuan Hsien, a city
situated on the extreme western edge of the plain, where the
Min River debouches from the mountains. The principle on
which the system is based is simple in conception, but very
intricate in detail. An obstructing hill called Li-tiu shan
was first cut through for the purpose of leading the waters
through and distributing them over the plain. The passage
having been excavated, the waters of the Min River were
divided, by means of an inverted V-shaped dyke, a little dis-
tance above the canal into two main streams, the ‘“ South ”’
and ‘‘ North”’ Rivers, as they are called. The waters of the
“North” stream are carried through the Li-tiu shan cut, and
after passing through the city of Kuan Hsien are divided into
three principal streams. The most southerly of the three,
called ‘“‘ The Walking Horse,” flows directly east, and irrigates
the districts of Pi Hsien and Chengtu. The central stream,
called the ‘‘ Cedar Stem River,’ flows north-east, and is
utilized to irrigate the western and northern parts of the
above-named districts. Branches of these two streams flow
past the south and north walls of Chengtu, uniting near the
east gate of the city. The third, or northern branch, known
as the ‘‘ South Rush River,” flows north towards the city of
Péng Hsien, and then south-eastwards past Han Chou. All
the subdivisions of this branch and its anastomosing canals
and ditches unite near Chao-chia-tu to form the head-waters of
the To River, which flows due south past the famous salt-
wells of Tzu-liu-ching, and finally enters the Yangtsze at
Lu Chou. This “South Rush River” is fed by numerous
torrents which descend from the ranges bounding the north-
west edge of the plain. These streams—broad, stony, irre-
sponsible things with no defined banks—exist only during
rains or the melting of the snow in spring. In crossing the
northern parts of the plain the traveller can form some estimate
of what the whole was like before the irrigation canals were

106 A NATURALIST IN WESTERN CHINA

dug and dikes erected. But to return to the system at Kuan
Hsien. The ‘“‘South”’ River, which occupies the original bed
of the Min River, is divided into four principal streams almost
immediately opposite the Li-tiu Hill. The most easterly
branch, named the “ Peaceful River,’ irrigates the districts
of Kuan Hsien, Pi Hsien, and Shuangliu Hsien. The next
branch, called the ‘“‘ Sheep Horse River,’’ irrigates other parts
of the above-named districts, uniting with the “ Peaceful
River,’ at Hsinhsin Hsien. The third stream, called “ Black
Stone River,” irrigates the department of Chungching Chou,
and unites with the other streams at Hsinhsin Hsien. The
fourth stream, called “‘ Sand Ditch,” flows south-west through
Tayi Hsien and Kiung Chou, joining the other streams at
Hsinhsin Hsien. All the streams which intersect the Chengtu
Plain, save those forming the upper waters of the To River,
unite at Chiangkou, a village at the extreme south-eastern edge
of the plain, some 45 English miles south of Chengtu city.

This system of anastomosing canals, ditches, artificial and
natural streams, forms a complex yet perfect network. The
current in all is steady and swift, the bunding secure, and
floods unknown. Not only are all these streams and canals
available for irrigation, but they are also utilized to generate
power required in various industries. Flour-mills abound,
driven by vertical or horizontally fixed water-wheels. Similar
mills are used for crushing Chinese rape-seed, preparatory to
pressing for the extraction of the oil.

It must not be supposed that Li-ping and his son completed
the system which obtains to-day. They were the originators,
and the lines they laid down have been followed and enlarged
upon by succeeding generations. These famous irrigation
works are perhaps the only public works in all China that are
kept in constant and thorough repair. Every year the bunding
is repaired and all silt removed from the bed of the channels.
An official styled Shui-li Fu—‘ Prefect of Water-Ways ’—
residing at Chengtu, has charge of the system. In late winter
the water is diverted at Kuan Hsien from the “ North”’ River
to admit of the removing of silt, etc. In the early spring,
conducted with much pomp, there is an annual ceremony of
turning on the waters. The motto of Li-ping, “‘ Shen tao t’an,

THE DIVIDED WATERS AND BRIDGE (AN-LAN CHIAO) 250 YDS. LONG

_.
iy
Se

CHANNEL CUT THROUGH THE LI-TIU SHAN BY LI-PING

THE CHENGTU PLAIN 107

ti tso yen’ (Dig the bed deep, keep the banks low), has become
an established law in these parts, and is rigorously carried inta
effect. Amidst so much that is decaying and corrupt in China
it is refreshing to find an old institution maintaining its
standard of excellence and usefulness through century after
century. The originators of this work have been deified, and two
magnificent temples overlooking their work at Kuan Hsien bear
witness to the gratitude of the millions who have enjoyed, and
continue to enjoy, prosperity from the labours of the famous
Li-ping and his son. The “‘ hero-worship’”’ here exemplified
would do credit to the people of any land.

The larger of the two temples merits some description.
It is by far the finest example I have seen in my travels, and is
probably not excelled by any temple in all China. It nestles
midst a grove of fine trees, facing the river on the side of a hill,
with broad flights of steps leading from terrace to terrace. The
buildings are of wood, finely carved and lacquered. The court-
yards of stone are broad and spacious, with ornaments in
bronze and iron of old and unique workmanship. There are
figures representing Li-ping, his wife and son, also many finely
gilded and inscribed votive boards, gifts of a long line of
succeeding emperors, viceroys, gentry, and guilds. Nota
weed is allowed to grow, the whole place being kept scrupulously
clean by the Taouist priests in charge. In the courtyards are
many interesting trees and shrubs, trained in Chinese manner
with consummate skill. Two magnificent specimens of the
Crépe Myrtle (Lagerstremia indica), trained into the shape of
a fan some 25 feet high by 12 feet wide, and said to be over
200 years old, are finer than anything of the kind I have
seen elsewhere.

The whole of the plain is subdivided into small fields, every
field or series of fields having its own level, differing (sometimes
only by one or two inches) from that of its neighbours. This
arrangement necessitates a complicated code of regulations,
which, sanctioned by custom and usage, determines the pro-
portions in which the water of any one canal is distributed into
its branches, and the order of succession in which proprietors
of different fields are allowed to make use of it. The system
has been so far perfected that each rice field receives, exactly

108 A NATURALIST IN WESTERN CHINA

at the right time, a sufficient supply of running water. So
complete is the whole arrangement that scarcity, much less
famine, is practically unknown on the Chengtu Plain.

There are no extremes of climate in this region. In summer
the temperature seldom reaches 100° F., in the shade ; in winter
it seldom falls below 35° F. It is humid at all times and
essentially cloudy, more especially in winter, when the sun is
rarely seen, owing to banks of mists. The land is always under
cultivation, yielding two main crops that ripen in April or
May, and August or September respectively. Catch crops are
obtained between these two main harvests. Rice is the chief
summer crop, but certain districts produce millet, sugar, pulse,
Indigo (Stvobilanthes flaccidtfolius), and tobacco in quantity,
Pi Hsien being noted in particular for the latter crop. Wheat
and Chinese rape are the chief winter crops with Broadbeans
(Victa Faba), peas, barley, and Hemp (Cannabis sativa),
common in certain districts. Wén-chiang Hsien is famous for
its hemp, which is grown in quantity as a winter crop and
exported largely to other parts of Szechuan and down river.
This product, known colloquially as ‘‘ Huo-ma,” has been
wrongly identified by many travellers. As summer crops,
Ramie or ‘‘ Hsien-ma ”’ (Behmeria nivea) and Abutilon hemp
or “ Tuen-ma”’ (Abutilon Avicenne) are both cultivated more
or less in quantity. The only Jute or ‘‘ Huang-ma’”’ (Corchorus
capsularis) I ever saw was in July IgI0, growing near Yao-
chia-tu. In the northern parts of the plain, Mienchu and
Teyang Hsiens, a little cotton is raised, but commercially the
crop is unimportant. Opium was never cultivated in quantity
on the plain.

All the Chinese vegetables and culinary oil-producing plants
are cultivated in quantity in the Chengtu Plain, and their
general excellence is not excelled elsewhere. To enumerate
them it would be necessary to give a complete list of such plants
cultivated in all but the coldest parts of China. This enumera-
tion is reserved for a subsequent chapter.

A striking feature of the plain is the enormous number of
large houses and farmsteads dotted here, there, and everywhere,
and shaded by groves of Bamboo, Nanmu, and Cypress. The
frequency of these houses, with their enveloping groves, gives a

THE CHENGTU PLAIN 10g

well-wooded appearance to the entire region, and the general
view is broken up in such a manner that from no point can
many miles of the plain be seen at one time.

The variety of trees is very great; fully fifty species could
easily be enumerated. Alongside the streams and ditches,
Alder, “‘ Ching-mu”’ (Alnus cremastogyne) abounds, and forms
one of the principal sources of fuel. In the more northern parts
of the plain the curious Camptotheca acuminata, with clean
trunk, grey bark, and globose heads of small white flowers,
displaces the Alder. Around the houses Bamboo, Oak, “‘Pride of
India,” Soap trees (Gleditsia), Cypress, and Nanmu are the com-
monest trees. The Nanmu is a special feature around temples.
Several species of the genus Machilus are called Nanmu, all
agreeing in being stately, tall, umbrageous evergreens. The
wood they yield is highly valued, and the trees are particularly
handsome. The Banyan tree, so abundant a little farther
south, is very rare here, and neither Pine nor Chinese Fir
(Cunninghamia) are common. Occasionally trees of the Red
Bean (Ormosia Hosiet) occur, always, however, in temple-yards
or shading wayside shrines. The great industry of Chengtu
Fu is sericulture, consequently Mulberry trees abound, and
Cudrania tricuspidata (Tsa shu), the leaves of which are also
used for feeding silkworms, is likewise fairly common.

In such a highly cultivated area the natural flora has, of
course, been destroyed. The few indigenous shrubs and herbs
that remain are relegated to the sides of streams and grave-
yards. In places the Chinese Pampas Grass (Miscanthus
sinensis and M. latifolius) is common; in autumn the fawn-
coloured plumes are most attractive. Occasionally thorny
shrubs like Barberry, Christ’s thorn, colloquially “‘ Teh-li-
pé kuo-tzu”’ (Paliurus ramosissimus), and “San-chia pi”’
(Acanthopanax aculeatum) are used as hedge plants. The
commonest fence, however, is made by bending down and
interlacing the bamboo-culms.

Since the plain is strewn with cities, villages, and farm-
steads, a network of roadways necessarily obtains. A main
artery extends north-north-east, through the plain and beyond
to Shensi province, and ultimately reaches far-distant Peking.
This road was commenced from the Shensi end by the great

IIO A NATURALIST IN WESTERN CHINA

Shih Hwang-ti (he who commenced building the Great Wall)
about 220 B.c. It extends from Chengtu in a south-westerly
direction to Kiung Chou, and thence to remote Lhassa. Other
highways connect the provincial capital with Chungking,
the great mart on the Yangtsze River to the south-east ; Kuan
Hsien in the west, and the Marches of the Mantzu beyond.
Roads of secondary importance link these highways with
other roads and connect the capital with all the principal
cities of the plain and regions beyond. Most of the roads were
originally paved with one or two slabs of stone laid lengthwise
down the middle, with bare earth on either side. The constant
wheel-barrow traffic, a feature of the entire region, has worn
deep grooves into these slabs. All too frequently the slabs
have disappeared altogether, leaving unpaved long stretches
of roadway. In dry weather these roads are dusty, but easy
to travel ; in wet weather they are from ankle to knee-deep
in sheer mud. Often they are practically impassable, and
travelling over them in ordinary rainy weather is an experience
beyond words to describe. They illustrate admirably the
contrariety of things which obtain in China generally. Here
in the wealthiest region of the west, if not of the whole of
China, the average road is of the meanest width, and in an
abominable state of repair. There is much talk of the need
of railways in China,—true, they are needed badly, but good
highways, roads, are an infinitely greater want. The highways
and byways on the Chengtu Plain are a disgrace to the entire
population of this fertile, wealthy region. ‘‘ What is every-
body’s business is nobody’s business ’’ is a saying that is as
applicable in China as in Western lands. The roads exist for
the good and welfare of all, but it is nobody’s real business to
protect them; they are, in consequence, neglected by all—
peasants, farmers, officials, and gentry alike.

Mean as these roadways are, they are spanned by hundreds
of large honorary portals and memorial arches, mostly con-
structed of red, or more rarely grey, sandstone, or occasionally
of wood. In the vicinity of the more wealthy cities (Han
Chou, for example) these portals and arches are extraordinarily
abundant. Many are masterpieces of Chinese architecture.
All are well built and covered with sculptures in relief, re-

VIEW IN THE MANCHU SECTION OF CHENGTU CITY

THE VILLAGE TEMPLE, KUNG-CHING CH’ANG

THE CHENGTU PLAIN III

presenting scenes of mythical or everyday life. The ends of
the ridge pole and the gable eaves are usually long drawn out
and revolutely upturned, adding additional lightness and
beauty to the whole. These long, exaggerated, upturned
eaves are a characteristic feature of the houses, temples, and
shrines met with all over this region.

The innumerable ditches, canals, and streams are all well
bridged. The bridges are kept in good repair, and reflect the
highest credit on the engineers who constructed them. They
are built of red or grey sandstone, more rarely of wood, as near
Han Chou. The stone bridges vary from one to a dozen or
more arches, sometimes hog-backed, but more usually the
“Roman Arch” is employed; others are of causeway or
trestle design, with or without balustrades, ranging from a
single slab laid across a narrow ditch to many such laid on
a series of piers built in the bed of the streams. Near Sintu
Hsien there is an example of one of these trestle or pier-
bridges 120 yards long. Outside the east gate of Chengtu is a
red-sandstone bridge of nine arches, which is generally regarded
as the bridge mentioned by Marco Polo. A similar bridge
exists near Yao-chia-tu, but this has some twenty arches. Im-
mediately outside Han Chou there is a covered wooden bridge,
120 yards long, 6 yards broad, resting on eight stone piers.
This bridge, known as the Chin-ying chiao (Bridge of the Golden
Goose), is the handsomest, most ornate wooden structure of
its kind I have met with in my travels.

In reference to the bunding of the streams and canals it
should be mentioned that cobble-stones enclosed within long
sausage-shaped, bamboo-latticed crates are universally em-
ployed for this purpose. This system is said to date back
to the later times of the Ming dynasty only. Previous to that
period the principal abutments and revetments were of iron,
fashioned into the shape of gigantic oxen, turtles, pillars, etc.
At the places where canals unite or divide, or where the water
cascades to a lower level, the earthworks are protected by walls
of stones firmly cemented together.

Another item, and one which astonishes every traveller,
is the enormous size of the blocks of stone used in the bridges,
more especially those erected on piers. I have no exact

112 A NATURALIST IN WESTERN CHINA

measurements by me, but these slabs would average at least
12 yards long by 20 inches square at the ends. Commonly
the blocks are of hard limestone, occasionally of conglomerate.
The slabs of sandstone when used are shorter. At Chao-
chia-tu sandstone slabs are used as fencing.

Any attempt to describe the cities on the Chengtu Plain
would necessitate more space than is at my disposal. They
differ, with the exception of the provincial capital, in no marked
particular from other cities of Szechuan. In size they vary
considerably, some of the large unwalled towns being com-
mercially more important than the walled cities. Most of
the cities and surrounding districts are noted for certain
things ; for example, Mienchu Hsien for its wheaten flour and
paper, P’i Hsien for tobacco, Wén-chiang Hsien for hemp,
P’éng Hsien for indigo, Shuangliu Hsien for straw-braid, and so
on. The majority of these cities are very ancient ; all contain
fine temples, as becomes such centres of wealth. Chengtu
(long. 104° 2’ E., lat. 30° 38’ N.) was described by Marco
Polo, who visited it during the thirteenth century, as a “‘ rich
and noble city.’ Modern travellers, and their name is well-
nigh legion, have all agreed with the great Venetian’s dictum.
In many respects Chengtu, with its population of 350,000
people, is probably the finest city in the whole of China. It
is built on a totally different plan from that of Peking, or
even Canton, so that comparisons are difficult. The present
city of Chengtu is comparatively modern, but occupies much
the same site as the capital of the aboriginal kingdom of Shu.
This kingdom was conquered by Shih Hwang-ti (the “ First
Emperor’) some time between 221-209 B.c., who nominally

added it to his dominions. The succeeding dynasty of Han.

(206 B.c. to A.D. 25) incorporated it as an integral part of
China. During the epoch of the Three Kingdoms the site (or
thereabouts) of the city was occupied as the capital of the
kingdom under Liu-pei. Succeeding dynasties have always
made it a most important seat of administration, and princes
of the imperial clan or viceroys have resided there. It is
still the seat of a Viceroy who governs the province of Szechuan
and nominally controls all Thibetan affairs.

Great Britain, France, and Germany have each established

ki»

THE CHENGTU PLAIN Eee

a Consulate-General there, but on the plea that the city is
not an “‘ open port,” the Chinese have successfully resisted
the purchase of land on which to erect suitable houses and
offices for the staffs representing these Powers. The result is
that these officers are housed in dilapidated Chinese quarters,
insanitary, dangerous to health, and unbecoming the dignity
of the Powers they represent. It is nothing short of a scandal
to thrust men into such abominable quarters. Chengtu Fu is
far removed from London, Paris, and Berlin, also from Peking,
but is it fitting to make backwoodsmen of these repre-
sentatives ? Missionaries of every denomination are firmly
entrenched at Chengtu, and can acquire all the property
their funds admit of either for residences, hospitals, schools,
or churches.

The city is surrounded by a magnificent wall, some 9 miles
in circumference, with eight bastions, pierced by four fine gates.
This wall is 66 feet broad at base, 35 feet high, and 40 feet broad
at top, along which runs a crenulated balustrade. It is faced
and paved with hard brick (the walls of all the other cities on
the plain are of sandstone), and is kept in thorough repair.
During Manchu times a Tartar garrison was stationed here,
a large area on the south-west side of the city being walled off
to form a Manchu city. Within the city walls are many fine
residences, private and official, temples, a large parade ground,
etc. The city is clean and orderly, with an efficient police. To
wander through the streets noting the varied industries carried
on is a liberal education in Chinese ways of doing things.
The wares on sale are of infinite variety, and are themselves —
indicative of the wealth which is everywhere apparent. The
shop-signs, lacquered and gilded, hang vertically downwards,
and proclaim in their large artistic characters the titles of the
shops and the wares on sale. The city is full of officials, both
in and out of office, who move about the streets in sedan-chairs
carried at a great speed. The chairs are peculiar in having
the long poles curved, with the body of the chair resting on
top of the curve. When carried, such a chair is well above
the heads of the crowd. The streets are always crowded
with pedestrians, chairs, and wheel-barrows. Different trades
occupy their own particular quarter. Certain streets are

VOL. 1.—8

114 A NATURALIST IN WESTERN CHINA

devoted to carpentry in all its branches, boot-shops, shops
devoted to hornware, skins and furs, embroideries, second-hand
clothes shops, silk goods, foreign goods, and so forth. Silk-
weaving is the great industry in Chengtu, hundreds of looms
being in use.

Evidences of Occidental influence abound. A provincial
university and many schools for imparting Western learning
exist. Two agricultural experimental farms, an arsenal, mint,
bazaar, and many buildings of semi-foreign design. The arsenal
and farms are outside the city. An electric lighting plant was
operating at the time of my last visit (1910), and the installation
of a telephone service was in progress. The Imperial Post
is strongly established here under control of Europeans, and
this is the only Western innovation really accomplishing good
work. The others (and I have not covered them all) are
experiments. pure and simple. These are controlled by
officials among whom jealousy is rife and peculation not un-
known. The good intentions of honest officials are easily
nullified by jealous-minded sycophants and ultra-conserva-
tives. The city exhibits numerous examples of blighted
experiments, some of them mere follies, but the majority
calculated to be beneficial if properly controlled and carried
through. The city-fathers and officials have exhibited mad
haste to acquire such Western knowledge as they deem useful.
They have no real idea of what they want, and there is little
co-ordination in any matter. The students rule the colleges ;
their fathers, the gentry, rule the province. “‘ China for the
Chinese,” and ‘‘away with all foreigners and foreign influence”
is their slogan. This cry is perfectly legitimate, but they should
move slowly. They think they are fully fledged men, whereas
they are mere babes in the knowledge of the things they covet
so much. The unfortunate Rebellion which has spread with
such rapidity and brought about so much disaster to the nation, _
originated with the hot-heads of Chengtu. Primarily it was
aimed not so muchagainst the dynasty as against foreign capital.
The Central Government had agreed to a foreign loan, which,
amongst other things, had for its object the construction of a
railway from Hankow to Chungking. It was this loan that was
the fat in the fire which produced the conflagration—the last

A MEMORIAL ARCH

THE CHENGTU PLAIN II5

straw, if you will, but the primary cause of the Rebellion. The
dynasty has been dethroned (it was effete, anyway, and
should have passed fifty years ago), a dictatorship under the
guise of a republic cleverly formed by the only man who can
save China from anarchy if not disruption—Yuan-shih-kai.
But foreign loans have become more absolutely necessary than
ever before. The present system of government can only be
transient, another dynasty must arise. I mentioned above
that the province was under a viceroy, and that the gentry
ruled the province. This is the keynote to the whole difficulty.
The Viceroy had to carry out the instructions of the Imperial
Government at Peking ; he had also to please the gentry. The
wishes of the two powers became diametrically opposed, and
not all the tact of the cleverest diplomats could save the situa-
tion. The Viceroy (Chao Erh-hsiin) was removed to Manchuria,
and his brother (Chao Erh-féng), recalled from the Thibetan
Marches (where China’s new toy, in the shape of an army
modelled on quasi-Western lines, had been indulging in an
altogether uncalled-for war of aggression), appointed to the
post. The new Viceroy arrived too late to check the revolt,
and was ultimately murdered. The gentry have declared
that no foreign capital, and the necessary foreign supervision
of such capital, shall enter into the construction of a railway
in Szechuan. With Chinese money and Chinese engineers the
scheme shall be accomplished, say these autocrats. The Cen-
tral Government thought otherwise and made other arrange-
ments. Then came the revolt, fulminated by the gentry of
the Chengtu Plain, which speedily got beyond their control,
and where it will really end is beyond prophecy. The Manchu
dynasty, when it ascended the Dragon throne in A.D. 1644,
immediately set to work and rescued Szechuan from the bloody
grip of the rebel and arch-destroyer, Chang Hien-tsung, and
brought peace to the land. Two hundred and sixty-seven
years later this dynasty has been dethroned by rebellion
initiated by the gentry of the Chengtu Plain. Dynasties and
republics may come and go, but in the future, as in the past,
industry, combined with agricultural skill, will continue to win
sustenance, derive wealth, influence, and power from this
fertile and beautiful region—the Garden of Western China.

CHAPTER X
NORTH-WESTERN SZECHUAN

NARRATIVE OF A Cross-MOUNTAIN JOURNEY TO
SUNGPAN TING

‘ FEW days after our arrival at Chengtu in rgro I deter-

mined upon a journey to the border-town of Sungpan

Ting, for the express purpose of securing seeds and
herbarium specimens of certain new coniferous trees previously
discovered by me in that region. During 1903 and again in
1904 I had visited this interesting town. On the first occasion
I travelled by the ordinary main road, via Kuan Hsien and the
Min Valley. The next year I followed the great north road
across the Plain of Chengtu to Mien Chou, then travelled via
Chungpa and Lungan Fu, by another recognized highway.
On these journeys I gleaned tidings of a by-road leading from
Shihch’uan Hsien across the mountains, finally connecting
with both the above routes. This route promised to be
interesting as well as novel. Only Roman Catholic missionaries
had previously traversed it, so far as I could learn. An Hsien
was selected as the real starting-point for this trip.

With this object in view we passed through the north gate
of Chengtu city early on the morning of 8th August. Follow-
ing the north road as far as the city of Han Chou, then branch-
ing off and travelling via Shihfang Hsien and Mienchu Hsien,
we reached the city of An Hsien, some 300 li from Chengtu,
after three and a half days. The road led us right through
the luxuriant Chengtu Plain to its extreme north-western limits
near Hsao-shui Ho. Afterwards we crossed some low foot-hills
to a small stream leading to An Hsien. The journey was very
easy, though fatiguing owing to the extreme heat of the season.

The city of An Hsien is small, of little importance, prettily

116

NORTH-WESTERN SZECHUAN 117

situated on the left bank of a stream backed by bare mountains
which rear themselves some 2000 feet above the level of the
river. Two streams uniting here form a river navigable during
high-water season to Mien Chou, a city on the Fou Ho—the
western branch of the Kialing River system. An Hsien is a
little beyond the north-western limits of the Chengtu Plain, and
its river gives it direct communication with Chungking, during
the summer at least.

Leaving by the north gate, we took a road that ascends
the main branch of the river which is kept from flooding the
city by a well-made low bund of stone slabs, firmly cemented
together. After traversing a small cultivated valley we plunged
into a rocky defile and crossed the river by an iron suspension
bridge, 110 yards long. This bridge is old and in poor repair,
and it swayed considerably as we walked singly across. A few
miles farther on we recrossed the stream by a similar bridge,
and reached Lei-ku-ping, our destination for the day, at
6 p.m. A certain amount of rice is cultivated hereabouts,
but maize is the staple crop. As an under-crop to maize,
Amorphophallus konjac (‘‘ Mo-yu’’) is commonly cultivated,
the tubers being used as food after their acrid properties have
been removed by washing in water. We met considerable
traffic, mostly coolies laden with sheep-skins and medicines
from Sungpan, which they put on boats at An Hsien for con-
veyance to Chungking; much potash (lye) in small tubs,
and oil-cakes consisting of the residue of Chinese rape-seed
after the oil is expressed. Coal of very poor quality, mostly
dust, is obtained in the surrounding mountains, and we met
scores of mules, ponies, and coolies engaged in transporting it.

Lei-ku-ping, alt. 2750 feet, is a large market village, pos-
sessing one principal street with gates at each end, which are
closed after sunset. The centre of a large and important
industry in tea, Lei-ku-ping largely supplies Sungpan Ting
and the country beyond. The tea is grown in the surrounding
districts and brought to the village for sale. Later we shall
have more to say concerning this industry.

It rained heavily during the early hours of the morning, and
though it was fair when we set out, showers fell the whole fore-
noon. On leaving Lei-ku-ping we ascended a few hundred

118 A NATURALIST IN WESTERN CHINA

feet to the head of a low divide, and then descended to the
village of Che-shan, situated on the right bank of a considerable
stream. This village shares in the tea industry for the Sung-
pan market, but is of less importance than Lei-ku-ping.

From Che-shan to Shihch’uan Hsien the road ascends the
right bank of the river, which flows between steep precipitous
mountains. The path is usually several hundred feet above
the stream, broad and fairly easy for the most part, but con-
stantly ascending and descending. The mountain-sides are
steep but, where not absolutely vertical, are all under cultiva-
tion, Maize being the staple crop. There is very little lime-
stone, the rocks being chiefly loose sandstone and mud shales.
These shales weather rapidly, and the steepest cultivated
slopes are usually composed of these rocks.

The river is broad, and could easily be made navigable
for boats during the high-water season. Even in its present
condition rafts could be floated down, but we saw no traffic
whatsoever on its waters. The water was dirty, and much
driftwood was strewn along the shores. This is collected,
dried, and stacked, forming apparently the principal source of
fuel. Trees are very scarce, but around houses occur Sophora,
Pistacia, Pteroceltis, Sterculia platanifolia (Wu-tung), Kelreu-
teria bipinnata, and Alder. The Keelreuteria was just coming
into flower; the flowers are golden yellow produced in large,
much-branched, erect panicles ; the leaves are very large and
much divided. Shrubs are not plentiful but, much to my
surprise, the Tea Rose (Rosa indica) is quite common, and evi-
dently spontaneous, by the wayside, on the cliffs, and by the
side of the stream.

A few li below the city of Shihch’uan Hsien the river is
spanned by a bamboo suspension bridge, about 80 yards
long, supported on cables made of split, bamboo culms plaited
together. These cables, eight in number, are nearly 1 foot in
diameter, and are fastened to stanchions fixed on either side of
the river. Two similar cables on either side of the bridge are
carried across at higher levels, and have attachments of
bamboo rope supporting those which form the base of the
structure. A capstan arrangement is used for making the
cables taut, and the lower ones are covered with stout wicker-

NORTH-WESTERN SZECHUAN 11g

work to forma footway. Like all such structures, this bridge
is heavy, sags very much in the middle, and is very unsteady
to walk across. The life of these bridges is only a few years,
and strong winds often make them very unsafe.

Shihch’uan Hsien is a small city charmingly situated at an
altitude of 2800 feet, on the left bank immediately below the
junction of two rivers. It is surrounded on all sides by steep,
more or less cultivated mountains. Inside the city are many
trees, which add considerably to the effect. A pavilion anda
small pagoda crown two prominent hills, and assure the “‘ luck ”’
of the place. A narrow suburb runs ribbon-like between the
river and the city wall. This wall is broken down in places,
and the gates are low and small. We found accommodation in
a large, curiously constructed inn remarkable for the strength
of its stinks and the abundance of vermin and mosquitoes it
sheltered. The day’s journey was given as 65 li, but the li were
long, consequently the coolies with their loads arrived late.
Cash was needed, but on opening a box to obtain some silver
for exchange we found that some one had stolen from it about
30 taels and 5 dollars. The load belonged to a coolie we had
engaged at Taning Hsien, and retained because he had given
unusual satisfaction! The previous day he had engaged a
local coolie to carry his load, on the ground that he was feeling
sick. He was last seen near Che-shan, still unable to carry
his load. Evidently he was the culprit, but he was thoughtful
enough to leave us about half the amount contained in the box.
Since he had about three-quarters of a day’s start I concluded
it was best to quietly cut the loss, my first and last in China.
The delays incident upon lodging a complaint with the official
would have involved me in further expense and trouble, with
but small chance of recovering the money lost.

The main road to Sungpan continues to ascend the right
bank of the river to its source, then crosses over a range and
enters the upper Min Valley at Mao Chou. I had been over
most of this route in 1908 when crossing the Chiuting range
from near Mienchu Hsien to Tu-mén, thence to Mao Chou. The
route we had in view leads to the north-west from Shihch’uan
Hsien. From Chengtu to this point we had travelled without
escort, but with the difficulties of an unknown route before us

120 A NATURALIST IN WESTERN CHINA

I thought it best to secure such at this city. Sending my card
to the Hsien’s yamén, in the ordinary way, I informed this
official of my project, and asked for the customary escort. Half
an hour afterwards my card was returned with the information
that there was trouble at Sungpan and no escort would be
supplied! The refusal was as curt as it was insolent, but
whether the Hsien was actually responsible I never found out.
In the whole of my eleven years’ travel in China this was the
first and last experience of official discourtesy. Thus two
annoying experiences, both unique in their way, yet, happily,
trivial and unimportant, marked my visit to Shihch’uan Hsien,
a town which, from the commencement of my travels in the
western Szechuan, I always had a keen desire to visit.

The next day we left Shihch’uan Hsien at sunrise, glad to
escape from the malodorous, vermin-infested inn. No one put
in an appearance from the yamén, and no attempt to prevent
our taking the route proposed was made. I had rather feared
this might happen, but my fears were fortunately groundless.
On leaving the city by the north gate we struck a stream nearly
equal in volume to the main river. The road ascends the left
bank, and almost immediately plunges into a narrow, wild
ravine, through which we continued the whole day. Like all
such roads it skirts the mountain-side, being usually several
hundred feet above the river, but is constantly descending to
the water’s edge, only to ascend again a few hundred yards
farther on. It is in good repair, although the rocks are of
soft mud shales, and signs of landslips were frequent. Wherever
possible maize is cultivated, but houses are few and far between.
The country strongly reminded me of that around Wénch’uan
Hsien in the upper Min Valley farther west. Trees are very
scarce, the Wu-tung (Sterculia) being perhaps the most common.
The shrubs denote a dry (xerophytic) climate, nearly all having
small leaves, either thick or covered with a felt of hairs. Of
these shrubs, Abelia parvifolia, Lonicera pileata, Ligustrum
strongylophyllum, and various kinds of Spirea, are common.
Bushes of the wild Tea Rose are not infrequent. Five li before
reaching Kai-ping-tsen, our destination for the day, we crossed
a clear-water tributary by a remarkably well-built stone-arch
bridge. During the day we passed several ‘‘ rope’’ bridges,

NORTH-WESTERN SZECHUAN 121

made of a single thick cable of plaited bamboo culms—sure
signs of difficult borderland country. Near Shihch’uan Hsien
we passed a bamboo suspension bridge, similar to the one
already described; at Kai-ping-tsen there is another such
bridge. There was a fair amount of traffic on the road. Potash
salts (lye), shingles, and oil-cake were the principal loads
encountered, all being carried on men’s backs, the first-named
being the most common.

Kai-ping-tsen, alt. 3200 feet, is a small village of about
fifty houses, situated on the left bank of a stream some 50 li
north of Shihch’uan Hsien. A new, empty house afforded us
comfortable lodgings ; the people were courteous, and made
our brief stay with them very pleasant. A remarkably fine
headstone, recently erected over the tomb of a much-respected
widow, was the chief thing of interest in the village.

On leaving Kai-ping-tsen we continued to ascend the left
bank of the stream through country similar to that of the
previous day, for 30 li to the market village of Hsao-pa-ti.
This village, all things considered, is of considerable size (about
one hundred houses), with many farmhouses scattered around.
The mountains are less rugged and steep, and are given over to
the cultivation of maize. The houses are low, built of mud
shales and roofed with slabs of slate. Market was in progress ;
food-stuffs, fuel, and potash salts being the principal goods on
sale. A bamboo suspension bridge spans the river and a road
leads across country, ultimately joining with the main road
between Shihch’uan Hsien and Mao Chou. On leaving Hsao-
pa-ti the road deserts the river and ascends through maize
fields over a rather low ridge. It then descends to a small
tributary, after crossing which a steep climb of 1000 feet leads
to the summit of another ridge. From this point we sighted
the main stream again, flowing through a smiling valley, at the
head of which nestles the village of Pien-kou, which was our
destination for the day. This village proved a good 20 li from
the ridge, though it looked close at hand. The road led through
fields of maize to the valley, and finally across the river by an
old, very shaky bamboo suspension bridge, which swayed
tremendously and was really unsafe.

Pien-kou (Yiian-kou of the maps), alt. 3800 feet, is a market

122 A NATURALIST IN WESTERN CHINA

village of some importance, but a fire had recently destroyed
half the houses. We had some difficulty in obtaining lodgings,
the only decent place being full and the occupants unwilling
to move. After a little time persuasive insistence won,
and we settled down comfortably, if crowded. One of the
occupants was down with fever. I dosed him with quinine,
and supplied him with enough to last several days, much to
his appreciation. This act got noised abroad, with the result
that applications for medicine quickly became too numerous.
Quinine is a drug much appreciated by Chinese, being about
the only foreign medicine they have real confidence in.

The day’s journey was said to be 70 li. It was long and
uninteresting. The flora is miserably poor; Alder being the
only really common tree.

The road we were following ultimately joined the Mao Chou-
Sungpan main road near Chén-ping kuan, about 160 li below
the town of Sungpan. We could get no tidings of a road
crossing to the Lungan-Sungpan highway, but all the same we
felt sure of finding one. Thus far the route indicated on my
map was all wrong, and we were left very much in the dark
as to our actual whereabouts. However, I was long since
accustomed to this state of affairs.

Leaving Pien-kou, a journey of 40 li brought us to Peh-
yang ch’ang, a village of a dozen scattered, dilapidated houses.
The road was distinctly bad in places owing to landslips. The
rocks are mainly mud shales standing on edge. We followed
the right bank of the river we had pursued from Shihch’uan
Hsien for the first 22 li, then crossed over to the left bank
by means of a shaky improvised bridge of two tree logs, the
bamboo suspension bridge which formerly crossed the stream
hereabouts having broken down. At this cross-over point
resides a Chinese official, locally styled a Tu-ssu. This official
was most courteous, helping us with advice and guidance to
cross the stream.

The journey generally was a repetition of the two former
days, through a rocky but uninteresting gorge. Wherever
possible, maize is cultivated, and we noted two odd patches
of rice. Houses are few and far between, and we met only
a few coolies laden with potash salts, charcoal, and shingles.

VIEW FROM HOSTEL, LAO-TANG-FANG

NORTH-WESTERN SZECHUAN 123

The flora was not interesting, Alder, Pterocarya, and Cornus
controversa are the only common trees. Buddleia Davidi
is abundant by the stream side, and was in full flower. The
Tea Rose alsoisfairlycommon. A Lily without bulbils, other-
wise very like Liliwm Sargentia, is plentiful in places. At
Peh-yang ch’ang, alt. 4100 feet, we found a road leading off
to the right, and connecting with the Lungan-Sungpan high-
way at Shui-ching-pu ; this we decided to follow.

Above our lodgings at Peh-yang ch’ang the river bifurcates,
one branch, a clear-water stream, being locally adjudged the
larger. It is up this stream the road connecting with the
Mao Chou-Sungpan highway ascends. The people told us
that this road was similar in character to the one we had
followed thus far, but more difficult, especially since the proper
bridges had nearly all been recently destroyed by floods.
The cross-over to the Min Valley is near a place called Hwa-tsze-
ling, where fine forests of Silver Fir and Spruce occur. Pien-kou
is a considerable wine market, much of the product finding
its way to Sungpan over this rough cross-country road.

A fatiguing march marked our first day’s journey towards
the Lungan-Sungpan highway. We made two long ascents
and descents, and commenced a third ascent, putting up for the
night at Hsao-kou, after covering 55 li. The second ascent
was fully 2000 feet, and very steep, through maize fields,
culminating in abandoned herb-clad areas. The descent was
mainly through coppice and brush. Houses occur scattered
here and there, wherever cultivation is practicable, maize
being the staple crop; the Irish potato and peas are also
grown. The road proved difficult, but I had traversed worse.

The forests have been destroyed, brushwood now covering
the uncultivated areas. Topping the loftier crags, and in
inaccessible places generally, a sprinkling of conifer trees still
exist, but we did not get near them. The vegetation generally
is that common to the 5000 to 6000 feet belt in west Szechuan,
but is less varied than in many parts I have visited. In the
valleys Alder was common, and on the slopes the Varnish tree
(Rhus verniciflua) and Walnut (Juglans regia) occur in quantity.
In coppices the Davidia, both the hairy and glabrous varieties,
is plentiful, but no large trees were noted. Throughout the

124 A NATURALIST IN WESTERN CHINA

bottom-lands and abandoned cultivated areas ‘‘ Summer
Lilac” (Buddleta Davidit) was a wonderful sight—thousands of
bushes, each one with masses of violet-purple flowers, delighting
the eye on all sides, the variety magnifica, with its reflexed
petals and intense coloured flowers, being most in evidence.
I gathered also an albino form, one small solitary bush, the
only one I have ever met with. Forming a much-branched
bush 4 to 8 feet tall, with rose-purple flowers, Hydrangea
villosa was, next to the Buddleia, the most strikingly orna-
mental shrub. On moist rocky slopes plants of Rodgersia
esculifolia occur in millions. It was in the fruiting stage,
but when in flower the acres of snow-white panicles must have
presented a bewitching sight. Nowhere else have I seen this
plant so abundant or luxuriant. The slender arching plumes
of white flowers, produced by Spive@a Aruncus, covered acres of
ground ; an apetalous Astilbe (A. rivularis) was also abundant,
and worthy of note.

The hamlet of Hsao-kou, alt. 5900 feet, consists of three
scattered houses, surrounded by maize plats, with remains
of other ruined houses near by. It is encompassed on all
sides by steep mountains, some of them culminating in lofty
limestone crags and rugged razor-like ridges with pinnacled
peaks—all of them inaccessible. At the back of the inn are a
few Larch trees, and near by several large trees of a flat-leaved
Spruce. The Hou-p’o (Magnolia officinalis) is cultivated
hereabouts, and also around all of the houses we passed dur-
ing the day. The innkeeper likewise cultivates a medicinal
Aconite (Aconitum Wilsonit), which is valued as a drug in
Chinese pharmacy.

We encountered only three men carrying goods during the
whole day ; two were laden with potash salts, the third with
the bark of a Linden (T7lia), used locally for making sandles.
Evidences of forest fires were all too frequent during the
day’s march.

The next day rain ruined what otherwise would have been
a more than ordinarily interesting march. From 7 a.m.
until 2 p.m. we struggled up some 4000 odd feet to the summit
of the pass leading across the Tu-ti-liang shan ; then descended
another 4000 feet to the hamlet of Hsueh-po, where we secured

NORTH-WESTERN SZECHUAN 125

lodgings in a large and good house. Rain commenced shortly
before Ir a.m., and continued the rest of the day. Our
perspective was limited to a few hundred feet ; now and again
a strong gust of wind would scatter the mists, admitting
momentary glimpses of cliffs and inaccessible peaks clothed
with jungle and with occasional Conifer trees, but such
views were rare.

The hamlet of Hsao-kou is very scattered, and we passed
two or three more houses soon after leaving our lodgings.
But after about 3 li houses and cultivation vanished, as did
also the Buddleia and Hydrangea previously so abundant.
The ascent, at first gradual, soon becomes precipitous, through
a jungle growth of shrubs and coarse herbs. The latter with
the thin brushwood is cut periodically and burnt. The ashes
so obtained are placed in wooden vats fitted with sieve bottoms,
boiling water is poured over them, and the liquid drains into
tubs, where it is evaporated and salts of potash (lye) left asa
residue. This product is packed in flasks and carried to
market towns for sale. We passed several rude huts where
men were engaged in this occupation. The road ascends a
small torrent and is nowhere easy. By throwing logs across
the stream and boggy places, lumber-men have succeeded in
making some sort of a path. But crossing these wet, slippery
logs was difficult. At one such crossing I slipped, but by
jumping into the rock-strewn torrent somehow managed to
avoid a nasty accident. Near the summit, and for some
distance down the Lungan side of the pass, are split pieces
of wood, arranged to form a long flight of shallow steps that
assist the roadway materially. The descent after a few hundred
feet becomes gradual, leading through open, park-like slopes,
quite unlike anything I have encountered elsewhere in China.
Now largely denuded of trees these glades are covered with
grass, and horses, goats, and pigs are raised here in some
quantity.

Formerly this range of mountains must have been covered
with conifers, but the lumber-man’s hands have been heavily
laid on these forests. We passed none but small, decrepit
specimens of no value. Hemlock, Spruce, and Silver Fir are
all represented. The outstanding feature of the march was

126 A NATURALIST IN WESTERN CHINA

the abundance of Cercidiphyllum trees. Throughout the moist
slopes and park-like areas on both sides of the range this tree
is common. Stumps of decaying giants abound, one of these,
which I photographed, measured 55 feet in girth! This
specimen had been broken off some 30 feet above the ground,
and was a mere hollow shell, but still supported many twiggy,
leafy branches. These stumps are relics of the largest broad-
leaved trees I have seen anywhere in China. Growing inter-
spersed with these remains were many specimens of the same
tree, 60 to 80 feet tall, 8 to 10 feet in girth, perfect in
outline, with myriads of neat, nearly round, bright green
leaves. One of these was in young fruit, and for the first
time in my travels I secured specimens of the fruit of this
beautiful and interesting tree. (Later I collected ripe seeds,
and this tree is now growing in the Arnold Arboretum, where
it promises to be quite hardy. It proved to be a variety
distinct from the Japanese species.)

This tree (Cercidiphyllum japonicum, var. sinense) attains
to greater size than any other broad-leaved tree known from
the temperate zone of eastern Asia. In size it is only
approached by its close ally, Tetracentron, which is also
common in the woods on the Tu-ti-liang shan. A _ local
name for the Cercidiphyllum is ‘‘ Peh-k’o,” a name strictly
applied all over China to the Maidenhair tree (Ginkgo biloba).

The summit of the range is composed of mud shales, which
seem favourable to the growth of vegetation generally. Be-
tween 8000 feet altitude and the summit Rhododendron
calophytum is extraordinarily abundant, trees 40 to 50 feet tall
and 5 to 7 feet in girth, with handsome cinnamon-brown bark,
cover many acres. Euptelea pleiosperma and Pterocarya hupe-
hensis are other interesting trees plentiful hereabouts. The
bark of the last-mentioned tree is used locally for roofing
purposes. Willows in many species are common; the bark
of certain of these and also that of Linden trees is used by
the peasants for making sandals. Viburnum erubescens, var.
Prattii, with pendulous panicles of white fragrant flowers,
followed by fruit which is at first scarlet and then changes
to black, is perhaps the commonest shrub. Various Araliads,
Sorbus, etc., grow epiphytically on all the larger trees that

NORTH-WESTERN SZECHUAN 127

have a rough humus collecting bark. Maples in variety,
Micromeles laden with fruit, and many other interesting
trees were striking constituents of these woods. Tall growing
herbs made a grand display, especially the apetalous A stilbe
vivularis, Spirea Aruncus, Anemone vitifolia with white and
pink flowers like the Japanese Anemone, Artemisia lactiflora
with large panicles of milk-white, fragrant flowers, Balsams
(Impatiens) with yellow, pink, and purple flowers; mixed with
them also were Meadow Rue (Thalictrum), Aconites, many
Senecios, and Meconopsis chelidonifolia growing about 3 feet
tall with clear yellow flowers, saucer-shaped and 24 inchesacross.
Acres of the country-side are covered by these various herbs.

There was indeed plenty to interest one; the flora of this
region is undoubtedly rich, and it was most unfortunate that
the rain prevented an exhaustive investigation.

Hsueh-po, alt. 6000 feet, consists of a few houses surrounded
by high mountains with a good-sized torrent, which rises near the
head of the pass, and flows through the narrow valley. Maize
is cultivated as the staple crop. The Hydrangea and Buddleia
previously noted ascend to this altitude, and were a wealth of
blossoms. Alder also extends to this point ; Poplar likewise.
This latter tree has a very graceful port and the leaves have
red petioles and veins when young.

Our lodgings were good and weather-proof, which was for-
tunate, since it rained heavily the night through, and until
eleven o'clock in the forenoon of the next day. Afterwards it
was fair, but threatening, heavy clouds and mists obscured the
country-side from our view. Around the inn are several trees
of a handsome, flat-leaved Spruce (Picea ascendens) with
pendulous branchlets. This tree, known locally as ‘‘ Mé-tiao
sha or sung,’ is the most esteemed timber tree in these parts.
The trees are felled, hewn into planks about 25 feet long,
5 inches thick, and 12 inches broad, and carried on men’s
backs to a point on the river whence it is possible to float down
rafts. Lumbering is a very considerable industry in these
mountains, the timber finding its way to Chungpa. This fine
Spruce was fruiting freely. (Later I secured plenty of seed,
and successfully introduced it into Western gardens.)

On leaving Hsueh-po, we crossed the torrent and descended

128 A NATURALIST IN WESTERN CHINA

the left bank. At K’ung chiao the torrent is joined by another
of equal size, the united waters forming a fine clear-water
stream. From this point downwards rice is cultivated. The
stream continues to receive affluents, a very considerable one
joining it at Tu-tien-tsze. At Peh-mu chiao, ro li above Tu-
tien-tsze, the timber logged in the surrounding mountains is
made into rafts and floated down. Just below Shui-ching-pu
the stream unites with the main branch of the Lungan River
(the Fou Ho), and the rafts are floated down past the city of
Lungan to Chungpa, a large village of vast commercial im-
portance, in direct water communication with distant Chung-
king, it being within the Kialing River system.

Tu-tien-tsze is a small market village and a Roman Catholic
Mission centre. This Church has a strong following through-
out the region we had traversed from An Hsien. The country
folk everywhere in this part were most courteous and civil.
This, I think, is probably due to the influence wielded by the
self-sacrificing priests of the Roman Faith. But whatever the
cause, I shall always retain pleasant memories of the people
encountered everywhere in this little-known region.

The road proved easy all day, usually skirting the moun-
tains well above the stream. At Tu-tien-tsze a cross-country
road leads to Lungan Fu, some 130 li distant. Ten li below
Tu-tien-tsze we crossed to the left bank of the stream by a
covered bridge. Descending a few miles and crossing a promon-
tory we reached the main river (Fou Ho) opposite Shui-ching-
pu. Ferrying across to this village, we found lodgings in a large
house owned by a Shensi man of the Mohammedan persuasion.

Shui-ching-pu, alt. 4200 feet, is a market village of about 200
houses, situated on an alluvial flat, surrounded by mountains
largely under cultivation. A river of considerable size, which
brings down an unusual quantity of detritus, joins the main
stream on the left bank immediately below the village. A
road ascends this stream, leading to Wén Hsien in Kansu
province. It was said to be difficult, traversing a mountainous
region peopled with Sifan. Iron is a local product of some
importance hereabouts. Gold is also mined in the neighbour-
hood. The quartz, after it has been broken into small pieces,
is pounded into dust in mills like those commonly used for

NORTH-WESTERN SZECHUAN 129

hulling rice. The dust is washed and the gold separated by
means of quicksilver. Placer mining is carried out all along
this Lungan River by unemployed peasants, but the yield is
small. In 1904, when I first journeyed to Sungpan by way of
Mien Chou, Chungpa, and Lungan, the officials were endeavour-
ing to put a stop to placer mining. Placards were posted
forbidding the people to wash for gold, on the ground that
landslips were caused through the removal of the rocks, etc.,
on the foreshore.

From Hsueh-po to Shui-ching-pu is said to be 60 li. The
valley which we traversed is all under cultivation ; farmsteads
are general after Peh-mu chiao is reached. Alder, Walnut, and
Poplar are the common trees, with Pear, Plum, and Peach
trees around houses. Ina garden I saw one magnificent speci-
men of the Crépe Myrtle (Lagerstremia indica), 25 feet tall,
24 feet girth, just one luxuriant mass of carmine-red flowers.
Here and there the moist rocks are beautifully carpeted with
ferns, Woodwardia radicans, Blechnum eburneum, and Maiden-
hair being particularly rampant. The Buddleia and Hydran-
gea, previously mentioned, are abundantly present, and were a
wealth of pleasing flowers.

At Shui-ching-pu we joined the highway between Lungan
Fu and Sungpan. The intrepid Captain W. J. Gill, in June
1877, was the first Occidental to traverse this route. Since
that date several travellers and missionaries have been over
this road, but the total is small.

My first journey over this highway was, as mentioned
above, in 1904. At that time I had no camera, and the
recollection of the wonderful scenery had much to do with
my second journey to these parts in 1910. I saw the country
through the eyes of a botanist, and for this reason I hope a
continuance of this narrative will prove justifiable.

Leaving Shui-ching-pu about 7 a.m., we saunteringly
covered the 50 li to Hsao-ho-ying by 4 p.m. The road ascends
the left bank of the stream for some 20 odd li to a point just
above the small village of Yeh-tang. At this place the river is
joined by another of nearly equal size on its right bank. A
by-road ascends this tributary and leads across the mountains

1 River of Golden Sand.
VOL. I.—9

130 A NATURALIST IN WESTERN CHINA

through country sparsely peopled with Sifans, and connects with
the Mao Chou—Sungpan highway a few miles below Sungpan
Ting. The road we followed crosses the left affluent of the
Fou River by means of an iron suspension bridge 24 yards long,
erected immediately above the union of the two streams. A
few li beyond this place the road plunges into a wild gorge.
The scenery is wonderful. Limestone cliffs clad with vege-
tation rear themselves 1000 to 2000 feet above the torrent
which hereabouts rushes headlong over huge rocks. Wherever
possible, maize is cultivated on the slopes and rice in the bottom-
lands. We crossed to the right bank by a covered wooden
bridge just below a place where landslips have produced a
series of cataracts. About 3 li below Hsao-ho-ying the gorge
suddenly opens out, leaving room for a small circular valley, in
the middle of which the walled village above named is situated.
Viewed from this point where there is an old gateway, the
village presents a charming picture of peace and plenty
locked in by precipitous mountains. On entering the village,
however, one is quickly disillusioned. Abject poverty is only
too apparent. The one main street is broad, flanked by more
or less ruined houses, with much of the land within the walls
given over to maize plats. The people are in keeping with
their dilapidated surroundings.

Hsao-ho-ying, alt. 5300 feet, signifies ““ Camp on the Small
River.” It is an ancient garrison village. Eighty years ago
some 700 soldiers were quartered here. This number was
speedily reduced as the surrounding country was conquered.
To-day the garrison is put down at 40 men, but it is doubtful if
even this number remains. Three yaméns belonging to military
officials of low rank are the only respectable buildings in the
place.

At Shui-ching-pu we were assured we could exchange
silver at Hsao-ho-ying. This proved a fable and landed us
in an awkward dilemma. However, ‘“ Mo-li-to”’ (Fata viam
invenient), as the locals have it !

The flora of the day’s journey was not particularly rich,
though we passed many plants of interest. Around Hsao-
ho-ying, the Walnut (Juglans regia), Varnish (Rhus verniciflua),
Poplar, Apple, Pear, Plum, Peach, and Tu-chung (Eucommia

NORTH-WESTERN SZECHUAN 131

ulmoides) are commonly cultivated. By the side of the torrent
the Buddleia was again a wonderful sight. In a temple yard
near Yeh-tang is a magnificent tree of Meliosma Beamana,
about 60 feet tall and 12 feet in girth, the head being fully 80
feet in diameter. The pinnate leaves produce abundant shade.
This tree was covered with small pea-like purple fruits which
later afforded me a supply of ripe seeds. (The pinnate-leaved
members of this small family are all handsome trees, and none
was in cultivation previous to my explorations. I have
succeeded in introducing three species, all of them promising
to thrive under cultivation. One, M. Veittchiorum, is now
flourishing just within the main entrance to Kew Gardens.)
From Hsao-ho-ying to Shuh-chia-pu, 30 li, the road ascends
a narrow valley which is without special interest, the bottom
lands and lower slopes being cultivated with maize and buck-
wheat. Houses occur at intervals. Just above Shuh-chia-pu,
a poverty-stricken hamlet of about a score of houses, the
river bifurcates. The road ascends the left and larger branch,
plunging immediately into a narrow gorge. The track, all
things considered, is good, though there is room forimprovement.
The scenery in this gorge, for magnificent, savage grandeur,
would be hard to surpass. The cliffs, chiefly limestone, are
mostly sheer, and 2000 to 3000 feet high. Wherever vegetation
can find a foothold it is rampant, and a luxuriant jungle of
shrubs clothes all but the most vertical walls of rock. By the
side of the torrent coarse herbs, shrubs, and small trees abound.
The mountain crests and ridges are covered with Spruce and
Pine. Now and again glimpses of vicious-looking, desolate
peaks, towering above thetree-line, were obtainable. The waters
of the torrent roar and dash themselves into foam in their
passionate endeavour to escape to more opencountry. In more
peaceful stretches the river describes a series of S-curves with
shingly areas covered with Myricaria germanica and Hippophaé
salicifolia (Sallowthorn), jutting out into the current. In one
place the cliffs recede somewhat, leaving room for a narrow
valley, where three or four peasants’ huts are pitched. Around
these cabins forlorn patches of maize, buckwheat, cabbage,
Rhubarb (Rheum palmatum, var. tanguticum), and Tang-kuei
(Angelica polymorpha, var. sinensis) are cultivated. The

132 A NATURALIST IN WESTERN CHINA

abandoned clearingsare covered with coarse herbs, among which
‘Senecio clivorum, growing 4 to 5 feet tall, with its golden yellow
flowers, was prominent. Asti/be Davidii also abounds ; like-
wise the Buddleia. A sub-shrubby Elder, growing 3 to 5 feet
tall, with masses of salmon-red fruits, was a pretty sight in
all the more open moist places. (The species proved to be
new and has been named Sambucus Schweriniana, Rehder, in
Plante Wilsoniane, Part II. p. 306 (1912).) The vegetation
indeed is rich and varied, and a large harvest of specimens
rewarded the day’s labours. After scrambling some 30 li
along this gorge we reached the hostel of Lao-tang-fang just
as night was closing in. We encountered considerable traffic
on the road. Coming from Sungpan were coolies laden with
medicines, sheep-skins, and wool. Journeying thither thecoolies
were laden chiefly with wine in specially constructed tubs,
preserved pork andrice. Lao-tang-fang, alt. 7600 feet, consists
of one large new hostel, not quite completed at the time of our
visit ; a long row of “‘bunks”’ are built along one side, with
benches for the accommodation of loads on the other. The
whole structure is of wood, the roofing being of shingles badly
laid. The mud floors were very damp, and vegetation was
- springing up in the corners and under the bunks. Skins of
Serow and Budorcas served as mattress on the bunks, or
settees, and no two of these skins exhibited the same coloration.
Both animals are said to be common in the neighbourhood,
more especially the Serow. The Parti-coloured Bear, or Giant
Panda, also occurs here in the Bamboo jungles.

The hostel was full to overflowing and undoubtedly supplies
a much-needed want. For the sake of future travellers, if for
no other reason, I heartily hope success attends the landlord’s
venture. Formerly a most miserable structure occupied this
site, and I have unpleasant memories of a night spent there in
1904. Except for a tiny cabbage-patch there was no sign
of cultivation around the hostel, but clearings were being
made for the purpose of cultivating Tang-kuei and other
medicines. The view from this spot is savage and grand
beyond power of words. On all sides are precipitous moun-
tains, towering 3000 feet and more above the torrent, all more
or less densely forested. Almost facing the river is a limestone

SSMONS HLIM ‘ONIL-Od HAOSH

SLISOddd AWIT HLIM GHALSAYONA WVAALS AO Gad

:
a 2,

“yt 4 ;

i)
tas

NORTH-WESTERN SZECHUAN 133

cliff with upturned strata on edge, sheer and bare of vegetation.
Behind this is another nearly vertical slope covered with stark,
dead Conifer trees. In the distance, looking back on the
road we had followed, bare, vicious-looking peaks, probably
14,000 to 16,000 feet high, were visible. All around the hostel
the lesser slopes are covered with impenetrable forest of broad-
leaved deciduous trees. The higher parts and the crags are
clothed with Conifers, tall, slightly branched trees of no great
size—altogether a wonderful scene of natural beauty, at present
undefiled by the hand of man.

It was cold ‘during the night; the wind played freely
through the unfinished structure, and the thickest of clothing
was needed in order to keep warm.

The next day wemade a later start than usual, and travelling
most leisurely covered the 40 li to San-tsze-yeh before 5 p.m.
The journey was one long scramble through a continuation
of the savage ravine. The chairs had to be carried piecemeal,
and all of us reached our destination very much fatigued. We
enjoyed a gloriously fine, sunny day, the narrow streak of sky
visible from the bed of the ravine being of the purest Thibetan-
blue. The camera was kept busy and I secured a fine set of
views, but so steep is the country and so dense the jungle that
it was impossible to photograph trees.

The rock-strewn torrent, with its thundering, seething
waters, occupies practically the entire bed of the ravine, leaving
scant room for the road which winds along its banks. We
crossed this torrent many times, either by fording it or by
means of half-rotten log bridges. Luckily the waters were
low and caused us notrouble. In 1904 J ascended this ravine
shortly after heavy rains, and have the liveliest recollections
of the difficulties encountered. Much of the road and many of
the bridges had been washed away, making it necessary to
hew a pathway through the jungle and improvise bridges by
felling trees in several places.

No words of mine can adequately depict the savage, awe-
inspiring scenery of this wild ravine. Stupendous limestone
cliffs, 3000 to 4000 feet high, often too steep for the scantiest
vegetation to find a foothold, but more generally sparsely
or plentifully forested, wall in the torrent and its accompanying

134 A NATURALIST IN WESTERN CHINA

roadway. Waterfalls abound, but lateral torrents are few.
The flora is very rich, but largely inaccessible. Practically
all the trees, shrubs, and herbs common to the 7000 to gooo
feet belt occur here. Conifers are the principal trees. Silver
Fir, Spruce, Hemlock, Larch, White Pine, Juniper, and Yew
are all represented. The Pine (P. Avymandt) is the commonest
tree up to 8500 feet, clinging to the sheer cliffs in a remark-
able manner. With its stunted branches and short leaves it
was hardly recognizable, suggesting a green Maypole rather
than a Pine tree! Many of the Spruce and Silver Fir were
fruiting freely, the erect, violet-coloured symmetrical cones of
the latter being very handsome. Larch (L. Potaniniz) abounds,
overtopping all the other Conifers, but the trees are small. All
the Conifers are hereabouts designated “‘ Sung-shu’’ (liter-
ally Pine trees), but the timber of the Larch, flat-leaved Spruce,
and White Pine, valued in the order given, are most prized for
building purposes generally. Of the broad-leaved deciduous
trees, Maple (Acer), Linden (T7zlia), and Birch (Betula) are the
most common. A few Poplar occur, but Oak is exceedingly
rare, the few noted being scrubby evergreens of no great
beauty. The variety of shrubs is very great, all the more
woodland genera being rich in species. Sorbaria, with its large
panicles of white flowers, was one of the most attractive.
Spireea, Viburnum Lonicera, Rubus, Philadelphus, Sorbus, and
many other families, made a fine display either with their
flowers or fruit. Strong-growing herbs, like the various species
of Senecio, Astilbe, Aconitum, and Anemone, cover miles of the
roadside. In shady places the handsome Maidenhair fern,
Adiantum pedatum, was a charming picture; in sunny spots
the lovely Gentiana purpurata, with intense carmine-red flowers,
was a Sight never to be forgotten.

About ro li below San-tsze-yeh the ravine widens out into a
narrow valley, with the mountain-slopes on the left bank of the
torrent less precipitous and grass-clad. We passed the ruins of
some old forts, and shortly afterwards a Sifan hamlet con-
sisting of three or four farmsteads, with numerous prayer-flags
fixed on the roofs. In the tiny valley wheat, barley, buck-
wheat, oats, peas, and broadbeans are cultivated, and the
crops were ready for harvesting.

NORTH-WESTERN SZECHUAN 135

San-tsze-yeh, alt. g200 feet, consists of ruinous hovels built
on a level with the infant stream which at this point breaks
up into three equal branches, all of which have their source in
the near neighbourhood. Looking back on the route we had
traversed we saw that all the higher peaks are barren and
desolate, the highest of all being flecked with snow. The whole
plexus is made up of the spurs and buttresses of the mighty
snow-clad Hsueh-po-ting. To the north-east from San-tsze-
yeh are other tremendous peaks, bare, barren, and uninviting
in appearance. The aspect of the country around this hamlet
is purely Thibetan. The scant crops and abject poverty of the
inhabitants speak plainly of a country where altitude and
climate set agricultural skill and industry at defiance. Such
regions the Chinese abhor and cannot colonize. The pastoral
Sifan, with their herds of cattle and sheep, remain masters of
the soil though politically subject to Chinese authority. The
conquest of this wild region must have been a most difficult
task and speaks volumes for the military genius which
accomplished it.

During the night at San-tsze-yeh I had a violent attack of
ague, probably caused by a chill, which culminated in a fit of
vomiting. This seizure and the howling of many dogs were
against a good night’s sleep. In consequence we took things
very gently the next day, and I used my chair much more than
usual.

Twenty-five li above San-tsze-yeh, to the right of the stream
which descends the narrow valley, there is a most interesting
place. A torrent heavily surcharged with lime descends from
the eternal snows of the Hsueh-po-ting, depositing along its
course thick lime encrustations of creamy white. The place is
considered holy by the Sifan, to whom any natural phenomenon
strongly appeals. A temple has been erected here and a
series of some fifty tarns constructed by leading the waters
from the stream and making small semicircular dams. All
are at slightly different levels, and the waters as they flow from
one to another continue to build up the dams by leaving
deposits of lime behind. The bed of each tarn is creamy white,
but owing to the light being reflected in different colours,
according to the varying depth of each, an attractive scene of

136 A NATURALIST IN WESTERN CHINA

many-coloured waters is presented. Some are clear azure blue,
others creamy white, pink, green, purple,andso on. The temple
is called ‘‘ Wang Lung-ssu’”’ (Temple of the Dragon Prince),
and it is fitting that the Sifan, children of nature as they are,
consider the place holy. Near the temple the waters have built
up a wonderful series of waterfalls, and every fallen tree and
bush obstructing the waters is speedily encrusted with lime.
Above the temple the stream is fully 80 yards wide, and the
bed is creamy white with soft encrustations of lime, the ripple
marks being beautifully defined. These lime-deposits extend for
a mile or two and present a most striking scene.

From the bed of this stream, a short distance above the
temple, a fine view of the snow-clad Hsueh-po-ting is obtain-
able. The face visible carries but little snow, and immediately
below the glaciers are wonderful cliffs of red-coloured rock. In
contrast the colour-effects are most remarkable. There was
said to be another temple some few li higher up towards the
snows, but I was too fatigued to visit it.

All around Wang Lung-ssu are fine forests of Spruce, Silver
Fir, Birch, with miscellaneous trees and shrubs. In the
vicinity of the lime-deposits the trees look very unhealthy,
many are bleached and dead, others yellow and dying. From
the vegetation it is evident that these lime-deposits are recent
and spreading rapidly. A few Rhododendrons occur on the
margins of the stream and in the woods, but are not happy.
Right by the water’s edge I gathered Arctous alpinus, var. ruber,
a tiny alpine shrub with red fruit closely allied to the Blueberries,
and found also near the glaciers in British Columbia! This
pretty little plant, only some 4 to 6 inches high, is quite common
hereabouts, but had not before been recorded from China.
Near the tarns Cypripedium luteum, a yellow-flowered counter-
part of the North American Moccasin flower (C. spectabile),
is very abundant. (Later I succeeded in introducing live roots
of this species to the Arnold Arboretum, where plants are now
growing.)

_ The forests of this immediate neighbourhood are rich in fine
Spruce trees, 80 to 150 feet tall and 6 to ro feet in girth, with
short branches producing a spire-like effect, are characteris-
tic of the region. The Silver Fir are less noteworthy, but, like

NORTH-WESTERN SZECHUAN 137

the Spruce, were fruiting freely. (Both were subsequently
introduced to cultivation.) Larch overtops all other trees,
reaching its limits at about 12,000 feet altitude. The vegetation
of the ranges flanking the narrow valley, up whichthe main road
ascends, presents a remarkable contrast. The range to the left
of the stream, above 10,000 feet altitude, is covered only with
scrub and grass; whereas the range on the right bank is
heavily forested up to altitude 12,000 feet. Early in the after-
noon, after covering 40 li, we reached the lonely hostel of San-
chia-tsze, alt. 12,800 feet, situate some 600 feet below the head
ofthe pass. During the first 25 li of the day’s march we passed
several large farmhouses, but nearly all are deserted and falling
into ruins. Around these houses a few plats of wheat, barley,
flax, and Irish potato are cultivated; also cabbage, garlic,
and other vegetables in minute quantities. Tobacco (Nicotiana
vustica), in small quantities for household use, is grown around
San-tsze-yeh, and the crop looked very happy. These sporadic
attempts at cultivation represent the vain and futile efforts of
the Chinese settlers to eke out an existence from the inhos-
pitable soil. This side of the pass is evidently much colder
than the Sungpan side, since there, at greater elevations, good
crops of wheat, barley, and peas can be raised.

Apart from the forests already mentioned, herbs dominate
the flora. A great variety were still in flower, the various
species of Senecio and Gentiana being most striking. Gentiana
detonsa, a slender plant a foot and more tall, with numerous
large deep blue flowers, looked particularly happy, flaunting
its blossoms in the sun. On rocky screes the yellow-flowered
Clematis tangutica is abundant and was covered with its
top-shaped blossoms. The hedges bordering the fields are
composed chiefly of Wild Gooseberry and Sorbaria arborea:
the latter was in full flower. In copses by the stream, up to
11,500 feet, Hornbeam, Cherry, Red Birch, Willow, Maple, and
Hazel-nut are common. The Hazel-nut is mainly Corylus
ferox, var. thibetica, a variety having a spiny fruit closely
resembling that of the Sweet Chestnut (Castanea).

The hostel of San-chia-tsze is maintained for the accommo-
dation of travellers, and a posse of soldiers is stationed here to
keep down banditti. The hostel is a roomy but miserable

138 A NATURALIST IN WESTERN CHINA

cabin, built of shales and roofed with shingles held down by
stones. The floor is of mud and very uneven; there is no outlet
for smoke, save the doorway, and no windows. At midday
a candle was necessary to avoid falling over things when moving
about indoors. During different visits I have suffered many
days and nights in this lonely spot, on one occasion being
snowed in for three consecutive days. The cabin is situated
on a narrow sloping valley running nearly east and west, a
mile or so above the tree-limit, flanked on the northern side
by a ridge of stark, crumbling rocks. To the south the range
culminates in bare peaks and eternal snows of the Hsueh-po-
ting. The moorland country all around is typical of Eastern
Thibet, so perhaps a few details are permissible. The treeless
spurs and valleys are covered with extensive heaths of scrub,
made up of several species of Spirzea (including S. mollifoha,
S. alpina, and S. myrtilloides), Sibirea levigata, Lomicera
ispida, L. chetocarpa, L. prostrata, L. thibetica, and others,
several Barberries, Currants, shrubby Potentillas, Astrag-
alus, Sallowthorn, small-leaved, twiggy Rhododendrons, and
Juniper. As the altitude increases, one by one these shrubs
give out until only the Juniper is left. This ceases about 15,000
feet ; alpine herbs ascend another rooo feet, and the limit of
vegetation is roughly 16,000 feet. The Juniper scrub is from
i to 24 feet tall, very dense and difficult to traverse, but
furnishes excellent fuel. Mixed with this scrub are herbs
in great variety, the Poppyworts (Meconopsis) being par-
ticularly abundant. Possibly the commonest herb between
12,500 feet and 14,000 feet is Meconopsis punicea, a lovely
species having large, dark-scarlet nodding flowers. (It was
from near this vicinity that I succeeded in introducing this
plant in 1903.) The violet-blue flowered M. Henrici is common
between 13,000 feet to 14,000 feet, but much less so than around
Tachienlu. The prickly M. racemosa, with blue flowers, is
plentiful in rocky places between 13,000 feet to 14,500 feet.
From 11,500 feet to 13,000 feet the gorgeous M. integrifolia,
growing 3 feet tall, with its peony-like, clear yellow flowers
8 to 11 inches across, occurs, but is not plentiful. The intense
colours among alpine flowers everywhere is well known, and
this region is no exception. The yellow is mostly supplied by

GANAOWANOA NI LNOA GHUNINAY ‘SSVd NVHS HHASH WOMA LSVAHLNOS LSVa ONIMOOT

NORTH-WESTERN SZECHUAN 139

Senecio, Saussurea and other Composite, slender growing
Saxifraga, etc. The blue and purple by various Aconites,
Larkspurs, and Gentians ; among the latter Gentiana Veitchio-
vum, with large erect flowers, covers large areas. The Lousewort
(Pedicularis) and Fumewort (Corydalis) are represented by
many species, having flowers embracing all the cardinal colours.
Primulas occur, but not in many species. Androsace, Sedum,
Cyananthus, and other alpine genera abound.

Large flocks of sheep are pastured on these uplands, but
yak are not kept in quantity hereabouts. There is not much
variety of game. Blue sheep are common, Budorcas are found
near the timber line ; on the higher crags occasional flocks of
Goa, or Thibetan gazelle, occur. Snow-partridge, Thibetan
Hazel-hen, Snow-cock and allied game-birds, together with
Thibetan Hares, are fairly numerous. Tke Wolf is the only
carnivorous animal really common.

The Hseuh-po-ting snows are visible on clear days from the
wall of Chengtu city, and are accounted the ‘‘ Luck of the
Plain.” The Chinese claim that so long as snow covers
this peak the prosperity of Chengtu and its surrounding plain
is assured. It was a perfect moonlight night on the occasion
of my last sojourn at San-chia-tsze, and late in the evening I
beheld the “Luck of Chengtu,” with its crown of eternal
snow lit by the radiant moonlight. The loneliness of the
region, the intense stillness on all sides, and the wonderful peak
with its snowy mantle, made a most impressive scene.

A glorious morning followed a perfect night. From the
head of the pass (alt. 13,400 feet) I obtained further good views
of the Hsueh-po-ting, bearing west-south-west and secured
some photographs. The peak is probably 22,000 feet high, in
shape an irregular tetrahedron, the south-west slopes carrying
enormous snow-fields. The north-east face is very steep
and carries but little snow. The surrounding peaks are bare
and desolate looking; no vestige of life was discernible, and
the whole scene was lonely, most forbidding, even awesome,
though bathed in brilliant sunshine.

Below San-chia-tsze are the stone ruins of an old fort and
stockade, relics of ancient warring times, but now covered
with various herbs, especially Saxifrages, which were masses

140 A NATURALIST IN WESTERN CHINA

of yellow, and other coloured flowers. The head of the pass
is marked by a ruined tower and fort, from the summit of which
Thibetan prayer-flags waved. That robbers still haunt these
regions was brought home to us by the sight of a partially
covered coffin near the head of the pass. A few weeks before,
a poor coolie, bound towards Lungan Fu to purchase rice, was
attacked here, robbed, and killed. The bandits got clear away.
The coolie’s “ pai-tzu’’ (a framework for carrying loads on)
and various appurtenances lay on top of the coffin and remain
to tell the story of the crime. All around are grassy areas,
covered at the season of our visit with blue and yellow alpine
flowers.

At the head of the pass small boulders of sandstone, marble,
granite, and other rocks lay scattered around. Just below are
beds, which resemble coal-ashes, probably of volcanic origin.

From the pass we dropped down into a valley which quickly
led to fields of golden wheat and barley. The crops were
ripening, and here and there the reapers were busy. Passing
a ruined fort, several Sifan farmsteads, and a lamasery, the
road led to the summit of a grassy ridge. Descending a few
hundred feet we sighted the city of Sungpan nestling in a
narrow, smiling valley, surrounded on all sides by fields of
golden grain, with the infant Min, a clear, limpid stream, wind-
ing its way through in a series of graceful curves. In the
fields the harvesters were busy, men, women, and children,
mostly tribesfolk, in quaint costume, all pictures of rude
health, laughing and singing at their work. Under a clear
Thibetan-blue sky, the whole country bathed in warm sun-
light, this busy scene of agricultural prosperity gladdened the
hearts of all of us, fatigued and exhausted as we were from the
hardships of our journey through savage mountains with their
sublime scenery and wonderful flora.

CHAPTER: XI
SUNGPAN TING

THE LAND OF THE SIFAN

HE city of Sungpan is situated on the extreme north-
west corner of Szechuan, about long. 103° 21’ E., lat.

32° 41’ N.,at an altitude of 9200 feet, and is the farther-
most outpost of Chinese civilization in this direction. The sur-
rounding country, more especially to south-west, west, andnorth-
west, is inhabited by Sifan, a people concerning which very little
is known. Originally established as a military post after the
conquest of the neighbouring regions by the Emperor Kienlung
about A.D. 1775, Sungpan has developed into a most important
trade entrepét. Itisacity of the second class (styled “‘ Ting’),
but the head civil official has the local rank of prefect, his
full title being “‘ Fu-I-Li Min-Fu,” which signifies ‘‘ the Bar-
barian-cherishing, Chinese-governing Prefect.’’ This fanciful
title has reference to the official’s control over the neighbouring
Sifan tribes—a control which is purely nominal. The military
importance of this stronghold is still fully recognized, and its
strategic value is beyond question. A Chinese general (Chen-
tai), in command of ten regiments, has his headquarters here,
with jurisdiction extending south to Kuan Hsien, east to
Lungan Fu, and north-east to Nanping in Kansu province.
The town is most picturesquely situated, occupying con-
siderable space in a narrow, highly cultivated valley flanked
by steep mountain-slopes 1000 to 1500 feet high. The Min (Fu)
River, which takes its rise some 35 miles to the north, winds
a circuitous course down the valley and flows through the
town in an §-curve, entering and leaving through the city
walls at unfordable points. On the western side the town is
backed by a steep slope, up two sides of which a wall is carried.

142 A NATURALIST IN WESTERN CHINA

The west gate of the city is situated at the top of this slope,
and is exactly 1000 feet above the river. Save for a yamén
and a temple or two the whole of the mountain-slope within
the walls is given over to terraced cultivation, the city proper
being clustered in the valley alongside the river. The wall
surrounding three sides of the city is very substantially built
of brick, being fully 20 feet thick and more high, but that which
ascends the mountain-sides is in places only 2 feet thick and
4 feet high; a steep ravine, however, immediately outside
this wall, affords additional protection. Since the Chinese first
established themselves here the town has undergone many
vicissitudes. Time and again the Sifan have swept down
upon it, captured it, and massacred all who fell into their hands.
So frequent have been these attacks, and so great is the Chinese
dread of treachery on the part of the Sifan, that it is only
within the last few years that any of these people have been
allowed to remain overnight within the city walls.

In 1910 Sungpan had a resident civil population of about
3000 people, and a floating population equalling, if not exceed-
ing, thisnumber. The houses are nearly all of wood, generally
well built, with rather curiously-carved porticoes ; the timber
employed for building is mostly Juniper, which is floated
down the Min River from a point some 15 miles to the north-
north-east. In October 1gor the city was two-thirds destroyed
by fire, but on the occasion of my last visit in 1910 the de-
vastated area had been practically rebuilt. The streets are
badly paved, ill-kept, and the city possesses no buildings of
architectural beauty. Near the south gate the military
section of the town is situated, and a considerable amount of
market-gardening is carried on there. The people are very
fond of flowers, nearly every house boasting some in pots,
on the walls, or in borders. Stately Hollyhocks, with multi-
coloured flowers, are a feature. With these are generally
planted Tiger Lilies, Chinese Asters, and small-flowered
Poppies, the whole making a bright and pleasing effect. The
Chinese Aster (Callistephus hortensis) is wild in the neigh-
bourhood ; the Poppy is a species closely allied to Papaver
alpinum. The population is mainly Mohammedan Chinese,
who carry on a remunerative barter-trade with the surrounding

THE CITY OF SUNGPAN TING

Scie ie A a Leak
, ja id ate

ro La 3 re 9 ‘ 2a i,
el pha? oa ow Ws ela .

en a wien
=o veage. ha

SUNGPAN TING 143

tribes. Tea is the all-important medium employed, this
commodity and a few odd sundries being taken in exchange
by the tribesmen for their medicines, skins, wool, musk, etc.
During the month of July a fair is held annually for trade
purposes. The people from far and near attend this fair, a
vast amount of business being transacted. Trading caravans
also make long journeys into the country north-west to the
borders of the Kokonor regions. Wool, sheep-skins, and various
medicines in great quantity are exported from Sungpan to
different parts of China.

The trade passing through Sungpan is, I am convinced,
not only greater than has been estimated, but is increasing
annually. In 1903, on the occasion of my first visit to this
town, I enjoyed the companionship of W. C. Haines-Watson,
Esq., then Commissioner of the Imperial Maritime Customs
at Chungking. This gentleman investigated the trade of this
region, estimating the exports to Thibet at Tls. 801,000, and
those into China at Tls. 512,000 (‘‘ Journey to Sungp’an,”
Jour. China Branch Roy. Aszat. Soc., 1905, xxxvi). Our
visit occurred before the city had recovered from the disastrous
fire of rgor, and trade was suffering in consequence. In 1910
trade was evidently booming. I have no figures to guide me,
but comparing the two visits I would put the trade with China
alone at a million taels. This trade has three outlets: one,
east, via Lungan Fu to Chungpa; another, south-east, via
Mao Chou, Shihch’uan Hsien to An Hsien ; the third through
Kuan Hsien to Chengtu. The first two routes afford water
communication from Chungpa and An Hsien respectively,
with Chungking on the Yangtsze River. By these routes
most of the goods intended for Chungking and beyond are
conveyed. The trade via Kuan Hsien is mainly with Chengtu
and other cities on the plain. This latter trade route has been
looked upon as the most important, whereas it is really less
so than either of the other two outlets.

The late Captain W. J. Gill in 1877 was the first Occidental
to visit Sungpan. Since that date several foreigners have
paid visits, and missionaries of Protestant denominations have
made abortive attempts to establish stations there. I have
visited this place three times, and on each occasion enjoyed

144 A NATURALIST IN WESTERN CHINA

the stay and departed with regret. Did the Fates ordain that
I should live in Western China I would ask for nothing better
than to be domiciled in Sungpan. Though the altitude is
considerable the climate is perfect, mild at all times, with, as a
general rule, clear skies of Thibetan-blue. During the summer
one can always sleep under a blanket, in winter a fire and extra
clothing are all that is necessary. Excellent beef, mutton,
milk, and butter are always obtainable at very cheap rates.
The wheaten flour makes very fair bread, and in season there
is a variety of game. Good vegetables are produced, such as
Irish potatoes, peas, cabbages, turnips, and carrots, and such
fruits as peaches, pears, plums, apricots, apples, and Wild
Raspberries (Rubus xanthocarpus). Nowhere else in interior
China can an Occidental fare better than at Sungpan Ting.
With good riding and shooting, an interesting, bizarre people
to study, to say nothing of the flora, this town possesses attrac-
tions in advance of all the other towns of Western China.

The valley, which varies from } to 4 mile in width, and the
mountain-slopes, rising from 1000 to 1500 feet above, are given
over to wheat and barley cultivation, with occasional fields
of peas and flax, the latter being valued for its seeds, which
yield an oil used as anilluminant. In the latter half of August
the whole country-side is one vast sheet of golden grain bending
to the wind. This grain is reaped, leaving a generous stubble,
which is immediately ploughed under. The ploughs are simple,
consisting of an iron-shod shear, a straight handle of wood, and
a long shaft, to which is harnessed a couple of oxen or half-bred
yak.

In harvesting the grain, tribesfolk (chiefly Po-lau-tzu),
who come from the upper reaches of the Tachin Ho, many
days’ journey to the west-south-west, play an all-important
part. Every year these people visit this region for the express
purpose of this work, and are, in fact, indispensable. As the
crop is reaped it is tied into little sheaves and stacked ears
downwards on high hurdle-like frameworks (Kai-kos) to await
threshing. The threshing is done by wooden flails, both men
and women taking part in the work. The corn is ground in
mills driven by water-power.

The name ‘‘ Sungpan’”’ has reference to forests of Spruce

TWO SIFAN

SUNGPAN TING 145

and Fir and the circuitous course of the river Min (Fu). The
river still pursues its winding course, but the forests have long
since disappeared. It is only in temple grounds and among
tombs that any trees remain. The mountains are absolutely
treeless, where not under cultivation they are covered with
scrub and long grass. The outer crust of the mountains con-
sists of a rich flaky loam, probably of glacial origin, ratherheavy,
but specially adapted for cereal cultivation. In the grass and
scrub Pheasants are very plentiful in the neighbourhood of
cultivation, so also is a long-eared, light-grey-coloured Hare.
Musk Deer, Wapiti, and White Deer occur in the neighbour-
hood. On the moorlands a Marmot, called ‘‘ Hsueh-chu”’
(Snow-pig), burrows in large colonies.

North-west of Sungpan is the Amdo country, a region of
grasslands. The Chinese designate it “‘ Tsaoti,’’ which may be
interpreted “‘ Prairie.”’ This region is made up of rolling
country above 11,000 feet altitude, where vast herds of cattle,
sheep, and many ponies are reared. A great part of this
region is peopled with pastoral Sifan, but the more remote
parts are in the hands of nomads belonging to Ngo-lok and
Nga-ba tribes, of evil reputation as robbers and bandits,
dreaded alike by Chinese and the more peaceably inclined
Sifan. These robber tribes are of Tangut origin, having their
headquarters around the Kokonor region. Being of nomadic
habit they wander far afield, and rob caravans and kill the
settlers weaker in numbers than themselves. When I arrived
in Sungpan in 1910 I found there some 200 soldiers from
Chengtu bent on a punitive expedition against these banditti.
About a year previously a Chinese official had been murdered
in the Amdo country, not many days’ journey from Sungpan,
and no redress had been obtained. Nine persons were held
guilty for this crime, but in spite of demands on the part
of the Chinese the clan would not give up these people.
The affair ended in the Chinese killing as many members of
this robber clan as the small army sent on the expedition
could capture. It is from the Amdo region that Sungpan
derives most of its wool, skins, and medicines, consequently
the trade depends very largely upon peace obtaining there.

The Sifan (Western people) are unquestionably of Thibetan

VOL. I.—Io

146 A NATURALIST IN WESTERN CHINA

origin. They are not nomads, but essentially a pastoral and
agricultural people. In dress, speech, and facial character-
istics they agree closely with the inhabitants of anterior
Thibet. Their houses are similarly constructed, and Lamaism
dominates their lives. As a people the Sifan are divided
into several tribal clans: those around Sungpan style them-
selves Murookai; those a little to the south-west of this town
Lapp&. Immediately around Sungpan the Chinese language
is generally understood, but away from the town colloquial
Thibetan only is spoken, each hamlet having an interpreter
to conduct all affairs with the Chinese. These people are ruled
by head-men who are held directly responsible for the proper
maintenance of law and order. The Chinese policy is one of
non-interference in so far as it is consistent with the status
of China as the paramount power.

The Sifan men as seen in the streets of Sungpan and
the immediate neighbourhood are swarthy in appearance and
average 5 feet 6 inches in height or rather more ; in walking
they have a clumsy gait and are generally awkward and sullen
when approached. Their dress is a sort of “‘ cover-all ’’ made
of grey or claret-coloured serge, confined around the waist
by a girdle; the right shoulder is generally uncovered. This
garment is often edged with fur ; sometimes it is made entirely
of sheep-skins, with the wool worn inside. Short trousers and
high felt boots cover the legs and feet, though in the streets
they frequently go barefooted. The head-gear is either a low,
stone-coloured, soft felt hat, with turned-up brim bordered
with black, or a high, cone-shaped, light grey felt hat edged
with white sheep-skin. Occasionally those living near Chinese
settlements affect a dirty turban. The hair is worn long and
gathered up inside the hat. The Lamas have their heads
close-cropped or shaven, and when seen in the streets are
usually hatless. In ceremonial dress they wear a sort of
cocked hat made of grey serge covered with a mass of fluffy
yellowish woollen stuff. Muleteers and men generally, when
travelling, go armed with swords, knives, and long guns, the
latter fitted with a fuse and a fork to rest the barrel on when
taking aim. All wear charm-boxes on their chests, and carry
a flint-box and tinder suspended from their girdle ; somewhere

SSVd NVHS HANSH NO SLIGNVi AT GHYAGCUNNAW NVW AO AWOL

an
ie

sss aa iy

SUNGPAN TING 147

about their person a wooden, often silver-lined, eating bowl
is also carried. The wealthy prize a leopard skin garment
most highly.

The young girls are occasionally passing fair to look upon,
but from hard work and exposure lose all charm of youth very
early. The women are generally flat-faced, very dirty, and
far from prepossessing. They have, however, considerable
character and an important voice in household and all business
matters. Toward foreigners they are timid, but amongst
themselves their manners are playful, free and easy, and they
laugh and sing at their work. Their outer dress consists of one
shapeless piece of serge, which envelops them down to their
ankles. Sometimes this is grey, more usually it is blue in
colour, with a fancy bordering of dark red or yellow in front and
around the bottom. High boots of untanned leather encase
their feet and lower legs. Their hair is long and black, worn
parted in the middle and collected into one large plait behind ;
around the forehead it is worn in a series of tiny plaits orna-
mented with coral-beads, amber-coloured stones, and small
shells. The large plait is usually wound around the head,
together with a piece of cloth to form a kind of padded turban,
the whole being decorated with shells and beads. Occasionally
saucer-shaped felt hats are worn. In holiday attire, silver rings
and gaudy red and yellow tassels are added to their coiffure.
They are very fond of silver rings, bracelets, and large ear-rings
ornamented with beads of turquoise and coral. In gala costume
the dress is decidedly picturesque.

The men assist in tilling the soil, and in sowing and harvesting
the crops, but the women do the bulk of the work around the
homestead, the men being away herding the flocks or onjourneys.
Though they lead hard lives they seem a happy and contented
people in spite of the fact that they are almost without exception
afflicted with goitre. Their houses are built of wood and shale-
rocks, being either one-storied, flat-roofed, with or without a
raised part behind, or, as is more usual, two-storied and
similarly roofed. They count their wealth in head of cattle,
horses, and sheep. Wheat, barley, and peas are the staple
crops. Meat, butter, and milk enter very largely into their diet.
Buttered tea is generally drunk, but they are very partial to

148 A NATURALIST IN WESTERN CHINA

a kind of small beer which they brew from barley ; they are also
fond of Chinese wine.

Monogamy is the rule, but polygamy is common, it being
merely a question of wealth. Polyandry is not practised, but
the morals are lax, as is the case everywhere else under Lamaism.
Marriage is by consent on the part of the girl, presents of oxen
and sheep being made on behalf of the bridegroom to the girl’s
parents; children are appreciated, but the Sifan are not a
prolific people. The second son generally enters a lamasery,
as is customary throughout Thibet. Widows are permitted
to remarry. The dead are disposed of by burial or by being
thrown into the rivers.

Abundant signs of Lamaism are everywhere apparent.
Prayer-flags flutter from the housetops, mountain-peaks,
across streams, and surmount cairns of rocks. Mani-stones are
heaped by the wayside ; praying-wheels, turned either by hand,
by the wind, or by the currents of streams, occur on all sides.
From the people at their work, either in low crooning tones or
in loud chorus, the mystic hymn, ‘‘Om mani padmi hom,” is
continually ascending to heaven. The chant of the Sifan is
decidedly musical, rising and falling in soft rhythmic cadence.
I have often listened to them with much pleasure, though from
a distance, since if one tried to approach closely they ran helter-
skelter away. They are naturally very superstitious, being
fond of charms, afraid of evil spirits, and reverence unusual
natural phenomena. Though my associations with the Sifan
were brief I always received the utmost courtesy at their hands,
and found much that was pleasing and interesting among these
happy, unsophisticated children of Nature.

CHAPTER’ XII
THE CHINO-THIBETAN BORDERLAND

‘THE MARCHES OF THE MANTZzU ”

political boundary between Szechuan and Thibet. Indeed,

no actual frontier has ever been agreed upon, consequently
it does not exist, except at one point, on the highway leading
from Tachienlu, via Batang, to Lhassa. There, on the Ning-
ching shan, three and a half days’ journey west of Batang, stands
a four-sided stone pillar, some 3 feet high, having been erected
in A.D.1728. The guide-book to Thibet says: ‘‘ All to the east
is under Peking; the territory to the west is governed by
Lhassa.’’ As to the regions north and south of this stone,
nothing is said.

For all practical purposes the Min River, from Sungpan Ting
inthe north-west to Kuan Hsien, maybe regarded as the frontier
thereabouts. From Kuan Hsien southwards an imaginary line
drawn through Kiung Chou, Yachou, Fulin to Ningyuan Fu,
and thence to the Yangtsze River, may be accepted as com-
pleting the frontier line. This constitutes a well-defined
ecclesiastical boundary between the peoples. Also it corre-
sponds very closely with the western limits of the Red Basin,
which constitutes an unmistakable physiographical frontier.
It is true that at certain points, such as Lifan Ting, Monkong
Ting, Tientsuan Chou, and Tachienlu, the Chinese have suc-
ceeded in establishing trading-centres and military depots.
But in all these places the population is mixed and the centres
themselves surrounded on two or more sides by non-Chinese
people. West of the boundary here indicated the Chinese
occupy a very limited aggregate area, being confined to the

high roads and to a few valleys suitable for rice and maize culti-
149

| T is impossible to define, with any approach toaccuracy, the

150 A NATURALIST IN WESTERN CHINA

vation. The largest of these areas is the region known as the
Chiench’ang Valley, of which the city of Ningyuan Fu is the capi-
tal. This narrow strip extends down to the Upper Yangtsze
River, being bounded on the east by the independent kingdom
of Lolo, which occupies the higher slopes of the Taliang shan
and has never been conquered by the Chinese. Immediately
to the west of the valley the country is peopled bysemi-indepen-
dent tribes akin to the Thibetans. Indeed, the Min River, with
such land to the immediate west suited to rice-culture, may
well be regarded as the real boundary of western Szechuan from
Sui Fu on the Yangtsze River to Sungpan Ting, in the extreme
north-west corner of the province. An arc-line, commencing
at Sungpan Ting and connecting with the boundary stone
west of Batang, thence southwards, skirting the right bank of
the Drechu (Upper Yangtsze), would form roughly the boundary
of Thibet proper. Nominally the whole of this region is con-
sidered by the Chinese part of Szechuan province. In certain
books and maps parts of this region are referred to as Eastern
Thibet, and much confusion has arisen from this misnomer.

The country included within the boundaries here given
constitutes the hinterland between Szechuan and Thibet, and
failing a more lucid term it may be designated the ‘‘ Chino-
Thibetan borderland,”’ a title which, if clumsy, has the merit
of being both descriptive and accurate. Several trade routes
traverse this borderland, but with one exception these have
been little travelled by foreigners—the exception being the
great highway between Chengtu Fu and Lhassa De, which
crosses this region from Yachou, via Tachienlu and Batang
to the boundary, and is closely controlled by Chinese. Apart
from this highway and the country in its immediate vicinity
as far west as Tachienlu, the whole borderland is very much a
terra incognita. It is made up of a series of stupendous moun-
tain ranges, separated by narrow valleys, well forested in the
lower parts with all the higher peaks extending above the
snow-line. These ranges are comparable only with the Hima-
layas, of which, indeed, they constitute a north-east extension.
This rugged region is populated by many independent or quasi-
independent tribes, more or less Thibetan in origin, with the
exception of the Lolo.

THE HIGHWAY TO TACHIENLU AND BEYOND, HERE BLASTED
FROM SOLID ROCK

THE CHINO-THIBETAN BORDERLAND 151

It is a region where altitude and climate, rather than longi-
tude and latitude, define the frontiers. In the north-west the
highlands of Central Asia abut more closely on the Red Basin
than they do in the south-west, and form uplands suitable as
grazing-grounds for herds of yak, cattle, horses, and sheep.
These areas are peopled by nomadic Thibetans, with whom
agriculture is relatively unimportant. The broken country,
made up of mountain-crag and valley, which forms the greater
part of this hinterland, is occupied by various tribes, with
whom agriculture is the paramount industry, and wheat,
barley, and buckwheat the staple food-stuffs. The forests of
this region contain much game, of which these people are skilled
hunters. Lastly, in the more fertile valleys, where rice and
maize can be successfully grown, Chinese settlers are found, but,
as mentioned earlier, away from trading-centres and the great
highway between Chengtu and Tachienlu, they are not much
in evidence

In the first chapter brief reference to the mountain chains
and rivers of this region has been made, but perhaps a few of
the more striking features may be given in detail here. Unlike
the mountains bordering the eastern limits of the Red Basin,
which are mainly of hard Carboniferous and Ordovician lime-
stones, those of the west are principally of mudshalesand granitic
rocks. Here and there, for example Mount Omei and its sister
mountains Wa and Wa-wu, hard limestones have been forced
up through the older rocks and form bold peaks and stupendous
precipices. There is indeed plenty of limestone throughout
the hinterland, but Pre-Cambrian rocks preponderate enor-
mously. These and the shales (probably Silurian) disinte-
grate very readily in their exposed parts and erosion is rapid.
In the deforested parts landslides are general. The region is
fairly rich in gold, silver, copper, lead, iron, and other minerals,
but very little mining is carried on. Coal is very rare, except
in a few localities where limestone predominates, as near Mount
Omei and the surrounding region. Salt is known from one
locality only (Pai-yen-ching in the Chiench’ang Valley). Around
Tachienlu hot springs of calcareous and chalybeate waters,
more or less rich in sulphur, are common. These springs are
usually found in close proximity to torrents, very often occur-

152 A NATURALIST IN WESTERN CHINA

ring in the actual bed of such streams. In many the waters
are actually boiling, and I have several times cooked eggs in
them. These hot springs are much resorted to by the people
of the surrounding regions for bathing purposes, the waters
being esteemed as a cure for rheumatism, skin affections, and
other complaints.

Three large rivers, Tung, Yalung, and Dre, flow through
this borderland, mainly from north to south, as necessitated by
the direction of the mountain axes. These rivers have tribu-
taries in abundance, and the majority of them, draining from
eternal snows, carry down enormous quantities of water and
detritus. None of these rivers is navigable save for rafts,
specially constructed boats, or skin coracles, over very short
and interrupted stretches. Bridges and ferries are few, never-
theless the highways and by-ways of this region skirt the banks
of these rivers and their main tributaries.

The valleys of all these streams, and for the purpose of
what follows the Min above Kuan Hsien may be included, are
deeply eroded, the waters flowing between steep slopes or pre-
cipices. These valleys are all very similar, being narrow, shut
in by lofty treeless mountains, and all enjoy a much hotter,
drier climate than their altitude warrants. Long stretches
are very barren and desert-like, more especially when the out-
cropping rocks are solely granitic. Owing to this dry, hot
climate, interesting anomalies obtain in these valleys. At
Hokou, on the Yalung, maize can be cultivated up to nearly
9500 feet altitude, whereas at Tachienlu, in the same latitude
and 1000 feet less altitude, it is impossible to bring this cereal
to maturity. Green parrots (Pal@ornis derbyana salvadort)
occur as summer migrants in the valleys of the Yalung and
Drechu up to 10,000 feet altitude. Rock pigeons occur in
multitudes throughout all these valleys above 4ooofeet altitude.
Monkeys also are common. The flora generally is specially
adapted to withstand drought, and is more closely allied to
that of the Yunnan plateaux than to the contiguous country.
Doubtless at one time the mountain-slopes flanking these
valleys were wooded, though it is improbable that the lower
slopes were ever heavily forested; but such timber as grew
there has long since disappeared, and to-day these slopes are

THE CHINO-THIBETAN BORDERLAND 153

clothed only with coarse grass and scrub. Landslides are a
feature of these regions, especially during the melting of the
snows or after heavy rains in the surrounding high mountains.
At such times travelling hereabouts is highly dangerous, as
nearly every traveller can testify from ocular proof. I have
witnessed several disastrous landslides, involving loss of life
and much destruction of property. In 1910, when descending
the Min Valley, I unfortunately got involved in a minor one,
and sustained a compound fracture of the right leg just
above the ankle. In many places rockslides are constantly
occurring, and warning notices to travellers not to tarry are
frequently displayed throughout the Upper Min Valley and
elsewhere.

Small villages and farmsteads are scattered through these
valleys where, goaded by stern necessity, the inhabitants
maintain a grim struggle to win a sustenance from the in-
hospitable soil. Where rice and maize can be cultivated
Chinese settlers are found, but above the altitudes admitting
of this the tribes are in full possession and cultivate crops of
wheat, barley, buckwheat, peas, and linseed—the latter for
its oil, which is used as an illuminant. Exceptionally good
Chilli peppers (Capsicum) are grown in these valleys, and certain
regions, for example Mao Chou, in the Min Valley, are renowned
for this produce. Around habitations a few trees, chiefly
Poplar, Alder, and Willow, are always present, affording a
welcome shade. Cupressus torulosa, a handsome timber tree,
often 80 to 100 feet tall, is very much at home in these valleys
and probably at one time covered quite considerable areas
hereabouts. This tree is well worth the attention of those en-
gaged in reafforestation work in dry, warm-temperate regions.
Other trees partial to these same conditions are Sophora
japonica, Diospyros Lotus, Pistacia chinensis, Erythrina indica,
Kelreuteria apiculata, Ailanthus Vilmoriniana, Celtis spp., and
the Soap trees (Sapindus mukorossi, Gleditsia spp.). Many
fruit trees occur, including the Pear, Apple, Peach, Apricot,
and Walnut; the latter (Juglans regia) is the commonest
tree up to 8000 feet. The natives hack the lower trunk
to make the tree fruitful, so they claim, showing that
the old adage—‘‘ beating the Walnut tree’’—is known out-

154 A NATURALIST IN WESTERN CHINA

side of Europe. Mulberry trees, the Cudrania tricuspidata,
and tall-growing Bamboos are common up to 4500 feet
altitude.

Many of the shrubs found growing in these valleys are
spinescent and nearly all are adapted to withstand drought.
In the majority the leaves are very small or covered with a
dense felt of hairs. These shrubs are usually scrubby in appear-
ance yet many produce ornamental flowers or fruit. The
“ Southernwood ”’ (Artemisia spp.), with silver-grey, elegantly
dissected foliage and yellow flowers, are perhaps the commonest
shrubs met with hereabouts. Barberries are another special
feature, and when laden with masses of red fruit and autumn-
tinted foliage present a most attractive picture. This same
remark applies to various species of Cotoneaster, all having
ornamental fruit. Many kinds of Rose occur, but often the
species are local. Common to all these valleys, though most
abundant in that of the Yalung, is Rosa Soulicana, with fragrant
flowers, opening sulphur-yellow and changing to white. So
also is Miss Willmott’s charming rose (R. Willmottie), with
its abundant straw-yellow prickles, neat glaucescent leaves,
rosy-pink flowers, and orange-red fruit. The beautiful R.
Hugonts is confined to a narrow stretch of the Min Valley
between 3000 to 5000 feet. This is the only rose with yellow
flowers I have met with in China. The fruit is black
and falls very early. R. multibracteata, an odd-looking species
having pretty pink flowers, is very common in the upper
reaches of the Min Valley and less so in that of the Tung.
Forms of the Musk Rose (R. moschata) and of R. sericea occur
but are local. With the exception of the “‘ Southernwood,”’
all these shrubs confine themselves closely to water-
courses. In more arid places Caryopteris incana and other
species, with intense blue flowers opening in late July, are
very abundant, so also are different species of Indigofera,
with pink to red-purple flowers. Several species of Buddleia
and two varieties of the lovely Clematis glauca, with glaucous
foliage and top-shaped, yellow, passing to bronze-coloured
flowers, ought not to be overlooked. The shingly and sandy
foreshores are covered with Willow, Sallowthorn, and False
Tamarisk (Myricaria germanica). In the Tung Valley, between

A PRICKLY PEAR (OPUNTIA DILLENII) NATURALIZED

THE CHINO-THIBETAN BORDERLAND 155

4000 and 5000 feet altitude, a “ Prickly Pear’’ (Opuntia
Dillenit) has become naturalized. This American colonist
has made itself very much at home, covering many miles of
barren rocky slopes. It grows 6 to ro feet tall, and when
covered with its yellow or pale orange flowers is very orna-
mental. The edible nature of the fruit is well known to the
natives but is little esteemed. An extract obtained by boiling
the fleshy stems is locally employed as a supposed cure for
hemorrhoids.

Amongst the coarse grass and scrub, the dominant features
of these regions, a variety of showy herbs occur, nearlyall having
bulbous or thickened rootstocks in some form or other. To
garden lovers everywhere these valleys are of special interest,
inasmuch as they are the home of many beautiful Lilies.
Each of these valleys has species or varieties peculiarly its
own, which range up to about 8000 feet altitude, yet whilst
very local these Lilies are numerically extraordinarily abundant.
In late June and July it is possible to walk for days through
a veritable wild garden dominated by these beautiful flowers.
In the Min Valley the charming Lilium regale luxuriates
in rocky crevices, sun-baked throughout the greater part of
the year. It grows 3 to 5 feet tall, and has slender leaves
crowded on stems bearing several large funnel-shaped flowers,
red-purple without, ivory-white suffused with canary-yellow
within, often with the red-purple reflected through, and is
deliciously fragrant. In the Tung Valley, Mrs. Sargent’s
Lily (L. Sargentia), a taller growing species than the foregoing,
with broader leaves, having bulbils in the axils, equally hand-
some flowers of similar shape, but varying from green to red-
purple without and from pure white to yellow within, is very
abundant in rocky places among grass and scrub. The
flowers of this species are collected, boiled, and dried in the sun,
then minced, fried with salt and oil, and eaten in the same
way as preserved cabbage. The bulbs of the Tiger Lily
(L. tigrinum) and its elegant ally, L. Thayere, which are
white, are cooked and eaten. Several other Lilies abound in
these valleys, including the lovely L. Bakerianum and other
species not yet named.

A herb very common in the Tung Valley is Thalictrum

156 A NATURALIST IN WESTERN CHINA

dipterocarpum. This Meadow-rue grows 6 to 8 feet tall, has
elegant, much-divided foliage, and multitudinous, large, lavender-
purple flowers—by common consent the handsomest member
of its family. In the Min and Tung Valleys, but very local,
Incarvillea Wilsonit, which grows nearly 6 feet tall and has
handsome flowers very like those of J. Delavay1, occurs. This
plant is monocarpic and has not yet flowered in cultivation,
although I introduced it into the Veitchian nurseries as long
ago as 1903. Salvia Przewalskii, with large purple flowers,
is another striking herb common in the valleys above 8000
feet altitude. This list of ornamental herbs could easily be
extended if any useful service would be served thereby. On
bare rocks various species of Selaginella abound; the Mullein
(Verbascum Thapsus), Deadly Nightshade (Hyoscyamus miger),
and Thornapple (Datura Stramonium) are common weeds by
the wayside. The poisonous properties of the two last named
are well known to the natives. From this brief sketch it will
be seen that these narrow, dry, almost desert-like valleys, with
their abnormally warm climate, possess a flora which, if limited
in number of species, contains many plants of more than
passing interest and horticultural value.

As mentioned earlier (p. 149), this hinterland is peopled
by various independent and semi-independent tribes about
which little is known. The whole region is analagous with
that separating India and Thibet, and this statement of fact
will perhaps convey a more intelligible idea than the most
voluminous details. These tribes are divisible into four
distinct groups, in accordance with their official status and
form of government.

I. States independent, non-tributary, hostile to both
Chinese and Lama authority, as the Lolo kingdom. I have no
intimate acquaintance with the Lolo—a people once spread
over much of Yunnan, but now relegated to the region of the
Taliang shan, where they have never been conquered by the
Chinese. This race possesses a written language peculiar to
itself and is probably indigenous.

2. States really independent and even hostile toward China,
directly controlled by the Dalai Lama and Council, whose policy
is supposed to be modified by High Commissioners appointed

THE CHINO-THIBETAN BORDERLAND 157

by China, as, for example, Chantui, Derge, and Sanai. The
territory occupied by these tribes is west of the Yalung River
and contiguous to that of Thibet proper; the people are in-
distinguishable from those inhabiting anterior Thibet generally.
These more western regions have been styled the ‘‘ Thibetan
Marches.” Some four years ago, an acting viceroy of Szechuan,
one Chao Erh-féng, was appointed Warden of these Marches.
With an army of Chinese soldiers he indulged ina most aggressive
policy and speedily subjected the whole region to Chinese con-
trol. He broke the Lama power, destroyed the principal
lamaseries, and beheaded the abbots and other dignitaries.
His task was rendered fairly easy owing to affairs in Lhassa,
consequent upon the British expedition to that city, and the
flight of the Dalai Lama, the whole making impossible any
concerted action by Lhassa De in support of their adherents
in the Marches. (In 191r Chao Erh-féng was appointed
Viceroy of Szechuan and was subsequently murdered in
Chengtu city by Chinese revolutionists.)

3. States tributary-controlled, governed by hereditary
native princes and subject to the Viceroy of Szechuan in
temporal affairs, but more or less strongly influenced by the
Dalai Lama, owing to Lamaism being the accepted religion.
Of these the kingdom of Chiala, the Horba states, and the
Chiarung tribes are the chief. They occupy most of the
territory between the Min and Yalung Rivers north of a line
connecting Yachou with Tachienlu and Hokou. The Chiala
kingdom I shall deal with separately when describing Tachienlu,
the capital city. The Chiarung are dealt with in the next
chapter.

4. A number of very small states, governed by quasi-
independent chiefs, indirectly controlled by Chinese officials
appointed for that purpose and by the surrounding tributary
kingdoms. They are, in fact, tiny buffer areas very useful to
the Chinese in maintaining the balance of power among the
larger, more independent kingdoms. Many of these princi-
palities are made up of people who may reasonably be looked
upon as remains of the aboriginal population of parts of
Szechuan and this hinterland. These petty states are
scattered through the more easterly parts of this hinterland

158 A NATURALIST IN WESTERN CHINA

from Mao Chou in the north through the Chiench’ang Valley
to borders of Yunnan province. The power exercised by
the chiefs varies according to their proximity to thickly
populated Chinese districts or otherwise. In the former it
is almost nominal, whereas in the latter case it is very
considerable.

In addition to the above are certain feudal states whose
overlord owes his office directly to Chinese influence, and who
is bound, if called upon, to render military service to China.
These feudal chieftainships are hereditary and were originally
bestowed as rewards for assistance rendered to the Chinese in
breaking up the Chiarung confederacy during the reign of the
Emperor Kienlung. Many of these, for example the Tsa-
ka-lao chief, have very considerable power and influence in
the temporal affairs of the surrounding tributary-controlled
kingdoms. The people are mainly of the same stock as
the Chiarung tribes. All the chiefs of these feudatory
states and tributary kingdoms are closely related by inter-
matriage.

The Chinese designate the inhabitants of this borderland
“‘ Mantzu,” a contemptuous term signifying “‘ Barbarian ’”’ and
of no ethnological value whatever. But the policy they have
pursued in dealing with these people has been shrewdly wise if
unscrupulous. With arms and money the Chinese have dis-
played their power and obtained what practically amounts to a
suzerainty over the whole borderland. A former emperor
said: ‘‘ Wardens of the Marches should seek to checkmate the
native tribes by becoming intimately acquainted with them and
their customs and thus able to prevent any united action. In
this way the tribes will remain weak and easy tomanage. They
should be encouraged to appeal to Chinese authorities for
advice and protection in their disputes with one another.
These authorities will, of course, be in no hurry to settle their
cases. If the tribes are taught to fear the Chinese, and the
officials act with energy, all trouble will be avoided.” This
crafty advice has long been acted upon by the Chinese, with
much success from their own view-point.

From this brief and very incomplete general account it
may be gathered that this hinterland is a fascinating region,

THE CHINO-THIBETAN BORDERLAND 159

presenting ethnological and other problems of great interest,
the solution of which is worthy of the attention of Western
scientists. It is hoped that a properly equipped expedition
will at no very distant date be organized and dispatched to
survey and investigate fully this little-known Chino-Thibetan
borderland.

CHAPTER XIII
THE CHIARUNG TRIBES

THEIR History, MANNERS, AND CUSTOMS

ITHIN the limits of the Chino-Thibetan borderland,
\ \ / asdefined in Chapter XII, from Sungpan Ting south-
wards to Yachou Fu, and west to the valley of the
Upper Tung or Tachin (Great Gold) River, the territory is divided
amongst numerous cognate tribes collectively spoken of by
Chinese as ‘“‘Chiarung.”’ These people are essentially agricul-
turists, making their homes in the upland valleys. They are all,
though tributary to China, ruled by their own hereditary
chiefs; each tribe occupies a properly defined area, with its
own capital town, the political centre of the entire region being
Monkong Ting. These tribes are non-Chinese and are not
indigenous to this region. They are also distinct from the
people found in anterior Thibet. They speak a difficult and
at first sound unpronounceable jargon, which, if it be the mother
of Thibetan dialects, is widely different from that spoken in
Thibet to-day. But Thibetan letters have, without difficulty,
been applied to it, and scholars, priests, officials,and merchants
both read and speak the Lhassa-Thibetan language with greater
or less fluency.

The origin of these people is obscure, yet there is good reason
to believe they come originally from the region around the
head-waters of the Tsang-po (Upper Brahmaputra River),
and probably have common origin with the people of Nepal
and Bhutan. Personally, I am of the opinion that they came
over with Genghis Khan, or his sonOk-Ko-Dai, at thecommence-
ment of the thirteenth century, and assisted in the conquest of
western Szechuan. As a reward for military services rendered

160

THE TACHIENLU RIVER

THE CHIARUNG TRIBES 161

they were given the territory they occupy to-day. During
the course of time they waxed powerful, menacing the territory
to the east of the Min River, and even taking possession of
certain parts. In Ming times the Chinese made war with them
on many occasions. They were a source of trouble to the
Manchu dynasty until the famous Emperor Kienlung deter-
mined upon crushing their power. After a very fierce struggle
this was accomplished by a Chinese general named A-kuei.
First he subjugated the region of the Hsaochin Ho (Little Gold
River), then, after much difficulty, he captured Lo-wu-wei
(modern Hsuching), the capital of the Tachin Ho (Great Gold
River), took the king prisoner, and made a map of the entire
region. The king, named Solomuh, was sent to Peking, where,
after a grand court ceremony, he was sliced to pieces. The
conquest was completed early in A.D. 1775. Military colonies
were then established by the Chinese in strategic places, the
more fertile regions were confiscated, and Chinese settlers
induced to take possession. In crushing this confederacy
the Chinese were assisted by the tribes, being to some extent
divided amongst themselves. Some of them fought on the
Chinese side, and as a reward certain areas situated at strategic
points were fiefed out and established as feudal states for
the benefit of these allies, an overlord with hereditary control
being appointed to each. The Chinese handled this campaign
with consummate skill, and the administrative system estab-
lished has remained unchanged down to the present day.
The power of the tribes was completely broken; and the feudal
states and the military colonies have safeguarded the Chinese
from any concerted action on the part of these people ever
since. It will, however, be readily understood that the tribes
farthest removed from regions fully occupied by Chinese enjoy
to-day greater independence than those in close contiguity.
Originally these ‘‘ Chiarung ’’ had one common language,
but time, isolation, and the dividing up into clans has produced
many very dissimilar dialects. These people are now split
up into eighteen tribes, occupying very unequal areas of terri-
tory, and though all are interrelated by marriage they are by
no means at peace with one another. Feuds are constant,
and fighting among themselves is very much the rule. Since
VOL. I.—II

162 A NATURALIST IN WESTERN CHINA

this keeps them weak in power the Chinese policy is to
intervene as seldom as possible. On the map are indi-
cated as accurately as our knowledge admits the positions
occupied by some of these tribes and feudal states. It is
almost impossible to render into English the guttural sounds ~
denoting the names of many of these tribes. But, fortunately,
the more important, namely, Mupin, Wassu, Somo, Damba,
Bati-Bawang, Wokje, are the least difficult to pronounce.
The whole territory occupied by these people is about 250
miles from north to south, and 200 miles east to west at broadest
point. The population is about half a million.

Two main roads, one from Kuan Hsien, the other from
Lifan Ting, cross this region and unite near Monkong Ting.
In addition, a network: of cross-country by-ways connects
the various villages and states.

The Chiarung are essentially agriculturists, cultivating
with much skill crops of wheat, barley, peas, buckwheat,
maize, Irish potato, and miscellaneous vegetables. Sheep,
cattle, ponies, and goats are kept by the more wealthy, often
in quantity. The horses are sold to Chinese traders, but the
wool is woven into cloth for their own use. Milk, butter, and
meat enter largely into their diet. They are also skilled gun-
and swordsmiths, more especially the Somo people, who manu-
facture most of these weapons in use among the tribes them-
selves and the people of eastern Thibet generally. Many are
also highly skilled masons, builders, and well-sinkers, and as
such have a reputation even amongst the Chinese. During
August many visit the upper reaches of the Min River every
year to take part in harvesting the crops; indeed, for this
purpose they supply most of the labour in that region. Often
they are in request in Chengtu and other cities for sinking wells
and such-like work.

The Chiarung live in settlements of from several to a
hundred families or even more, always in positions admirably
suited for defence. These settlements usually crown some
bluff or eminence ; very often they are perched like an eagle's
aerie high up on the steep mountain-side. The architecture
which obtains throughout is characteristic and peculiar. Each
settlement is dominated by one or more tall, chimney-like

IMA-VL TO GOVITIA ALMOM AHL

THE CHIARUNG TRIBES 163

towers, either square, hexagonal, or octagonal in shape, 60 to 80
feet high, and resembling from a distance the stack of some
large factory in Western lands. The exact significance of these
towers it is difficult to fathom, but it is evident that they can
serve as storehouses, watch-towers, and harbours of refuge in
times of stress and war. They have also some obscure con-
nexion with religious matters, possibly in this they have some
remote affinity with the pagodas of China and Burmah. The
houses are more or less square, flat roofed, solidly built of
shale-rock and mud. Those belonging to the chiefs and men
of property are three or four stories high. The walls are thick,
pierced with loopholes and several narrow latticed windows.
At all four corners of the roof turrets 3 to 4 feet high are built,
sometimes there are more in different patterns. From these
prayer-flags are displayed, often with the green branches of
Juniper. On the roof also is fixed an incinerator for the sacri-
ficial burning of fragrant juniper branches as incense. Part
of the roof is frequently occupied by a hurdle-like framework
called ‘‘ Kai-kos,”’ 10 to 15 feet high, which is employed for
drying grain upon. The rest of the roof is used for religious
exercise, eating, sleeping, and recreation; in harvest-time it
serves as a threshing-floor. The ground story is made up of
a courtyard surrounded by sheep and cattle-pens, the kitchen,
and usually a guest-room.

The turrets, upper rim of the walls, edges of the window-
spaces, base and base angles of the walls, are washed white,
commonly white lines stretch diagonally up the walls, and the
swastika cross, with other devices and symbols, are displayed
in white on these walls. Crowning the edges of the roof, or
arranged on separate structures, symbols denoting a globe,
upturned crescent, and the swastika are commonly displayed.
The lamaseries are similarly constructed, only larger, and
usually with more stories. The houses of the peasants also are
on the same plan, but of one or two stories only. All these
structures are closely packed together with one to several
towers reared above the whole assemblage. The different
emblems and symbols of Nature worship may occur in the
structure of Thibetan houses and lamaseries, but the tall
tower is peculiar to the Chiarung.

164 A NATURALIST IN WESTERN CHINA

Another interesting feature of these regions is the bridges.
All these structures are of designs differing from those found
throughout China proper, but agreeing closely with those in
use throughout Sikhim, Bhutan, and Nepal, thus furnishing
additional evidence of the affinity of these peoples. All
the smaller streams and torrents are bridged by logs arranged
on a semi-cantilever principle, and call for no special remark.
But the larger streams are crossed by suspension bridges con-
structed of split and plaited bamboo cables. These bridges
are very similar to the cane bridges of Sikhim and Bhutan.
They are found throughout the territory occupied by these
tribes and the narrow strip of territory wedged in between the
Min Valley and the western limits of the Red Basin. This
latter strip was formerly occupied by these tribes, and is to-day
largely peopled by their descendants or half-caste Chinese. As
mentioned in Chapter X, pp. I17, 130, iron suspension bridges
occur in one or two places in the north-west corner of Szechuan.
This style of bridge is common from the valley of the Ya River,
and the Tung at Luting chiao, southward to the frontier of
Burmah, and is probably of Shan origin. Similar bridges of
iron rods and chains are met with in Bhutan, where they are
considered to be of Chinese origin (White, Sikhim and Bhutan,
p. 191). Throughout the Chino-Thibetan Borderland iron and
bamboo are equally common, yet it is a singular fact that their
use in bridge-building is restricted to definite areas.

Cable or rope bridges are abundant throughout the entire
region, and extend much farther west and south than the Chia-
rung territory. These simple but extremely useful structures
consist of a bamboo hawser stretched across the stream usually
from a higher to a lower point; if the stream is moderately
narrow the question of incline is of less importance. The
hawser may be anything from 8 inches to 1 foot thick, and
being heavy sags considerably in the middle, unless the stream
is very narrow, as around Tachienlu, where a rather different
method of crossing than that about to be explained is in vogue.
To cross one of these cable bridges a person is supplied with
a length of strong hempen rope hanging free from a saddle-
shaped runner of oak or some other tough wood. The runner
clips the cable, and the hempen rope is fastened under and

THE CHIARUNG TRIBES 165

around the legs and waist to form a “cradle.’” When all is
properly secured the person throws one arm over the top of
the runner, gives a slight spring, and glides down the inclined
cable at increasing speed. The impetus obtained in the down-
ward rush carries the passenger over the central dip and more
or less up the lesser incline on the opposite side. If the
momentum is insufficient to land the person, the remaining
distance has to be traversed by taking hold of the hawser and
hauling hand over hand. Crossing these bridges is fearsome work
until one is accustomed to it. It is speedily accomplished, and
there is practically no danger so long as one keeps a cool head
and the ropes do not break. It is a common sight to see men
with loads and women with children on their backs cross these
bridges. But heavy loads are usually fixed to the runners and
hauled across by a rope attached to them.

None of the rivers traversing Chiarung territory is navigable
in the ordinary sense of the term, but skin coracles, broadly
oval in shape, descend certain stretches of the Upper Tung
River. These frail boats serve also to ferry over goods and
passengers at certain necessary places. They are made of
cattle hide stretched over ribs of tough, light wood. The whole
coracle is very easily carried by one man, and closely resembles
pictures of the boats used by ancient Britons prior to the
Roman invasion. They are steered by a man seated in the
stern operating a paddle, and accommodate about two
passengers. A passage down or across stream in one of these
coracles consists very largely in describing, more or less rapidly,
a series of wide circles and half circles. As a novelty, pro-
ductive of excitement, not unmixed with danger, these coracles
and cable bridges can with confidence be recommended to
“ World’s Fair’ promoters and showmen generally. The skin
coracle is in general use at ferries throughout Eastern Thibet
and the Marches, and is not strictly a Chiarung specialty.

In height the tribesmen average about 5 feet 7 inches or
rather more; the face is usually oval, with rather pointed
chin, straight nose, sometimes inclining towards aquiline.
They dress ordinarily in undyed serge cloth of local make,
worn in the same manner as that of the Sifan. The legs
are swathed in felt putties; the head-gear is either a turban

166 A NATURALIST IN WESTERN CHINA

or black pudding-basin-shaped felt hat. Those living near
Chinese settlements and the highways have their head in part
shaven, and wear their hair in a queue Chinese fashion. On
holiday occasions their garments are brightened with red
bordering, and high felt boots are worn. The women are
short in stature (about 5 feet), sturdy and buxom, somewhat
gipsylike, with dark olive complexions, and when young are
often good looking. Their ordinary outer dress is a garment
of grey native serge of no definite shape, reaching to just
below the knee and bound around the waist with a scarf.
The legs and feet are bare or encased in top-boots. Commonly
they go bare-headed with their long black hair parted down
the middle and hanging down the back in one large plait.
They are fond of large bangles, ear-rings, etc., made of silver
inlaid with turquoise and coral. On festive occasions
garments edged with red and very often made of blue cloth
are worn. The more wealthy dames decorate themselves very
lavishly with silver ornaments, and wear covering their heads
a piece of cloth held down by means of their large plait of hair,
which is wound around and decorated with silver and beads
of coral and turquoise ; the lower part of the piece of cloth
hangs free over the back of the neck and shoulders. These
dames are women of character, and have a ruling voice in
household and family matters generally; also, from what I
saw of them, they appear to conduct most of the business.
These women lead a strenuous life; they cultivate the fields,
tend the flocks, take the farm produce to market, hew wood,
and carry water. The domestic duties of cooking, making
and mending clothes and general household work devolve
upon the men. Yet the women are not unkindly treated,
and are far from being down-trodden. Being of cheerful
disposition, they seem well suited to the free outdoor life they
lead, and laugh and sing as they ply their task. Among
themselves these people are frank and easy in manner, and the
women enjoy a freedom of position unknown amongst the
Chinese. A party of dames and men were fellow-travellers
with me once for a couple of days. When the time came to
separate they made merry over cups of wine; the women
officiated, and cordially invited me to join them. With their

HTdWHL VdNOd V NI S'IOCI

Dg Gene
Sea

oe
tS ~

Yv¢

THE CHIARUNG TRIBES 167

laughter and song they made cheery companions, and I was
sorry to part from them.

The families are small, but the children are usually strong
and healthy. Girls marry between the ages of seventeen and
twenty, polygamy is common, but polyandry is unknown except,
perhaps, in the upland regions bordering Thibet proper.
Temporary marriages, so general in Thibet, are also unknown
amongst the Chiarung. Nevertheless, the standard of morals
in vogue among these people is a very low one. In certain
states hetairism precedes maternity. In Badi-Bawang the
unmarried girls and childless women wear only two sporran-like
fringes of woollen threads or pieces of fur, suspended from a
girdle passed around the body above the hips. The legs are
exposed, but the upper parts of the body are usually covered
by a coarse serge garment. Only after their first child is born
may they wear skirts, since the gods have then purified them.
A pregnant damsel selects from among her lovers a husband,
who thus becomes the accepted father of her child, her word
in this matter being final. Maternity alone ratifies marriage,
and indeed saves women from promiscuity. The defloration
of virgins is the prerogative of chiefs and head-men, but is not
always exacted. In many ways these people are apparently
shameless, according to Chinese and Occidental ideas alike.
It is no uncommon sight to see women of all ages, quite nude,
bathing in streams by the wayside. This same custom is
also common at Tachienlu, where the hot springs are favourite
bathing-places for both sexes. But after maternity the women
are said to remain constant ; divorce or legal separation after
ratified marriage are not practised.

The explanation of the above and other curious customs
of these interesting people is found in their religious beliefs.
Although orthodox Lamaism is more or less paramount the
mysterious Bonpa religion, with its marked tendency toward
phallic worship, lurks throughout the lonely valleys of the
Chiarung tribes. In Badi-Bawang it is the recognized state
religion. It should also be remembered that these regions
constituted the famous matriarchal kingdoms of Chinese
historians. Indeed, even to-day, certain states have queens
holding nominal or actual authority, and in these in some

168 A NATURALIST IN WESTERN CHINA

capacity a woman must always rule. Occasionally the
difficulty is overcome by styling the ruling head a “ Queen ”’
quite irrespective of sex !

Lamaism appears in three forms, the Yellow, Red, andBlack,
the latter representing the Bénpa cult. The religious centre
is Tsong-hua on the Tachin River, about 60 miles west of
Monking Ting. But lamaseries are scattered over the land,
occurring separately by themselves or in association with the
residences of the hereditary chiefs. The Yellow or orthodox
sect is first in importance and numbers, and is controlled directly
from Lhassa. The ritual differs in no way from that practised
throughout the hierarchy of Lamaism. The same remark
applies to the unorthodox Red sect, which is of much less
importance, and whose priests are allowed to marry.

The Black or Bénpa sect has a ritual bearing an outward
resemblance to orthodox Lamaism, but apart from this there is
little else in common. In many things the Bonpa are the avowed
enemies of the orthodox. They turn their praying wheels
from left to right instead of from right to left; they pass
sacred objects on the right instead of on the left; also they
refuse to repeat the mystic Mantra, ‘‘Om mani padmi hom,”
replacing it with one peculiarly their own. As to the origin of
this Bonpa it is difficult to say. My friend, Mr. J. Hutson
Edgar of the China Inland Mission, who has travelled among
and studied these Chiarung tribes more closely than any one else
living, inclines to regard it as the remains of the old Nature
worship of Thibet, which probably underlies all the religious
systems of Eastern Asia.

In the state of Wassu are several temples belonging to this
Boénpa sect. Through the courtesy of the chieftain I was
allowed to inspect some of these temples, and succeeded
in obtaining fair photographs of the idols. These latter,
made of stone, wood, straw, and plaster, represent giants
and demons with their female energies; the walls are
decorated with paintings depicting erotomania. Hideous and
disgustingly obscene are the contents of these temples, where
phallic worship holds unblushing sway. The Wassu chief
informed me that the Mantra used by these Bénpa priests is
“Hom ma-te ma-tsi ma-yéor tsa-lien doo.” He kindly gave

THE CHIARUNG TRIBES 169

me a copy of this hymn, but I have not yet succeeded in getting
it translated into intelligible English. The principal symbol
in use is the Fylfot or swastika, which they call “ Yungdrung.”’
A mystical bird, ‘‘ Chyong”’ or ‘‘ Garuda,” is also regarded
with great favour as an emblem of fruitfulness. In the Bonpa
temples at Tung-ling shan, near the residence of the Wassu
chief, I also recognized the image of Kwanyin (Goddess of
Mercy), the God of Wealth, and many demons similar in
appearance to those found in ordinary Buddhist temples
throughout China proper. It would thus appear from the
catholic nature of the contents of their temples that these
people accept a measure of Buddhism, and Lamaism both
orthodox and unorthodox, and the Bonpa in its entirety. An
atmosphere of secrecy and mystery enshrouds the Bénpa
temples, which are frequently built in places difficult of access.
The cult has been subjected to much persecution at the hands of
Lamaists, yet, notwithstanding, it retains a firmer hold on the
people of most of the Chiarung states than any other form of
religion. In their hearts children of nature, their daily life
one constant struggle against an inhospitable soil and climate
to win a crop necessary for their sustenance, these people very
naturally incline most toward the gods of Increase and
Fecundity.

CHAPTER XIV
ACROSS THE CHINO-THIBETAN BORDERLAND

KuAN HSIEN TO ROMI CHANGO; THE FLORA OF THE
PAN-LAN SHAN

] Jims the summer of 1908, when in Chengtu, I de-

termined upon a journey to Tachienlu. Previously, in
1903 and again in 1904, I had visited this town by three
different routes. This time I decided upon following the road
leading from Kuan Hsien via Monkong Ting and Romi Chango.
The only published account of this route that I have knowledge .
of is in a Report by Mr. (now Sir) Alexander Hosie,* erstwhile
H.B.M.’s Consul-General at Chengtu, who returned from
Tachienlu over this road in October 1904. What is written in
this Report about the forests of that region created a desire
within me which nothing short of actual experience could
satisfy. Again, this route promised further acquaintance with
the tribesfolk inhabiting the hinterland. Sir Alexander’s
description of the road portrayed a difficult journey, but I
felt sure that by taking time and but lightly burdening my
men I could get through all right. This confidence was fully
justified, as events proved, and what I saw of the forests and
mountain scenery, together with the quantity and variety of
plants discovered and collected, abundantly repaid me for the
hardships experienced. The journey is estimated at 1326 h,
approximately 330 English miles, but, whilst mere mileage is
of little moment in mountainous countries, I should consider
250 miles a more accurate figure.
With Tachienlu as my goal I left the city of Chengtu on
the morning of 15th June, and at noon the next day reached

1Journey to the Eastern Frontier of Thibet, presented to both Houses of
Parliament, August 1905.

170

LONG

TO) NDS:

rik

RID¢C

OO SUSPENSION

AMB

re ge ne NRC
at Shee. oe Verb OP

ACROSS THE CHINO-THIBETAN BORDERLAND 171

the city of Kuan Hsien. An afternoon sufficed to complete
my arrangements. The caravan consisted of eighteen carrying
coolies and one head coolie, two chairs, two handy men, an
escort of two soldiers, my Boy, and self, making a party of
thirty all told. The journey occupied twenty-three days from
Kuan Hsien.

What follows is compiled from my diary :—

The famous bamboo bridge, known as the An-lan chiao,
over which the road to Monkong Ting passes, was having its
annual overhauling; in consequence, on leaving Kuan Hsien we
had to journey down stream some 5 li to a point where it was
possible to cross the various arms of the Min River by im-
provised bridges and ferry. Inso doing we had an opportunity
of realizing, somewhat hazily be it confessed, what this area
must have been like before Li-ping’s wonderful irrigation works
came into existence. Without counting the streams flowing
Chengtu-wards we crossed five distinct arms of the Min River
proper scattered over an area a mile wide, covered with sand,
shingle, and coarse grass (Miscanthus sinensis). The detour
involved 15 li, and it was not until 9 o'clock that we were
opposite the An-lan chiao. This most remarkable structure
is about 250 yards long, 9 feet wide, built entirely of bamboo
cables resting on seven supports fixed equidistant in the bed of
the stream, the central one only being of stone. The floor of
the bridge rests across ten bamboo cables, each 21 inches in
circumference, made of bamboo culms, split and twisted
together. Five similar cables on each side form the “ rails.”
The cables are all fastened to huge capstans, embedded in
masonry, which are revolved by means of spars and keep the
cables taut. The floor of the bridge is of planking held down
by a bamboo rope on either side. Lateral strands of bamboo
keep the various cables in place, and wooden pegs driven
through poles of hard wood assist in keeping the floor of the
bridge in position. Not a single nail or piece of iron is used in
the whole structure. Every year the cables supporting the
floor of the bridge are replaced by new ones, they themselves
replacing the “‘rails.’”’ This bridge is very picturesque in
appearance, and a most ingenious engineering feat.

From the An-lan chiao the road ascends the right bank

172 A NATURALIST IN WESTERN CHINA

of the Min River, and is broad, in good repair, but with many
awkward gradients. We found lodgings for the night at
Hsuan-kou, alt. 2640 feet, a market village of some 300 houses,
situated on a tributary immediately above its union with the
main stream, which describes a very sharp turn on leaving a
narrow gorge. The Min River from the An-lan chiao to this
point is full of minor rapids, and the current is very swift.
Near Hsuan-kou timber is made into rafts and floated down to
Kuan Hsien, thence to Chengtu and elsewhere.

During the day’s march we passed some good-sized trees of
Black Birch, Nanmu (Machilus spp.), Hog-plum (Spondias
axillaris), and small trees of Cryptomeria japonica, the latter
obviously planted. A large trumpet-flowered Lily was
abundant in rocky places by the wayside. Rice occurred
sporadically, but the principal crop was maize. Around the
inn Tea-bushes are abundantly planted.

On leaving Hsuan-kou we crossed the tributary by a small
bamboo suspension bridge, and ascended the left bank by an
easy road for 30 li to Shui-mo-kou. Throughout this stretch
Cryptomeria iscommon. All the trees are small and obviously
planted, yet I cannot rid myself of the idea that it must be
indigenous somewhere in this vicinity. It occurs scattered
over a large area, always near habitations, yet it is scarcely
feasible to suppose that this tree has been brought from Japan
for the purpose of planting it hereabouts.

Shui-mo-kou is an ordinary Chinese market village of some
350 houses lining either side of the main street. It is interest-
ing, however, as being the last purely Chinese village in this
direction, also the last place wherein supplies can be purchased
or silver exchanged until Monkong Ting is reached. I hired
an extra man, and all my followers laid in a stock of rice and
food-stuffs generally. At Kuan Hsien, appreciating fully the
difficult road before us, I had reduced all loads to two-thirds
the normal weight. In spite of this the carriers were heavily
laden with extra supplies, and could hardly stagger along on
leaving Shui-mo-kou.

A short distance beyond the above village there is a steep
ascent, but after a few li the road becomes easy and winds
around the mountain-side. Scrub Oak and unhappy-looking

NOMNVOSH JO HOVITIA AHL

ACROSS THE CHINO-THIBETAN BORDERLAND 173

trees of Cunninghamia are abundant, but the flora generally
is poor. Wild Strawberries cover the more grassy slopes, and
were laden with white and red luscious fruit. We passed a few
houses, and finally reached the top of the ridge, alt. 5600 feet,
which is known as the Yao-tszeshan. Crossing over we entered
the territory under the jurisdiction of the Wassu chieftain,
who resides at Tung-ling shan, near Wénch’uan Hsien in the
Min Valley.

Descending by a path, which at first easy soon becomes
very precipitous and difficult owing to the abundance of loose
rocks, we reached Hei-shih ch’ang, our destination for the day,
at 6 p.m. In this descent, near the head of the pass, the
“Yang-tao”’ (Actinidia chinensis) is abundant, and was laden
with a wealth of large, white, fragrant flowers. By the wayside,
Rosa microphylla is very plentiful, and bushes 2 to 4 feet tall
were covered with large pink blossoms. One small tree of
Carrieria calycina, laden with curiously-shaped, waxy-white
flowers borne in erect panicles, was also worthy of note. But
the flora generally has been destroyed to make way for crops
of maize, oats, and pulse.

Hei-shih ch’ang, alt. 4000 feet, is considered to be 60 li
from Hsuan-kou, and consists of three or four houses, situated
in a ravine alongside a torrent, with wild mountains on every
side. Our lodgings were roomy, and the people both courteous
and attentive.

Rain fell heavily next morning when we started out, but
ceased about 9 a.m. ; the weather remained dull the rest of the
day until 4 p.m., when rain recommenced to fall and con-
tinued far into the following night. Crossing the torrent by
means of a covered wood bridge the road immediately ascends
a steep mountain called Che shan from the abundance of
Varnish trees growing thereon. The ascent, though very steep,
is short, and afterwards for the next 30 li the road skirts the
mountain-sides until the summit of the Chiu-lung shan is
reached. Descending this ridge it ultimately enters a narrow
grassy valley. Here we found lodgings for the night in the
solitary hostel of Hoa-tzu-ping, alt. 6100 feet, having covered
50 li during the day.

Until reaching the valley the country generally was either

174 A NATURALIST IN WESTERN CHINA

under maize or covered with a dense jungle. The flora was of
passing interest only, being similar in character to that found
everywhere in western Szechuan between 4000 and 6000 feet
altitude. The more interesting shrubs collected were a yellow-
flowered Schisandra, a white-flowered Clematoclethra, and the
Yunnan Holly (Ilex yunnanensis) with small, neat leaves,
clusters of purplish, fragrant flowers, and hairy shoots.
Actinidia Kolomikta, a large climber with white, fragrant
flowers and added beauty in the shape of a multitude of white
leaves, is excessively common. Nearly all the species of
Actinidia and the allied genus Clematoclethra, other than those
clothed with rufous hairs, have these white leaves, which
usually become pinkish as the season advances. All the species
are handsome climbers, and the majority bear very palatable
juicy edible fruit.

The trees of this region, though not numerous or of any
great'size, include such remarkable subjects as Davidia, Ptero-
styrax, Tapiscia, Tetracentron, Beech, and Horse Chestnut.
Occasional trees of Cornus kousa occur, and were a wealth
of white flower-heads enlivening the country-side. Walnut
trees are common around houses and wild strawberries by
the wayside. In the grassy valley the beautiful Ilex Pernys
occurs with Rodgersia esculifolia and Lilium giganteum in
quantity. Around Hao-tzu-ping odd patches of maize are
cultivated, but where clearings have been made the ground
is mostly covered with grass and coarse herbs.

During the day we met many men laden with huge logs
of Teih-sha (Tsuga, Hemlock Spruce) and Hung-sha (Lamx,
Larch) timber. These logs were dressed, and carried on a
wooden framework. I measured one with a tape; it was
18 feet 6 inches long, 7 inches thick, and 9 inches broad. It is
astounding how such loads are carried over vile mountain
roads. As fellow-travellers during the day we had some
tribesmen in charge of a small mule caravan of tea, bound
for the state of Wokje.

After leaving Hoa-tzu-ping we soon reached the head of
the valley which merges into a narrow jungle-clad ravine.
After a precipitous climb of 30 li we reached the summit of
the Niu-tou shan, alt. 10,000 feet, where dense mists blotted

ACROSS THE CHINO-THIBETAN BORDERLAND 175

out the landscape. A similarly precipitous descent of 20 li
brought us to Chuan-ching-lou, where we put up for the night.
The flora was very interesting, but owing to a thick pall of
mist I was able to observe only the plants immediately along-
side the pathway. Perhaps the commonest shrub of the day
was Salix magnifica, which is abundant everywhere, but more
especially near the watercourses. This extraordinary Willow
has leaves up to 8 inches long and 5 inches wide, with
catkins 1 foot or more long. It forms a straggling bush 5 to
20 feet tall and, except when in flower or fruit, would scarcely
be taken even by the closest observer for a Willow. (I first
discovered this plant in 1903, and in 1908 succeeded in intro-
ducing living plants into cultivation.) Many other kinds of
Salix, varying from prostrate shrubs to small trees, occur on
the Niu-tou shan ; indeed, this mountain is remarkable for its
wealth in Willows (subsequently I succeeded in introducing
into cultivation about a dozen species from this locality).
The Actinidia and Clematoclethra previously noted again
very abundant. Clematis montana, var. grandiflora, with large
white flowers, was a pleasing sight; so also was a Deutzia
(D. rubens) with pretty rose-tinted flowers. I saw no deciduous
broad-leaved trees of any size, but herbs were luxuriant every-
where, especially the Rodgersia, which covers acres of the moun-
tain-side. The Conifers were the most interesting plants of the
day. In the ascent, save for odd trees of Silver Fir and
Yew, I saw nothing but Hemlock Spruce. This tree delights
in rocky country, clinging to the cliffs in a most remarkable
manner. In the descent, however, Silver Fir, Spruce, Larch,
Hemlock, and White Pine all occur, but the trees are being
rapidly felled, and no large specimens were to be seen. From
this place come the logs of timber noted yesterday. The
Larch (L. Mastersiana) is first met with below T’ang-fang,
alt. 9400 feet, where it is common more especially to the right
of the road, and descends to 7200 feet altitude.
Chuan-ching-lou, alt. 7000 feet, 50 li from Hoa-tzu-ping,
consists of one large, dirty hostel, and three other houses, situ-
ated in a narrow ravine, walled in by lofty mountains. A noisy
torrent which descends from the Niu-tou shan flows past the
inn, and vegetation is rampant on all sides. The road over

176 A NATURALIST IN WESTERN CHINA

the Niu-tou shan is difficult, and in many places dangerous.
Here and there steps have been cut in the hard rock to assist
the traveller, but in the main the road is strewn with loose
stones and boulders—vile to walk on or over.
We were unfortunate in the matter of weather, for it again
rained as we continued our journey. Following the torrent
through a narrow ravine for 5 li we reached Erh-tao chiao,
where the torrent connects with a very considerable stream
which flows from the Pan-lan shan. The united waters form a
river which, after traversing very wild country, joins with the
Min near the foot of the Niangtsze-ling on the Wénch’uan
Hsien side of the pass. Turning sharply to the left at Erh-
tao chiao we ascended the stream, which is called Pi-tao Ho,
and soon crossed over by a wooden semi-cantilever bridge to
the left bank. From this point the next 25 li to Wu-lung-kuan
is easy, going through a narrow valley where occasional houses
occur and a certain amount of cultivation obtains. Above
Wu-lung-kuan the road becomes increasingly difficult, and in
many places is execrable. The river is joined by numerous
lateral torrents, some of large size, and as the valley narrows
into a ravine becomes an untamable, roaring torrent. The
scenery, such as the mists permitted of our seeing, is savage
and grand. Here and there perpendicular cliffs of limestone
cropped out through the mists, their summits covered with
Pine trees. We crossed and re-crossed the torrent many times,
and after covering 65 li reached Ta-ngai-tung, which was our
destination for the day. This hamlet, alt. 7600 feet, consists
of one large hostel, which was in moderately good repair, and is
completely surrounded by steep mountains heavily clad with
mixed shrubs and small trees, the upper parts being covered
with forests of Conifers. The flora generally is very similar
to that of the Niu-tou shan, though scarcely as rich. All the
Conifers except Silver Fir are present, though Larch only puts
in its first appearance near the hostel. At Erh-tao chiao
I photographed a magnificent Juniper tree, 75 feet tall, 22
feet in girth, with graceful pendent branches, and a Black
Pine which retains its cones over many years. (It proved to
be a new species, and has been named Pinus Wilsonit). This
Pine is common on the cliffs, but White Pine (P. Avmandt)

ACROSS THE CHINO-THIBETAN BORDERLAND 177

is rare, although we passed the largest specimens of this tree I
have ever met with. Deuizia longifolia with lovely rosy-lilac-
coloured flowers, Spivea Henryi with yard long, flat sprays
of pure white, and Neillia longeracemosa with rose-coloured
flowers were perhaps the commonest shrubs in blossom.
Poplar is the only large deciduous tree hereabouts. Maple
is not uncommon, and near Ta-ngai-tung I gathered specimens
of a Black Birch having short, stout erect catkins.

Early next morning we continued our journey, spending
the whole day toiling up the ravine through wild and savage,
yet wondrous, scenery, with a profusion of vegetation on all
sides. Coniferous trees preponderate, the species being the
same as those previously mentioned, with a couple of new
Spruces added. Yew is less abundant, but Larch much more
so, though large trees are very scarce. To my astonishment
the Larch cones were ripe, and I collected a quantity of seed.
A Poplar with large leaves, silver-grey on the under side, is
very common, and we passed some very large specimens. A
Rose with large bright red flowers made a fine display, so
also did the pink-flowered Deutzia mentioned above. Two
Lady-slipper orchids (Cypripedium Franchetti and C. luteum),
with rosy-purple and yellow flowers respectively, occur, but are
rare. In the bed of the torrent Hippophaé salicifolia (Sallow-
thorn) is common, and varies from dwarf spiny bushes to
trees 25 feet tall, the long slender foliage silvery-grey below
forming a pleasing contrast to the brighter greens of surround-
ing trees and shrubs. Many kinds of Maple (Acer), Linden
(Tia), and Mountain Ash (Sorbus) are plentiful, and Tetra-
centron sinense, an interesting tree exceeding in size all other
deciduous trees of this particular region, occurs sparingly.
Hydrangeas, Spireeas, Honeysuckles, Mock-orange, Brambles,
Roses, Actinidia, Clematoclethra, Viburnum, and other orna-
mental shrubs struggle for possession of every available spot.
The variety and wealth of bloom was truly astonishing, and
I know of no region in Western China richer in woody plants
than that traversed during the day’s march.

The weather continued exasperatingly showery, but luckily
no great quantity of rain fell, otherwise the route would have
been impassable. Heavy mists limited our view, but whenever

VOL. IL—I2

178 A NATURALIST IN WESTERN CHINA

the clouds lifted we saw nothing but steep mountain-sides,
beetling crags or cliffs, bare here and there but mostly clothed
with mixed vegetation, giving place ultimately to forests
of Conifers. The road is vile beyond the power of language
to describe. In several places poles have been fixed hori-

zontally into holes made in the face of the cliffs and half-rotten |

planks laid on these to form a roadway. Such bridges as
exist are of logs, often rotten, and were always difficult to
cross. The river is simply a roaring torrent, cascading over
huge boulders and madly endeavouring to escape to less
savage regions. At one point it receives a torrent, which,
judging from the colour and temperature of the waters, evi-
dently comes down from eternal snows.

During the day we passed a few miserable hovels, but
there is no room for cultivation, and the people are wretchedly
poor. We stayed for the night at Yii-yii-tien, alt. 8800 feet,
42 li from Ta-ngai-tung, where there are two poor hostels.
These useful if squalid structures are all alike on this route,
being one-storied, constructed of wood, and roofed with shingles
held down by stones. A portion is sectioned off as private
quarters for the family in charge, and near by the kitchen is
located. A series of bunks is built around all sides of the
place, the central part being occupied by benches for the
accommodation of loads. Travellers furnish their own food-
supplies, since nothing is obtainable at the hostel except,
perhaps, some green vegetables in minute quantities. Shelter
for the night and a fire to cook food and dry clothing are all
these places afford. But the foreign traveller enjoys a welcome
quietude and freedom from curious crowds. A sound night’s
sleep rewards the labours of the day, and he awakens refreshed,
perfectly fit, and all eager to drink in more of the wondrous
scenery, the charm of woodland, crag, and stream.

At Téng-shéng-t’ang, 8 li beyond Yii-yii-tien, the ravine
widens out into a shallow valley, and the road boldly ascends
the grassy, scrub-clad mountains to the left of the stream.
Hereabouts Barberries in great variety luxuriate. After a
severe ascent we crossed over a shoulder, and for the rest of
the day skirted the side of a grassy ridge carpeted with
brilliantly coloured alpine flowers.

IMULA VEITCHII

ACROSS THE CHINO-THIBETAN BORDERLAND 179

The main stream takes its rise in some snowclad peaks,
of which we obtained a glimpse and a photograph, but a
considerable tributary flows down from the Pan-lan shan
Pass. The mountains to the right of this affluent, and also
to the right of the main stream, are forested up to 11,500
feet altitude with Spruce, Silver Fir, and Larch. The bed
of the valley is covered with bushes of Willow, Hippophe,
and Barberries. Up to 10,000 feet altitude Cypripedium
luteum is not uncommon on humus-clad boulders and in
the margins of woods.

The flora of the grassy ridge leading up to the Pan-lan shan
Pass is strictly alpine in character, and the wealth of herbs
was truly amazing. Most of the more vigorous growing had
yellow flowers, and this colour predominated in consequence.
Above 11,500 feet altitude, the gorgeous Meconopsis integrifolia,
which has huge, globular, incurved, clear yellow flowers,
covers miles of the mountain-side. Growing on plants from
2 to 24 feet tall the myriads of flowers of this wonderful Poppy-
wort presented a magnificent spectacle. Nowhere else have I
beheld this plant in such luxuriant profusion. The Sikhim cow-
slip (Primula sikkimensis), with deliciously fragrant pale yellow
flowers, is rampant in moist places. Various kinds of Senecio,
Trollius, Caltha, Pedicularis, and Corydalis added to the over-
whelming display of yellow flowers. On boulders covered with
grass and in moderately dry loamy places, Primula Veitchit
was a pleasing sight with its bright rosy-pink flowers. All
the moorland areas are covered so thickly with the Thibetan
Lady-slipper Orchid (Cypripedium tibeticum) that it was
impossible to step without treading on the huge dark red
flowers reared on stems only a few inches tall. Yet the most
fascinating herb of all was, perhaps, the extraordinary
Primula vinceflora, with large, solitary, violet flowers, in
shape strikingly resembling those of the common Periwinkle
(Vinca major), produced on stalks 5 to 6 inches tall. This
most unprimrose-like Primula is very abundant in grassy
places. The variety of herbs is indeed legion, and the whole
country-side was a feast of colour. Silence reigns in these
lonely alpine regions, a silence so oppressive as to be almost
felt and only broken on rare occasions by the song of some lark

180 A NATURALIST IN WESTERN CHINA

soaring skywards. We flushed an occasional Snow-partridge
and saw one or two flocks of Snow-pigeons, but bird-life
generally was extremely sparse. Save a few voles and mice we
saw no animals, but Bharal and Wolves were said to occur
here, the former in quantity.

After travelling 38 li we reached the hostel at Hsiang-
yang-ping, alt. 11,650 feet, and remained there for the night.
This place is part temple, part inn, and is kept by a priest, to
whose clothing and person water was evidently astranger. The
medicine Pei-mu (Fritillaria Roylet and other species) is
common in this region, and as fellow-guests for the night we
had a number of people engaged in digging up the tiny white
corms of this plant. Some Chinese traders also were there
buying up this medicine at 60 cash per ounce. In Chengtu
it is worth, wholesale, 400 cash per ounce, so their profit is
a handsome one. Among the medicine-gatherers were several
Wokje tribesfolk, about 5 feet 8 inches tall, sturdily built, with
straight noses and fearless expression. Two of their women
were with them, and had they been clean and decently dressed
they would have been decidedly handsome and attractive.
We enjoyed during the day a certain amount of sunshine, in-
terrupted by occasional showers, but soon after our arrival at
Hsiang-yang-ping it commenced raining in torrents, and con-
tinued to do so far into the night.

It ceased raining before daylight, to our great joy. Making
an early start we toiled slowly over the dreaded Pan-lan shan,
crossing the pass in a dense, driving, bitterly cold mist. The
ascent is nowhere difficult, and none of us suffered seriously
from the effects of the rarefied atmosphere, in spite of the
evil reputation this pass has for mountain-sickness. The
ridge is narrow, razor-backed, the summit being composed of
sandstone, with marble embedded, piled up at an acute angle
and devoid of vegetation. Snow, unmelted from the winter,
lay in odd patches immediately below the pass, and on all
sides there was much fresh snow. The dense mists prevented
any extended view, but what little of the region was visible
was bare and desolate. Two or three of the lovely Snowbird,
Grandala celicolor, were flitting around the snowy patches,
their intense blue plumage contrasting remarkably with the

ACROSS THE CHINO-THIBETAN BORDERLAND 181

white carpet around. I made the pass, 14,250 feet altitude, and
the tree-limit about 11,800 feet.

The flora above 12,000 feet altitude is purely alpine and
similar in character to that of the region around Sungpan and
elsewhere throughout the Chino-Thibetan Borderland at the
same altitude. Meconopsis integrifolia occurs in countless
thousands; also, to my pleasant surprise, the dark scarlet-
flowered M. punicea. Although by no means so plentiful as
around Sungpan, there were many thousands of this beauti-
ful herb scattered around. Primroses are most abundant ;
Primula vinceflora ascends to 13,000 feet, where its place is
taken by the lovely P. nivalis and another closely allied species.

On crossing over I photographed the pass and then
descended with all possible speed to the miserable hostel of
Wan-jén-fén, alt. 13,700 feet, where our lunch awaited us.
A little below this hostel a few bushes of Willow, small-leaved
Rhododendrons, and Caragana spp. first appeared and
became abundant as we descended. Soon Larch and occa-
sional Spruce appear, and at 11,300 feet altitude trees are
fairly numerous. A shrubby, evergreen Prickly Oak is char-
acteristic of these wind-swept mountain-sides, the golden-
brown undersurface of its leaves rendering it most conspicuous.
(This Oak is almost as beautiful as the Golden Chestnut of
California (Castanopsis chrysophylla), and I am very pleased to
report its successful introduction to cultivation.)

In addition to the shrubs mentioned above, dwarf Juniper,
Spirea, and Sallowthorn alsc abound. This moorland country
is very interesting and shows unmistakable signs of a drier
climate than that enjoyed by the regions on the opposite side
of the pass.

A torrent which rises near the head of the pass is soon
augmented by tributaries and quickly becomes a roaring un-
fordable stream. The mountain-slopes close in, and at the tiny
hamlet of Kao-tien-tzu the road plunges into a ravine. The
sides of this ravine are wooded, Larch and Spruce being abun-
dant, with miscellaneous shrubby vegetation. The elegant
Syringa tomentella, a Lilac with branching panicles of fragrant
flowers, is very common. On issuing from this ravine we
crossed a tributary torrent, more turbulent in character than

182 A NATURALIST IN WESTERN CHINA

even the main stream, and found in front of us open country
largely under cultivation.

Our caravan was to have stopped for the night at the
hamlet of Kao-tien-tzu, but with greater zeal than knowledge
pushed on 20 li farther to Reh-lung-kuan. This blunder
upset my plans and put all things awry. The collecting work
had to be curtailed; it was Io p.m. before I got any supper,
and much of our work had to remain over until the morrow.

The Pan-lan shan is the boundary between two Chiarung
states. On crossing over we quitted the state of Wassu and
entered that of Wokje. The Wassu territory is wildly moun-
tainous, well forested, and but little suited to agriculture. In
consequence it is sparsely populated, and we encountered
very few of the inhabitants en route. The hostels and houses
on the main road are in the hands of Chinese or half-castes.
The men of Wassu are tall (5 feet 8 inches or thereabouts),
with large, muscular frames, frank, open countenances, and
are noted hunters of the beasts of forest and crag. The women
are sturdy, buxom, and engagingly frank. Both men and
women are darker complexioned than the Chinese, and, I am
sorry to say, infinitely less cleanly in appearance. They are
very fond of jewellery, both sexes wearing bangles of silver
and copper, and silver rings studded with coral and turquoise.
The women also wear large silver ear-rings, usually having
insets of coral and turquoise. The men are addicted to opium-
smoking, though possibly this is strictly true only of those
engaged near the main roads as porters and muleteers who
have come in close contact with the Chinese.

Reh-lung-kuan, alt. 10,900 feet, is a Wokje village con-
sisting of about a score of houses, a small lamasery, and a tall
square tower. We found here a spacious and very fair inn,
and the people were courteous and obliging. Our carrying
coolies were able to purchase opium and a certain amount
of food-stuffs. This explained their anxiety to cover 75 li,
instead of stopping 20 li short at Kao-tien-tzu.

On leaving Reh-lung-kuan we descended the right bank
of the river, which rises near the Pan-lan shan pass, for 33 li
to the hamlet of Kuan-chin-pa, a short day’s march being
necessary in order to accomplish the work left over. The day

ACROSS THE CHINO-THIBETAN BORDERLAND 183

was fine and warm, with a strong, cool breeze. Looking back
on our route the snows of the Pan-lan shan were visible the
whole day. The road was in good repair, and skirts the
mountain-sides well above the stream. In ancient times this
valley was filled with glacial detritus, through which the strong
torrent has cut a deep, narrow bed. This stream, known
locally as the Nei chu, is really the principal branch of the
Hsaochin Ho (Little Gold River). Formerly gold in con-
siderable quantities was mined in this valley, and we passed
many old workings during the march.

The country generally reminded me forcibly of the Upper
Min Valley, near Sungpan, above 8000 feet altitude. On the
left bank of the stream the mountain-sides are very steep and
largely covered with woods composed of Spruce, Silver Fir,
and a few Pine trees. On the right bank the mountains are
more sloping and mainly under cultivation. Wheat is the
staple crop and ripens in early August; buckwheat ranks
next in importance, followed at a respectable distance by peas,
beans, and Irish potato. The Wokje people are evidently
skilled agriculturists and in their own way fairly well-to-do.
The prosperous condition of this state was evidenced by the
plenitude of large houses, lamaseries, and by the relatively
dense population. The hostels, however, are all in the hands
of half-breeds, descended from early Chinese colonists. The
larger houses and lamaseries are usually perched on some bluff
composed of glacial mud, grits, and boulders. They are more
or less square, two-storied, with flat mud roofs, having small
turrets at each corner, from which prayer-flags flutter; a
branch of some kind of Conifer is usually in evidence near
these flags. Chortens and other Lamaist monuments occur
here and there, while inscribed Mani-stones are common.
The peasants’ houses are low, one-storied, built of sandstone
shales, the roof either flat or with very slight slope.

That the climate of this valley is relatively dry and warm
is clearly shown by the flora, which is markedly xerophytic.
Two species of Cotoneaster, several Clematis, the Sallowthorn,
Prickly Oak, Barberries, and Roses are the chief constituents.
A curious Bush Honeysuckle, with small leaves and tubular,
white, fragrant flowers borne in pairs, is locally abundant.

184 A NATURALIST IN WESTERN CHINA

(This proved to be a new species and has been named Lonicera
tubuliflora.) Another common plant is the shrubby Clematis
Jruticosa, with simple oblong leaves and golden yellow, nodding
flowers. A Lilac (Syringa Potaninit), with erect panicles of
rose-purple flowers, is another interesting shrub, plentiful in
this valley. Poplar, a Hard Pine (Pinus prominens), with
almost prickly cones, and a White Birch, the bark of which is
used for lining straw hats, are the more common trees by the
wayside. I also gathered a few late flowers of Incarvillea
Wilsonit. Ina general way this Incarvillea resembles Delavay’s
species, but averages 4 to 6 feet in height. Another new plant
collected was a Primrose akin to Primula sibirica, but with
taller scapes and longer pedicels.

Kuan-chin-pa, alt. 9500 feet, consists of two small and rather
poor inns, with the ruins of a large square tower near-by.

Twelve li below Kuan-chin-pa, and also on the right bank
of the river, is the village of Ta-wei, a considerable place for
this region, boasting a large lamasery. This place has an evil
reputation, but no ill-will was displayed toward me. Many
Lamas clad in claret-coloured serge crowded around and
watched me as I photographed the village, and displayed
much interest in my camera, dog, and gun. Nevertheless,
the reputation of this village is well founded, and I would advise
travellers to avoid staying overnight there. From Ta-wei a
road leads across the river and over the mountains to Mupin.

On continuing our journey we followed the right bank
of the stream for a further 27 li to Mo-ya-ch’a, where, owing to
an old landslide, it was necessary to cross over to the left bank.
This was accomplished by means of a wooden semi-cantilever
bridge. Such bridges have been fairly common en route, but
this was the first our road had led over. From this bridge the
road descends the left bank, keeping high up above the river to
Kuan-chai, which was our destination for the day. The whole
valley is very arid, though a considerable area was under
wheat. A few Poplar and Willow trees occur near the river,
otherwise only high up on the mountain-sides were any trees
discernible. The flora is similar to that of all the principal
river-valleys of this hinterland, as described in Chapter XII.
Rosa Soulieana is very abundant. I gathered several new

a
ae

ACROSS THE CHINO-THIBETAN BORDERLAND 185

plants, but the country is too arid to be of much interest
botanically.

Situated at an altitude of 8500 feet, Kuan-chai is a small
village and the residence of the Wokje chieftain. The chief’s
house is very large, the upper structure, all of wood, is well
built, and the whole is dominated by several tall towers, and
fine Walnut trees occur scattered around. The prosperous
condition of this little state was further evidenced during the
day’s march. Large houses are frequent, many being perched
high up on the steep mountain-sides. Wheat is the principal
crop grown, and at Kuan-chai was just bursting into ear.
Maize and the Irish potato are likewise commonly cultivated.
A little flax and Hemp (Cannabis) also occur, the oil expressed
from the seeds of these plants being in general use as an illumi-
nant. We passed odd fields of opium poppy, the plants being
only a few inches tall. On the fan-shaped slope, at the head
of which the village of Kuan-chai is situated, all the crops were
remarkably luxuriant.

At Ma-lun-chia a considerable torrent joins the Nei chu
on the right bank. A by-road ascends this tributary, leading
to Fupien and thence to Lifan Ting. Our road was for the
greater part good and we easily covered the 67 li, enjoying
bright sunshine the whole day.

Immediately beyond the chief’s residence the road mounts
over a steep bluff, where is situated the hamlet of Hsao-kuan-
chai. This place is reputed to have offered a stern resistance
to the Chinese in their conquest of this valley a hundred odd
years ago, and was only captured after a long siege. The
remains of the sangars and old forts are still to be seen. From
this point the road continues to wind along the left bank of
the river for 40 li to the town of Monkong Ting. Both sides of
the valley are very arid, and the flora poor and uninteresting.
Very few houses occur in the valley, but high up on the moun-
tain-sides we saw many scattered about and surrounded by
wheat fields. At Laoyang the river is joined on the right bank
by another of almost equal volume. The main road from
Lifan Ting, via Fupien, descends this tributary stream and joins
at this point the road we were following. From what little
we could see of the valley of this Fupien stream it appeared

186 A NATURALIST IN WESTERN CHINA

to be as arid and barren as the one we had descended from
Reh-lung-kuan. Continuing our journey and on rounding a
bend in the river, we suddenly sighted, perched on a rocky
promontory, the town of Monkong Ting. After passing
through a gateway we noticed a separate township, rather
prosperous looking, situated in a lateral valley a little to the
left of the main road. This is the official town of Monkong
Ting, where reside the principal officials, civil and military.
Crossing a torrent by a wooden bridge we entered the place
first sighted from the bend in the river. This proved to be
an old military camp of poverty-stricken, dilapidated houses,
scattered alongside a street about 100 yards long. Two hun-
dred yards beyond this camp we reached the thriving business
town known as Hsin-kai-tsze. Monkong Ting, therefore,
consists of three distinct towns or villages: (1) the official
town, (2) an old military camp, (3) the business town. All
three are unwalled, though a gateway has to be passed on
entering each. The situation is most picturesque and strategi-
cally very strong. Monkong Ting is the political capital of
this region and a place of very considerable importance. The
two Chiarung states of Wokje and Mupin have their boundaries
at this point, and the rest of the valley to Romi Chango is
divided into feudal states.

The streets of Hsin-kai-tsze were thronged with people,
chiefly tribesfolk, selling medicines and buying various articles
for their own use. They made a very picturesque crowd, the
women being especially noticeable by reason of their display
of silver dress-ornaments, bangles, and ear-rings. The inns were
all crowded, but the head official obligingly secured a couple
of rooms for us and treated us with much courtesy and good-
will. The people were naturally curious and grouped them-
selves around us, but their manners were deferential.

Hsin-kai-tsze, alt. 8200 feet, is a most important medicine
mart, being famous for its “ Pei-mu” (Fvitillaria spp.),
‘“‘ Rhubarb,” ‘‘ Ch’ung-tsao”’ (a caterpillar infested with the
fungus Cordyceps sinensis), and ‘‘Chung-hoa”’ (an Umbelliferous
plant, possibly Ligusticum Thomsonit). All of these are
collected and brought in for sale by the tribesfolk. Musk and
deer-horns also figure in the trade.

ACROSS THE CHINO-THIBETAN BORDERLAND 187

Several roads radiate from this centre; one of these leads
from the official town to Mupin, over the pass of Chia-chin shan,
which was said to be higher than that of the Pan-lan shan and
surrounded by snow-clad peaks.

The Wokje state preserved its prosperous appearance to the
end, and is evidently a thriving, happy little country. The
people strongly resemble the Wassu folk, though possibly
they are scarcely as tall and have slightly sharper features.
The Chinese language is understood and in common use along
the main road, where the people imitate the Chinese in shaving
their heads and wearing a queue. Lamaism evidently has a
strong hold on these people, judging by the number of lama-
series we saw.

I had intended remaining a day at Monkong Ting, but owing
to the crowded condition of the town decided to defer this
holiday until we reached Romi Chango. The inn in which
rooms were provided for us was crowded with persons who
were noisy over their cups and business dealings far into the
night, rendering sleep well-nigh an impossibility.

Just outside Hsin-kai-tsze the road crosses over by a log
bridge to the right bank of the stream. This bridge was
being repaired, and only two very uneven logs were in position.
A thin rope was stretched across to serve as a hand-rail on
the left side. Crossing was really dangerous, the waters
below being deep and turbulent. The official kindly provided
local experts to carry our gear over, and the way these men
accomplished the task filled me with admiration. I rewarded
them with rooo cash, to their astonished delight. My dog
was lashed firmly to a flat board and carried across on a man’s
back. He struggled violently, and the man only just managed
to get him over before he got half loose. I walked over behind
the dog and was relieved when the 30 yards across the yawning
gulf were safely passed. Everything came over all right, but
my followers clung to the local men like grim death, the
majority shaking in their nervous fright. Such dangerous
experiences are not desirable, and I heartily hoped that we
had no more such bridges to cross. From this bridge we
descended 60 li to the hamlet of Shéng-ko-chung, alt.
7600 feet, through arid country and over a bad road. The

188 A NATURALIST IN WESTERN CHINA

river is here a broad and turbulent stream, flowing between
steep banks composed of loose rocks. A few Poplar, an
occasional Cypress (C. torulosa), and the Kcelreuteria (the latter
' was covered with masses of small yellow flowers) are the only
trees of note. The region is very sparsely populated, but
high up, on the left bank more especially, are a few houses of
the same architecture as those of Wokje. :

As travelling companions during the day we had a party
of tribesfolk, chiefly women in holiday attire. They were
very cheerful, laughing and singing most of the time. On
parting company at Shéng-ko-chung they made merry over cups
of Chinese wine, the dames officiating as to the manner born.

It rained heavily during the night, and it was cool and
delightfully fresh in the morning when we recommenced our
journey down the valley of the Hsaochin Ho. Thirty li below
Shéng-ko-chung we passed the large lamasery of Gi-lung,
coloured white and picturesquely situated on the right bank
of the river. Over a hundred Lamas reside here and exercise
considerable authority over the neighbourhood. About ro li
beyond this lamasery the river suddenly develops into a series
of boiling, roaring cataracts. The fury of the waters was most
fearsome to behold, and a wilder stretch of river is scarcely
imaginable. Earlier in the day we had crossed to the left
bank, and just below the very worst bit of this savage waterway
we recrossed to the right bank over a rotten and most unsafe
wooden bridge. Some 7 li below this point we reached the
hamlet of Pan-ku chiao, alt. 7100 feet, where we found
accommodation for the night, having covered 70 li. Just
above the hamlet a torrent joins the river on its left bank,
and up this lateral valley mountains clad with snow were
plainly visible. Bridges are scarce and the few that exist look
as if they had not been renewed since this region was conquered,
well over a hundred years ago. One thing is certain, they
cannot possibly last much longer: the two we crossed during
the day were all askew and decidedly dangerous.

The district is rather less arid than that around Monkong
Ting, yet the flora is very poor. Poplar is a common tree,
so also is the Kcelreuteria, which was a fine sight, with a wealth
of flowers, and it evidently enjoys a dry, hot situation. The

AMUSVANVT ONNTIOS AHL

ACROSS THE CHINO-THIBETAN BORDERLAND 189

sub-shrubby Incarvillea variabilis and Amphicome arguta,
both with large, tubular, pink flowers, are very abundant by
the roadside. Other common shrubs are Bauhinia, Sophora
vicitfolia, Ceratostigma, with lovely blue flowers, Ligustrum and
Rosa Soulieana. On the cliffs Cupressus torulosa is dotted
about. Maize is the principal crop, occupying in season almost
every inch of available land. Houses are fairly numerous,
but most of them are relegated to the higher slopes well above
the valley. The scenery in places is rugged and grand. In
front of the inn at Pan-ku chiao limestone cliffs rear themselves
some 2000 feet, abutting on a cultivated slope where Walnut
trees are scattered around. Crowning a bluff is a tall tower,
and near-by another in ruins, telling of glories now departed.

On leaving Pan-ku chiao we descended the right bank of
the Hsaochin Ho, some 42 li to the point where it joins the
Tachin Ho or Upper Tung River. This final stretch is little
else but one long succession of cataracts and strong rapids,
the turbulent waters being thick with brown mud. High
bare cliffs predominate, but here and there occur more or less
flat fan-shaped areas under cultivation, with houses shaded by
Poplar, Willow, and Walnut trees. Duzospyros Lotus, Hovemia
dulcis, and the large-leaved Ligustrum lucidum are other
trees common hereabouts. Maize is evidently the chief
summer crop in these regions, but wheat is grown, a red,
beardless variety, with stout ears, and harvesting was in
progress. Rock-pigeons are very abundant, and were busily
engaged in exacting their toll of the ripening grain.

After passing the hamlet of Yo-tsa we sighted on the
opposite (left) bank a large lamasery sequestered midst a fine
grove of trees. A little beyond this is the village of Tsung-lu,
a curious-looking place, boasting of a score or more tall towers.
Skin coracles are employed to ferry over to these places.

The Hsaochin Ho is prevented from joining the Tachin Ho
at right angles by a rocky spit which at times is evidently
flooded over. Marble and granite are common rocks hereabouts,
the latter being full of mica flakes which glistened in the sun.
Ascending the left bank of the Tachin Ho for a couple of li, then
crossing over a bamboo suspension bridge go yards long, we
soon reached the small town of Romi Chango. The whole day’s

190 A NATURALIST IN WESTERN CHINA

journey was only 45 li, but owing to the heat and rough road we
all arrived very much fatigued and in sore need of a day’s rest.

From all I could learn it would appear that the region in
the vicinity of the river from Monkong Ting to Romi Chango,
after its conquest by the Chinese about A.D. 1775, was
divided into feudal states, and certain chieftains installed in
possession as rewards for services rendered during the struggle.
The chiefs, styled Shao-pés, hold hereditary office and are
directly responsible to Chinese authority for the good behaviour
of the people under their rule, also, if necessity arises, they
are bound to supply armed men to assist the Chinese cause.
Lamas alone are exempt from such military duties ; ordinarily
the people of these feudal states are agriculturists. These
Shao-pés are subordinate to the Chinese military commander
stationed at Monkong Ting. The two chief Shao-pés reside,
one at Monkong Ting, the other at Che-lung, a village in the
mountains, 20 li removed from the left bank of the Hsaochin
Ho and 60 li below Monkong Ting. Another Shao-pé resides
at Ta-ching, 120 li to the north-east of Monkong Ting; a
fourth at A’n-niu, a place in the mountains to the south-west
of the region controlled by the Che-lung Shao-pé. Beyond the
original grant of territory these feudal chiefs receive no rewards,
monetary or otherwise, from the Chinese. The system has
much to recommend it and evidently works very well. It keeps
the Chinese authority supreme, while it allows the native
people to be governed by their own recognized chiefs. The
difference between the chieftain of a semi-independent Chiarung
state and a Shao-pé appears to be that, whereas the former
is an absolute ruler over a territory long hereditary to his
tribe, the latter is more in the nature of an alien ruling over a
tract of country fiefed to his forbears by the Chinese, after they
conquered this region and broke up the Chiarung confederacy.
The territory occupied by these feudal states formerly belonged
to the Chiarung tribes, and the people are principally derived
from that stock. Chinese settlers have intermarried with
the natives, and in the vicinity of the main road the population
is mixed. The people living in the lower stretches of the
Hsaochin Ho are an inferior race, of poor physique, and most
abominably filthy.

CHAPTER XV
ACROSS THE CHINO-THIBETAN BORDERLAND

Romi CHANGO TO TACHIENLU; THE FORESTS OF THE TA-P’AO
SHAN

a poor, unwalled, straggling town of about 130 houses,

It is without rank, but a magistrate, subordinate to the
Tachienlu Fu, and a military official, controlled from Monkong
Ting, reside there. The town is really a Chinese settlement,
situated in the extreme north-east corner of the state of Chiala.
It is built on the right bank of the Tachin Ho, at a point where
the river, making a right-angled turn from the northward, is
joined by a very considerable torrent from the west. The
Tachin, a river 100 yards broad, with a steady current and
muddy water, sweeps round majestically. High cliffs on the
left bank, steep mountain-slopes on the right, lofty mountains
to the east and west wall in the town, at the western entrance
to which a massive square tower stands sentinel. Chango is a
very poverty-stricken place, with a small trade in medicines
and sundries. It draws its supplies of rice, paper, and Chinese
commodities generally from Kuan Hsien, and everything is
phenomenally dear. This is only natural when the distance and
difficulties of the journey are duly considered.

A small road descends the right bank of the Tachin Ho, by
means of which Luting chiao may, with great difficulty, be
reached. A road ascends the right bank of the Tachin Ho
and leads to the interesting Chiarung states of Badi and
Bawang, where the Bonpa religion holds full sway. Badi, the
capital of these now united principalities, is only 60 li from
Romi Chango. The chieftain is dead, but his widow, assisted

by asteward, acts as regent for her infant son. Badi-Bawang is
191

R OMI CHANGO, or Chango, as it is commonly called, is

192 A NATURALIST IN WESTERN CHINA

one of the ancient matriarchal kingdoms of Chinese historians,
and at all times a woman holds an important place in its
government. Badi, the larger of the two states, is very rich
in gold, which, though unworked during recent years, is
jealously guarded. Chinese visitors, rich or poor, are cross-
questioned as to their business and closely watched during
their sojourn in this state. The Badi-Bawang folk often visit
Chango on business, and during our stay there we saw several.
Most of them were peasant girls and women, dressed so scantily
as to scarcely hide their nakedness. They were short in
stature, and apparently unwashed from birth! However,
since these were “‘ hewers of wood and drawers of water ”’ of
the poorest class, it would be unfair to judge the whole race
by them.

In Chango we lodged at a comfortable inn, having a clean
room, well removed from the street and overlooking the river.
We spent a quiet day resting and refitting for the final stage
of our journey to Tachienlu. The people were not over-
inquisitive and those in charge of the inn were exceedingly
obliging. Soon after our arrival the magistrate sent me word
that he was suffering from pains in the stomach and vomiting,
and would be grateful for some medicine to relieve his suffering.
I sent him some Epsom-salt and an opiate. The next day
word came that he was much better, only too tired to leave
his room. A traveller gets many such requests for medicine,
and I have generally found quinine, Epsom-salt, and opium pills
most useful cures, for which the people were always grateful.

On leaving this lonely town of Chango, which I made
6700 feet altitude, the road to Tachienlu ascends the right
bank of the tributary torrent. We were warned that the road
was very difficult, leading through forests and over high moun-
tains. It was not long before these statements were verified.
The torrent quickly develops into an angry, irresponsible
stream; the road in many places had been washed away and
much wading was necessary. Our carriers had great diffi-
culty in getting along, and had the waters of the torrent been
a few feet higher the road would have been quite impassable.
All the bridges were rotten and most insecure. High up on the
mountain-sides we saw several large hamlets, but there are very

OONVHO INOW AO NMOL AHL

ACROSS THE CHINO-THIBETAN BORDERLAND 1093

few houses in the valley—quite sufficient, however, for when-
ever the road led past a house we had to traverse an open
sewer, often a foot deep in dung and refuse. Such filthy
surroundings are characteristic of Thibetan houses. The
Chinese would collect all this sewage for their fields, but the
Thibetans, who are but poor agriculturists at best, have not
yet learned the value of manure. At such places I usually
climbed over the fences and walked through the crops, but my
men waded through the filth and gave vent to their wrath in
loud, angry imprecations. The people of Chiala are typical
Thibetans and use the lower stories of their flat-roofed houses
as pens for horses and cattle. A few li above Chango the flora
begins to lose its purely xerophytic character, and becomes
more and more luxuriant as the ascent proceeds. The higher
slopes are well forested with mixed trees, but near-by the road
trees are scarce. The mountain-sides flanking the stream
are very steep, being often sheer cliffs. Such places are
dotted with Cypress (Cupressus torulosa) and prickly leaved
Evergreen Oak.

After journeying 60 li we reached the village of Tung-ku,
alt. 7800 feet, where there are several large Thibetan houses,
decorated with prayer-flags, but only two or three hostels, and
these very poor in character. The owner of the one we stayed
in is a noted hunter, and many pelts of the Budorcas, Serow,
and Black Bear were in use as bed-mattresses. His family
told us the hunter was away after Musk-deer; they also
informed us that both the Thibetan-eared and Lady Amherst
Pheasants are common hereabouts. Around the village
Walnut trees are most abundant. Wheat is a common crop
and was just ripening. Maize too was plentiful and is evidently
the staple summer crop everywhere in these regions.

The next day we covered another 60 li, putting up for the
night at the poor hamlet of T’ung-lu-fang. We crossed the
river four times by wooden bridges, each more rotten than the
other. The river was in partial flood, and a goodly portion of
the road was either washed away, obliterated by landslides,
or under water. Often we had to make a path for ourselves
up the mountain-side. The under-water portions of the road
I traversed on the back of one of the soldiers we had with us

VOL. I.—I3

194 A NATURALIST IN WESTERN CHINA

from Chango, until he stumbled and gave me a ducking. After
this I waded. There was no traffic on the road so called, and
I marvelled how my coolies managed to get their loads along.
Our chairs were carried piecemeal and even then with difficulty
over the worst places. The river was a roaring torrent through-
out the whole day’s journey, in places really awesome to be-
hold, dashing itself headlong over enormous boulders, or boiling
as if forced up by some malignant spirit. In many places our
path actually overhung this torrent, and one false step meant
death.

About to li above Tung-ku the river makes a right-angled
turn and is joined at this point by another stream of almost
equal volume from the westward. From this place the road
skirts the river through a narrow, savage, magnificently
wooded ravine. Maple, Ash, Hornbeam, Birch, Poplar,
Hemlock Spruce, and Prickly Oak are the chief constituents
of these woods, followed by Evodia, Rhus, Cypress, Willow,
Elm, Sallowthorn, Bamboo, and miscellaneous shrubs. The
Maples (Acer Davidit and A. pictum, var. parviflorum) are
larger trees than I have seen elsewhere. The Ash and Horn-
beam are all fine trees, and the Hemlock Spruce in many cases
over 100 feet tall, with a girth of 12 to 15 feet.

On leaving this magnificent fragment of virgin forest the

country became less interesting. Where the cliffs are not sheer

and bare the mountain-slopes have been cleared to a very
large extent. The ravine widens into a narrow valley which is
covered with scrub. The cliffs and mountain-slopes high up
are sparsely clad with Cypress, White and Hard Pine, Spruce,
Silver Fir, and Hemlock. The scenery is sublime.

We passed very few houses and these of the meanest
description. Very little land is under cultivation; Maize
is the chief crop, with patches of wheat and oats here and
there. The country is not suited to cultivation, and one mar-
vels how the few people living there manage to find even the
most miserable subsistence. Yesterday we noticed herds of a
small breed of cattle. The people are shorter in stature than the

eee eee

average, and perfectly proportioned dwarfs are fairly common. _

Since leaving Monkong Ting, goitre has been manifest among
the inhabitants, and in this river-valley it is very prevalent.

OREGROUND

E IN

O, A POPLAR TRI

FORESTS WEST OF CHAN(

ACROSS THE CHINO-THIBETAN BORDERLAND 195

T’ung-lu-fang, alt. 8800 feet, consists of about half a dozen
scattered houses. The one we stayed in is of Thibetan archi-
tecture, fairly clean, and owned by a Chinese settler. None of
these houses affords any bedding for the coolies, and of course
nothing is purchasable—all food-stuffs have to be carried by the
travellers themselves.

The people at T’ung-lu-fang informed us that we should not
be able to reach Mao-niu, as the road had been badly washed
away in several places, and under the lee of some cliffs was
flooded to a depth of 4 feet or more. This gratuitous and
discouraging information proved, luckily for us, to be scarcely
accurate, since, after a struggle, we managed to get through.
My head coolie declared it was the very worst road we had ever
traversed, and I was inclined to agree with him. Worse it
could not have been and constitute a roadway at all! For
fully half the distance the track was under water or washed
completely away, and we were forced to wade or make a new
path over the mountain-side. Just how we got over the 30 li
I cannot describe, but we all came through with nothing worse
than a severe wetting.

Mao-niu is a fair-sized village for the country, and is mainly
perched on a flat 200 feet above the torrent, and surrounded by
a considerable area under wheat—a veritable oasis, in fact,
surrounded by high mountains. Formerly it was the principal
village of a petty state to which it gave its name. It now
belongs to the state of Chiala. As far as Mao-niu the scenery
and flora is similar to that around Tung-ku and calls for no
special remark. The outstanding feature is the woods of Hard
Pine (Pinus prominens). The steeper the country the happier
this Pine appeared to be. The bark of the trunk is deeply
furrowed, often red in the upper parts of the tree ; the cones are
quite prickly, and are retained for many years. The wood is
very resinous, and is evidently much esteemed for building
purposes. The Hemlock Spruce is common, and all the trees
are of great size.

At Mao-niu the main stream leads off in a westerly direction
to Th’ai-ling, a large village of over 100 houses and several
lamaseries. It is also the centre of a considerable gold-mining
industry, and has the reputation of being a lawless district.

196 A NATURALIST IN WESTERN CHINA

We were informed that the road thither was in a dreadful state
of disrepair, and that most of the bridges had been washed away
by recent floods.

On clearing the cultivated area around Mao-niu we plunged
immediately into a narrow, heavily forested ravine, down which
a considerable torrent thundered. Conifers preponderate in
these forests, Spruce being particularly abundant. We noticed
some huge trees, but the average was about 80 to 100 feet tall.
White and Red Birch are common, and I was fortunate enough
to secure seeds of the latter. The Sallowthorn (Hippophé
salicifolia) is exceedingly common, forming trees 30 to 50 feet
tall with a girth of 4 to 10 feet. The size of these trees very
much surprised me. Willows, Cherries, and different species of
Pyrus are also plentiful. Deutzia, Hydrangea, Philadelphus,
Rosa, and Clematis are the principal shrubs, and many were
in flower. Primula Cockburniana, which has orange-scarlet
flowers, is the most noteworthy herb hereabouts.

After wandering several miles through the forests we
reached the hamlet of Kuei-yung, alt. 10,100 feet, and 60 li
from T’ung-lu-fang. This place consists of half a dozen houses,
purely Thibetan in character, built on a slope and surrounded by
a considerable area under wheat, barley, and oats. The moun-
tains all around are heavily forested with coniferous trees, and in
the far distance a snow-capped peak glittered on the horizon.

The house we lodged in is three-storied with the usual flat
mud roof. The walls built of shale-rock are most substantial.
Entering through a low doorway we had first to traverse a yard
filled with cattle dung, then a piggery where a steep ladder led
upwards to a couple of dark empty rooms in which we installed
ourselves. A ladder from these rooms led to the roof, where
I should have preferred to sleep had it not been raining. The
house boasts neither table, stool, nor chair, and we had to
improvise as best we could. The Thibetans squat on the
floor for their meals, and therefore have no use for tables or
chairs. The housewife, a most cheery if dirty person, had a
very musical laugh. Things generally appeared a joke to her,
and incited her to frequent laughter, which it was pleasant to
hear. My followers were oddly amused at the strangeness of
things, and appeared to enjoy the novelty.

a ies

ACROSS THE CHINO-THIBETAN BORDERLAND 1097

Yet it was not out of love for our quarters that I stayed
over a day at Kuei-yung, but to photograph various trees and
investigate the Conifers. Photography in the forests is no
mere pastime. It took over an hour on three occasions clearing
away brushwood and branches so as to admit of a clear view
of the trunk of the subject. I secured a dozen photographs,
which entailed a hard day’s work. The trees of Larch and
other Conifers, Birch, and Poplar are very fine. The Larch
(L. Potaninii), though not plentiful, is of great size, and trees
100 feet by 12 feet in girth occur. But the most astonishing
feature of these forests is the large trees of Sallowthorn
(Hippophé salicifolia). I had never imagined it could attain
to the size of specimens I saw during the day. I photographed
two old trees 50 feet tall, 12 and 15 feet in girth respectively.
I saw others taller but less in girth. Another interesting tree
hereabouts is a Cherry (Prunus serrula, var. tibetica), which
has a short, very thick trunk, and wide-spreading head. The
leaves are willow-like, 3 to 4 inches long; the fruit is red,
ovoid, on pendulous stalks. The tree averages about 30 feet
in height, the head being 60 feet and more through.

The next morning we bade farewell to our cheery hostess
at Kuei-yung, and continued our journey. The road immedi-
ately plunges into the forest, and winds through and among
magnificent timber. The forests are very fine, and coniferous
trees 100 to 150 feet tall, with a girth of 12 to 18 feet, are quite
common. The latter consist of four species of Spruce, three
of Silver Fir and one of Larch. The handsomest of the Silver
Firs is Abies squamata, which has purplish-brown bark, ex-
foliating like the bark of the River Birch. The Larch becomes
general in the ascent, and ultimately overtops all other trees
and extends to the tree-limit. White and Red Birch, Poplar
and Sallowthorn are the only broad-leaved deciduous trees
really common. An Evergreen Oak (Quercus Ilex, var. rufescens),
with prickly leaves like a Holly, is abundant. In the shelter
of the forests this Oak makes a good-sized tree, but in the
more exposed places it is reduced to a small shrub. The wood
is very hard and makes the finest of charcoal. Shrubs are not
rich in variety, but Bush Honeysuckles, Barberries, Spirzas,
and Clematis are plentiful. Herbs, especially the Sikhim

198 A NATURALIST IN WESTERN CHINA

cowslip (Primula stkkimensis), P. involucrata, Anemone, Caltha,
Trollius, and various Composite luxuriate on all sides, and the
glades and marshy places were nothing but masses of colour.
The men who were in front of me saw several troupes of
monkeys and some Eared-pheasants, but I saw no animals and
very few birds.

We camped near the tree-limit, at about 12,000 feet altitude,
and erected a small hut of spruce boughs under a large Silver
Fir tree. My Boy preferred to pass the night in his chair,
and the men arranged themselves around a log fire. The
neighbourhood has an evil reputation for highway robbers,
but we felt sure there was small possibility of any attack on
us being made. It rained a little during the day, and a sharp
shower fell in the early evening, but the night proved fine.
The altitude, however, affected our sleep; it was also very
cold, and we were all glad when morning broke. My dog
suffered as much as any of us; he refused to eat his supper,
and I never saw him so utterly miserable. The coolies looked
a most woebegone crowd, shivering with cold and generally
wretched. They seemed to have no idea of making themselves
comfortable ; it would have been a simple matter for them to
have rigged up a shelter of spruce boughs, but they were too
indifferent to do this or even to collect firewood. We brought
with us from Kuei-yung, as guide, a Thibetan, and it was he
who got together all the wood required for a fire.

There was a slight frost and a heavy dew, but the sun,
which rose like a ball of fire, soon warmed us and dispersed the
dew. The road is of the easiest, winding through timber and
brush alongside a small stream, up to within 1000 yards of the
head of the Ta-p’ao shan Pass, where the ascent becomes
steeper. It is, however, only the last 500 feet that make
any pretence of being difficult. Above the place where we
camped the Conifer trees rapidly decrease in size, Larch
becomes more and more abundant, and ultimately forms pure
woods. It overtops every other kind of tree, and extends up
to 13,500 feet altitude. Just below the limits of the Larch a
dwarf Juniper appears and ascends to near the head of the
pass. The scaly-barked Silver Fir (Abies squamata) ascends
to 12,500 feet and two species of Spruce to 13,000 feet. This

SYAMOTA ANICTV YWHHLO GNV UVAGNVXUTVY WOHHA

ACROSS THE CHINO-THIBETAN BORDERLAND 199

side of the pass enjoys a moist climate, and the tree-line
(13,500 feet approximately) is remarkably high. Above the
tree-line the mountain-sides, to within a few hundred feet
of the pass, are covered with scrub composed, as usual in
these regions, of Willow, Berberis, small-leaved species of
Rhododendron, Spirea, Juniper, Potentilla Veitchit, P. fruticosa,
and Rhododendron Przewalskii, the latter being the most
alpine of all the large-leaved members of its family. Herbs,
of course, made a wonderful display of colour. In addition
to those previously mentioned, other species of Primula, the
yellow and violet-blue Poppyworts (Meconopsis integrifola
and M. Henrict), various Stone-crops (Sedum spp.), and Saxi-
frages are abundant. But the most striking of all the herbs
is a Rhubarb (Rhewm Alexandre), an extraordinary plant,
with a pyramidal inflorescence 3 to 4 feet tall, arising from
a mass of relatively small, ovate, shining, sorrel-like leaves,
and composed of broad, rounded, decurved, pale yellow bracts
overlapping one another like tiles on a house-roof. The local
name of this plant is “‘Ma Huang’’ (Horse Rhubarb); it
prefers rich boggy ground where verdure is luxuriant and
yak delight to feed. Such places were studded with its most
conspicuous tower-like spikes of flowers. The Rhubarb and
yellow Poppywort (Meconopsis integrifola) are always most
rampant around places where yak have been herded.
Unmelted snow of the preceding winter was lying in
patches just below the summit of the pass, a bare, narrow ridge
crowned by a cairn of stones surmounted with many prayer-
flags, and 14,600 feet above sea-level. This narrow neck is
composed of slate and sandstone, with a certain amount of
marble rock scattered about, and connects two massive ranges
clad with eternal snows. The day was gloriously sunny, and
we had a rare opportunity of enjoying and appreciating the
delights of this alpine region. Except for a feeling of giddi-
ness when stooping, and a general shortness of breath, I
suffered no inconvenience from the altitude. In spite of their
loads only two or three of my men were seriously affected ;
the gradual ascent was, I think, responsible for our good
fortune in this matter. From past experience I had rather
dreaded the effects this pass might have on my followers, and

200 A NATURALIST IN WESTERN CHINA

was pleasantly surprised at the ease with which they negoti-
ated it.

With the weather conditions so favourable the view from
the summit of the pass far surpassed my wildest dreams. It
greatly exceeded anything of its kind that I have seen, and
would require a far abler pen than mine to describe it ade-
quately. Straight before us, but a little to the right of our
viewpoint, was an enormous mass of dazzling eternal snow,
supposed to be, and I can well believe it, over 22,000 feet high.
Beneath the snow and attendant glaciers was a sinister-looking
mass of boulders and screes. In the far distance were visible
the enormous masses of perpetual snow around Tachienlu.
In the near distance, to the west-north-west of the pass,
another block of eternal snow reared itself. Looking back
on the route we had traversed we saw that the narrow valley
is flanked by steep ranges, the highest peaks clad with snow,
but in the main, though bare and savage-looking, they scarcely
attain to the snowline. On all sides the scenery is wild,
rugged, and severely alpine. A cold wind blew in strong
gusts across the pass, and we were glad when our photographic
work was finished, and we could hurry down. Several fine
Eagles and Lammergeiers were soaring aloft, but we saw no
animals, though Wild sheep and Thibetan gazelle were said to
frequent this region.

Descending by a precipitous, break-neck path, over loose
slate, sandstone shales and greasy clayey-marls for 15 li, we
reached the head of a broad valley. The pass on this side
offers a far more severe climb than the side we had ascended.
On reaching the valley the track we followed connects with
the main road to Th’ai-ling, Chantui, and Chamdo. Com-
mercially speaking this is the highway into Thibet from
Tachienlu. It leads through grasslands, affording good pastur-
age for animals, and though the mean elevation is very con-
siderable the passes are less steep than those on the political
highway via Litang and Batang. This Ta-p’ao shan region is
notorious for its highway robberies. We met five tribesmen
who told us that in the previous night their camp had been
rushed by an armed band and everything they possessed
carried off. Every Thibetan is by nature a robber, and

ACROSS THE CHINO-THIBETAN BORDERLAND 2o1

behaves as such when he fancies he can do so with impunity.
They rob one another freely, but the tribesmen are their
favourite victims.

From the head of the valley to Hsin-tientsze, the first
habitation, is reckoned as 30 li. The road is broad but uneven,
winding through a valley, and keeping close to a torrent which
descends from the Ta-p’ao shan snows. The mountains on
either side of the valley in all their higher parts range above
the snowline; their lower slopes are covered with grass,
small Conifer trees, and brushwood. In the valley itself shrubs
of large size, chiefly Willows, Honeysuckles, Barberries, and
Sallowthorn abound. Odd trees of Larch and Spruce occur,
all of small size. Flocks of Snow-pigeons were plentiful, and
I shot several of these birds for our larder.

From Kuei-yung, 120 li, there is no house of any descrip-
tion save Hsin-tientsze, alt. 10,800 feet, a filthy and miserable
hostel. Near Kuei-yung we passed a charcoal-burning camp
where a few men were engaged, otherwise we did not meet a
living soul, until we had crossed the pass. It is indeed a
most lonely region, but of great interest to a Nature lover. I
count myself particularly fortunate in being favoured by per-
fect weather for crossing the pass, more especially as it was the
first day without any sign of rain since leaving Kuan Hsien.

The thermometer registered 36° F. when we turned out
next morning, and our ears and fingers tingled with cold, even
though it was 8th July. The smoke inside the inn was too
much for my eyes, so I breakfasted out in the middle of the
roadway. I think everybody was glad to quit Hsin-tientsze
with its vermin and stinks. There was an odd patch of wheat
around the hostel, but it looked miserable; the season is too
short and the climate too severe for cultivation hereabouts at
this altitude.

We followed a broad, uneven road, which had suffered
much from animal traffic, for 60 li to Jé-shui-t’ang (Hot-water
pond), alt. g800 feet. The descent is gradual, and the day’s
journey proved a delightful loiter through a shrub-clad valley.
We met several hundreds of yak and ponies, all laden with
brick tea encased in raw hides and bound for interior Thibet.
The Thibetans in charge were an unkempt, wild-looking lot

202 A NATURALIST IN WESTERN CHINA

of men, with long guns, swords, and conspicuous charm-boxes.
Many of them wore their hair in a long plait with a sort of black
yarn braided in, the whole being wrapt around their heads to
form a turban ; a few wore felt hats with high conical crowns.
One or two women were with these caravans tending the
animals exactly in the same way as the men. Ability to
whistle and heave rocks with sure aim seemed to be the essential
parts of a yak-muleteer’s profession. Yak are slow, phlegmatic
animals, and on sighting any unusual object they stand stock-
still for a little time, and then make a mad rush forward. They
appeared to be docile enough, but their long horns looked
dangerously ugly, and we got out of their way as much as was
possible. Each caravan was accompanied by one or more
large dogs. These animals trot alongside the caravan and take
no notice of any one, but when tethered and on guard in camp
will allow no stranger to approach. They are massively-built
dogs, and their savage appearance is heightened by a huge
red-coloured collar of woollen fringe, with which they are
commonly decorated.

The flora was merely a repetition of that of the previous
afternoon’s journey. The valley and contiguous hill-sides are
covered with scrub, except for clearings here and there which
serve as yak-camps. In addition to the shrubs mentioned as
occurring around Hsin-tientsze, Prickly Oak, Juniper, several
kinds of Rose, and the Thibetan Honeysuckle (Lomnicera
thibetica) are common; Barberries in variety are a special
feature. Conifers are scarce and all of small size; all the
larger timber has been felled and removed long ago At the
hamlet of Lung-pu, reckoned 40 li from Hsin-tientsze, crops
of wheat, barley, oats, and peas put in an appearance, and
became more general as we descended the valley. Around
Jé-shui-t’ang the cereals were just coming into ear.

During the day, which was beautifully fine, we had grand
views of the snowclad peaks around Tachienlu and the steep
ranges with pinnacled peaks to the east-south-east of that
town. Around Jé-shui-t’ang there are several hot springs, in
some of which the water was actually boiling. These springs
are rich in iron, but in those I examined no sulphur was evident.

Our quarters at Jé-shui-t’ang were a considerable improve-

SNVAd ‘NV. N-VIHO-VA AHL

ACROSS THE CHINO-THIBETAN BORDERLAND 203

menton those of Hsin-tientsze, but we were, nevertheless, glad to
leave very soon after day dawned. It is considered to be go li
from this place to Tachienlu, but I should say 60 li is a nearer
estimate. We enjoyed another sunny day. The road is
easy and leads through a continuation of the valley that we
entered on descending from the Ta-p’ao shan Pass. The valley
and mountain-sides for some 300 to 500 feet above it become
more and more under cultivation. Cereals, peas, and Irish
potato are the principal crops. The potatoes were being
harvested, and I noticed that red ones predominated. The
region generally has been denuded of its trees, and where not
under crops is covered with scrub and coarse herbs. In rocky
places small trees of White and Hard Pine (Pinus Armandt,
P. prominens) occur, also a few comparatively large trees of a
very distinct-looking Peach having narrow, lance-shaped, long
pointed leaves, rather small fruits, downy on the outside.

Around habitations tall trees of Poplar are common, and
an occasional Spruce and White Birch occur. The Spruce
(Picea aurantiaca) is a particularly handsome species, with
square, dark green needles on spreading branches and red-
brown pendulous cones clustered near the top of the tree.
The Apple, Apricot, Peach, Plum, and a few Walnut trees are
cultivated. The fields are fenced with hedges of Wild Goose-
berry (Ribes alpestre, var. giganteum) and the handsome Sorbaria
arborea, which has large erect masses of snow-white flowers.
Over these and other shrubs various species of Clematis trail,
the most common being C. nutans, var. thyrsoidea, which was
laden with a multitude of creamy-yellow nodding flowers.
The most beautiful shrub, however, was a Lilac growing 12 to
15 feet tall, and covered with huge panicles of pink or white
fragrant flowers. (It proved anew species, and has been named
Syringa Wilsonit.)

1 At the time I paid no further attention to this Peach, but in 1910 I
secured ripe fruit, and found to my astonishment that the stones were per-
fectly smooth, free, and relatively very small—characters denoting a distinct
species of Peach. It proved to be new, and has since been named Prunus
miva. I regard this as among the most remarkable of the discoveries I have
been privileged to make. This new Peach is now in cultivation, and by cross-
breeding with the old varieties of the garden Peach (P. Persica) may result in
the production of entirely new and improved races of this favourite fruit.

204 A NATURALIST IN WESTERN CHINA

We crossed the stream by a wooden cantilever bridge and,
on rounding a bend, the goal of our long journey came into view.
We were all well-nigh dead beat, and it was with thankful and
joyous hearts that we greeted the cluster of closely packed
houses, which, nestling in a narrow valley, constitute the
important border town of Tachienlu.

CHAPTER -XV1I
TACHIENLU, THE GATE OF THIBET

THE KINGDOM OF CHIALA, ITS PEOPLE, THEIR MANNERS AND
CUSTOMS

lat. 30° 3’ N. circa, at an altitude of about 8400 feet.
By the most direct route it is twelve days’ journey from
Chengtu Fu, the provincial capital, on the great highway which
extends westwards to Lhassa. It is the Ultima Thule of China
and Thibet, where a large and thriving trade is done in the wares
of both countries. It is also the residence of the King of Chiala,
who governs a very considerable tract of country and exercises
a strong influence over conterminous states peopled with
Thibetans. The first Occidental other than Roman Catholic
priests to visit Tachienlu was the late Mr. T. T. Cooper in 1868.
Since that date it has been visited by many scores of travellers,
and has become fairly well known to the outside world. Itisa
more than ordinarily interesting place, and though muchhasbeen
written concerning it the subject is far from being exhausted.
The present town is built in the narrowest of valleys at the
head of a gorge, down which the river Lu cascades, falling some
4000 feet before it joins the river Tung, 18 miles distant. A
branch of the Lu River bisects the town, being crossed by means
of three wooden bridges, and is joined immediately below the
north gate by another stream, which flows from the Ta-p’ao
shan snows. The town is hemmed in on all sides by steep,
treeless mountains whose grassy slopes and bare cliffs lead up
to peaks culminating in eternal snow. On the whole, the
situation is about the last in the world in which one would
expect to find a thriving trade entrepét. Formerly, Tachienlu
occupied a site about half a mile above the present town, but

205

o: HE town of Tachienlu is situated in long. 102° 13’ E.,

206 A NATURALIST IN WESTERN CHINA

about I00 years ago it was totally destroyed by a landslip,
due to a moving glacier. Some day a similar fate will
doubtless overtake the existing town.

Notwithstanding its great political and commercial im-
portance Tachienlu is a meanly built and filthy city. It is
without a surrounding wall, save for a fragment which runs
across near the south gate, and it has no west gate. The
narrow, uneven streets are paved with stone in which pure
marble largely figures, though this is only evident after some
heavy downpour has washed away the usual covering of mud
and filth. The houses are low, built of wood resting on founda-
tions of shale rocks. The principal shops are by no means
of imposing appearance, and, indeed, the only places note-
worthy are two Chinese temples and the palace of the local
king. The latter consists of several lofty semi-Chinese build-
ings of wood with sloping roofs and curved eaves surmounted
by gilded pinnacles, the whole structure being situated in a
large compound and surrounded by a high stone wall. The
residences of the Chinese officials are poor, ramshackle places,
and the same is true of the various inns. In the latter most of
the business is transacted. Some inns that I visited contained
valuable collections of porcelain and bronze-ware, and an
extraordinary number of old French clocks. Very few of the
clocks were in working order, but many were of large size, and
how they all reached this remote place is a mystery to me.

The population of Tachienlu consists of about 700 Thibetan
and 400 Chinese families and, with its floating members, is
reckoned at gooo people. In and near the town are eight
lamaseries boasting 800 lamas and acolytes. The population
is very mixed, consisting of pure Thibetans, pure Chinese,
and half-breeds. Very few purely Chinese women are to be
found in Tachienlu.

As seen in and around Tachienlu the Thibetans are a
picturesque people, Of medium height and lithely but muscu-
larly built, they have an easy carriage and independent mien.
The young women are usually sprightly in manner, always
cheery, with dark-brown eyes and finely cut features. Both
sexes are fond of jewellery ornamented with turquoise and
coral, but they are strangers to soap and water, and personal

GNNOWOANOT NI NMOL YWANWMOA AO WLIS ‘MN INATHOVL

i :

(he
‘ a

, ¢

he
in uy, 5 }
X oH

TACHIENLU, THE GATE OF THIBET 207

cleanliness is neither appreciated nor practised. Meat, milk,
butter, barley-meal, and tea constitute the favourite food of
these people ; they are also fond of Chinese wine. Everybody
carries on his or her person a private eating-bowl, and the
average Thibetan disseminates an odour strongly suggestive
of a keg of rancid butter! The everyday dress of these people
is a loose, shapeless garment of dull red or grey woollen serge,
sometimes sheep-skins are substituted in part. Top-boots
of soft hide with the hair inside usually encase the feet and
lower legs of both sexes. The men wear their hair in a queue
wound round the head and ornamented with beads and rings
of silver, coral, and glass. A large silver ear-ring with a long
silver and coral pendant usually decorates the left ear. The
women wear their hair parted down the middle and made up
into a number of small plaits, which are gathered into a queue,
bound at the end by a bright red cord, and wound around the
head. Silver and coral are lavishly used in their coiffure and
about their persons generally. When in holiday attire these
people are more gaily dressed, red-coloured trimmings to their
garments being then much in evidence, whilst the wealthy
affect silk and fur robes, Ornaments of silver and gold, inset
with coral and turquoise, are most profusely worn. The lamas
shave their heads and wear a raiment of coarse serge of a dull
red or brownish colour. This has no shape, being simply a
large piece of cloth thrown over the right shoulder, leaving the
left bare. A similar piece of cloth is wound two or three times
round the waist and reaches down to the ankle, forming a kind
of pleated skirt. They are usually bareheaded and bare-
footed, and each lama carries in his hand a rosary and a small
praying-cylinder. They swagger through the streets with an
insolent mien, and lack the good manners so delightful in
the ordinary unsophisticated Thibetan. The lamaseries are
usually very richly endowed with land, and most charmingly
situated midst groves of Poplar and other trees. Nearly all
Thibetan families of affluence maintain a lama on the premises
to perform by proxy their religious duties. Many other
lamas find employment as temporary chaplains to less wealthy
families on occasion of marriage, illness, or death.
Commercially, Tachienlu is a most important centre,

208 A NATURALIST IN WESTERN CHINA

enjoying a monopoly of the trade between this part of China
and Thibet. The value of the trade is estimated at about one
and three-quarter million taels. The Thibetans bring in musk,
wool, deer horns, skins, gold-dust, and various medicines, and
take in exchange brick tea and miscellaneous sundries. The
trade is largely one of barter, but much less so than that of
Sungpan Ting. Sycee and Indian rupees were formerly the
only coinage current, but the Chinese during the last few years
have been minting in Chengtu a rupee of their own for the
special purposes of this trade-centre. Its use has been insisted
upon, and, in consequence, the Indian coin has been ousted from
the field. Most of the “ bigger’’ trade is in the hands of the
lamaseries on the one hand, and Chinese from the province of
Shensi on the other. About 30 li to the north-east of Tachienlu
gold is found at an altitude of about 11,000 feet, and placer-
mining is carried on there. The gold-washing is done in exactly
the same way as elsewhere in Western China, but the method
of paying the miners is peculiar,—the arrangement being six
baskets for the owner of the mine and a seventh for the miners.
Silver also occurs at this same place. The Thibetans hold the
view that gold and other precious metals grow, and that their
death may result if too much is removed at any one time.
How far they actually believe in this superstition is a moot
point, but at times it serves as an unanswerable argument.
Nine years ago a difference of opinion in the matter of assessing
the profits arose between the Chief of Chiala, owner of the
mine, and the head Chinese official at Tachienlu, who was
apparently over-avaricious in the matter. The Chief very
quietly advanced the above theory, and closed down the mines
for an indefinite period! Gold in great quantity occurs in
the state of Litang, west of Tachienlu; much also is mined
around Th’ai-ling to the north of this town.

Being on the great highway from Peking via Chengtu to
Lhassa, officials are constantly passing through Tachienlu, and
the political importance of the town is very great. Although
only a city of the second class the head Chinese official has the
local rank of Prefect (Chiung-Liang Fu), and holds the post of
commissary for the Chinese troops stationed in Thibet. Al-
though Batang, 18 days’ journey westwards, is more accurately

TACHIENLU, THE GATE OF THIBET 209

the frontier town, Tachienlu is actually the “‘ Gate of Thibet.”
The country around and beyond is physically purely Thibetan
in character, and is ruled by native chieftains. Garrisons of
soldiers and a few resident Chinese officials protect the interests
of the Celestial empire and keep a sharp eye on the actions of
the local rulers.

It was stated at the commencement of this chapter that the
King of Chiala resides at Tachienlu, and perhaps a few details
concerning this kingdom and its people may be of interest.
According to the Guide Book of Thibet this State came under
Chinese influence during the Ming Dynasty, about a.p. 1403,
and its Chief was given the rank of a second-class native official,
with control over the tribes west of the river Tung and south-
wards to Ningyuan Fu. ‘“ The Manchu Dynasty, in considera-
tion of the above, made the then Chief a third-class native
official, with power over three trading companies. New chiefs,
chiliarchs, and centurions to the extent of fifty-six were created.
This illustrious Chief now controls six subsidiary chiefs, one
chiliarch, and forty-eight centurions.”’ Since the date of this
appointment the Chinese have increased their grip over these
regions, to the curtailment of the Chief’s power and authority.
Nevertheless, the Thibetans of this region acknowledge this
Chief as their supreme ruler, and in domestic affairs his authority
is absolute. His native title is ‘‘ Chiala Djie-po”’ (King of
Chiala) ; his Chinese title, “ Ming-ching Ssu,” which may be
translated ‘‘ Bright-ruling official.” The King and Chinese
Prefect (Fu) are supposed to be colleagues, but in reality the
King is subordinate, and when paying official visits must make
obeisance before the Fu. In what little dealings I had with
them I found both to be courteous and obliging, but suspicious
and jealous of one another.

The present King is a slimly built, intelligent man, about
forty odd years of age. He took considerable interest in our
collecting work around Tachienlu, and with his brother, who is a
hunter of much renown, paid us many unofficial visits. He was
never tired of watching my companion, Mr. Zappey, fixing up
his birds’ skins. My own work amongst flowers interested him
but little. As a parting gift Mr. Zappey stuffed and mounted
a Hoopoe for the King, who evinced almost childlike pleasure

VOL. I.—I4

a10 A NATURALIST IN WESTERN CHINA

on receiving it. In return, he made Mr. Zappey and myself
several presents, and urged us to visit his country on a future
occasion. We found that these Thibetans possessed keen and
accurate knowledge of the birds and animals of their country,
which made them enthusiastic hunting companions. During
the reign of the former king, his brother, the present Ruler,
was banished, and suffered dire hardships during his exile,
and often wanted for food. The missionaries stationed in this
neighbourhood on more than one occasion assisted him, and I
understood from them that he had not forgotten their kindly
help. The history of the family is a tragic one. The present
King’s brother was supposed to have been poisoned, and two

sisters died early deaths, the result, it is said, of immoral —

associations with lamas.

The state of Chiala is of considerable size, comprising
practically the whole of the territory lying between the
Tung River, Chiench’iang Valley, and the Yalung River
from lat. 28° to 32° N. The five Horba states in a measure
also come under the influence of the King of Chiala. From
all I can learn this region has the best right to be considered
the kingdom of Menia, or ‘ Miniak,” of European maps.
The whereabouts of “ Miniak’’ has considerably puzzled
geographers, but the evidence seems to point to the kingdom
of Chiala as representing it in the greater part. North-east of
Chiala is the large and prosperous state of Derge, famed for
its copper, silver, and swordsmiths. Monsieur Bons d’Anty,
French Consul-General at Chengtu, visited Derge in the
autumn of 1910, and on his return gave me a most interesting
account of this region. He informed me that Derge is a region
of much cultivation, surrounded on three sides by snowclad
ranges. The various industries for which the state is famous
are not carried on in towns, but by the peasants individually
in their homes, and from thence carried to towns for sale.
In the valley of the Upper Yalung, abutting on the north-west
frontier of Chiala and the south-east frontier of Derge, is a
wedge of country known as Chantui, peopled by a race of

Ishmaels, whose hands are ever turned in conflict against their _

neighbours. A similar people occupy a wedge of country in
the Drechu valley north of Batang, where they are known as

MANI STONES

TACHIENLU, THE GATE OF THIBET 211

the Sanai tribe. Monsieur Bons d’Anty considers that these
people are of Shan origin, and remnants of an aboriginal popula-
tion of this region. This authority has spent many years
in studying the ethnological problems of this borderland, and
is most competent to express an opinion. It is well known that
the Shans formerly ruled in western Yunnan, and there is
no reason why they should not, in the distant past, have
ascended the valleys of the Yalung and Drechu and established
themselves there. But whatever the origin of these people
of Chantui and Sanai, they are dreaded by their neighbours,
who regard them all as robbers and murderers (Ja-ba) quite
beyond the pale.

The religion of the people of Chiala is Lamaism, both the
orthodox “ yellow’’ and unorthodox “‘red”’ sects being re-
presented, but the former are the more numerous and powerful.
Some one has described Lamaism as “‘ mechanical,’ a most
descriptive term, since the religion consists in the main of
turning praying-wheels by hand, water, or wind, counting
beads, and the continual muttering or chanting of the mystic
hymn, ‘Om mani padmi hum.” Lamaism draws its inspira-
tion from Lhassa, where all the priests repair for study, the
head of the sect being the Dalai Lama. Aided and abetted by
Chinese authority, the King of Chiala has never submitted
to the Dalai Lama in temporal affairs ; he has maintained his
freedom and right to govern his own people untrammelled by
Lhassa interference, in spite of the dire threats and treachery
on the part of lamaseries within his jurisdiction. In 1903 the
Dalai Lama issued an ultimatum to the King of Chiala threaten-
ing to take from him and the Chinese by conquest all the
territory west of the Tung Valley. The British Expedition
prevented the carrying out of this threat. The Dalai Lama
undoubtedly had designs of territorial expansion at the expense
of China’s vassal states. The Chinese knew this, and it was
fortunate for them that Great Britain stepped in and broke
the power of Lhassa De. I was in Tachienlu during 1903 and
1904, and from what I saw and heard there it was plain that
the British were unwittingly pulling China’s “‘ chestnuts from
the fire.’”’ The Chinese were not slow to perceive the advan-
tageous position they were in after the power of the Dalai

212 A NATURALIST IN WESTERN CHINA

Lama was dissipated. Almost immediately a “‘ Wardenship of
the Thibetan Marches ”’ was established, and a war of conquest
engaged upon against certain wealthy lamaseries in Litang
and other states, who owned direct allegiance to Lhassa, and
heretofore had boasted their independence of China. This
war was relentlessly and victoriously pursued under the
leadership of Chao Erh-féng, and resulted in the extension of
Chinese authority over a very considerable tract of country.
Indirectly the King of Chiala’s position has been very much
weakened as the outcome of these conquests.

The state of Chiala is made up of mountain, dale, and
plateau, being essentially a highland country affording good
pasturage for yak, sheep, and horses. A chain of snowclad
peaks traverses it near its eastern boundaries. It is a region
where altitude regulates the mode of life, the wealth, and
marriage customs of its people. The inhabitants are less
nomadic than the people to the north and west, but, in common
with all other Thibetans, their wealth is represented by herds
of yak, horses, and cattle, and flocks of sheep and goats. They
are great hunters of Musk-deer, Wapiti, Bear, and other animals,
the commercial products of which they trade to the Chinese.
The same is true of the medicinal roots and herbs, which
grow abundantly in these uplands. Where altitudes admit,
agriculture is practised, but is supplementary to grazing
and relatively unimportant. Wheat, barley, oats, buckwheat,
peas, and Irish potato are the chief crops. During the winter
months these Thibetans live in well-built houses situated
in the valleys, and in the spring they migrate to the uplands.
The nomads do not move about aimlessly, but have clearly
defined regions and are subject to responsible head-men.
Where agriculture is carried on the womenfolk mostly remain
to look after the crops and to do other work pertaining to the
farmstead.

Wealth and convenience decide which form the matri-
monial alliance shall assume among these people, and polyg-
amy, monogamy, and polyandry obtain. Above 12,000 feet
altitude polyandry is the rule, and in many places women
so united wear distinguishing and honorary badges. Such
women are usually the business and ruling heads of their

TACHIENLU, THE GATE OF THIBET 213

establishments. This custom of polyandry is characteristic
of Thibet, and the following note on the subject, written by a
friend who has spent many years of his life among these people,
is worthy of much thoughtful study :—

‘“So many able men have written about polyandry that
what follows will be without interest to those who have studied
the system; but to the great mass who are comparatively
unacquainted with Thibet and her customs these notes may be
of some value. The writer has spent several years among
Thibetans and cognate tribes, and has lived for months alone
on the wild steppes as well as in the more civilized and well-
cultivated valleys.

“‘ The term ‘ polyandry ’ is here applied (a) to women living
permanently, and cohabiting legally, with more than one man ;
(b) to those who have been, or are, married temporarily to more
than one man or companies of men.

“The former, true polyandry, is confined to the pastoral
nomads of the grassy plateaux; the latter, quasi-polyandry,
is rampant in all the commercial and political centres on the
border and throughout Thibet. In both cases a low con-
ception of the relation of the sexes has made it possible; and
climate and political conditions have made it desirable.

“The past hints, and the present proves, that indifference
to female virtue connotes the people known as Thibetans
and tribes of common origin, and I understand it to be
the indirect cause of polyandry. From time immemorial the
Thibetan has been taught that the female is a kind of Pan-
dora’s box, in which are all the evils that have cursed mankind.
All down the ages woman seems to have been the slave of man :
dangerous because of latent evil, but also valuable on account
of her ability torender him service. In the old barbaric days,
when prowess was the ‘prime virtue and a_ thoroughgoing
communism the rule, woman was only a tribal asset, like the
animals she tended. Then came religion, a deification of all
that rude minds could not explain. It was probably the
mysterious Bonpa of to-day which lingers in the lonely valleys
where nations meet, and which could have been no friend of
virtue if the accounts of orgies in its temples before indecent
idols are true, and the unseemly dress of young women and

214 A NATURALIST IN WESTERN CHINA

barren wives either demanded or sanctioned by it. Explain it
as we may, the fact remains that Thibetan women are to-day,
as they seem to have been in the time of Marco Polo, the most
immodest of their sex, and the Thibetan men strangely in-
different about matters which other races demand as essentials.

‘« All outside work is done by the women, who represent the
coarser element of Thibetan society, and their language is often
filthy in the extreme. The domestic arrangements make no
provision for privacy. Men and women must eat, live, and sleep
perforce in the same apartment, and there is no effort on the
part of the male to shield the female from conditions which are
inimical to virtue.

“The morality of the Thibetans has made such a system
possible. This will not be denied by any one who knows them
even slightly ; but it will sound strange to many when I say
that the climatic and political conditions are such that the
reformer is puzzled to think of anything to offer as a sub-
stitute! To the untutored Thibetan mind it must seem
absolutely necessary. Undoubtedly the high altitudes are
unfavourable to women. The Thibetan views woman very
much as he does an animal, 7.e. she can do so much work.
Living and working at 12,000 feet altitude and upwards requires
the strongest material. Woman very imperfectly fulfils these
requirements, and maternity and nursing, apart from unfitting
her for work, would be well-nigh useless, since infant mortality
would be abnormally high. On the relatively thinly populated
plateaux the conditions obtaining are emphatically against
woman being wanted in numbers. Here robbing and escaping
from robbers is the normal condition. It will be evident
at once that family duties are not only inconvenient, but
interfere with the woman’s efficiency personally, and at the
same time misdirect the energies of the male portion of the
community.

‘The nomad is a herdsman, continually moving to and fro
with his flocks and belongings. The woman, and the centre
she forms, would impair the necessary freedom of movement ;
it would also follow that she and her belongings would often be
unprotected for long periods. Polyandry, by not encouraging
permanent settlements and at the same time being the best

4) gi ee a aR

TACHIENLU, THE GATE OF THIBET 215

security against marauding bands, must seem eminently
rational to the nomad.

‘“Polyandry also entails the family property. This is very
important, as division of the flocks or grazing-grounds would
soon ruin every one. Whatever the ideal system for these
Thibetans may be, the one which provides one wife, one family,
and one flock for all the male members of the family is the most
convenient. Anything else would be suicidal. Both polygamy
and monogamy presuppose racial increase and the formation
of new and independent centres, but polyandry promises the
great desideratum of the Thibetan—an almost stationary
community and an intact patrimony.

“In a land of polyandry, priestly celibacy, and nondescript
roving, the number of unmarried women must be large. This
class, with the Chinese, Lamas, and Thibetan merchants, is
responsible for the quasi-polyandry of the plain, which only
differs from prostitution inasmuch as it has the sanction of the
country and carries with it no odium. The priest is a celibate,
as a rule, by profession, but an inveterate roué in practice.
Quite a large number of women are required wherever lamaseries
exist. In Lhassa, where thousands of students from all parts
of the country study for years, the number of women married
temporarily, openly, or in secret, to individuals or small com-
munities is very great. The wandering Thibetan merchants
form another class who demand a supply of temporary wives for
longer or shorter periods. These may often be men who have
formed polyandrous unions in the mountains, but the exigencies
of circumstances demand their presence on the plain. In other
words, there is no reason why a man may not be a polyandrian
legally, and in practice a polygamist.

“But the most interesting phase of this system arises from
peculiarities of Chinese domination. Chinese soldiers, officials,
and merchants residing temporarily in Thibet form a very
large body. These victims of circumstances leave their wives
in far-away China. There is a legend that the Lamas have put
an embargo on the dainty Chinese woman: but, more prob-
ably, her lord and owner has neither the mind nor the money
to introduce her to the dangers and hardships of a Thibetan
journey. But he rarely, if ever, pines for the wife of his

216 A NATURALIST IN WESTERN CHINA

youth. Polyandry and polygamy meet, and temporary
marriages, for one month to three years, are the rule. The
highest official and the meanest soldier take advantage of the
system. With the former it is temporary monogamy or polyg-
amy, but with the latter, owing to pecuniary limitations, one
woman often becomes, pro tempore, the wife of a small com-
munity of soldiers. These wives or their children, for obvious
reasons, are seldom, if ever, brought out from Thibet; the
former make new alliances and the children are claimed by the
Lamas.

“The question of Thibetan morality is a very complex
one, and it is almost impossible to disentangle the cause
from the effect. True polyandry is owing, indirectly, to a
low moral perception; but it might be correct to blame
it, in a measure, for the more degenerate quasi-polyandry.
Whatever we may think of the former, from the standpoint
of absolute morality, it is relatively a moral system and solves
many problems. To change it without changing the conditions
would be tantamount to driving the brave nomad women
into the towns to become the temporary wives of Chinese
rabble, priestly roués, and peripatetic Thibetans. Perhaps my
hinting that polyandry as a system is in many ways well suited
to the plateaux will evoke much unfavourable comment, but
there are good men, Roman Catholic and Protestant, priest and
layman, who have noticed the same difficulty.

“ The effect of the system on the women is another ques-
tion about which we cannot afford to be dogmatic. When
young the Thibetan women are often very pretty, but they
age quickly and become as weirdly ugly as the mediaeval
witches. To say that polyandry is alone responsible for this
change would be sentiment unsupported by facts; but un-
doubtedly this system, combined with hard work, loathsome
uncleanliness, and often grotesque head-dress tends to give a
great many women an inhumanly vile expression.

“The families on the plateaux are very small and many
women are barren. This is a blessing in disguise, owing to
the impossibility of the nomad country supporting more thana
very limited population, and the small amount of arable land
capable of relieving the congested centres. Polyandry is both

ea Oe Slr,

THE LAMASERIES JUST OUTSIDE TACHIENLU

TACHIENLU, THE GATE OF THIBET ary

directly and indirectly the cause of this limitation of offspring.
A glance at the system will show how these uncultured Mal-
thusians obtain their end: Three men, for instance, centre
their affections on one woman, who in her lifetime rears two
or three children. As monogamists each of these men would
have had his own wife and probably a total of fifteen
children. But another factor has to be taken into con-
sideration: polyandry not only limits a woman’s natural
fecundity, but in a great number of cases is the direct cause
of barrenness.

“About the domestic arrangements I cannot speak
authoritatively, but I have never heard internal discord used
as an argument against polyandry. It must often happen
that one or two husbands are away tending flocks, worshipping
at holy mountains, or robbing travellers. But this is an
accident ; the domestic equilibrium is rarely disturbed by
petty jealousies. The defloration of the bride or brides—
for there is no reason why two or more sisters should not
come into the community—is the right of the elder brother,
and the first child is, by courtesy, assigned to him; but the
child or children of the union are, in reality, a joint possession.
The girls in the community either follow their mother’s
example, or go into the towns and become the temporary
wives of Chinese, Lamas, or wandering merchants. In the
former case a dowry is given to the parents, but in the latter
the ‘fair one’ makes the most of her time and the simplicity
of her husband or husbands.

“ Polyandry in one form or other is probably practised
whenever Thibetan communities are found. Its existence
may be denied emphatically, but closer investigation will
only prove the wide distribution of the ‘ Miinchausen’
family. However, an exception may be allowed in the deep,
populous valleys of Eastern Thibet. Here individualism is
the rule, and new centres are formed and thrive without the
shadow of a grim Frankenstein disturbing them. So com-
pletely has the old dread of offspring been effaced that
marriage is always preceded by a tentative period, and
maternity alone establishes a girl’s right to be admitted
into her husband’s family. Here the quondam upholder of

218 A NATURALIST IN WESTERN CHINA

polyandry, realizing that the fruitful earth and the fruitful
woman bring wealth and strength respectively, becomes a
confirmed polygamist. To the student of ethnology this
metamorphosis suggests the permanency of the valley Thibetan
and the gradual absorption or total extinction of his mountain
brother.”

“CHAPTER XVII
SACRED OMEI SHAN

Its TEMPLES AND ITS FLORA

or Omei shan, is situated about long. 103° 41’ E., lat.

29°32’ N.,one day’s journey from the cityof Kiating. A
gigantic upthrust of hard limestone, it rises sheer from the plain
(alt. 1300 feet) to a height of nearly 11,000 above sea-level.
From the city of Kiating a fine view of this remarkable moun-
tain is obtainable during clear weather, the mirage of the
plain seemingly lending it additional height. Viewed from
a distance it has been aptly likened to a couchant lion de-
capitated close to the shoulders, the fore-feet remaining in
position. The down-cleft surface forms a fearful, well-nigh
vertical precipice, considerably over a mile in height! It is
one of the five ultra-sacred mountains of China, but the origin
of its holy character is lost in antiquity. We are told that
in a monastery here the patriarch P’u (an historical personage)
served Buddha during the Western Ts’in Dynasty (A.D. 265-
317). P’u-hsien Pu’ssa (Samantabhadra Bodhisattva), Mount
Omei’s patron saint, descended upon the mountain from the
back of a gigantic elephant possessed of six tusks. In one
of the temples (Wan-nien-ssu) there is a life-sized elephant
cast in bronze of splendid workmanship which commemorates
this manifestation. Upwards of seventy Buddhist temples
or monasteries (either word is applicable, since the buildings
are really a combination of both) are to be found on this
mountain. On the main road to the summit there is a temple
every 5 li, and they become even more numerous as the ascent
finally nears the end. These temples are controlled by abbots .
and contain upwards of 2000 priests and acolytes. The

219

“1 HE lofty and sacred eminence known as Mount Omei,

220 A NATURALIST IN WESTERN CHINA

whole of the mountain is, or rather was, church property,
much of the land on the lower slopes suitable for cultivation
having from time to time been sold away from the church.
Voluntary subscriptions are now the chief sources of revenue
of the religious houses, though many of the temples have
money as well as land endowments.

Many thousands of pilgrims, coming from all parts of the
Chinese Empire, visit this mountain annually. At the time of
my visit there were several pilgrims who had walked all the
way from Shanghai, some 2000 miles distant, for the express
purpose of doing homage before the shrines of Mount Omei.
Thibetans and even Nepalese make pilgrimages here. The
images and sacred objects are numberless, many of them being
of pure bronze or copper. Three mummified holy men,
lacquered, gilded, and deified, the elephant above mentioned,
and a tooth of Buddha are among the more interesting objects.
The tooth is about a foot long and weighs 18 English lb., and
is in all probability a fossil-elephant’s molar. On the extreme
summit of the mountain, the Golden Summit, as it is called,
are the ruins of an ancient temple which was built of pure
bronze. It is said to have been erected by the Emperor Wan-li
(A.D. 1573-1620), and was destroyed by lightning in 1819.
Since this catastrophe nine or ten abbots have come and
gone, but none has been able to collect enough money to
rebuild it. The mass of metal at present heaped around,
consisting of pillars, beams, panels, and tiles, is all of bronze.
The panels are particularly fine pieces of work. I measured
one panel which had dimensions as follows: 76 inches high,
20 inches wide, 14 inches thick ; some of the panels are slightly
smaller than this. All are ornamented with figures repre-
senting seated Buddhas, flowers, and scroll-work, and on the
reverse with hexagonal arabesques. Many of the panels have
been incorporated in one of the two small temples which now
stand on the crest of the precipice. Wan-li’s tablet, which
was contained in the ancient bronze temple, is to-day accommo-
dated in an outhouse along with fuel. The crown-piece is
detached and lies outside. This tablet is of bronze, but is
hollow. With crown-piece and pedestal it measures go inches
high, 32 inches wide, and 7 inches thick. Another grand relic

gt eee

‘LA 4 HLAID “TIVE “La 04 (VIMOLOUANT SOOM) NVANVA ASHNIHO AHL

SACRED OMEI SHAN 221

left to the tender mercies of the elements is a huge bell which
stands 54 inches high and is 120 inches round the middle.
On the edge of the cliff are two bronze pagodas, each about
12 feet high, and the remains of a third, which formed
part of the ancient temple. It is a saddening sight to gaze
around on these most interesting relics so ignominiously
neglected.

From the summit of Mount Omei, when the sky is clear and
clouds of mist float in the abyss below, a natural phenomenon
similar to that of the Spectre of the Brocken is observable. I
have never seen it myself, since rain fell almost continuously
during the week I spent on the summit, but it has been described
as a “golden ball surrounded by a rainbow floating on the
surface of the mists.’”’ This phenomenon is known as the
“ Fo-kuang ’”’ (=“‘ Glory of Buddha”). Devotees assert that it
is an emanation from the aureole of Buddha and an outward
and visible sign of the holiness of Mount Omei. The edge of the
precipice is guarded by chains and wooden posts, but pilgrims
in a state of religious fervour have been known to throw
themselves over on beholding the Fo-kuang. From this cause
the point isealled the ‘‘ Suicide’s Cliff.”’ It is the highest and
most vertical part of the precipice, which extends in a nearly
southerly direction for a couple of miles.

The first foreigner to ascend this famous mountain was
the late E. Colborne Baber, who visited it in July 1877, and
whose incomparable and accurate account of this region has
never been equalled.» Unfortunately Baber paid little or
no attention to the flora, nor did the equally distinguished
traveller and writer Hosie,*” who ascended Omei shan in 1884.
It was not until 1887 that any plants were collected on this
mountain. In that year it was visited by a Rhenish missionary,
who was also an industrious botanical collector—the late Dr.
Ernst Faber. During a fortnight’s stay this enthusiast made
a most interesting collection; which was found on critical
examination to contain no fewer than seventy novelties. In
1890 an English naturalist, Mr. A. E. Pratt, visited the moun-
tain and collected a few plants. Since Baber’s visit many

1 Royal Geographical Society, Supplementary Papers, vol. i.
2 Sir Alexander Hosie, K.C.M.G., H.B.M.’s Consular Service in China.

222 A NATURALIST IN WESTERN CHINA

hundreds of foreigners have ascended Mount Omei, but with the
exception of those of Faber and Pratt, there is no record of
any one having collected plants during their visits. For this
reason alone I hope this chapter will find justification. The
mountain and its temples have been well described by Baber
and others, and I have no desire to attempt to repeat
descriptions which have been made by abler pens than
mine. With this prelude I append the following record of
my visit :—

It was on the morning of 13th October 1903 that I set
out from the city of Kiating intent on investigating the flora
of this famous mountain. Traversing the highly cultivated
plain, which is intersected here and there by low hills, charmingly
wooded, the little town of Omei Hsien (alt. 1270 feet) was
reached at the close of the day. The next morning, after
journeying to li across the plain along a road shaded with trees
of Alder and Nanmu, we reached the village of Liang-ho-kou,
situated at the foot of the sacred mountain. Here the road
bifurcates and both paths lead by different routes to the
summit. They are paved with blocks of stone throughout, an
undertaking that must have entailed a vast expenditure in
labour and money, but it would be impossible to traverse
certain of the steeper parts unless paving existed. I ascended
by one of the routes and returned by the other, so that I saw
as much as was possible of the mountain and its rich flora.

Between Omei Hsien and Liang-ho-kou are a number of
truly magnificent Banyan trees (Ficus infectorta), known
locally as Huang-kou-shu. These trees shelter some old
temples and are of enormous size. I measured one, which
appeared to be the largest specimen ; it was about 80 feet
tall, and had a girth of 48 feet at 5 feet from the ground. We
also passed some fine trees of Oak (Quercus serrata) and Sweet
Gum (Liguidambar formosana). The sides of the rice fields are
studded with thousands of pollarded trees of the Chinese
Ash (Fraxinus chinensis) on which an insect deposits a valuable
white wax. The ditches were gay with the spikes of cream-
coloured, fragrant flowers of a species of Hedychium, the
golden-flowered Senecio clivorum, flowers of many kinds of
Impatiens, and other moisture-loving herbs.

Cs 7
—SS OE eT ee eee

SACRED OMEI SHAN 223

On leaving Liang-ho-kou the ascent began, and journeying
slowly three days’ hard climbing brought us to the ‘‘ Golden
Summit.”

For the purpose of grouping the flora it is convenient to
divide the mountain into two regions—(r) from the base to
6000 feet, and (2) 6000 feet to the summit (10,800 feet). Thus
divided the flora falls into two well-defined altitudinal zones.
The lower zone is made up of such plants as enjoy a warm-
temperate climate. Evergreen trees and shrubs predominate,
and in the shady glens and ravines Selaginellas and Ferns
luxuriate. Of these latter I, in one day, collected over sixty
species! The upper zone consists entirely of plants requiring
a cool-temperate climate. With the exception of Rhododendron
and Silver Fir it is composed almost entirely of deciduous
trees and shrubs and herbaceous plants. The belt between
4500 feet and 5500 feet may be termed the Hinterland.
Here the struggle for supremacy is most keen and the fusion
of the zones most marked. At 6000 feet the boundary line
is unusually well defined.

Cultivation extends up to 4000 feet, maize and pulse
being the principal crops, with rice relegated to the valleys
and bottom-lands. Plantations of Ash trees for the culture of
insect-wax extend up to 2600 feet. The foot-hills around the
base of the mountain are covered with Pine (Pinus Massoniana),
Cypress (Cupressus funebris), and Oak (Quercus serrata). The
sides of the streams which meander among these hills are
clothed with Alder (Alnus cremastogyne), Pterocarya stenoptera,
and the curious Camptotheca acuminata. Around the temples
and farmsteads Nanmu and tall-growing Bamboos abound ;
on the more exposed hillsides the climbing fern Gleichenia lin-
earis forms impenetrable thickets, and Onychium japonicum,
Melasioma candida, Musseendra pubescens, are common road-
side plants. At 3000 feet all these plants drop out and give
place toothers. Cunninghamia lanceolata, which occurs sparingly
in the valleys, gradually increases in number, and between
2500 and 4500 feet large areas are covered solely with this
invaluable Conifer. Apart from the Cunninghamia, the family
of Laurinee forms, between 2000 and 5000 feet, fully 75 per
cent. of the arborescent vegetation. This ‘‘ Laurel zone,”

224 A NATURALIST IN WESTERN CHINA

as it may be termed, is composed chiefly of evergreen trees
and shrubs, the genera Machilus, Lindera, and Litsea being
exceptionally rich in species. Within this zone also occur
the following interesting monotypic trees: Tapiscia, Carrieria,
Itoa, Emmenopterys, and Idesia. The evergreen Viburnum
cortaceum, with blue-black fruits, and five species of evergreen
Barberries are also met with found here.

In ascending any high mountain, more especially in these
latitudes, it is most instructive and interesting to note the
aggressiveness of the temperate flora. Mount Omei offers special
facilities for studying this phenomenon. Everything around
us looks so smiling that all nature seems to be at peace.
In these days, however, every one is alive to the fact that a
stern and relentless war of conquest is being continually waged
on all sides, and that every inch of ground is contested. It
is well that plants cannot speak, or the exultations of the victors
and the groans of the vanquished would be too much for human- —
ity to bear! But to note the struggle: The large-leaved
Cornel (Cornus macrophylla) manages to extend its area nearly
to the base of the mountain, being closely attended by several
species of Maple, among which Acer Davidii, with white-
striped bark, is particularly prominent. A Black Birch
(Betula luminifera), several species of Viburnum, Pyrus,
Malus, Rubus, and Prunus are also well to the fore ; but it is in
the Hinterland (4500 to 5500 feet) that the main battle between
the zones is fought. This narrow belt is extraordinarily rich
in woody plants. Of those peculiar to it I may mention
Pterostyrax hispidus, Pterocarya Delavay1, Euptelea pleiosperma,
Decaisnea Fargesii, Horse Chestnut (4sculus Walsonit),
and the monotypic genera Tetracentron, Emmenopterys, and
Davidia. At least five species of Maple occur with many fine
specimens of each. Several species of Evonymus, Holbeellia,
Actinidia, and Holly (I/ex) are also common. The bulk of the
Laurinee have given up the struggle, their place being taken
by Evergreen Oak and Castanopsis. In this belt monkeys are
common, and are fond of the blue pod-like fruit of the
Decaisnea, the shining black, flattened seeds of which, however,
I noticed they cannot digest.

On clearing a dense thicket and emerging on to a narrow

HYDRANGEA XANTHONEURA VAR. WILSONII, 15 FT. TALL

SACRED OMEI SHAN 225

ridge, 6100 feet above sea-level, a magnificent view presented
itself. Above towered gigantic limestone cliffs nearly a mile
high ; below spread valleys and plains filled with a dense,
fleecy cumulus, through which the peaks of mountains peered
like rocky islands from the ocean’s bed; to the westward
the mighty snowclad ranges of the Thibetan border, 80 miles
distant as the crow flies, presented a magnificent panorama
stretching northward and southward as far as the eye could
range. The contrast between the floral zones was equally
startling and impressive. Below, until lost in the clouds, was a
mass of rich, sombre, green vegetation ; above were autumnal
tints of every hue, from pale yellow to the richest shades of
crimson, relieved by clumps of dark green Silver Fir. The
whole scene was bathed in sunlight, a gentle zephyr stirred the
air, and gorgeous butterflies flitted here and there seemingly
unconscious of winter’s near approach. The stillness and quiet
was most solemn, and was broken only by the warbling of an
occasional songster in some adjacent tree or bush. It was
indeed a never-to-be-forgotten scene !

At 6200 feet the Cunninghamia gives up the fight, having
struggled nobly until reduced to the dimensions of an insignifi-
cant shrub. A Silver Fir (Abies Delavayi) next assumes the
sway, and right royally does it deserve the sceptre, for no more
handsome Conifer exists in all the Far East ; its large, erect,
symmetrical cones are violet-black in colour and are usually
borne in greatest profusion on the topmost branches. The
temples on the higher parts of the mountain are constructed
almost entirely of the timber of this tree. It is first met
with on Mount Omei, at 6000 feet, at which altitude it is of no
great size and unattractive in appearance; at 6500 feet itis a
handsome tree. It is, however, between 8500 and 10,000 feet
that this Silver Fir reaches its maximum size. In this belt
hundreds of trees 80 to 100 feet tall, with a girth of 10 to 12
feet, are to be found. Hemlock Spruce (Tsuga yunnanensis)
occurs sparingly, but always in the form of large and shapely
trees. An occasional Yew tree (Taxus cuspidata, var. chinensis)
and, on the summit, dwarf Juniper (J. squamata) complete
the list of Conifers growing on the higher parts of this moun-
tain. The unspeakably magnificent autumnal tints already

VOL. I.—I5

226 A NATURALIST IN WESTERN CHINA

referred to are principally due to numerous species of
Viburnum, Vitis, Malus, Sorbus, Pyrus, and Acer, together
with Enkianthus deflexus, which surpasses all in the richness
of its autumn tints of orange and crimson.

At 6200 feet the ascent becomes increasingly difficult,
and having surmounted a formidable flight of steps, 800 feet
high, we were glad to rest at the temple of Hsih-hsiang-chiih.
All the temples on Mount Omei occupy lovely and romantic
situations, but none more so than this, which has one side
flush with the edge of a precipice, and the others sheltered by a
grove of Silver Fir. The hospitable priests regaled us with tea
and sweetmeats and entertained us with much that was curious
and amusing. They claimed that it was at this particular
place that P’u-hsien Pu’ssa alighted from his elephant to allow
the footsore animal to bathe in a near-by pool; the spot
to-day is marked by a cistern.

Immediately on leaving this temple two steep flights of steps,
followed by a slight descent, led us to a small wooded plateau
which shelves away from a vertical precipice. Hereabouts
Sorbus munda, with white fruits, was a most conspicuous
shrub. A climbing Hydrangea (H. anomala) reaches to the top
of the tallest trees. Several other species of Hydrangea grow
epiphytically on the larger trees and so also do two or three
species of Sorbus. Rhododendrons are fairly abundant,
more especially near the edge of the precipice. The first few
Rhododendron bushes were noted growing at 4800 feet, and
altogether I gathered thirteen species on this mountain.
But as compared with the region to the westward Mount Omei
is poor in Rhododendrons. The same is true of Primulas,
of which four species only were met with.

At gooo feet the most difficult stairway of all occurs, and
I was fairly exhausted when the top of it was reached at
10,100 feet. Winter had laid his stern hand heavily here,
and most of the woody plants were leafless. At 10,000 feet
Bamboo-scrub puts in an appearance and increases as the
summit is neared until finally it crowds out nearly everything
else and forms an impenetrable jungle about 4 to 6 feet high.

From the top of the last stairway an easy pathway of
planking leads to the summit, which we reached just as the

SACRED OMEI SHAN 227

sun was setting behind the snowclad ranges of the Thibetan
border.

A perfect night succeeded the day, and our hopes were
high for the morrow. Alas! a thick fog and a drizzle of rain
was what we awoke to find. A terrible precipice in front and
a more or less shelving away behind was all we could make
out of the lay of the land. To find out what the summit is
really like, a long walk was undertaken, but resulted in little
beyond a thorough drenching. The mountain-top is somewhat
uneven, sloping away from the cliffs by a fairly easy gradient.
It is everywhere covered with a dense scrub, composed mainly
of dwarf Bamboo, with bushes of Willow, Birch, Sorbus, Bar-
berry, Rhododendron, Spirza, and Rosa ometensis interspersed.
Near the watercourses these shrubs are more particularly
abundant. Trailing over the scrub Clematis montana, var.
Wilsonit,is verycommon. At least five species of Rhododendron
grow on the summit, but, judging from the paucity of fruits,
they flower but sparingly. In places sheltered from the winds
fine groves of Silver Fir remain, but in the more fully exposed
sites these trees are very stunted and weather-beaten. The
dwarf Juniper, with twisted, gnarled stems, is also plentiful in
rocky places.

Around the temples small patches of cabbage, turnips, and
Irish potato are cultivated, and several favourite medicines
are grown in quantity, such as Rhubarb, “‘ Huang-lien”’
(Coptis chinensis), “‘ Tang-shén,”’ and ‘‘ Tang-kuei.”’

Here and there on the mountain we passed hucksters’
stalls, on which various local products were exposed for sale.
These consist chiefly of medicines, porcupine quills, crystals
of felspar, sweet tea, and pilgrim staves. The latter, made
from the wood of an Alder (Alnus cremastogyne), are carved
in representation of fantastic dragons and Buddhas. The
sweet tea isa peculiarity of Mount Omei, being prepared from
the leaves of Viburnum theiferum.

CHAPTER XVIII
THROUGH THE LAOLIN (WILDERNESS)

NARRATIVE OF A JOURNEY FROM KIATING TO MALIE,
vIA WA-wu SHAN

EAVING the city of Kiating on 4th September 1908, we
followed the main road to Yachou Fu and stayed for the

night at Kiakiang Hsien, a small city, altitude 1200
feet, 70 li from our starting-place. It had rained heavily in
the early morning, but cleared just before we set out, and was
cool and fine, although dull the whole day. The road is broad,
mostly well paved, and leads through a rich and highly cultivated
region. Around Kiating the rice had been harvested, much
of the land reploughed, and another crop, chiefly buckwheat
and turnips, planted. A few miles beyond this city, however,
the rice crop was not so forward, and though a portion was
being reaped the bulk would not be ripe for some weeks.

Around the margins of rice fields trees for the culture
of insect white-wax are abundantly planted. Pollarded Ash
(Fraxinus chinensis) were chiefly noticeable, but in places trees
of Privet (Ligustrwm lucidum) are used for this industry.
Much of the wax had been collected, but in one place we were
fortunate enough to witness the process and obtain photo-
graphs. (In Vol. II., Chapter X., this interesting industry is
fully described.) Sericulture was very much in evidence, and all
the alluvial flats are planted with Mulberry trees, but trees of
Cudrania are not common. In this region in particular the
silkworms are fed on the leaves of both these trees ; the people
claim that this mixed diet results in a stronger kind of silk.

The Szechuan Banyan (Ficus infectorta) is the most striking
tree hereabouts ; its widespreading umbrageous head usually
shelters some wayside shrine. Venders of cakes, pea-nuts,

228

THROUGH THE LAOLIN (WILDERNESS) 229

and fruit are also to be found occupying some temporary
stall under these beautiful trees. The road skirts the sides of
low hills of red sandstone for considerable distances, and is
mainly parallel with, and in full view of, the Ya River. The
hills are clad with common Pine (Pinus Massoniana), Cypress
(Cupressus funebris), a jungle-growth of low shrubs, and the
scandent Gleichemia linearis. Small trees of Oak and Sweet
Chestnut and larger ones of Alder are also common. Groves
of tall-growing Bamboos, of course, are everywhere abundant.
In the sandstone cliffs are very many square-mouthed Mantzu
caves ; the scenery is distinctly pretty and pleasing.

We left Kiakiang at 6.30 a.m. the following day, and
quickly reached a ferry, where we crossed over the Ya River,
a broad, stony, shallow stream. Quite near this place are
two really fine and very large old temples known as Ping-
ling-ssu and Kuei-ling-ssu. The first named, in particular,
contains some very fine idols; both, however, have a very
deserted and neglected appearance, and give the impression of
“glories departed.’”’ The sandstone cliffs at the ferry are
highly sculptured, but are rapidly weathering away, much
of the work being undecipherable and hidden by vegetation.

The li proved very long, and we did not reach Che-ho-kai
until 7 p.m., going steadily the whole day. The distance is
80 li, and three ferries, which hinder considerably, have to be
crossed. Near the city of Hungya Hsien, which we sighted
in the late afternoon, large plantations of Ash trees for the
culture of insect white-wax abound. Rice is everywhere
the great crop; the yield was heavier than usual, and the
people were busy reaping and threshing it. Fine Banyan
trees are plentiful, Alder is abundant, and handsome Nanmu
trees are not infrequent around temples and houses. We
also noted a small tree of the Hog-plum (Spondias axillaris)
bearing quantities of its oblong, yellow, edible fruits. The
vegetation generally is similar to that around Kiating, but
the Chinese Fir (Cunninghamia lanceolata) is more common
and Pine and Cypress less so.

Che-ho-kai, alt. 1400 feet, is a large and important
market village, situated on the right bank of the Ya River.
The inn is very fair. I occupied a large room overlooking

230 A NATURALIST IN WESTERN CHINA

the river, but, as I discovered later, with a piggery and
latrine below.

The next day we began our real journey. Instead of
following the main route to Yachou Fu by crossing over the
river, we ascended the right bank for a couple of li beyond
Che-ho-kai, and then crossed a considerable affluent of the
main stream. Rafts of good-sized poles of Chinese Fir
descend this tributary from Liu ch’ang, a market village, and
ordinary bamboo-rafts ascend to this place. After climbing
to the tops of some low hills the road zigzags around con-
siderably through fields of rice and wooded knolls, and
affords an unusually fine view of the Ya Valley. © Passing
the tiny market village of Tung-to ch’ang we reached Kuang-
yin pu (or ch’ang) at 10.45 a.m., having covered 30 li.

From Kuang-yin pu we engaged in a steep ascent over a
well-paved if narrow road, and after four hours’ climbing
reached the summit of the Fung-hoa-tsze, alt. 4100 feet.
This ridge is of red sandstone throughout, and is well timbered
with small trees of the Chinese Fir. This conifer abounds
on the slopes flanking the roadway to the top of the pass and
forms pure woods. Though the timber is of no great size,
the area covered with this tree compares most favourably
with any other I have seen. Where timber is scarce the
jungle growth is very thick, warm-temperate in character,
and of little interest.

Descending, at first steadily, through knolls covered with
Chinese Fir and the densest fern jungle composed of
Gleichenia linearis I have ever seen, we soon reached an area
under maize. From this point a steep descent led to a
cultivated flat, then, after winding through rice fields with
tiny wooded hillocks on all sides, we crossed a neck and
entered the hamlet of Liang-ch’a Ho, alt. 2350 feet, and
65 long li from our starting-point. We found very decent
accommodation, all things considered, but mosquitoes were
most unpleasantly numerous and hungry.

It rained very heavily during the early morning of the
next day, so we delayed our starting until eleven o’clock.
We found all the streams in flood, and to cross one larger
than the ordinary we had to engage local assistance. After a

NA ONILVIN “TIVM ALIO AHL NO WIdNAL

THROUGH THE LAOLIN (WILDERNESS) 231

rather steep ascent of 500 feet from Liang-ch’a Ho we crossed
a narrow ridge and descended to the market village of N’gan
ch’ang. This is a poor place, partly in ruins, situated on
the right bank of the stream which unites with the Ya Ho,
2 li above Che-ho-kai. On leaving N’gan ch’ang we ascended
the right bank of the stream to Pao-tien-pa, alt. 2600 feet.
This scattered hamlet possesses no inn, but we found quarters
in a schoolhouse devoted to the ‘‘ New Learning”’ (1.e.
Western Knowledge). A scholar from this place had recently
gone to Japan to increase his store of knowledge, and the
dominie was very proud of this success. This hamlet boasts
a ruined pavilion, a temple, and a stone gateway, evident
signs of former prosperity.

During the short journey of 25 li the road led through
fields of rice, bounded by wooded knolls and sandstone
bluffs. The flora was of little interest ; Idesia polycarpa and
Kalopanax ricinifolium are fairly common in places, but the
trees are of small size. Alongside the ditches and roadway
the handsome Lycoris aurea abounds, and the golden-yellow
flowers with recurved, wrinkled, perianth-segments made
a gay display. Its red-flowered counterpart, L. radiata, also
occurs, but is much less frequent. The local name for this
plant is ‘‘ Lao-wa-suan,” which signifies “‘ Crow’s foot Onion,”
a very apt term in so far as the shape of the flower is con-
cerned.

The following day was fine but hot, and more or less
cloudy. With only 35 li to cover, we journeyed slowly after
making an early morning start. A moderately steep ascent
of 15 li brought us to the summit of the Tsao shan, alt.
4100 feet. This ridge is covered with an uninteresting jungle
of coarse grass and scrub, with odd trees of Chinese Fir, but
in the ascent I gathered specimens of a fine new species of
Castanopsis.

From the summit of Tsao shan we obtained our first view
of the Wa-wu shan, an extraordinary-looking massive mountain,
singularly like Wa shan in contour, resembling a huge ark
floating above clouds of mist. Following an easy path which
led through fine woods of Evergreen Oak, Nanmu, and
Castanopsis we descended to Ma-chiao-kou, where there is an

232 A NATURALIST IN WESTERN CHINA

iron suspension bridge over a wide torrent. This hamlet
consists of one large house and a mill, where a specially good
and tough bamboo paper is made, which is used at Yachou
Fu for wrapping up brick tea. The bamboo is obtained
from the surrounding mountains, and is a species with dull
green culms about the thickness of a man’s thumb, growing
12 to 15 feet tall. Oncrossing over the bridge, I photographed
a fine specimen of Alniphyllum Fortunei, one of the rarest of
Chinese trees. A short steep ascent, then a rather drawn-
out descent, ultimately brought us to the banks of a clear-
water stream of considerable size, which we crossed by an
iron suspension bridge 50 yards long, and soon reached the
market village of Ping-ling-shih, alt. 2900 feet. This is
a small and dirty place of about 50 houses, situated on the
left bank of a stream which joins the Ya Ho, some ro li below
Yachou Fu. It is in Hung-ya Hsien, in full view of Mount
Wa-wu, and the most important place in the Laolin
(Wilderness), as this region is denominated.

The flora of the day’s journey was rather more interesting
than heretofore. Wooded knolls are the rule. Evergreen
trees, more especially Oak and Castanopsis, are very general,
and of large size. I gathered four species belonging to the
latter genus, all handsome umbrageous trees. A fine specimen
of the curious Hazel-nut (Corylus heterophylla, var. crista-gallt),
60 feet tall, 5 feet in girth, was one of the most interesting trees
noted. The nut in this variety is hidden in a crested cup.
The Chinese Fir is most abundant, being the only Conifer met
with. The absence of Pine and Cypress since leaving the
valley of the Ya River has been a most remarkable feature.
The country generally is very broken, the sandstone bluffs
bold, and clad with the usual jungle growth wherever trees
are sparse.

In order to ascend Mount Wa-wu from Ping-ling-shih it was
necessary for us to make a detour from our intended route.
The summit was said to be 70 li distant, but, owing to the steep
and difficult road, two days are required to cover this. We
left behind all our spare gear and arranged what it was necessary
to take into light loads. The road on clearing Ping-ling-shih
ascends a rock-strewn tributary of the main stream, through a

THROUGH THE LAOLIN (WILDERNESS) = 233

region given over to rice fields and cultivation generally. At
eleven o’clock in the forenoon after traversing 30 li we reached
the large temple of Tsung-tung-che, alt. 4000 feet, situated
at the foot of the real ascent of Mount Wa-wu. This temple is
built of wood, very old, and in poor repair. A priest and one
attendant were in charge ; the rooms, though dingy and damp,
were alive with fleas. But since there is no other accom-
modation between this place and the summit it was necessary
to make the best of things. I had my bed arranged in a large
hall where three huge images of Buddha looked down benignly
upon me. During the morning occasional showers fell, but
in the afternoon a steady downpour set in, which added to the
cheerlessness of our roomy but dilapidated quarters.

Just before reaching the temple we passed through the
hamlet of Tung-ch’ang Ho, where there is a very large iron
foundry employing a considerable number of men. Iron ore
is common in the surrounding mountains, and costs 12,000 to
13,000 cash per 10,000 catties. Every 10,000 catties of ore
yields about 4000 catties of pig iron, which was said to be of
good quality, and sells for 2500 to 3000 cash per picul of 100
catties. The smelting is done in furnaces heated by charcoal,
which costs at the foundry 12 to 13 cash per catty. Most of
the smelting is done during the winter, the summer months
being given to the collecting of charcoal and iron ore. Large
iron cooking-pans are also made here in considerable quantities.

Copper is also found in the same range as the iron ore, but
on the opposite side. Formerly it was worked and smelted
here, the name Tung-ch’ang signifying ‘“‘ copper-shop’’ or
factory. From what I could learn the industry was abandoned
some ten years or more ago when copper mining became a
Government monopoly controlled by the officials. The people
told me that they could not produce copper on paying lines
under Tls. 35°00 to Tls. 36°00 per picul. The officials would
only pay Tis. 28-00, consequently copper smelting was given
up and replaced by that of iron. A hard, smokeless coal occurs
in the neighbourhood, but is not much used. Altogether, this
Tung-ch’ang Howith its iron foundry, coal mines, andabandoned
copper workings constitutes an interesting mining centre.

Around the temple are many fine trees of Castanopsis,

234 A NATURALIST IN WESTERN CHINA

and the finest specimen of the interesting monotypic Tapiscia
sinensts I have seen. This tree is fully 80 feet tall, with a girth
of 12 feet. Many fine trees of the Kuei-hwa (Osmanthus
fragrans) are planted in the temple grounds, and were in full
flower, scenting the atmosphere all around. Near streams
Alder (Alnus cremastogyne) is abundant, and on the hills the
Chinese Fir is common.

It rained heavily all night, and a drizzle fell when we set
out next morning at 6.30 a.m. This drizzle developed into
a steady downpour as we advanced, and continued with in-
creased violence the whole day. The road is atrocious from
the very beginning. For the first 2500 feet there is a semblance
of a track, some of it being made by laying pieces of split
timber crosswise. The next 2500 feet is a rough scramble
upwards through cane-brake and brushwood until the summit
is reached. The ascent is up the north-north-east angle of the
mountain, and though never really dangerous is always very
difficult. We dragged ourselves upward by grasping shrubs,
and it was a marvel to me how the coolies with their loads
managed to overcome the ascent. The foothold was pre-
carious, and it was often a case of one foot forward and two
backward !

On reaching the summit we followed a winding path
for 12 li to the temple of Kwanyin-ping, alt. gtoo feet. The
mountain-top is undulating, park-like, and covered with an
impenetrable jungle of Bamboo-scrub about 6 feet tall, arising
from a floor of Sphagnum moss. Silver Fir (Abies Delavayi),
called Lien sha, z.e. Cold Fir (signifying that it is only found in
cold regions), is scattered through in quantity, but I saw no
really handsome trees, all of them showing the effects of wind-
storms, age, and decay. The pathway across the summit is
about 24 feet wide, paved throughout with split timbers, though
here and there fallen Silver Fir trees, slightly notched and
flattened, have been utilized in making this roadway. We
passed three temples in absolute ruins, but saw no signs of
life of any description. The heavy rain and dense mists
obscured all views, and I saw nothing of the country; or
scenery except what was encompassed in a perspective of 30
yards. Drenched to the skin but mildly describes the plight

VILLAGE OF PING-LING-SHIH, MT. WA-WU IN DISTANCE

THROUGH THE LAOLIN (WILDERNESS) = 235

in which we reached the temple. Our gear arrived equally
wet some two hours afterwards, and we were some time getting
things dry and shipshape.

The temple of Kwanyin-ping is very large, with many
outhouses, and is built entirely of wood. It contains many
scores of idols, but is in a poor state of repair. The main road
hither is from Yungching Hsien, distant 120 li. During the
Chinese fifth and sixth moons (June, July) some two to three
thousand pilgrims visit this temple, but for the rest of the year
it has scarcely a visitant. The priests reside at Yungching
Hsien except at the pilgrim season, a novice being left in charge.
This novice lives all alone, without even a dog for a com-
panion. As a reward he receives 1} catties of rice per diem
as rations and 2000 cash (say, half a crown) per annum
salary! In spite of his lonely life, and he has been in charge
for three years, this novice was a very cheery person. He
moved around quickly, had a ready smile, and chanted hymns
and prayers wherever he went. He speedily made a fire for
us to dry ourselves and clothing, and made himself generally
useful. His cheery influence made itself felt, and my men soon
ceased their grumbling over the vileness of the road and my
madness in wanting to visit such a place. The novice told us
that-the first temple was built on this mountain during the
Eastern Han Dynasty (A.D. 25-87). At one time there were as
many as 40 temples here, but during Ming times the majority
were destroyed, and the temple ornaments melted down. To-
day there are only two in any sense habitable, and in one
only is aman kept the year round. This same authority vouch-
safed the information that the heavy rains were due to the fell-
ing of timber ; the country folk holding this view were opposed
to further cutting, but the Magistrate at Yungching Hsien
pooh-poohed the idea, and insisted on the slaughter being
continued, with the result that torrential rains fell every day
except in winter, when snow took their place.

The next morning opened dull and threatening, but eventu-
ally the sun came out and we enjoyed a fine day. The temple
stands in Hungya Hsien, and is situated on the edge of a
precipice. The views looking north-east over the Ya Valley
and west to the Thibetan alps are very fine; some almost

236 A NATURALIST IN WESTERN CHINA

vertical limestone cliffs near by the temple are covered in a
remarkable manner with Silver Fir. The whole surroundings
are wildly romantic, and it is small wonder that the place is
deemed sacred and holy.

Wa-wu shan or Wa shan, as it is much more frequently but
erroneously called, is one of three sacred mountains, forming
the three corners of and enclosing a triangular tract of wild,
sparsely inhabited country known as the Laolin (Wilderness).
On even the most recent maps the term Lolo is written across
this region, but as a matter of fact no Lolos live here. The few
people found here are Chinese—peasants, charcoal-burners,
miners, and medicine-gatherers. The other two mountains,
Omei shan and Wa shan, have been described by former
travellers, but, with the possible exception of some Roman
Catholic priest, my visit was the first undertaken by any
foreigner to the summit of Wa-wu shan.

Like its sister mountains, Wa-wu is a gigantic upthrust of
hard limestone, but of lesser altitude than they, being only
9200 feet above sea-level. It is a huge oblong mass, composed
of a series of vertical cliffs 2000 feet and more sheer, reared on a
base of red sandstone rocks. The summit is flat with sand and
mudstone shales scattered about, and is said to be 60 li long by
40 li wide, but this is an exaggeration—30 li by 15 li being,
probably, nearer the truth. Its appearance from a distance
has already been given, and the nearer the approach the more
impressive become the perpendicular walls of rock. The
similarity in appearance between this mountain and the real
Wa shan has also been alluded to, and I strongly suspect that
the mountain seen from the summit of Omei shan and called
Wa shan is really this Wa-wu shan. Their extraordinary
vertical sides and flat summits make these two peaks unique
among the mountains of Western China.

From a botanical standpoint Mount Wa-wu proved dis-
appointing. In the first place, its altitude was some 1500 less
than I had hoped for. Secondly, all the mixed timber has been
felled for making charcoal and other purposes, leaving only a
dense shrubbery in which variety is not great. Thirdly, the
paucity of Conifere on the summit other than Silver Fir and
the impenetrable thickets of slender Bamboos which render any

THROUGH THE LAOLIN (WILDERNESS) = 237

extended exploration impossible. The flora generally is that
common to every mountain in this region of similar altitude,
but, of course, it has a certain number of species peculiarly its
own in the same way as every other mountain in China has.
The outstanding feature is its wealth of Bamboo-scrub ; its
speciality, the abundant carpet of Sphagnum moss on the
summit. This moss occurs on Wa shan and virtually on all
the other mountains of this region, between 8000 and 11,500
feet, but nowhere have I seen it so luxuriantly plentiful as on
Wa-wu shan.

The day being fine and clear I obtained good views of every-
thing. The summit is made up of low, wooded hillocks, tiny
dales, and glades. Here and there it is a morass, and on one
occasion from such a place we flushed a Solitary Snipe. The
feathery Bamboo-culms are very beautiful, and the scattered,
often sentinel-like, old trees of Silver Fir quite picturesque. A
few trees of Hemlock Spruce occur, but their number is in-
finitesimal. Some of the Silver Fir were roo feet tall, and ro
to 12 feet in girth, but all such trees contain much dead wood.
_Here and there saplings are common, but they can scarcely
compete with the Bamboo in the struggle for possession.
At one time Davidia (both hairy and glabrous-leaved forms),
Tetracentron, Magnolia, various species of Acer, Pyrus, Cas-
tanopsis, Evergreen Oak, and Laurinew covered the lesser
slopes, but, to-day, these are all represented only by bushes
which have sprung up from the felled trees. Rhododendrons
are fairly numerous, and I noted about ten species. One of
these forms a tree 25 feet tall and 3 to 4 feet in girth. (It
proved to be new, and has been named in honour of the
Rey. Harry Openshaw, of Yachou Fu.) Various Araliads are
plentiful, and were mostly in ripe fruit. The Chinese Fir
ascends to 4500 feet altitude, and very few of the evergreens
other than Rhododendron extend above 6000 feet. Herbs,
of course, occur, but none of any great value or interest.

A local industry of considerable importance at the season
of the year my visit occurred, and for six weeks previously, is
the collecting and preparing of young Bamboo shoots for
culinary purposes. The species in request is one having
culms the thickness of a man’s thumb, and growing Io feet tall.

238 A NATURALIST IN WESTERN CHINA

The young shoots are culled when 8 to 12 inches above the
ground, stripped of their sheaths and apices, leaving only the
white, brittle succulent central part. These are boiled in
water, then removed, suspended from rafters in a close chamber
and dried by means of heat from steady-burning fires fed
from locally made briquettes. When thoroughly dry they are
packed in bales and carried to Chengtu and other cities, where
they are esteemed a great delicacy. We saw fully a score of
rude shanties where this industry was in full swing. On the
spot the raw shoots are bought for 6 cash per 16-oz. catty,
the collecting being done by contract. The prepared article,
known as “ Tsin-tzu,”’ sells at Ping-ling-shih for 8 to 9 Tls. per
100 catties of 20 oz. each. This region is.famed far and wide
for its product of dried Bamboo shoots, and the industry affords
employment for a large number of people.

Many wild animals, including Budorcas, Serow, Goral,
Leopard, and Bear were said to occur on Wa-wu, but hunting
them would be almost an impossibility. We saw no animals
of any kind, but I do not doubt the reports given as to their
presence on this jungle-clad mountain.

A day sufficed for our investigations, and leaving the next
morning about nine o’clock, a hard day’s march brought us
back to Ping-ling-shih at 5.45 p.m.

Our object being to traverse this Laolin country through
its greatest width to some point in the valley of the Tung
River, we readjusted our loads, and the following day continued
our march. Crossing the tributary stream by a rickety iron
suspension bridge, we soon left Ping-ling-shih behind. The
path ascends the right bank of the main stream frequently
high above its waters, and at times some little distance removed.
As soon as it enters limestone country the river becomes
gorged. The li were long, the road rough, and it took us five
hours to cover 30 li to Yiieh-ch’a-ping. This place consists of
a single house, situated near where the stream bifurcates.
One branch and a companion roadway leads off in a south-
easterly direction, and by this track it is possible to reach
Huang-mu ch’ang. The path we followed ascends the branch
which swings round from the south-west, skirting the base of
Wa-wu shan. After crossing a cultivated shoulder we plunged

QAVAWTHC SHldv)
Ula AHATIS HLIM GHYHLOTID SHAITO ‘AM-VM ‘LIN JO LININAS NO WIdNHL NOMA ATA

THROUGH THE LAOLIN (WILDERNESS) = 239

into a deep, narrow gorge, traversing a difficult roadway usually
high up above the stream. The scenery is very fine—steep
cliffs, either bare or clothed with shrubs, on every side.
Journeying slowly we reached the solitary house at Chang-ho-
pa, alt. 4000 feet, about 5 p.m., having covered 50 li.

During the day’s march we saw a number of interesting
trees, and obtained specimens and photographs. Carrieria
calycina, a widespreading flat-topped tree, is very common in
rocky places by the stream-side, and was laden with its torpedo-
shaped, velvety-grey fruit which was not ripe. The Tapiscia
is fairly numerous, but the trees are of no great size. Perhaps
the most noteworthy tree of this region is Meliosma Kirkit,
which has a shapely port, rigid branches, and handsome pinnate
leaves, 2 feet long. Evergreen Oak, various Laurinea, tall-
growing Bamboos, and a Fan Palm (Tvachycarpus excelsus)
are abundant, denoting a mild, moist climate. The Chinese
Fir is the only Conifer. The quantity of this useful tree and
the many fine and shapely specimens were among the leading
features of this trip. We had left rice behind at last, and
entered a region where only maize is grown. Every available
bit of land is under cultivation, but the district is very sparsely
populated. A certain amount of tea is grown around Ping-
ling-shih, but the industry is of little importance commercially.

The people at Chang-ho-pa informed us that the road before
us was much worse than that which we had traversed. For
the first ro li after leaving our lodgings I thought they had
dissembled, but afterwards the truth of their statement was
only too evident The stream flows through a narrow, wild
gorge or succession of gorges ; the road is either some hundreds
of feet above the stream, or down by the water’s edge. The
“ups and downs”’ repeat themselves with monotonous and
irritating frequency. The path is very much overgrown with
weeds and brush, always very narrow, the ascents and descents
precipitous and difficult. It is misleading and foolish to term
it a “road.” Goats would make a better pathway, did they
travel it frequently !

The scenery is grand, though mists and a drizzle of rain
did their best to rob us of its enjoyment. The cliffs are in
the main clothed with shrubby vegetation, but alongside the

240 A NATURALIST IN WESTERN CHINA

stream large trees are common. The climate is evidently
very moist and warm, since broad-leaved evergreens abound.
Perhaps the most common shrub or small tree is a Walnut
or Chinese Butternut (Juglans cathayensis), which has six to
twelve fruits arranged in a raceme, and leaves up to a yard
in length. The Horse Chestnut (4sculus Wilson), Yellow-
wood (Cladrastis sinensis), Hornbeam, and various Maples
are among the more interesting trees hereabouts. Clearings
and abandoned cultivated areas are overgrown with the
handsome Anemone vitifolia, var. alba, which was 4 to 5 feet
tall, and bore myriads of large attractive flowers. This herb
made a wonderful display, and I do not remember having seen
it so luxuriant elsewhere in my travels. Beneath cliffs drip-
ping with moisture, Begonias, Impatiens, Ferns, and various
Cyrtandrece in masses made pretty effects. The Chinese Fir
ceases at 4800 feet altitude, but limestone country is not to
its liking, and the trees quickly become scarce on quitting the
red sandstone.

Houses and patches of cultivation are few and far between,
but it is surprising that any should be found in such a pre-
cipitous country. We put up for the night at one of the three
small houses which collectively form the hamlet of Peh-sha Ho,
altitude 5000 feet, 40 li from Chang-ho-pa. The house is built
on a steep bank, overlooking a point where the stream divides,
the larger branch flowing from a southerly direction.

On leaving Peh-sha Ho we headed for the source of the
lesser of the two streams—a mere mountain torrent. Our
difficulty all day was in discerning the track and keeping to it.
I lost it early in the morning, and wasted two hours in a jungle

of Bamboo ; my Boy had the same misfortune in the afternoon. —

The collecting of Bamboo shoots is an industry here as on the
other side of Wa-wu, and the tracks made by men engaged
in this are many. The path we endeavoured to follow was
frequently less well-defined than these tracks and, moreover,
was overgrown with vegetation. It crossed the torrent many
times, but the fords were difficult to discover. We passed
neither house nor person, and perforce had to explore our own
route. It rained heavily the whole day, increasing our diffi-
culties and discomforts.

ae ne a ee

THROUGH THE LAOLIN (WILDERNESS) = 241

Our objective for the day was some lead mines, but early
in the afternoon it became evident that we could not reach
them before night was well advanced. Darkness overtook us,
and we had visions of spending the night in the woods, which
bound the torrent; suddenly, however, the welcome glare
from acharcoal-burner’s hut gladdened our hearts. Scrambling
somehow down the steep slope, and across the torrent, we
quickly reached this haven of shelter. It proved a wretched
hovel, but the warmth from the charcoal pit was comforting
since we, and all our belongings, were wet through. My bed
was fixed up in a shed where prepared charcoal was stored,
the men taking possession of the hut, thankful that a refuge
of some sort had been found.

Much of the day’s journey had consisted in struggling
through brush and Bamboo, and by way of variety wading
the torrent was thrown in. Whenever the mists lifted, cliffs
and crags, densely covered with vegetation, were to be seen
onallsides. The flora is apparently rich, but it was impossible
for us to investigate it. All the larger trees have been cut
down and converted into charcoal. Davidia, Tetracentron,
Cercidiphyllum, and Cornus sinensis are common as_ bushy
trees by the wayside ; Maples are plentiful, and stout climbers,
such as Actinidia, Clematoclethra, and Holbeellia are rampant.

Two men were in charge of the charcoal pits. They told us
the place is called Tan-yao-tzu, and that we had only covered
30 li! All the hardwood trees having been felled they are
now forced to use the softwood of Silver Fir and Hemlock
Spruce, which, they said, grow in quantity on the higher crags.
The charcoal is all used for smelting lead at the mines.

The roof of the shed leaked freely, but an arrangement of
oil-sheets kept my bed fairly dry, and I enjoyed a good night’s
sleep. Awaking soon after daybreak we found it was still
raining. Leaving the hut (alt. 7250 feet), we crossed two
branches of the stream and scrambled up the mountain-side
to rejoin the track. Soon afterwards we entered a narrow
scrub-clad valley, at the head of which a precipitous, circuitous
ascent brought us to the top of a ridge where the lead mines
are situated. In the ascent, Rhododendron Hanceanum and
two other species are particularly abundant, forming thickets ;

VOL. I.—16

242 A NATURALIST IN WESTERN CHINA

Lonicera deflexicalyx is also plentiful, and was a wealth of
orange-coloured fruit. On humus-clad rocks a pretty little
prostrate Gaultheria with snow-white fruits iscommon. The
hovels at the lead mines are miserable structures, but we were
glad of their shelter from the rain and cold. The whole moun-
tain appears to be full of lead, the ore (galena) being very rich.
Well-shored adits are carried for considerable distances into
the mountain-side, and the ore is brought out in baskets fitted
onrunners. The galena is pounded by hand labour into small
particles; the lead is obtained by levigation and stored in
large wooden vats. Subsequently it is melted into large
oblong ingots, in which form it is carried to Chengtu
and Sui Fu. The freight down to the nearest waterway is
very considerable. Lead has been worked in this neighbour-
hood for many years, and the mines are owned by a man who
resides at Kiating. The labourers are paid 1800 cash per
month. We were told that the previous year’s output was
10,000 catties, but little reliance can be placed on this state-
ment. Suchan output is very small, but the primitive methods
employed are slow and expensive. For smelting and other pur-
poses the mountain has been denuded of its timber, and is now
in its upper parts a grassy, scrub-clad wilderness. I made the
altitude of the mines 9400 feet, that is to say, 2000 feet above
the charcoal pits whence the fuel necessary to melt down the
lead is drawn. The sides of the workings are bare and gravelly,
and were covered with rich yellow flowers of a Sedum-like
plant, which was new and is unknown to me.

On leaving the lead mines and crossing a slight dip we
reached a babbling brook which forms the roadway for the
next few li. On deserting this we made a very steep ascent to
the top of a grassy ridge, alt. 10,400 feet, only to find that a
deep ravine separated us from the watershed proper. After a
most precipitous descent of 1600 feet over a rocky and difficult
path, we reached the bed of a torrent, which I take to be the
stream we noted at Peh-sha Ho flowing from a southerly
direction.

On reaching this stream the rain ceased, the mists cleared
away rapidly, and the sun showed itself for the first time in four
days. The surrounding country is savage, and is made up of

(TAVAVIAd SUIAV) Mia WAATIS GNV WIONAL OOUWVA

THROUGH THE LAOLIN (WILDERNESS) = 243

a magnificent series of limestone cliffs, their steepest crags
clothed with weather-worn trees of Silver Fir. Everywhere
else the trees have been cut down.

From the torrent we struggled up a severe ascent of 1000
feet, and reached the summit of the watershed, alt. 10,100 feet.
Here we got a very fine view of the country, which is simply
a succession of cliffs and crags capped by rugged trees of Silver
Fir, and with a dense growth of broad-leaved trees in the more
inaccessible pockets.

The rest of the day’s journey was all downhill over a vile
pathway. We reached the tiny hamlet of Yang-tientsze,
alt. 7600 feet, at 6 p.m., having occupied eleven hours in cover-
ing 30 li. Two men who carried our food-stuffs arrived just as
darkness closed in, and reported the rest of our gear far behind.
Our lodgings were poor enough in all conscience, but most
acceptable after such a fatiguing tramp. After dinner I tried
to sleep on an oil-sheet spread over one of the native beds, but
was soon discovered by hungry, tormenting fleas, and, tired
as I was, sleep proved impossible. About one o’clock my bed
and some other gear arrived. The carriers had been forced
to wait after darkness fell until the moon was up in order to see
the path. I could not complain; they had done their best over
a most heart-breaking road. The rest of our loads turned up
soon after daybreak, and we left Yang-tientsze at 7.30 a.m.
Descending by a comparatively easy road for 30 li we reached
before noon the village of Malie, alt. 5300 feet, a very poor
place, situated on the main road between Omei Hsien and
Fulin via Wa shan.

Thus had the Laolin been crossed from north-east to
south-west, and, personally, I have no desire to repeat the
journey. The continued rains increased considerably the
difficulties of the bad roads and made what, under the most
favourable weather conditions, must always be a fatiguing
journey, an exceedingly arduous and miserable one. The
rain and dense mists robbed the trip of its greatest charm,
namely, the scenery. Except on odd occasions I saw nothing
outside a radius of 50 yards. The unpropitious weather also
prevented any investigation of the flora other than that along-
side the pathway. In so far as it came under my observation

244 A NATURALIST IN WESTERN CHINA

this region possesses very little in the way of woody plants
beyond what are common to the same altitude everywhere in
western Szechuan. For richness in species it does not compare
favourably with Mount Omei or Mount Wa. However, there are
some points of interest. The region evidently enjoys a warm,
wet climate, and the belt of broad-leaved evergreens, especially
Oak and Laurinee, extends to a greater altitude than usual.
The abundance of Chinese Fir and such interesting trees as
Davidia, Tetracentron, Cladrastis, Magnolia, Aésculus, Cercidi-
phyllum, and Chinese Butternut (Juglans cathayensis) is perhaps
the outstanding feature. Strong-growing climbers such as Hol-
beellia, Actinidia, and Clematoclethra abound, and I obtained
seeds of several species. Many kinds of Sorbus with white,
red, and purple fruits occur, and seeds of these were also secured.
Honeysuckles, Brambles,and Rhododendrons are alsoabundant.
The scarcity of Birch, Beech, deciduous Oak, and Sweet Chest-
nut, and the entire absence of Pine, Cypress, and Poplar are
marked features of the region. Throughout the higher altitudes
Silver Fir and Hemlock Spruce are the only Conifers, although
in one place I thought I detected some Spruce trees high up
on the cliffs. I saw no fine trees of either of these Conifers ;
all that now remain grow on the crags and other equally
inaccessible places, and have suffered much from the winds and
weather generally. The jungle growth of Gleichenia on the
sandstone, and the impenetrable Bamboo thickets everywhere
between 6000 and 10,000 feet altitude, are the most striking
floral characteristics of the entire region. The mining in-
dustries have been the cause of the wholesale felling of the
timber.

The entire absence of decent roads, the sparse population,
wretchedly poor accommodation, the savage cliffs, and jungle-
clad mountain-sides sufficiently entitle this region to be termed
‘‘ Laolin,”’ z.e. a ‘‘ Wilderness.”

CHAPTER XIX

WA SHAN AND ITS FLORA
Fr \ HE sister mountain to the sacred Omei is Wa shan,
situated about long. 103° 14’ E., lat. 29° 21’ N., six days’
journey (roughly 80 miles) from the city of Kiating. The
intervening country is very rough, wild, and mountainous.
The road is execrable. Baber, the first foreigner to visit and
ascend this mountain, as well as Mount Omei, gives its altitude
as 10,545 feet above the sea-level, 4560 feet above the neigh-
bouring valleys. My readings were 11,250 feet above the sea,
5150 feet above the surrounding country. Allowing for error
in the barometer, I think the mountain cannot be less than
11,000 feet. The flora—always a fair guide as to altitude—
proves it to be higher than Mount Omei (10,800 feet) ; and this
agrees with the opinion of the natives, who assert that it is the
higher of the two mountains.

As seen from the top of Mount Omei it resembles a huge
Noah’s Ark, broadside on, perched high up amongst the clouds.
Viewed from a near distance it is seen to consist of a succession
of tiers of vertical limestone cliffs, only seriously broken at one
point, with a peculiarly flat summit. From the hamlet of
Ta-t’ien-ch’ih (6100 feet), which is situated in a depression at
its base, the mountain is remarkably square looking, its four
sides being more or less perpendicular. It appears to be no
more than 2000 feet above the hamlet, and yet it is really 5000
feet higher. When it was first pointed out to me, 20 miles or
so distant, I could not believe it was Wa shan—it looked so like
a huge precipice, its massiveness belittling its height.

As already stated, the first foreigner to visit Wa shan was the
late E. Colborne Baber, who made the ascent on 5th June 1878.
The description of this mountain, given by him, is so accurate
and beautiful that I cannot do better than quote it: “ The

245

upper storey of this most imposing mountain is a series of twelve
or fourteen precipices, rising one above another, each not much
less than 200 feet high, and receding very slightly on all four
sides from the one next below it. Every individual precipice
is regularly continued all round the four sides. Or it may be
considered as a flight of thirteen steps, each 180 feet high and ~
30 feet broad. Or, again, it may be described as thirteen layers
of square, or slightly oblong, limestone slabs, each 180 feet thick
and about a mile on each side, piled with careful regularity
and exact levelling upon a base 8000 feet high. Or, perhaps, it
may be compared to a cubic crystal, stuck amid a row of
irregular gems. Or, perhaps, it is beyond compare. Some day
the tourist will go there and compose‘ fine English’; he could
not choose a better place for a bad purpose ; but if he is wiser
than his kind he will look and wonder, say very little, and
pass on.”

It was on the afternoon of 30th June 1903 that I arrived
at the scattered hamlet of Ta-t’ien-ch’ih, from whence the
ascent can be made. This tiny hamlet is situated in an oval
depression, locked in by high mountains on all sides. The
depression is about a mile long and rather less than half a mile
broad at its widest point, a small lake surrounded by a luxuriant
greensward occupies the lower end. A species of Delphinium,
with lovely blue flowers, is very abundant. The Chinese call it
“ Wu-tzu,” and say that it is poisonous to man and cattle alike.
Around the farmhouses, maize, peas, beans, buckwheat, and
Irish potato are cultivated. The people here mostly profess
Christianity, and a Roman Catholic mission-house is the only
decent building in the hamlet.

Having procured a guide, I left the inn at 5.45 a.m. on Ist
July, to ascend the mountain. Mists obscured everything as
we set out, and it felt veryraw andcold. The path is the merest
track—very sinuous, steep, and difficult. Rain commenced at
2.30 p.m., and continued during the whole of the descent.
We reached our inn at 6.30 p.m., drenched through and through.

At one time a dense forest of Silver Fir covered the mountain,
but this has long since been felled, and the majority of the trees
still lie rotting where they fell. It is a common sight to see
bushes of Rhododendrons, 20 feet or more tall, growing on the

‘La 00¢'TT ‘NVHS VA

WA SHAN AND ITS FLORA 247

rotting trunks. Some of these Firs could not have been less
than 150 feet in height and 20 feet in girth. On the summit
there are still a number of trees left, but none of great size, and
nearly all have their tops broken off, either by the wind or by
thesnow. This mountain, in common with others I have visited,
shows only too plainly the destructive nature of the Chinese.
Fifty years more, under the present regime, and there will not
be an acre of accessible forest left in all central, southern, and
Western China. The making of charcoal alone imposes a very
heavy toll on hardwood trees and shrubs. The preparing of
potash salts is a common industry on the mountains west, and
is another means of clearing away the vegetation in a ruthless
manner. It is to the charcoal-burning industry that I attribute
the marked absence of Oak, Beech, and Hornbeam.

Besides the Silver Fir (Abies Delavayi), the only other
Conifers are Tsuga yunnanensis, Juniperus formosana, and
Picea complanata. Rhododendrons constitute the conspicuous
feature of the vegetation, and their wood is, luckily, not
esteemed for making charcoal. They begin at 7500 feet, but are
most abundant at 10,000 feet and upwards. In the ascent I
collected 16 species. They vary from diminutive plants 4 to 6
inches high, to giants 30 feet or more tall. Their flowers,
also, are of all sizes and colours, including pale yellow. It was
most interesting to watch the displacement of one species by
another as we ascended. One of the commonest species is
R. yanthinum, which has flowers of various shades of purple.

The ascent of the mountain commences 100 yards or so
from the inn; cultivation ceases at 6200 feet. Above this,
for 1000 feet, is a belt, which has at some time been cleared
for cultivation, but is now densely clad with coarse weeds.
Among these occur quantities of Rodgersia pinnata, var. alba,
Spirea Aruncus, Astilbe, and Pedicularis, with a few bushes
of Deutzia longifolia, Philadelphus Wilsonii, and Poison Ivy
(Rhus orientalis) interspersed. Above this, for 500 feet,
comes a wellnigh impenetrable thicket of Bamboo scrub.
The species (Avundinaria nitida) is of remarkably dense growth,
with thin culms, averaging 6 feet in height. Next above this,
till the plateau is reached, is a belt of mixed shrubs and herbs,
conspicuous amongst which are Syringa Sargentiana, Hy-

248 A NATURALIST IN WESTERN CHINA

dvangea anomala, H. villosa, Neila affinis, Dipelta ventricosa,
Ribes longeracemosum, var. Davidii, Enkianthus deflexus,
Styrax voseus, Deutzia (2 spp.), Rubus (5 spp.), Viburnum
(4 spp.), Spireea (4 spp.), Acer spp., Malus spp., Sorbus spp.,
Meconopsts chelidonifolia, Fragania fiipendula, Lilium gigan-
teum, and the herbs of the lower belt. A few Rhododendrons
occur chiefly on the cliffs.

The plateau (8500 feet) is about half a mile across, marshy
in places, and densely clad with shrubby vegetation and Bamboo
scrub. In addition to those already noted as occurring in
the belt below, we here found Hydrangea xanthoneura, Rosa
sericea, and Avalia chinensis, also a species of Caltha and a
few Conifers. Rhododendrons become more abundant as we
advanced. Crossing this plateau we reached the north-west
angle of the upper storey, and scrambled upwards by a narrow,
rocky, tortuous path through dense thickets of mixed shrubs,
which gradually give place to Rhododendrons as the narrow
ledge at 10,000 feet is reached. Rosa sericea, which was past
flowering below, was here a mass of lovely white. Two or
three species of Lonicera and various Labiate occur within
this belt, and on shady rocks at least three species of Primula,
including P. ovalifolia.

From 10,000 feet to the summit of the mountain Rhodo-
dendron accounts for fully 99 per cent. of the ligneous vegeta-
tion. A few Conifers, Lonicera, Rosa sericea, Clematis montana,
var. Wilsonit, Pieris, and Gaultheria make up the remaining
one per cent. Of the herbs, Primula is the most noteworthy.
Five fresh species of this genus occur, and amongst them, though
uncommon, the lovely yellow-flowered P. Prattit. A blue-
flowered Corydalis, Cypripedium luteum, with large yellow
flowers; Rubus Fockeanus and another herbaceous species
are other pleasing plants. On shady rocks the curious Ber-
neuxta thibetica abounds. ‘This interesting plant was first
referred to the genus Shortia by Franchet, and was later made
the type of a new genus by Decaisne. The flowers are small
and insignificant, white or pale pink in colour. On bare
rocks I gathered the pretty white-belled Cassiope selaginotdes.

My attention and interest, however, were chiefly taken
up with the Rhododendrons. The gorgeous beauty of their

ee ee ee ee ee

;
5
“
4
;
‘
4
“i
3
:

WA SHAN AND ITS FLORA 249

flowers defies description. They were there in thousands
and hundreds of thousands. Bushes of all sizes, many fully
30 feet tall and more in diameter, all clad with a wealth of
blossoms that almost hid the foliage. Some flowers were
crimson, some bright red, some flesh-coloured, some silvery-
pink, some yellow, and others pure white. The huge rugged
stems, gnarled and twisted into every conceivable shape, are
draped with pendant Mosses and Lichens, prominent among
the latter being Usnea longissima. How the Rhododendrons
find roothold on these wild crags and cliffs isa marvel. Many
grow on the fallen trunks of the Silver Fir and some are
epiphytic. Beneath them Sphagnum moss luxuriates and
makes a pretty but treacherous carpet. On bare exposed
cliffs I gathered two diminutive species of Rhododendron,
each only a few inches tall, one with deep purple and the
other with pale yellow flowers.

Dense mists obscured our view, though about ten o’clock
the sun broke through and made a temporary rift in the clouds
of mist, disclosing a scene which made us hunger for more.
In one place we leant over a precipice and could hear the roar
of a torrent some 2000 or 3000 feet below. Near the summit
three precipices, each 40 or 50 feet in height, have to be
ascended by means of wooden ladders. Up these I carried
my dog, never thinking of the descent. On returning he got
frightened, and though we blindfolded him, he struggled
hard, and on one occasion his struggles all but upset my
balance. I was heartily thankful when safe ground was
reached. It requires all one’s nerve to mount a ladder with
no balustrade, fixed to a vertical cliff 40 feet high, and on either
side a yawning abyss lost in the clouds. It is at 10,700 feet—
a narrow ridge not 8 feet broad—that the first ladder is en-
countered. From here to within a few feet of the summit
the path is terribly steep, difficult, and dangerous. On clearing
the topmost ladder and the remains of another, we unexpectedly
reached the summit by the easiest path imaginable—for all
the world like a woodland path at home.

The summit is a slightly undulating plateau, many acres
in extent, with thickets of tall Rhododendrons festooned with
Clematis montana, var. Wilsoni, and clumps of Silver Fir,

250 A NATURALIST IN WESTERN CHINA

the remnant and offspring of giants which once clothed this
magnificent mountain alternating with glades carpeted with
Anemones and Primulas and tiny streamlets meandering
hither and thither. Baber aptly describes it as ‘‘ the most
charming natural park in the world.”

In times past several temples existed on the summit,
of which ruins only now remain. At present there is but
one temple, which contains an image of P’u-hsien Pu’ssa
seated on a plaster elephant. It is built of the timber of the
Silver Fir (Abies Delavayi) and was in excellent repair. Near
the temple a small patch of medicinal Rhubarb, a few
cabbages, and Irish potatoes are cultivated.

The partly shrubby Sambucus adnata and several herbs,
including Pedicularis, Microula, Fragaria filipendula, and F.
elatior, range from base to summit. Fragaria filipendula is a
new Strawberry worthy of note; the fruit is red, more or
less cylindrical in shape, often an inch in length, and of very
good flavour. It is widely distributed in Western China, and
at Tachienlu I have enjoyed many a dish of this fruit with
cream from yak’s milk.

Two days later I ascended a lofty spur (10,000 feet) of this
mountain and added several new plants to my collection.
Of these I may mention Peéonia Veitchiit, Rubus tricolor,
Clematis Faberii, Ribes laurifolium, Potentilla V ettchi, Pyrola
rotundifolia, Styrax Perkinsia, Aristolochia moupinensts, Acer,
Anemone, Pyrus, Sorbus, Berberis, and Primula. High up
on the cliff Leontopodium alpinum and several species of
Anaphalis abound. Amongst the Sphagnum at least three
species of Lycopodium occur. On dripping, shady rocks and
trunks of the Rhododendrons, a filmy Fern (Hymenophyliwm
ometense) is abundant.

During the four days I botanized on this mountain I added
some 220 odd species to my collection. On each of these
days the work was excessively hard, and “drenched to the
skin’ but mildly describes our condition each evening as we
reached our inn. On one occasion, through treading on some
loose debris, I was only saved from being precipitated over
a steep cliff by the presence of mind of a coolie who happened
to be near me at the moment.

WA SHAN AND ITS FLORA 251

Zodlogically, Mount Wa and the surrounding wilderness
is particularly interesting as being one of the places where
wild cattle (Budorcas tibetanus) are found. I saw their foot-
prints only; they were nearly as large as those of a cow.
Ornithologically, it is interesting as being the home of at
least five species of Pheasant, including the ‘“ Blood”’ and
“ Amherst ”’ varieties.

I have climbed and botanized on many mountains in
different parts of China, some much higher than this, but
none have I found richer in cool-temperate plants, and more
especially flowering shrubs. Altogether, with its rich flora,
peculiar fauna, its singular geological formation, and its
magnificent natural park on the summit, Wa shan has many
claims on the attention of the naturalist.

END OF VOLUME I

| @K355
4 Wilson, E

New York Botanical Garden aaa

f iT i

turalist in west

4]
stite hai
iyi

—

Hi init
Ht)

;
iy
{

Wb dhia eld!
“y CUE beet
} i}

HE it
Veit
tay

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "A naturalist in western China, with vasculum, camera, and gun; being some account of eleven years' travel, exploration, and observation in the more remote parts of the Flowery kingdom"

See other formats