全 文 ：西藏地衣物种多样性的海拔梯度分析*
黄满荣1, 郭摇 威2,3
(1 北京自然博物馆, 北京摇 100050; 2 中国科学院微生物研究所真菌地衣系统学重点实验室, 北京摇 100101;
3 韩国国立顺天大学 韩国地衣研究中心, 顺天摇 540鄄742)
摘要: 根据西藏地衣分类和区系的文献资料, 对西藏地衣物种多样性的海拔梯度进行了分析。 267 个分类
群按照生长型或者基物被分为六组。 每一组地衣的物种丰度随海拔增加呈现单峰曲线形式的变化。 多数组
物种丰度的峰值出现在海拔 3 400 ~ 3 900 m之间, 对应于山地寒温带针叶林带的上半部分, 比尼泊尔对应
类群出现极值的海拔要高。 此植物带谱内复杂的生态系统可能是物种多样性高的主要原因。 壳状地衣物种
丰度的峰值出现在 5 100 ~ 5 400 m区间则可能是因为该区内高大的树木的消失以及具有充裕的阳光。 西藏
的地衣物种多样性远低于尼泊尔, 两地共有的物种数量很少。 对西藏地衣物种多样性的调查不充分应当是
其主要原因, 因此今后中国地衣学工作者应当加强西藏地衣多样性的研究。
关键词: 物种丰度; 海拔; 西藏; 单峰模型
中图分类号: Q 948摇 摇 摇 摇 摇 摇 摇 文献标识码: A摇 摇 摇 摇 摇 摇 摇 文章编号: 2095-0845(2012)02-192-07
Altitudinal Gradients of Lichen Species Richness in Tibet, China
HUANG Man鄄Rong 1, GUO Wei 2,3
(1Beijing Museum of Natural History, Beijing 100050, China; 2 Key Laboratory of Systematic Mycology
and Lichenology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China;
3 Korean Lichen Research Institute, Sunchon National University, Sunchon 540鄄742, Korea)
Abstract: Altitudinal gradients of lichen species richness in Tibet, China were investigated based on taxonomic and
floral literatures. Two hundred and sixty鄄seven lichen taxa were classified into six groups based on their growth鄄forms
or substrates. The species richness of all groups varied according to altitudinal gradient in a unimodal way. The ma鄄
ximum species richness of most lichen groups were detected between 3 400-3 900 m, corresponding to the upper part
of montane boreal coniferous forest, and the extreme height of the occurrence of theses lichen groups in Tibet is hig鄄
her than that in Nepal. This was probably due to the complex ecosystems in this vegetation zone. The species rich鄄
ness of crustose lichens reached its peak at 5 100-5 400 m, which may be attributed to the disappearance of trees
and ample sunlight. However, comparisons showed that the species diversity in Tibet was much lower than that in
Nepal. This suggests poor understandings of lichen diversity in the area, and further investigations are need.
Key words: Species richness; Altitudinal; Tibet; Unimodal
摇 Altitudinal patterns of species diversity are of
particular importance in theoretical and applied ecol鄄
ogy as well as in conservation biology ( Sergio and
Pedrini, 2007; Pryke and Samways, 2010). The
relationships of species richness and elevation were
investigated by various authors in various areas yet
different patterns were found (Rohde, 1992; Gryt鄄
nes et al., 2006). Some researchers reported that
植 物 分 类 与 资 源 学 报摇 2012, 34 (2): 192 ~ 198
Plant Diversity and Resources摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 DOI: 10. 3724 / SP. J. 1143. 2012. 11159
* Foundation items: Flora of the Cryptogamics of China (NSFC 31093440), Beijing Academy of Science and Technology Mengya Project
(2009), Beijing Natural Science Foundation (5123044)
Received date: 2011-11-11, Accepted date: 2012-02-09
作者简介: 黄满荣 (1975-) 男, 博士, 主要从事地衣的系统和分类学研究。 E鄄mail: hmanrong@ yahoo. com. cn
species richness decrease with increasing altitude
(MacArthur, 1972; Rohde, 1992), while others
argued that species richness increase with increasing
altitude until they reach their maximum at the mid鄄
dle elevation and then they decrease with increasing
altitude ( Grytnes et al., 2006; Pinokiyo et al.,
2008). Tibet is the highest area in China, vegeta鄄
tion zonations are obvious and distinctive there, and
this study aimed to determine the altitudinal patterns
of lichen there.
Prof. Wei Jiang鄄Chun took part in the scientific
expedition in Tibet, China, organized by Chinese A鄄
cademy of Sciences about four decades ago. He and
other scientists collected a great deal of lichen speci鄄
mens during the expeditions, and the reports on tax鄄
onomy and flora based on these specimens were pub鄄
lished subsequently. These publications are cited
here as data source of vertical range of species.
1摇 Investigation areas and the vegetation
zonations
The study areas are exactly the same as in Wei
and Jiang (1986; Fig. 1 therein), covering from the
Qamdo, Zogang regions and Zayu in the East and
Gyirong in Shigatse regions in the West, to the Co鄄
na, Yadong and Nyalam regions in the South and the
Lhasa, Nagqu and Baen regions in the North, corre鄄
sponding to E85毅18. 43忆-97毅50. 20忆, N27毅28. 44忆-
31毅55. 29忆. The altitude of this area is between
1 680 m a. s. l. and 6 100 m a. s. l. Seven vegetation
zones were recognized in this area as following, and
compositions and structures of higher plants and li鄄
chen populations vary in different zones accordingly
(Wei and Chen, 1974; Zheng and Chen, 1981;
Wei and Jiang, 1986).
Subtropical evergreen broad鄄leaved forest is ele鄄
vated from 1 600 m to 2 500 m. In this zonation,
Castanopsis hystrix and Cyclobalanopsis kiukiangensis
are the constructive and dominant higher plants,
while Parmotrema tinctorum, Parmelina wallichi鄄
ana, Cetrelia pseudolivetorum and Usnea arborea are
the common lichen species there.
Montane temperate mixed forest ranges from
2 500 m to 3 100 m, and Tsuga dumosa is the domi鄄
nant species there. The most abundant lichen spe鄄
cies in this zone are Parmelina irrugans, Parmelina
aurulenta, Cetraria pallescens, Ramalina sinensis,
Cladonia aggregate, etc.
Abies delavayi and Ab. delavayi var. motuoensis
dominate at montane boreal coniferous forest (3 100
-3 900 m), and Rhododendron basilicum, Rh. war鄄
dii and Acanthopanax evodiaefolia occupy the under鄄
story. Thus, the community structure is rather com鄄
plicated. At this range, the lichen species diversity
is most abundant. Alectoria acanthodes, Bryoria
smithii, Usnea longissima, U. thomsonii, Punctelia
borreri, Parmotrema arnoldii, Parmelia grayana,
Menegazzia terebrata, Cladonia gracilis, Peltigera
scabrosa, Umbilicaria indica, etc. are the main spe鄄
cies here.
Flowering plants, such as Rh. campylogynum,
Salix oreophila, Cassiope wardii, etc. are the main
components of subalpine boreal shrubs and meadows
(3 900-5 000 m), while Juncus bracteatus, Polygo鄄
num viviparum, etc. are components of alpine mea鄄
dow belt (5 000-5 600 m), no tall trees appear here
and beyond. The common lichen species at these
two zonations are Evernia mesomorpha, Parmelia
saxatilis, Thamnolia vermicularis, etc.
The alpine lichen belt (5 400-6 100 m) is be鄄
tween alpine meadow belt and alpine nival belt
(above 6 100 m). At this zonation, mean annual
temperature is between -4 益 and -8 益, and soil is
poorly developed, higher plants are seldom seen.
Thus lichens especially the crustose ones dominate at
this zone. Lecidea auriculata, Rhizocarpon superfici鄄
ale, Lecanora polytropa, Th. vermicularis, etc. are
the most dominant species at this belt.
2摇 Data and methods
The elevation data adopted in this study were
obtained from the literatures exclusively (Wei and
Chen, 1974; Wei and Jiang, 1980, 1982, 1986). Li鄄
chen nomenclature followed Wei (1991). We grouped
3912 期摇 摇 摇 摇 摇 摇 HUANG and GUO: Altitudinal Gradients of Lichen Species Richness in Tibet, China摇 摇 摇 摇 摇 摇 摇 摇
these data according to growth forms and substrates.
Crustose, foliose and fruticose lichens were consi鄄
dered to be their own natural groups according to
growth forms, and corticolous, saxicolous and terri鄄
colous lichens were grouped according to their sub鄄
strates. As an intermediate growth form, the squa鄄
mulose lichens were grouped into crustose lichens.
Lichens growing on twigs, stumps, trunks and barks
were recognized as members of corticolous group,
while those growing on soil, grasses and mosses were
grouped into terricolous group.
Two principles were introduced to deal with
these data. Firstly, the lowest elevation for each ta鄄
xon was lowered by 10% from the original data in
literatures, while the highest elevation was height鄄
ened by 10% . After all, data from literatures were
not as accurate as from sample investigations. The
collectors may not be biased and collected whatever
they met in collection activities, but it was very im鄄
possible for them to collect the specimen of every
taxon at its lowest or highest limits of distribution. In
addition, many taxa reported in the literatures were
cited only one or two specimens for each. This situa鄄
tion made us very difficult to determine their lowest
and highest elevation, and thus, such adjustment
was necessary. Secondly, we assumed that, each
taxon was distributed consecutively between its two
extreme elevations. That was, if we found a species
at 1 000 m and 2 000 m, then we regarded that it al鄄
so occurred at every 100 m band in this altitudinal
range. The necessary of this assumption, we called
“continuity assumption冶, was self鄄evident, because
there were many taxa reported in literatures, which
were cited only two or three specimens for each in a
large altitudinal range, and it was unlikely that such
taxa were distributed at their two extreme elevations
only.
Despite the obvious necessity of these two prin鄄
ciples, we carried out partial correlation tests con鄄
trolling for the effect of elevation between two data
sets for each group to evaluate whether these two
principles were reasonable: one set was the number
of taxa in each 100 m belt derived according to these
two principles (referred as derived data hereafter),
the other was the number of taxa in each 100 m belt
that exactly recorded in literatures (referred as origi鄄
nal data hereafter) .
According to these two principles, we divided
the whole altitudinal range of the investigated areas
into 53 consecutive zones, with each zone covering
100 meters. To be convenient, all species, subspe鄄
cies, varieties, and forms were treated as different
species, and species richness were estimated by
counting the total number of all taxa occurring in
each 100 m band. Thus gamma diversity sensu Whit鄄
taker (1972) were dealt in this study. These treat鄄
ments were processed in Microsoft Office Excel
2003. The obtained data then were transferred to
SPSS 16. 0 for Windows (SPSS Incorporated, Chica鄄
go, Illinois) for further analyses, which included
plotting and correlation analyses.
Lichens in Nepal (Baniya et al., 2010) were
also integrated into this study in order to make a
comparison between the altitudinal patterns of li鄄
chens from these two areas.
3摇 Results and discussion
Totally we collected data for 267 lichen taxa
(actual number of species was 228) represented 58
genera from Tibet, China, distributed between 1 680 m
a. s. l. and 6 100 m a. s. l. The number of taxa and
altitudinal range of each group can be seen in Table
1. Crustose lichens had the least abundant species
and they also occupied the highest altitudinal ran鄄
ges, while foliose lichens had the most abundant
species and saxicolous lichens were distributed
throughout the investigated altitudinal ranges.
Results of partial correlation tests for the origi鄄
nal data and derived data can be seen in Table 2.
Except for the crustose group, the data in other
groups were strongly correlated, which also can be
referred from the similarities between the topography
of their histograms of original data and scatter plots
of derived data ( Figs. 1, 2). The low correlation
491摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 34 卷
occurring in crustose lichens probably is because too
few species were adopted in analysis. In any case,
our two principles got a solid support from statistics
and were applicable in following analyses.
Table 1摇 Number of taxa and altitudinal range in each group of lichens in Tibet, China
Groups No. of taxa Altitudinal rangea / m
Altitudinal range of the peaks of
species richness in Tibetb / m
Altitudinal range of the
peaks in Nepalc / m
Crustose 22 2 850-6 100 5 100-5 600 4 100-4 200
Foliose 155 1 680-5 667 3 400-3 800 2 400-2 500
Fruticose 89 1 800-5 800 3 400-3 900 3 200
Corticolous 118 1 680-5 800 3 400-3 800 2 500-2 700
Saxicolous 76 1 680-6 100 3 400-3 700 3 900-4 200
Terricolous 90 2 050-5 700 3 500-3 900 3 900-4 200
Totald 267 1 680-6 100 3 500-3 900 3 100-3 400
Note: aOriginal data; baccording to derived data; cdata from Baniya et al. (2010); d the sum of number of species of growth鄄form鄄based groups is
little larger than the number of total species for some species have more than one growth forms; it is also true for the groups based on substrate
Table 2摇 Partial correlation tests for original and derived data of each group*
Group Fruticose Foliose Crustose Terricolous Saxicolous Corticolous
Coefficient 0. 736 0. 760 0. 235 0. 748 0. 694 0. 788
2鄄tailed sig. <0. 01 <0. 01 0. 181 <0. 01 <0. 01 <0. 01
Note: *Controlling for the effect of altitude
Fig. 1摇 Histograms and scattered plots of species richness of
crustose, foliose and fruticose lichens along altitude.
Left: histograms of original data; right: scatter plots of
derived data. A-B: crustose lichens; C-D: foliose
lichens; E-F: fruticose lichens
Fig. 2摇 Histograms and scattered plots of species richness of
corticolous, saxicolous and terricolous lichens along altitude.
Left: histograms of original data; right: scatter plots of
derived data. A-B: corticolous lichen; C-D: saxicolous
lichens; E-F: terricolous lichens
5912 期摇 摇 摇 摇 摇 摇 HUANG and GUO: Altitudinal Gradients of Lichen Species Richness in Tibet, China摇 摇 摇 摇 摇 摇 摇 摇
摇 摇 Baniya et al. (2010) cited 524 lichen taxa to
investigate the altitudinal gradient of lichen species
richness in Nepal from literatures. These lichens
were distributed between 200-7 400 m, occupying a
much larger altitudinal range than that in this study.
Among these taxa, eighty were common in Tibet and
Nepal, the similarity coefficient was rather small
(0. 10 sensu, Sorensen, 1948). Among them, 43
were distributed in the similar altitudinal ranges or
partly overlapped at least. There were 11 taxa dis鄄
tributed at higher elevations in Nepal than in Tibet,
while 26 other taxa were lower. For the lichens spe鄄
cies in Nepal, 487 were distributed between 1 680
m and 6 100 m, an altitudinal range corresponding
to that in this study. There were 63 species shared
by Tibet and Nepal at this range, which made the
similarity coefficient further lower (0. 08). This low
similarity may be partly attributed to different com鄄
position of lichen flora due for complete difference of
ecological environments in these two areas. But the
underestimated species diversity for these two areas,
especially for Tibet, is also an important reason.
Though R. P. David had a field trip to Tibet as early
as the year of 1869, and Zahlbruckner and other li鄄
chenoloigsts reported some species from Tibet, it was
until 1980s that Prof. Wei made systematic investi鄄
gations on the flora (Wei, 1991 and citations there鄄
in). Later, few taxonomic and floral studies were
made concerning this area, and the resulted under鄄
estimation of species diversity was not surprising.
The total species richness showed a unimodal
pattern according to altitudinal gradients, with its
peak at 3 500-3 900 m, whereas the observed maxi鄄
mum richness (187 taxa) was at about 3 600 m
(Fig. 3, Table 1). The maximum species richness
of crustose lichens was at 5 100-5 600 m, that of fo鄄
liose lichens was at 3 400 -3 800 m, while that of
fruticose lichens was at 3 400-3 900 m (Fig. 1, Ta鄄
ble 1). The maximum species richness of terricolous
lichens was at 3 500-3 900 m, that of corticolous li鄄
chens was at 3 400-3 700 m, while that of saxico鄄
lous lichens was at 3 400 -3 800 m (Fig. 2, Table
1). It was obvious that, except for the crustose
group which reached its peak of species richness at
much higher elevations, the altitudinal patterns of
species richness of the other five groups were highly
accordance with the total species richness which was
supported by correlation tests (Table 3). The topog鄄
raphy of their scattered plots along the altitude also
resembled to each other (Fig. 3).
The peak of species richness of each group fell
into different altitudinal ranges in Tibet and Nepal
(Table 1). Both in Tibet and Nepal, the maximum
species richness of crustose lichens was at the high鄄
est elevations among these six groups. But it was
somewhat surprising that, except for terricolous and
Fig. 3摇 Scattered plots of total species richness along altitude. Scattered plots of the 6 groups
are also reproduced here to make convenient comparisons
691摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 34 卷
Table 3摇 Correlation tests for the altitudinal distributions of total species and six groups
Group Terricolous Saxicolous Corticolous Fruticose Foliose Crustose
Tatal taxa 0. 973 0. 972 0. 953 0. 983 0. 988 0. 016
2鄄tailed sig. <0. 01 <0. 01 <0. 01 <0. 01 <0. 01 0. 908
saxicolous groups, the species richness reached their
peaks at lower elevation in Nepal than in Tibet.
Maybe it can be explained by the more complicated
micro鄄ecological conditions in Tibet. What it may
be, this results was accordance with the low similari鄄
ty of lichen diversity between this two areas.
The peaks of species richness of the groups,
except for crustose group, fell into the upper part of
the mountain boreal coniferous forest (3 400-3 900
m). This result coincided with the geographical ele鄄
ment analyses by Wei and Jiang (1986). This was
because the factors of latitudinal zonality have been
weakened here due to the continuously elevation of
Tibet Plateau, and the distributions of organisms
were strongly effected in turn. The fruticose and ter鄄
ricolous lichens had obvious plateaus at this belt
(Fig. 3). Below this belt, the total species richness
increased rapidly with the increasing elevation. This
may be mainly attributed to the increase of species
richness of foliose lichens or corticolous lichens
(Fig. 3). Species richness of other groups increased
relatively slowly. The species richness of crustose li鄄
chens remained in a low plateau until the elevation
reaches 3 000 m and thus made less contribution to
the increase of total species richness. Above the
mountain boreal coniferous forest, the total species
richness decreased steeply with increasing elevation
until the elevation reached ca. 4 800 m where it ar鄄
rived a plateau (Fig. 3). This phenomenon was in鄄
duced by the dramatic decrease of the species rich鄄
ness of all groups except the crustose one. The spe鄄
cies richness of crustose lichens, on the contrary,
slightly increased from 3 800 m to 5 400 m and ar鄄
rived its maximum there and then decreased slowly
with increasing altitude. The species richness of sax鄄
icolous lichens also increased slightly at 4 300-5 400
m and formed a secondary peak there, but this
didn爷t influenced the drastic decline of total species
richness. This increase of species richness of saxico鄄
lous lichen in this altitudinal range may be due to
the disappearance of trees and the exposure of rocks
to sunlight.
As a whole, the species richness of all groups
had a unimodal altitudinal pattern. This pattern has
been reported by many other authors in different or鄄
ganisms (Bruun et al., 2006; Grytnes et al., 2006;
Pinokiyo et al., 2008; Huang, 2010). It is charac鄄
terized by a peak of species richness at intermediate
altitudes. However, a slightly bimodal peak was
detected for saxicolous lichens ( Fig. 2: D) be鄄
tween 4 800 - 5 600 m, the transition from alpine
meadow to lichen belt. Baniya et al. (2010) also
detected bimodal peaks for saxicolous and terricolous
lichens in Nepal. These secondary peaks are proba鄄
bly induced by complex ecological environments in a
local scale, and it doesn爷t influence the topography
in a whole.
4摇 Conclusions
Altitude is believed a key factor to determine
the patterns of biodiversity (McCune et al., 1998;
Sergio and Pedrini, 2007). Lichens are important
indicators of air quality and forest health (McCune,
2000; Jovan and McCune, 2006) and their altitudi鄄
nal patterns have been received some attentions re鄄
cently. Based on Prof. Wei爷s significant works on
lichen taxonomy and flora of Tibet, our analyses sug鄄
gested the lichen species richness were peaked in
the mountain boreal coniferous forest.
However, much less taxa were reported from
Tibet than from Nepal, and there were only a few
taxa common in these two areas. This phenomenon
may be explained by the different ecological condi鄄
tions partly, but it is also necessary to point out that
7912 期摇 摇 摇 摇 摇 摇 HUANG and GUO: Altitudinal Gradients of Lichen Species Richness in Tibet, China摇 摇 摇 摇 摇 摇 摇 摇
the flora of Tibet is seriously underestimated and
thus deserves intensive investigations in future.
Acknowledgements: Huang Man鄄Rong wants to express his
sincere thanks to his friend Nicholas Whipps for improving
the English writing of this manuscript.
Refferences:
Baniya CB, Solh覬y T, Gauslaa Y et al., 2010. The elevation gradient
of lichen species richness in Nepal [ J] . The Lichenologist, 42:
83—96
Bruun HH, Moen J, Virtanen R et al., 2006. Effects of altitude and
topography on species richness of vascular plants, bryophytes and
lichens in alpine communities [ J] . Journal of Vegetation Sci鄄
ence, 17: 37—46
Grytnes JA, Heegaard E, Ihlen PG, 2006. Species richness of vascu鄄
lar plants, bryophytes, and lichens along an altitudinal gradient in
western Norway [J] . Acta Oecologica, 29: 241—246
Huang MR, 2010. Altitudinal patterns of Stereocaulon ( Lichenized
Ascomycota) in China [J] . Acta Oecologica, 36: 173—178
Jovan S, McCune B, 2006. Using epiphytic macrolichen communities
for biomonitoring ammonia in forests of the Greater Sierra Nevada,
California [J] . Water, Air, & Soil Pollution, 170: 69—93
MacArthur RH, 1972. Geographical Ecology: Paterns in the Distribu鄄
tion of Species [M]. New York: Harper & Row, 1—269
McCune B, 2000. Lichen communities as indicators of forest health
[J] . The Bryologist, 103: 353—356
McCune B, Rogers P, Ruchty A et al., 1998. Lichen communities for
forest health monitoring in Colorado, USA [R]. A report to the
US Department of Agriculture, Forest Service, Forest Health
Monitoring National Office, Southern Research Station, Re鄄
search Triangle Park, NC, 1—30
Pinokiyo A, Singh KP, Singh JS, 2008. Diversity and distribution of
lichens in relation to altitude within a protected biodiversity hot
spot, northeast India [J] . The Lichenologist, 40: 47—62
Pryke JS, Samways MJ, 2010. Significant variables for the conserva鄄
tion of mountain invertebrates [ J] . Journal of Insect Conserva鄄
tion, 14: 247—256
Rohde K, 1992. Latitudinal gradients in species diversity: the search
for the primary cause [J] . Oikos, 65: 514—527
Sergio F, Pedrini P, 2007. Biodiversity gradients in the Alps: the o鄄
verriding importance of elevation [J] . Biodiversity and Conserva鄄
tion, 16: 2343—3254
Sorensen T, 1948. A method of establishing groups of equal ampli鄄
tude in plant sociology based on similarity of species content
[J] . Biologiske Skrifter, 5: 1—34
Wei JC, 1991. An Enumeration of Lichens in China [M]. Beijing:
International Academic Publishers, 1—278
Wei JC, Chen JB, 1974. Materials for the lichen flora of the Mount
Jolmo Lungma region in southern Tibet, China [A]. In: Report
on the Scientific Expedition of the Mt. Jolmo Lunga Region
[M]. Beijing: Science Press, 173—182 (in Chinese)
Wei JC, Jiang YM, 1980. Species novae Lichenum e Parmeliaceis in
regione Tibetensi [J] . Acta Phytotaxonomica Sinica (植物分类
学报), 18: 386—388 (in Chinese)
Wei JC, Jiang YM, 1982. New materials for lichen flora from Tibet
[J] . Acta Phytotaxonomica Sinica (植物分类学报 ), 20:
496—501 (in Chinese)
Wei JC, Jiang YM, 1986. Lichens of Tibet [M]. Beijing: Science
Press, 1—130 (in Chinese)
Whittaker RH, 1972. Evolution and measurement of species diversity
[J] . Taxon, 21: 213—251
Zheng D, Chen WL, 1981. A preliminary study on the vertical belts
of vegetation of the eastern Himalayas [J] . Acta Botanica Sinica
(植物学报), 23: 228—234
891摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 植 物 分 类 与 资 源 学 报摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 摇 第 34 卷