全 文 ：基因流介导的种间渐渗与物种界定*
杜摇 芳, 徐摇 放
(北京林业大学计算生物学中心, 林木遗传育种国家工程实验室, 生物科学与技术学院, 北京摇 100083)
摘要: 物种界定是生物学中最基本的问题之一。 近年来随着分子生物学技术的进步如分子条形码的发展, 物
种界定也越来越引起人们的兴趣。 界定一个物种或相似的一组物种时最重要的一个原则就是选择适合的分子
标记。 然而, 植物中广泛存在的不完全谱系筛选与种间渐渗却常常会阻碍准确鉴定物种。 最近, 有关基因流
介导的种间渗入和物种界定在理论和实验研究中都取得了重大进展。 本文综述了基因流介导的物种形成; 评
价了种间渐渗与不完全谱系筛选的区别; 最后总结出应该利用基因流速度较快的分子标记去鉴定物种。
关键词: 物种界定; 分子标记; DNA条形码; 种间渐渗; 不完全谱系筛选; 基因流
中图分类号: Q 949, Q 943摇 摇 摇 摇 摇 文献标识码: A摇 摇 摇 摇 摇 文章编号: 2095-0845(2012)03-257-06
Gene Flow Dependent Introgression and Species Delimitation
DU Fang, XU Fang
(Center for Computational Biology, National Engineering Laboratory for Tree Breeding, College of Bioscience
and Technology, Beijing Forestry University, Beijing 100083, China)
Abstract: Species delimitation is one of the most fundamental issues in biology and has recently drawn significant
interest. A main reason for the increasing interests was the barcoding initiative associated rapid development of mo鄄
lecular techniques. One of the most important principles to diagnose species or species groups is to choose appropriate
markers. However, incomplete linkage sorting and introgression, which are widespread phenomena in plants, present
major obstacles in species delimitation. Recently, significant progress in our understanding of gene flow dependent in鄄
trogression and species delimitation has been made both theoretically and empirically. In this paper, we reviewed the
gene flow mediated speciation; evaluated the difference of introgression and incomplete linkage sorting; and finally
concluded that species delimitation should be more effective with markers experiencing high levels of gene flow.
Key words: Species delimitation; Molecular Marker; DNA barcoding; Species introgression; Incomplete linkage
sorting; Gene flow
Gene flow mediated speciation
Speciation refers to the evolution in reproduc鄄
tive barriers (as well as phenotypic, behavioral and
genetic differences) between populations, eventually
leading to distinct species ( Coyne and Orr 1997;
Rieseberg et al., 2006 and references therein ).
However, the mechanisms by which species are
formed remain incompletely understood and the topic
of intense research and debates. One major reason
for the continued debates relates to different opinion
on species concepts (Wiens, 2004).
Species concepts have played a major role in
植 物 分 类 与 资 源 学 报摇 2012, 34 (3): 257 ~ 262
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. 11187
* Foundation item: Young Scientist Fund (2010BLX01) and the Fundamental Research Funds (YX2011鄄23) in BJFU Research Fund for the
Doctoral Program of Higher Education of China (2011DD14120014) and an open funding (K1004) from Beijing Normal
University to FKD
Received date: 2011-12-16, Accepted date: 2012-02-02
作者简介: 杜摇 芳 (1981-) 女, 讲师, 主要从事植物群体遗传学及数量遗传学研究, 特别关注森林物种间的基因流和种间渗
透等问题。 E鄄mail: dufang@ bjfu. edu. cn
evolutionary biology during the past 250 years, as
summarized by De Queiroz (2007) in his review pa鄄
per “ Ernst Mayr and the modern concept of spe鄄
cies冶. The term “species冶 originally came from the
Latin word for “ kind冶 and its use has been made
more precise following the work of Carolus Linnaeus
(1753 and 1758). In this view, species is “ the
basic unit of biological classification冶 (Flexner and
Hauck, 1993). A major shift took place after the
seminal publication of Charles Darwin爷s On the Ori鄄
gin of Species ( 1859 ). The modern evolutionary
synthesis provided the foundation for systematics and
evolutionary biology ( Dobzhansky, 1937; Mayr,
1942). Ernst Mayr, the coiner of the biological spe鄄
cies concept ( BSC ), proposed that species are
‘groups of actually or potentially interbreeding natu鄄
ral populations, which are reproductively isolated
from other such groups爷. This definition presents two
characteristics of species: (i) that they have a con鄄
tinuous gene pool (i. e. all individuals can, or have
the potential to, interbreed) and (ii), that the indi鄄
viduals of a species are reproductively isolated from
individuals of other species ( Mayr 1963; Niklas
1997 ). This definition suggests that interspecific
gene flow should be low. Although the BSC is widely
accepted by zoologists, lots of botanists prefer using
morphological characters as main features to diag鄄
nose species and criticized the use of reproductive
barrier for species delineation because such a barrier
is not as strict in plants as in animals ( e. g. many
plant taxa are potentially interfertile or parthenogeni鄄
si is common in plants) . In addition, Mayr pointed
out in further statements later that ‘ the steady and
high genetic input caused by gene flow is the main fac鄄
tor responsible for genetic cohesion among the popula鄄
tions of a species爷 (Mayr, 1963). This later argu鄄
ment emphasizes the importance of intraspecific gene
flow, which unites all individuals and populations of
one species together. While most researchers agreed
that interspecific gene flow ought to be limited to
keep species distinct, arguments were made against
the importance of intraspecific gene flow, as species
cohesion did not seem to be always dependent on in鄄
traspecific gene flow (Ehrlich and Raven, 1969).
Recently Morjan and Rieseberg (2004) demon鄄
strated that intraspecific gene flow, even when limit鄄
ed, is essential to keep species genetically coherent,
while acknowledging that other evolutionary forces
such as selection are also important in maintaining
species cohesion. Both components, inter鄄 and in鄄
traspecific gene flow, therefore lie at the root of the
biological species concept (Mayden, 1997). How鄄
ever, their interaction and the way they affect spe鄄
cies delimitation have drawn attention only very re鄄
cently (Petit and Excoiffer 2009; Du et al., 2011)
Species delimitation has involved many methods
to identify the actual boundaries of species and at the
same time determine the number of species ( De
Queiroz, 2007). Delimiting species has traditionally
relied on morphological characters supplemented
with geographic and ecological information ( Briggs
and Walters, 1997). Interest for species delimita鄄
tion has fluctuated through time: it drew a lot of at鄄
tention in the middle of the last century, thanks to
the emergence of modern systematics ( Sites and
Marshall, 2004), and now experiences a phase of
renaissance thanks to the rapid development of mo鄄
lecular technologies (Wiens, 2004). In particular,
DNA barcoding ( the use of a short standardized
DNA sequence to identify and discover species) has
attracted much attention ( Hebert et al., 2003,
2004; Hollingsworth et al., 2011; Li et al., 2011).
A key point for DNA barcoding and more generally
for species delimitation is to choose a “good冶 marker
showing species鄄specific variation.
Incomplete lineage sorting and introgression
However, numerous studies have revealed shared
DNA polymorphisms between closely related spe鄄
cies. This situation can be caused by two main rea鄄
sons: ( 1 ) retention of ancestral polymorphisms,
caused by incomplete lineage sorting ( also called
sharing of ancestral polymorphisms) during and fol鄄
lowing speciation (Heckman et al., 2007; Willyard
852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
et al., 2009 and references therein); (2) introgres鄄
sion, caused by genetic exchange after secondary
contact between two previously geographically sepa鄄
rated species ( Liston et al., 1999; Gay et al.,
2007). Distinguishing these two mechanisms is dif鄄
ficult; the most common approach is to use coales鄄
cent modeling to compare divergence time and an鄄
cestral population sizes of the two species. Several
studies have used this approach (Hey, 2001; Ran鄄
nala and Yang, 2003; Smith and Farrell, 2005;
Burgess and Yang, 2008; Joly et al., 2009). How鄄
ever, for closely related species that have diverged
very recently, this type of approach seems to have
limited utility, unless spatial information is taken in鄄
to account (McGuire et al., 2007). Incomplete lin鄄
eage sorting and introgression can be differentiated
by studying the geographic variation and demograph鄄
ic history of the species using molecular markers. If
shared polymorphisms are randomly distributed,
then retention of ancestral polymorphisms might be
involved. However, if shared haplotypes occur only
in sympatric populations, then introgression is more
likely (Fig. 1).
Fig. 1摇 Possible interpretations of the mechanisms underlying cases of
shared haplotypes between species: If shared haplotypes are randomly
distributed, incomplete lineage sorting is more likely ( top left) .
If introgression is involved, following contact between species,
shared haplotypes will be restricted to the areas of sympatry
(bottom left) . In the illustration, only the distribution of
haplotypes of species 2 is illustrated
摇 摇 Introgression is a widespread phenomenon with
potentially profound evolutionary consequences (An鄄
derson and Hubricht, 1938; Anderson, 1953; Ander鄄
son and Stebbins, 1954; Rieseberg and Brunsfeld,
1992). Edgar Anderson defined introgression as the
process of infiltration of genes from one species to
another through regular mating events involving
backcrosses with one of the parental species. The in鄄
trogression process can be divided into three steps:
generation of F1 hybrids; backcrossing with one or
both of the parents; incorporation of this new genetic
variation into the genome of the backcrossing spe鄄
cies, possibly following screening by natural selec鄄
tion ( Anderson and Hubricht, 1938; Anderson,
1953; Anderson and Stebbins, 1954). Indeed, in鄄
dividuals with introgressed genetic materials can se鄄
lectively retain ( or “ filter冶) advantageous genes,
while disadvantageous genes can be eliminated by
purifying selection (Key, 1968; Harrison, 1986).
The possibilities offered by introgression were real鄄
ized early on by breeders willing to incorporate in a
domesticated species a given attribute of a wild rela鄄
tive (Bessey, 1906; Gur and Zamir, 2004). Intro鄄
gression processes may differ for different genomes.
In the nuclear genome, the F1 hybrid gets 50% of
the genes from each parent, and the proportion of
additional introgressed genetic material is halved af鄄
ter every generation of backcrossing. For uniparen鄄
tally inherited genomes, the situation is strikingly
different. Each F1 hybrid receives a complete unal鄄
tered version of the genome from one of its parent,
so that there is no dilution of the contribution of the
donor species or population after several generations
of backcrossing (Fig. 2). The formation of introgres鄄
sion can be very quick ( several generations), in
contrast, incomplete lineage sorting typically repre鄄
sents much more ancient events.
Note that genetic drift is greatly reduced in a
subdivided population compared to a single random
mating population of similar census size (Wright,
1943; Gilpin, 1991). Hence, under the condition of
similar population size reduced intraspecific gene flow,
9523 期摇 摇 摇 摇 摇 摇 摇 摇 DU and XU: Gene Flow Dependent Introgression and Species Delimitation摇 摇 摇 摇 摇 摇 摇 摇 摇 摇
which implies increased genetic drift and hence in鄄
creased subdivision, means that sorting of ancestral
variation will take longer, making species diagnosis
more challenging. Thus, species delimitation should
be easier if it were based on molecular markers expe鄄
riencing high rates of gene flow (Fig. 3).
Fig. 2摇 Illustration of the introgression process for three different genomes: Biparentally inherited nuclear genome (left); ma鄄
ternally inherited organelle genome (middle) and paternally inherited organelle genome ( right) . The red circles represent
genes with maternal ancestry and the blue circles represent genes with paternal ancestry in the offspring. For the nuclear genome
(left) the genetic material inherited from the donor parent is reduced to 1 / 32 after four generations of backcrossing. For the
maternally inherited genomes, the genetic material from the father does not contribute at all (middle) . In contrast, for a pater鄄
nally inherited genome, the offspring retains the entire genome from the father, there is no dilution effect (right)
Fig. 3摇 Species delimitation ability at two markers experiencing contrasted rates of gene flow under the condition of similar population size.
The marker with lower levels of gene flow (marker 1) has lower taxonomic resolution than the marker with higher gene flow (marker 2)
Recent progress on gene flow dependant
introgression and species delimitation
To date, many empirical studies have already
demonstrated that incomplete linkage sorting is wide鄄
spread in plants ( Du et al., 2009; Zhou et al.,
2010; Wang et al., 2011a; Palma鄄Silva et al.,
2011), however, studies on introgression in plant is
lacking ( but see Arnold et al., 2010). Recently,
significant progress in our understanding of introgres鄄
sion has been made with the development of a neu鄄
tral demo鄄genetic model (Currat et al., 2008). This
model predicts that, when one species invades an ar鄄
ea already occupied by a related species, introgres鄄
sion of neutral genes takes place mainly from the
native species towards the invading species. In addi鄄
tion, following contacts between two hybridizing
species, the model predicts that introgression should
be particularly frequent for genome components ex鄄
periencing little gene flow. In line with this neutral
model, Petit and Excoffier (2009) suggested that
062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
markers experiencing high rates of gene flow should
be better suited for species delimitation than those
experiencing low rates of gene flow, in part because
high rates of intra鄄 specific gene flow can prevent in鄄
trogression.
Empirical studies also support the above predic鄄
tions in both gymnosperm and angiosperm plants.
Du et al. (2011) use molecular markers from two or鄄
ganelle genomes (mtDNA and cpDNA) with con鄄
trasting rates of gene flow to examine genetic exchan鄄
ges between two morphologically distinct spruce Pi鄄
cea species growing in the Qinghai鄄Tibetan Plateau.
They found that all sympatric populations of the ex鄄
panding species had received their maternally inheri鄄
ted mitochondrial DNA (mtDNA) ( transferred by
seed, low gene flow) from the resident species,
whereas for paternally inherited chloroplast ( cpD鄄
NA) ( transferred by pollen and seeds, high gene
flow) introgression is more limited and not strictly u鄄
nidirectional (See their schematic model in Fig. 1 of
Du et al., 2011). In angiosperm plants, however,
after comparative analysis of a large dataset on both
chloroplast DNA ( rbcL, matK and trnH鄄psbA) and
nuclear internal transcribed spacer (ITS), The China
Plant BOL Group et al. (2011) discovered that the
later performed relatively well in angiosperm plant
species delimitation. This conclusion based on a large
dataset represents another step forward towards rou鄄
tine use of DNA barcoding ( Hollingsworth et al.,
2011) as well as using markers with fast rate of gene
flow to diagnose species (Wang et al., 2011b).
Conclusion
Recent progress on both theoretical and empiri鄄
cal studies suggested that the important role of gene
flow should not be ignored no matter the study focus
is on the species demographic history or diagnostics.
If the studies were focusing on revealing the phylo鄄
geographic history of the species then the markers
with low rate of gene flow should be used, i. e.
mtDNA in gymnosperm or organic (cp) DNA in an鄄
giosperm. However, if the studies were designed to
delimitate related species or species groups (barcod鄄
ing for example), then the markers with fast rate of
gene flow should be chosen i. e. cpDNA in gymno鄄
sperm or nuclear DNA in angiosperms.
Acknowledgements: This review was part of FK Du爷s doctor鄄
al thesis. The authors thank Prof. XueJue Ge, South China
Botanical Garden; Prof. Jianquan Liu in Lanzhou University;
Prof Zhixiang Zhang from Beijing Forestry University; Martin
Lascoux from Uppsala University, Sweden and R佴my J. Petit
from INRA, France who give comments for this manuscripts.
References:
Anderson E, Hubricht L, 1938. Hybridization in tradescantia. III.
The evidence for introgressive hybridization [J] . American Jour鄄
nal of Botany, 25 (6): 396—402
Anderson E, 1953. Introgressive hybridization [ J] . Biological Re鄄
views, 28 (3): 280—307
Anderson E, Stebbins GLJ, 1954. Hybridization as an evolutionary
stimulus [J] . Evolution, 8 (4): 378—388
Arnold ML, Tang SX, Knapp SJ et al., 2010. Asymmetric introgres鄄
sive hybridization among louisiana iris species [ J] . Genes, 1:
9—22
Bessey CE, 1906. Crop improvement by utilizing wild species [ J] .
Journal of Heredity, os鄄2 (1): 112—118
Briggs D, Walters SM, 1997. Plant Variation and Evolution [M].
Cambridge: Cambridge University Press
Burgess R, Yang ZH, 2008. Estimation of hominoid ancestral popula鄄
tion sizes under bayesian coalescent models incorporating muta鄄
tion rate variation and sequencing errors [J] . Molecular Biology
Evolution, 25 (9): 1979—1994
China Plant BOL Group, Li DZ, Gao LM et al., 2011. Comparative
analysis of a large dataset indicates that internal transcribed spac鄄
er ( ITS) should be incorporated into the core barcode for seed
plants [J] . Proceedings of the National Academy of Science, 108
(49): 19641—19646
Coyne JA, Orr HA, 1997. “Patterns of speciation in Drosophila冶 re鄄
visited [J] . Evolution, 51 (1): 295—303
Currat M, Ruedi M, Petit RJ et al., 2008. The hidden side of inva鄄
sions: massive introgression by local genes [ J] . Evolution, 62
(8): 1908—1920
De Queiroz K, 2007. Species concepts and species delimitation [J] .
Systematic Biology, 56 (6): 879—886
Dobzhansky T, 1937. Genetics and the Origin of Species [M]. New
York: Columbia University Press, 364
Du FK, Petit RJ, Liu JQ, 2009. More introgression with less gene
flow: chloroplast vs. mitochondrial DNA in the Picea asperata
complex in China, and comparison with other Conifers [J] . Mo鄄
1623 期摇 摇 摇 摇 摇 摇 摇 摇 DU and XU: Gene Flow Dependent Introgression and Species Delimitation摇 摇 摇 摇 摇 摇 摇 摇 摇 摇
lecular Ecology, 18 (7): 1396—1407
Du FK, Peng XL, Liu JQ et al., 2011. Direction and extent of organ鄄
elle DNA introgression between two spruce species in the Qing鄄
hai鄄Tibetan Plateau [ J] . New Phytologist, 192 (4): 1024—
1033
Ehrlich PR, Raven P, 1969. Differentiation of populations [J] . Sci鄄
ence, 165 (3899): 1228—1232
Flexner SB, Hauck LC, 1993. Random House Webster忆s Unabridged
Dictionary [M]. German: Random House
Gay L, Neubauer G, Zagalska鄄Neubauer M et al., 2007. Molecular
and morphological patterns of introgression between two large
white鄄headed gull species in a zone of recent secondary contact
[J] . Molecular Ecology, 16 (15): 3215—3227
Gilpin M, 1991. The genetic effective size of a metapopulation [J] .
Biological Journal of the Linnean Society, 42 (1-2): 165—175
Gur A, Zamir D, 2004. Unused natural variation can lift yield barri鄄
ers in plant breeding [J] . Plos Biology, 2 (10): e245
Harrison RG, 1986. Pattern and process in a narrow hybrid zone
[J] . Heredity, 56 (3): 337—349
Hebert PDN, Ratnasingham S, deWaard JR, 2003. Barcoding animal
life: cytochrome c oxidase subunit 1 divergences among closely
related species [J] . Proceedings of the Royal Society B: Biolog鄄
ical Sciences, 270 (Suppl 1): S96—S99
Hebert PDN, Stoeckle MY, Zemlak TS et al., 2004. Identification of
birds through DNA barcodes [J] . Plos Biology, 2 (10): e312
Heckman KL, Mariani CL, Rasoloarison R et al., 2007. Multiple nu鄄
clear loci reveal patterns of incomplete lineage sorting and com鄄
plex species history within western mouse lemurs (Microcebus)
[J] . Molecular Phylogenetics and Evolution, 43 (2): 353—367
Hey J, 2001. The mind of the species problem [J] . Trends in Ecolo鄄
gy & Evolution, 16 (7): 326—329
Hollingsworth PM, Graham SW, Little DP, 2011. Choosing and u鄄
sing a plant DNA barcode [J] . Plos One, 6 (5): e19254. doi:
10. 1371 / journal. pone. 0019254
Joly S, McLenachan PA, Lockhart PJ, 2009. A statistical approach
for distinguishing hybridization and incomplete lineage sorting
[J] . American Naturalist, 174 (2): E54—E70
Key KHL, 1968. The concept of stasipatric speciation [J] . System鄄
atic Zoology, 17 (1): 14—22
Liston A, Robinson WA, Pi觡ero D et al., 1999. Phylogenetics of Pi鄄
nus ( Pinaceae) based on nuclear ribosomal DNA internal tran鄄
scribed spacer region sequences [J] . Molecular Phylogenetics and
Evolution, 11 (1): 95—109
Mayden RL, 1997. “A hierarchy of species concepts: the denoue鄄
ment in the saga of the species problem冶 [J] . Species the Units
of Biodiversity, 54: 381—424
Mayr E, 1942. Systematics and the Origin of Species [ M]. New
York: Columbia University Press
Mayr E, 1963. Animal Species and Evolution [M]. Cambridge MA:
Harvard University Press
McGuire JA, Linkem CW, Koo MS et al., 2007. Mitochondrial intro鄄
gression and incomplete lineage sorting through space and time:
phylogenetics of crotaphytid lizards [J] . Evolution, 61 (12):
2879—2897
Morjan CL, Rieseberg LH, 2004. How species evolve collectively:
implications of gene flow and selection for the spread of advanta鄄
geous alleles [J] . Molecular Ecology, 13 (6): 1341—1356
Niklas KJ, 1997. The Evolutionary Biology of Plants [M]. Chicago:
The University of Chicago Press
Palma鄄silva C, Wendt T, Pinheiro F et al., 2011. Sympatric bromeli鄄
ad species ( Pitcairnia spp. ) facilitate tests of mechanisms in鄄
volved in species cohesion and reproductive isolation in Neotropi鄄
cal inselbergs [J] . Molecular Ecology, 20 (15): 3185—3201
Petit RJ, Excoffier L, 2009. Gene flow and species delimitation
[J] . Trends in Ecology & Evolution, 24 (7): 386—393
Rannala B, Yang ZH, 2003. Bayes estimation of species divergence
times and ancestral population sizes using DNA sequences from
multiple loci [J] . Genetics, 164 (4): 1645—1656
Rieseberg LH, Brunsfeld SJ, 1992. Molecular evidence and plant in鄄
trogression [A]. In: Soltis DE, Soltis PS, Doyle JJ (eds) . Mo鄄
lecular Systematics of Plants [ M]. New York: Chapman and
Hall Inc., 151—176
Rieseberg LH, Wood TE, Baack EJ, 2006. The nature of plant spe鄄
cies [J] . Nature, 440 (7083): 524—527
Sites JW, Marshall JC, 2004. Operational criteria for delimiting spe鄄
cies [J] . Annual Review of Ecology, Evolution, and Systematics,
35 (1): 199—227
Smith CI, Farrell BD, 2005. Phylogeography of the longhorn cactus
beetle Moneilema appressum Lecont (Coleoptera: Cerambycidae)
was the differentiation of the Madrean sky islands driven by Pleis鄄
tocene climate changes? [ J] . Molecular Ecology, 14 ( 10 ):
3049—3065
Wang J, Wu YX, Ren GP et al., 2011a. Genetic Differentiation and
Delimitation between Ecologically Diverged Populus euphratica
and P. pruinosa [ J] . PLoS One, 6 (10): e26530. doi:10.
1371 / journal. pone. 0026530
Wang Q, Yu QS, Liu JQ, 2011b. Are nuclear loci ideal for barcoding
plants? A case study of genetic delimitation of two sister species
using multiple loci and multiple intraspecific individuals [ J] .
Journal of Systematics and Evolution, 49 (3): 182—188
Wiens JJ, 2004. What is speciation and how should we study it?
[J] . The American Naturalist, 163 (6): 914—923
Wilyard A, Cronn R, Liston A, 2009. Reticulate evolution and in鄄
complete lineage sorting among the Ponderosa pines [J] . Molec鄄
ular Phylogenetics and Evolution, 52: 498—511
Wright S, 1943. Isolation by distance [J]. Genetics, 28 (2): 114—138
Zhou YF, Abbott RJ, Jiang ZY et al., 2010. Gene flow and species
delimitation: a case study of two pine species with overlapping
distributions in southeast China [ J ] . Evolution, 64 ( 8 ):
2342—2352
262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