全 文 ：书本地种云南柳与外来种旱柳 (杨柳科)的同倍体自然杂交
*
吴 杰1，2，3，4，王东超1，2，3，杨永平1，2，3
＊＊，陈家辉1，2，3＊＊
(1 中国科学院昆明植物研究所东亚植物多样性与生物地理学重点实验室，昆明 650201;2 中国科学院
昆明植物研究所西南野生生物种质资源库，昆明 650201;3 中国科学院昆明植物研究所
青藏高原研究所昆明部，昆明 650201;4 中国科学院大学，北京 100049)
摘要:对分布于云南的旱柳 (Salix matsudana)和云南柳 (Salix cavaleriei)之间的一个自然杂交种进行了研
究。野外观察表明疑似杂交种异蕊柳 (Salix × heteromera)形态上介于旱柳和云南柳之间，并得到了基于叶
形态特征的主成份分析的印证。核基因 ITS 序列数据表明这三个种存在 ITS 序列的种内和个体内的多态
性，且疑似杂交种的 ITS序列的基因型总是疑似亲本的嵌合体。因此可以判定异蕊柳是旱柳和云南柳的自
然杂交后代。流式细胞分析表明这三个种均为四倍体，因而，本杂交事件为同倍体杂交。基于四个叶绿体
序列片段的数据表明本自然杂交事件是不对称的，云南柳是异蕊柳的母本。常见外来种旱柳与稀有本地种
云南柳的杂交可能导致稀有种云南柳的濒危甚至灭绝。研究表明柳属植物的引种应非常谨慎。
关键词:柳属;同倍体杂交;不对称杂交;分子证据
中图分类号:Q 943 文献标志码:A 文章编号:2095－0845(2015)01－001－10
Homoploid Hybridization between Native Salix cavaleriei
and Exotic Salix matsudana (Salicaceae)
WU Jie1，2，3，4，WANG Dong-chao1，2，3，YANG Yong-ping1，2，3＊＊，CHEN Jia-hui1，2，3＊＊
(1 Key Laboratory for Plant Diversity and Biogeography of East Asia，Kunming Institute of Botany，Chinese Academy of Sciences，
Kunming 650201，China;2 Germplasm Bank of Wild Species in Southwest China，Kunming Institute of Botany，
Chinese Academy of Sciences，Kunming 650201，China;3 Institute of Tibetan Plateau Ｒesearch at Kunming，
Kunming Institute of Botany，Chinese Academy of Sciences，Kunming 650201，China;
4 University of Chinese Academy of Sciences，Beijing 100049，China)
Abstract:Natural hybridization between Salix matsudana and Salix cavaleriei was investigated based on populations
from Yunnan，China． Field observations revealed that the putative hybrid，S. × heteromera had intermediate morpho-
logies between S. matsudana and S． cavaleriei． This was further confirmed by principal component analysis． Sequence
data of nuclear rDNA internal transcribed spacer region showed both intraspecific and intragenomic polymorphisms in
all the three species，and S. × heteromera showed a strong additive pattern between its suspected progenitors at all
nucleotide sites of the genotypes identified． Therefore，S. × heteromera was confirmed to be a natural hybrid between
S. cavaleriei and S. matsudana． Flow cytometry analysis indicated that all the three species are tetraploid，and the hy-
bridization was homoploid． Sequence data from four chloroplast datasets indicated that the hybridization was asym-
metric，with S. cavaleriei as the maternal parent． The hybridization between the exotic common species S. matsudana
and native rare species S. cavaleriei might increase the risk of endangerment and even extinction，indicating that the
植 物 分 类 与 资 源 学 报 2015，37 (1):1～10
Plant Diversity and Ｒesources DOI:10．7677 /ynzwyj201514052
＊
＊＊
Funding:The National Natural Science Foundation of China (NSFC 31270271 to J. H． Chen)，the Project of Knowledge Innovation Program
of the Chinese Academy of Sciences (Grant No．:KSCX2-EW-J-24) ，Yunnan Natural Science Foundation (2010CD109 to J. H．
Chen) ，and the Youth Innovation Promotion Association，CAS
Author for correspondences;E-mail:yangyp@mail. kib. ac. cn;chenjh@mail. kib. ac. cn
Ｒeceived date:2014－03－27，Accepted date:2014－06－10
作者简介:吴杰 (1988－)女，硕士研究生，主要从事植物分类学及系统学研究。
introduction of Salix species should be made very cautiously．
Key words:Salix;Homoploid hybridization;Asymmetry;Molecular evidence
Natural hybridization is prevalent and plays an
important role in the evolution of plants;the possible
outcomes of hybridization include breakdown of iso-
lating barriers;introgression;increased genetic di-
versity;and the origin of adaptations，ecotypes，and
species (Coyne and Orr，2004;Soltis and Soltis，
2009;Abbott et al．，2013)． Natural hybrids usually
show mosaic morphological characters of their paren-
tal species;therefore，the intermediate of parental
characters in morphology can be used to identify nat-
ural hybrids． However，this may not always be true，
because not all morphological characters have genet-
ic basis． Further，when there is introgression，a hy-
brid may be similar to one of its parental species and
therefore difficult to identify (Ｒieseberg et al．，1993;
Ｒieseberg and Wendel，1993)． Besides，some mor-
phological intermediates might form via convergent
evolution (Ｒieseberg and Wendel，1993;Ｒieseberg
et al．，1999)． Molecular studies have shown that in-
terspecific hybridization is more prevalent than that
indicated by morphological and cytogenetic evi-
dence，as reviewed by Ｒieseberg (1997)and Ar-
nold (1997)． Many natural hybrid species have
been confirmed by molecular investigations，and nu-
merous historical hybridizations have also been re-
vealed (e. g．，Hardig et al．，2000;Kaplan and Fe-
hrer，2007;Zha et al．，2008)．
The genus Salix L．，collectively known as wil-
lows，is a well-known taxonomically difficult plant
taxon that consists of some 460－520 species world-
wide，which are mainly distributed in the north tem-
perate areas． Willows have high economic value;
species of this genus can be used in ornamentals，
fuel，and medicines，and are good sources of energy
biomass as well (Fang et al．，1999;Skvortsov，1999;
Argus，2010)． Salix is taxonomically difficult be-
cause of common natural hybridization，simple flow-
ers that seldom present stable reproductive traits，di-
oecism，and large phenotypic variation (Ｒechinger，
1992;Skvortsov，1999;Argus，2010)． As reviewed
by Argus (2010)，there are about 120 Salix hybrids
that have been recognized in the North American flo-
ra (113 native Salix species are recorded in North
American flora) ，and about half of these are rela-
tively common． Indigenous species hybridize not only
with each other，but also with introduced willows
species． For example， the introduced Old World
species Salix alba L． is documented to form natural
hybrids with indigenous species Salix lucida Muhlen-
berg and Salix nigra Marshall in New World (Ar-
gus，2010)． Despite the prevalent natural hybridiza-
tion in Salix，most Salix hybrids were identified by
morphological evidence，which might not be reliable
as mentioned above，and seldom have been con-
firmed by molecular evidence． A morphological and
molecular study by Hardig et al． (2000)revealed
that about one-third plants originally identified as
Salix eriocephala were possible introgressants． An a-
symmetrical natural hybridization of Populus， a
closely related genus of Salix，was identified by
Hamzeh et al． (2007)．
China is rich in Salix species，with about 275，
having been recorded (Fang et al．，1999) ;some of
the species described in Flora of China might be nat-
ural hybrids． In our previous study，we found that
Salix heteromera Handel-Mazzetti，a tree willow dis-
tributed in limited areas of Yunnan province，China，
always and almost only coexists with other two tree
willows，the invasive Salix matsudana Koidzumi and
the indigenous S. cavaleriei H． Léveillé under natural
conditions (Fig. 1)． Moreover，S. × heteromera is mor-
phologically (e. g．，leaf morphology，stamen num-
ber，ovary stipe)intermediate between S. matsudana
and S. cavaleriei (Table 1，Fig. 2)． Therefore，we
suspected that S. × heteromera might be a natural hy-
brid between S. matsudana and S. cavaleriei．
In this study，we used morphological and molec-
ular methods to elucidate whether Salix × heteromera
2 植 物 分 类 与 资 源 学 报 第 37卷
Fig. 1 Distribution of Salix × heteromera，S. cavaleriei，and
S. matsudana in and around Yunnan province，China
is a natural hybrid between S. matsudana and S. cava-
leriei，the direction of the possible hybridization and
its impact on natural populations of these species．
1 Materials and methods
1. 1 Plant material
We selected five samples of the putative hybrid
Salix × heteromera and three samples each of its sus-
pected parents S. cavaleriei and S. matsudana from
Lashihai，Lijiang，Yunnan，China (26°5351″ N，
100° 7 56″ E) for sequencing of nuclear rDNA
(nrDNA)internal transcribed spacer (ITS)and cp-
DNA rbcL，matK，trnD-T，and atpB-rbcL regions．
One sample each of the above species was used for
Table 1 Morphological comparison of the putative Salix × heteromera with the suspected parents S. cavaleriei and S. matsudana
Taxa
Characters
Leaf (length × width)/ cm Number of stamens Gynophore
Flowering time
S. cavaleriei 4－11 × 2－4 6－8 long March to the end of June
S. × heteromera 5－7 × 1. 2－1. 4 2－5 short March to April
S. matsudana 1. 5－3 × 0. 6－0. 8 2 none March to April
Fig. 2 Leaf morphologyof S. cavaleriei，S. × heteromera and S. matsudana
31期 WU Jie et al．:Homoploid Hybridization between Native Salix cavaleriei and Exotic Salix matsudana …
flow cytometry analysis． In all，107 specimens from
four localities were used for morphological analysis，
i. e．，Lashihai，Heilongtan (26°5254″ N，100°14
1″ E)，Suhe (26°4729″ N，99°4858″ E) ，all in
Lijiang，and Xizhou (25°5114″ N，100°8 E)in
Dali，Yunnan province，China (see Table 2 for de-
tails)． All voucher specimens are deposited at the
Herbarium of Kunming Institute of Botany，Chinese
Academy of Sciences (KUN)．
1. 2 Morphological Analysis
Principal components analysis (PCA)of leaf
morphology was based on five leaf characters，i. e．，
maximum blade length (L) ，maximum blade width
(W) ，petiole length (PL) ，blade length from the
base to the point of maximum width (BL) ，and
blade length-width ratio (LWＲ)． At least five ma-
ture leaves of each specimen were measured，and
the average values were used in PCA．
The PCA included data standardized for each
trait and was performed using a correlation matrix
without rotation;factor axes that described less than
11% of the overall variation in leaf morphology were
excluded from the analysis． The results were ana-
lyzed using software SPSS 11. 5 (SPSS，2002)．
1. 3 Molecular analysis
1. 3. 1 DNA extraction，PCＲ amplification，and se-
quencing
Total DNAs were isolated using the cetyltrimeth-
ylammonium bromide (CTAB)method of Saghai-Ma-
roof et al． (1984)as modified by Doyle and Doyle
(1987)． The nrDNA ITS regions were amplified by
polymerase chain reaction (PCＲ) using primers
“ITS-a”and“ITS-d”(Leskinen and Alstrom-Ｒapa-
port，1999)． The direction of hybridization was de-
termine by amplifying partial sequences of the chlo-
roplast trnD-T， atpB-rbcL intergenic region， and
rbcL-matK gene by using the following primes:
“trnDGUCF”and“trnTGGU”for trnD-T (Demesure et
al．，1995)，“atpB-1”and“rbcL-1”for atpB-rbcL
(Chiang et al．，1998) ，“1F”and“1024Ｒ” for
rbcL (Lledo et al．，1998) ，and“3F_KIM f”and
“1Ｒ_KIM r”for matK (Janzen，2009)．
PCＲ was performed using a PTC-100TM pro-
grammable thermal cycler (MJ Ｒesearch，Inc．)in a
total volume of 25 μL containing 15 μL Power Taq
PCＲ MasterMix (BioTeke Corporation)，8. 5 μL
ddH2O，1 μL of each primer，and 1. 5 μL DNA
template． The PCＲ conditions included an initial de-
naturation for 3 min at 94 ℃，followed by 35 cycles
of 30 s at 94 ℃ for template denaturation，30 s at
50 ℃ for primer annealing，1 min at 72 ℃ for exten-
sion，and a final extension period of 10 min at 72 ℃ ．
The PCＲ products were purified using the PCＲ
Products Purification Kit (Biotype Corporation)，
Table 2 Samples used for sequencing and GenBank accession numbers
Taxon Voucher*
GenBank accession numbers
ITS rbcL matK atpB-rbcL trnD-T
Salix cavaleriei
C518
C519
C1038＊＊
KF209139-146
KF209147-155
KF209128-138
KF209231
KF209232
KF209230
KF209254
KF209255
KF209253
KF209243
KF209244
KF209242
KF209265
KF209266
KF209264
S. × heteromera
C1030
C1047
C1048＊＊
C1056
C1058
KF209156-165
KF209166-174
KF209175-184
KF209185-193
KF209194-203
KF209233
KF209235
KF209236
KF209237
KF209238
KF209256
KF209257
KF209258
KF209259
KF209260
KF209245
KF209246
KF209247
KF209248
KF209249
KF209267
KF209268
KF209269
KF209270
KF209271
S. matsudana
C523
C1034
C1042＊＊
C1099
KF209222-229
KF209204-212
—
KF209213-221
KF209241
KF209239
KF209240
—
KF209263
KF209261
KF209262
—
KF209252
KF209250
KF209251
—
KF209274
KF209272
KF209273
—
* All specimens collected by Jiahui Chen in Lashihai，Lijiang，Yunnan，China，and deposited in Herbarium of Kunming Institute of Botany，
Chinese Academy of Sciences (KUN);＊＊ Samples used for flow cytometry
4 植 物 分 类 与 资 源 学 报 第 37卷
following the manufacturer’s instructions． The puri-
fied PCＲ products of nrDNA ITS were ligated to the
pUC18 plasmid vector，and the recombinant plas-
mids were cloned into competent Escherichia coli
DH5α cells (Biotype Corporation)． The bacteria
that contained recombinant plasmid were sequenced
directly，and at least 8 different clones of each sam-
ple were used． Sequences were assembled by Ge-
neious (Drummond et al．，2011)and aligned with
Muscle (Edgar，2004)，followed by manual correc-
tion using Geneious 5. 4 (Drummond et al．，2011)．
Average sequence divergence (i. e．，pairwise dis-
tance)was estimated using Kimura’s (1980)two-
parameter method in Mega 5. 2 (Tamura et al．，
2011)．
1. 3. 2 Determination of ploidy using flow cytometry
DNA ploidy level of the putative hybrid Salix ×
heteromera and its suspected progenitors S. cavaleriei
and S. matsudana was determined by comparing
their DNA contents with those of taxa of known
ploidy level;we used S. gracilistyla Miquel as a ref-
erence sample，which has been reported to be dip-
loid with 2n = 2x = 38 (Ｒudyka，1990)． To avoid
the risk of error due to instrument drift，we simulta-
neously chopped the test samples and reference sam-
ple． Leaves (50 mg silica gel dryed leaf)of refer-
ence and test plants were placed in a plastic petri
dish containing 1 000 μL WPB (0. 2 mol·L－1 Tris
HCl，4 mmol·L－1 MgCl2·6H2O，2 mmol·L
－1 ED-
TA Na2·2H2O，86 mmol·L
－1 NaCl，10 mmol·L－1
sodium metabisulfite，1% PVP-10. 1% Triton X-
100，pH 7. 5)for 30 min． Next，they were chopped
using a razor blade，passed each sample through a
30-μm filter，and added 150 μL of staining solution
(500 μg·mL－1 ＲNase A，1. 12 mg·mL－1 PI)for 15
min in dark． Each sample was run for 2－3 min on
Partec CyFlow Space flow cytometer． The peak of the
reference sample was adjusted to be located approxi-
mately at channel 100，so that the relative ploidy of
the unknown samples could be determined by com-
paring the peak positions of reference sample and the
test sample by using the following ratio (Doleel et
al．，2007):
Sample ploidy = Ｒeference ploidy ×
mean position of the sample peak
mean position of the reference peak
2 Ｒesutls
2. 1 Morphological analysis
The result of PCA of leaf morphology (Fig. 3)
indicated that the putative hybrid Salix × heteromera
is an intermediate between and separate from its sus-
pected parents S. cavaleriei and S. matsudana along
the first factor，which accounted for 84% of the vari-
ance observed．
2. 2 Genotypes of ITS
The complete ITS regions of Salix × heteromera，
S. cavaleriei，and S. matsudana were sequenced;the
length varied from 593 to 599 base pairs (bp)，and
the aligned length was 603 bp (Fig. 4)． Both intra-
specific and intra-individual polymorphism were de-
tected in all the three species sequenced except for a
specimen of S. cavaleriei (c518) ，which had only
one ITS repeat type． In all，11 ITS DNA variations
(6 in ITS1 and 5 in ITS2 region;2 are indels and
the other 9 are point mutations)were recognized in
the three species． The putative hybrid S. × hetero-
mera showed nucleotide additivity of its suspected
parents S. cavaleriei and S. matsudana at all variation
sites of the 11 genotypes． Further，it showed the
most average sequence divergence that equaled the
sum average sequence divergence of its suspected
parents (see Table 3 and Fig. 4 for details)．
2. 3 Chloroplast haplotypes
In all，11 haplotypes (4，1，2，4 for atpB-
rbcL，matK，rbcL，trnD-T，respectively)were de-
tected in the chloroplast regions sequenced;10 of
the haplotypes of the putative hybrid Salix × hetero-
mera were identical to those of S. cavaleriei，and one
haplotype (a deletion in the atpB-rbcL region)was
exclusive to S. × heteromera (see Table 4 for de-
tails)． Therefore，the hybrid S. × heteromera had
S. cavaleriei as the plastid donor parent，and the hy-
bridization was unidirectional，i. e．，asymmetrical．
51期 WU Jie et al．:Homoploid Hybridization between Native Salix cavaleriei and Exotic Salix matsudana …
2. 4 Ploidy level
Flow cytometry analysis of intact leaf nuclei in-
dicated that all the three species were tetraploid
(Fig. 5)． The diploid standard sample (Salix graci-
listyla)nuclei produced a single peak that appeared
at channel 100，with average coefficient of variation
(CV)of 9. 24%，and the peak mean channel of the
three test samples were around 200 (S. cavaleriei:
X = 198. 22，CV = 4. 50%;S. × heteromera:X =
208. 22，CV= 4. 19%;S. matsudana:X = 209. 57，
CV= 5. 22%)． Thus，the three species were conclu-
ded to be tetraploid with the chromosome number of
2n= 4x= 76． Our result is consistent with the repor-
ted ploidy of S. matsudana (Suda，1958，1963)．
Fig. 3 Plot of leaf characters according to the first and second factor
scores derived from PCA (factor 1 described 84% and factor 2
described an additional 11% of the overall variation)
Fig. 4 Schematic diagram of sequence alignments of the ITS region in Salix × heteromera，S. cavaleriei，and S. matsudana． Sequence names
are presented as“species name_voucher_clone number”(cav=S. cavaleriei，het =Salix × heteromera，mat=S. matsudana)
Fig. 5 Estimation of nuclear DNA content by using flow cytometry． (A)Simultaneous analysis of nuclei isolated from standard diploid species
(Salix gracilistyla)and from S. cavaleriei;(B)Simultaneous analysis of nuclei isolated from standard diploid species (S. gracilistyla)and
from S. × heteromera; (C)Simultaneous analysis of nuclei isolated from standard diploid species (S. gracilistyla)and from S. matsudana
6 植 物 分 类 与 资 源 学 报 第 37卷
书书书
!
#$% &’ () *$
%+
,-.
%
/+
0-,
-+
10
-1
,2
%
$-3 1% . -1 ,% 41$
,4
10
*4
-#
%.
0/
*
%4
!5
!6

0%
7)
%1
*%
0
-1
!
#$% 8 &’ (’) *+ ’) # !9 , - #’)$’$# 1 . !9 + ( ./ 0 1 ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ 6% 7) %1 *% 4% 3- +1 # 3% 1+ ,: /% 1) ; #% 4# 1 . 1) *$
%+
,-.
%
/+
0-,
-+
10
’
’
’
’
’
’
’
’
’
’
’
’
’
6/
%*
-%
0
()
;
#%
4
+<
0/
%*
-;
%1
0
()
;
#%
4
+<
*$+1 %0 5! 6= = > & ? @ A ?B C @ D EA FF C = D? => A =E ? >= F 5! 6> B E F =D == ?@ D ?E @ ?F > @F D @F F 5! 6 4% /% , 0 GH %4 3 % ; % 1 .- 0, 1* % ! #$%
,
-
#’)
$’$
&
>E
GI
J!
!
K
!L
J
!
!L
I
J
J
I
G
!L
J
A
D9
DD
&
!9
8
&’
(’)
*+
’)

@
?E
M
!L
J
K
!L
J
!L
J
!L
I
!L
J
!L
J
G
L I
G
L I
!L
J
F
D9
DD
B
!9
+
(
./
0
1
&
>A
M
!L
J
K
!
!L
J
I
!L
J
!L
J
G
L I
G
L I
!
E
D9
DD
?
M
N
OOO
O
L G
IJ
!$K N JI JI JJ L OOO OOO P ! #$
%
?’
J2
$+ 4+ /$
0
,2
/
$+ ,: /% ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ 6% 7) %1 *% 4% 3- +1 1 . 1) *$
%+
,-.
%
/+
0-,
-+
10
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
’
6/
%*
-%
0
()
;
#%
4
+<
0/
%*
-;
%1
0
(
23
4)5
,6
@E
=?
=
>>
D
>?
=
+
(
7
BF
)5
,6
&&
&
?@
>
()1
8
49
FA
C F
F
>=
D
&E
>
B=
?C
B?
>
!9
,
-
#’)
$’$
&
G
G
!
J
I
J
!
C
G
C
Q
!9
8
&’
(’)
*+
’)

@
G
G
!
RC
J
I
J
!
C
G
C
Q
!9
+
(
./
0
1
&
C
J
!
!
G
G
J
S
C
!
C
S
N
G!
G!
$
Q
N
!J
GG
!G
IG
G!
IG
GJ
GI
!!
!!
!!
IG
GG
!!
I
3 Discussion
The biparental inherited nrDNA ITS variation of
the putative hybrid Salix × heteromera and its suspec-
ted parents S. cavaleriei and S. matsudana were inves-
tigated． Our results showed that concerted evolution is
not completed in these three species． Both intraspecif-
ic and intra-individual polymorphisms were detected
for all the three species． Multiple nrDNA repeats were
common at interspecific，intraspecific，and intraindi-
vidual levels，arising both from organismal processes
such as hybridization and polyploidization and by ge-
nomic processes such as gene and chromosome seg-
ment duplication and various forms of homologous and
non-homologous recombination(Alvarez and Wendel，
2003)． The putative hybrid S. × heteromera showed
both intraspecific and intraindividual polymorphism of
nrDNA ITS region，and all the variant nucleotide
sites were perfectly additive (i. e．，chimera)of its
suspected parents S. matsudana and S. cavaleriei (Ta-
ble 3，Fig. 4)． Stochastic genomic processes men-
tioned above are not likely to produce such an addi-
tive nucleotide pattern． Moreover，the ITS sequence
diversity of S. × heteromera (0. 007)was considerably
higher than those of S. matsudana (0. 004) and
S. cavaleriei (0. 003)． Further，S. × heteromera occurs
only where both S. cavaleriei and S. matsudana are
present;it is a morphological intermediate between
S. cavaleriei and S. matsudana as indicated by the
features of leaf，stamen，ovary stipe，and PCA anal-
ysis of morphological traits． Taken together，these
findings suggest that S. × heteromera is a natural hy-
brid between S. cavaleriei and S. matsudana．
Salix matsudana and S. cavaleriei also showed
intraspecific and intraindividual polymorphism in the
ITS region． Considering that both species are tetra-
ploid，as shown by our flow cytometry analysis，it is
possible that both species are of allopolyploid (hy-
brid)origin and have merged and maintained both
ITS repeat types of their progenitors． Divergent re-
peats of ITS have been reported to be clearly main-
tained over tens of millions of years (Baumel et al．，
2001;Alvarez and Wendel，2003)．
71期 WU Jie et al．:Homoploid Hybridization between Native Salix cavaleriei and Exotic Salix matsudana …
Flow cytometry analysis revealed that Salix ×
heteromera is tetraploid and thus is a homoploid hy-
brid． In nature，homoploid hybrid speciation might
be a rare phenomenon;in that，parent species must
be closely related for the homoploid hybrid to be via-
ble，or the differences in chromosome arrangement
might affect mitosis (Ｒieseberg，1997;Coyne and
Orr，2004;Abbott and Ｒieseberg，2012)． The pa-
rental species S. cavaleriei and S. matsudana belong
to Salix subgenus Salix，and sections Wilsonia K. S．
Hao ex C. F． Fang ＆ A. K． Skvortsov and Salix，re-
spectively． A molecular phylogenetic study showed
that these two sections belong to different clades
(Azuma et al．，2000;Chen et al．，2010)，indica-
ting that S. cavaleriei and S. matsudana are not
closely related species． This is consistent with their
morphological differences． For example，they showed
differences in stamen number，which is an important
character in systematics of Salix (Ding，1995a;
Skvortsov，1999;Argus，2010) :S. cavaleriei has 6
－ 12 stamens，and S. matsudana has 2 stamens．
They also differed ecologically: S. matsudana is
somewhat adapted to arid and semiarid environment
(Skvortsov，1999)and grows along rivers or depres-
sions amidst sand in basin and plain areas，whereas
S. cavaleriei is a typical riparian species，requiring
high moisture，and only occurs around streams and
can even grow in water． Our field observations re-
vealed that many individuals of S. × heteromera are
big trees and apparently viable，and the hybrids
share their habitat closely with S. matsudana;how-
ever，the coexistence of the hybrid and its parents
can also be observed at one site (e. g．，along a
stream)． Moreover，the flowering time of S. × hetero-
mera overlaps with that of its parents． These findings
indicate that S. × heteromera is not significantly di-
vergent in ecology from its progenitors． All ho-
moploid hybrid species that have been documented
thus far are ecologically divergent from their parental
species (Abbott and Ｒieseberg，2012)． In addition，
the restricted distribution of S. × heteromera (only
occurs when both its parents are present)suggests
that it is sterile or its progeny is sterile and /or invia-
ble and therefore lacks the ability to expand beyond
its distribution range (i. e．，it is not a true species)．
However，this needs to be further investigated．
Chloroplast DNA is usually maternally transmit-
ted in angiosperms (Mogensen，1996) ，and se-
quencing can be used to determine hybrid origin
(Zhou et al．，2008)． Our results from the four chlo-
roplast sequence datasets indicated that the hybrid-
ization is unidirectional， i. e．， asymmetric， with
Salix cavaleriei as the maternal parent． Hybridization
tends to be unidirectional at sites where one of the
parental species is rare，because the pollens deliv-
ered to the rare species would consist mainly of pol-
len from the common species (Ｒieseberg，1995;
Zhou et al．，2008)． In the habitat of our putative
hybrid，S. cavaleriei is rare and S. matsudana is
more abundant，and S. × heteromera shared its habi-
tat closely with S. matsudana． Under such circum-
stance，the rare species is usually the maternal par-
ent of the hybrid (Ｒieseberg，1995) ;this would
have been the possible reason for the asymmetric hy-
bridization observed．
Hybridization is also often associated with habi-
tats that have been altered by anthropogenic disturb-
ance (Abbott and Ｒieseberg，2012)． This might be
the case in our current study;Salix matsudana is
not documented in Floras as native species in Yun-
nan province (Ding，1995b;Fang et al．，1999)．
This species has long been used as an ornamental
plant and cultivated almost all over the temperate
zone in the world． It is widely cultivated around
farmlands，villages，and deforestation areas． The
cultivated plants might escape and naturalize;in-
deed，natural population of S. matsudana is at pres-
ent quite common in north-central Yunnan． There-
fore，crossing occurred between the previously allo-
patric，common and widespread S. matsudana and
the rare and narrowly distributed S. cavaleriei． Hy-
bridization between common and rare species might
have severe consequences for the rare species:if fit-
ness of the hybrid is lower relative to either parental
8 植 物 分 类 与 资 源 学 报 第 37卷
species or even sterile (i. e．，outbreeding depres-
sion)，the growth rate of the rare species may de-
cline below that required for replacement (i. e．，de-
mographic swamping)． If，however，the hybrid is
fertile or fitness decline is negligible，the hybrid
tends to backcross more frequently with the common
species and may displace the rare species (i. e．，ge-
netic assimilation) (Ｒieseberg and Wendel，1993;
Ｒhymer and Simberloff，1996;Ellstrand et al．，
1999;Wolf et al．，2001)． In our case，S. cavaleriei
did not seem to be seriously threatened by its hybrid-
ization with the exotic S. matsudana at present，be-
cause although S. cavaleriei is rare and highly re-
quires a moist habitat，some of its distribution range
is not invaded by S. matsudana (e. g．，Tengchong of
Yunnan province and south Sichuan province)．
However，if S. matsudana continues to invade the
distribution range of S. cavaleriei by means of human
intervention and becomes numerically superior com-
pared to S. cavaleriei，S. cavaleriei might become in-
creasingly endangered or even extinct through genet-
ic assimilation and /or outbreeding depression，re-
gardless of the fitness of hybrids． Therefore，our
study indicated that willows should be introduced for
purposes such as ornamentation and afforestation
with caution，since they may cross with indigenous
willow species and increase the risk of rare species
becoming extinct．
Acknowledgments:We thank Zhikun Wu，Detuan Liu of
Kunming Institute of Botany for their kind assistance in field
work．
Ｒeference:
Abbott Ｒ，Albach D，Ansell S et al．，2013． Hybridization and specia-
tion ［J］． Journal of Evolutionary Biology，26:229—246
Abbott Ｒ，Ｒieseberg L，2012． Hybrid Speciatioin ［M］． Chichester:
John Wiley ＆ Sons，Ltd
Alvarez I，Wendel JF，2003． Ｒibosomal ITS sequences and plant phy-
logenetic inference ［J］． Molecular Phylogenetics and Evolution，
29:417—434
Argus GW，2010． Salix ［A］． / / Committee FoNAE ed．，Flora of
North America ［M］． Oxford and New York:Oxford University
Press，23—162
Arnold ML，1997． Natural Hybridization and Evolution ［M］． New
York:Oxford University Press，USA
Azuma T，Kajita T，Yokoyama J et al．，2000． Phylogenetic relation-
ships of Salix (Salicaceae)based on rbcL sequence data ［J］．
American Journal of Botany，87:67—75
Baumel A，Ainouche M，Levasseur J，2001． Molecular investigations
in populations of Spartina anglica CE Hubbard (Poaceae)inva-
ding coastal Brittany (France)［J］． Molecular Ecology，10:
1689—1701
Chen JH，Sun H，Wen J et al．，2010． Molecular phylogeny of Salix
L． (Salicaceae)inferred from three chloroplast datasets and its
systematic implications ［J］． Taxon，59:29—37
Chiang TY，Schaal BA，Peng CI，1998． Universal primers for ampli-
fication and sequencing a noncoding spacer between the atpB and
rbcL genes of chloroplast DNA ［J］． Botanical Bulletin of Acade-
mia Sinica，39 (4) :245—250
Coyne JA，Orr HA，2004． Speciation ［M］． Sunderland，MA:Sinau-
er Associates
Demesure B，Sodzi N，Petit ＲJ，1995． A set of universal primers for
amplification of polymorphic non-coding regions of mitochondrial
and chloroplast DNA in plants ［J］． Molecular Ecology，4:
129—134
Ding T，1995a． Origin，divergence and geographical distribution of
Salicaceae ［J］． Acata Botanica Yunnanica，17:277—290
Ding T，1995b． Salicaceae ［M］． Beijing:Science Press
Doleel J，Greilhuber J，Suda J，2007． Estimation of nuclear DNA
content in plants using flow cytometry ［J］． Nature Protocols，2:
2233—2244
Doyle JJ，Doyle JL，1987． A rapid DNA isolation procedure for small
quantities of fresh leaf tissue ［J］． Phytochemical Bulletin，19:
11—15
Drummond A，Ashton B，Buxton S et al．，2011． Geneious v5. 4
［CP］． Auckland，New Zealand:Biomatters Ltd
Edgar ＲC，2004． MUSCLE:multiple sequence alignment with high
accuracy and high throughput ［J］． Nucleic Acids Ｒesearch，32:
1792—1797
Ellstrand NC，Prentice HC，Hancock JF，1999． Gene flow and intro-
gression from domesticated plants into their wild relatives ［J］．
Annual review of Ecology and Systematics，539—563
Fang Z，Zhao S，Skvortsov AK，1999． Salicaceae ［M］． Flora of Chi-
na，4:139—274
Hamzeh M，Sawchyn C，Perinet P et al．，2007． Asymmetrical natural
hybridization between Populus deltoides and P. balsamifera (Sal-
icaceae) ［J］． Botany，85:1227—1232
Hardig T，Brunsfeld S，Fritz Ｒ et al．，2000． Morphological and mo-
lecular evidence for hybridization and introgression in a willow
(Salix)hybrid zone ［J］． Molecular Ecology，9:9—24
Janzen DH，2009． A DNA barcode for land plants ［J］． Proceedings
of the National Academy of Sciences of the United States of Ameri-
ca，12794—12797
91期 WU Jie et al．:Homoploid Hybridization between Native Salix cavaleriei and Exotic Salix matsudana …
Kaplan Z，Fehrer J，2007． Molecular evidence for a natural primary
triple hybrid in plants revealed from direct sequencing ［J］． An-
nals of Botany，99:1213—1222
Kimura M，1980． A simple method for estimating evolutionary rates of
base substitutions through comparative studies of nucleotide se-
quences ［J］． Journal of Molecular Evolution，16:111—120
Leskinen E，Alstrom-Ｒapaport C，1999． Molecular phylogeny of Sali-
caceae and closely related Flacourtiaceae:evidence from 5. 8 S，
ITS 1 and ITS 2 of the rDNA ［J］． Plant Systematics and Evolu-
tion，215:209—227
Lledo MD，Crespo MB，Cameron KM et al．，1998． Systematics of
Plumbaginaceae based upon cladistic analysis of rbcL sequence
data ［J］． Systematic Botany，21—29
Mogensen HL，1996． The hows and whys of cytoplasmic inheritance in
seed plants ［J］． American Journal of Botany，383—404
Ｒechinger KH，1992． Salix taxonomy in Europe-problems，interpreta-
tions，observations ［J］． Proceedings of the Ｒoyal Society of Ed-
inburgh Section B-Biological Sciences，98:1—12
Ｒhymer JM，Simberloff D，1996． Extinction by hybridization and intro-
gression ［J］． Annual Ｒeview of Ecology and Systematics，83—109
Ｒieseberg L，Ellstrand N，Arnold M，1993． What can molecular and
morphological markers tell us about plant hybridization?［J］．
Critical Ｒeviews in Plant Sciences，12:213—241
Ｒieseberg LH，1995． The role of hybridization in evolution:old wine
in new skins ［J］． American Journal of Botany，82:944—953
Ｒieseberg LH，1997． Hybrid origins of plant species ［J］． Annual Ｒe-
view of Ecology and Systematics，359—389
Ｒieseberg LH，Archer MA，Wayne ＲK，1999． Transgressive segrega-
tion，adaptation and speciation ［J］． Heredity，83:363—372
Ｒieseberg LH，Wendel JF，1993． Introgression and its consequences
in plants ［J］． Hybrid Zones and the Evolutionary Process，70—
109
Ｒudyka E，1990． Chromosome numbers of vascular plants from the va-
rious regions of the USSＲ ［J］． Bot Zhurn，75:1783—1786
Saghai-Maroof MA，Soliman KM，Jorgensen ＲA et al．，1984． Ｒiboso-
mal DNA spacer-length polymorphisms in barley:Mendelian in-
heritance，chromosomal location，and population dynamics ［J］．
Proceedings of the National Academy of Sciences，81:8014
Skvortsov AK，1999． Willows of Ｒussia and Adjacent Countries ［M］．
Joensuu，Finland:University of Joensuu
Soltis PS，Soltis DE，2009． The role of hybridization in plant specia-
tion ［J］． Annual Ｒeview of Plant Biology，60:561—588
SPSS，2002． SPSS 11. 5 for Windows ［CP］． Illinois，USA:SPSS
Inc Chicago
Suda Y，1958． The number of chromosomes in some Japanese salica-
ceous plants． I ［J］． Institutes Tohoku University，Ser 4:1—4
Suda Y，1963． The chromosome numbers of salicaceous plants in rela-
tion to their taxonomy ［J］． Science Ｒeports of the Ｒesearch，29:
413—430
Tamura K，Peterson D，Peterson N et al．，2011． MEGA5:molecular
evolutionary genetics analysis using maximum likelihood，evolu-
tionary distance，and maximum parsimony methods ［J］． Molec-
ular Biology and Evolution，28:2731—2739
Wolf DE，Takebayashi N，Ｒieseberg LH，2001． Predicting the risk of
extinction through hybridization ［J］． Conservation Biology，15:
1039—1053
Zha HG，Milne ＲI，Sun H，2008． Morphological and molecular evi-
dence of natural hybridization between two distantly related Ｒho-
dodendron species from the Sino-Himalaya ［J］． Botanical Jour-
nal of the Linnean Society，156:119—129
Zhou Ｒ，Gong X，Boufford D et al．，2008． Testing a hypothesis of u-
nidirectional hybridization in plants:Observations on Sonnera-
tia，Bruguiera and Ligularia ［J］． BMC Evolutionary Biology，
8:149
01 植 物 分 类 与 资 源 学 报 第 37卷