全 文 :红豆蔻专性异交主导下的主动自交机制
崔煜文1ꎬ2ꎬ 李庆军1∗
(1 中国科学院西双版纳热带植物园ꎬ 云南 勐腊 666303ꎻ 2 中国科学院大学ꎬ 北京 100049)
摘要: 主动自交一直以来都被认为是植物在面对极端环境ꎬ 例如传粉者稀缺的情况下所采取的繁殖保障机
制ꎬ 它不利的一面主要表现为近交衰退和花粉或胚珠贴现ꎮ 在姜科植物部分属ꎬ 如山姜属和豆蔻属当中ꎬ
存在着一种被称作花柱卷曲 (flexistyly) 的特殊机制ꎮ 过去的研究认为这种机制的意义在于避免自交和性
别干扰ꎬ 但是根据我们的野外观察ꎬ 如果下垂型花朵的花粉在其花柱开始向下弯曲时仍然留在花药之上ꎬ
柱头就很有可能接触到花粉从而完成主动自交ꎮ 为此我们探索了红豆蔻 (Alpinia galanga) 的繁殖系统ꎬ 研
究结果表明: (1) 红豆蔻自交亲和ꎬ 并存在一定的近交衰退ꎻ (2) 上举型为异交ꎬ 较之下垂型投入了更多
的资源ꎻ (3) 上举型的花粉胚珠比显著低于下垂型ꎬ 由于红豆蔻的胚珠数是固定的六枚ꎬ 故两种表型花粉
胚珠比的差异反映了其花粉数目的差异ꎮ 这可能与上举型仅仅为下垂型提供花粉ꎬ 而下垂型除为上举型提
供花粉外ꎬ 还需要有足够的花粉以备自交所需ꎮ 因此ꎬ 红豆蔻当中存在主动自交机制ꎬ 并且其花柱卷曲机
制在避免不必要的自交和性别干扰的同时ꎬ 也为作为繁殖保障所必需的延迟主动自交创造了有利条件ꎮ 同
时通过控制主动自交发生的时间达到避免三种功能之间的相互冲突ꎮ 这种只在姜科植物当中存在的特殊机
制充分展示了其进化过程中对不利环境的适应ꎮ
关键词: 红豆蔻ꎻ 主动自交ꎻ 繁殖保障ꎻ 花柱卷曲ꎻ 适应性
中图分类号: Q 945 文献标志码: A 文章编号: 2095-0845(2015)06-793-08
Autonomous Self ̄pollination under Dominant Flexistylous Outcorssing
Mechanism in Alpinia galanga (Zingiberaceae)
CUI Yu ̄wen1ꎬ2ꎬ LI Qing ̄jun1∗
(1 Xishuangbanna Tropical Botanical Gardenꎬ Chinese Academy of Sciencesꎬ Menglaꎬ Yunnan 666303ꎬ Chinaꎻ
2 University of Chinese Academy of Sciencesꎬ Beijing 100049ꎬ China)
Abstract: Autonomous self ̄pollination is considered to be a reproductive assurance mechanism for plants when fa ̄
cing extreme environments that scarce in pollinatorsꎬ and it also has disadvantages like inbreeding depression and
pollen / ovule discounting. In several genera of Zingiberaceae like Alpinia and Amomumꎬ a specialized mechanism
named flexistyly existsꎬ precedent studies consider its significance as preventing unnecessary self ̄pollination and sex ̄
ual interference. However according to our field observationꎬ if pollen grains on cataflexistylous (CATA) flowers re ̄
main on the anther when the style curves downwardlyꎬ potential exists that the stigma contacts the pollenꎬ thus au ̄
tonomous self ̄pollination occurs. Hereꎬ we studied the breeding system of Alpinia galangaꎬ the results show that
(1) Alpinia galanga is self ̄compatible in which inbreeding depression occurs to some degreeꎻ (2) The anaflexi ̄
stylous (ANA) morph of Alpinia galanga allocates more resource into outcrossing than the CATA morphꎻ (3) The
P / O ratio of the ANA morph is significantly lower than that of the CATA morphꎬ as Alpinia galanga has constant six
ovules in each ovaryꎬ the significant difference in P / O value reflects the contrast in pollen production. This phenom ̄
enon may be interpreted as the ANA morph provides pollen exclusively for the CATA morph. But the CATA morphꎬ
植 物 分 类 与 资 源 学 报 2015ꎬ 37 (6): 793~800
Plant Diversity and Resources DOI: 10.7677 / ynzwyj201515070
∗ Author for correspondenceꎻ E ̄mail: qjli@xtbg ac cn
Received date: 2015-04-23ꎬ Accepted date: 2015-08-04
作者简介: 崔煜文 (1992-) 男ꎬ 硕士研究生ꎬ 主要从事进化生态学研究ꎮ E ̄mail: capebulbs@foxmail com
despite delivers pollen to the ANA morphꎬ also need its own pollen for potential autonomous self ̄pollination. In sum ̄
maryꎬ autonomous self ̄pollination exists in Alpinia galangaꎬ and while flexistyly functions to avoid unnecessary self ̄
pollination and sexual interferenceꎬ it also provides advantages for delayed autonomous self ̄pollination as a necessary
reproductive assurance and preventing conflict among these three major features by controlling the time of autonomous
self ̄pollination. This peculiar mechanism in Alpinia galanga thoroughly demonstrates its adaptation to unfavorable
surrounding during the evolutionary process.
Key words: Alpinia galangaꎻ Autonomous self ̄pollinationꎻ Reproductive assuranceꎻ Flexistylyꎻ Adaptive significance
Self ̄ and cross ̄pollination are two primary re ̄
productive strategies in plants. In self ̄compatible
speciesꎬ self ̄pollination always coexists with cross ̄
pollination (Lloyd and Schoenꎬ 1992)ꎬ comprising
mixed mating system. The recurrent mixed mating
mechanisms consist of pollination by flowers on the
same plant including autogamy and geitonogamyꎬ
plus the combination of cleisogamy and chasmogamyꎬ
along with facilitated autogamy (Richardsꎬ 1986).
Among those varying mating systemsꎬ the autonomous
self ̄pollination is relatively less commonꎬ which oc ̄
curs in flowers that are able to cross ̄pollinationꎬ
whose occurrence does not need facilitation from an ̄
ything but the flower itself. Autonomous self ̄pollina ̄
tion can be categorized in terms of the beginning
time relating to outcrossing into three types: priorꎬ
competing and delayed autonomous self ̄pollination
(Lloyd and Schoenꎬ 1992)ꎬ in whichꎬ the delayed
modeꎬ comparing with the other twoꎬ for its occurrence
succeeds the culmination of outcrossingꎬ has the least
impact on cross ̄pollination.
With regard to the adaptive significance of au ̄
tonomous self ̄pollinationꎬ Darwin (1876) claims the
mechanism ubiquitously functions as a reproductive
assurance in the situations where scarce pollinator
activityꎬ harsh environment or competitors for polli ̄
nators exist. For instanceꎬ Epipogium roseum ( D.
Don) Lindl.ꎬ an obligate self ̄pollinated saprophytic
orchidꎬ its breeding system may possibly be evolved
due to its ephemeral flowering period owing to its sa ̄
prophytic natureꎬ plus inhabiting in understory that
suffers from paucity of pollinator activity since being
achingly shaded (Zhou et al.ꎬ 2012). Though plants
benefit from the reproductive assurance introduced
by autonomous self ̄pollinationꎬ they are handicapped
by its drawbacks primarily involving inbreeding de ̄
pressionꎬ namely the self ̄pollinated offspring is low ̄
er in fitness than those pollinated by allogamyꎬ a ̄
longside ovule discountingꎬ that is autogamous polli ̄
nated ovulesꎬ which produce low ̄fitness seedsꎬ com ̄
pete for resources with other cross ̄pollinated ones
that produce normal offspringꎬ decreasing fitness of
the latter.
The ginger familyꎬ Zingiberaceaeꎬ is an outcross ̄
ing dominated groupꎬ exhibiting various breeding
systems (Gao et al.ꎬ 2005aꎬbꎻ Wang et al.ꎬ 2005ꎻ
Gaoꎬ 2008) where multifold species harbor autono ̄
mous self ̄pollination. For instanceꎬ Roscoea schnei ̄
deriana (Loesener) Cowleyꎬ an alpine speciesꎬ a ̄
chieves its self ̄pollination process by means of arc ̄
hing its terminus of style towards the anther at the
incipience of blooming (Zhang and Liꎬ 2008)ꎻ Ad ̄
ditionallyꎬ Caulokaempferia coenobialis (Hance) K.
Larsen performs autogamy via secreting two oil ̄like
pollen drops through the dehiscence of two pollen
sacs which fuse into a larger one and gradually slide
towards the horizontally grown stigma (Wang et al.ꎬ
2005).
In a few genera of Zingiberaceaeꎬ e g. Alpinia
and Amomumꎬ there is a mechanism called flexistyly
(Li et al.ꎬ 2001aꎬbꎻ Gao et al.ꎬ 2005ꎻ Li and Xiaꎬ
2006ꎻ Liu et al.ꎬ 2007ꎻ Liuꎬ 2007ꎻ Chen and Liꎬ
2008)ꎬ plants with this mechanism have anaflexi ̄
stylous (ANA) and cataflexistylous (CATA) individ ̄
uals within populationꎬ each individual have only
one kind of flowers. The style tip of ANA morph
curves downwards and being receptive in the morn ̄
ing with pollen sacs remain intactꎬ and the flower is
in its female state at this time. Then the style tip
bends upwards until situated on the upper side of the
497 植 物 分 类 与 资 源 学 报 第 37卷
anther in the afternoonꎬ following the dehiscence of
pollen sacsꎬ dispersing pollen and the flower is in
its male state. The other morphꎬ CATAꎬ displays the
very reverse pattern of ANA morph. Moreoverꎬ all
individuals of identical morph in a population are
synchronous in stylar movementꎬ hence preventing
pollen deposition onto stigmas of the same morphꎬ
fulfilling outcrossing. Thereforeꎬ flexistylyꎬ the idio ̄
syncratic strategyꎬ realizes dichogamy by setting the
flower in disparate sexual states through different
periods of its lifetimeꎬ together with implementing
herkogamy via relative movement of style and anther
(Cui et al.ꎬ 1995ꎻ Li et al.ꎬ 2001aꎻ Sun et al.ꎬ
2007ꎬ 2011).
As for the adaptive significance of flexistylyꎬ
precedent studies solely regarded it as evading self ̄
pollination through synchronous heterodichogamy (Li
et al.ꎬ 2001b). Neverthelessꎬ recent studies (Sun et
al.ꎬ 2007ꎬ 2011) proposed that merely fostering
outcrossing is inadequate for understanding the adap ̄
tive significance of flexistyly. Firstꎬ dichogamy is
sufficient for hindering autogamyꎬ thus it seems re ̄
dundant for the flower to bear herkogamy. Further ̄
moreꎬ the style of ANA morphꎬ after the stigma hav ̄
ing received pollen in the morningꎬ will curve to up ̄
per side of the anther in the afternoon where is inac ̄
cessible to pollinatorsꎬ which is also superfluous if
just for avoiding self ̄pollination. Nonethelessꎬ exclu ̄
sively dichogamy in the absence of herkogamy is not
efficacious in preventing sexual interferenceꎬ Sun et
al. (2011) substantiated that stylar movement is ca ̄
pable of preventing interference between male and
female functions as well as increasing pollen outputꎬ
consequently increasing male fitness via comparing
the seed paternity of CATA morph as pollen receptor
with the stylar movement of supplier ANA morph had
been restrained with the CATA seed paternity with
normal provider ANA flower and found significantly
less seed set in manipulated group.
Conventionallyꎬ taxa bearing flexistyly are con ̄
sidered to be outcrossing ̄dominatedꎬ even those car ̄
rying a portion of autogamous features are considered
as sporadically occurred ( Chen and Liꎬ 2008 ).
Howeverꎬ according to our field observationꎬ the
CATA morph of greater galangal ( Alpinia galanga
(L.) Willd.) curves its style so extensively that the
stigma is able to touch the pollen sacs. Thus if no
pollinator visitation occurs during its male periodꎬ
the pollen would remain on the pollen sacsꎬ as a
consequence autonomous self ̄pollination is in pros ̄
pect. In additionꎬ Alpinia galanga blossoms during
the monsoon seasonꎬ when pollinator activity has
been limited by the weatherꎬ hence autonomous self ̄
pollination may be chosen as reproductive assurance.
Whereas many studies on flexistyly in company with
some peculiar autogamous mechanisms in Zingiber ̄
aceae have been published and dramatic advance
been acquiredꎬ the adaptive significance of autono ̄
mous self ̄pollination under dominance of flexistylyꎬ
an unparalleled strategy in ginger familyꎬ is awaiting
further investigation. This study aims at answering
two questions: (1) Does the CATA morph of Alpinia
galanga possess autonomous self ̄pollination mecha ̄
nism? (2) If soꎬ what is the adaptive significance?
1 Materials and Methods
1 1 Study Object
The greater galangal (Alpinia galanga (L.) Willd.
(Fig 1: A - B)ꎬ a rhizomatic perennial in ginger
family (Zingiberaceae) which habitually inhabits in
understoryꎬ may erect as tall as 2 m. It flowers from
April to Septemberꎬ the raceme holding more than
200 flowers raise from the apex of pseudostemꎬ each
flower is made up by an apical petal and two lateral
ones plus a pair of degenerated staminodes and a la ̄
bellum with elongated stem. Seed pods are scarlet
when thoroughly ripen. Alpinia galanga distributes
in tropical secondary forests and has been semi ̄culti ̄
vated by ethnic minorities in Southeastern Asia like
Dai for culinary use as an essential flavorꎬ thus be ̄
ing economically significant.
The study employed cultivated communities grown
in partly shaded understory of rubber plantation at
Xishuangbanna Tropical Botanical Garden (XTBG)
5976期 CUI and LI: Autonomous Self ̄pollination under Dominant Flexistylous Outcorssing Mechanism in
(21°55′Nꎬ 101°15′E) as the research material.
1 2 Methods
1 2 1 Basic reproductive features One flower was
randomly specified in each morph and their changes
with timeꎬ especially the angle between style and
antherꎬ were recorded and photographed sidewise at
one hour interval for 26 hours from 6 ∶ 00 am in the
first morningꎬ with pollinator behavior as well as
blooming pattern of flowers in the vicinity also ob ̄
served and documented concurrently.
Fig 1 A. Alpinia galanga (L.) Willd.ꎬ showing the plant and habitatꎻ B. Platynopoda magnifica pollinating Alpinia galangaꎬ showing
massive pollen grains on its backꎻ C-E. Status of a single ANA flower in the morningꎬ afternoon and night of the same dayꎻ F-H. Status of
a single CATA flower in the morningꎬ afternoon and night of the same dayꎬ showing the stamens of both morphs have bent towards the label ̄
lum at nightꎬ preventing pollinators from pollinating the flowersꎻ I. If pollen remains on the anther of CATA flower after the style curves
downwardsꎬ it is possible that stigma contacts the remaining pollenꎬ completing autonomous self ̄pollinationꎻ J. The infructescence of Alpinia
galanga in its natural stateꎻ K. Infructescence of bagged ANA morphꎬ on which no flowers set fruitꎻ L. Infructescence of bagged CATA mor ̄
phꎬ showing some fruit set but much less than the natural stateꎬ suggesting autonomous self ̄pollination occurred with inbreeding depression
697 植 物 分 类 与 资 源 学 报 第 37卷
1 2 2 Floral characteristics One flower per inflores ̄
cence was designated at random in different plants of
both morphs for more than 20 inflorescencesꎬ each part
of which was measured with a vernier caliper. Addi ̄
tionallyꎬ on each individual of either morph one single
flower bud that about to open in the next morning was
arbitrarily cut off at duskꎬ later the ovaries and anthers
were isolated and fixed by 2 mL of 70% ethanol in two
Eppendorf tubes respectively for Pollen / Ovule (P / O)
ratio measurements under stereoscopeꎬ more than 20
plants per morph were sampled for the experiment.
1 2 3 Bagging experiment Inflorescences with few
or no blossoms opened were randomly selected and
blooming flowers were removedꎬ the numbers of re ̄
maining flowers were counted. Pollinators were pre ̄
cluded from visiting the flower by bagging the inflo ̄
rescences with fine nylon mesh bags. More than 30
inflorescences were processed in either morph. Be ̄
sidesꎬ more than 30 extra inflorescences in each mor ̄
ph were also marked at the same time with flowers on
which counted as the control group. Infructescences
were collected one month after all flowers faded and
fruit setꎬ seed set along with average seed set per
fruit were countedꎬ in which fruit set was measured
via dividing overall fruit set per inflorescence by total
number of flowers on the same inflorescenceꎬ while
seed set was defined as division of the collective num ̄
ber of ovules produced by the actual number of normal ̄
ly ̄developed seeds of identical inflorescenceꎬ and seed
set per fruit was equal to average number of seeds yiel ̄
ded by a single fruit in a single inflorescence.
1 2 4 Data analysis All the statistical tests were
implemented in R by t ̄testꎬ and the outcomes of bag ̄
ging experiments were plotted in Microsoft Excelꎬ
the photos of relative movement between style and
anther was processed in Adobe Photoshop for meas ̄
uring the angle between the two floral organsꎬ then
the result was plotted in R.
2 Results
2 1 Basic reproductive features
The flower bud of Alpinia galanga cracks in ap ̄
proximately 20 pm at duskꎬ and completely opens in
roughly 4 am of next morningꎬ at about 5 am the pol ̄
len sacs of CATA morph dehisce with pollen start to
spread out. The style of ANA morph commence ris ̄
ing at nearly 12 pm in the middayꎬ while that of CA ̄
TA morph begins to bend down at 1 pmꎬ the ANA
morph does not release pollen until 15 pm or so.
Each flower lasts one dayꎬ self ̄compatible and nec ̄
tar ̄providing. The result of stylar movement of the
two morphs through time is shown in Fig 2. Anthers
of both morphs snap to the labellum at nightꎬ shiel ̄
ding from pollinator contact (Fig 1: C-H). In CA ̄
TA morph if pollen still stay on the anther when style
curves downwardsꎬ possibility exists that stigma con ̄
tact the pollenꎬ hence autonomous self ̄pollination
takes place (Fig 1: I).
Fig 2 Change of the angle between style and stamen over timeꎬ
in which the value is positive when the style positions
above the stamenꎬ and negative when below
2 2 Floral characteristics
The result of floral characteristics are demon ̄
strated in Table 1ꎬ showing the ANA morph signifi ̄
cantly outcompetes its counterpart in features of flow ̄
er parts like flower pedicel lengthꎬ ovary sizeꎬ label ̄
lum stem as well as filament sizeꎬ but the CATA
morph is significantly higher in P / O ratio.
7976期 CUI and LI: Autonomous Self ̄pollination under Dominant Flexistylous Outcorssing Mechanism in
Table 1 The result of floral character measurement of two morphs of Alpinia galanga
Characteristics
ANA
mean se
CATA
mean se
t df P
flowers per inflorescence 210 10 207 10 -0 1977 75 0 8348
pedicel length 15 464 0 449 11 114 0 312 8 1716 58 3 16E ̄11
ovary length 5 537 0 128 4 303 0 087 8 1959 58 2 88E ̄11
ovary diameter 3 170 0 055 2 644 0 028 8 9424 58 1 65E ̄12
corolla tube diameter 2 985 0 049 2 715 0 062 3 3462 41 0 001762
top petal width 6 86 0 167 6 375 0 078 2 7887 58 0 007146
side petal width 4 98 0 129 4 675 0 088 2 0062 58 0 04951
labellum stem length 11 277 0 326 9 761 0 234 3 8648 58 0 0002832
filament thickness 0 674 0 010 0 627 0 014 2 4946 58 0 01548
pollen / ovule ratio 2288 90 2847 114 -3 7896 43 0 0004649
2 3 Pollinator behavior
The chief pollinator of Alpinia galanga is car ̄
penter bee (Platynopoda magnifica) (Fig 1B)ꎬ When
visiting the flowers of one inflorescenceꎬ it does not
switch to another inflorescence until all opened flow ̄
ers on the current one have been visitedꎬ during its
visitation on one inflorescence it starts from the lower
flowers and moving upward. One flower can be visi ̄
ted several times by more than one pollinators during
its lifetime and conspicuous variance exists in the
duration of visitation on flowers by a pollinator.
2 4 Bagging Experiment
None of the bagged ANA group set any fruit
(0 00 ± 0 00)% (N = 32) (mean ± SE)ꎬ conse ̄
quently its seed set rate and seed set per fruit are all
zero (Fig 1K). Yet bagged CATA group produces
the fruit set of (9 78 ± 1 71)% (N= 35ꎬ Fig 1L)ꎬ
the t ̄test yields t =5 4726ꎬ df = 65ꎬ P < 0 0001∗∗∗ .
On the other handꎬ the natural ANA renders (44 5
± 2 4)% (N= 41) while the CATA delivers (40 8
± 2 3)% (N = 36ꎬ Fig 1J)ꎬ the outcome of t ̄test
is: t= 1 1222ꎬ df= 75ꎬ P= 0 2654 (ns)ꎬ the rami ̄
fication of comparison between CATA natural and
bagged groups is t=10 775ꎬ df = 69ꎬ P < 0 0001∗∗∗ .
With regard to seed setꎬ ANA natural group gives
out (19 7 ± 1 0 )% (N = 38) and CATA natural
group ( 27 8 ± 2 2 )% ( N = 35)ꎬ while CATA
bagged produces (6 70 ± 1 26)% (N= 30). The t ̄
test demonstrates that in between natural ANA and
CATA groups t= -3 438ꎬ df= 71ꎬ P < 0 01∗∗ꎬ and
CATA natural and bagged groups t = 8 0023ꎬ df =
63ꎬ P < 0 0001∗∗∗ (Fig 3A)ꎻ As for seed set per
fruitꎬ the natural ANA and CATA groups return
2 90 ± 0 17 (N = 38) and 3 89 ± 0 14 (N = 35)
respectively. In contrastꎬ the output of CATA bagged
group is 3 27 ± 0 27 (N= 30). The outgrowth of t ̄
test is: natural ANA vs CATA: t= 4 3825ꎬ df= 71ꎬ
P < 0 0001∗∗∗ꎬ CATA natural vs bagged: t=2 1149ꎬ
df= 63ꎬ P < 0 05∗(Fig 3B).
3 Discussion
Alpinia galanga harbors representative flexistyly
mechanismꎬ both ANA and CATA morphs occur in
every population. The outcome of floral characteris ̄
tics measurement evidently claims that ANA flowers
sampled significantly surpass its counterpart in flow ̄
er pedicel lengthꎬ ovary sizeꎬ labellum stem as well
as filament sizeꎬ these increased floral investments
may be deemed as the ANA morph exceeds the CA ̄
TA morph in expenditure in floral presentationꎬ
which can be ascribed to that ANA morph is only a ̄
ble to process cross ̄pollination so it has to attract
more pollinators for reproductive success.
In accordance with classic P / O value theory (Cr ̄
udenꎬ 1977)ꎬ those who bear a lesser P / O value
are prone to self ̄pollination and vice versa. Nonethe ̄
lessꎬ as there are only fixed 6 ovules in each ovary
of Alpinia galangaꎬ therefore the ANA morphꎬ with
smaller P / O valueꎬ disperses less pollen than its coun ̄
terpart that maintains a larger P / O value. This may
897 植 物 分 类 与 资 源 学 报 第 37卷
Fig 3 A. Fruit and seed set rateꎬ in which the bagged ANA group doesn’t set any fruit so its data in fruit and seed
set rate plus seed set per fruit are all zeroꎻ B. Seed set per fruit of Alpinia galanga
be attributed to the fact that there are no other pollen
receptor of the ANA morph except the CATA morphꎬ
and despite satisfying the ANA morph for outcross ̄
ingꎬ pollen from CATA morph are prepared for de ̄
layed autonomous self ̄pollination when pollinator ac ̄
tivity is insufficientꎬ so the CATA morph allocates
more resource in generating pollen. As a resultꎬ the
fact in Alpinia galanga that is seemingly contradict to
the classical theory exactly champions our viewpoint
that the ANA morph inclines to outcrossing while its
counterpart favors self ̄pollination. Herein the P / O
valueꎬ in consonance with what Queller (1984) pro ̄
posesꎬ mirrors the sexual allocation of the plant.
Moreoverꎬ through our observationꎬ the ANA ̄
morph disperse pollen at roughly 15 amꎬ when polli ̄
nators are far less active than the morningꎬ sugges ̄
ting pollen limitation lies in the CATA morph. The
ANA morphꎬ howeverꎬ is at its female state in the
morning when pollinators are vigorousꎬ giving rise to
more pollen reception that pertinent to high fruit set.
Furthermoreꎬ in association with the floral measure ̄
ment data mentioned previouslyꎬ the ANA flower
produces notably less pollen than the CATA morphꎬ
but carries significantly larger ̄sized ovaryꎬ indica ̄
ting in Alpinia galangaꎬ the ANA morph invests more
resource in female function in guaranteeing offspring
fitnessꎬ while its counterpart allocates more energy
in male function ( pollen production)ꎬ embodying
the functional diversification in the two morphs of Al ̄
pinia galangaꎬ that isꎬ the ANA morph is more out ̄
crossing biased for providing excellent offspring
while its counterpart is self ̄pollination inclined as
reproductive assurance and serves as the only pollen
supplier for the ANA morph. The ANA morph is able
to spend more energy in nurturing ovaries that guar ̄
antees offspring fitness because of the CATA morph
provides pollen exclusively for the ANA while satis ̄
fying its own demand for autonomous self ̄pollination.
Accordingly the resource distribution pattern of Al ̄
pinia galangaꎬ in company with its mixed mating
systemꎬ are optimum measurements for adapting the
surrounding and maximizing the benefit of plants. In
Alpinia galanga populations investigatedꎬ the ANA
invests more in cross ̄pollination and female func ̄
tionꎬ its counterpart are self ̄pollinated as a repro ̄
ductive assurance while providing more resource in
male function. Considering the flower of Alpinia ga ̄
langa lasts only one dayꎬ plus the anther snaps to
the labellumꎬ impeding pollinators from getting into
the right place and pollinate the flower. Alpinia ga ̄
9976期 CUI and LI: Autonomous Self ̄pollination under Dominant Flexistylous Outcorssing Mechanism in
langaꎬ for ensuring reproductive successꎬ ingenious ̄
ly circumvents unnecessary or detrimental autogamy
and sexual interference ( Sun et al.ꎬ 2007ꎬ 2011)
while accommodating for necessary delayed autono ̄
mous self ̄pollination as reproductive assurance by a
single mechanismꎬ which is the result of long term
evolutionary process in coping with unfavorable con ̄
ditions in its surrounding.
References:
李庆军 (Li QJ)ꎬ 夏永梅 (Xia YM)ꎬ 2006. 姜科植物的花柱卷曲
性: 生态、 遗传与进化 [A] / / 全国系统与进化植物学研讨
会暨第九届系统与进化植物学青年研讨会 [C]ꎬ 中国陕西
西安
王英强 (Wang YQ)ꎬ 张奠湘 (Zhang DX)ꎬ 陈忠毅 (Chen ZY)ꎬ
2006. 姜科植物繁育系统与传粉生物学 [A] / / 全国系统与
进化植物学研讨会暨第九届系统与进化植物学青年研讨会
[C]ꎬ 中国陕西西安
Chen XC (陈绪超)ꎬ Li QJ (李庆军)ꎬ 2008. Outcrossing rates anal ̄
ysis of three Amomum species in Zingberaceae [ J] . Journal of
Yunnan University (Natural Science Edition) (云南大学学报:
自然科学版)ꎬ 30 (5): 531—534
Cruden RWꎬ 1977. Pollen ̄ovule ratios ̄conservative indicator of breed ̄
ing systems in flowering plants [J] . Evolutionꎬ 31 (1): 32—
46
Cui XL (崔晓龙)ꎬ Wei RC (魏蓉城)ꎬ Huang RF (黄瑞复)ꎬ
1995. A preliminary study on the genetic system of Amomum tsao ̄
ko [J] . Journal of Yunnan University (Natural Science Edition)
(云南大学学报: 自然科学版)ꎬ 17 (3): 290—297
Darwin CRꎬ 1876. The Effects of Cross and Self Fertilization in the
Vegetable Kingdom [M]. Murrayꎬ Londonꎬ UK
Gao JYꎬ 2008. Study on Reproductive Ecology of Zingiberaceae: Flo ̄
ral longevity do Matter [ D]. Mengla: Xishuangbanna Tropical
Botanical Gardenꎬ Chinese Academy of Sciences
Gao JY (高江云)ꎬ Ren PY (任盘宇)ꎬ Li QJ (李庆军)ꎬ 2005.
Advances in the study of breeding system and pollination biology
of gingers (Zingiberaceae and Costaceae) [ J] . Acta Phytotaxo ̄
nomica Sinica (植物分类学报)ꎬ 43 (6): 574—585
Li QJꎬ Xu ZFꎬ Kress WJ et al.ꎬ 2001a. Flexible style that encourages
outcrossing [J] . Natureꎬ 410 (6827): 432
Li QJ (李庆军)ꎬ Xu ZF (许再富)ꎬ Xia YM (夏永梅) et al.ꎬ
2001b. Study on the flexistyly pollination mechanism in Alpinia
plants (Zingiberaceae) [J] . Acta Botanica Sinica (植物学报)ꎬ
43 (4): 364—369
Liu M (刘敏)ꎬ 2007. Study on the adaptive significance of flexistyly
in Alpinia (Zingiberaceae) [D]. Mengla: Xishuangbanna Tropi ̄
cal Botanical Gardenꎬ Chinese Academy of Sciences
Liu M (刘敏)ꎬ Sun S (孙杉)ꎬ Li QJ (李庆军)ꎬ 2007. The relation
between stigma position and receptivity in two flexistylous gingers
[J] . Biodiversity Science (生物多样性)ꎬ 15 (6): 639—644
Lloyd DGꎬ Schoen DJꎬ 1992. Self ̄fertilization and cross ̄fertilization
in plants. 1. functional dimensions [J] . International Journal of
Plant Sciencesꎬ 153 (3): 358—369
Queller DCꎬ 1984. Pollen ̄ovule ratios and hermaphrodite sexual allo ̄
cation strategies [J] . Evolutionꎬ 38 (5): 1148—1151
Richards AJꎬ 1986. Plant Breeding Systems [M]. Allen and Unwinꎬ
Londonꎬ UK
Sun Sꎬ Gao JYꎬ Liao WJ et al.ꎬ 2007. Adaptive significance of flexi ̄
styly in Alpinia blepharocalyx ( Zingiberaceae): a hand ̄pollina ̄
tion experiment [J] . Annals of Botanyꎬ 99 (4): 661—666
Sun Sꎬ Zhang DYꎬ Ives AR et al.ꎬ 2011. Why do stigmas move in a
flexistylous plant? [ J ] . Journal of Evolutionary Biologyꎬ 24
(3): 497—504
Wang YQꎬ Zhang DXꎬ Renner SS et al.ꎬ 2005. Self ̄pollination by
sliding pollen in Caulokaempferia coenobialis ( Zingiberaceae)
[J] . International Journal of Plant Sciencesꎬ 166 (5): 753—
759
Zhang ZQꎬ Li QJꎬ 2008. Autonomous self ̄pollination provides repro ̄
ductive assurance in an alpine ginger Roscoea schneideriana
(Zingiberaceae) [J] . Annals of Botanyꎬ 102 (4): 531—538
Zhou Xꎬ Lin Hꎬ Fan XL et al.ꎬ 2012. Autonomous self ̄pollination
and insect visitation in a saprophytic orchidꎬ Epipogium roseum
(D. Don) Lindl. [ J] . Australian Journal of Botanyꎬ 60 (2):
154—159
008 植 物 分 类 与 资 源 学 报 第 37卷