Skip Navigation


AOBPreview originally published online on August 11, 2004
Annals of Botany 2004 94(4):583-591; doi:10.1093/aob/mch177
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
94/4/583    most recent
mch177v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Agricola
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


Annals of Botany 94/4, © Annals of Botany Company 2004; all rights reserved

Pollen Histochemistry and Pollen : Ovule Ratios in Zingiberaceae

YING-QIANG WANG1,2, DIAN-XIANG ZHANG1,* and ZHONG-YI CHEN1

1 South China Botanical Garden, The Chinese Academy of Sciences, Guangzhou 510650, China and 2 Zhongkai Agrotechnical College, Guangzhou 510225, China

* For correspondence. E-mail dx-zhang{at}scib.ac.cn

Received: 26 March 2004    Returned for revision: 13 May 2004    Accepted: 29 June 2004    Published electronically: 11 August 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

• Background and Aims Pollen grains of 37 species from 11 genera in the family Zingiberaceae were examined to assess qualitatively starch or lipid contents; the pollen grain and ovule numbers per flower and pollen : ovule ratios were also counted and calculated. Pollen : ovule ratios were studied to reveal patterns of variation in the Zingiberaceae.

• Methods Freshly open flowers with dehiscing anthers were collected at random from plants growing in natural habitats or in botanical gardens. Presence of lipids or starch in pollen grains was tested by Sudan solution and IKI solution, respectively, and examined under a microscope. To estimate the pollen and ovule numbers per flower, one anther from each bud was carefully dissected and all pollen grains were counted; ovaries were carefully dissected out of each flower and counted. Whenever possible, at least 10–15 buds were used in the determination.

• Key Results Thirty-three of all the 37 species examined had starchy pollen. Starch was not found in only four species and lipid was not found in only one species; among the four tribes in subfamily Zingiberoideae, all species of Zingibereae and Globbeae had pollen with no starch, Alpineae and Hedychieae had pollen with and without starch, whereas, all species of subfamily Costoideae had starchy pollen with abundant lipids. The mean pollen : ovule ratios in the members of the Zingiberaceae investigated range from 3·25 ± 1·56 to 616·52 ± 117·83.

• Conclusions The pollen nutrition types seemed not related to mating systems. The pollen : ovule ratios in members of the Zingiberaceae with the same breeding system are noticeably lower than those recorded by previous authors. The lower pollen : ovule ratios in this family are presumed to be related to the highly efficient pollination systems, mediated by pollen which can be quite glutinous and the relatively large stigma area. In most of the Alpinia species the anaflexistylous flowers have much larger numbers of pollen grains and higher pollen : ovule ratios than the cataflexistylous flowers. There are significant differences in mean pollen grain numbers and pollen : ovule ratios between different life forms but ovule numbers are approximately the same.

Key words: Zingiberaceae, pollen, starch, lipid, pollen-ovule ratio, breeding system


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Dafni (1992)Go commented that pollen histochemistry is possibly related to pollination mode, pollinator foraging behaviour and phylogeny. All angiosperm pollen grains contain stored food reserves in the form of starch and/or lipids and can be classified as two classes, ‘starchy’ and ‘starchless’ (Baker and Baker, 1979Go). Many studies have shown that all angiosperm pollen contains some lipids, while starchs are not always present; possible ecological and taxonomic (including phylogenetic) correlations with their presence or absence have also been discussed (Baker and Baker, 1979Go). The tendency for whole families characterized by having either starchy or starchless pollen, and hence the potential phylogenetic usefulness of this character, was recognized at an early date (Grayum, 1985Go). Baker and Baker (1979)Go refuted the suggestion that ‘primitive’ angiosperms would be characterized by starchy pollen and revealed that there is no evidence that ‘primitive’ families are any more likely to be made up of taxa with starchy pollen than others. However, Grayum (1985)Go suggested that starchless pollen is the primitive type in Araceae, and the same is true of monocots as a whole.

Generally, starchy pollen has been considered a feature of wind-pollinated (anemophilous) flowering plants, whereas insect-pollinated (entomophilous) species show a greater or lesser replacement of starch by sugar or lipids (Baker and Baker, 1979Go). Baker and Baker (1979)Go also concluded that selection in favour of starchless (oil-rich) pollen grains is where these are part or all of the reward received by bees and flies visiting the flowers and, conversely, where insect nutrition is not involved selection is in favour of starchy pollen. They also proposed that the mean diameter of the starchy grains is significantly larger than that of the starchless grains. Darwin (1877)Go put forward that pollen size was related to the length of the style in connection with pollen dimorphism. Baker and Baker (1979)Go elucidated that both starchy and starchless groups of species showed that mean diameters are significantly larger than those of their shorter-styled counterparts and they suggested that this was an indication that copious reserves are necessary to produce long pollen tubes required in these long-styled species.

After Cruden (1977)Go suggested that the pollen : ovule (P/O) ratio indicates plant breeding systems, many subsequent studies tried to relate P/O ratios with species' breeding systems, and P/O ratios are widely used in breeding system studies. Nevertheless, it is still unclear what the exact factors are that determine P/O ratios (Wyatt, 1984Go). It has been suggested that the P/O ratios reflect the likelihood of sufficient pollen grains reaching each stigma to result in maximum seed set, and thus, the more efficient the transfer of pollen, the lower the P/O ratio should be (Cruden, 1977Go). Charnov (1982)Go, however, argued that pollen-ovule ratios can be explained more directly as a result of local mate competition, with allocation ratios in hermaphroditic plants being driven by male/female gain curves for fitness. Cruden and Miller-Ward (1981)Go added that pollen grain size is a factor influencing pollen-ovule ratios and stated that species with relatively large stigma areas compared with the pollen-bearing area of the pollinator often have lower P/O ratios. Furthermore, Preston (1986)Go suggested that the interpretation of the P/O ratio must take into account sources other than the breeding system and proposed there can be no single general standard for evaluating P/O ratios, and any study that includes P/O ratios must discuss the criteria used to evaluate them. Vasek and Weng (1988)Go commented that, in any event, the standards for evaluating breeding systems need to be set at the level of family (or perhaps tribe) and each genus or other major taxon requires its own standard, and its breeding system cannot necessarily be judged on the standard for a mixture of other major taxa. Recently, pollen grain numbers and ovule numbers were found to correlate with style number, life form and breeding system, and that there are significant differences between some taxonomic groups in pollen grain numbers and ovule numbers but not in P/O ratios (Jürgens and Gottsberger, 2002Go). In fact, many different factors influence the pollen-ovule ratio, and species with different evolutionary histories may have very different P/O ratios despite having similar breeding systems (Preston, 1986Go).

The shift from predominant cross-pollination to predominant self-pollination is a recurrent theme in the evolutionary history of the angiosperms (Stebbins, 1970Go). The evolution of inbreeding in flowering plants has commonly been accompanied by a reduction in the number of pollen grains produced per flower (Vasek and Weng, 1988Go).

Zingiberaceae is a large family of animal-pollinated tropical monocotyledons, and members of this family display a broad range of pollination and breeding systems (Sakai et al., 1999Go; Zhang et al. 2003Go). A new behavioural outbreeding mechanism, ‘flexistyly’, was found recently in some species from the genera Amomum and Alpinia (Zingiberaceae) (Cui et al., 1996Go; Li et al., 2001Go). In this paper, pollen grains of 37 species from 11 genera in Zingiberaceae, were examined to assess starch or lipid contents, the pollen grain and ovule numbers per flower, and the P/O ratios to reveal patterns of variation in the family, aiming to shed more light on the relationship between pollination mechanism, phylogenetic constrains, breeding systems and the pollen nutrition types and the P/O ratios.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Pollen histochemistry
Between April and August 2003, freshly open flowers with dehiscing anthers were collected at random from plants. The pollen samples were immersed in a drop of IKI solution or a drop of Sudan IV solution, and examined under a microscope. A dark bluish-black colour (stained with IKI) indicates the presence of starch, and a red colour (stained with Sudan IV) indicates the presence of lipids (Dafni, 1992Go). The colour of stained starch or lipid was recorded and more than one sample was examined for each species. Baker and Baker (1979)Go have emphasized the necessity of using only mature pollen to test for starch content, inasmuch as even starchless pollen may accumulate starch at premature stages of development. For this reason, only fresh and mature pollen from dehiscing anther was surveyed during this study. In all, 32 species from 10 genera of four tribes in subfamily Zingiberoideae, and five species from one genus of subfamily Costoideae were tested. Information on aperture type, exine sculpturing and the diameters of pollen grains were cited from Liang (1988aGo, bGo).

Pollen : ovule ratios
During March 2002 to August 2003, flower buds 1 or 2 d prior to anthesis, were collected for a total of 22 Zingiberaceae taxa from natural or cultivated populations in Guangdong and Yunnan provinces in China and were placed in FAA solution, and refrigerated until needed. Only a single population was sampled for most of the taxa and only one bud was collected per individual (but more than one population was sampled for Amomum villusom and Caulokaempferia coenobialis). One anther from each bud was carefully dissected and all pollen grains were transferred to calibrated tubes and made up to 0·5 or 1 mL with staining solution. The suspension was stirred with a vortex mixer for 60 s, then 10–15 separate samples of 1 µL each were transferred onto slides and the pollen grain numbers were counted with the aid of a microscope. Ovaries were carefully dissected out of each flower and placed in a drop of water on a microscope slide. The entire placenta with attached ovules was removed via a longitudinal incision in the ovary wall. The ovules were carefully loosened from the placenta and spread in the drop of water to be counted at x40 magnification under a dissecting microscope. For each flower, the P/O ratio was calculated as the number of pollen grains in one anther divided by the number of ovules. Whenever possible, at least 10–15 buds were used in the determination.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Pollen histochemistry
The results of pollen histochemistry are presented in Table 1. Thirty-three of all the 37 species examined had starchy pollen. Starch was not found in only four species (Amomum villosum, Globba racemosa, Hedychium coronarium and H. flavum) and lipid was not found in only one species (Stahlianthus involucratus). Except for these five species, pollen of all the other species examined had both starch and lipid, and the starchy pollen of some species always stored rich lipids at the same time, e.g. the phylogenetically primitive taxa, Alpinia guinanensis, A. platychilus, A. zerumbet, and the phylogenetically derived taxa Pyrgophyllum yunnanense, Curcuma kwangsiensis and all species of Costus (Wu, 1994Go).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Pollen histochemistry of some plants in Zingiberaceae

 
Among the four tribes in subfamily Zingiberoideae, all species of Zingibereae (only two species examined) and Globbeae (only once species examined) showed entirely starchless pollen, Alpineae and Hedychieae show both starchless pollen and starch pollen. Whereas, all species of subfamily Costoideae showed starchy pollen with abundant lipids. Except for Amomum and Hedychium, all species within the same genera studied showed identical pollen nutrition reservoir types.

Pollen : ovule ratios
The pollen grain numbers, ovule numbers, and P/O ratios for the species investigated are given in Table 2. The mean P/O ratios in Zingiberaceae investigated range from 3·25 ± 1·56 in Caulokaempferia coenobialis to 616·52 ± 117·83 in the anaflexistylous flower of Alpinia stachyodes. The maximal mean pollen grain number is 51270 ± 1197·51 in the anaflexistylous flower of Alpinia hainanensis; and the minimal pollen grain number per flower, 113·75 ± 41·63, is found in Caulokaempferia coenobialis. Ovule numbers ranged from 7·1 ± 1·3 in cataflexistylous flower of Alpinia stachyodes to 182·22 ± 28·36 in the cataflexistylous flower of Alpinia hainanensis. The data show that pollen grain numbers, ovule numbers and P/O ratios from different species in Zingiberaceae even in a same genus are greatly varied (Table 2). But the mean pollen grain numbers, ovule numbers and P/O ratios of genera from the same tribe are not significantly different (Table 3). The degree of intraspecific differences in the P/O ratios from different populations varies greatly in different species: the pollen grain numbers, ovule numbers and P/O ratios of Caulokaempferia coenobialis from different populations are very different, but the pollen grain numbers and P/O ratios of Amomum villosum from different populations are approximately equivalent, although their ovule numbers are different (Table 4).


View this table:
[in this window]
[in a new window]
  		TABLE 2. The pollen grain numbers, ovule numbers, and P/O ratios for each of the species investigated (mean ± s.d.)

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Comparison of P/O ratios from taxonomic groups (mean ± s.d.)

 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Comparison of P/O ratios from different populations in the same species (mean ± s.d.)

 
The species investigated were classified as autogamous, facultatively xenogamous and obligately xenogamous based on the observations given here and Cruden's outcrossing index (Cruden, 1977Go), which is the sum of assigned values for three characteristics of the flower and floral behaviour: diameter of the flower; temporal separation of anther dehiscence and stigma receptivity; and spatial relationship. According to our data, the pollen grain numbers, ovule numbers, and P/O ratios in species with the same breeding system are very different (Table 2). The mean P/O ratios of species with obligate xenogamy, facultative xenogamy and autogamy are 345·39 ± 125·07, 108·07 ± 62·93 and 58·70 ± 91·78, respectively (Table 5). Caulokaempferia coenobialis is classified as actively autogamous based on its mucilaginous pollen moving onto the stigma and carrying out selfing (Y.Q. Wang, D.-X. Zhang and Z.-Y. Chen, unpubl. res.) and Pyrgophyllum yunnanense as passively autogamous based on the fact that its anthers and stigmas have a very close position which leads to selfing (Y.-Q. Wang, D.X. Zhang and Z.-Y. Chen, unpubl. res.). Their P/O ratios are 5·73 ± 3·50 and 164·63 ± 36·15, respectively.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Comparison of breeding-system P/O ratios (mean ± sd)

 


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Starch in pollen
Franchi et al. (1996)Go proposed that a positive starch reaction was categorized by three colour reactions, brown, blue and black, together with the presence or absence of birefringence. In our studies, starch reaction with IKI shows two colours (blue and black), and an unwanted phenomenon occurred with most or all pollen gains of Globbeae racemosa, Zingiber striolatum, and a few pollen grains of Hedychium coronarium, H. flavum and H. spicatum stained red. Previous studies revealed that starch reserves stored during pollen development give rise to carbohydrates at maturity and combinations of different types of carbohydrates in mature pollen may depend on the extent of starch hydrolysis (Speranza et al., 1997Go). After total or partial starch hydrolysis, insoluble periodic acid–Schiff reaction (PAS)-positive oligo/polysaccharides were found in the cytoplasm associated with much soluble sugar, and the pollen grains were dehydrated at dispersal (Speranza et al., 1997Go). The red colour in the pollen may suggest that they contain some protein or pollen starch hydrolysis.

Baker and Baker's (1979)Go studies on developing pollen grains showed that Fuchsia pollen, passes through a starchless phase, followed by one of starch accumulation, followed by its reduction to virtual absence by the time the anthers dehisce. They also mentioned that, generally, in the Rosaceae there is little difficulty in classifying pollen as starchy or starchless. In cases where starch had a very strong presence in immature grains but not disappear entirely from mature grains it was decided that the grains should be classified as ‘starchless’ (Baker and Baker, 1979Go). In Zingiberaceae, three species of Hedychium (H. coccineum, H. flavum and H. coronarium) exhibited phenomena similar to those they described, although the pollen of the other two species (H. spicatum and H. spicatum var. acuminatum) was typically starchy.

Previous studies indicated that members of a particular family are usually uniform regarding the possession of starchy or starchless pollen (Baker and Baker, 1979Go). Zona (2001)Go revealed that the order Zingiberales is the most uniform order as regards starchy pollen within the Commelinoid monocots, and all of the taxa in the Zingiberaceae he studied (19 species from ten genera) had starchy pollen, stained either brown or black, with two species of Alpinia having birefringent starchy pollen. The present results, however, are only in partial agreement with theirs. Of all the species in this study, four (Amomum villosum, Globba racemosa, Hedychium coronarium and H. flavum) have entirely starchless pollen, while all other species have pollen at least some reserves of starch. Also, no lipid was found in the pollen of one species (Stahlianthus involucratus), a result which contradicts Baker and Baker (1979)Go, who considered that all pollen grains contain some lipid.

In rare instances, there may be two kinds of pollen produced by the same flower (Baker and Baker, 1979Go). Mangelsdorf (1932)Go suggested that within a single anther there may be segregation of alleles for starch production when the sporophyte itself is heterozygous for a mutation; also there may be a proportion of ‘bad’ pollen admixed with the ‘good’ pollen in the anther as the result of some upset in the meiotic process, and the ‘bad’ pollen may give no starch reaction while the ‘good’ pollen gives a positive reaction. Among the species examined, Zingiber zerumbet shows two types of pollen (large pollen and small pollen), with the much fewer smaller ones showing a starch reaction, although both types apparently contain lipids. In Curcuma zanthorrhiza, C. wenyujin and C. phaeocaulis, roughly half of the pollen grains are starchy and the other half starchless.

Among 37 species studied, two (Caulokaempferia coenobialis and Pyrgophyllum yunnanense) were regularly self-pollinated, 15 were exclusively cross-pollinated and 20 were facultatively xenogamous. The two autogamous, all the 15 obligate xenogamous and 14 of 20 facultatively xenogamous species have starchy pollen, while all the lipid pollen comes from the facultatively xenogamous species. The results shown here indicate that there is no correlation between mating system and pollen nutrition reserve type, although previously Baker and Baker (1979)Go commented that the pollen of the selfers are more frequently starchy than those of outcrossers.

Almost all species studied (only Caulokaempferia coenobialis and Pyrgophyllum yunnanense being self-pollinated) were entomophilous, with most of the pollinators being bees and, in a few instances, butterflies or moths. Baker and Baker (1979)Go elucidated that, when pollen grains are part or all of the reward received by bees and flies visiting flowers, selection favoured starchless (oil-rich) pollen and, when insect nutrition is not involved, it favoured starchy pollen. Insects visiting flowers of Zingiberaceae in this study mostly forage on nectar and only accidentally on pollen grains (bees), with the exception of Hedychium species, Amomum villosum and Globba racemosa to which the pollinators came for both nectar and pollen grains. As almost all species of Zingiberaceae in the present study produced a large amount of nectar, according to Baker and Baker's hypothesis, species in this family should mostly produce starchy pollen grains. Our results are in full accordance with this hypothesis.

Grayum (1985)Go reported that in Araceae, 91 % of all spinose pollen is starchy, 83 % of all psilate genera are exclusively starchy and the figures for verrucate and striate pollen types are comparable, but genera in the most primitive, foveolate-reticulate category are equally likely (10/10) to be starchy or starchless. In the studies carried out by the authors, 89·5 % of all spinose pollen is starchy (17 of 19 species), 50 % of all psilate pollen is starchy (six of 12 species, with the exception that three produce starchless pollen and the other three produce a mixture of starch and starchless). All the verrucate and striate pollen types (100 %) are starchy. This result is does not agree with what Grayum (1985)Go suggested as the correlation between exine sculpture and pollen nutrition type.

Baker and Baker (1979)Go also suggested that in many cases, pollen grain size followed family classification. In Zingiberaceae, the pollen grain sizes are highly correlated with generic classification (Table 1). Generally, the mean pollen grain diameter of the species with starchy pollen is significantly greater than that of the species with starchless pollen (Grayum, 1985Go). The present study, however, shows that there is no significant difference in size between starchy and starchless pollens in Zingiberaceae, although there are more variations in starchless pollens (see Table 6).


View this table:
[in this window]
[in a new window]
  		TABLE 6. Comparison of P/O ratios and pollen size (diameters) in species of Zingiberaceae with starchy and starchless pollen

 
Pollen : ovule ratios
According to Cruden (1977)Go, the P/O ratios range from 2·7 to 6·7 in cleistogamous flowers, from 18·1 to 39·0 in obligate autogamous flowers, from 31·9 to 396·9 in facultative autogamous flowers, from 244·7 to 2558·6 in facultative xenogamous flowers, and from 2108 to 19523 in xenogamous flowers. The present data show that P/O ratios of xenogamous species in the family are above 184·00 (mean 345·39 ± 125·07), while those of most facultatively xenogamous species (with the exception of Amomum villosum) are less than 168 (mean 108·07 ± 62·93). The P/O ratios of autogamous species range from 5·73 ± 3·50 to 164·63 ± 36·15. These data are obviously lower than the numbers in the respective breeding systems summarized by Cruden (1977)Go. Previous studies have shown that species with special pollen-transporting mechanisms (pollinia, viscin threads, polyads) have much lower P/O ratios than species lacking these mechanisms (Cruden, 1977Go; Cruden and Jensen, 1979Go; Mehrhoff, 1983Go; Kopture, 1984Go; Vasek and Weng, 1988Go; Wyatt et al., 2000Go). The low P/O ratios may be due to very efficient pollinating mechanisms ensuring that sufficient numbers of pollen grains are deposited on the stigmas (Cruden, 2000Go; Jürgens and Gottsberger, 2002Go). The lower P/O ratios in the family may be another indication of its much specialized, sophisticated pollination mechanisms. Cruden and Miller-Ward (1981)Go postulated that species with relatively larger stigma areas compared with the pollen-bearing area of the pollinator often have lower P/O ratios; the low P/O ratios in Zingiberaceae could also be explained by the large stigma areas.

Dimorphic flowers (cataflexistylous vs. anaflexistylous) have been described recently in Amomum and Alpinia (Cui et al., 1996Go; Li et al., 2001Go). The present observations on the P/O ratios of the two phenotypes in Alpinia showed that in most taxa in the genus (except A. zerumbe), the anaflexistylous flowers have larger pollen numbers and higher P/O ratios than the cataflexistylous flowers; the ovule numbers in the anaflexistylous flowers, however, are approximately equivalent to, or less than those in the cataflexistylous flowers (except in A. guinanensis) (see Table 2), a result in contradict with Li (2002)Go and Zhang (2003)Go who also measured the P/O ratios of the two phenotypes and further speculated that it may indicate that flexistyly be an intermediate from hermaphrodite to dioecy.

While studying the P/O ratios in Caryophyllideae, Jürgens and Gottsberger (2002)Go found significant differences between some taxonomic groups in pollen grain numbers and ovule numbers but not in P/O ratios. The same phenomenon is found in Zingiberaceae (see Table 3). It was also noticed that there is a correlation between the breeding system and life-form types in the family, although with significant differences in mean pollen grain numbers and P/O ratios, approximately constant ovule numbers were recorded in species with different life forms (see Table 7). The result is consistent with that of Caryophylloideae as reported by Jürgens and Gottsberger (2002)Go. All of the two autogamous and most facultative xenogamous species are deciduous perennials. Most of the obligated xenogamous species (the only exception being Roscoea blanda) are evergreen perennials. The pollen number and the P/O ratios were obviously higher in evergreen perennials than in deciduous perennials (see Table 7).


View this table:
[in this window]
[in a new window]
  		TABLE 7. Comparison of P/O ratios from different life forms

 


   ACKNOWLEDGEMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
LITERATURE CITED
 
We are indebted to Dr S. Zona and an anonymous reviewer for critically reading the manuscript, and Dr Qing-jun Li and Dr Da Luo for valuable discussion. This project is supported by the National Key Program for Basic Research of China (2001CCA00300) and the Chinese Academy of Sciences Knowledge Innovation Program (South China Botanical Garden Director's Fund).


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

    Baker HG, Baker I. 1979. Starch in angiosperm pollen grains and its evolutionary significance. American Journal of Botany 66: 591–600.[CrossRef]

    Charnov EL. 1982. The theory of sex allocation. Princeton: Princeton University Press.

    Cruden RW. 1977. Pollen-ovule ratios: a conservative indicator of breeding systems in flowering plants. Evolution 31: 32–46.

    Cruden RW. 2000. Pollen grains: why so many? Plant Systematics and Evolution 222: 143–165.[CrossRef]

    Cruden RW, Jensen KG. 1979. Viscin threads, pollination efficiency and low pollen-ovule ratios. American Journal of Botany 66: 875–879.[CrossRef]

    Cruden RW, Miller-Ward S. 1981. Pollen-ovule ratio, pollen size, and the ratio of stigmatic area to the pollen-bearing area of the pollinator: a hypothesis. Evolution 35: 964–974.[CrossRef]

    Cui X-L, Wei R-C, Huang R-F. 1996. A study on the breeding system of Amomum tsaoko. In: Wu T-L, Wu Q-G, Chen Z-Y, eds. Proceedings of the second symposium on the family Zingiberaceae. Guangzhou: Zhongshan University Press.

    Dafni A. 1992. Pollination ecology, a practical approach. New York: Oxford University Press.

    Darwin C. 1877. The different forms of flowers on plants of the same species. London: John Murray.

    Franchi GG, Bellani L, Nepi M, Pacini E. 1996. Types of carbohydrate reserves in pollen: localization, systematic distribution and ecophysiological significance. Flora 191: 143–159.

    Grayum MH. 1985. Evolutionary and ecological significance of starch storage in pollen of the Aracea. American Journal of Botany 72: 1565–1577.[CrossRef]

    Jürgens A, Gottsberger TWG. 2002. Pollen grain numbers, ovule numbers and pollen-ovule ratios in Caryophylloideae: correlation with breeding system, pollination, life form, style number, and sexual system. Sexual Plant Reproduction 14: 279–289.[CrossRef]

    Kopture S. 1984. Outcrossing and pollinator limitation of fruit set: breeding systems of neotropical Inga trees. Evolution 38: 1130–1143.[CrossRef]

    Li Q-J, Xu Z-F, Xia Y-M, Zhang L, Deng X-B, Gao J-Y. 2001. Study on the flexistyly pollination mechanism in Alpinia plants (Zingiberaceae). Acta Botanica Siica 43: 364–369 [in Chinese].

    Li Q-J. 2002. Study on the flexistyle outcrossing mechanism in Alpinia plants (Zingiberaceae). PhD Thesis, Kunming Institute of Botany, Chinese Academy of Sciences [in Chinese].

    Liang Y-H. 1988a. Pollen morphology of Alpinia (Zingiberaceae) from China. Acta Botanica Austro Sinica 4: 103–107 [in Chinese].

    Liang Y-H. 1988b. Pollen morphology of the family Zingiberaceae in China—pollen types and their significance in the taxonomy. Acta Phytotaxonomica Sinica 26: 265–281 [in Chinese].

    Mangelsdorf PC. 1932. Mechanical separation of gametes in maize. Journal of Heredity 23: 288–295.

    Mehrhoff LA. 1983. Pollination in the genus Isotria (Orchidaceae). American Journal of Botany 70: 1444–1453[CrossRef]

    Preston RE. 1986. Pollen-ovule ratios in the Cruciferae. American Journal Botany 73: 1732–1740.[CrossRef]

    Sakai S, Kato M, Inoue T. 1999. Three pollination guilds and variation in floral characteristics of Bornean gingers (Zingiberaceae and Costaceae). American Journal of Botany 86: 646–658.[Abstract/Free Full Text]

    Speranza A, Calzoni GL, Pacini E. 1997. Occurrence of mono- or disaccharides and polysaccharide reserves in mature pollen grains. Sexual Plant Reproduction 10: 110–115.[CrossRef]

    Stebbins GL. 1970. Adaptive radiation in angiosperms. I. Pollination mechanisms. Annual Review of Ecology and Systematics 1: 307–326.

    Vasek FC, Weng V. 1988. Breeding systems of Clarkia sect. Phaeostoma (Onagraceae). I. Pollen-ovule ratios. Systematic Botany 13: 336–350.[CrossRef]

    Wu T-L. 1994. Phytogeography of the Zingiberaceae. Journal of Tropical and Subtropical Botany 2: 1–14 [in Chinese].

    Wyatt R. 1984. Evolution of self-pollination in granite outcrop species of Arenaria (Caryophyllaceae). III. Reproductive effort and pollen-ovule ratios. Systematic Botany 9: 432–440.[CrossRef]

    Wyatt R, Steven BB, Sara RL. 2000. Pollen-ovule ratios in milkweeds (Asclepiadaceae): an exception that proves the rule. Systematic Botany 25: 171–180.

    Zhang L, Li Q-J, Deng X-B, Ren P-Y, Gao J-Y. 2003. Reproductive biology of Alpinia blepharocalyx (Zingiberaceae): another example of flexistyly. Plant Systematics and Evolution 241: 67–76.[CrossRef]

    Zona S. 2001. Starchy pollen in commelinoid monocots. Annals of Botany 87: 109–116.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
94/4/583    most recent
mch177v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Agricola
Right arrow Articles by WANG, Y.-Q.
Right arrow Articles by CHEN, Z.-Y.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?