Annals of Botany

AOBPreview originally published online on November 16, 2004
Annals of Botany 2005 95(2):295-304; doi:10.1093/aob/mci025

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 95/2/295    most recent mci025v1 E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (6) Request Permissions Google Scholar Articles by TORIMARU, T. Articles by TOMARU, N. Search for Related Content PubMed PubMed Citation Articles by TORIMARU, T. Articles by TOMARU, N. Agricola Articles by TORIMARU, T. Articles by TOMARU, N. Social Bookmarking          What's this?

Annals of Botany 95/2 © Annals of Botany Company 2004; all rights reserved

Fine-scale Clonal Structure and Diversity within Patches of a Clone-forming Dioecious Shrub, Ilex leucoclada (Aquifoliaceae)

TAKESHI TORIMARU and NOBUHIRO TOMARU^*

Laboratory of Forest Ecology and Physiology, Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya 464-8601, Japan

^* For correspondence. E-mail tomaru{at}agr.nagoya-u.ac.jp

Received: 15 March 2004    Returned for revision: 18 August 2004    Accepted: 14 September 2004    Published electronically: 16 November 2004

	ABSTRACT

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

 

• Background and Aims The mode of reproduction (sexual vs. asexual) is likely to have important effects on genetic variation and its spatial distribution within plant populations. An investigation was undertaken of fine-scale clonal structure and diversity within patches of Ilex leucoclada (a clone-forming dioecious shrub).

• Methods Six patches were selected in a 1-ha plot previously established in an old-growth beech forest. Two of the selected patches were composed predominantly of stems with male flowers (male patch), and two contained stems with predominantly female flowers (female patch). The remaining two patches contained stems with male flowers and stems with female flowers in more or less equal proportions (mixed patch). Different genets were distinguished using random amplified polymorphic DNA (RAPD) markers.

• Key Results One hundred and fifty-six genets with different RAPD phenotypes were identified among 1928 stems from the six patches. Among the six patches, the male patches had the lowest clonal diversity, and the mixed patches had the highest. Distribution maps of the genets showed that they extended downhill, reflecting natural layering that occurred when stems were pressed to the ground by heavy snow. In every patch, there were a few large genets with many stems and many small genets with a few stems.

• Conclusion The differences in clonal diversity among patches may be due to differences in seedling recruitment frequencies. The skewed distribution of genet size (defined as the number of stems per genet) within patches may be due to differences in the timing of germination, or age (with early-establishing genets having clear advantages for acquiring resources) and/or intraspecific competition.

Key words: Ilex leucoclada, dioecy, patch, genet, ramet, RAPD, clonal structure, clonal diversity, layering, clonal growth, seedling recruitment

	INTRODUCTION

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

 
A variety of life-history traits, such as life form, breeding system and seed dispersal mechanism, influence genetic variation and its spatial distribution within populations of plant species (Heywood, 1991), causing the genetic architecture within populations to change through time (Gillespie, 1998). The mode of reproduction (sexual vs. asexual) is likely to have important effects on genetic variation and its spatial distribution within plant populations (Harada and Iwasa, 1996; Winkler and Stöcklin, 2002), since sexual reproduction is accompanied by genetic recombination and asexual reproduction is not. Genetic recombination leads to the continuous emergence of new genotypes and thus can buffer the loss of genotypic diversity. On the other hand, vegetative reproduction or clonal growth processes, such as the production of subterranean stems, stolons, propagules and (in woody plant species) sprouts and root suckers, leads to spatial clumping of ramets of identical genets within populations. However, spatial patchiness can be caused by many factors other than asexual reproduction, e.g. restricted seed dispersal, microhabitat variation and gap recruitment (Berg and Hamrick, 1994). The spatial arrangement of genets can affect mating patterns (Handel, 1985). Clusters of genetically identical plants, for example, limit the number of mates available to self-incompatible species with limited pollen dispersal (Eckert and Barrett, 1993). Therefore, knowledge of fine-scale clonal structure within populations is essential for understanding micro-evolutionary processes in clonal plant species.

Many perennial plants combine sexual reproduction through seeds with reproduction through vegetative propagation (Richards, 1986). However, the relative proportions of sexual vs. asexual progeny produced and recruited may often vary widely within a species, due to variations in ecological and/or genetic factors that limit or enhance one or other reproductive mode (Eckert, 2002). This, in turn, may directly affect the genotypic diversity within natural populations. Since vegetative reproduction yields offspring that are genetically identical to both the maternal plant and each other, the resultant patches are expected to exhibit no genotypic diversity. By contrast, frequent seedling recruitment within patches promotes high genotypic diversity.

Molecular markers are the most promising tools for genet identification. Random amplified polymorphic DNA (RAPD) is a PCR-based marker method that increases the number of markers without limit (Williams et al., 1990). Therefore, many studies have used them to study clonal structure and diversity in plant species (Hsaio and Rieseberg, 1994; Sydes and Peakall, 1998; Tani et al., 1998; Esselman et al., 1999; Kreher et al., 2000; Moriguchi et al., 2001; Persson and Gustavsso, 2001).

Ilex leucoclada M. (Aquifoliaceae) is an evergreen broad-leaved dioecious shrub, reaching up to 2 m in height. It is distributed in mountainous regions, where there is heavy snowfall, in Honshu and the southern part of Hokkaido, Japan. This species is a common shrub of the deciduous broad-leaved forests in the cool temperate zone along the Sea of Japan, which are dominated by Japanese beech, Fagus crenata (Fujita, 1987). Avian seed dispersal and insect-mediated pollination are typical of the Aquifoliaceae, including I. leucoclada (Watanabe, 1994). Previous field observations of I. leucoclada by the authors indicate that this species forms distinct patches and reproduces both sexually and asexually by layering when stems are pressed to the ground by heavy snow. Substantial variation in the sex ratios of ramets (stems) within the patches has also been observed. Some patches are unisexual, consisting solely of stems with male flowers or stems with female flowers, while other patches contain both stems with male flowers and stems with female flowers. Furthermore, in a previous study high clonal diversity within patches and high genetic differentiation among patches, with no hierarchical genetic structure, had been detected (Torimaru et al., 2003). Therefore, the spatial (patch formation) and reproductive (dioecy) characteristics of I. leucoclada make the species an attractive subject for studying clonal structure and diversity in relation to sexes in clone-forming plants.

In this paper, the fine-scale clonal structure within patches of I. leucoclada is illustrated, and their clonal diversity compared, by distinguishing genets using random amplified polymorphic DNA (RAPD) markers. Ecological processes promoting clonal structure and diversity within patches of the species are then discussed.

	MATERIALS AND METHODS

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

 
Study site and field methods
The study site was part of the Forest Reserve on Mt Daisen in the north-central Chugoku Mountains in south-western Japan. Beech forests occur between 650 and 1350 m a.s.l. in the reserve, but are now rare below 800 m because of past human land use. Old-growth beech stands occur between 800 and 1200 m. Dwarf bamboos, such as Sasa kurilensis and S. palmate, dominate the understorey of some stands, although they often have a patchy distribution.

A 4-ha permanent plot (200 x 200 m) was established at about 1100 m a.s.l. on the south-east-facing slope in the reserve during 1987 and 1988, and tree censuses were performed in 1990, 1992, 1997 and 2002 for all stems with a diameter at breast height (dbh) 4 cm. In 1988, there were 3749 living stems in the plot; the dominant species in terms of stem number included Fagus crenata (200 stems ha^–1), Acer japonicum (168 stems ha^–1) and Acanthopanax sciadophylloides (88 stems ha^–1) (Yamamoto et al., 1995). The community structure of this old-growth beech forest is heavily influenced by natural disturbances (the occurrences of canopy gaps) and the heterogeneity of the forest floor (differences in soil conditions and density of dwarf bamboo plants) (Yamamoto et al., 1995).

A 1-ha subplot (200 x 50 m) was established in the upper quarter of the permanent plot in 2000 to estimate clonal diversity and genetic variation of I. leucoclada (Torimaru et al., 2003). At that time, since each cluster of I. leucoclada stems had a distinct edge, patches could easily be identified, and the length, width and spatial coordinates (x, y) of the centre of each patch determined (Torimaru et al., 2003).

To investigate fine-scale clonal structure within the I. leucoclada patches, six patches within the plot were selected. A patch dominated by stems with male flowers was defined as a male patch. Similarly, a patch dominated by stems with female flowers was defined as a female patch, and a patch including both stems with male flowers and stems with female flowers was defined as a mixed patch. Two male patches and one mixed patch were selected in 2001 (hereafter referred to as ‘male 1’, ‘male 2’ and ‘mixed 1’, respectively) and an additional three patches, two female and one mixed, in 2002 (designated ‘female 1’, ‘female 2’ and ‘mixed 2’, respectively). These patches were located under the closed canopy, on mature soil without the Sasa cover. Each of the patches was divided into 1 x 1 m contiguous quadrats, and the inclination of the surface in each patch was measured at each corner of the quadrats by level measurement. The spatial coordinates (x, y), stem length, absence/presence of flowers, and sex of each stem were recorded. Leaves were collected from all stems in the patches for RAPD analysis, and were stored at –30 °C until DNA was extracted.

DNA extraction and RAPD analysis
The techniques used for DNA extraction from the leaves and RAPD amplification by polymerase chain reaction (PCR) are described in detail by Torimaru et al. (2003). Eighteen primers (derived from Operon 10-mer Kit, Operon Technologies, Inc.) that were found to yield reproducible and unambiguous polymorphic fragments by screening were used in this study (Table 1). The presence or absence of RAPD fragments (bands) was scored and used for subsequent data analysis.

View this table: [in this window] [in a new window]   TABLE 1. Eighteen RAPD primers that yielded reproducible and unambiguous polymorphic fragments in the six patches of Ilex leucoclada

 
Data analysis
Determination of genets by RAPD phenotype. Single or multiple stems with an identical RAPD phenotype within a patch were considered members of the same genet. To assess the validity of distinguishing between different genets using RAPD phenotypes, the probability was calculated of observing at least n stems of each RAPD phenotype by chance in a sample of N stems within patches, if all stems resulted from independent recombination [modified from Park and Werth (1993) for dominant markers], using a binominal probability function:

where

and p_i = the estimated frequency of band presence or absence at loci i among genets in the population. We estimated the frequencies of band presence or absence by RAPD analysis using the same samples as used by Torimaru et al. (2003). L = the number of loci used within patches.

Estimation and test for clonal diversity. Three frequently used measures of clonal diversity for each patch were calculated. The first was the number of genets (G) relative to the number of stems (N) sampled (Ellstrand and Roose, 1987) [note that G/N is a biased estimator influenced by the number of stems (N) but frequently used]. The second was Simpson's diversity index (D) (Pielou, 1969), which represents the probability that two stems selected at random from a patch of N stems are from different RAPD phenotypes (genets):

where n_i = the number of individual stems with RAPD phenotype i. This equation yields a measure of clonal diversity ranging from 0 to 1, with 1 being the maximum possible diversity. Simpson's D is influenced by the numbers of stems and RAPD phenotypes. To compare directly clonal diversity among patches with different numbers of stems and RAPD phenotypes, the evenness indicator (E) of Fager (1972) was calculated as the third measure of clonal diversity:

where

and

Fager's E ranges from zero (if all stems possess the same phenotype) to 1·0 (if all RAPD phenotypes are equally represented among stems).

These three measures of clonal diversity are sensitive to deviations from random association of alleles within loci (i.e. from Hardy–Weinberg equilibrium) and between loci (i.e. from linkage equilibrium). Thus, clonal diversity in a sample can be compared with that expected under random mating to provide evidence of processes such as asexual reproduction (Hoffmann, 1986). To estimate expected values of the clonal diversity measures, Monte Carlo simulations (Hoffmann, 1986; McFadden, 1997; Ceplitis, 2001) were conducted briefly as follows. For each patch, a number—equal to the sample size—of RAPD phenotypes was generated by drawing markers (band presence or absence) at random from the distribution of observed marker frequencies for the population. With this procedure, Hardy–Weinberg proportions of RAPD genotypes are not required as long as populations are in equilibrium. The process was repeated 1000 times and each observed value of diversity measures was compared with the 5 % confidence limits of the simulated distribution. If observed values of the diversity measures were found to be lower than the lower specified rejection limit, the hypothesis of free recombination was rejected.

Spatial association among stems of genets. To assess the spatial association among stems of genets in each patch, the average distance between stems of the same genet and the probability of the nearest neighbour being of the same genet (P_IN) were calculated. P_IN indicates the extent to which the stems of the same genet intermingle spatially with those of other genets, and the extent of intermingling increases with decreasing the P_IN value.

	RESULTS

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

 
Determination of genets by RAPD phenotypes
Over the six patches, 2013 stems were found in total, averaging 335·5 ± 188·0 (s.d.) per patch (range = 178–671) (Table 2). Polymorphic loci, ranging from six to ten, were used to distinguish different genets in each patch, and 156 different RAPD phenotypes were determined among 1928 analysed stems, averaging 26·0 ± 12·0 (s.d.) per patch (range = 15–46). The number of loci used was not correlated with the number of RAPD phenotypes detected (Spearman's coefficient of rank correlation r_s = 0·147, P > 0·05), indicating that the differences in the number of loci used in each patch did not influence the number of RAPD phenotypes detected.

View this table: [in this window] [in a new window]   TABLE 2. The number of male, female and sexually undetermined stems of the six patches of Ilex leucoclada

 
Eighty-four of the 156 RAPD banding phenotypes were detected in two or more stems, and each of the other 72 phenotypes appeared only in a single stem (Appendix). For the 84 phenotypes with multiple stems, the validity of grouping them as different genets was assessed by calculating P_sex values. Fifty-three (63·1 %) had P_sex values <0·05, and the remaining 31 (36·9 %) had values >0·05. There were no bands that were exclusively associated with one sex.

Clonal diversity within patches
Varying numbers of genets were distinguished in the six patches (Table 3). Observed values of clonal diversity measures varied among patches. Among the six patches, G/N, Simpson's D and Fager's E values ranged from 0·005 to 0·245, 0·193 to 0·937 and 0·149 to 0·915, respectively (Table 4). There were no trends in terms of either the numbers of genets detected or G/N ratios among male, female and mixed patches after taking account of differences in the numbers of stems analysed among patches. Although there was some caution about the results not being statistically tested due to the relatively low number of patches sampled, female and mixed patches tended to be more diverse than male patches in terms of Simpson's D and Fager's E values (Table 4).

View this table: [in this window] [in a new window]   TABLE 3. The number of male, female and sexually undetermined gentes of the six patches of Ilex leucoclada

 

View this table: [in this window] [in a new window]   TABLE 4. Clonal diversity measures for the six Ilex leucoclada patches

 
Expected values of clonal diversity measures generated by simulation showed that all patches had significantly lower clonal diversity than would be expected under free recombination (Table 4).

Fine-scale clonal structure within patches
Assuming each distinct banding phenotype corresponds to a genet, the average genet, amongst the 156 detected, consisted of 12·6 ± 3·3 (s.e.) stems (max. = 353) (Appendix). In every patch, there were a few large genets and a large number of small genets (Fig. 1 and Appendix). The sex ratios (male/female) of the genets were determined and found them to be 7 : 1 and 5·99 : 1 in male 1 and male 2, respectively. The female genets within male patches were all small. In the two female patches, there were no male genets. In mixed patches 1 and 2, the sex ratios of the genets were 1·16 : 1 and 0·33 : 1, respectively (Table 3). Male, female and sexually undetermined genets consisted, on average, of 22·4 ± 11·4 (s.e.), 22·7 ± 7·3 and 1·9 ± 0·2 stems, respectively. Sexually undetermined genets were all small, and accounted for 39 to 61 % of the genets within each patch.

View larger version (15K): [in this window] [in a new window]   FIG. 1. Frequency distribution of the number of stems per genet. Black bars, male genets; white bars, female genets; shaded bars, sexually undetermined genets.

 
Distribution maps of genets and stems within patches showed that genets extended downhill (Fig. 2). Male patches showed high P_IN values (Table 5), indicating the nearest neighbour being most likely of the same genet. This was also confirmed visually by the maps that genets tended to form distinct clusters of stems and did not intermingle with each other (Fig. 2). By contrast, female and mixed patches showed moderate or low P_IN values, indicating that genets intermingled, to varying extents, with each other (Table 5 and Fig. 2).

View larger version (38K): [in this window] [in a new window]   FIG. 2. Spatial distribution of genets and stems in male, female and mixed patches (male 2, female 1 and mixed 1, respectively) of Ilex leucoclada. Different large genets are represented by different symbols (circles, triangles, squares and diamonds). Filled and open symbols represent male and female genets, respectively. Plus signs represent small genets and crosses represent stems for which RAPD phenotypes could not be determined due to the low quality of the templete DNA.

 

View this table: [in this window] [in a new window]   TABLE 5. The number of multi-stemmed genets, average distances between stems of the same genet, and PIN for the six patches of Ilex leucoclada

 
The maximum distance between stems within genets was significantly correlated with the number of stems per genet (r_s = 0·837, P < 0·001). Furthermore, the maximum distance between stems of the largest genet within patches was significantly correlated with both the total number of stems within patches (r_s = 0·829, P < 0·05) and the patch area (r_s = 0·943, P < 0·01). However, neither the stem density nor the number of genets detected within patches was associated with the patch area.

	DISCUSSION

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

 
Determination of genets and estimation of clonal diversity
Single stems with unique RAPD banding phenotypes were considered distinct genets. For multiple stems with identical phenotypes, P_sex was calculated, which means the probability of observing at least n stems of each RAPD phenotype by chance in a sample of N stems within patches, if all stems resulted from independent recombination. All sets of multiple stems with identical phenotypes that had P_sex values lower than 0·05 were also considered distinct genets. However, sets of multiple stems that had identical phenotypes with P_sex values exceeding 0·05 were also detected. Such phenotypes may represent more than one genet. Therefore, the true number of genets in the six patches may be >156, and the values of the clonal diversity measures may have been underestimated. The number of such phenotypes was higher in the female 2 patch than in any other patch, suggesting that the clonal diversity estimates were the most underestimated for this patch. However, relative frequencies of stems with such phenotypes within patches were very low (<0·05). According to Peet (1974), Simpson's D is much more strongly influenced by phenotypes with large stem numbers than by phenotypes with small stem numbers. This attribute of Simpson's D was also true for Fager's E because E is derived from D. Thus, even if some banding phenotypes with small stem numbers actually represented more than one genet, the resulting changes in the values of D and E would be minor. Therefore, our estimates of clonal diversity would be valid for comparing patches.

Clonal diversity within patches
Clonal diversity within patches was found to be substantial for I. leucoclada; the mixed patches showed high, female patches moderate or high, and male patches moderate or low levels of clonal diversity, compared with figures presented in reviews of clonal diversity among >20 clonal plant species by Ellstrand and Roose (1987) (G/N = 0·17, Simpson's D = 0·62, and Fager's E = 0·68, on average) and Hangelbroke et al. (2002) (G/N = 0·44, Simpson's D = 0·74). As discussed by Torimaru et al. (2003), the abundant clonal diversity within I. leucoclada patches may be explained by multiple founders, seedling recruitment during patch formation, and/or somatic mutation. There have been similar reports of high levels of fine-scale clonal diversity in various clonal woodland herbs, e.g. Anemone nemorosa (Holderegger et al., 1998; Stehlik and Holderegger, 2000), Uvularia perfoliata (Kudoh et al., 1999) and Viola riviniana (Auge et al., 2001).

However, the Monte Carlo simulations indicated that, although the patches harboured substantial clonal diversity, there were significant reductions in clonal diversity from expectations assuming free recombination. Genetic differentiation among patches (Torimaru et al., 2003) may partly explain the deviation from the expected values of clonal diversity measures. However, the significant reductions in clonal diversity obviously reflected substantial effects of clonal growth within the patches.

The levels of clonal diversity within patches were higher in the female and mixed patches than in the male patches of I. leucoclada. As discussed by Torimaru et al. (2003), the input of genets into patches through seed dispersal most likely accounts for the high level of clonal diversity observed in this I. leucoclada population. For plant species whose seeds are dispersed by frugivorous birds, the activities of the animals concerned may affect the survival and establishment of seeds and seedlings (Herrera et al., 1994; Jordano and Herrera, 1995). Therefore, birds' seed dispersal behaviour should be considered when discussing possible mechanisms that could have created the clonal diversity observed. Several studies on dioecious plant species with bird-dispersed seeds have shown that the birds' behaviour may generate sex-biased dispersal patterns (Herrera et al., 1994; Verdú and García-Fayos, 2003). The mechanism responsible for the sex-biased dispersal has been described in the cited studies as follows: only female plants produce fruits in dioecious plant species, so the birds concentrate their frugivorous activities on the female plants and tend to egest seeds under female plants more often than under male plants. In the present study, fruit production was observed within the female and mixed patches. The differences in clonal diversity between male patches and other patches (female and mixed patches) found in this study support the hypothesis that seed rain and subsequent seedling recruitment occur more frequently in patches with female plants than in patches with only male plants.

Fine-scale clonal structure within patches
In the present study, I. leucoclada genets extended downhill. Natural layering, which is one of the modes of clonal growth, occurs in this species as follows. Stems pressed to the ground by heavy snow produce adventitious roots in the humid litter layer. The roughly linear distribution of I. leucoclada stems of identical genets that run downhill is most likely due to such layering. In Pinus pumila, which also regenerates asexually by layering, Kajimoto (1992) demonstrated that stems leaned towards the downside of the slope, and trees moved down the slope by repetitive layering of the prostrate stems, also resulting in linear distributions of stems representing identical ramets of the same genets.

Clonal plant species display a continuum of growth forms, from ‘phalanx forms’, with a solid advancing front of ramets to ‘guerilla forms’, in which widely spaced ramets infiltrate the surrounding vegetation (Lovett Doust, 1981). In the present study, although most large genets formed distinct clusters of stems within male patches (e.g. male 1), there were overlaps among genets within patches, especially in the female 2 and mixed 2 patches, suggesting that I. leucoclada adopts a loose or moderate mode of the phalanx form. Once established in favourable sites, the phalanx form would be advantageous for local persistence (Herben and Hara, 1997). In a previous study (Torimaru et al., 2003), most of the 38 I. leucoclada patches found were located near the middle of the investigated plot on sites where there was mature soil and no Sasa cover (Yamamoto et al., 1995), indicating that sites on mature soil and without Sasa are favourable for I. leucoclada. Therefore, the phalanx form may be advantageous for persistence in such favourable sites for I. leucoclada.

The I. leucoclada genets consisted of many spatially extending stems, indicating that layering allowed them to occupy extensive areas. Layering may be a manifestation of phenotypic plasticity that facilitates the occupation of two-dimensional space and thus increases resource capture, analogously to the occupation of three-dimensional space achieved by many non-clonal plant species through branching and the proliferation of root and shoots in response to light and below-ground resources (Slade and Hutchings, 1987a, b; de Kroon and Hutchings, 1995).

The distribution of genet size (defined as the number of ramets per genet) was similar among the six I. leucoclada patches; each patch consisted of a few large genets with many stems and many small genets with a few stems, with a skewed distribution of genet size within patches. Moreover, the patch area was not associated with the number of genets within patches, but was correlated with the extension of the largest genet within patches. These results suggest that clonal growth contributed more than seedling recruitment to patch development in the studied population. Skewed distributions of genet size have also been found in Polygonum viviparum (Diggle et al., 1998), Quercus havardii (Mayes et al., 1998), and Uvularia perfoliata (Kudoh et al., 1999).

Differences in the timing of establishment among genets may account for the skewed distribution of genet size within the I. leucoclada patches. Genets that establish early would be more likely to acquire sufficient resources to expand than those that establish later (Ross and Harper, 1972; Firbank and Watkinson, 1987). Another possible reason for the skewed distribution of genet size within patches is intraspecific competition among the genets. Support for this hypothesis is provided by a study of Rhododendron ferrugineum, where a significant positive correlation was found between the annual shoot growth rate and genet area, indicating that successful genets were favoured during population closure (Pornon et al., 2000). Further demographic studies are necessary to characterize the level of intraspecific competition among genets of I. leucoclada.

	CONCLUSIONS

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

 
Using RAPD analysis, based on 18 polymorphic loci, multiple genets were identified within patches of I. leucoclada. Levels of clonal diversity differed among male, female and mixed patches. It was also found that the species had explicit clonal structure within patches with genets being of varying size. The genets grew clonally downhill, creating a linear distribution of ramets of the same genet. Although formation of the patches likely involves both clonal growth processes and seedling recruitment, the former probably contributes more strongly to patch enlargement, while the latter is more responsible for the differences in clonal diversity among patches and the skewed distribution of genet size within the patches.

	APPENDIX

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

 
The number of stems and P_sex for each RAPD phenotype in the six patches of Ilex leucoclada. P_sex is the probability of observing at least n stems of each RAPD phenotype by chance in a sample of N stems within patches, if all stems resulted from independent recombination. The proportions of stems in each patch are shown in parentheses.

Patch Phenotype no. No. of stems Psex Male 1 1 293 (0·89) <10–300 2 20 (0·06) 1·581 x 10–19 3 3 (0·01) 7·240 x 10–4 4 2 (0·01) 5·453 x 10–4 5 1 (<0·01) 0·217 6 1 (<0·01) 0·330 7 1 (<0·01) 0·144 8 1 (<0·01) 0·265 9 1 (<0·01) 0·113 10 1 (<0·01) 0·076 11 1 (<0·01) 0·369 12 1 (<0·01) 0·023 13 1 (<0·01) 0·581 14 1 (<0·01) 0·265 15 1 (<0·01) 0·113 Male 2 1 91 (0·28) 2·537 x 10–298 2 44 (0·13) 6·372 x 10–77 3 4 (0·01) 7·472 x 10–14 4 4 (0·01) 2·630 x 10–12 5 4 (0·01) 2·070 x 10–7 6 3 (0·01) 1·950 x 10–7 7 3 (0·01) 2·660 x 10–6 8 3 (0·01) 1·068 x 10–5 9 3 (0·01) 1·376 x 10–5 10 3 (0·01) 2·677 x 10–4 11 1 (<0·01) 0·011 12 1 (<0·01) 0·012 13 1 (<0·01) 0·332 14 1 (<0·01) 0·049 15 1 (<0·01) 0·050 16 1 (<0·01) 0·017 17 1 (<0·01) 0·019 18 1 (<0·01) 0·001 Female 1 1 89 (0·36) 7·337 x 10–147 2 59 (0·24) 2·780 x 10–66 3 37 (0·15) 3·234 x 10–23 4 21 (0·08) 9·556 x 10–22 5 8 (0·03) 0·003 6 7 (0·03) 2·754 x 10–5 7 6 (0·02) 0·378 8 5 (0·02) 0·985 9 5 (0·02) >0·999 10 3 (0·01) 0·334 11 2 (0·01) 0·058 12 2 (0·01) 0·677 13 1 (<0·01) 0·994 14 1 (<0·01) >0·999 15 1 (<0·01) 0·878 16 1 (<0·01) 0·989 Female 2 1 353 (0·57) <10–300 2 119 (0·19) 6·860 x 10–94 3 23 (0·04) 2·704 x 10–10 4 21 (0·03) 1·979 x 10–7 5 20 (0·03) 0·221 6 14 (0·02) 0·939 7 8 (0·01) >0·999 8 6 (0·01) 0·837 9 6 (0·01) >0·999 10 5 (0·01) >0·999 11 4 (0·01) 0·064 12 4 (0·01) >0·999 13 4 (0·01) >0·999 14 3 (<0·01) 0·985 15 3 (<0·01) 0·996 16 3 (<0·01) 0·998 17 2 (<0·01) 0·541 18 2 (<0·01) 0·812 19 2 (<0·01) 0·995 20 2 (<0·01) 0·998 21 2 (<0·01) >0·999 22 2 (<0·01) >0·999 23 1 (<0·01) >0·999 24 1 (<0·01) 0·999 25 1 (<0·01) >0·999 26 1 (<0·01) >0·999 27 1 (<0·01) >0·999 28 1 (<0·01) 0·940 29 1 (<0·01) 0·999 30 1 (<0·01) >0·999 31 1 (<0·01) 0·92 Mixed 1 1 118 (0·31) 1·110 x 10–176 2 92 (0·24) 3·810 x 10–145 3 38 (0·10) 2·459 x 10–50 4 35 (0·09) 9·772 x 10–48 5 27 (0·07) 3·638 x 10–27 6 14 (0·04) 8·751 x 10–9 7 14 (0·04) 3·294 x 10–8 8 13 (0·03) 6·636 x 10–6 9 4 (0·01) 0·773 10 3 (0·01) 0·033 11 3 (0·01) 0·352 12 2 (0·01) 0·244 13 2 (0·01) 0·476 14 1 (<0·01) 0·172 15 1 (<0·01) 0·833 16 1 (<0·01) 0·959 17 1 (<0·01) 0·464 18 1 (<0·01) 0·312 19 1 (<0·01) 0·662 20 1 (<0·01) 0·480 21 1 (<0·01) 0·307 22 1 (<0·01) 0·991 23 1 (<0·01) 0·789 24 1 (<0·01) 0·437 25 1 (<0·01) 0·884 26 1 (<0·01) 0·299 27 1 (<0·01) 0·532 28 1 (<0·01) 0·331 29 1 (<0·01) 0·648 30 1 (<0·01) 0·814 Mixed 2 1 38 (0·21) 4·115 x 10–75 2 12 (0·06) 3·038 x 10–17 3 11 (0·06) 4·503 x 10–27 4 11 (0·06) 4·615 x 10–6 5 10 (0·05) 1·253 x 10–11 6 8 (0·04) 5·866 x 10–11 7 8 (0·04) 5·866 x 10–11 8 7 (0·04) 6·409 x 10–13 9 7 (0·04) 6·899 x 10–11 10 7 (0·04) 0·011 11 6 (0·03) 1·044 x 10–9 12 6 (0·03) 5·644 x 10–5 13 4 (0·02) 1·340 x 10–5 14 4 (0·02) 0·040 15 4 (0·02) 0·057 16 4 (0·02) 0·577 17 3 (0·02) 0·001 18 2 (0·01) 7·291 x 10–5 19 2 (0·01) 9·043 x 10–5 20 2 (0·01) 9·043 x 10–5 21 2 (0·01) 8·879 x 10–4 22 2 (0·01) 0·074 23 2 (0·01) 0·270 24 1 (0·01) 0·011 25 1 (0·01) 0·083 26 1 (0·01) 0·118 27 1 (0·01) 0·136 28 1 (0·01) 0·004 29 1 (0·01) 0·296 30 1 (0·01) 0·001 31 1 (0·01) 0·013 32 1 (0·01) 0·697 33 1 (0·01) 0·468 34 1 (0·01) 0·974 35 1 (0·01) 0·025 36 1 (0·01) 0·026 37 1 (0·01) 0·069 38 1 (0·01) 0·040 39 1 (0·01) 0·028 40 1 (0·01) 0·103 41 1 (0·01) 0·505 42 1 (0·01) 0·505 43 1 (0·01) 0·233 44 1 (0·01) 0·884 45 1 (0·01) 0·884 46 1 (0·01) 0·679

	ACKNOWLEDGEMENTS

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

 
The authors are grateful to S. Yamamoto for his valuable suggestions, and to other members of both the Laboratory of Forest Ecology and Physiology, Nagoya University, and the Laboratory of Forest Ecology and Ecosystem Management, Tottori University, for their field and laboratory assistance. We thank the Tottori Distinct Forest Office for permitting this study, which was supported by a Grant-in-Aid for Scientific Research (No. 14206017) and JSPS Fellowship (No. 15000944) from the Japanese Society for the Promotion of Science.

	LITERATURE CITED

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

 

Auge H, Neuffer B, Erlinghagen F, Gruppe R, Brandl R. 2001. Demographic and random amplified polymorphic DNA analyses reveal high levels of genetic diversity in a clonal violet. Molecular Ecology 10: 1811–1819.[CrossRef][Medline]

Berg EE, Hamrick JL. 1994. Spatial and genetic structure of two sandhill oaks: Quercus laevis and Quercus margaretta Fagaceae. American Journal of Botany 81: 7–14.

Ceplitis A. 2001. The importance of sexual and asexual reproduction in the recent evolution of Allium vineale. Evolution 55: 1581–1591.[CrossRef][ISI][Medline]

de Kroon H, Hutchings MJ. 1995. Morphological plasticity in clonal plants: the foraging concept reconsidered. Journal of Ecology 83: 143–152.[CrossRef]

Diggle PK, Lower S, Ranker TA. 1998. Clonal diversity in alpine populations of Polygounm viviparum Polygonaceae. International Journal of Plant Sciences 159: 606–615.[CrossRef]

Eckert CG. 2002. The loss of sex in clonal plants. Evolutionary Ecology 15: 501–520.[CrossRef][ISI][Medline]

Eckert CG, Barrett SCH. 1993. Clonal reproduction and patterns of genotypic diversity in Decodon verticillatus (Lythraceae). American Journal of Botany 80: 1175–1182.[CrossRef]

Ellstrand NC, Roose ML. 1987. Pattern of genotypic diversity in clonal plant species. American Journal of Botany 74: 123–131.[CrossRef][ISI]

Esselman EJ, Jianquiang L, Crawford DJ, Winduss JL, Wolfe AD. 1999. Clonal diversity in the rare Calamagrostis porteri ssp. Insperata (Poaceae): comparative results for allozyme and random amplified polymorphic DNA (RAPD) and inter simple sequence repeat (ISSR) markers. Molecular Ecology 8: 443–451.[CrossRef]

Fager EW. 1972. Diversity: a sampling study. American Naturalist 106: 293–310.[CrossRef][ISI]

Firbank LG, Watkinson AR. 1987. On the analysis of competition at the level of the individual plant. Oecologia 71: 308–317.[CrossRef]

Fujita N. 1987. One-sided distribution of the component species of the Japanese beech forest along either the Pacific or the Japan Sea. Acta Phytotaxonomica et Geobotanica 38: 311–329 [in Japanese with English summary].

Gillespie JH. 1998. Population genetics: a concise guide. London: Johns Hopkins Press.

Handel SN. 1985. The intrusion of clonal patterns on plant breeding systems. American Naturalist 125: 367–384.[CrossRef]

Hangelbroke HH, Ouborg NG, Santamaría L, Schwenk K. 2002. Clonal diversity and structure within a population of the pondweed Potamogeton pectinatus foraged by Bewick's swans. Molecular Ecology 11: 2137–2150.[CrossRef][Medline]

Harada Y, Iwasa Y. 1996. Analyses of spatial patterns and population processes of clonal plants. Researches on Population Ecology 38: 153–164.[CrossRef]

Herben T, Hara T. 1997. Competition and spatial dynamics of clonal plants. In: de Kroon H, van Groenendael J, eds. The ecology and evolution of clonal plants. Leiden: Backhuys Publisher, 331–357.

Herrera CM, Jordano P, López-Soria L, Amat J. 1994. Recruitment of a mast fruiting, bird dispersed tree: bridging frugivore activity and seedling establishment. Ecological Monographs 64: 315–344.

Heywood JS. 1991. Spatial analysis of genetic variation in plant populations. Annual Reviews of Ecology and Systematics 22: 335–355.

Hoffmann RJ. 1986. Variation in contributions of asexual reproduction to genetic structure of populations of the sea anemone Metridium senile. Evolution 40: 357–365.[CrossRef]

Holderegger R, Stehlik I, Schneller JJ. 1998. Estimation of the relative importance of sexual and vegetative reproduction in the clonal woodland herb Anemone nemorosa. Oecologia 177: 105–107.[CrossRef]

Hsaio L-H, Rieseberg LH. 1994. Population genetic structure of Yushania niitakayamensis (Bambusoideae, Poaceae) in Taiwan. Molecular Ecology 3: 201–208.

Jordano P, Herrera CM. 1995. Shuffling the offspring: uncoupling and spatial discordance of multiple stages in vertebrate seed dispersal. Ecoscience 2: 230–237.

Kajimoto T. 1992. Dynamics and dry matter production of belowground woody organs of Pinus pumila trees growing on the Kiso mountain range in central Japan. Ecological Research 7: 333–339.[CrossRef]

Kreher SA, Fore SA, Collins BS. 2000. Genetic variation within and among patches of the clonal species, Vaccinium stamineum L. Molecular Ecology 9: 1247–1252.[CrossRef][Medline]

Kudoh H, Shibaike H, Takasu H, Whigham DF, Kawando S. 1999. Genet structure and determinants of clonal structure in a temperate deciduous woodland herb, Uvularia perfoliata. Journal of Ecology 87: 244–257.[CrossRef]

Lovett Doust L. 1981. Population dynamics and local specialization in a clonal perennial Ranunculus repens. 1. The dynamics of ramets in contrasting habitats. Journal of Ecology 69: 743–755.