HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters 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 Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (17) Citing Articles via Google Scholar Google Scholar Articles by Sun, M. Articles by Lee, J. S.Y. Search for Related Content PubMed Articles by Sun, M. Articles by Lee, J. S.Y. Agricola Articles by Sun, M. Articles by Lee, J. S.Y.

Reproductive biology and population genetic structure of Kandelia candel (Rhizophoraceae), a viviparous mangrove species¹

^a Department of Botany and Zoology, The University of Hong Kong, Pokfulam Road, Hong Kong; and ^b School of Applied Science, Griffith University Gold Coast, Gold Coast MC PMB 50, Queensland 9726, Australia

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The pollination biology, mating system, and population genetic structure of Kandelia candel were investigated. Field observations on its pollination and reproductive biology suggested that this species is pollinator dependent for fruit set, and bee activities can lead to substantial geitonogamous selfing. Quantitative analysis of the mating system parameters was performed using progeny arrays assayed for allozyme markers. Multilocus outcrossing rates (t_m) were estimated to be 0.697 ± 0.091 and 0.797 ± 0.062 in two populations. In comparison to other plant species with mixed-mating system, the level of allozyme variation was very low in the 13 populations sampled along the coastlines of Hong Kong. At the species level, the proportion of polymorphic loci was 20%, number of alleles per locus was 1.2, and heterozygosity was 0.0362. The total gene diversity was primarily distributed within populations (H_S = 0.0339), and the coefficient of genetic differentiation among populations was low (G_ST = 0.064). This pattern of population genetic structure suggests that gene flow, primarily in the form of water-dispersed seedlings in viviparous mangrove species, is not as limited as previously thought. However, microgeographic pattern in allele frequency at the marker loci could still be detected between the western and eastern coastal populations.

Key Words: Kandelia • mangrove • outcrossing rate • pollination biology • population genetic structure • Rhizophoraceae

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mangrove species are characteristic of tropical and subtropical coastlines. Approximately 80 species from ~ 20 plant families have been recognized, consisting primarily of trees and shrubs that normally grow above mean sea level in the intertidal zone of marine coastal environments or estuarine margins (Duke, 1992). The ecological and economic importance of mangroves cannot be overestimated. However, overexploitation in many parts of the tropics is depleting mangrove resources, resulting in the loss in many tropical nations of >1% of mangrove forest area per year (Umali et al., 1987; Hatcher, Johannes, and Robertson, 1989). Therefore, the conservation and management of mangrove resources have become an urgent task. Sound scientific knowledge of the basic biology of these species is essential to provide guidance for their conservation planning and management.

One of the important attributes of many mangroves is the production of viviparous offspring. Vivipary in flowering plants is defined as the precocious and continuous growth of the offspring when still attached to the maternal parent. In many viviparous mangrove species, the embryo penetrates through the fruit pericarp and grows to a considerable size before dispersal, such as in Kandelia candel and many other species in the Rhizophoraceae. Vivipary could be an adaptive feature for the plant to overcome the harsh tidal environment for seedling establishment in the parental sites, but it is not considered adaptive for dispersal either in time or space (Elmqvist and Cox, 1996). The rare successful long-distance dispersal event and rare opportunity for seedling establishment would cause founder effects in the resultant populations. The subsequent migration rate through dispersal of the viviparous seedlings at a microgeographical scale would determine the extent of genetic differentiation among local populations. However, the effect of this unique dispersal mechanism on the level of genetic variation within species and its distribution among populations have not previously been reported for any viviparous mangrove species.

In addition to the rate of migration or gene flow through seed or seedling dispersal, the pollination mechanism and mating system of a species also play an important role in determining population genetic structure. Although the mating systems of mangrove species may be inferred from their reproductive features, genetic marker analysis can provide more accurate information on the actual level of outcrossing in different populations. A few previous studies have used the offspring segregation ratios of chlorophyll-deficient heterozygotes to estimate outcrossing rates in natural populations of Rhizophora mangle (Lowenfeld and Klekowski, 1992; Klekowski, Lowenfeld, and Hepler, 1994) and Kandelia candel (Chen, Lin, and Lin, 1996). However, allozyme marker-based multilocus estimates of mating-system parameters have apparently not been obtained for any mangrove species.

The objective of this study was to investigate population genetic attributes of a typical species of mangrove, Kandelia candel (L.) Druce (Rhizophoraceae), in relation to its reproductive biology and viviparous seedling dispersal. Kandelia candel is one of the dominant species of mangroves along the Chinese coast (Li and Lee, 1997). Although considered to be of limited abundance throughout its range (Tomlinson, 1986), K. candel demonstrates reproductive habits typical of the Rhizophoraceae. Information on the mating system parameters will help explain the pattern of genetic diversity in the species. Knowledge of population genetic structure of this important mangrove species will also contribute to its conservation and management. In Hong Kong, mangrove stands have been greatly reduced to ~15% of their original distribution over the past few decades, mainly due to reclamation and urban development (Yipp, Hau, and Walthew, 1995). Up to the present, only limited ecological surveys on these mangroves have been conducted (e.g., Tam and Wong, 1997). No genetic information is available for incorporation in developing conservation strategies. This study provides needed genetic data for guiding conservation and management practice that would maximize genetic diversity in the species.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Species description
Kandelia candel is a small tree. It has been recorded from Indo-Malaya, southern China, and the Ryukyus with a northern limit in southern Japan. In comparison to other mangroves, the species has a limited distribution globally. It gives the northern limit of global mangrove distribution at 31°23'N but has an unusually restricted southern limit in Java. Throughout its range, the species was considered to be nowhere abundant (Tomlinson, 1986). In Hong Kong, however, it can be found in most of the 45 known mangrove locations along the coastlines of the New Territories, Hong Kong Island, and Lantau Island. In most of these locations, the population size is over a few hundreds of individuals. The plants may vary in height among the sites, from small trees of ~1 m to a maximum of ~8 m tall. In Hong Kong, the species flowers from early June until late August, and fruits and viviparous seedlings (propagules) grow on the trees from September to June.

Pollination ecology and bagging experiment
For studying population genetic structure, 13 populations of K. candel throughout Hong Kong were sampled in May and June 1997. Flower buds or the youngest leaf buds were collected from ~30 individuals per population site (see Fig. 1 for population names and locations). For estimating outcrossing rates, two of the population sites, Site A and Maipo, were sampled. These sites were chosen for having abundant propagules still attached to the trees at the time of collection. An average of eight propagules per tree and ~30 trees per site were sampled to provide materials for progeny arrays used in mating system analysis. The propagules collected from the same tree were kept separate on a family basis and half-immersed in water with a salinity similar to that at the sites of collection, until the completion of isozyme electrophoresis.

View larger version (28K): [in this window] [in a new window]   Fig. 1. Map showing location of the populations of Kandelia candel sampled. Population names are 1, Site A; 2, Maipo; 3, Tsimbeitsui; 4, Sheungpaknai; 5, Hapaknai; 6, Lukkeng; 7, Tingkok; 8, Naichung; 9, Pakshawan; 10, Tai Tam; 11, Tungchung; 12, Tai O; and 13, Tongfuk.

Isozyme electrophoresis
Flower buds or the youngest leaf buds were collected from the field and kept cool and moist until electrophoresis, normally conducted on the day following the collection. The fresh materials were ground with an extraction buffer as described in Sun and Ganders (1990). Two buffer systems were used to survey 14 enzyme systems in 12.5% starch gels. The "H" buffer system (histidine-tris-citrate system, pH 7.0; Sun and Corke, 1992) was used to resolve isocitrate dehydrogenase (IDH; E.C. [Enzyme Commission number] 1.1.1.42), malate dehydrogenase (MDH; E.C. 1.1.1.40), malic enzyme (ME; 1.1.1.40), phosphoglucomutase (PGM; E.C. 2.7.5.1), 6-phosphogluconate dehydrogenase (6PGD; E.C. 1.1.1.44), and shikimate dehydrogenase (SKDH; E.C. 1.1.1.25). The "L" buffer system (pH 7.8; Shields, Orton, and Stuber, 1983) was used to resolve aspartate aminotransferase (AAT; E.C. 2.6.1.1), aconitase (ACO; E.C. 4.2.1.3), esterase (EST; E.C. 3.1.1.1), leucine aminopeptidase (LAP; E.C. 3.4.11.1), phosphoglucoisomerase (PGI; E.C. 5.3.1.9), and triose-phosphate isomerase (TPI; E. C. 5.3.1.1). Fluorescent esterase (F-EST; E.C. 3.1.1.1) and ß-glucosidase (GLU; E.C.3.2.1.21) were resolved by either of the buffer systems. The protocols of Wendel and Weeden (1989) were used for enzyme activity staining. Genetic interpretation of band patterns followed standard principles (Weeden and Wendel, 1989; Wendel and Weeden, 1989).

Data analysis
Measures of within-population genetic variation include polymorphism, allelic diversity, and heterozygosity. The percentage of polymorphic loci (P ) was calculated based on the 99% criterion. The P values were not significantly different when the 95% criterion was used. Allelic diversity was measured as the number of alleles per locus (A). Gene diversity (H), the mean expected heterozygosity assuming Hardy-Weinberg equilibrium, was calculated as H = 1 - 1/n_jip_ij, where p_ij is the frequency of the ith allele at the jth locus, and n is the total number of isozyme loci surveyed. Heterozygosities observed (H_o) and expected (H_e) were calculated only for two commonly polymorphic loci, Skdh-1 and Tpi-2, in these populations.

The proportion of genetic variation distributed within (H_S) vs. between populations (D_ST) was measured using Nei's (1973) gene diversity statistics. The coefficient of genetic differentiation among populations, G_ST, was used to estimate the level of gene flow, Nm (the number of migrants exchanged between local populations per generation), based on the relationship G_ST = 1/(4Nm+1) where G_ST is Nei's (1973) estimator of F_ST (Wright, 1951). The UPGMA (unweighted pair-group arithmetic averaging) dendrogram of Nei's (1972) genetic distance was constructed using the GDD computer program provided by K. Ritland.

For the estimation of outcrossing rates, two polymorphic allozyme loci, Skdh-1 and Tpi-2, were used as gene markers in Site A, and two more loci, Mdh-2 and Pgm-1, were sufficiently polymorphic as markers in Maipo. Single-locus Mendelian inheritance at these loci was verified by the conformity of progeny array patterns to Mendelian predictions, a useful method when controlled crosses are not performed (Kephart, 1990). Mating-system parameters were estimated using Ritland's MLTR program following Sun and Ritland (1998). The mating-system parameters estimated included: (a) the multilocus population outcrossing rate t_m (Ritland and Jain, 1981); (b) the average of single-locus population outcrossing rate t_s; (c) the inbreeding coefficient of maternal parents F; (d) the correlated selfing rate r_s; and (e) the correlation of paternal genotypes between outcrossed sibs r_p. After it was determined that pollen and ovule population gene frequencies did not significantly differ, these frequencies were pooled to increase statistical power to estimate mating system parameters. The Newton-Raphson method was used to find the maximum likelihood estimates. Standard errors of population estimates, including t_m - t_s, were calculated using 100 bootstraps, where the unit of resampling was the progeny array.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reproductive biology
The number and type of pollinators visiting K. candel show that the species has nonspecialized floral mechanism. During a 30-min period of observation in early July, we recorded two species of bees (Apis mellifera L. and Xylocopa iridipennis Lep.) and four species of butterflies (Delias pasithoe L., Euchrysops cnejus Fabricius, Euploea midamus L., and Heliophorus epicles Godart) visiting the flower of K. candel. Bees tend to visit more than one flower on the same inflorescence or different inflorescences on the same plant before flying away to visit the neighboring plants and may come back again to visit the same flowers later. Butterflies, however, tend to visit only one flower at a time before flying away to visit other plants. Apparently substantial geitonogamous selfing may result from bee activity.

Of the 20 inflorescence branches bagged in late June, we found only two fruits in early September, one each from two bags. The sample estimate of the number of flowers per bagged branch is 18.6 ± 5.0, and thus the fruit set in the bagged inflorescences was only ~0.54%. A sample of 28 open-pollinated inflorescences on the same set of trees produced an average of 4.54 ± 2.0 fruits each, giving an estimate of 24.41% natural fruit set at the site. A common feature among the viviparous mangroves is that only one seedling per fruit emerges while still attached to the mother tree, and there are six ovules per flower, so the actual fertility of the bagged flowers is as low as 0.09%. These data suggest that automatic self-fertilization events are rare in the species, and natural fruit setting is almost completely pollinator dependent.

The mating system analysis showed that the inbreeding coefficient of maternal parents (F) was not significant in the populations (Table 1). Multilocus outcrossing rates ranged from 70 to 80%, and averaged 75 ± 7% for the two populations studied. The single-locus outcrossing estimates varied within populations, and the standard error (SE) estimates at each marker locus appear to be affected by the allele frequency distribution (Table 2). The marker loci having skewed allele frequencies, such as Mdh-2 and Pgm-1, tend to have much higher SE values than the markers having more even allele frequencies, such as Skdh-1 and Tpi-2. It is apparent that more accurate outcrossing estimates could be obtained from these more informative marker loci. Genetic substructuring, in terms of biparental inbreeding (detectable by the difference between multilocus and single-locus selfing rates), was not significant in either population. The correlation of paternity (i.e., probability that sibs shared the same father), r_p, was significant at Site A, indicating a small "paternal mating pool" of ~2–3 plants in this population (Table 1). The "paternal mating pool" is calculated as the inverse of r_p and is the number of pollen donors that gives rise to this correlation, assuming all paternal trees have equal mating probabilities and consecutive matings are independent (Ritland, 1989).

View larger version (15K): [in this window] [in a new window]   Fig. 2. Dendrogram of Nei's genetic distances among populations of Kandelia candel (see Fig. 1 for population locations).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mating system has traditionally been characterized by selfing or outcrossing rates, which influence patterns of population genetic structure. The pattern of paternity, measured by the parameters of the "correlated mating model" of Ritland (1989), is often missing in plant mating system studies. These parameters describe the genetic relatedness among progeny of a plant. One parameter is the correlation of outcrossed paternity within progeny arrays, r_p (defined as the proportion of pairs of progeny that are full-sibs, as opposed to half-sibs; see Ritland, 1989). This correlation arises from repeated matings to the same near-neighbors or from the tendency of pollinator pollen loads to be derived from single plants. In partially selfing populations, correlations among progeny also arise from variation of selfing rates among plants. This is measured by a second parameter, the correlation of selfing within progeny arrays, r_s (Ritland, 1989). In addition to selfing or outcrossing rates, correlations among progeny affect local patterns of genetic variation.

This study revealed a significant level of selfing in K. candel, with a mean multilocus selfing rate of 25.3% in the populations assayed. Based on the results of the bagging experiment, automatic self-pollination is obviously negligible, if any, in the species. However, as the plant produces many flowers concurrently at anthesis, and bees often visit several flowers consecutively on the same inflorescence or different inflorescences on the same plant, the potential for geitonogamous selfing is high. The estimated 25.3% selfing rate could be partly or mainly a result of geitonogamous selfing. As no significant differences were detected between t_m and t_s, biparental inbreeding is apparently not an important factor in the populations of K. candel. The occurrence of outbred parents, as shown by nonsignificant F and f values detected in these populations (Tables 1, 4), is in agreement with nonsignificant biparental inbreeding. Although we detected a significant heterozygote excess in one population and no significant departure from Hardy-Weinberg expectations in 12 other populations studied, the estimated selfing rates in the populations showed a significant departure from random mating. This suggests the presence of selection against inbred progeny, which can occur if genetic load is high, such as in cases of traditionally outcrossing species.

Based on a marker gene for chlorophyll deficiency, the estimated single-locus selfing rates ranged from 17.7 to 19.7% in two populations of Kandelia candel in Fujan (Chen, Lin, and Lin, 1996), not significantly lower than the average single-locus selfing rate (23.9%, based on Table 2) detected in the present study. In contrast, using the same marker gene for chlorophyll deficiency, the selfing rates were estimated to be 71.2 to 95.2% in populations of Rhizophora mangle (Lowenfeld and Klekowski, 1992; Klekowski, Lowenfeld, and Hepler, 1994), indicating that this mangrove species is a predominant selfer.

Sharing of paternity within a family, as measured by the correlations of outcrossed paternity, r_p, could arise from multiple deposits of pollen from a single male parent, or from mating to a small number of near neighbors. In K. candel, the extent of correlated paternity between flowers of the same plant is significant at Site A (0.351), equivalent to random mating to an effective pool of 1/0.351 or about three neighbors. The large standard error of r_p at Maipo made the estimate insignificant, but it can still be concluded that, on average, at least six neighboring trees contributed genes to the progeny of the maternal parent in the population. Differences in the degree of relatedness among progeny could contribute to the observed variation in genetic structure between local populations (Table 4).

Population genetic consequences
Nearly all species with true vivipary tend to inhabit shallow marine habitats, which are normally characterized by extraordinarily coarse-grained environments for seedling establishment (Elmqvist and Cox, 1996). Selection in such environments would favor vivipary since the seed germinates while still attached to the mother tree and the young seedling grows to a considerable length before falling. Thus, seedling survival and establishment in the parental sites would be easier to achieve. However, viviparous seedlings are considered ill-adapted for dispersal since they lack the protection and nutritional support from the maternal tissue. However, many of the viviparous seedlings are buoyant, capable of long-distance dispersal by ocean currents, such as in the case of K. candel, whose propagules can grow to 20–47 cm long and 1.0–1.3 cm wide, depending on population locations and environmental conditions (Maxwell, 1995). Propagule size may reach 70 cm in Rhizophora mucronata, another member of the Rhizophoraceae. Colonization through long-distance dispersal by ocean current may frequently occur in these species, but likely involves a small number of founders. In addition, the probability of an offspring being dispersed to a site better than the parental patch is considered very low in the marine environment (Elmqvist and Cox, 1996). Thus, the rare success of a few founders in establishing a new population would lead to a genetic erosion. This probably accounts for the observed amount of genetic variation in K. candel in Hong Kong, which is much lower than the mean levels of genetic variation in species with mixed-mating systems (see review in Hamrick and Godt, 1996). Since very little information is available on other viviparous mangrove species regarding their mating systems and population genetic diversity, no conclusion can be made as to whether the pattern of genetic diversity in K. candel is representative of those existing in other viviparous mangrove species. A recent study of a nonviviparous mangrove species, Acanthus ilicifolius Linn. (Acanthaceae), revealed a very low level of within-population genetic variation but substantial genetic polymorphism between populations (Lakshmi et al., 1997). However, no information on its pollination biology or mating system is given in the paper. It is not clear whether the pattern of genetic variation in this species is primarily affected by its mating system or largely due to its being an opportunistic colonizer.

The high genetic identities among the populations of K. candel in Hong Kong suggest their recent coancestry and/or lack of polymorphism in the species. A recent coancestry and/or a high level of gene flow among the local populations could result in the lack of population genetic differentiation (G_ST = 0.064). However, at the two common marker loci, Skdh-1 and Tpi-2, microgeographic differentiation in allele frequencies is apparent, resulting in the clustering of populations into two major groups separating the west coast from the east coast. Thus, it can be concluded that seedling dispersal in K. candel is somehow restricted between coastlines in Hong Kong. Also, genetic differentiation occurs much faster at the isolated sites. For example, individuals of Site A are located in a shallow impoundment at the Maipo marshes long-established (>50 yr) by farmers for culturing shrimps. Natural dispersal of propagules by tidal waves or ocean current is apparently limited as can be seen by the accumulation of propagules under the canopy at the site. Lee (1989) estimated that <1% of the litter (including propagules) produced by K. candel inside the ponds was exported away from the mangrove stand. The level of migration or gene flow through viviparous seedlings in K. candel apparently played an important role in shaping population genetic structure.

Due to the recent construction and development of an airport, highways, and ports on Lantau Island, many mangroves have been destroyed and those remaining are scattered on all sides of the island (Tam and Wong, 1997). Of the three populations of K. candel sampled from this island, two (Tungchung and Tongfuk) clustered with the West Coast populations, whereas one (Tai O) clustered with the East Coast populations (Fig. 2). Historical factors may account for this pattern of genetic divergence. The sampled site in Tai O was located in an abandoned salt pan that was used to produce salts for several centuries (Tam and Wong, 1997). Some salt pans were later converted to fish ponds and seawalls were built to protect the ponds from strong ocean currents and erosion. Long isolated from the sea, natural gene flow through propagules between this site and others could be very restricted. Genetic drift in allele frequencies in this small population likely accounts for its clustering by chance with the East Coast populations. Due to rapid development in Tai O, very few mangroves remain at the site, and the small patches still present are fragmented and extremely polluted with sewage from neighboring restaurants and households. However, a recent ecological survey suggests that the abandoned salt pans can be made suitable for artificial replanting of mangroves (Tam and Wong, 1997). Enhanced gene flow through transplanting propagules from other populations to this site would also help to maintain the total gene diversity in the species.

Conservation implications in Hong Kong
Compared to the mangroves in the tropics, fewer mangrove species are present in Hong Kong, with smaller stature and limited abundance. Of the eight mangrove species occurring in Hong Kong, four produce viviparous seedlings. Among them, K. candel, Avicennia marina, and Aegiceras corniculatum are the main components in the seaward mangroves. In recent years, urban development in Hong Kong has exerted great pressure on mangroves. Mangroves have been cut and reclaimed at an alarming rate (Morton and Morton, 1983). The remaining mangroves can be found at ~45 locations along the coastlines of the mainland territories and the two major islands, Hong Kong Island and Lantau Island. K. candel is one of the three seaward species and constitutes the major "pioneer" species responsible for seaward extension of the mangrove communities. Research on mangroves on the Chinese coast has been limited to the documentation of species composition and their ecology (see Li and Lee, 1997 for a review). Research on population genetic diversity of mangrove species is still in its infancy worldwide. Such knowledge is important to the management of the mangrove resources. Knowledge of genetic composition of populations and the degree of interpopulational divergence can help managers to make informed decisions regarding how to protect the existing genetic resources. As the difference among populations of K. candel in Hong Kong is mostly in allele frequencies rather than in gene composition, loss of populations at certain locations may not cause immediate loss in genetic diversity but more damage may occur in terms of long-term genetic consequences due to reduced number of populations and smaller population size. Replacement of the lost populations at certain locations can be achieved by transplanting a large number of propagules from long-established, large populations.

	FOOTNOTES

 
¹ The authors thank Mr. Alfred Lai and Ms. Sandy Kwok for their technical assistance, Ms. W. Y. Law for identification of pollinators, and Dr. Harold Corke for helpful comments on the manuscript. This research was supported by a CRCG grant from the University of Hong Kong.

² Author for correspondence.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Baker, H. G.,1955Self-compatibility and establishment after "long-distance" dispersal. Evolution 9: 347–348. [CrossRef][ISI]

Chen, X-Y., P. Lin, and Y-M. Lin.1996Mating systems and spontaneous mutation rates for chlorophyll-deficiency in populations of the mangrove Kandelia candel. Hereditas 125: 47–52. [CrossRef][ISI]

Duke, N. C.1992Mangrove floristics and biogeography. In A. Robertson and D. Alongi [eds.], Tropical mangrove ecosystem, 63–100. American Geographical Union, Washington, DC.

Elmqvist, T., and P. A. Cox.1996The evolution of vivipary in flowering plants. Oikos 77: 3–9.

Hamrick, J. L., and M. J. W. Godt.1996Effects of life history traits on genetic diversity in plant species. Philosophical Transactions of the Royal Society of London B 351: 1292–1298.

Hatcher, B. G., R. E. Johannes, and A. I. Robertson.1989Review of research relevant to conservation of shallow tropical marine ecosystems. Oceanography and Marine Biology: An Annual Review 27: 337–414.

Kephart, S. R.1990Starch gel electrophoresis of plant isozymes: a comparative analysis of techniques. American Journal of Botany 77: 693–712. [CrossRef][ISI]

Klekowski, E. J., JR., R. Lowenfeld, and P. K. Hepler.1994Mangrove genetics. II. Outcrossing and lower spontaneous mutation rates in Puerto Rican Rhizophora. International Journal of Plant Sciences 155: 373–381. [CrossRef]

Lakshmi, M., S. Rajalakshmi, M. Parani, C. S. Anuratha, and A. Parida.1997Molecular phylogeny of mangroves. I. Use of molecular markers in assessing the intraspecific genetic variability in the mangrove species Acanthus ilicifolius Linn. (Acanthaceae). Theoretical and Applied Genetics 94: 1121–1127. [CrossRef][ISI]

Lee, S. Y.1989Litter production and turnover of the mangrove Kandelia candel (L.) Druce in a Hong Kong tidal shrimp pond. Estuarine, Coastal and Shelf Science 29: 75–87. [CrossRef][ISI]

Li, M. S., and S. Y. Lee.1997Mangroves of China: a brief review. Forest Ecology and Management 96: 241–259. [CrossRef][ISI]

Lowenfeld, R., and E. J. Klekowski, Jr.1992Mangrove genetics. I. Mating system and mutation rates of Rhizophora mangle in Florida and San Salvador Island, Bahamas. International Journal of Plant Sciences 153: 394–399. [CrossRef]

Maxwell, G. S.1995Ecogeographic variation in Kandelia candel from Brunei, Hong Kong and Thailand. Hydrobiologia 295: 59–65. [CrossRef][ISI]

Morton, B., and J. Morton.1983The sea shore ecology of Hong Kong. Hong Kong University Press, Hong Kong.

Nei, M.1972Genetic distance between populations. American Naturalist 106: 283–292. [CrossRef][ISI]

———.1973Analysis of gene diversity in subdivided populations. Proceedings of the National Academy of Sciences, USA 70: 3321–3323. [Abstract/Free Full Text]

Ritland, K.1989Correlated matings in the partial selfer, Mimulus guttatus. Evolution 43: 848–859.

———, and S. K. Jain.1981A model for the estimation of outcrossing rate and gene frequencies using n independent loci. Heredity 47: 35–52. [ISI]

Shields, C. R., T. J. Orton, and C. W. Stuber.1983An outline of general resource needs and procedures for the electrophoretic separation of active enzymes for plant tissue. In S. D. Tanksley and J. J. Orton [eds.], Isozymes in plant genetics and breeding, part A, 443–468. Elsevier Science, Amsterdam.

Sun, M., and H. Corke.1992Population genetics of colonizing success of weedy rye in Northern California. Theoretical and Applied Genetics 83: 321–329. [ISI]

———, and F. R. Ganders.1990Outcrossing rates and allozyme variation in rayed and rayless morphs of Bidens pilosa. Heredity 64: 139–143. [ISI]

———, and K. Ritland.1998Mating system of yellow starthistle (Centaurea solstitialis), a successful colonizer in North America. Heredity 80: 225–232. [CrossRef][ISI]

Tam, N. F. Y., and Y-S. Wong.1997Ecological study on mangrove stands in Hong Kong, vol. 1. Agriculture and Fisheries Department, Hong Kong Special Administrative Region Government.

Tomlinson, P. B.1986The botany of mangroves. Cambridge University Press, Cambridge.

———, R. B. Primack, and J. S. Bunt. 1979Preliminary observations on floral biology in mangrove Rhizophoraceae. Biotropica 11: 256–277. [CrossRef][ISI]

Umali, R. M., P. M. Zamora, R. R. Gatera, R. S. Jara, A. S. Camacho, and M. Vannucci.1987Mangrove of Asia and the Pacific: status and management. Technical Report of the UNDP/UNESCO Research and Training Pilot Programme on mangrove ecosystems in Asia and the Pacific (RAS/79/002), Quezon City, The Philippines.

Weeden, N. F., and J. F. Wendel.1989Genetics of plant isozymes. In D. E. Soltis and P. S. Soltis [eds.], Isozymes in plant biology, 46–72. Dioscorides Press, Portland, OR.

Wendel, J.F., and N. F. Weeden.1989Visualization and interpretation of plant isozymes. In D. E. Soltis and P. S. Soltis [eds.], Isozymes in plant biology, 5–45. Dioscorides Press, Portland, OR.

Wright, S.1951The genetical structure of populations. Annals of Eugenics 15: 323–354. [ISI]

Yipp, M. W., C. H. Hau, and G. Walthew.1995Conservation evaluation of nine Hong Kong mangals. Hydrobiologia 295: 323–333. [CrossRef][ISI]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters 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 Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (17) Citing Articles via Google Scholar Google Scholar Articles by Sun, M. Articles by Lee, J. S.Y. Search for Related Content PubMed Articles by Sun, M. Articles by Lee, J. S.Y. Agricola Articles by Sun, M. Articles by Lee, J. S.Y.