Abstract. Camellia japonica L. (Theaceae), an insect-and bird-pollinated, broad-leaved evergreen tree, is widely distributed in Japan and the southern Korean peninsula. The species has a relatively even age distribution within populations, which may influence the spatial genetic structure of different age classes relative to species with typical L-shaped age distributions. To determine whether the internal spatial genetic structure found in seedlings and young individuals carries over into adults, we used allozyme loci, F-statistics, spatial autocorrelation statistics (Moran's I), and coancestry measures to examine changes in genetic structure among seven age classes in a population (60-m ϫ 100-m area) in southern Korea. In seedlings, weak but significant positive values of Moran's I-statistics and coancestry measures were found for distances less than 14 m, which is consistent with a mechanism of limited seed dispersal combined with overlapping seed shadows. This spatial structure, however, dissipates in older age classes, and in adults genetic variation has an essentially random spatial distribution. Morisita's index of dispersion of individuals in each age class showed that seedlings and juveniles are more highly clustered than are older individuals. These results suggest that self-thinning changes the spatial relationships of individuals, and thus genotypes. A multilocus estimate of F ST (0.008) shows a small but statistically significant difference in allele frequencies among age classes. In summary, intrapopulation genetic structure within and among age classes of C. japonica was significant but weak. Despite presumably limited seed dispersal, weak spatial genetic structure in juveniles suggests overlapping seed shadows followed by self-thinning during recruitment. The present study also demonstrates that studies of spatial genetic structure focusing on limited numbers of generations may not be sufficient to reveal the entire picture of genetic structure in populations with overlapping generations. Theoretical models have shown that the genetic structure of local populations is affected by various nonequilibrium processes (Slatkin 1977;Wade and McCauley 1988). Genetic and demographic factors, such as pollen and seed dispersal from neighboring parents, past major reproductive events, selection, and other processes determining the life history of a species, influence population genetic processes (Jain and Bradshaw

No abstract

Orchid seeds are unusual for being the smallest among flowering plants. These dust-like seeds are wind-borne and, thus, would seem to have the potential for long-distance dispersal (a common perception); this perception has led to a prediction of near-random spatial genetic structure within orchid populations. Mathematical models (e.g., simple ballistic model) for wind-dispersed seeds and wind-tunnel experiments, in contrast, indicate that most seeds of orchids should fall close to the maternal plant (<6 m), supporting a prediction of significant fine-scale genetic structure within populations. In reality we do not know much about seed dispersion in orchids. To determine which of these two predictions is more appropriate, Wright's F statistics and spatial autocorrelation analysis were used to examine the genetic structure within two adult populations of the terrestrial orchid Cephalanthera longibracteata (Orchidaceae) in southern Korea. In results comparable to those of other self-compatible, mixed-mating plant species, C. longibracteata populations exhibited low levels of genetic diversity (mean H(e) = 0.036) and a significant excess of homozygosity (mean F(IS) = 0.330), consistent with substantial inbreeding via selfing and/or mating among close relatives in a spatially structured population. Spatial autocorrelation analysis revealed significant positive genetic correlations among plants located <10 m, with relatedness at <3 m comparable to that expected for half sibs and first cousins. This genetic structure supports the prediction that the majority of seed dispersal occurs over distances of less than 10 m and is responsible for generating substantial overlap in seed shadows within C. longibracteata populations.

We investigated the potential for gene flow and genetic assimilation via hybridization between common and rare species of the terrestrial orchid genus Liparis, focusing specifically on sympatric and allopatric populations of the common Liparis kumokiri and the rare Liparis makinoana. We utilized analyses of genetic diversity, morphology, and the spatial distributions of individuals and genotypes to quantify the dynamics of interspecific gene flow at within- and among-population scales. High levels of allozyme genetic diversity (HE) were found in populations of the rare L. makinoana (0.317), whereas the common L. kumokiri (N = 1744 from 14 populations) revealed a complete lack of variation. This contrast may reflect different breeding systems and associated rates of genetic drift (L. makinoana is self-incompatible, whereas L. kumokiri is self-compatible). At the two known sympatric sites, individuals were found that recombined parental phenotypes, possessing floral characteristics of L. kumokiri and vegetative characteristics of L. makinoana. These putative hybrids were the only individuals found segregating alleles diagnostic of both parental species. Analysis of these individuals indicated that hybrid genotypes were skewed towards L. kumokiri and later generation recombinants of L. kumokiri at both sympatric sites. Furthermore, Ripley's bivariate L(r) statistics revealed that at one site these hybrids are strongly spatially clustered with L. kumokiri. Nonetheless, the relatively low frequency of hybrids, absence of ongoing hybridization (no F1s or first generation backcrossess), and strong genetic differentiation between morphologically 'pure' parental populations at sympatric sites (FST = 0.708-0.816) indicates that hybridization was not an important bridge for gene flow. The results from these two species suggest that natural hybridization has not played an important role in the diversification of Liparis, but instead support the view that genetic drift and limited gene flow are primarily responsible for speciation in Liparis. Based on genetic data and current status of the species, implications of the research for conservation are considered to provide guidelines for appropriate conservation and management strategies.

Effective population size ( N e ) influences the degree to which random genetic drift changes allele frequencies, increases inbreeding, and decreases genetic diversity, and thus is a parameter of direct relevance to the conservation of rare species. Few empirical data are available, however, concerning the effects of clonal structure on N e in plant species reproducing both sexually and asexually. Using genetic markers and spatial autocorrelation analysis, we quantified the statistical significance and spatial scale of clonal spread of six populations of the rare terrestrial orchid Cremastra appendiculata (D. Don) Makino in a large (180 ha), undisturbed landscape on Oenaro Island, located off of the southeastern coast of South Korea. We used this information to calculate three demographic estimators of N e : the number of ramets within a population, the number of genets, and an estimator that incorporates information on both the number of genets and variation in the number of ramets per genet. Taking clonal structure into account results in up to a fivefold decrease in estimates of N e relative to the ecologically apparent number of individuals within populations. Levels of standing genetic variation are in fact greater than expected given our estimates of N e , leading us to consider historical factors resulting in N e being greater in the past than in present-day populations. Like C. appendiculata, many terrestrial orchids occur in relatively small, spatially isolated populations and are of special concern for conservation. Our results indicate that efforts aimed at the long-term preservation of these species should be based on a sound understanding of the potential for clonal structure and its implications for the sensitivity of populations to losses of genetic diversity and fitness via random genetic drift.Resumen: El tamaño poblacional efectivo ( N e ) influye en la proporción en la que las frecuencias alélicas son cambiadas por deriva génica aleatoria y disminuye la diversidad genética, y por lo tanto es un parámetro de relevancia directa para la conservación de especies raras. Sin embargo, existen pocos datos empíricos sobre los efectos de la estructura clonal sobre N e en especies de plantas de reproducción sexual como asexual. Con el uso de marcadores genéticos y análisis de autocorrelación espacial cuantificamos la significación estadística de la expansión clonal de seis poblaciones de la orquídea terrestre rara Cremastra appendiculata (D. Don) Makino en un paisaje extenso (180 ha) no perturbado en la Isla Oenaro, localizada cerca de la costa sur de Corea del Sur. Utilizamos esta información para calcular tres estimadores demográficos de N e : el número de rámulas en una población, el número de "genets" y un estimador que incorpora información tanto del número de "genets" y la variación en el número de rámulas por "genet." Tomando a la estructura clonal en consideración resulta en un decremento de hasta cinco veces en las estimaciones de N e en relación con el ‡Address correspondence to M. G. Chung, 151...