No abstract

No abstract

No abstract

Camellia japonica and Eurya japonica, broad‐leaved evergreen woody plants, are widely distributed in eastern Asia. Eurya japonica is dioecious, whereas flowers of C. japonica are perfect. We investigated the spatial distribution of genotypes among 15 and 21 Korean populations of C. japonica and E. japonica using spatial autocorrelation analysis of enzyme polymorphisms to test the previous prediction that dioecy is a primary factor which increases interpopulational gene flow. The patterns of spatial genetic distribution among populations of both species are very similar: percentage of significant Moran's /values (15 vs. 17 for C. japonica and E. japonica) and percentage of significant overall correlogram (18 vs. 20 for C. japonica and E. japonica). For both species, the patterns of average Moran's/values for each distance class were very similar as in an isolation by distance model. It may be concluded that dioecy may not a primary factor in shaping spatial genetic structure in E. japonica. Instead, abundant and continuously distributed patterns observed in E. japonica and an occasional pollen transfer by birds between isolated island populations of C. japonica in Korea may play roles in shaping spatial genetic structure among Korean populations of the two species.

Starch-gel electrophoresis was used to resolve nine polymorphic enzyme loci from leaf tissue collected from 20 Korean populations of Eurya japonica in order to determine differences in allele frequencies between male and female trees. In addition, 84 adults were sampled and mapped in a population located on Naenaro Island in Korea to examine spatial genetic structure using spatial autocorrelation analysis. Allele frequencies between males and females gave few contribution to the genetic structuring within populations. Only nine (5%) of 180 cases were significantly different from both sexes in allele frequencies. On the other hand, Moran's I was significantly different from the expected value in 31 (23.5%) of 132 cases. In the shortest distance (0 < 5 m), I was significantly positive in 10 (22.7%) of 44 cases. The results indicate that a significant small scale genetic structure was detected in the population and patch widths were inferred to be approximately 5-7 m. A nonrandom distribution of genotypes may be indicative of restricted gene flow via seed and pollen dispersal, and patchy establishments of genetically distinct individuals. These factors are responsible for shaping population genetic structure of E. japonica.