Abstract
To determine the structural changes of a tree population in an isolated stand of lucidophyllous forest, we carried out a tree census in a 0.47-ha plot in a 3.4-ha primary forest in 1989 and 2000, and analyzed the 11-year change in the stem number for canopy and subcanopy species. The stems above 1.3 m in height in the study plot were classified into three size classes: canopy, subcanopy, and sapling. For the 52 species combined, the ratio of stems in the canopy class decreased (-1.0% year1^<-1>). The dominant canopy species, Distylium racemosum, showed little change in population structure and had abundant stems in all size classes in both 1989 and 2000. The three dominant canopy Cyclobalanopsis species (Quercus salicina, Q. sessilifolia, Q. gilva) had few stems in the subcanopy class in 1989. The stem number in the canopy class of each of these three Quercus species decreased considerably (more than -1.0% year^<-1>) and there was a small decrease in the subcanopy class in 2000. For Quercus, most of the stems newly recorded in the sapling class in 2000 were sprouts from stools recorded in 1989. For the Lauraceae and Theaceae, major families in the flora of lucidophyllous forests, the populations of most canopy and subcanopy species in 2000 were the same as in 1989. Of the major canopy and subcanopy species other than Quercus species, 43-97% of the stems newly recorded in the sapling class in 2000 were main stems that were recruited via seed reproduction. The population changes observed in this study site, which in 2000 had been undisturbed for at least 40 years, suggest that the population of Quercus species will decrease in the near future, while that of the other major species will remain stable. Although no structural characteristics or population changes were regarded as due to forest fragmentation, the loss of five species from a small population in 11 years suggests that there will be a further decrease in the number of species in the study plot. The tree population and the area of isolated stands with high species diversity, such as the study plot, should be increased by disseminating species to neighboring areas before many of the species in the stand lose their ability to regenerate by active seed reproduction.
