EN
To determine the optimum plot size for bryophyte-biocoenosis investigations and identify the sampling methods that can provide sufficient and representative data for bryophyte diversity, we explored two sampling methods and investigated the species composition and their coverage in three sites located within the subtropical- temperate zone in China: mixed coniferous and deciduous broadleaved forest with species of Quercus, Betula, Pinus, Abies (Guanyinshan Nature Reserve), mixed evergreen and deciduous broadleaved forest, broadleaved forest with species of Phoebe, Castanopsis, Lithocarpus, Acer, Tilia, Dacidia, Abies (Mabian Dafengding Nature Reserve) and temperate deciduous broadleaved forest with species of Quercus, Betula, Populus (Dongling Mountains). Two sampling methods are applied: the systematic-sampling method based on nested quadrates of five different sizes (10 × 10 cm, 25 × 25 cm, 50 × 50 cm, 1 × 1 m, and 2 × 2 m) in each 2 × 2 m grid within a 10 × 10 m plot (total 25 grids) and the microcoenose-sampling method (sampling with the minimum area quadrate at the center of the largest fragment in each of the 25 grids). The minimum area of sampling was determined based on the similarity-area curve, the coverage-area curve, and the importance- value-area curve through the systematicsampling. The appropriate sampling method and quadrate number were determined by analyzing the species diversity and evenness. We compared two sampling methods by assessing the species number at two different sites. Both the similarity- area curve and importance-value-area curve showed that the turnover point of sampling size occurred at 50 × 50 cm where the similarity and importance-value were closer to the actual. We concluded that a quadrate of 50 × 50 cm could be used as the minimum area of sampling. However, the systematic-sampling method was not suitable for analyzing the diversity of bryophytes. A viable alternative is the microcoenose-sampling method which allows to obtain sufficient information in terms of species richness and their distribution.