Abstract:Habitat fragmentation poses a wide spread threat to biodiversity by disrupting seed production and dispersal. To understand the seed bank and natural regeneration patterns of the evergreen broad-leaved forest (EBLF), we selected five isolated fragments and one continuous forest of the typical EBLFs in the mountain ranges at the boundary of Zhejiang, Fujian, Jiangxi and Anhui Provinces in Eastern China. The six fragments are of similar age and have similar constructive species (Castanopsis eyrei and Schima superba).The seed bank were investigated during the early winter 2005 (December), later spring 2006(May), early winter 2006 (December) and later spring 2007(May), the sampling times (early winter and later spring) considered to be critical periods. At each sampling time, five 10 m intervals parallel lines were laid out through each of the patches at all six sites and each line had five 5 m intervals soil sampling frames. The sampling frame measured 10 cm×10 cm by 10 cm deep. Soil was carefully excavated in two 5-cm layers and bagged separately. The twenty-five samples, each comprising two depth subsamples, were taken for each site. Thus, 600 soil samples were obtained during the study:25 samples × 6 sites × 2 replicates × 2 seasons. At the same time, we had set twenty-five 1 m×1 m plots at each site to assess the natural regeneration pattern, each plot was 2 m away from the placement of the soil sampling frame. We censused and marked the seedlings of each plot in 2006, and all of the plots were recensused after one year. Soil samples were dried and passed through sieves to remove large stones, litter and sieved out larger seeds. The remaining soils were placed in plastic trays. They were held in a greenhouse with 12 ∶ 12 photoperiod, and mean temperatures of 25℃.Samples were watered daily and checked for emerging seedlings every two days for 6 months. All the seeds and seedlings were identified to species level. The result shows: (1) Significant differences were found among fragments in seed densities with values ranging from (207±38) to (1332±337) ind./m2, corresponding to 29-57 species. However, the seed density was (1776±344) ind./m2 in the continuous forest which located in Gutian Mountain National Nature Reserve, and corresponding to 46 species. The seed banks were mainly composed of tree species and shrub species, and there were rare herb species. In general, the plant species composition of the soil seed banks did not closely reflect the composition of the above-ground communities. (2) The mean proportion of the upper layer (0-5cm) seeds density was 3.6 times higher than that of the lower layer (5-10cm) in the five fragments, which was higher than the continuous forest's 2.5 times. It suggested that the continuous forest was more suitable for seeds to sink down and could storage more kinds of seeds. (3) Seedling regeneration appeared to be quite sensitive to seed bank species richness rather than seed density (P=0.005, 0.159, respectively). (4) Seed-seedling conflicts had occurred during some of these habitats if conditions or traits that were favorable for seeds were unfavorable for seedlings. Habitat fragmentation had reduced the density and species richness of seeds, however, because of more and more gaps caused by fragmentation and the following lower canopy coverage, it had positive effects on seedling establishment.

Seed rain as a propagule source for restoration of semi‐arid floodplain old fields

Quantifying seed rain patterns in a remnant and a chronosequence of restored tallgrass prairies in north central Missouri

Contribution of soil seed banks to vegetation resilience in coastal freshwater wetlands of subtropical Australia

Nutrient-acquisition strategy influences seed nutrient concentration and seed-to-seedling transition in ecological restoration in a regional dryland flora

Systematic reduction in seed rain of large‐seeded and endozoochorous species in pastures compared to forests along a tropical elevational gradient

The role of seed rain, seed bank, and clonal growth in plant colonization of ancient and restored grasslands

Linking ex situ germination to in situ direct seeding for landscape scale restoration efforts in the semiarid Mallee region of Victoria, Australia

Seed banks trigger ecological resilience in subalpine meadows abandoned after arable farming on the Tibetan Plateau.

Soil seed bank and natural regeneration of fragmented evergreen broad-leaved forest

Revegetated riparian areas are dominated by weeds, and lack structural diversity and natural recruitment: lessons for restoration practice

Impact of land-use intensity, productivity, and aboveground richness on seed rain in temperate grasslands

Wetland restoration by natural succession in abandoned pastures with a degraded soil seed bank

Basic characteristics and influencing factors of seed rain of woody plant in fragmented habitats in the Thousand Island Lake

Seed rain and soil seed bank in Chinese fir plantations and an adjacent natural forest in southern China: Implications for the regeneration of native species

Seed quality and the true price of native seed for mine site restoration

Look to seedling heights, rather than functional traits, to explain survival during extreme heat stress in the early stages of subtropical rainforest restoration

Floodplain meadow restoration revisited: Long-term success of large scale application of diaspore transfer with plant material in restoration practice

Transfer seeds, hay, or soil blocks? The importance of the completeness of biological inputs to address dispersal and establishment limitations during the restoration of plant assemblages in floodplain grasslands

River restoration effects on dispersal and the development of riparian seed bank: do poor seed banks limit restoration of boreal riparian zones?

Environmental flows drive sediment and seed supply in regulated rivers