Authors:H. Bu, X. Chen, Y. Wang, X. Xu, K. Liu, and G. Du
In this paper, 633 species (involving 10 classes, 48 families, 205 genera) collected from the alpine meadow on the eastern Qinghai-Tibet plateau were studied. We tested potential factors affecting variation in mean germination time (MGT), i.e., plant traits (adult longevity, dispersal mode and seed size) or phylogeny, to evaluate if these factors were independent or they had interaction. Nested ANOVA showed that taxonomic membership accounted for the majority of MGT variation (70%), and in the generalized linear model, family membership could explain independently the largest proportion of MGT variation (29%). The strong taxonomic effect suggests that MGT variation within taxonomic membership is constrained. The other plant traits could also explain MGT variation independently (1% by adult longevity and dispersal mode, respectively, and 2% by seed size). Thus, the phylogeny was an important constraint to maintain the stability of species, and we could simplify the question if we regarded the phylogeny as an individual factor, but we could not negate the adaptive significance of the relationship between other plant traits and seed MGT. In addition, a large percentage of the variance remained unexplained by our model, thus important selective factors or parameters may have been left out of this analysis. We suggest that other possible correlates may exist between seed germination time and additional ecological factors (for example, altitude, habitat and post-dispersal predation) or phylogenetic related morphological and physiological seed attributes (e.g., endosperm mass) that were not evaluated in this study.
Authors:J. Wang, L. Huang, H. Ren, Z. Sun, and Q. Guo
Soil seed banks can act as an important source in forest regeneration, and the information on the seed bank composition is vital for determining the resilience of plant communities under severe environments such as urban settings. In this study, we examined the seed bank density and functional composition, and their relationships with aboveground vegetation in three remnant evergreen broad-leaved forests, i.e., PuGang (PG), LuoGang (LG), and DaLingShan (DLS) under urbanization in Guangzhou, South China. In both years of our study (2010–2011), seed density and species richness for overall soil seed banks and each classified life forms (tree, shrub, herb and grass) significantly differed among the forests and were much higher in the PG forest. The prevailing life forms in the seed banks were herbs and grasses, and the proportion of tree species Importance Value index (IV) of the seed banks was low. We did not detect significant difference in the percentage of exotic species seeds in the seed banks among the forests. The proportion of species with animal dispersal mode was much higher in the DLS forest than in the PG and LG forests. The similarity in species composition between standing vegetation and seed banks was low with the lower value in the DLS forest than in PG and LG forest. Our findings suggest that the regeneration potential of the soil seed banks is limited for the remnant forests in urban areas. Therefore, greater proactive and enhanced conservation efforts are thus needed.