Physical and chemical effects of litter on plant establishment in semi-natural grasslands

A field experiment was conducted in a semi-natural grassland to study the interspecific variation in the effect of litter on seedling emergence and establishment and separate physical from chemical effects. Seeds of seven forb species were sown in plots subjected to either litter amendment (0, 400 or 900 g m⁻²) or water extracts of litter (corresponding to 400 and 900 g litter m⁻²). In addition, an extract was treated with activated carbon to estimate the possible effects of secondary chemical compounds. The response to plant litter differed amongst species: negative, neutral and positive responses were observed. Anthriscus sylvestris was the only species with a strong positive response to litter. We found no consistent relation between seed size and response to plant litter. Physical effects of litter were generally stronger than chemical effects. However, water extract of litter inhibited emergence in three species. Activated carbon removed the negative effect of the litter extract, which suggests that the effect was caused by an inhibitory chemical compound rather than by increased competition in response to nutrients added via the extract. The balance between facilitative and inhibitory effects of litter depended on species identity and litter quantity. Facilitative effects dominated at low and intermediate quantities of litter, and inhibitory effects at high litter quantities. One species, Campanula rotundifolia, showed a switch from positive to negative responses with increasing quantities of litter. However, we found no general threshold for litter quantity valid across species.