DECOMPOSITION OF LEAF-LITTER IN A MULL BEECH FOREST - COMPARISON BETWEEN CANOPY AND HERBACEOUS SPECIES

Patterns of decomposition of leaf litter of canopy and understory herbaceous species were compared in the Gottinger Wald, a beech forest on mull soil. Impacts of the meso- and macrofauna on rates of litter disappearance and changes in C and N content of the litter during decomposition were determined by exposing the litter in bags of 45 mum, 1 mm and 3.5 mm mesh for up to 1.4 years. Litter of both herbaceous species (Anemone nemorosa and Mercurialis perennis) decomposed rapidly. Half of the litter had disappeared within 48 days from the 45 mum bags (the treatment that excluded the fauna). Both meso- and macrofauna accelerated decomposition of herb litter. Within 10 days, 50% had disappeared from the 3.5 mm bags (the treatment that permitted access to both meso- and macrofauna). Litter of canopy species decomposed more slowly. Fifty percent of a mixture of ash (Fraxinus excelsior) and maple (Acer spp.) litter had disappeared within 280 and 60 days from the 45-mum and 3.5-mm mesh bags, respectively. Freshly fallen leaves of beech (Fagus sylvatica) and year-old beech litter decomposed much more slowly than litter of herbs or ash-maple. Over 50% of both categories of beech litter remained after 1.4 years, even in the 3.5-mm mesh bags. The fauna had no significant impact on the rate of disappearance of fresh beech leaves, and only the macrofauna accelerated the disappearance of aged beech. Nitrogen content increased, and the C/N ratio decreased, of all categories of litter in the 45-mum mesh bags. The impact of the fauna on nutrient content of each litter category tended to mirror the extent to which the fauna fed on that category. Feeding by the fauna on herbs, ash, maple and aged beech retarded or prevented the increase in N content. This pattern suggests that nitrogen content influenced palatability of the litter to the fauna, although the correlation coefficient between initial nitrogen content and the decomposition coefficient [k (yr-1)] was not statistically significant (r = 0.63, df = 3; 3.5-mm mesh bags). Our findings support the hypothesis that herbaceous litter plays an important role in maintaining high densities of the fauna in beech forests on mull soils. We come to this conclusion because the fauna did not attack fresh beech leaves, but readily consumed herb litter, which previous research has demonstrated comprises a substantial fraction of the litter input to the Gottinger Wald. We discuss the evidence for this hypothesis in light of contradictory evidence from several studies regarding the extent to which animals consume fresh beech litter.