HERBIVORY AND GAP REGENEMTION IN SOUTHERN BOTTOMLAND HARDWOOD FORESTS

We conducted 2 experiments to examine the influence of herbivory by white-tailed deer (Odocoileus virginianus) and swamp rabbits (Syluilagus aquaticus) on plant regeneration within canopy gaps in southern bottomland hardwood forests. One study was conducted at Moro Bottoms Natural Area in southern Arkansas in a bottomland forest containing windthrow gaps. The other was conducted at the Savannah River Site (SRS), South Carolina in a forest containing gaps created by group selection timber harvest. Herbivore exclosures and control plots were constructed in 30 gaps at Moro Bottoms and 36 gaps at the SRS. General patterns of deer use and effects on vegetation were similar for both sites. Counts of deer fecal pellet groups indicated little use of bottomland forests in spring and summer, with increased use during fall and winter when oak mast was available. Rabbit herbivory was not detected in the SRS gaps. Rabbits were more abundant in southern Arkansas, but apparent use of sampled vegetation in study plots was negligible. Plant abundance and diversity varied among gap sizes at the SRS, though these patterns likely were not influenced by deer herbivory. Dif- ferences in plant species diversity, richness, and number of unique plant species in the Arkansas gaps were not detected among treatments, but grasses and sedges were more abundant in plots available to deer. Landscape context may influence patterns of habitat use, so it may be necessary to index deer densities prior to timber harvest in some locations to assess their potential impacts on regeneration. We conclude that other factors. such as degree of inundation and soil disturbance from timber harvesting, were more important in determining the initial vegetative response than herbivory or gap size on these study areas.