COLONIZATION DYNAMICS OF HERBS AND SHRUBS IN A DISTURBED NORTHERN HARDWOOD FOREST

(1) Modes of recruitment of herbs and shrubs, and the extent to which availability of propagules affected distribution of the various species after disturbance were examined in a field experiment of canopy removal at the Hubbard Brook Experimental Forest, New Hampshire, U.S.A. (2) The most common shrubs after canopy removal were Rubus idaeus and Viburnum alnifolium; the most common herbs were Aster acuminatus, Dennstaedtia punctilobula and Dryopteris spinulosa. All of the first-year recruitment of Rubus occurred through germination of buried seeds. In contrast, first-year recruitment of the other four species occurred through survival of pre-existing stems or expansion of pre-existing patches, or both. Recruitment of Viburnum, Dennstaedtia and Dryopteris during the second and third years of regrowth continued to be strictly vegetative, whilst recruitment of Aster occurred through newly dispersed seeds as well as rhizomatous growth. (3) The spatial distribution of herbs and shrubs after overstorey removal was strongly related to the location of predecessors before removal. During the first year of regrowth, the locations of almost all stems of all species except Rubus were associated with presence of pre-disturbance individuals on that site. (4) Collections from seed traps indicate that seed production in the site with overstorey removal was higher than in the surrounding, intact forest for most species. Some unoccupied sites were colonized by these seeds during the second and third years of regrowth. (5) The overall distribution of woodland species such as Viburnum alnifolium, Dryopteris spinulosa and Lycopodium lucidulum was unaffected by overstorey removal. After three years of regrowth, the spatial distribution of these species was entirely restricted to locations where predecessors grew in the pre-disturbance forest. (6) Life-history traits of herbs and shrubs at the Hubbard Brook Experimental Forest closely corresponded to Grime's 'competitive', 'stress-tolerant' and 'ruderal' strategies. It appears that ruderals obscured but did not affect spatial patterns of stress-tolerators and competitors that were present in the forest before overstorey removal. This suggests that, in the absence of physical destruction to existing herbs and shrubs in the forest, 'disturbance' has, in the short term at least, little effect on spatial patterns of some understorey species.