Natural regeneration of yew (Taxus baccata L): Microsite, seed or herbivore limitation?

1 This study integrates results of field experiments with historical data to determine the extent and circumstances under which regeneration of English yew (Taxus baccata L.), a long-lived tree, may be seed-, microsite- or herbivore-limited. 2 Native yew populations within Fraxinus-Acer woodlands in two coastal denes, (Castle Eden Dene and Hawthorn Dene) and two limestone gorges (Greta Gorge and Horsleyhope Ravine) all in County Durham, north-east England were selected for study. 3 The coastal denes reflect a unimodal age-distribution with peak regeneration occurring between 150 and 250 years ago. In contrast, the age-distribution of trees in Greta Gorge appeared to show low but relatively constant levels of regeneration over the last 550 years. Too few yews were found to provide comparable data for Horsleyhope Ravine. 4 The sites differed considerably in patterns of recent regeneration but patterns were consistent with the dynamics depicted by the age-distributions. Over 30% of individuals in Greta Gorge were saplings, while in Hawthorn Dene this figure was only 8% and no saplings were found in either of the two remaining sites. Yew seedlings were only found in the coastal denes. Comparison with other tree species in these sites reveal yew to be the only species which suffers consistently poor seedling recruitment. 5 Rates of seed predation were similarly high across all four sites. However, there was significant variation between microsites with seed predation twice as intense beneath shrubs as in the open. 6 Examination of the age-distribution revealed regeneration to become increasingly microsite limited as the yew population developed. Therefore, although sites with the highest rates of seed predation had fewest yew seedlings, these were also the oldest sites and the pattern could be better explained by limitation of regeneration by microsites rather than herbivores. 7 A continuum was found to exist with sites where current regeneration is primarily herbivore-limited (Hawthorn Dene) to sites which were microsite limited (Greta Gorge) with regeneration in intermediate sites (Castle Eden Dene) being limited by both microsites and seed predators. The problems of extrapolating results from shortterm experiments to the dynamics of long-lived plant species are discussed.