Influence of seed size and seedling ecological attributes on shade-tolerance of rain-forest tree species in northern Queensland

1 Growth, leaf anatomy and stomatal responses were examined in seedlings of 12 rainforest tree species in northern Queensland. Plants were kept in a shade house for a period of 15 months at 37%, 10% and 2.5% photosynthetically active radiation (PAR), roughly corresponding to light environments in large gaps, small gaps, and forest understorey, respectively. The species were chosen to represent a wide array of taxa, ecological, and morphological characteristics, and were assigned regeneration niche (RN) index values based on observations of life history and spatial distribution of their seedlings and saplings in various forest microsites. 2 Growth, biomass allocation patterns, leaf anatomy and stomatal responses varied appreciably among species, as well as in response to shading. All species showed reduced growth with decreasing light intensity. Plants grown at 37% and 10% PAR exhibited morphological and physiological responses typical of plants found at forest edges and in large gaps: i.e. low leaf area ratio (LAR), high specific leaf weight (SLW), high root-shoot ratio (RSR), high relative growth rate (RGR), high net assimilation rate (NAR), high leaf conductance, and high chlorophyll a/b ratio. Those grown at 2.5% PAR showed the opposite trends. Comparing species within each light level, NAR and LAR showed positive correlation with RGR, but pooled data for all light levels suggests that the contribution of LAR and NAR to RGR are in opposite directions, depending on light conditions. 3 Sensitivity of seedling growth to shading was significantly correlated with mean seed reserved weight of the species: small-seeded species were more affected than large-seeded ones. Relationship between seed reserve weight and species RN index value, though linear, was weak and nonsignificant. RGR was significantly correlated with seed reserve weight at 37% and 10% PAR but not at 2.5% PAR, while LAR showed the opposite trend. 4 Plasticity of most parameters examined appeared highest in species that regenerate mostly in large gaps (> 200 m squared) and at forest edges, and least in species that regenerate mostly in forest understorey and small gaps (< 50 m squared). However, ordination of the study species, using seed size and various shade tolerance criteria for the seedlings, yielded a continuum rather than discrete groups. The first axis, which explained most of the variation (65.6%), was influenced by NAR, RGR, stomatal conductance and total biomass, and was strongly correlated with species regeneration status.