The geographic distribution of seed-dispersal mutualisms in North America

It is often difficult to determine the relative importance of different sorts of species interactions in shaping community structure or how communities function because we lack basic information on patterns of occurrence of biotic interactions. Here we determine the geographic distribution of seed-dispersal mutualisms across North America, and then test the hypotheses that those patterns are correlated with mean annual precipitation and latitude. We analyze the floras of 197 sites across North America to identify which species of native seed plants are dispersed by animals in one of three types of plant-animal mutualisms: frugivory, scatter hoarding, and ants. We identified 1655 plant species that are involved in these seed-dispersal mutualisms. 16.5 ± 6.5% of all seed plants across North America are dispersed by animals; 10.0 ± 4.2% by frugivores, 3.7 ± 1.7% by scatter hoarders, and 3.9 ± 2.2% by ants. Secondary dispersal by a different mode and vector (e.g., wind dispersal followed by scatter hoarding, or ballistic dispersal followed by myrmecochory) occurred in 16.8% of all plant mutualist records. Although each mode of dispersal showed a different pattern, the prevalence of seed-dispersal mutualisms increased with mean annual precipitation. The prevalence of seed dispersal by frugivory and scatter-hoarding decreased with increasing latitude. The prevalence of seed-dispersal interactions varies dramatically across North America. The center of greatest diversity for all three types of seed-dispersal mutualisms is the eastern United States, roughly coincident with the eastern deciduous forest. Knowing the distribution of species interactions improves our understanding of how the structure and functioning of communities varies across environmental gradients.     

Phosphorus conservation during post‐fire regeneration in a Chilean temperate rainforest

Nitrogen and phosphorus are the main elements limiting net primary production in terrestrial ecosystems. When growing in nutrient‐poor soils, plants develop physiological mechanisms to conserve nutrients, such as reabsorbing elements from senescing foliage (i.e. nutrient retranslocation). We investigated the changes in soil N and P in post‐fire succession in temperate rainforests of southern Chile. In this area, forest recovery often leads to spatially scattered, discrete regeneration with patches varying in age, area, species richness and tree cover, representing different degrees of recovery from disturbance. We hypothesized that soil nutrient concentrations should differ among tree regenerating patches depending on the progress of forest regeneration and that nutrient resorption should increase over time as colonizing trees respond to limited soil nutrients. To evaluate these hypotheses, we sampled 40 regeneration patches in an area of 5 ha, spanning a broad range of vegetation complexity, and collected soil, tree foliage and litter samples to determine N and P concentrations. Nutrient concentrations in leaf litter were interpreted as nutrient resorption proficiency. We found that soil P was negatively correlated with all the indicators of successional progress, whereas total soil N was independent of the successional progress. Foliar N and P were unrelated to soil nutrient concentrations; however, litter N was negatively related to soil N, and litter P was positively related with soil P. Finally, foliar N:P ratios ranged from 16 to 25, which suggests that P limitation can hamper post‐fire regeneration. We provide evidence that after human‐induced fires, succession in temperate forests of Chile can become nutrient limited and that high nutrient retranslocation is a key nutrient conservation strategy for regenerating tree communities.