Understanding the phylogenetic structure provides evidence about the importance of the different processes that shape an ecological community. We addressed the following questions: i) What is the intensity of the role of the phylogenetic component on variations in the tree species composition in a large biogeographic transition zone in the Neotropics? ii) Have forest and savanna habitats similar answers to the above questions? We built an ultrametric phylogenetic tree for 2475 tree species and obtained eigenvectors for each site using the PCPS method (Principal Coordinates of Phylogenetic Structure). We used a variation partitioning framework coupled with Moran’s Spectral Randomization with three explanatory tables - a novel methodological approach - from canonical models expecting that the phylogenetic component would better explain the floristic variation than spatial and environmental components. The model containing the selected phylogenetic, spatial and environmental predictors explained 18.98 % of the floristic variation; the explanation provided by the phylogenetic component, after partitioning out the influence of spatial and environmental factors, accounted for 17.61 % of the total variation. The predominance of phylogenetic component was also found at other spatial levels (habitats and vegetation types). Variations in species composition in tree communities are thus strongly influenced by phylogenetic structure, and this pattern is consistent for both savanna and forest habitats. We confirmed that evolutive relationships are a synthetic indicator that can be correlated with patterns of floristic variation. This approach is unprecedented for tropical communities and brings promising contributions not only to understanding beta diversity in biogeographic transitions, but to the most varied ecological groups.

了解系统发育结构可提供有关塑造生态群落的不同过程的重要性的证据。我们解决了以下问题：i）在新热带地区一个大的生物地理过渡带中，系统发育成分对树木物种组成变化的作用的强度是多少？ii）森林和热带稀树草原的栖息地是否对上述问题有相似的答案？我们为2475个树种构建了超系统的系统树，并使用PCPS方法（系统发生结构的主坐标）获得了每个位点的特征向量。我们使用变异划分框架，结合Moran光谱随机化和三个解释表（一种新颖的方法论方法），来自规范模型，期望系统发育成分比空间和环境成分能更好地解释植物区系变异。包含所选系统发生，空间和环境预测因子的模型解释了18.98％的植物区系变异；系统发育成分提供的解释，在排除了空间和环境因素的影响之后，占了总变异的17.61％。在其他空间水平（栖息地和植被类型）上也发现了系统发育成分的优势。因此，树木群落物种组成的变化会受到系统发育结构的强烈影响，这种模式对于热带草原和森林栖息地都是一致的。我们确认进化关系是可以与植物区系变化模式相关的综合指标。对于热带社区来说，这种方法是前所未有的，不仅为了解生物地理转变中的β多样性，而且为最多样化的生态群体带来了可喜的贡献。