The Amazon, the world's largest tropical rainforest, exhibits great complexity in patterns of species diversity and distribution. Understanding these patterns involves investigating an intricate network of ecological and historical processes, which include species' relationships with past and present environments and requires methodologically complex stages. These range from systematic field sampling, with inventories representative of the local fauna and an ideal sampling effort, to covering an extensive geographical breadth necessary to address large-scale questions. Macroecological studies frequently use species distribution data (range maps) or data compiled from multiple studies. However, each of these approaches has advantages and disadvantages, such as the errors of commission and omission associated with range maps, and the lack of standardised sampling effort between different studies. Therefore, understanding the environmental, spatial, and historical determinants of species diversity requires an integrated approach. In this context, this thesis is structured in three chapters, each focusing on a specific dimension of this problem: methodological, environmental, and historical. In Chapter 1, we propose a minimum sampling effort to representatively sample understorey avifauna, establishing a standardised protocol based on a broad compilation of mist-net capture data. We suggest capturing at least 129 individuals per site (and ideally 179) and using at least 830 net-hours to accumulate this number of individuals. For species-richer areas (>60 species), the minimum sample size should be at least 234 individuals. Our results emphasise the importance of standardising sampling protocols to increase comparability between studies, optimising time, personnel, resource allocation, and logistics in future field studies. In Chapter 2, we assess the methodological and environmental determinants of local species richness and its variation across spatial gradients. Sampling effort plays a fundamental role in representing local diversity, particularly in communities with many uncommon and rare species. Despite pronounced local variation in species richness, we found no evidence of a longitudinal gradient in understorey species richness towards the Andes. Instead, our results reiterate a latitudinal pattern, with species richness significantly higher on the southern margin of the Amazon River. Our findings suggest that the tropical forest understorey represents an environment with a degree of microclimatic stability, which helps maintain the modest variation in species richness across large spatial scales. Finally, in Chapter 3, we investigate how Amazonian rivers, geographical distance, and geological context influence the taxonomic turnover of understorey birds. Geographical distance was the most important predictor of taxonomic replacement. This change was more pronounced in birds with low dispersal capacity than in those considered good dispersers. The major Amazonian rivers (which delineate the areas of endemism) explained a substantial portion of the variation in taxonomic composition, while smaller rivers (within the areas of endemism) also contributed considerably. Furthermore, taxonomic replacement was greater on more recent geological surfaces. Our results highlight that the effect of distance is not merely a function of geographical separation but emerges from the combined effects of historical divergence, driven by riverine barriers, geological history, and species' dispersal capacity.