Explaining species richness patterns is a central issue in ecology, but a general explanation remain elusive. Environmental conditions have been proposed to be important drivers of these patterns, but we still need to better understand the relative contribution of environmental factors. Here we aim at testing two environmental hypotheses for richness gradients: energy availability and environmental seasonality using diversity patterns of the family Leguminosae across Mexico. We compiled a database of 502 species and 32,962 records. After dividing Mexico into 100x100 km grid cells, we constructed a map of variation in species richness that accounts for heterogeneity in sampling effort. We found, the cells with the highest species richness of legumes are in the Neotropical region of Pacific coastal and southern Mexico, where the legume family dominates the tropical rain forests and seasonally dry tropical forests. Regression models show that energy and seasonality predictors can explain 25% and 49% of the variation in richness, respectively. Spatial autocorrelation analysis showed that richness has a strong spatial structure, but that most of this structure disappears when both energy and seasonality are used to account for richness gradient. Our study demonstrates multiple environmental conditions contribute complementarily to explain diversity gradients. Moreover, it shows that in some regions, environmental seasonality can be more important than energy availability, contradicting studies at coarser spatial scales. More basic taxonomic and floristic work is needed to help describe patterns of diversity for many groups to allow for testing the underlying mechanisms responsible for diversity gradients.