In this brief report, the relative importance of different climatic variables on structuring the spatial richness patterns of endemic tree species are quantified for Western Ghats of India, a biodiversity hotspot of the world. Path analysis and spatial multiple regression analysis are employed to measure the relative impacts of different climatic variables on the richness of tree species. Four climatic predictors, annual mean temperature (TEM), annual mean precipitation (PREC), annual evapotranspiration (AET), and elevation range (TOPO), are used to represent different perspectives of climatic variability. TEM represents temperature tolerance, PREC water availability, AET energy and TOPO habitat heterogeneity. Results show that, without removing spatial autocorrelation, AET is the most influential variable. However, the effect of AET was detected to be an artifact of spatial autocorrelation and after controlling such a misleading effect using simultaneous autoregressive models, the effects of other variables became evident and important in explaining the richness of endemic trees. All the statistical models consistently identified that TOPO and PREC are the equally influential variables for structuring richness patterns of endemic trees in Western Ghats. In conclusion, habitat heterogeneity and water availability are the most important predictors of richness in tropical endemic tree communities in Western Ghats.