The climatic variability hypothesis (CVH) states that species are geographically more widespread at higher latitudes because individuals have a broader range of physiological tolerance or phenotypic flexibility as latitude and climatic variability increase. However, it remains unclear to what extent climatic variability or latitude, acting on the phenotype, account for any observed geographical gradient in mean range size. In this study, we analyzed the physiological flexibility within the CVH framework by using an intraspecific population experimental approach. We tested for a positive relationship between digestive-tract flexibility (i.e., morphology and enzyme activities) and latitude and climatic and natural diet variability in populations of rufous-collared sparrows (Zonotrichia capensis) captured in desert (27°S), Mediterranean (33°S), and cold-temperate (41°S) sites in Chile. In accordance with the CVH, we observed a positive relationship between the magnitude of digestive-tract flexibility and environmental variability but not latitude. The greatest digestive flexibility was observed in birds at middle latitudes, which experience the most environmental variability (a Mediterranean climate), whereas individuals from the most stable climates (desert and cold-temperate) exhibited little or no digestive-tract flexibility in response to experimental diets. Our findings support the idea that latitudinal gradients in geographical ranges may be strongly affected by the action of regional features, which makes it difficult to find general patterns in the distribution of species.