Flowers function as billboards, using signals such as color and scent to attract pollinators. Flowers with more appealing signals are more likely to be visited by pollinators, resulting in higher rates of pollination and reproduction. Consequently, pollinator preferences are the primary force shaping the evolution of floral traits, although other ecological influences also play a role. While the evolutionary impact of insects and birds is well documented, much less is known about the role of bats, one of the most distinctive pollinator groups. To address this knowledge gap, our project will conduct extensive fieldwork in the Caatinga, a vast Neotropical dry forest notable for its many chiropterophilous, or bat-pollinated, plants. We will assess the intraspecific variation of floral signals in different species and populations. By connecting measures of floral variation with reproductive outcomes, we will construct phenotypic selection models to evaluate the extent and direction of evolutionary changes in these signals. Finally, we will directly investigate the mechanisms underlying patterns of selection by conducting behavioral experiments with bats in flight tents.