Summary Geographic variation in abiotic factors and species interactions is widespread and is hypothesized to generate concomitant patterns of species trait variation. For example, higher rates of herbivory at lower latitudes are thought to select for increased plant defences, although latitudinal variation in defences may also be influenced directly by abiotic factors and indirectly by predators and parasitoids reducing herbivore pressure. We measured defences of the herb Ruellia nudiflora among 30 populations spanning a latitudinal gradient from northern Yucatan to southern Belize that vary substantially in leaf herbivory (fourfold), seed herbivory (25‐fold) and seed herbivore parasitism (14‐fold). These surveyed populations span one‐third of the species’ latitudinal distribution (5° of latitude), the entire precipitation gradient of its distribution, and one‐third of the temperature gradient of its distribution. Our prior work showed that leaf herbivory decreased with latitude and that seed herbivory increased with latitude. Here, we measured leaf trichome density and leaf and seed phenolics and tested whether latitudinal variation in climate, herbivory and parasitism explained latitudinal variation in these defensive traits. Patterns of variation in leaf trichomes fully supported predictions, with trichome density increasing with a parallel increase in herbivory towards lower latitudes. While seed phenolics were positively associated with herbivory, and seed herbivory tended to increase with latitude, the predicted (positive) association between latitude and defence was not detectable. There was no detectable association between parasitoids and seed defences. In addition, the association between leaf herbivory and phenolics was weak, and leaf phenolics were not associated with latitude. Importantly, variation in the abiotic environment was associated with plant defence, indicating that abiotic factors can play a major role in shaping plant defences, independently of herbivory. Synthesis. Latitudinal variation in abiotic factors may drive concomitant patterns of variation in plant defences, independently of herbivory. Collectively, these findings highlight the need for assessing geographic variation in plant defences from a multi‐factorial perspective, testing for the simultaneous influence of biotic and abiotic factors.