1.Connectivity is a pivotal feature of landscapes that affects the structure of populations and the functioning of ecosystems. It is also a key consideration in conservation planning. But the potential functional effects of landscape connectivity are rarely evaluated in a conservation context. 2.The removal of algae by herbivorous fish is a key ecological function on coral reefs that promotes coral growth and recruitment. Many reef herbivores are harvested and some use other habitats (like mangroves) as nurseries or feeding areas. Thus, the effects of habitat connectivity and marine reserves can jointly promote herbivore populations on coral reefs thereby influencing reef health. 3.We used a coral reef seascape in eastern Australia to test whether seascape connectivity and reserves influence herbivory. We measured herbivore abundance and rates of herbivory (on turf algae and macroalgae) on reefs that differed in both their level of connectivity to adjacent mangrove habitats, and their level of protection from fishing. 4.Reserves enhanced the biomass of herbivorous fish on coral reefs in all seascape settings and promoted consumption of turf algae. Consumption of turf algae was correlated with the biomass of surgeonfish that are exploited outside reserves. By contrast, both reserve status and connectivity influenced herbivory on macroalgae. Consumption of macroalgae was greatest on fished reefs that were far from mangroves, and was not strongly correlated with any fish species. 5.Our findings demonstrate that landscape connectivity and reserve status can jointly affect the functioning of ecosystems. Moreover, we show that reserve and connectivity effects can differ markedly depending on resource type (in this case turf algae vs macroalgae). The effectiveness of conservation initiatives will therefore depend on our ability to understand how these multiple interactive effects structure the distribution of ecological functions. These findings have wider implications for the spatial conservation of heterogeneous environments and strengthen the case that the impact of conservation on ecosystem functioning is contingent on how reserves are positioned in landscapes.