The spatial complexity in plantherbivore dynamics has been identified as an area requiring more detailed studies1,2. Among the key attempts to explain relationships between plant diversity and herbivore populations, from the herbivores’ point of view, Elton’s ‘enemies hypothesis’ states that less diverse plant communities undergo greater herbivory due to presence of fewer predators in contrast to complex communities3. A corresponding ‘resource concentration hypothesis’ argues that specialist herbivores get attracted to and stay longer in patches with high host plant density4,5. Conversely, Feeny’s ‘plant apparency theory’ suggests that evolution of plant defence mechanisms and their allocation are a function of risk of discovery by herbivores, cost of defence and value of that plant part3,5,7,8. The ‘resource availability hypothesis’ suggests that defensive capabilities of plants are determined by their growth rate, photosynthetic capabilities and nutrient availability2. Both of these are based on resource allocation from a plant’s point of view. Subsequent studies suggest that soil characteristics, nutrient availability, light conditions and community composition play a significant role in determining the extent and direction of herbivory1,9,10. Several studies and meta-analyses have examined various hypotheses on herbivory2,11,12; however, few studies have empirically tested them in tropical evergreen forests of India. Studies in a gradient of disturbance where plants and herbivores respond differently could result in a better understanding of the effects of conversion of primary forests on plant communities through changing patterns of herbivory.