This dataset contains measurements of herbivory and the potential controls on herbivory for nitrogen-fixing and non-fixing trees in a mature tropical forest of Panama. Data include herbivory measures on 1,626 leaves from 350 seedlings belonging to 43 species, 23 nitrogen-fixing species, 20 non-fixing species. Herbivory metrics are presented at the leaf and seedling level. The data also includes leaf chemical and physical traits hypothesized to influence herbivory at the species level, and seedling-level traits such as stem length and growth rate. Data were collected in 2017 by measuring leaf area missing on seedlings in the BCI 50 ha plot seedling census in order to determine whether fixers have higher rates of herbivory than non-fixers, and what traits govern herbivory. Data were collected by W. Barker, S.J. Wright, L. Comita, B. Sedio and colleagues. Funders of research generating the data include the Natural Environment Research Council, U.S. National Science Foundation, Leverhulme Trust, British Council, Society of Experimental Biology Company of Biologists, University of Leeds Priestly International Centre for Climate and Smithsonian Tropical Research Institute