Plant species height (mature height), which strongly correlates with other measures of size, is widely considered an important functional trait. Taller species contribute greatly to ecosystem function and are frequently identified as strong competitors. Additionally, important ecological processes like self-thinning scale with plant species size. In spite of the clear functional importance of species size, evidence for size-based plant community assembly rules has been scant. While herbivory can be species specific and there are reasons why herbivores may target larger plant species, it is unknown whether herbivory has size-biased impacts on species composition. In an old-field plant community, we compared plant species composition in herbivore exclusion (caged) and control plots over five years to test whether herbivore activity favors species at one end of the size spectrum. Across 5 years, smaller species were modestly, but consistently, underrepresented in caged plots compared to control plots. This underrepresentation of relatively smaller plant species in caged plots, where seedling herbivory was greatly diminished, was driven by lower small plant species richness and abundance in those plots; the abundance and richness of tall species did not differ across treatments. Our findings demonstrate that reducing herbivore activity in our study community modestly but consistently altered plant species composition, slightly favoring larger species. There was no evidence that larger species are preferentially targeted by herbivores. Some research suggests, although it will require further investigation, that seedling herbivory, regardless of the species eaten, produces vegetation gaps that small plant species are more successful in colonizing.