Premise of research. African grass communities are dominated by two distinct functional types: tall, caespitose bunch grasses and short, spreading lawn grasses. Functional type coexistence has been explained by differences in defoliation tolerance, because lawn grasses occur in intensively grazed areas while bunch grasses are less associated with heavy grazing. If different responses to tissue loss explain their distribution, expectations are that biomass production and leaf-level physiology will be negatively impacted in bunch relative to lawn grasses. Methodology. We tested the influence of defoliation on three lawn and three bunch grasses from Tanzania and South Africa by quantifying growth and measuring physiological response of these grasses to simulated herbivory in a glasshouse experiment. Specifically, we measured photosynthesis, transpiration, stomatal conductance, leaf dry matter content (LDMC), specific leaf area (SLA), leaf nitrogen, and leaf pigment concentrations in leaves of bunch and lawn grasses that were clipped or unclipped. Pivotal results. In contrast to our expectations, clipped lawn and bunch grasses did not differ in photosynthesis, leaf nitrogen, or biomass production, and both lawn and bunch grasses upregulated photosynthesis in response to clipping. However, defoliated bunch grasses had higher rates of stomatal conductance and transpiration compared with defoliated lawn grasses. Also, leaf carotenoid concentrations increased in response to clipping for both functional types but much more in bunch than in lawn grasses. An analysis of leaf-level physiological relationships with structural equation modeling showed that lawn and bunch grasses exert control over carbon gain in different ways. In bunch grasses, net carbon gain was associated with leaf-level structural properties (LDMC and SLA) that varied in response to defoliation, while in lawn grasses, increased carbon gain was the result of increased leaf [N] subsequent to defoliation. Concl