Field exclosure studies have shown that mammalian browsers such as white-tailed deer (Odocoileusvirginianus) can have pervasive effects on forest dynamics in eastern North America. Direct experimental tests of the effects of browsing on growth and survival of a wide range of tree species, however, have yielded conflicting results. This study was designed to assess the effects of variation in the frequency, seasonal timing, and intensity of browsing (simulated by mechanical clipping) on the growth and mortality of three of the major tree species of the Hudson Valley, New York. The clipping treatments were applied to seedlings grown under two different light regimes (full sun and 8% of full sun) to examine seedling responses under different levels of shade-induced carbon stress. Our results demonstrate that even 2 successive years of heavy winter clipping (75% of new shoot growth removed) has little immediate effect on growth or survival of any of the three species. It is possible that winter browsing only has significant negative effects when seedlings are browsed repeatedly over long periods of time. However, comparable levels of summer browsing for only 2 years significantly reduced both growth and survival of all three species. While most natural browsing occurs in the dormant season, our results suggest that it is the less frequent browsing during late spring and early summer that has the greatest immediate effect on tree seedlings. Shading reduced growth and increased mortality in all three species; however, there was only a limited interaction between light level and the simulated browsing treatments. The effects of browsing on survival were similar in all three species; however, the effects of browsing on cumulative height and annual growth varied enough among the species to suggest that browsing could cause significant variation among these species in their rate of invasion in old fields and rights of way, and their rate of regeneration following logging or disturbance of forests.