Six-week-old half-sib seedlings of Ulmus americana L. were subjected to different amounts of flexure daily for 3 weeks under controlled greenhouse conditions. The daily flexure treatments were: no flexing in a staked stem, minimal flexing in a non-staked stem, and five, 10, 20, 40, and 80 flexures. Seedling height and diameter growth and average leaf area were determined before and after the treatments. The ratio of the change in height growth (DeltaH) to the change in diameter growth (DeltaD; (DeltaH:DeltaD)) before and after the 3-week treatments were calculated. At the end of the 3-week experiment, staked seedlings were significantly taller and had smaller stem diameters than all of the flexed seedlings. Height growth tended to decrease exponentially with increased flexure, with significant differences between the extremes of treatment. All of the flexure treatments significantly increased stem diameter compared to staked seedlings. The DeltaH:DeltaD ratio exhibited an exponential function in response to increased flexure. Average leaf area decreased with increased flexure, and seedlings in the 40x and 80x flexure treatments had significantly less leaf area than seedlings in all of the other treatments. These data are similar to the dose responses previously observed in herbaceous species. The finding that trees exhibit greater sensitivity to low doses of flexure than to high doses of flexure indicates that slight exposure to wind may result in a large initial alteration in stem morphology, producing a thigmomorphogenetic effect. Trees will continue to respond to increasing amounts of mechanical stress, but at an exponentially declining rate. Declining leaf areas in response to increasing amounts of mechanical stress may result in a decrease in available photosynthate, resulting in a tree of smaller stature compared to trees exposed to lower amounts of mechanical loading.