I incubated eggs of the smooth softshell turtle (Apalone mutica) at 260, 280, and 30 C in 2 yr to investigate the physiological significance of temperature on devel- oping embryos and hatchlings. In particular, I tested the hypothesis that tempera- ture has a direct effect on physiology of developing embryos independent of the hydric environment, specifically as expressed through its effects on hatchling size and locomotor performance. Hatching success was lowest at 260C but did not dif- fer significantly among treatments or between years. Survivorship of turtles to 2 wk posthatching varied significantly among temperature treatments, because rel- atively few individuals from 260C survived Duration of incubation was negatively related to incubation temperature. However, sex determination of hatch- ling turtles was independent of incubation temperature. All but one measure of body size of turtles varied positively with incubation temperature in both years, which suggests that temperature directly affected embryonic growth. Body size of hatchlings also differed significantly among clutches, even after accounting for initial egg mass, which implies maternal or genetic variation in embryonic de- velopment. Measures of locomotor performance (running and swimming speed) were significantly influenced by incubation temperature in both years, such that warmer temperatures produced faster hatchlings. In 1989, but not in 1990, larger turtles ran and swam faster than smaller individuals. Clutch exhibited a significant influence on running and swimming speed in both years, which sug- gests maternal or genetic effects underlying performance. Both measures of loco- motor performance were positively correlated in both years, and the correlations were significant, which indicates that faster runners were also faster swimmers. Running and swimming speed were highly repeatable within the 1-d trials in both years and exhibited significant heritable variation. These findings highlight the pervasive influence of environmental and genetic factors on the physiological ecology of embryonic and hatchling turtles and indicate that growth and locomo- tor performance may possess the genetic variation necessary to evolve in response to selection pressures in nature. Experimental field studies of the ecological and evolutionary consequences of variation in these traits would prove valuable in elucidating the pervasive impact of environmental conditions during embryonic development.