Eight female games players (GP) and eight female endurance athletes (EA) ran intermittently at high-intensity and for prolonged periods in hot (30 degrees C) and moderate (16 degrees C) ambient temperatures. The subjects performed a two-part (A, B) test based on repeated 20-m shuttle runs. Part A comprised 60 m of walking, a maximal 15-m sprint, 60 m of cruising (90% maximal oxygen uptake, VO(2max)) and 60 m of jogging (45% VO(2max)) repeated for 75 min with a 3-min rest every 15 min. Part B involved an exercise and rest pattern of 60-s running at 100% VO(2max) and 60-s rest which was continued until fatigue. Although the GP and EA did not respond differently in terms of distances completed, performance was 25 (SEM 4)% less (main effect trial, P < 0.01) in the hot (HT) compared with the moderate trial (MT). Sprints of 15 m took longer to complete in the heat (main effect, trial, P < 0.01), and sprint performance declined during HT but not MT (interaction, trial x time, P < 0.01). A very high correlation was found between the rate of rise in rectal temperature in HT and the distance completed [GP, r =-0.94, P < 0. 01; EA (n = 7), r = -0.93, P < 0.01]. Blood lactate [La(-) ](b) and plasma ammonia [NH(3)](p1) concentrations were higher for GP than EA, but were similar in HT and MT [La(-) ](b), HT: GP vs EA, 8.0 (SEM 0. 9) vs 4.9 (SEM 1.1) mmol x l(-1); MT: GP vs EA, 8.0 (SEM 1.3) vs 4.4 (SEM 1.2) mmol x l(-1); interaction, group x time, P < 0.01; [NH(3)](p1), HT: GP vs EA, 70.1 (SEM 12.7) vs 43.2 (SEM 6.1) mmol x l(-1); MT: GP vs EA, 76.8 (SEM 8.8) vs 32.5 (SEM 3.8) micromol x l(-1); interaction, group x time, P < 0.01. Ad libitum water consumption was higher in HT [HT: GP vs EA, 18.9 (SEM 2.9) vs 13.5 (SEM 1.7) ml x kg(-1) x h(-1); MT: GP vs EA, 12.7 (SEM 3.7) vs 8.5 (SEM 1.5) ml x kg(-1) x h(-1); main effect, group, n.s.; main effect, trial, P < 0.01]. These results would suggest that elevated body temperature is probably the key factor limiting performance of prolonged, intermittent, high-intensity running when the ambient temperature is high, but not because of its effect on the metabolic responses to exercise.