The aim of this study was to examine whether ingesting water alone, or dextrose (7.5 g x 100 ml(-1)) with electrolytes, or fructose/corn solids (7.5 g x 100 ml(-1)) (400 ml every 20 min) would reduce the perceived exertion associated with 16 km (3 h) walking/running in the heat compared with that perceived during exercise with no fluid intake. Perceived exertion was assessed at 1-h intervals during exercise. Blood samples, required for analysis of blood glucose, plasma sodium, plasma osmolality and plasma volume, were obtained prior to exercise and at 1-h intervals during the exercise; further samples were obtained 1-h intervals for 3 h following the exercise. Drinking fluids at regular intervals reduced the level of perceived exertion. In the test during which no fluid was ingested, body mass decreased by 4.9 (0.4) kg [mean (SEM)], but decreased less with ingestion of either the dextrose/electrolytes or fructose/corn solids solutions, or water alone [1.3 (0.2) kg, 1.6 (0.3) kg and 2.0 (0.1) kg, respectively]. Plasma volume fell by 17% when taking no fluid, but fell less when ingesting fluids. Blood glucose fell significantly (P < 0.01) when taking no fluid and rose to 8.4 (1.3) mmol x l(-1) (P < 0.001) and 6.8 (1.1) mmol x l(-1) (P < 0.01) with ingestion of the dextrose/electrolytes or fructose/corn solids solutions, respectively. Urine output was greater with ingestion of water than with any of the other drinks. Six subjects experienced fatigue during exercise with no fluid and failed to complete the exercise. These results suggest that fatigue was caused by several interacting factors: a fall in blood glucose and plasma volume, dehydration, and neuroglycopenia. Taking fluids during exercise reduced the strain and the rating of perceived exertion; this was better achieved by ingesting a dextrose/electrolytes solution.