Following a 2.0 +/- 0.1% body mass loss induced by intermittent exercise in the heat, seven male volunteers ingested either a carbohydrate-electrolyte solution (CE) or skimmed milk (M) in a volume equal to 150% of body mass loss. At the end of the 3 h recovery period, subjects were essentially in positive fluid balance on trial M (191 +/- 162 mL), and euhydrated on trial CE (-135 +/- 392 mL) despite being in negative sodium balance on both trials and negative potassium balance on trial CE. This difference of 326 +/- 354 mL or 0.4% body mass approached significance (P = 0.051). Subjects ingested 137 +/- 15 and 113 +/- 12 g of CHO during the CE and M trials, respectively, as well as 75 +/- 8 g of protein during the M trial. At the end of the 3 h recovery period, an exercise capacity test was completed at 61% VO(2peak) in warm (35.3 +/- 0.5 degrees C), humid (63 +/- 2%) conditions. HR (P = 0.020) and rectal temperature (P = 0.045) were higher on trial M, but no difference in exercise time to exhaustion was observed between trials (39.6 +/- 7.3 min vs. 39.7 +/- 8.1 min on trials CE and M, respectively). The results of the present study suggest that milk can be an effective post-exercise rehydration drink, with subjects remaining in net positive fluid balance throughout the recovery period. Despite the effect on fluid retention, exercise capacity was not different between skimmed milk and a commercially available carbohydrate-electrolyte drink 4 h following exercise/heat-induced body mass loss.