Eight well-trained male cyclists were used to determine the influence of carbohydrate feedings on exercise performance and muscle glycogen use. Two days prior to each trial, the subjects performed a 60-min "depletion ride" at 70% VO2max, which was followed by the ingestion of a high carbohydrate diet (approximately 500 g X -1). During the experimental trials, the men performed 2 h of cycling exercise and consumed 150 ml of 1 of 4 solutions at 24-min intervals. The drinks were: H2O (artificially flavored and sweetened); maltodextrin (5 g X 100 ml-1) and fructose (5 g X 100 ml-1); maltodextrin (7.7 g X 100 ml-1) and high fructose corn syrup (2.3 g X 100 ml-1); maltodextrin (3 g X 100 ml-1 and glucose (2 g X 100 ml-1). The amount of work completed during the four trials was not significantly different. Initial glycogen levels were high, and glycogen values were not significantly different at the beginning of exercise or at 90 min (185.35 +/- 3.26 and 91.93 +/- 3.39, respectively). Blood glucose was greater at 60 min in trial maltodextrin and glucose (5.70 +/- 0.36 mmoles X l-1), maltodextrin and high fructose corn syrup (6.05 +/- 0.54), and maltodextrin and fructose (6.03 +/- 0.42) compared to H2O (4.97 +/- 0.35) (P less than 0.05). Blood glucose remained elevated at 90 min during the maltodextrin and fructose and maltodextrin and high fructose corn syrup trials and at 120 min in the maltodextrin and fructose trial. No differences were observed between trials in blood lactate, serum glycerol, respiratory exchange ratio, or the subjects' perception of effort.(ABSTRACT TRUNCATED AT 250 WORDS)