This study compared the physiological responses and performance following the ingestion of a waxy starch (WS), resistant starch (RS), glucose (GL) and an artificially-sweetened placebo (PL) ingested prior to exercise. Ten college-age, male competitive cyclists completed four experimental protocols consisting of a 30 min isokinetic, self-paced performance ride preceded by 90 min of constant load cycling at 66% VO2max. Thirty min prior to exercise, they ingested 1 g.kg-1 body weight of GL, WS, RS, or PL At rest, GL elicited greater (p < 0.05) serum glucose and insulin responses than all other trials. During exercise, however, serum glucose, insulin, blood C-peptide and glucagon responses were similar among trials. The mean total carbohydrate oxidation rates (CHOox) were higher (p < 0.05) during the GL, WS, and RS trials (2.59 +/- 0.13, 2.49 +/- 0.10, and 2.71 +/- 0.15 g.min-1, respectively) compared to PL (2.35 +/- 0.12 g.min-1). Subjects were able to complete more work (p < 0.05) during the performance ride when they ingested GL (434 +/- 25.2 kj) or WS (428 +/- 22.5 kj) compared to PL (403 +/- 35.1 kj). They also tended to produce more work with RS ingestion (418 +/- 31.4 kj), although this did not reach statistical significance (p < 0.09). These results indicate that preexercise CHO ingestion in the form of starch or glucose maintained higher rates of total carbohydrate oxidation during exercise and provided an ergogenic benefit during self-paced cycling.