The aims of this study were to evaluate the effects of caffeine supplementation on sprint cycling performance and to determine if there was a dose-response effect. Using a randomized, double-blind, placebo-controlled design, 17 well-trained men (age: 24 ± 6 years, height: 1.82 ± 0.06 m, and body mass(bm): 82.2 ± 6.9 kg) completed 7 maximal 10-second sprint trials on an electromagnetically braked cycle ergometer. Apart from trial 1 (familiarization), all the trials involved subjects ingesting a gelatine capsule containing either caffeine or placebo (maltodextrin) 1 hour before each sprint. To examine dose-response effects, caffeine doses of 2, 4, 6, 8, and 10 mg·kg bm(-1) were used. There were no significant (p ≥ 0.05) differences in baseline measures of plasma caffeine concentration before each trial (grand mean: 0.14 ± 0.28 μg·ml(-1)). There was, however, a significant supplement × time interaction (p < 0.001), with larger caffeine doses producing higher postsupplementation plasma caffeine levels. In comparison with placebo, caffeine had no significant effect on peak power (p = 0.11), mean power (p = 0.55), or time to peak power (p = 0.17). There was also no significant effect of supplementation on pretrial blood lactate (p = 0.58), but there was a significant time effect (p = 0.001), with blood lactate reducing over the 50 minute postsupplementation rest period from 1.29 ± 0.36 to 1.06 ± 0.33 mmol·L(-1). The results of this study show that caffeine supplementation has no effect on short-duration sprint cycling performance, irrespective of the dosage used.