The aim of the present study was to determine the effects of 40 km of cycling on the biomechanical and cardiorespiratory responses measured during the running segment of a classic triathlon, with particular emphasis on the time course of these responses. Seven male triathletes underwent four successive laboratory trials: (1) 40 km of cycling followed by a 10-km triathlon run (TR), (2) a 10-km control run (CR) at the same speed as TR, (3) an incremental treadmill test, and (4) an incremental cycle test. The following ventilatory data were collected every minute using an automated breath-by-breath system: pulmonary ventilation VE, l x min[-1]), oxygen uptake (VO2, ml x min(-1) x kg[-1]), carbon dioxide output (ml x min[-1]), respiratory equivalents for oxygen (VE/VO2) and carbon dioxide (VE/VCO2), respiratory exchange ratio (R) respiratory frequency (f, breaths x min[-1]), and tidal volume (ml). Heart rate (HR, beats x min[-1]) was monitored using a telemetric system. Biomechanical variables included stride length (SL) and stride frequency (SF) recorded on a video tape. The results showed that the following variables were significantly higher (analysis of variance, P < 0.05) for TR than for CR: VO2 [51.7 (3.4) vs 48.3 (3.9) ml x kg(-1) x min(-1), respectively], VE [100.4 (1.4) l x min(-1) vs 84.4 (7.0) l x min(-1)], VE/VO2 [24.2 (2.6) vs 21.5 (2.7)] VE/VCO2 [25.2 (2.6) vs 22.4 (2.6)], f[55.8 (11.6) vs 49.0 (12.4) breaths x min(-1)] and HR [175 (7) vs 168 (9) beats x min(-1)]. Moreover, the time needed to reach steady-state was shorter for HR and VO2 (1 min and 2 min, respectively) and longer for VE (7 min). In contrast, the biomechanical parameters, i.e. SL and SF, remained unchanged throughout TR versus CR. We conclude that the first minutes of the run segment after cycling in an experimental triathlon were specific in terms of VO2 and cardiorespiratory variables, and nonspecific in terms of biomechanical variables.