The aim of this study was to devise a laboratory-based protocol for a motorized treadmill that was representative of work rates observed during soccer match-play. Selected physiological responses to this soccer-speci® c intermittent exercise protocol were then compared with steady-rate exercise performed at the same average speed. Seven male university soccer players (mean ± s: age 24 ± 2 years, height 1.78 ± 0.1 m, mass 72.2 ± 5.0 kg, V Ç O 2m ax 57.8 ± 4 ml ´kg -1 ´min -1 ) completed a 45-min soccer-speci® c intermittent exercise protocol on a motorized treadmill. They also completed a continuous steady-rate exercise session for an identical period at the same average speed. The physiological responses to the laboratory-based soccer-speci® c protocol were similar to values previously observed for soccer match-play (oxygen consumption approximately 68% of maximum, heart rate 168 ± 10 beats ´min -1 ). No signi® cant diþ erences were observed in oxygen consumption, heart rate, rectal temperature or sweat production rate between the two conditions. Average minute ventilation was greater (P < 0.05) in intermittent exercise (81.3 ± 0.2 l ´min -1 ) than steady-rate exercise (72.4 ± 11.4 l ´min -1 ). The rating of perceived exertion for the session as a whole was 15 ± 2 during soccer-speci® c intermittent exercise and 12 ± 1 for continuous exercise (P < 0.05). The physiological strain associated with the laboratory-based soccer-speci® c intermittent protocol was similar to that associated with 45 min of soccer match-play, based on the variables measured, indicating the relevance of the simulation as a model of match-play work rates. Soccer-speci® c intermittent exercise did not increase the demands placed on the aerobic energy systems compared to continuous exercise performed at the same average speed, although the results indicate that anaerobic energy provision is more important during intermittent than during continuous exercise at the same average speed.