The effect of an active warm-up on maximal voluntary contraction (MVC), maximal instantaneous power output and surface EMG (sEMG) parameters was studied. Eight volunteers [mean (SD) 22 (4) years] completed two trials on the same day, one control (CO), and the other preceded by an active cycling warm-up (WU) at 70% ventilatory threshold determined by the ventilatory equivalent method. Quadriceps muscle temperature, measured from vastus lateralis with a flexible thermistor, was 33.8 (0.4) degrees C in CO compared to 36.8 (0.5) degrees C in WU ( P<0.05). Aural temperature, measured by an infrared tympanic thermistor, was not different between conditions. Experimental trials consisted of three knee-extension maximum voluntary contractions at a 90 degree angle with simultaneous recording of sEMG from the vastus lateralis, followed by three squat jumps performed on a force platform. SEMGs were analysed in the frequency domain as median frequency (MDF) and in the time domain as root mean square (RMS). MDF was 59.2 (14.1) Hz in CO compared to 67.2 (11.8) Hz in WU ( P<0.05), while RMS was higher in CO compared to WU [0.65 (0.28) mV vs. 0.56 (0.19) mV; P<0.05]. MVC was not different [465.7 (107.6) N vs. 490.1 (117.2) N], whilst instantaneous power output during the squat jump was significantly higher in the WU trial [3324 (866) W vs. 3569 (919) W; P<0.05]. These data show MDF to be altered with an active warm-up, which would relate to a greater conduction velocity. This may translate into faster activation of the muscle fibres, thus partly explaining the increase in power output.