This study was undertaken to assess the reproducibility of a variable intensitycycling protocol using subjects of varying abilities, under warm humid conditions. Eleven subjects (Age 21.4+/-2.6 years; VO2peak 3.30+/-0.9 l x min(-1); peak power 322.8+/-86.3 W; mean+/-SD) performed a 60 min cycling trial punctuated with six one-min "all-out" sprints at 10-min intervals on three occasions 5-14 days apart. Ambient temperature and relative humidity were set at 33+/-0.7 degrees C and 63+/-2.0%, respectively. Subjects used their own bicycle mounted to an electromagnetic trainer and were only permitted to monitor elapsed time and heart rate. Repeatability was assessed using the limits of agreement which were best between trials 2 and 3 where the distance cycled was -0.54 km below and 1.34 km above the distance cycled for trial 2. The co-efficient of variation (CV) for distance for three trials was 3.58%. For trials 1 and 2 the CV was 3.54% (r = 0.97, p< 0.001) decreasing to 1.34% (r = 0.99, p< 0.001) for trials 2 and 3. The intra-class correlation for three trials was 0.93. Distance for trial 1 (26.3+/-5.0 km; p< 0.05) was less than trials 2 (27.7+/-5.7 km) and 3 (28.1+/-5.6 km). It was concluded that repeatability for this performance protocol with cyclists of varying abilities In warm humid conditions was acceptable given at least one familiarisation trial. However, it is not yet known whether other protocols designed for moderate environments are applicable to less favorable conditions. Further studies are needed before results of treatment effects under differing ambient conditions can be fully understood and assigned appropriate significance.