The present study investigated whether increasing morning rectal temperatures (T) to evening levels, or increasing morning and evening T to an "optimal" level (38.5°C), resulting in increased muscle temperatures (T), would offset diurnal variation in repeated sprint (RS) performance in a causal manner. Twelve trained males underwent five sessions [age (mean ± SD) 21.0 ± 2.3 years, maximal oxygen consumption (V̇Omax) 60.0 ± 4.4 mL.kg min, height 1.79 ± 0.06 m, body mass 78.2 ± 11.8 kg]. These included control morning (M, 07:30 h) and evening (E, 17:30 h) sessions (5-min warm-up), and three further sessions consisting of a warm-up morning trial (M, in 39-40°C water) until T reached evening levels; two "optimal" trials in the morning and evening (M and E, in 39-40°C water) respectively, until T reached 38.5°C. All sessions included 3 × 3-s task-specific warm-up sprints, thereafter 10 × 3-s RS with 30-s recoveries were performed a non-motorised treadmill. T and T measurements were taken at the start of the protocol and following the warm-up periods. Values for T and T at rest were higher in the evening compared to morning values (0.48°C and 0.69°C, p < 0.0005). RS performance was lower (7.8-8.3%) in the M for distance covered (DC; p = 0.002), average power (AP; p = 0.029) and average velocity (AV; p = 0.002). Increasing T in the morning to evening values or optimal values (38.5°C) did not increase RS performance to evening levels (p = 1.000). However, increasing T in the evening to "optimal" level through a passive warm-up significantly reduced DC (p = 0.008), AP (p < 0.0005) and AV (p = 0.007) to values found in the M condition (6.0-6.9%). Diurnal variation in T and T is not wholly accountable for time-of-day oscillations in RS performance on a non-motorised treadmill; the exact mechanism(s) for a causal link between central temperature and human performance are still unclear and require more research.