Objectives: To determine the effects of heat stress on ankle proprioception and running gait pattern. Design: Counterbalanced repeated measures. Methods: 12 trained runners performed a proprioception test (active movement discrimination) before and immediately after a 30 min, self-paced treadmill run in HOT (39 °C) and COOL (22 °C) ambient conditions. Velocity was imposed during the first and last minute (70% of maximal aerobic velocity, 13.3 ± 0.8 km h−1) for determination of running mechanics and spring–mass characteristics. Results: Rectal (39.7 ± 0.4 vs. 39.4 ± 0.4 °C), skin (36.3 ± 1.1 vs. 31.8 ± 1.1 °C) and average body (38.3 ± 0.2 vs. 36.4 ± 0.4 °C) temperatures together with heart rate (178 ± 8 vs. 174 ± 6 bpm) and thermal discomfort (6.5 ± 0.5 vs. 4.3 ± 1.3) were all higher at the end of the HOT compared to COOL run (all p < 0.05). Distance covered was lower in HOT than COOL (−5.1 ± 3.6%, p < 0.001). Average error during the proprioception test increased after running in HOT (+11%, p < 0.05) but not in COOL (−2%). There was no significant difference for most segmental and joint angles at heel contact, except for a global increase in pelvis retroversion and decrease in ankle dorsi-flexion angles with time (p < 0.05). Step frequency decreased (−2.5 ± 3.6%) and step length increased (+2.6 ± 3.8%) over time (p < 0.05), independently of condition. Spring–mass characteristics remained unchanged (all p > 0.05). Conclusions: Heat stress exacerbates thermal, cardiovascular and perceptual responses, while running velocity was slower during a 30 min self-paced treadmill run. Heat stress also impairs ankle proprioception during an active movement discrimination task, but it has no influence on gait pattern assessed at a constant, sub-maximal velocity.