Evidence suggests that core temperatures of approximately 40 degrees C can induce fatigue, although this may be confounded by coincident elevations in skin temperatures and maximal cardiovascular strain. In an observational field study to examine core temperature threshold for fatigue, we investigated whether running performance is impaired when rectal temperature (T(re)) is >40 degrees C and skin temperature remains modest. Seventeen competitive runners (7/10 women/men: 8 km best 1,759 +/- 78/1,531 +/- 60 s) completed 8-km track time trials in cool (WBGT approximately 13 degrees C; n = 6), warm (WBGT approximately 27 degrees C; n = 4), or both (n = 7) conditions. T(re), chest skin temperature, and heart rate were logged continuously; elapsed time was recorded every 200 m. Running velocity for T(re) >40 degrees C was compared with that for T(re) <40 degrees C for each runner. Changes in running velocity over the last 600 m were compared between runners with T(re) >40 degrees C and <40 degrees C. Twelve runners achieved T(re) >40.0 degrees C with >or=600 m remaining (range 600-3,400 m). Average running velocity for T(re) <40 degrees C (282 +/- 27 m/min) was not different from that for T(re) >40 degrees C (279 +/- 28 m/min; P = 0.82). There were no differences in running velocity during the final 600 m between runners with final T(re) >40 degrees C or <40 degrees C (P = 0.16). Chest skin temperature ranged from 30 to 34 degrees C, and heart rate was >95% of age-predicted maximum. Our observation that runners were able to sustain running velocity despite T(re) >40 degrees C is evidence against 40 degrees C representing a "critical" core temperature limit to performance.