We determined the effect of a high-fat diet and carbohydrate (CHO) restoration on substrate oxidation and glucose tolerance in 7 competitive ultra-endurance athletes (peak oxygen uptake [V̇O_2peak] 68 ± 1 ml · kg⁻¹ · min⁻¹; mean±SEM). For 6 days, subjects consumed a random order of a high-fat (69% fat; FAT-adapt) or a high-CHO (70% CHO; HCHO) diet, each followed by 1 day of a high-CHO diet. Treatments were separated by an 18-day wash out. Substrate oxidation was determined during submaximal cycling (20 min at 65% V̇O_2peak) prior to and following the 6 day dietary interventions. Fat oxidation at baseline was not different between treatments (17.4 ± 2.1 vs. 16.1 ± 1.3 g · 20 min⁻¹ for FAT-adapt and HCHO, respectively) but increased 34% after 6 days of FAT-adapt (to 23.3 ± 0.9 g · 20 min⁻¹, p < .05) and decreased 30% after HCHO (to 11.3±1.4 g · 20 min⁻¹, p < .05). Glucose tolerance, determined by the area under the plasma [glucose] versus time curve during an oral glucose tolerance (OGTT) test, was similar at baseline (545±21 vs. 520±28 mmol · L⁻¹ · 90 min⁻¹), after 5-d of dietary intervention (563 ± 26 vs. 520 ± 18 mmol · L⁻¹ · 90 min⁻¹) and after 1 d of high-CHO (491 ± 28 vs. 489 ± 22 mmol · L⁻¹ · 90min⁻¹ for FAT- adapt and HCHO, respectively). An index of whole-body insulin sensitivity (S_I 10000/÷fasting [glucose] × fasting [insulin] × mean [glucose] during OGTT × mean [insulin] during OGTT) was similar at baseline (15 ± 2 vs. 17 ± 5 arbitrary units), after 5-d of dietary intervention (15 ± 2 vs. 15 ± 2) and after 24 h of CHO loading (17 ± 3 vs. 18 ± 2 for FAT- adapt and HCHO, respectively). We conclude that despite marked changes in the pattern of substrate oxidation during submaximal exercise, short-term adaptation to a high-fat diet does not alter whole-body glucose tolerance or an index of insulin sensitivity in highly-trained individuals.