Poor exercise tolerance (the inability to sustain physical tasks) is indicative of morbidity, mortality, and low quality of life. The aim of this thesis was to elucidate the determinants of exercise tolerance, which remain incompletely understood. The primary outcome measure was maximal voluntary isokinetic power (Piso), from which locomotor neuromuscular fatigue (exercise-induced reduction in Piso) and 'power reserve' (difference between task power and Piso at intolerance) were quantified. First, a potential association between power reserve magnitude and severity of inspiratory muscle fatigue following short- and long-duration exercise was investigated. Inspiratory rib cage muscle and diaphragm fatigue were inferred from reductions in oesophageal (Poestw) and gastric (Pgatw) twitch pressures, respectively, in response to cervical magnetic stimulation. Global inspiratory muscle fatigue was assessed via transdiaphragmatic pressure (Pditw; difference between Pgatw and Poestw). Inter-trial differences (Δ) in power reserve were not related to ΔPditw (r=-0.47; P=0.17) or ΔPgatw (r=0.53; P=0.11), but were negatively associated with ΔPoestw (r=0.76, P < 0.01), suggesting that a larger power reserve results from a greater contribution to intolerance from rib cage muscle fatigue. Next, whether task power influences locomotor neuromuscular fatigue was assessed. Compared to constant-power exercise, intermittent exercise allows higher powers to be performed without increasing intensity (metabolic stress). Motor unit recruitment and non-oxidative energy provision were greater during intermittent than intensity-matched constant-power exercise (both P < 0.05), but the reduction in Piso was not different between protocols (-13±6% vs. -14±6%, respectively; P=0.69). Finally, adaptations to intensity-matched intermittent and constant-power protocols were investigated. Preliminary findings indicated that traininginduced increases in Piso and V̇O₂peak were not different between protocols (both P > 0.05). However, there was a greater increase in lactate threshold (P=0.02) and 'large' additional increase in exercise tolerance (effect size=2.16) following intermittent compared with constant-power training. Overall, the investigations presented herein may aid optimisation of exercise training strategies to improve exercise tolerance.