The primary aim of this thesis was to develop a better understanding of the physiology and perceptual responses associated with the performance of continuous (CE) and intermittent exercise (IE) in patients with moderate chronic obstructive pulmonary disease (COPD). A secondary aim was to examine factors that could potentially limit exercise tolerance in COPD patients, particularly in relation to the dynamics of the cardiovascular system and muscle metabolism. The results of the four studies conducted to achieve these aims are presented in this thesis. In Study 1, the physiological, metabolic and perceptual responses to an acute bout of IE and CE were examined in 10 individuals with moderate COPD. Each subject completed an incremental exercise test to exhaustion on a cycle ergometer. Subjects then performed IE (1 min exercise: 1 min rest ratio) and CE tests at 70% of peak power in random order on separate days. Gas exchange, heart rate, plasma lactate concentration, ratings of breathlessness, inspiratory capacity and the total amount of work completed were measured during each exercise test. Subjects were able to complete a significantly greater amount of work during IE (71 ± 32 kJ) compared with CE (31 ± 24 kJ). Intermittent exercise was associated with significantly lower values for oxygen uptake, expired ventilation and plasma lactate concentration when compared with CE. Subjects also reported a significantly lower rating of breathlessness during IE compared to CE. The degree of dynamic lung hyperinflation (change in end-expiratory lung volume) was lower during IE (0.23 ± 0.07 L) than during CE (0.52 ± 0.13 L). The results suggest that IE may be superior to CE as a mode of training for patients with COPD. The greater amount of total work performed and the lower measured physiological responses attained with intermittent exercise could potentially allow greater training adaptations to be achieved in individuals with more limited lung function. The purpose of Study 2 was to compare the adaptations to 8 wk of supervised intermittent and continuous cycle ergometry training, performed at the same relative intensity and matched for total work completed, in patients with COPD. Nineteen subjects with moderate COPD were stratified according to age, gender, and pulmonary function, and then randomly assigned to either an IE (1 min exercise: 1 min rest ratio) or CE training group. Subjects trained 3 d per week for 8 wk and completed 30 min of exercise. Initial training intensity, i.e., the power output applied during the CE bouts and during the exercise interval of the IE bouts, was determined as 50% of the peak power output achieved during incremental exercise and was increased by 5% each week after 2 wk of training. The total amount of work performed was not significantly different (P=0.74) between the CE (750 ± 90 kJ) and IE (707 ± 92 kJ) groups. The subjects who performed IE (N=9) experienced significantly lower levels of perceived breathlessness and lower limb fatigue during the exercise-training bouts than the group who performed CE (N=10). However, exercise capacity (peak oxygen uptake) and exercise tolerance (peak power output and 6-min walk distance) improved to a similar extent in both training groups. During submaximal constant-load exercise, the improved (faster) phase II oxygen uptake kinetic response with training was independent of exercise mode. Furthermore, training-induced reductions in submaximal exercise heart rate, carbon dioxide output, expired ventilation and blood lactate concentrations were not different between the two training modes. Exercise training also resulted in an equivalent reduction for both training modes in the degree of dynamic hyperinflation observed during incremental exercise. Thus, when total work performed and relative intensity were the same for both training modes, 8 wk of CE or IE training resulted in similar functional improvements and physiological adaptations in patients with moderate COPD. Study 3 examined the relationship between exercise capacity (peak oxygen uptake) and lower limb vasodilatory capacity in 9 patients with moderate COPD and 9 healthy age-matched control subjects. While peak oxygen uptake was significantly lower in the COPD patients (15.8 ± 3.5 mL·min-1·kg-1) compared to the control subjects (25.2 ± 3.5 mL·kg-1·min-1), there were no significant differences between groups in peak calf blood flow or peak calf conductance measured 7 s post-ischemia. Peak oxygen uptake was significantly correlated with peak calf blood flow and peak conductance in the control group, whereas there was no significant relationship found between these variables in the COPD group. However, the rate of decay in blood flow following ischemia was significantly slower (p less than 0.05) for the COPD group (-0.036 ± 0.005 mL·100 mL-1·min-1·s-1) when compared to the control group (-0.048 ± 0.015 mL·100 mL-1·min-1·s-1). The results of this study suggest that the lower peak exercise capacity in patients with moderate COPD is not related to a loss in leg vasodilatory capacity. Study 4 examined the dynamics of oxygen uptake kinetics during high-intensity constant-load cycling performed at 70% of the peak power attained during an incremental exercise test in 7 patients with moderate COPD and 7 healthy age-matched controls. The time constant of the primary component (phase II) of oxygen uptake was significantly slower in the COPD patients (82 ± 8 s) when compared to healthy control subjects (44 ± 4 s). Moreover, the oxygen cost per unit increment in power output for the primary component and the overall response were significantly higher in patients with COPD than in healthy control subjects. A slow component was observed in 5 of the 7 patients with COPD (49 ± 11 mL·min-1), whereas all of the control subjects demonstrated a slow component of oxygen uptake (213 ± 35 mL·min-1). The slow component comprised a significantly greater proportion of the total oxygen uptake response in the healthy control group (18 ± 2%) than in the COPD group (10 ± 2%). In the COPD patients, the slow component amplitude was significantly correlated with the decrease in inspiratory capacity (r = -0.88, P less than 0.05; N=5), indicating that the magnitude of the slow component was larger in individuals who experienced a greater degree of dynamic hyperinflation. This study demonstrated that most patients with moderate COPD are able to exercise at intensities high enough to elicit a slow component of oxygen uptake during constant-load exercise. The significant correlation observed between the slow component amplitude and the degree of dynamic hyperinflation suggests that the work of breathing may contribute to the slow component in patients with COPD.