Patients with Peripheral Arterial Disease have a reduced capacity for exercise, the exact causes of which are poorly understood. This thesis investigated alternative testing procedures that aim to provide a more complete and precise description of the exercise capacities of these patients. Furthermore, the potential roles of gastrocnemius muscle fibre morphometry, capillary supply and glycogen stores in the exercise tolerance of PAD patients were studied. Study one aimed to determine the effect of test repetition on maximal exercise performance and test-to-test variability in PAD patients using an incremental treadmill walking test (T) (n=5), an incremental cycle test (C) (n=5), and incremental endurance (PF-endurance) and maximal strength (PF-strength) plantar flexion tests (n=5). Tests were conducted once per week for eight weeks. Performance was stable on the T (~530 s) and C (~500 s) tests across the eight weeks. Test-to-test variance on T decreased from 16%CV (CV: coefficient of variation) to 6%CV (p=.21,NS), and from ~8%CV to 2%CV on C (p<.05) over the eight week period. Variance of peak gas exchange variables tended to decrease with performance variance on both tests; however, other physiological variables, and the associated variance levels, were stable throughout the study. PF strength (635-712N) gradually increased over the initial 2-3 weeks (p<.05) which was accompanied by a reduction in variance from ~8%CV to ~3%CV (p<.05). Similarly, PF endurance increased over the first two weeks (~32,000 to 41500 N.s-1) while variance of this measure fell from ~21%CV to ~10%CV (p<.05) over the study duration. It is concluded that the implementation of familiarisation sessions leads to a reduction in whole body and local calf muscular performance variance in patients with PAD. Using a randomised crossover design, study two aimed to compare performance and the physiological and symptomatic responses between a T test and a C test in 16 patients with PAD. Peak exercise time on C (690 s) was greater than that on T (495 s); however the two were significantly correlated (n=16, r=.69, p<.05). Peak HR (120 bpm), VO2 (~1.22 l.min-1) and rate pressure product (~20') did not differ between the two tests, nor did the post exercise ankle pressure (T: 56; C: 61 mmHg). In two subjects with lower back pain during C, the ankle pressure of their "worst" limbs failed to fall by >10mmHg. Performance on both the T and C tests was closely related to the onset of leg symptoms; however the site of pain during C was much more variable than during T. Incremental cycle testing would overcome some of the limitations of treadmill testing (e.g. measurement of mechanical work), and it appears to be an acceptable alternative for measuring the exercise capacity and physiological exercise responses in known claudicants. Use of cycle ergometry for the diagnosis of PAD requires testing in the general population. Study three aimed to compare whole body (T test and C test) and local calf muscular (PF strength and endurance) exercise performance between 16 PAD patients (age: 63 ± 2; BMI: 25.9 ± 1.1) and 13 healthy, sedentary control (CON) subjects (age: 62 ± 1; BMI: 25.9 ± 0.4), and to describe relationships between the whole body and local calf muscular exercise capacities within the two groups. Furthermore, this study aimed to compare several histochemical characteristics of the medial gastrocnemius muscle fibres between PAD and CON, and to establish whether these factors were related to the exercise capacities of both groups. Maximal performance on T was 59% lower in the PAD group compared with the CON group, as was performance on C (50%), PF strength (25%), and PF endurance (58%). Compared with CON, PAD patients had a lower estimated calf muscle mass and a slight reduction (10%) in muscle fibre size (p=.14, NS). They also had a lower proportion of type I fibres (PAD: 49%; CON: 62%) that was offset by a greater proportion of type IIA fibres (PAD: 27%; CON: 16%), and a reduction in the capillary contacts per muscle fibre (PAD: 1.63; CON: 2.12) compared with CON. When expressed relative to fibre area there were no differences in capillarisation between PAD and CON; however this index was significantly related to resting and post exercise ABI in the PAD patients. There were no differences in the mixed muscle [glycogen], nor the optical density of glycogen in the individual fibres, between the two groups. PF endurance was poorly predictive of walking performance, and did not correlate with any of the morphological variables in both groups. Calf muscle mass correlated with PF strength (r=.59 - .62), and strength was correlated with T performance (r= .61 - .63) in both groups. In the PAD patients, T performance was correlated with the cross sectional area (n=12, r=.72, p<.05), capillary contacts (n=10, r=.81, p<.05) and glycogen density (n=9, r=.81, p<.05) of type I fibres. This study confirms that a reduction in calf strength, which appears to be mediated through muscle atrophy, plays some role in the reduced exercise capacity of claudicants. While both fibre area and capillary supply seem to be of relevance to the exercise capacity of PAD patients, these two factors are closely linked and further research is required to establish the determinants, and relative importance of both. An important, and possibly limiting role of carbohydrate oxidisation in PAD patients is supported by the strong relationship between type I glycogen stores and whole body exercise capacity.