One hundred and one non-diabetic and 54 diabetic patients suffering from lower limb ischaemia were divided into (i) asymptomatic subjects, (ii) claudicants, (iii) critically ischaemic patients, i.e. Fontaine III or IV patients with either an ankle pressure < 51 mmHg or a toe pressure < 31 mmHg, and (iv) Fontaine III or IV patients in whom ankle and toe pressures could not be assessed due to vessel wall sclerosis or skin ulceration. Skin microcirculation was investigated to assess (a) the compounding effect of diabetes in leg ischaemia and (b) the additive value of microcirculatory investigation in the appreciation of the severity of the ischaemic disease. The techniques used included capillary microscopy, transcutaneous oximetry and laser Doppler fluxmetry. The severity of ischaemia was readily discernable using microcirculatory techniques. The presence of diabetes appeared to change skin microcirculatory perfusion, but especially in critically ischaemic patients, the microcirculation was no more compromised than non-diabetics. Using skin oxygen tension measurements, a positive predictive value of 77% was obtained in the detection of critical ischaemia, when a cut-off value of 30 mmHg was applied. Seventy per cent of patients, in whom the severity of ischaemia could not be classified using blood pressure measurements, could be classified as critically ischaemic on the basis of microcirculatory investigation. In conclusion, the influence of diabetes on the microcirculation is outweighed by the effects of atherosclerosis when vascular disease becomes severe. Techniques to investigate skin microcirculation are a useful way of assessing the severity of lower limb ischaemia in the presence of diabetes mellitus or if peripheral blood pressures cannot be obtained.