The reliability of thermography for the determination of the level of amputation for an ischaemic lower limb was compared with that of the doppler flowmeter and the clinical judgement of an experienced surgeon. The results showed that thermography tended to overestimate the degree of ischaemia and to indicate a higher level of amputation than was necessary. The advantages and the limitations of each method are discussed. It is concluded that the clinical judgement of the surgeon was the most reliable method and that the other investigations are only an adjunct to clinical assessment.