The aim of this study was to investigate the reproducibility of skin surface infrared thermography (IRT) measurements and determine the factors influencing the variability of the measured values. While IRT has been widely utilized in different clinical conditions, there are few available data on the values of the skin temperature patterns of healthy subjects and their reproducibility. We recorded the whole body skin temperatures of sixteen healthy young men with two observers on two consecutive days. The results were compared using intra-class correlations analyses (ICC). The inter-examiner reproducibility of the IRT measurements was high: mean ICC 0.88 (0.73-0.99). The day-to-day stability of thermal patterns varied depending on the measured area: it was high in the core and poor in distal areas. The reproducibility of the side-to-side temperature differences (deltaT) was moderately good between the two observers (mean ICC 0.68) but it was reduced with time, especially in the extremities, mean ICC 0.4 (-0.01-0.83). The results suggest that the IRT technique may represent an objective quantifiable indicator of autonomic disturbances although there are considerable temporal variations in the measured values which are due to both technical factors such as equipment accuracy, measurement environment and technique, and physiological variability of the blood flow, and these factors should be taken into account.