1. Seven thermal conditions were imposed on male sitting subjects (slightly clothed: 0.6 clo). 2. A thermal mannikin was also used to determine the exact operative temperature, T0. 3. Conditions were: uniform (UN: all parameters at 24.5°C, air velocity at 0.15 ms−1), heated ceiling (HC at 45°C), heated floor (HF at 34°C), cold floor (CF at 14°C), two conditions of one cold wall at 6°C (CW1 and CW2 respectively with and without air temperature compensation) and increased air velocity (AV at 0.4 ms−1). 4. Local skin temperatures and answers to questionnaires were obtained. 5. Skin temperature variations were affected by conditions and slight T0 changes. 6. Comfort judgments were fairly well related to T0, especially when expressed as differences between actual non-uniform environment and the uniform one. 7. It is concluded that, in case of non-uniform environments close to thermoneutral zone, thermal comfort or discomfort reflects the climate alterations better than the thermal sensation does.