The aim of the present study was to investigate the relevance of BTPS (gas at body temperature, atmospheric pressure and saturated with water vapour) conversion of volumes measured with the Vitalograph bellows spirometer. The Vitalograph bellows were tested against a MicroMedical turbine spirometer in extreme temperatures (0-37 degrees C) using a biological control to deliver expired gas at BTPS. Before testing, it was shown that the accuracy of the DairyCard turbine was stable in the relevant temperature range. In a clinical trial six patients with emphysema performed home spirometry b.i.d for 1 month using both the Vitalograph and the turbine. Both the DairyCard and the Vitalograph showed stable accuracy at extreme temperatures when results were reported without any BTPS conversion. These findings were supported by the clinical trial but the conclusions from the clinical setting were weakened by the surprising fact that domiciliary temperatures showed almost no variation. We conclude that the Vitalograph bellows, during dynamic spirometry, measures expired volume at conditions closer to BTPS (than to ATPS) gas at ambient temperature, atmospheric pressure and saturated with water vapour). The use of the BTPS correction based on ambient temperature seems unjustified at office temperatures close to 23 degrees C and at extreme temperatures the conversion of volume will introduce significant over or underestimation.