The accuracy, precision and linearity of a new portable flow-volume meter, the Microspiro HI-298 (Chest Corporation, Tokyo, Japan), was investigated using a Fleisch no. 4 pneumotachograph as a standard. After connection and calibration of the pneumotachograph and the Microspiro, a healthy subject performed 44 forced vital capacity (FVC) manoeuvres at different levels of lung inflation. The FVC of these expirations ranged from 2.5-5.1 l. Linear regression of Microspiro values (dependent variable) on Fleisch pneumotachograph values (independent variable) showed that a good linear relationship existed: Pearson correlation coefficients ranged from 0.938-0.985. Linearity of the Microspiro was good except for the peak expiratory flow rate (PEFR) and the maximal expiratory flow at 25% of the expired volume (MEF75). The random error (measure of precision) of all flow-volume (F-V) indices was lower than 5%. The systematic error (measure of accuracy) was low for the forced expiratory volume in one second (FEV1) and the FVC (1% and 4.6%, respectively) but much higher for the instantaneous expiratory flows (PEFR 11.0%; MEF75 7.0%; MEF50 8.5%; MEF25 11.4%). Only the total error in FEV1 complied with the tolerance of 4% of the European Community for Coal and Steel (ECCS). When the measured values were adjusted according to the regression equations of this study, all F-V indices were accurate and precise within 5%. It was concluded that the portable Microspiro HI-298 is a useful instrument for bedside, work-site spirometry and for use in general practice.(ABSTRACT TRUNCATED AT 250 WORDS)