In 18 clinically stable asthmatic patients histamine inhalation challenges were performed with a Wright's nebuliser and tidal volume breathing for two minutes on two to four occasions for each subject at a maximum interval of two weeks. The response was measured in terms of specific lung conductance (sGL) by the subtraction technique, maximum partial and maximum complete expiratory flow at 40% and 50% of vital capacity respectively (Vmax40p and Vmax50c), and FEV1 from the maximum flow-volume curve. Dose-response curves were analysed for (1) provocative concentration (PC) of histamine causing a 20% fall in FEV1 and a 40% change in the other measurements; (2) threshold concentration (TC)--the concentration at which changes in the measurement exceed 2 SD from control values; (3) reactivity (R)--the slope of the dose-response curve beyond TC. We found that PC20-FEV1 was the most reproducible index, the 95% confidence intervals based on a single determination being +/- 1.6 single two-fold concentration difference. PC20-FEV1 was more reproducible than PC values for other measurements and more reproducible than any of the TC values. The 95% confidence intervals based on a single determination of R varied from +/- 52% to +/- 74% change/log histamine concentration. Both sGL and Vmax40p detected the bronchoconstrictor response assessed by PC and TC at a significantly lower histamine concentration than FEV1 (p less than 0.01 and p less than 0.05 respectively). PC and TC results showed a significant correlation, but neither were correlated with R.