The aim of the present investigation was to study the dose response relationship of allergen, histamine and histamine releasers in skin prick test (SPT) and the precision of the SPT method. In one experiment timothy allergen, histamine HCl, rabbit anti-human-IgE and compound 48/80 were studied in seven patients. In a second experiment timothy allergen and histamine and dog allergen and histamine were investigated in two groups of 10 patients. Histamine HCl 1 and 10 mg/ml induced weals about 15 and 25 mm2 (4.5 and 5.5 mm in diameter), respectively. The precision as expressed by the coefficient of variation was about 25% for histamine and 40% for allergen for weal areas greater than 10 mm2. Calculations of the regression lines to test the dose response relationships were based on the method of least squares. The best fit was to a log/log model. The slopes of allergen, histamine and histamine releasers were essentially parallel within patients. The median slope of allergen was estimated to about 0.4 based on weal areas and 0.2 based on mean weal diameters. Furthermore, no significant differences were found between the lower and upper parts of the dose response curves of allergen and histamine, although there was a tendency towards steeper slopes at lower concentrations. These results show that histamine concentrations greater than or equal to 1 mg/ml should be used as positive control in SPT and that histamine releasers do not offer advantages over histamine as reference substances in SPT. A common slope for the dose response relationship of allergen and histamine can be used for the estimation of skin sensitivity.