Endogenously released nitric oxide (NO) in airways might contribute to physiological bronchodilation; induced production of NO might play a role in the pathogenesis of asthma, although it could also be a compensatory mechanism to other factors that cause bronchoconstriction or inflammation. To investigate the efficacy of NO donors on bronchial tone, the bronchorelaxing efficacies of NO donors, new experimental GEA compounds 3268 and 5145 (oxatriazole sulphonylamides) were compared with those of sodium nitroprusside and SIN-1 (3-morpholinosydnonimine) and to the standard beta 2-adrenergic agonist, salbutamol, in bronchi of guinea-pigs and rats in-vitro. Their relaxing effects were also studied in rat mesentery arteries to compare the selectivity for airways. The capacity of the NO donors to produce nitrites and nitrates was assayed by the Griess reaction. The novel NO donors GEA 3268 and GEA 5145 were more potent bronchorelaxing agents than the old NO donors sodium nitroprusside and SIN-1. In guinea-pig bronchi, however, salbutamol was most potent. In rat bronchi the GEA compounds induced the strongest relaxation effect when compared with the old NO donors or with salbutamol. The airway selectivity of the drugs studied decreased in the order of salbutamol, SIN-1, GEA 5145, GEA 3268, sodium nitroprusside. The nitrites and nitrates produced spontaneously did not correlate with the efficacy of the relaxants. The results obtained suggest that NO is only partly responsible for the relaxation and the potency is dependent on the animal species and constricting agents used.