To determine the predominant site of action of methacholine (MCh) on lung mechanics, two groups of open-chest Sprague-Dawley rats were studied. Five rats were measured during intravenous infusion of MCh (i.v. group), with doubling of concentrations from 1 to 16 micrograms.kg-1.min-1. Seven rats were measured after aerosol administration of MCh with doses doubled from 1 to 16 mg/ml (ae group). Pulmonary input impedance (ZL) between 0.5 and 21 Hz was determined by using a wave-tube technique. A model containing airway resistance (Raw) and inertance (Iaw) and parenchymal damping (G) and elastance (H) was fitted to the ZL spectra. In the iv group, MCh induced dose-dependent increases in Raw [peak response 270 +/- 9 (SE) % of the control level; P < 0.05] and in G (340 +/- 150%; P < 0.05), with no increase in Iaw (30 +/- 59%) or H (111 +/- 9%). In the ae group, the dose-dependent increases in Raw (191 +/- 14%; P < 0.05) and G (385 +/- 35%; P < 0.05) were associated with a significant increase in H (202 +/- 8%; P < 0.05). Measurements with different resident gases [air vs. neon-oxygen mixture, as suggested (K.R. Lutchen, Z. Hantos, F. Peták, A. Adamicza, and B. Suki J. Appl. Physiol. 80: 1841-1849, 1996)] in the control and constricted states in another group of rats suggested that the entire increase seen in G during the i.v. challenge was due to ventilation inhomogeneity, whereas the ae challenge might also have involved real tissue contractions via selective stimulation of the muscarinic receptors.