heparin was administered. The left ventricular muscle along the anterior descending branch of the left coronary artery was then excised and immersed in cold Ringer's solution at about 4°C. The wet weight of the isolated left ventricular preparations varied from 8 to 15 g. The anterior descending branch was cannulated and perfused with heparinized blood conducted from the common carotid artery of the support dog with the aid of a peristaltic pump (model 1210; Harvard Apparatus). A pneumatic resistance was placed in parallel with the perfusion system so that a constant perfusion pressure of 100 mmHg could be maintained. The venous effluent from the preparation was led to a blood reservoir and returned to the support dog through the external jugular vein. The isolated ventricular muscle was driven by an electrical stimulator (SEN 7103; Nihon Kohden, Tokyo, Japan) with a pulse duration of 1 ms and 4 V pulse amplitude at a frequency of 2 Hz. The superior part of the ventricle was connected to a forcedisplacement transducer (AP620G; Nihon Kohden) by a silk thread to measure the isometric tension. The muscle was loaded with a resting tension of 2 g.In the first series of experiments, we investigated the changes in heart rate and arterial blood pressure of the support dog and the concomitant changes in contractile force of the isolated left ventricle (n ϭ 5) when propofol (0.3-3 mg·kg Ϫ1 ) was administered to the external jugular vein of the support dog. Each dose of propofol was given cumulatively at 10-min intervals. In the second series, to compare the negative inotropic effects of propofol and thiopental, both drugs (30-1000 µg) were injected into the anterior descending branch of the left coronary artery of the isolated ventricle (n ϭ 6).The results are shown as maximal percentage changes from predrug values and are expressed as mean Ϯ SEM. The data were analyzed by an analysis of variance and Bonferroni's method for multiple comparisons of data. P values less than 0.05 were considered statistically significant.