Previously, we have shown that pericardial constraint cannot be measured by true (hydrostatic) pressure except when an excess of pericardial fluid is present and that a device such as a balloon (which reflects radial contact stress as well as hydrostatic pressure) must be used. Since radial contact stress is the major component of the constraint exerted by the pericardium when little pericardial liquid is present, it follows that the pressure measured by the balloon might be different over different parts of the heart. In an attempt to test this hypothesis, in 11 anesthetized dogs we placed pericardial balloons over the right and left ventricular free walls, instrumented the animals to measure ventricular dimensions (sonomicrometry) and pressure, mounted pneumatic constrictors on the aortic and pulmonary artery, reapproximated the pericardium, and closed the chest under suction. We studied the transient effects of constrictions of the ascending aorta and pulmonary artery and of angiotensin infusion before and after intravenous saline infusion. Aortic constriction and, to a lesser degree, angiotensin increased pericardial pressure over the left ventricle more than over the right ventricle. Pulmonary artery occlusion increased pericardial pressure over the right ventricle but significantly decreased pericardial pressure over the left ventricle. We conclude that there are significant local differences in pericardial pressure (recorded by balloon) over the lateral ventricular surfaces during acute changes in afterload. These observations may be explained in part by decreased venous return to the contralateral ventricle, the tendency of the heart to resist lateral displacement, and the limited mobility of the pericardium.