Several geometric algorithms have been applied to estimate left ventricular wall volume (V(wall)) from two-dimensional echocardiograms but have not been validated in eccentrically hypertrophied hearts. These algorithms can be fitted to the general formula: V(wall) = k · A(o) · L(o) - k · A(i) · L(i), where A(o) and A(i) are the outer (epicardial) and inner (endocardial) short-axis areas, L(o) and L(i) are the corresponding long-axis lengths, and k is a constant. The simplifying assumption that L(o) and L(i) are equal yields V(wall) = k · A(wall) · L(o), where A(wall) = A(o) - A(i). In 20 unsedated dogs (10-30 kg), including 10 with aortic regurgitation of 1-18 wk duration, the relationship between actual V(wall) (determined postmortem) and A(wall) · L(o) was not significantly different from the line of identity (V(wall) = 1.01 A(wall) · L(o) + 0.5 ml, r = 0.98, SEE = 3.5 ml), indicating k was not significantly different from 1. There was no significant difference between predicted and actual V(wall) over a range of 31-105 ml, and interobserver variability was 4.1%. The simple area-length product, A(wall) · L(o), accurately predicts V(wall) of both normal and volume-overloaded hypertrophied canine left ventricles and is thus suitable for serial observations of hypertrophic adaptation to volume overload.