Wavefront distortion produced by transmission through abdominal wall, breast, and chest wall has been measured in a two-dimensional aperture and the abdominal wall measurements have been compared to distortion simulated using sound speed and density determined from stained cross sections. The results show that wavefront distortion varies with tissue type, scattering contributes significantly to distortion, and arrival time and energy level fluctuations can be large compared to increments in current beamformers. Focus degradation and compensation has also been investigated by reconstructing the time history of a virtual point source without compensation, with time-shift compensation in the aperture, and with backpropagation followed by time-shift compensation. Backpropagation followed by time-shift compensation was found to be more effective than time-shift compensation in the aperture. A comparison of wavefront distortion and compensation in one- and two-dimensional apertures has shown that distortion is smoothed increasingly by one-dimensional apertures with larger elevation dimensions and that one-dimensional compensation becomes less effective than two-dimensional compensation for larger elevations. Pulse-echo studies using different f-number transmitters have shown that wavefront aberration is significantly underestimated with a broad incident beam and demonstrated that a sharply focused incident beam is necessary for good estimation of time shifts and for effective compensation.