The reproducibility and accuracy of routine echocardiographic measurements made by an inexperienced doctor using tele-instruction were evaluated. Thirty-eight patients were first examined at a local hospital by an inexperienced doctor instructed by a specialist 450 km away at a university hospital. The specialist then examined the patients at the local hospital using the same equipment, after an average of 50 days. The accuracy of M-mode and quantitative Doppler measurements was comparable to that observed in reproducibility studies made under normal examination conditions. There were no systematic measurement errors. No important M-mode information was missed except evidence of left ventricular hypertrophy in six patients. In the two-dimensional examination there were differences of clinical significance in only three patients. There were no clinically important differences in the Doppler quantification of mitral and aortic regurgitation. Tele-instructed echocardiography is also an excellent educational tool, allowing an inexperienced examiner gradually to take responsibility for the local echocardiographic service.