The theoretical possibilities and the practical limitations of the Xenon-133 (133Xe) method for the study of regional myocardial perfusion in man are discussed. The techniques for data acqusition and processing developed over the past 5 years are described in detail. Illustrative examples of experimental findings are reported. The practical interpretation of the data, at the light of the influence of injection site, initial tracer distribution, constancy of counting geometry, spatial resolution, and Xenon retention in fat, is presented.