Positron emission tomography (PET) makes possible the quantitation of the regional distribution of positron-emitting radioisotopes in a selected organ (Ter Pogossian et al.: Radiology 114:89-98, 1975). Since a large variety of physiologic compounds can be labeled with positron emitters (Wolf: Seminars in Nuclear Medicine 11:2-12, 1981; Fowler and Wolf: Nuclear Science Series, 1982), the technique can be used to monitor various biochemical and physiological processes of living tissues (Phelps et al.: Annals of Internal Medicine 98:339-359, 1983; Brodie et al.: Handbook of Neurochemistry, 1983). Indeed, research in the clinical application of PET has expanded considerably over the past 5 years and PET has been used in a variety of clinical problems in psychiatric, neurologic, and oncologic disorders (Phelps and Mazziotta: Science 228:799-809, 1985; Patronas et al.: Journal of Neurosurgery 62:816-822, 1985; Volkow and Tancredi: International Journal of Technology Assessment in Health Care 2:577-594, 1986). Interpretation of the findings by the clinician has sometimes been confounded by the complexity of the technique and the many parameters which can influence the data. A simple knowledge of the methodological background of PET is needed in order to appreciate the potential applications and the limitations of this technique. There are three basic components in PET: instrumentation, radiopharmaceuticals, and mathematical modeling. A number of reviews address varied aspects of the last two components (Fowler and Wolf: Positron Emission Tomography and Autoradiography: Principals and Application of the Brain and Heart, 1986; Karlstrom and Christman: Brookhaven National Laboratory Associated Universities, Inc., 1983; Huang and Phelps: Positron Emission Tomography and Autoradiography: Principles and Applications for the Brain and Heart, 1986), and in this paper we will focus on PET instrumentation as it pertains to its clinical application. The aim of this article is to give an outline of the basic principles behind PET instrumentation, such as positron detection, resolution, quantitation, sensitivity, and image reconstruction as it affects the data presented to the clinician.