As measurement of glomerular filtration rate (GFR) is now generally the responsibility of departments of nuclear medicine, it is important for nuclear medicine physicians and scientists to understand the pharmacokinetics of the indicators and radiotracers that are used, generally known as filtration markers. The single-injection, non-steady state technique is almost universally used, departments varying in how many blood samples are taken: rarely multisample clearance, which does not assume a single compartment of tracer distribution, commonly clearance based on a limited number of blood samples between 2 and 4 h after injection, which assumes a single compartment of distribution, and often a single sample at a defined time point. The volume of distribution, V(d), of a filtration marker is close to extracellular fluid volume (ECFV). GFR and ECFV are both overestimated by the assumption of a single compartment by amounts that are functions of the rate of plasma clearance, Z. Residence time, T, of tracer in its V(d) is equal to V(d) divided by Z. Z and T can both be measured from a multisample clearance curve, whereupon V(d) is the product of Z and T. GFR is usually indexed to patient size by expressing it in relation to body surface area (BSA), which in turn is calculated from an equation based on the patient's height and weight. An equation in common use was described by Haycock et al. and is BSA=0.024265 x weight(0.5378) x height(0.3964). An alternative indexation variable is ECFV. GFR per unit ECFV is close to the rate constant, alpha(3), of the terminal exponential of the plasma clearance curve. It is in fact slightly higher than this rate constant by an amount that is a function of the rate constant itself. The discrepancy between GFR/ECFV and alpha(3) arises from the development of a concentration gradient between interstitial fluid and plasma, which in turn produces an extrarenal veno-arterial gradient throughout the body. Indexing GFR to ECFV not only has physiological attractions (especially in children) but is technically simple because it requires measurement only of alpha(3) (slope-only technique). A disadvantage, however, is a lack of robustness in comparison with the conventional slope/intercept method, which measures tracer dilution as well as alpha(3). Nevertheless, the advantages of indexation to ECFV can still be exploited by changing the constants of an equation of the Haycock type so that the equation becomes a predictor of ECFV rather than BSA. A recently described equation is ECFV=0.02154 x weight(0.6469) x height(0.7236). Indexation to ECFV abolishes differences that arise between children and adults when GFR is indexed to BSA.