This paper identifies and discusses potential sources of error in evaluating the adsorptive capacity of granular activated carbon (GAC), with special attention given to carbon sampling and preparation, preparation of test solutions, and selection of GAC dosages, adsorbate concentration, and equilibration time. Adsorptive capacity can vary with particle size, but data are presented to show that this may not always occur. Heterogeneous solutions require careful selection of GAC dosages and adsorbate concentration, and the data for such solutions require careful interpretation. Model simulations are used to illustrate the importance of closely approaching equilibrium and to estimate the time required to approach equilibrium under various conditions. For large GAC particles and slowly diffusing adsorbates, several years may be required to reach equilibrium. Failure to reach equilibrium can result in a significant underestimation of adsorptive capacity. Pulverizing GAC greatly reduces the time required to reach equilibrium, thus reducing the possibility of biodegradation of the adsorbate. (Abstract also appears in Spanish) Includes 36 references, tables, figures.