Current understanding of beta2-microglobulin (beta2M) and phosphate (or inorganic phosphorus) kinetics during hemodialysis is reviewed. The postdialysis:predialysis concentration ratio for beta2M is determined by dialyzer clearance for beta2M, treatment time, patient body size (specifically, extracellular fluid volume), and total ultrafiltration volume during the treatment. Evaluation of these treatment parameters can be used to calculate dialyzer clearance for beta2M; however, such calculated values are only approximations, since they neglect intradialytic generation, nonrenal (nondialyzer) clearance, and postdialysis rebound of beta2M. The detailed kinetics of beta2M during hemodialysis are best described using a two-compartment model. Theoretical predictions from such two-compartment models suggest that the product of dialyzer clearance for beta2M and weekly treatment duration, independent of treatment frequency, is the main determinant of plasma beta2M concentrations. The kinetics of phosphate removal during hemodialysis are incompletely understood. Phosphate is removed from both extracellular and intracellular compartments during hemodialysis; the plasma phosphate concentration levels off after the first 1 or 2 hours of treatment and plasma concentrations can rebound even before therapy is complete. Increases in dialyzer clearance of phosphate have been previously achieved only by increasing dialysis membrane surface area or by the use of hemodiafiltration. A four-compartment model of phosphate kinetics proposed recently by Spalding et al. suggests that the major barrier to phosphate removal is limited transfer of phosphate between the intracellular and extracellular compartments, although other complex factors also play important roles. Theoretical predictions using the model of Spalding et al. suggest that increasing either treatment frequency or treatment duration can increase phosphate removal. The kinetics of beta2M are representative of middle molecules whose removal during hemodialysis is governed predominantly by clearance at the dialyzer. In contrast, phosphate removal is limited primarily by its sequestration in the intracellular compartment (and possibly other compartments), not by its clearance at the dialyzer. The kinetics of phosphate may therefore be representative of uremic toxins whose removal is limited by sequestration into compartments or by protein binding. Enhanced removal of both of these uremic toxins using a given therapy will require treatments of increased frequency and longer duration.