Efficiency of peritoneal dialysis (PD) is dependent on adequate ultrafiltration (UF) and purification (solute clearance). These two goals apparently seem to conflict in terms of duration of dwells: short dwell time enhances UF capacity and, conversely, long dwell time enhances solute clearance. Peritoneal equilibration test (PET) allows an approach to the ultrafiltration time: the point at which the over time dialysate urea saturation and glucose desaturation curves cross, call APEX time. PET also allows an approach to the purification time: the point at which dialysate-to-plasma (D/P) concentration ratios over time are high. Because of the value of phosphate as a uremic factor of morbidity, we have chosen the time at which D/P phosphate is equal to 0.6 as a purification phosphate dwell time (PPT). A total of 17 patients were studied, over a five-year period, allowing 142 determinations. APEX times (range 18-71 min) and PPT (range 105-238 min) were spread over a wide distribution. PPT and APEX times were significantly shorter in children younger than three years of age than in children older than ten years of age. PPT were nearly four times longer than APEX times. Knowledge of these conflicting ultrafiltration and purification times should help, in our view, in the individual choice of the PD modality: if UF is the major goal, short dwell times should be used (automated PD); if purification is the major goal, long dwell times should be used, as in continuous ambulatory peritoneal dialysis; if both are the target goal, tidal PD should be discussed.