Objective To determine whether paediatric hydroceles result entirely from a small-calibre patent processus vaginalis, allowing free communication between the abdominal cavity and hydrocele sac, or whether there are other mechanisms. Patients and methods Twenty-®ve hydroceles were studied prospectively in 24 boys (aged 18±132 months). Consent for the intraoperative measurements was obtained before surgery. The hydrocele was repaired under general anaesthesia with endotracheal intubation, using a standard approach, taking care not to open the sac during mobilization. Intra-abdominal pressures during surgery were measured indirectly via a nasogastric tube after gastric decompression. The pressure in the sac was measured via a 20 G intravenous cannula inserted via a purse-string suture. The relative pressure was then calculated by subtracting the intra-abdominal from the sac pressure. The effects of age and laterality were evaluated. Results The median (range) intra-abdominal, sac and relative pressures were 8 (2±18), 11 (3±30) and 4 (3±30) cmH 2 O, respectively. The sac pressure in the sac was greater than the intra-abdominal pressure in 17 of 25 (68%; P=0.004) patients. Age or laterality had no signi®cant in¯uence on any of the pressures. Conclusions These results suggest that in a signi®cant proportion of hydroceles in children the pressures are higher than the intra-abdominal pressure. Therefore, they cannot be explained simply as a freely communicating, narrow-calibre processus. In addition, the pressures may reach levels which are potentially damaging to the testis.