The aim of this study was to assess oral premedication with midazolam in paediatric anaesthesia. Sedation, quality of induction, recovery time, acceptance and effects on gastric contents were analysed. This prospective, double blind, at random and controlled study was performed in 107 children, aged between three and ten years. They were divided into: group 1 (control, n = 29), group 2 (placebo) receiving 5 ml of water in the preoperative stage (n = 40), and group 3 (midazolam) with 0.75 mg.kg-1 midazolam by mouth (n = 38). Two children refused to take medication. In children aged five years or more (n = 48) of groups 2 and 3, acceptance of premedication was evaluated. The midazolam group showed a better level of sedation as compared with the placebo (P < 0.05). The recovery time was similar for the two groups. There were no statistically significant differences in gastric pH or residual volume among the three groups. It is concluded that midazolam given by mouth is an efficient and safe drug for premedication in paediatric anaesthesia.