Two hundred and forty-four surgical patients who received no anticholinergic premedication were compared with 160 patients who had received atropine or scopolamine before the induction of anaesthesia. Infants and patients undergoing heart surgery were excluded. Eleven anaesthetists participated in the study. They were asked to report problems with oropharyngeal and tracheobronchial secretions. Two per cent of unpremedicated patients experienced problems with secretions of a degree sufficient to require treatment. This small percentage appears insufficient to warrant routine preoperative anticholinergic medication.