jections, a literature review showed that the effect of position on discomfort had never been tested systematically. Therefore, we undertook a controlled study with 44 patients at a university medical center who were to receive preoperative injections(2). Seventeen men and 27 women, aged 20 to 65 years (mean 42.6 years), agreed to participate. We selected only patients who were oriented to time, place, and person; who could easily assume a prone position; who were to receive preoperative medications of atropine, morphine, and diazepam; and who could safely receive injections in the dorsogluteal site. Each patient received one injection of atropine and morphine, and a second injection of diazepam. To rotate their femurs internally, patients were told to point their toes inward. To rotate them externally, they were instructed to point their toes outward. Each patient received one injection with femurs internally rotated, the other with them externally rotated. The study conditions were specified before the study began and provided all possible combinations of the following factors: position, medication, and order of injection. The injections were given by two nurse researchers. All injections were given with a 23gauge needle. Needle length varied from 1 to 1-/2 inches, depending on the patient's size and weight. Each patient received an equal amount of medication (1.5 to 2 cc.) in each of the two injections. After each injection, patients rated their discomfort on a five-point scale: 1-no discomfort, 2-slightly uncomfortable, 3-moderately uncomfortable, 4-very uncomfortable, and 5extremely uncomfortable. The scale was purposefully kept simple to minimize demands on patients awaiting surgery. We anticipated that diazepam, an irritating drug, would cause more discomfort than atropine-morphine injections(3). Therefore, we examined the effect of patient position on discomfort scores separately for each medication.