Three milliliters of plain bupivacaine 0.5%, previously warmed to 37 degrees C, was injected into the subarachnoid space in two groups of 20 patients. In Group 1, speed of injection of the bupivacaine solution was as fast as possible (mean, 0.54 ml/second), whereas in Group 2 the bupivacaine solution was administered at a rate of 0.05 ml/second. Patients were kept sitting for 3 minutes after completion of the subarachnoid injection. All observations were made by an observer unaware of the speed of injection. There were no statistically significant differences in onset to maximal level of sensory block (17.2 minutes in Group 1 versus 19 minutes in Group 2) or complete motor block of the lower limbs (9 minutes in both groups). The difference in the maximal level of sensory block as determined by pinprick (T4.9 in Group 1 versus T3.4 in Group 2) was statistically significant; this difference may be explained by the fact that the plain solution of bupivacaine 0.5% at 37 degrees C is hypobaric and patients in Group 2 were in the sitting position for 1 minute longer when measured from the start of the subarachnoid injection. It is concluded that a tenfold difference in speed of injection does not significantly affect subarachnoid spread or onset of sensory and motor block with 0.5% bupivacaine.