Continuous electrocardiography was performed during eighty operations on the posterior fossa of the skull and fifty-seven other neurosurgical operations, all in the sitting position. There was little difference between the two groups in the incidence of arrhythmia before operation and during the surgical approach; more serious irregularities were almost entirely confined to periods when the surgeon was operating in the vicinity of the pons, the medulla, and the roots of the fifth, ninth and tenth cranial nerves. Severe arrhythmias may indicate or be the cause of medullary failure in this position and may also give an early warning of air embolism. Therefore continuous electrocardiography is valuable provided that its limitations are understood.