Abstract In cat and monkey a comparison was made of simultaneous records of eye movements obtained by the electro-oculographic and the magnetic search coil techniques. The major drawback of the electro-oculographic method is the slow d.c. drift, still present despite careful preparation of the electrodes. The d.c. level varies considerably with lighting conditions over long periods of time. In addition, the EOG picks up blink artifacts in monkey and other movement artifacts which are described in detail. The objective of this study was to document and evaluate the relative importance of these technical problems. Because of its simplicity and convenience the EOG method still has some applications in neurophysiological experiments. Its use requires the possibility of making frequent calibrations of gain and d.c. level.