Tungsten stimulating microelectrodes have been tested in monopolar, bipolar and concentric configurations for the extent to which unwanted current spread occurred. Current spread from monopolar electrodes in close conformity with the predictions of the inverse square law, both in vitro in vivo. The bipolar and concentric configurations, tested only in vitro, had current-spread characteristics which did not follow the inverse square law so closely. The bipolar configuration gave little reduction in spread, compared wpith a monopolar electrode, but the concentric configuration did reduce the extent of stimulus spread. The extent of spread depends greatly upon a number of experimental variables, including the dimensions of the microelectrode tip. For studies requiring precise localization with low stimulus currents, it is advisable to determine the inverse square law constant for each experimental situation. For more generalized stimulation purposes within the mammalian C.N.S., employing monopolar microelectrodes of moderate tip size, brief pulses of 10 muA can be expected to stimulate myelinated axons with about 0.15 mm of the tip, and 100 muA WILL Stimulate with a radius of about 0.5 mm.