The mode of activation of nerve cells by extracellular stimuli was investigated while recording from a selected cell with one electrode, and applying current pulses around this cell with another electrode. The analysis was done on motoneurones and on spinal border cells from lower lumbar segments in the cat. 2. Directly evoked action potentials were defined by their appearance in an all-or-none fashion with stable latencies of less than 0-5 ms. The lowest thresholds for their generation were 0-15-0-20 muA in the spinal border cells and 0-35-0-40 muA in the motoneurones. In the main series on motoneurones a correlation has been established between different positions of the extracellular stimulating electrode in relation to the cells and the thresholds for the direct excitation of these cells. The position of the electrode were defined on the basis of an analysis of the IS and SD components of the action potentials recorded extracellularly around the cell when evoked by current pulses applied through the intracellular electrode; both the amplitudes of these IS and SD components and their timing with the IS and SD spikes, which were simultaneously recorded with the intracellular electrode, were then taken into account. The lowest thresholds (less than 2 muA) for the direct activation of cells were found nearest the initial segment of the axon. Their values increased to about 5 mu A at near-soma positions and to greater than 10 muA at near-dendrites positions about 150 mum away. 3. Transsynaptically evoked action potentials which were clearly set up by the preceding e.p.s.p.s appeared with latencies greater than 0-7 ms. When single current pulses were used, the lowest thresholds for transsynaptic spike activation were usually greater than 5-10 muA but they considerably decreased with repetitive stimuli. These thresholds were higher than the thresholds for the direct activation of cells within the region of the initial segment, of the same order of magnitude near the soma, and lower when the stimulating electrode was nearer the dendrites than the soma and generally at all larger distances from the cells. 4. All the observations on direct excitation of cells by extracellular stimuli (generation of the IS spike before the SD spike, lowest thresholds near the region of the initial segment of the axon, similar rates of increase in these thresholds with distance as for fibres) lead to the conclusion that the effects of the extracellular stimuli are exerted primarily via spread of current to the initial segment of the axon and its depolarization. 5. Late extracellular negativities presumably related to dendritic activation were observed in a few cells. These negativities were synchronous with late components of the intracellulary recorded action potentials.
