Experiments have been performed in order to get information about the phenomena preceding the electrical breakdown in small vacuum gaps. Most experiments have been made with impulse voltages of different rise times; some complementary results obtained with alternating voltage are also presented. The effect of surface layers on the breakdown voltage and on the pre-breakdown current is discussed. It has been found that the rise time of the voltage affects both the breakdown voltage and the pre-breakdown current. The experiments seem to indicate that breakdown in the underlying circumstances is the result of a discharge in metal vapour, originating from the anode. The vapour is thought to be generated by the heating of the anode by a bombardment of field-emission electrons. The transition of the pre-breakdown current to a sudden discharge may occur when the vapour density passes a critical value.