If a voltage is applied across the diode with the p-type end positive (the "forward bias" case), any current that flows must flow from the p-type to the n-type material. The positive holes and the negative electrons both travel toward the boundary. When they meet there, the electrons happily fill the holes, so a steady current can easily flow: the resistance is small. The energy released when the electrons jump into the holes may even be released in the form of light, as in the familiar LED (Light Emitting Diode) device.

On the other hand, if a voltage is applied across the diode with the p-type end negative (the "reverse bias" case), any current that flows must travel in the opposite direction, so electrons and holes both travel away from the boundary. But there is no good source of electrons and holes at the boundary, so very little current flows: the resistance of the diode is large. (If sufficient energy is made available at the boundary region, it could spring an electron loose and thereby create a free electron and hole which would contribute to a current. That energy could for example be supplied by light, as in a photocell or video camera; or it could be caused by thermal fluctuations, as in a solid-state thermometer.)