Abstract
In the course of determinations of the surface resistivity of hard-glass and silica rods from observations of the discharge of a Wulf electrometer, it was found that the voltage decay was initially non-exponential. This is attributed to the effect of capacitance between rod and case, distributed along the length of the rod. The effect is treated mathematically and it is shown that, even if the resistance is determined from the exponential portion of the curve, a correction is required. A formula for the correction is given and an approximate method of applying it referred to. An indication is given of the circumstances, probably not uncommon, in which values of resistance determined by the leakage method, might require appreciable correction.