The differential capacitance measurement for determining the height of a Schottky barrier at a metal—semiconductor contact is based upon a number of assumptions. In practice, one or more of these assumptions may not be valid. Some of these deviations from the ``ideal case'' are examined in order to determine the effects of each upon the interpretation and validity of measurements on such contacts. Specifically, the effects of each of the following are considered: series resistance, traps in the depletion layer, effective contact area variation with depletion layer width, an insulating interfacial layer between the metal and semiconductor, semiconductor surface charge variation with bias voltage, and the reserve layer at the edge of the barrier. Experimental data for gold plated contacts to conducting cadmium sulfide single crystals are given to illustrate some of the results.