With the purpose of subjecting the electronic method of contact potential measurement to a thorough test for consistency with independent photoelectric work function determinations, the Volta potential Ba-Mg has been measured for some 30 pairs of surfaces and the results compared with recent careful photoelectric studies of these metals. Each surface was prepared by fractionally distilling the metal in a gettered vacuum and revaporizing a middle fraction to form a thick film on glass at room temperature. Each film was measured a few seconds after deposition and tubes of two different designs used to minimize the possibility of errors originating in tube geometry. The majority of the observed Volta potentials fell within the range of values predicted by the photoelectric data, 1.08-1.16 v, and no values below this range were found. The magnesium films were largely responsible for variations in the Volta potential and a few gave potential settings sufficiently low (work functions sufficiently high) to raise the observed Volta potentials to a maximum of 1.26 v. Since all probable contaminations should lower the work function the maximum value is regarded as the most reliable. Assuming a work function of 2.52 ev for Ba this gives 3.78 ev for the work function of Mg. Observations on the optical reflection of the magnesium surfaces suggest, however, that a work function of 3.65\ifmmode\pm\else\textpm\fi{}0.05 ev may be characteristic of mirror-like surfaces of the metal; 3.78 ev of macrocrystalline (matte) surfaces.