Voltammetric studies on the oxidation of ferrocene in acetone + 0.1M Bu4NPF6 have been carried out at platinum and mercury microelectrodes over a wide temperature range which included the freezing point of mercury. The system is close to ideal at all electrode types. The dependence of the calculated diffusion coefficient on temperature obeys the Stokes-Einstein equation and enables a hydrodynamic radius of 3.2 Angstrom to be calculated for ferrocene which may be compared with a crystallographic radius of 2.7 A. This system, where the electrode is not intimately involved in the electrode process, provides reference data against which processes which do involve the mercury electrode may be assessed. It is known that the mercury electrode is involved in the electrochemical behavior of the Hg(Et(2)dtp)(2)/[Et(2)dtp](-)/[Hg(Et(2)dtp)(3)](-) system (Et(2)dtp = S2P(OEt)(2), O,O-diethyidithiophosphate). Under conditions where surface based effects are minimized, data are consistent with the reversible reaction scheme 2Hg degrees + 6[Et(2)dtp](-) reversible arrow 2[Hg(Et(2)dtp)(3)](-) + 4e(-) (process 1) and 2[Hg(Et(2)dtp)(3)](-) + Hg degrees reversible arrow 3Hg(Et(2)dtp)(2) + 2e(-) (process 2); overall Hg degrees + 2[Et(2)dtp](-) reversible arrow Hg(Et(2)dtp)(2) + 2e(-) irrespective of whether the mercury electrode is in the liquid or solid state. Thus, there is no abrupt change in the rate or nature of the electron transfer step at a mercury electrode at the freezing point of mercury in both non-interacting (ferrocene) and interacting systems. Data for the mercury dithiophosphate system were obtained at a conventional hanging mercury drop, a solid mercury drop, a dropping mercury electrode and at liquid and solid mercury coated platinum disk microelectrodes. Except for the dropping mercury electrode experiments, all data showed signs of a surface reaction, especially at low temperatures. Observation of the mercury microelectrodes by optical microscopy during voltammetric scans revealed that the non-ideality was caused by the formation of an insoluble product which passivated the electrode surface. Optical microscopy also confirmed that mercury microelectrodes formed on platinum disk substrates cover the entire surface and that a hemispherical model is appropriate for theoretical calculations. Digital simulations of the cyclic voltammograms obtained at a hanging mercury drop electrode are in moderate agreement with the proposed reaction scheme, with deviations from ideality being attributed to the presence of surface based reactions. Electrospray mass spectrometry of solutions containing equimolar concentrations of Hg(Et(2)dtp)(2) and [Et(2)dtp](-) confirmed the existence of [Hg(Et(2)dtp)(3)](-) in solution at ambient temperature.