Naturally anoxic waters containing Fe(II) and S(−II), and their synthetic analogues, have been shown to contain a species which is reduced at the mercury electrode at −1.1 V (SCE). A combination of field and laboratory studies involving several voltammetric techniques has shown that the observed signal is due to the reduction of both a solution species and a surface species which may be adsorbed from solution, or formed in situ at the mercury surface as HgS is reduced to liberate S(−II). Normal pulse polarography is the most elegant technique for separating the effects of HgS reduction and adsorption. When an initial potential more positive than −0.6 V is used, the polarographic signal reflects reduction of an iron sulphide species formed at the electrode surface due to the release of sulphide from the reduction of HgS. If the initial potential is more negative than −0.6 V, the surface species cannot form in situ, but there is a contribution to the polarographic signal from the adsorption of a soluble iron sulphide species. The soluble species forms slowly (minutes–hours) in solution, suggesting that its stoichiometry is not simple, and may possibly be polynuclear. However, the species must be small, as it diffuses as rapidly as Fe2+. A neutral species such as Fe2(HS)4 would be consistent with its ready adsorption to mercury. It is probable that this slowly forming polynuclear species is a precursor to the formation of colloidal and ultimately particulate FeS. When techniques that scan potential at a single mercury drop are used, such as cyclic and square wave voltammetry, the observed signal will contain a contribution from the reaction of Fe2+ with sulphide released at the surface from reduction of HgS. Although it is difficult to estimate the magnitude of this contribution, it will lessen as scan rates are increased. It vanishes if the initial potential is more negative than −0.6 V (SCE), or with differential pulse or dc sampled polarography at a dropping mercury electrode.