It was the scope of this project to collate, compile and summarise the scientific literature on the toxicity of methyl mercury, inorganic mercury and total mercury in humans and experimental animals published since 2002 and not evaluated in former JECFA risk assessments. Epidemiological data on MeHg toxicity confirmed former findings on an association between prenatal exposure and developmental neurotoxicity. Recent findings also support an influence of methyl mercury on the cardiovascular and immune system. Existing epidemiological data reveal several shortcomings limiting their suitability for risk assessment. Developmental neurotoxicity seems to be the most critical effect of MeHg in animals. A LOAEL of 0.01 mg/kg b.w. per day has been reported in young mice for locomotor activity exposed in utero. Rats seem to be less sensitive towards this endpoint: a NOAEL of 0.04 mg/kg b.w. per day has been reported in this species for locomotor activity. Systemic effects other than neurotoxicity have been mostly observed at higher doses than those producing neurotoxicity. No qualified epidemiological data appropriate for the evaluation of the toxicity of inorganic mercury after oral intake have been identified due to several limitations of the studies, e.g. small study group, insufficient control for confounders, inadequate exposure assessment. Toxicological studies in experimental animals widely confirmed former findings with respect to the target organs (neuro-, liver-, kidney-, immune-and reproductive toxicity, oxidative stress). The animal studies indicate that there might be relevant toxicological effects (developmental toxicity, liver toxicity) of inorganic mercury at dose levels similar or below the BMDL 10 value used for the derivation of the existing PTWI. However, due to some limitations of the studies no final conclusions can be drawn from these studies and the findings need further confirmation.