The utility of fishes a s biomonitors for mercury in aquatic environments is widely recognised, with skeletal muscle a s the tissue normally chosen for most monitoring programmes. The liver may also be a suitable candidate tissue for monitoring, as it concentrates many pollutants to higher levels than muscle, and is closely involved with processes of metal dynamics, storage, and detoxification. Little consideration has been given previously to the validation of the use of liver for mercury biomonitoring, with regard to quantifying natural temporal vanation. This study considered such variation in the eelpout Zoarces viviparusL., a resident estuarine fish species collected from a subtidal site near a formerly major industrial mercury discharge in the Forth Estuary, eastern Scotland Llver and skeletal muscle mercury concentrations, liver weight, and body size variables were measured for 196 individual eelpout in seven 3-month periods. Liver Somatic Index (LSI) was significantly hlgher, and liver mercury concentrations significantly lower, in the summer than In other seasons. Llver mercury burden (the mass of liver mercury per fish), in contrast, showed only llmited sign~ficant seasonal variation relative to body size, implying that the observed seasonal variation of liver mercury concentratlon (by around a factor of 3 between summer and winter) results largely from the dilution of simllar burdens by a seasonally growing and shrinking liver. Such natural seasonal variation may confound the use of the muscle/liver mercury concentration ratio, which has been proposed as evldence of fish migrat~on between waters with different levels of mercury contamination. These results also suggest that, as variation in liver mercury concentrations is closely related to seasonal vanation in LSI, liver t~s s u e may not be a reliable monitor for mercury in temperate species of marine fishes with seasonally fluctuating l~v e r size.