Magnesium is a commonly measured element in otolith chemistry analysis and is often included in the suite of elements used to discriminate fish from different environments. Poor relations between Mg in water and otolith chemistry are, however, often found. We examined the uptake of Mg into the otoliths of a freshwater fish (silver perch Bidyanus bidyanus), to determine the extent to which this element can be used to record previous environmental conditions. Silver perch fingerlings were reared for 30days in water with four concentrations of Mg (14.5, 36.6, 52.1 and 69.6mgL -1) and fed on a diet supplemented with a combination of natural Mg and enriched 26Mg at five concentrations (1496, 1626, 1902, 2005 and 2036?gg -1). Enriched 26Mg was added to the diet to achieve a 26Mg/ 25Mg ratio that was different from the natural ratio, such that the relative contribution of water and diet to Mg incorporated into otoliths could be determined. Enriching the diet with 26Mg resulted in an isotope shift in the otolith of silver perch from the natural 26Mg/ 25Mg ratio of 1.10-1.42; however, this was not as high as the ratio in the diet (> 3.7) suggesting that the fish did not fully incorporate Mg from the diet. Water was the primary source of otolith Mg, contributing on average>80% to otolith Mg (range 74-95%). The fact that Mg concentrations in the otolith did not change in response to Mg concentrations in the water or diet, indicates that Mg is likely physiologically regulated and therefore is not a reliable environmental indicator.