A preliminary account of metal levels in eggs of farmed and wild Atlantic salmon and their relation to egg viability

Abstract

Levels of copper, zinc, iron and manganese in newly fertilised eggs of Atlantic salmon, Salmo salar, were determined during the 5 years 1982–1986. Eggs of 97 wild fish from two rivers and 173 farmed fish from seven farms in Scotland were examined. Egg copper levels were consistently 4.8–5.6 times higher in wild salmon than in farmed salmon. Egg zinc levels were usually significantly higher in farmed salmon. Iron levels did not usually differ, and manganese levels were too low for meaningful comparison. Metal levels varied significantly between farms and between rivers. There were small but significant differences in egg metal levels between farmed fish of 1, 2 and 3 seawinters. Liver copper levels were higher in farmed salmon than wild salmon, but liver zinc and iron levels did not differ. Between fertilisation and the end of the yolk-sac stage (start of feeding), the quantity of copper in wild eggs decreased 10-fold, that of farmed eggs two-fold, so that at the start of feeding both groups had similar amounts of copper. There were no such changes in quantities of zinc and iron. Nine correlations, between initial egg levels of copper, zinc or iron, and egg mortality at three successive stages of incubation, were sought in eggs from three sources. None of the nine correlation coefficients was significant, either for wild eggs, or for eggs from a farm with very high egg mortality attributable to other causes. However, two of the nine correlation coefficients were significant in eggs from a farm with low egg mortality where other circumstances were favourable: a positive correlation between zinc level and mortality up to the eyed stage ( r=0.50, P < 0.05) and a negative correlation between copper level and mortality between the eyed stage and hatching ( r= −0.51, P < 0.05). The correlations suggest that unnaturally high zinc and low copper of farmed eggs lead to increased egg mortality. However, both correlations were weak, and factors such as time of stripping, and salinity in which broodstock are held, are known to exert much greater influence on egg viability.