Human activities over the last several centuries have transferred vast quantities of mercury (Hg) from deep geologic stores to actively cycling earth-surface reservoirs, increasing atmospheric Hg deposition worldwide. Understanding the magnitude and fate of these releases is critical to predicting how rates of atmospheric Hg deposition will respond to future emission reductions. The most recently compiled global inventories of integrated (all-time) anthropogenic Hg releases are dominated by atmospheric emissions from preindustrial gold/silver mining in the Americas. However, the geophysical evidence for such large early emissions is equivocal, because most reconstructions of past Hg-deposition have been based on lake-sediment records that cover only the industrial period (1850-present). Here we evaluate historical changes in atmospheric Hg deposition over the last millennium from a suite of lake-sediment cores collected from remote regions of the globe. Along with recent measurements of Hg in the deep ocean, these archives indicate that atmospheric Hg emissions from early mining were modest as compared to more recent industrial-era emissions. Although large quantities of Hg were used to extract New World gold and silver beginning in the 16th century, a reevaluation of historical metallurgical methods indicates that most of the Hg employed was not volatilized, but rather was immobilized in mining waste.

Recent laboratory studies have demonstrated that environmentally realistic concentrations of dietary methylmercury can impair reproduction of fish. To evaluate relations between reproductive success and biomarkers of methylmercury exposure, we fed juvenile fathead minnows (Pimephales promelas) one of three diets contaminated with methylmercury: 0.06 (control), 0.87 (low), and 3.93 (medium) microg of Hg g(-1) dry weight. At sexual maturity, fish were paired, allowed to reproduce, and then analyzed for total mercury, plasma testosterone (T), and 17beta-estradiol (E2). Diets did not affect survival or growth of fathead minnows. Methylmercury suppressed levels of T in males and E2 in females. Male fathead minnows fed the control diet had mean T concentrations 20% and 106% greater than those fed the low and medium diets; control females had mean E2 concentrations 149% and 402% greater than those fed the low and medium diets. Methylmercury also inhibited gonadal development of females; the gonadosomatic index (GSI) of females fed the medium diet was 40% less than that of females fed control or low diets. Plasma levels of T in males and E2 in females were positively related to GSI. Methylmercury reduced the reproductive success of fathead minnows. Spawning success was 32% for pairs fed the control diet, 12% for pairs fed the low diet, and 0% for pairs fed the medium diet. Pairs fed the low diet required, on average, 5 d longer to spawn a clutch of eggs than the controls. Concentrations of methylmercury fed to fathead minnows in this study are also encountered by invertivorous and piscivorous fish in some methylmercury-contaminated aquatic ecosystems. This suggests that reproduction of wild fishes may be adversely affected by methylmercury and that suppressed hormone levels may be used to indicate diminished reproduction of fish.

Threshold concentrations associated with adverse effects of dietary exposure to methylmercury (MeHg) were derived from published results of laboratory studies on a variety of fish species. Adverse effects related to mortality were uncommon, whereas adverse effects related to growth occurred only at dietary MeHg concentrations exceeding 2.5 µg g(-1) wet weight. Adverse effects on behavior of fish had a wide range of effective dietary concentrations, but generally occurred above 0.5 µg g(-1) wet weight. In contrast, effects on reproduction and other subclinical endpoints occurred at dietary concentrations that were much lower (<0.2 µg g(-1) wet wt). Field studies generally lack information on dietary MeHg exposure, yet available data indicate that comparable adverse effects have been observed in wild fish in environments corresponding to high and low MeHg contamination of food webs and are in agreement with the threshold concentrations derived here from laboratory studies. These thresholds indicate that while differences in species sensitivity to MeHg exposure appear considerable, chronic dietary exposure to low concentrations of MeHg may have significant adverse effects on wild fish populations but remain little studied compared to concentrations in mammals or birds.

No abstract