Dental wastewater (DWW) was collected over two months from a 12-chair clinic and a single-chair office to identify conditions that may affect Hg methylation. DWW was settled for 24 h and samples were collected from the top and bottom of the supernatant to simulate a range of particles that may escape in-line traps. Total Hg spanned 5 orders of magnitude (0.02-5000 microM), following a log-normal distribution with p10, p50, and p90 concentration values of 0.24, 31 and 4000 microM, respectively; typically well in excess of free aqueous Hg solubility. Methyl Hg was present in high levels (2-270 nM), also following a log-normal distribution with p10, p50, and p90 concentration values of 2.8, 17, and 100 nM, respectively. There were no statistically significant differences (90% CI) in p50 methyl Hg or total Hg between the clinic and office. Methyl Hg was predicted from total Hg data by (+/- 95% CI): Log (Me-Hg) = 0.33 (+/- 0.06) x Log (T-Hg) - 2.27 (+/- 0.13). Total methyl Hg from DWW to U.S. wastewater collection systems is estimated to be 2-5 kg yr(-1). Equilibrium speciation modeling predicted that DWW Hg was primarily in sulfide-Hg complexes, except at high total Hg levels where organo-Hg complexes become significant. DNA extracts amplified by quantitative polymerase chain reaction with primers for total eubacteria and sulfate-reducing bacteria (SRB) indicated that the total eubacterial DNA was composed primarily of SRB, and highly significant correlations were found between methyl Hg and both amplified Desulfobacteraceae (p < 0.0001) and Desulfovibrionacaea DNA (p < 0.00001). Both are known Hg methylators. In marked contrast, there was no significant correlation between methyl Hg and amplified Desulfobulbus DNA, a genus generally not known to methylate Hg at high rates. These results strongly suggest that SRB are implicated in DWW Hg methylation.