Recent research has recognized the utility of the peat record in ombrotrophic bogs to reconstruct past changes in the atmospheric deposition of, for example, lead and mercury. Frequently, these reconstructions rely only on single peat records to make assessments of regional deposition rates. The surface of a bog is not a uniform feature; rather, vegetation and micro-topography vary over small spatial scales, which can affect the interception and retention of atmospheric deposition, and over relatively short timescales. Analyses of nine hummock cores collected from a 2000 m2 area on one bog, Store Mosse, show that concentrations and cumulative inventories (past 110 years) of lead and mercury vary by a factor of 2 for lead (0.75 to 1.40 g Pb m-2, three cores) and 4 for mercury (0.85 to 3.4 mg Hg m-2, nine cores). The range of values within Store Mosse is greater than the difference between any one core and the metal inventories from either Dumme Mosse (0.95 g Pb m-2 and 2.0 mg Hg m-2) or Trollsmosse (1.6 mg Hg m-2), each about 60 km distant. Although the general temporal trends are similar in all cores, our data indicate that single reconstructions do not necessarily provide a representative flux for the bog as a whole. To overcome the potential constraints of single records, we suggest incorporating data from multiple sites or at least multiple cores in order to scale up to regionally valid models of past metal deposition.