and Zr. A substantial number of cases had values below detection limits for Ni, Sr, U, or Zr, and these elements were excluded from consideration. Element concentrations were transformed to log(10) values for statistical analyses. 2.2. Ceramic Petrography Petrographic analysis was conducted on a subset of 40 pottery thin sections representing the main ceramic composition groups and the full range of ceramic wares included in the study. Both qualitative and quantitative data were collected for each thin section using a standard petrographic microscope, following methods detailed in Minc and Sherman (2011). Qualitative analysis began with the identification of the overall suite of minerals evident in the thin section based on their optical properties under both plane-polarized and cross-polarized light. Quantitative data were collected using a point-counting technique similar to that described by Stoltman (1989, 1991). Using a movable stage, between 150 and 300 points were assessed within a grid of 1-mm intervals, and the point under the crosshairs recorded as matrix, void, or inclusion. If the crosshairs landed on an inclusion, the mineral was identified and the grain size recorded as silt (< .0624 mm), fine sand (.0625-.249 mm), medium sand (.25-.49 mm), coarse sand (.50-.99 mm), very coarse sand (1.00-1.99 mm), or gravel (> 2.00 mm) (following Stoltman, 1989:149, 1991:108). A grain size index based on an ordinal scale of 1 to 5 (fine = 1, medium = 2, coarse = 3, very coarse = 4, gravel = 5) was also calculated to represent the mean of the sand-and gravel-sized grains in each sample (Stoltman, 1991:108-109). 2.3. Assessing Temper vs. Natural Inclusions Determining whether mineral inclusions are naturally occurring or cultural additions has important implications for establishing provenance, since the addition of temper significantly modifies the trace-element and mineralogical signature of natural clays (Neff et al. 1988, 1989; Sterba et al. 2009). While it can be notoriously difficult to distinguish temper from natural inclusions, several lines of evidence have been proposed to identify cultural additions or mixing of clays. These include (1) the coexistence of minerals from distinct geological contexts; (2) a distinctive suite of minerals in the coarser sand-sized fraction as compared to the silt-and fine sand-sized fractions; and (3) the predominance of one size-class of inclusions, visible as a narrow unimodal or strongly bimodal size distribution within the sand-sized particles (