||P.246 right column bottom paragraph: "A key question is how many binding sites for a particular protein exist in the E. coli genome. One approach for establishing an upper bound is to determine the number of protein molecules in the cell, although it is believed that DNA-binding proteins are present in excess of the number of biologically relevant sites (Berg, 1978, 1988a). [Researchers] have identified such values from the published literature and from a 2-D electrophoresis survey in [their] laboratory (Link et al., 1997) for 16 of the proteins in [their] study (Table 4). Comparison of [researchers’] predictions to the observed protein abundances suggests that many of our matrices are underspecific- many more sites are predicted than are expected from protein abundance levels. There is no significant correlation between [researchers’] predictions and the observed abundances. However, there is also no consistency in the conditions under which the experimental measurements in Table 4 were performed." Note-table gives value of 200 ArcA copies/cell, primary source Link et al 1997 PMID 9298646. BNID 105139 gives value of 40 copies/cell.