The very high concentration of macromolecules within cells can potentially have an overwhelming effect on the thermodynamic activity of cellular components because of excluded volume effects. To estimate the magnitudes of such effects, we have made an experimental study of the cytoplasm of Escherichia coli. Parameters from cells and cell extracts are used to calculate approximate activity coefficients for cytoplasmic conditions. These calculations require a representation of the sizes, concentrations and effective specific volumes of the macromolecules in the extracts. Macromolecule size representations are obtained either by applying a two-phase distribution assay to define a related homogeneous solution or by using the molecular mass distribution of macromolecules from gel filtration. Macromolecule concentrations in cytoplasm are obtained from analyses of extracts by applying a correction for the dilution that occurs during extraction. That factor is determined from experiments based upon the known impermeability of the cytoplasmic volume to sucrose in intact E. coli. Macromolecule concentrations in the cytoplasm of E. coli in either exponential or stationary growth phase are estimated to be approximately 0.3 to 0.4 g/ml. Macromolecule specific volumes are inferred from the composition of close-packed precipitates induced by polyethylene glycol. Several well-characterized proteins which bind to DNA (lac repressor, RNA polymerase) are extremely sensitive to changes in salt concentration in studies in vitro, but are insensitive in studies in vivo. Application of the activity coefficients from the present work indicates that at least part of this discrepancy arises from the difference in excluded volumes in these studies. Applications of the activity coefficients to solubility or to association reactions are also discussed, as are changes associated with cell growth phase and osmotic or other effects. The use of solutions of purified macromolecules that emulate the crowding conditions inferred for cytoplasm is discussed.