1. A survey of the literature indicates that the apparent capacitance per unit area (Cm) of biological membranes is in general significantly greater than is that of 'artificial' phospholipid (black lipid) membranes (BLM). It is not possible, by quantitative arguments alone, to exclude that this simply reflects the idea that protein-containing biological membranes have a greater thickness than BLM. 2. The temperature-dependence of the membrane capacitance for both solvent-containing and solvent-free BLM is negative. However, where appropriate data are available, it appears that the capacitance of biological membranes has a positive temperature-dependence, indicating a qualitative difference between natural and artificial membranes. 3. Using a 4-terminal dielectric spectrometer, and the fitting program and electrode polarisation correction described in the accompanying paper, we have carried out a careful study of the temperature-dependence of the/?-dielectric dispersion of a unicellular eukaryote (Saccharomyces cerevisiae) and a prokaryote (Escherichia coli), 4. In the range 15-40 °C, the temperature-dependence of the/?-dielectric dispersion (and thus in principle of Cm) in S. cerevisiae and E. coli is respectively +0.13 and +0.35% (°C) -1. 5. Flow cytometric measurements indicate that the cell size of E. coli is unchanged in the temperature range studied. 6. These data strongly suggest that the/?-dielectric dispersion in cell suspensions is not due solely to the charging of a 'static' membrane capacitance. It is proposed that the positive temperature coefficient of the /?-dispersion reflects the contribution of temperaturedependent, partially restricted, lateral motions of the charged lipid and protein components of the cytoplasmic membrane.