1. Simple thermodynamic expressions are used to describe the properties of uncharged binary and ternary polymer solutions, in particular the sedimentation equilibrium of binary systems and the osmotic pressures and ;incompatible' phase separations of ternary systems. 2. Sedimentation-equilibrium experiments were performed on four samples of dextran and two of polyethylene glycol. The critical points of the phase diagrams were determined for the mixed solutions of polyethylene glycol-dextran-water and of polyethylene glycol-bovine serum albumin-0.2m-sodium chloride solution. Osmotic pressures were measured on a single-phase mixed solution of a polyethylene glycol and a dextran. By use of the simple thermodynamic expressions consistent values of second virial and interaction coefficients for the materials used were obtained from these experiments. 3. The interpretation of the values of the second virial and interaction coefficients, on the basis of three models of molecular interaction, is discussed.

1. Sephadex in bead form shows reversible changes of inner volume when immersed in solutions of a non-penetrating solute. These changes are in accordance with the theory of Flory (1953) for the swelling of gels. This makes possible the use of single beads for measuring the osmotic pressures of polymer solutions, up to or beyond 1kg./cm.(2). 2. Measurements of the inner volume of Sephadex G-200 by equilibrium dilution, at various concentrations of dextran 500, gave values in agreement with those obtained from the dimensions of single beads. 3. Measurements by chromatography of the inner volume of Sephadex G-200 gave values that differed slightly but significantly from those obtained by the other methods. Similar small disagreements were found between the dilution and chromatographic values for bovine serum albumin measured in the presence of various concentrations of dextran 500. The major factor that affects the dependence of the chromatographic elution volume (of dextran 500) on concentration is shown to be the change of the inner volume of the Sephadex, with dynamic factors playing a minor role. 4. Contrary to the findings of Ackers (1964), the equilibrium dilution and chromatographic methods gave closely agreeing values for the distribution coefficient, K(av.), of bovine serum albumin on Sephadex G-200; this cast doubt on his explanation of the nature of chromatographic separations on Sephadex.

(1) We have measured the incompatible phase separation that occurs in a polyethylene glycol-sodium dextran sulphate-sodium chloride-water system and have determined a critical point. (2) We have measured the activity coefficients of sodium chloride in critical-point concentrations of polyethylene glycol and sodium dextran sulphate respectively, and the osmotic coefficient of sodium dextran sulphate at the critical-point concentration. (3) With use of the relevant thermodynamic equations for a quaternary ionic system, we have determined the interaction coefficients between polyethylene glycol and dextran sulphate and between polyethylene glycol and sodium chloride. The former could be due mainly to volume exclusion, but the latter is too large to be explained on that basis.