344 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS given to the w/re and then locked in position. The lower cylinder lock is released, a stop-watch started, and the difference between the positions of the upper and lower needles is noted periodically. From the shape of the time-decay of torque (stress) curve and the final equilibrium difference of angles, semi-quantitative information of the growth of plastic viscosity and static rigidity may be estimated (Fig. 11). It will be appreciated that since measurements are not made purely statically, the relative rigidity as deter- mined by the end position is smaller than the true value. For fairly rough determination of rigidity the thixotrometer serves as a handy' instrument for comparing rigidities of different specimens and their rate of thixotropic build-up. Time -* Fig. 11 Fig. 10. Pryce-Jones •hixotrometer. INFLUENCE OF CONSTITUTIVE FACTORS ON RHEOLOGICAL DATA Increase in concentration of disperse phase effects an increase in: (a) plastic or residual viscosity, in proportion to the specific surface (or degree of dispersity) of the disperse phase (b) the curvature of the stress/shear rate line (c) the rigidity as measured by thixotrometer, penetrometer, or parallel- plate plastometer.

INTRODUCTION TO THE RHEOLOGY OF DISPERSE SYSTEMS 345 The degree to which such non-Newtonian properties are conferred decreases markedly as the dispersing power of the continuous phase for the disperse phase is increased. Thus, a stiff, rigid paste decreases in plastic viscosity, curvature of the line, and in rigidity, when a deflocculator is added. The rate of thixotropic regain after deflocculation by shear increases with decrease in viscosity in the dispersing medium, but this again is modified by the degree of the deflocculating power of the medium, thixotropic regain being very slow when this degree is high. Finally, a few words on the application and interpretation of rheological data. In the first place, rheological figures afford, perhaps, a unique means of classifying the properties of plastic bodies, checking possible variation from a known standard, and observing stability on storage by periodical testing. Thus, changes in residual viscosity and rigidity clearly indicate internal structural changes arising from coalescence, coagulation, gel formation, etc. A study of the rheologicM data of materials subjectively recognised to exhibit marked characteristics of shortness, heavy body, stickiness or tack, etc., often permits of short cuts to formulation of those systems, the subjective properties of which depend not only on quantitative but on structural com- positions. The rheological curve has also proved useful in determining the efficacy of different types of mixers, grinders and emulsifiers. APPENDIX Whilst the interpretation of rheological data into basic units demands somewhat involved mathematical treatment, the older de Waele-Ostwald equation • can be usefully employed for most practical purposes. It may be stated that in spite of the publication of the writer's later dimensionally sound equation, • many rheologists still prefer to use the earlier treatment as yielding usefully indicative information. Reduced to its simplest iorm, the de Waele-Ostwald equation is ß Shearing stress (Effiux velocity)4 = O' 4 being a positive exponent less than unity, (1 -- 4) being a measure of the non-Newtonianism of the material and •' a coefficient of flow. •', not possessing the dimensions of a true viscosity •, must not be regarded as anything but a coefficient of flow. It is thus only necessary to plot log shearing stress against effiux velocity, when ß /k log shearing stress • log effiux velocity