The modeling of the surface of viscosity in composition and temperature at atmospheric pressure for methanol-water and acetonitrile-water was performed by interconnecting various elementary functions and testing them in a systematic manner to minimize the least-squares error, The systematic approach involved developing expressions that were the sum or product of elementary functions in temperature and composition, then visually observing their fit and quantitating the error. To reduce the error, transformation of the data using elementary functions was necessary to create a modified surface simpler in form. For MeOH-H(2)O, the least-squares error was 0.6 and 0.01 for untransformed data and transformed data, respectively. Similarly, for ACN-H(2)O, the error was 0.3 and 0.008 for untransformed and transformed data, respectively. The expression describing the viscosity-composition-temperature relationship for both the transformed and untransformed data was a quadratic in x with exponential functions of T as coefficients where T degrees is a nondimensionalizing term of value one. The transformed temperature is log(T/T degrees) + 2, while the transformed viscosity was n(1/8).