Understanding the main phenomena involved in the controlled-release kinetics of herbicides in a water bath is a very important requisite for diffusive-parameter estimation, because, some mathematical models based on Fick's second law for diffusion have been developed to describe the controlled-release kinetic data. However, the validity of these models is restricted to the following assumptions: (1) the formulation is an isothermal slab; (2) the release occurs through the two faces of the slab; (3) the herbicide is dissolved in the water contained in the slab pores at a concentration less than the saturation concentration (c(is)); (4) the total sum of the individual volumes of the pores is epsilonAL (epsilon is the slab porosity, A is the slab area, and L is the slab thickness); and (5) the initial concentration of herbicide in the pores is M-0/epsilonAL (M-0 is the initial amount of herbicide in the matrix). The fourth assumption may be invalid for mathematical description of systems in which the herbicide concentration in the slab may be above the saturation concentration. If this were true, the final assumption would also be invalid, because the initial concentration of herbicide in the pores is c(is) in this case. This work presents a study of the solubility effect on the controlled-release kinetics of herbicides from lignin matrices.