Two views of tailing in pump-and-treat systems, which can significantly increase remediation time, were put forth by Malone et al. [1993] and Wise et al. [1992]. The former paper, based on the failure of LEA to predict tailing and the desire to predict dissolution rates observed in long-term studies, split NAPL into two phases in a model, one with fast mass exchange with water and the other slow. The latter contended that the inability of LEA to explain tailing had been based on linear partitioning relationships, and that this could be remedied within LEA by use of favorable nonlinear ones. The study was based on chromatographic analysis of published laboratory data on leaching of NAPLs into a flushing water phase.