It has been theoretically and experimentally shown that rate-limited sorption/desorption can have a profound effect upon the transport of sorbing contaminants. The advection/dis-persion equation that has been traditionally used to model contaminant transport uses a retardation factor to account for sorption, thereby implicitly assuming local equilibrium between contaminant in the sorbed and aqueous phases. This assumption fails to consider the possibly large effects of rate-limited sorption/desorption. The mass release characteristic of arsenic-contaminated soils at the Crystal Chemical site in Houston, TX, was examined. Soils were collected from beneath two former waste-water ponds that were the source of arsenic in the uppermost aquifer. Samples were typical of those found within the fine-grained components of local alluvial overbank deposits that comprise the bulk of the site. The dynamic test applied a continuing head of water, operating in an upflow mode, through 4-inch-diameter by 12-inch-long soil columns repacked to in-situ density. Three columns were constructed The mass release characteristic of arsenic-contaminated soils at the Crystal Chemical site in Houston, TX, was examined. Soils were collected from beneath two former waste-water ponds that were the source of arsenic in the uppermost aquifer. Samples were typical of those found within the fine-grained components of local alluvial overbank deposits that comprise the bulk of the site. The dynamic test applied a continuing head of water, operating in an upflow mode, through 4-inch-diameter by 12-inch-long soil columns repacked to in-situ density. Three columns were constructed— two containing predominantly clay, and the other containing clayey silt. Leachate from the most permeable column was collected over 42 pore volumes (equivalent to 120 years of extraction). Sharp declines in arsenic concentrations in the leachate were measured after just four pore volumes. A biphasic response was evident, consistent with published research on kinetically limited mass transfer of retarding solutes. The most responsive column was pulsed to elucidate the effects of diffusion and pulsed pumpingtwo containing predominantly clay, and the other containing clayey silt. Leachate from the most permeable column was collected over 42 pore volumes (equivalent to 120 years of extraction). Sharp declines in arsenic concentrations in the leachate were measured after just four pore volumes. A biphasic response was evident, consistent with published research on kinetically limited mass transfer of retarding solutes. The most responsive column was pulsed to elucidate the effects of diffusion and pulsed pumping. Arsenic concentrations returned to baseline levels in less than three pore volumes. These studies ultimately led to a joint assessment between Southern Pacific Lines (SPL) and EPA Region VI, which concluded that extraction and treatment of the shallow aquifer beneath the site was not feasible, in light of the aggressive restoration goal.