The research for this project dealt with diffusive retardation: solute moving through a fracture diffuses into and out of the rock matrix. This diffusive exchange retards overall solute movement, and retardation both dilutes waste being released, and allows additional decay. Diffusive retardation involves not only fracture conductivity and matrix diffusion, but also other issues and processes: contaminants may sorb to the rock matrix, fracture flow may be episodic, a given fracture may or may not flow depending on the volume of flow and the fractures connection to the overall fracture network, the matrix imbibes water during flow episodes and dries between episodes, and so on. The objective of the project was to improve understanding of diffusive retardation of radionuclides due to fracture/matrix interactions. Results from combined experimental modeling work were to (1) determine whether the current understanding and model representation of matrix diffusion is valid, (2) provide insights into the upscaling of laboratoryscale diffusion experiments, and (3) help in evaluating the impact on diffusive retardation of episodic fracture flow and pore connectivity in Yucca Mountain tuffs.