A two-well tracer test carried out in fractured chalk was analyzed using a three-dimensional finite-difference model for flow and transport which, was constructed on the basis of the geological and hydraulic information collected at the field site. The model was developed as a dual-porosity continuum model, in which advection was assumed to occur only in the fractures, and the water in the porous matrix was assumed to be static. The exchange of solute between the fractures (mobile phase) and the porous matrix (immobile phase) was assumed to occur as a diffusion process in response to the local concentration difference of solute between the two phases. Simulations from the dual-porosity model reproduced the shape of the observed breakthrough curves, although some portions of the tail were not accurately represented. The model was also applied as a single-porosity model for advection and dispersion in the fractures with no solute exchange with the porous matrix. The simulations from the single-porosity model greatly overestimated the observed lithium concentrations, and showed very little tailing effect in the falling limb. The study shows that, based on the given tracer test, solute transport in a fractured chalk cannot be represented by a single-porosity approach and hence when dealing with contaminant transport in such systems, both a fractured and a porous domain need to be considered.