A series of laboratory tracer migration experiments in a single rock f
racture have been performed, and the breakthrough curves have been int
erpreted using mathematical modelling. Discrepancies were observed bet
ween the experimental data and the predictions made using a simple adv
ection-dispersion model. The potential reasons for these discrepancies
have been investigated by applying more complex models: one model inc
orporates channelling of flow within the fracture, the other couples d
ispersion and advection in the fracture with rock-matrix diffusion. It
is concluded that channelling of flow can adequately explain the obse
rved spreading behaviour; rock-matrix diffusion is not a significant m
echanism influencing transport in these experiments.