Recent studies of the host rock palaeomagnetism of the lead-zinc deposits on the Cévennes margins pointed towards regional fluid circulation from the early to middle Eocene. The hypothesis has therefore been put forward that mineralising fluids might have migrated as a consequence of the Pyrenean uplift. Based on this assumption, a digital model was developed to describe this palaeo-circulation along two reconstituted cross-sections, in early Eocene times. One of them extends from the Gulf of Lion to the Les Malines deposits at the southern end of the Cévennes mountains; the other one connects the Montagne Noire to Les Malines in order to test the hypothesis of a more localised fluid circulation. The modelling of heat and fluid circulation along these cross-sections is constrained mainly by fluid-temperature data, derived from analyses of fluid inclusions. The maximum recorded temperatures are about 150°C. The METIS code (Ecole des Mines, Paris) was used to test the transport scenarios while prescribing hydrodynamic characteristics in the series that would allow fluid flow. Gravity-driven flow is initiated at a high point, either the Montagne Noire or the Pyrenees. Drainage occurs at depth. The permeable formations concerned are: Cambrian dolomite in the cross-section beginning in the Montagne Noire, and Triassic and Liassic carbonate or sandstone formations in the other one. The fluids converge at the deposit site through faults on the margins of the Cévennes horst. The highest temperatures reproduced by the digital simulations in a steady-state regime are in the order of 80°C at the deposit site for each pathway. A sensitivity test showed that higher temperatures, in the order of 150°C, could only be reached with a heat flux of 120 mW.m−2 and by optimising such parameters as permeability, aquifer geometry and thermal conductivity. However, such a parameter set does not seem geologically feasible. The modelling demonstrates that circulation must have occurred at greater depths in the case of gravity-driven fluid flow. The most probable explanation is that the fluid migrated in the deep crustal basement and that, during its ascent along the faults bordering the Cévennes heights, it mixed with basinal brines migrating through shallower aquifers.