The biogeochemical cycle of copper and cadmium has been studied in two forested ecosystems for two years. These two sites are: a) a Norway spruce forest (Picea abies) on an acid brown soil (FAQ: Dystric cambisol) and b) a mixed coniferous forest (Picea abies, Pinus cembra, Larix decidua and Abies alba) on a podzol. Both of them are located on low heavy metal polluted sites: the Swiss Plateau (850 m a.s.l.) for the first one and the oriental edge of the Mont-Blanc Massif (1650 m a.s.l.) for the second one. The Cu and Cd stocks in the vegetation and in the soil are evaluated. Heavy metal fluxes through the ecosystems are determined by analysis of concentrations and quantities of rain, throughfalls, stemflows, lysimetric soil solution and litterfall. In addition, the forms of Cu and Cd immobilization in soils are studied by means of sequential extraction, whereas immobilization forms in solution are studied by means of ion exchange chromatography and selective electrode. The results obtained allowed us to draw conclusions concerning the general pollution level in the experimental sites and the annual and/or detailed behaviour of Cu and Cd in the ecosystems. The conclusions are the following: As expected, the sites are not polluted since concentrations and fluxes are low. The chemical composition of solutions is modified during transition through the ecosystem, especially the copper concentrations. The acid brown soil humic layer is more efficient than the podzol humic layer in modifying this composition. There is no significant seasonal variation of concentrations and fluxes. However, the metal diffusion through the ecosystem is almost immediate. The soil plays an essential role in the immobilization of the heavy metals. The organic and organo-mineral layers are particularly active in this process. Exportations out of the ecosystems are limited. They are relatively more important for the podzol than for the acid brown soil and higher for Cd than for Cu. Transfer and stocking forms, and the ways of input to the soil depend on the metal considered: Copper input to the soil is mainly due to litterfall. In the soil, the metal has a preference to be associated with organic matter. In solution, it forms complexes with dissolved organic matter (mainly fulvic acid of MW < 1000 dalton). So, it is strongly retained within the soil and it cannot be easily drained out of the soil profile. At the opposite, cadmium is mostly transported by liquid means (throughfalls). In solution, it is present as free cadmium ions. In the soil, it is less strongly fixed than Cu since Cd is mostly associated with Fe and Mn oxides. For both of the metals, the free ion concentration in solution depends on pH and the saturation rate of organic matter. The latter has a heterogenous and polyfunctional character which explains a similar complexing capacity for all the samples studied. The differences observed in copper and cadmium behaviour can explain the variable diffusion of a polluting Cu and Cd flux through a given ecosystem and to the aquifer. The soil intervenes as a discriminating factor when the two experimental sites are compared.