Transplantation of fetal neural cells represents an attractive replacement strategy for the treatment of certain neurodegenerative diseases. This is the case with Parkinson's disease, which results from a selective loss of dopaminergic neurons of the substantia nigra. Experimentation with animal models has demonstrated the feasibility of this approach. Grafting studies in patients have shown that intrastriatal implantation of solid grafts or cells obtained from human fetal mesencephalon usually results in a clinical benefit in patients. Despite continuous methodological progress, transplantation requires both conceptual and technical improvements. Current research aims at preventing the extensive death of donor dopaminergic neurons during the grafting procedure. However, the possibility of new sources of cells is currently being investigated. These include xenogeneic porcine neurons, or human cells programmed to produce dopamine or neurotrophic factors. A promising approach is based on the use of pluripotent stem cells derived from the brain, the bone marrow or early embryos. It is hoped that it will be possible to tightly control their proliferation and differentiation into dopaminergic neurons. Hence, it seems possible that transplantation will be widely used in the clinic in the future.