Huntington's disease is a genetic disease, autosomal and dominant, that induces motor disorders, an inexorable deterioration of higher brain functions and psychiatric disturbances. At present, there are no known therapeutics against Huntington's disease. The Network of European CNS Transplantation and Restoration (NECTAR) has begun a program aimed at defining the conditions under which intrastriatal transplantation of fetal striatal cells could be attempted as an experimental treatment for Huntington's disease. This review presents the reasons why our group is considering participating in these trials. The validity of this therapeutic approach is supported by three main series of data: (i) neuropathological, clinical and imaging data indicate that Huntington's disease is, above all, a localized affection of a specific neuronal population ("medium-spiny" neurons) in the striatum; (ii) a large body of experimental results, obtained in rats and non-human primates, demonstrates that transplanted fetal striatal cells are able to integrate the host brain and to substitute for previously lesioned host striatal neurons; (iii) expertise in clinical neural transplantation has now been acquired from the treatment of patients with Parkinson's disease. These different sets of data are presented and discussed in this review. There are a number of problems which do not yet appear to be entirely resolved, nor are they likely to be using the experimental models currently available. These problems are identified and explicitly presented as working hypotheses. (1) Anatomo-functional results obtained in rodents and non-human primates with excitotoxic striatal lesions can serve as a basis for the extrapolation of what can be obtained from patients with Huntington's disease. (2). Huntington's disease can be efficiently fought by substituting degenerated striatal neurons alone. (3) Huntington's disease is due to a genetic defect which either hits the neurons that carry it directly or hits them indirectly only after several decades. Transplanted neurons, because they do not carry the gene or because they are of fetal origin, will not be rapidly affected by the ongoing disease process. Given the current state of knowledge, intracerebral transplantation appears to be the most serious opportunity (if not the only one that has been experimentally validated) for clinical improvement to be obtained in patients with Huntington's disease. The purpose of this review is to open a scientific discussion on its experimental bases before actual clinical trials start.