Previous rules of allocation of livers for transplantation were based mainly on local priorities, with final management left to the local team. This created substantial regional disparities. A prospective survey of waiting list deaths and dropouts due to aggravation of liver disease (2003-2005) validated the MELD (Model for End-stage Liver Disease) score on French data. A new allocation score (Liver Score) for liver transplants, based on specific variables for each liver disease, was introduced in March 2007. An initial evaluation, based on the first 5 months of practice, clearly shows that the Liver Score reduces the rates of deaths, dropouts, and futile transplantations; it also accelerates access to transplantation for the sickest patients. Several points remain unresolved: both the MELD and Liver scores may be improved. The variability of the MELD score related to different laboratory assay methods requires harmonization between laboratories.