INTRODUCTION Molecular anthropology - the study of human genetic polymorphisms - is now often used to investigate the accuracy of archaeological and/or linguistic hypotheses. One of the classic examples is the use of genetics in an attempt to discriminate between two alternative models for the spread of agriculture in Europe - the demic and the cultural diffusion models - which finally led to a general approval of the former by geneticists, who regard this spread as possibly linked to the expansion of Indo-European languages (Ammerman and Cavalli-Sforza 1984, Renfrew 1992, Barbujani et al. 1995, Weng and Sokal 1995, Chikhi et al. 2002). More generally, because genetic clines can give evidence for population migrations (Barbujani 2000), the analysis of genetic patterns is particularly interesting for the analysis of early agriculturalist diasporas and their link to the diffusion of human languages (Barbujani and Pilastro 1993, Bellwood 2001). Molecular anthropology can also be useful in estimating the contribution of different gene pools to the make-up of present-day populations, when attempting to ascertain the origin of specific linguistic families (such as the Austronesian family, see further in this chapter); to test the permeability of linguistic boundaries to gene flow (Dupanloup de Ceuninck et al. 2000); or to investigate precise linguistic hypotheses (Excoffier et al. 1987; Poloni et al., this volume; this study), although genetics alone cannot be used to discriminate between alternative linguistic models. The present work aims at bringing genetic evidence to bear on the vexing question of East Asian linguistic relationships. The phylogenetic links between the main language phyla of this region (ST, AA, Tai-Kadai, AN and Altaic) are still deeply controversial (see the introduction to the volume for a review of the main theories). To investigate these relationships from a genetic point of view, we report here on the results of a population genetics analysis of one molecular polymorphism, HLA-DRB1. The DRB1 locus of the major histocompatibility complex (MHC) in humans is a cell surface protein-encoding gene, located on the short arm of chromosome 6 and surrounded by other HLA loci. Its allelic variability is amongst the highest known in the human genome thus far, with 418 DRB alleles detected by DNA oligotyping and sequencing techniques (IMGT/HLA sequence database 2003). Besides this high level of polymorphism, the DRB1 locus also has the advantage of having been extensively tested at the DNA level in human populations for at least 15 years (mostly using the HLA International Workshop typing kits), and abundant population data with high-resolution allelic definition are thus available. In this study, we analyse this polymorphism to explore a possible congruence between genetic and linguistic relationships in East Asia.