Beyond its theoretical impact, the rise of molecular biology has brought about great progress also in the historical classification of species and populations. Sokal (1988) and Cavalli Sforza et. al. (1988, 1994) advocated a correspondence between some proposed language families and classical genetic markers, i.e. between biological evolution and transmission of cultural traits (languages), in agreement with Darwin’s (1859) prediction of an eventual isomorphy between trees of languages and populations. Their conclusions were variously criticised, and are especially undermined by the weakness of available language classification methods, mainly based on lexical (in a broad sense) material: these are either too narrow in scope (classical method) or probabilistically unreliable (Greenberg’s mass comparison) to provide solid long-range taxonomies to be matched against those of geneticists. We will address such issues with a radically new comparative method (Longobardi and Guardiano 2009) based entirely on grammatical evidence and on recent theoretical advances in formal and typological linguistics. The method ensures unprecedented standards of testability/replicability and can measure linguistic distances between even remote populations. Linguists and biologists will for the first time jointly select the populations most significant from either perspective for language/gene sampling. We will also adopt the newest powerful genetic tools of the last decade, both in terms of data (genome-wide studies of Single Nucleotide Polymorphisms, including uniparentally transmitted markers of the Y chromosome and of mitochondrial DNA) and of biostatistical methods to describe DNA diversity and interpret its relationships to linguistic diversity.