Randy McDonald points me to this fascinating post, Genetic clues to the Ossetian past. In the post author outlines phylogeographic inferences one can make from uniparental lineages; maternal and paternal lines of descent. Specifically, they are in interested in the origins and relationships of the Ossete people. I assume that one reason Randy pointed me to this post is that the Ossetes are assumed by many to be the descendants or fragments of the Alans. More broadly they're remnants of a broad array of North Iranian peoples, of whom the Scythians were the most prominent, which have been erased from the pages of history because of the expansion of the Slavs and Turks. The main lacunae in the above analysis is that it does not cover results from autosomal studies. Some of that has been performed by Dienekes, but more is necessary for a region characterized by as much ethnographic diversity and density as the Caucasus. One peculiarity that emerges in analyses of autosomal data sets is that the Caucasus looms relatively large in a wide array of dispersed populations. For example, there is a genetic signature which ties Indo-Aryan and Caucasian populations together, and others which seem to connect the latter to some Balkan groups. These are possible hints that the Caucasus is the "mother of nations," and that the old idea of the "Caucasian race" may have some reality to it. But I would bet on something else: the Caucasus is not the mother of nations, but the repository of forgotten peoples. The Ossetes themselves are presumed to be just such a population. I offer up the hypothesis that one reason that disparate Caucasian populations have diverse and wide-ranging connections has less to do with outward expansion, and more to do with the fact that on the margins of the Caucasus a great range of historic genetic diversity erased by later demographic events (e.g., the Slavic and Turkic expansions from two directions in on the North Iranian peoples) is preserved, as the defeated take refuge.