As another comment says, the author of this New York Times article, Carl Zimmer, is a very solid science writer who knows a lot about evolution. I have occasion

As another comment says, the author of this New York Times article, Carl Zimmer, is a very solid science writer who knows a lot about evolution. I have occasion to discuss human population genetics with psychologists who study behavior genetics, and when population genetics issues come up in those discussions, some of the new discoveries are surprising even to them. It's hard for reporters, us, and even the working researchers to keep up with the new findings in human population genetics. I'll share a finding here that surprised some researchers I discussed it with this week. The first human genome to be sequenced was that of Craig Venter, who would be called a "white" man in the United States. The second human genome to be sequenced was that of James Watson, who is also regarded as "white" in either the United States or Britain. One of the first dozen or so human genomes to be sequenced was that of Korean scientist Seong-Jin Kim,[1] and his genome, even though "Asian," is more similar to Venter's than Venter's is to Watson's, and more similar to Watson's than Watson's is to Venter's.[2] Even though lineages are traceable in broad outlines by looking at human genes, there are quite a few human beings who resemble someone in someone else's group more than they resemble people in their own group. "The fact that, given enough genetic data, individuals can be correctly assigned to their populations of origin is compatible with the observation that most human genetic variation is found within populations, not between them. It is also compatible with our finding that, even when the most distinct populations are considered and hundreds of loci are used, individuals are frequently more similar to members of other populations than to members of their own population. Thus, caution should be used when using geographic or genetic ancestry to make inferences about individual phenotypes."[3] So what's going in these studies of early hominid DNA is that we necessarily have tiny sample sizes, and we actually have no idea how much genetic variation there was in the population we call Neanderthal, how much in the population we call Denisovan, and so on. We will want to sequence as much ancient DNA as we can, and meanwhile sequence more modern human DNA from more people from more places around the world. But we won't be able to trace back the ancestry relationships unerringly, no matter what we do. [1] http://genome.cshlp.org/content/early/2009/05/26/gr.092197.109 http://genome.cshlp.org/content/early/2009/05/26/gr.092197.1... [2] http://www2.webmatic.it/workO/s/113/pr-1611-file_it-Barbujani%20Pigliucci%20CurrBio.pdf http://www2.webmatic.it/workO/s/113/pr-1611-file_it-Barbujan... [3] http://www.genetics.org/content/176/1/351.full http://www.genetics.org/content/176/1/351.full