The availability of Neanderthal genomic sequence data is allowing us to date key events in our own evolution.

So far, only one gene has been convincingly associated with human language and speech — forkhead box P2 ( FOXP 2 ). Despite being a highly conserved gene, the human and chimapanzee versions differ at two positions in exon 7, and these substitutions have been implicated in our unique ability for speech.

Krause et al. found both substitutions in two Spanish Neanderthal samples. Further, they concluded that, for both substitutions, at least one of the individuals was homozygous for the derived allele that is predominant in modern populations.

Next, the authors analysed the Neanderthal samples for evidence of a selective sweep close to exon 7 of FOXP2 that, from modern human genetic diversity data, was previously proposed to have occurred within the past 200,000 years. Seven polymorphic regions from the intronic region upstream of exon 7 were successfully amplified from the Neanderthal samples and, for six of these, all products from both Neanderthals represented the derived allele. These results suggest that the selective sweep in the FOXP2 region began before the split from the human–Neanderthal common ancestor, which existed 300,000 to 400,000 years ago — much earlier than previously thought.

So, Neanderthal sequences can provide a useful tool to investigate our own evolution, making the extreme care that must be taken when retrieving genetic information from ancient samples worthwhile.