Indeed, HAR1F is rather mysterious. It codes only for an RNA, while most genes ultimately code for a protein. While other RNA coding genes have been identified -- RNA genes sometimes have a regulatory function, altering production of a protein -- this sequence resembles none of the known RNA genes. "The gene seems to be a gene in its own universe," says Lahn.

The gene is also distinct from previously identified genes that might play a role in brain evolution. Last year, Lahn's group identified two brain genes as being under recent evolutionary pressure. Unlike HAR1F, these genes have a known function -- they're both involved in control of brain size. In addition, different variations of these genes are found in different regions of the world, while everyone tested so far seems to carry the same 18 variations in the rapidly evolving chunk of HAR1F.

Haussler's group was able to identify the novel piece of DNA because of their unique approach to genome analysis. Previous studies focused on the coding regions of the genome: the section of DNA that directs the production of proteins. Haussler's group, on the other hand, did a genome-wide search for rapidly evolving sequences within the conserved regions of DNA. "Their method seems to be a very elegant way of picking up the potential regulatory elements that acquired unique functions in the human lineage," says Ganeshwaran Mochida, a neuroscientist and geneticist at Harvard Medical School in Boston.

However, some scientists are skeptical that the piece of DNA will turn out to be the key to human brain evolution. "It is a bit hard to fathom how a 118 base pair piece of RNA could have been so stone-cold evolutionarily, with two changes in the last 318 million years, then undergo 18 successive independent substitutions in the human lineage since the split with chimpanzee," says Andrew Clark, a population geneticist at Cornell University in Ithaca, NY. He suggests the recent changes may be the product of an unusual and not-yet-understood mutational process, rather than of evolutionary pressure. And he adds that no proof yet exists for either possibility. "In any event, the work serves as strong motivation to understand the biological function of this gene, and that has always been the excitement of this kind of work," he says.

Haussler and collaborators are now trying to figure out what role HAR1F plays during brain development. Among other experiments, they plan to create a mouse that expresses the human form of the gene. "We don't know if the mouse will start writing Shakespeare," says Haussler. "It's unlikely to function like a human gene because it's being expressed in the mouse cortex with other mouse genes, but it's something to try."

Haussler predicts the findings will spark a wave of research into the role of HAR1F. For example, he hopes that scientists who study the genetics of schizophrenia and other disorders will search their patient populations for mutations in the gene, which might shed light on its role in brain development and cognitive function.