Chinese fossil finds and the lamprey genome both provided insight into a quest to understand the evolution of fins. In the course of their research, developmental biologist Martin Cohn, based at the University of Florida, and his graduate students Renata Freitas and GuangJun Zhang even attempted to fluorescently label shark embryos. Their efforts identified a surprise 'recycling' of genetic mechanisms involved in the formation of median and, much later in evolution, paired fins (see page 1033).

Before this, most fin-development work had focused on the zebrafish. But Cohn and his colleagues wanted to understand the gene-expression patterns that give rise to paired fins, so they turned to the shark, the most primitive living vertebrate with such fins. During their initial gene-expression screen, they noticed that genes expressed in paired fins are also active in median fins.

While they were redirecting their efforts to study median fins, an exciting find of fossil fishes in China confirmed that median fins arose about 100 million years before paired fins. This suggested that median fins might hold the key to the origin of fins and limbs, and, therefore, the first steps towards vertebrate locomotion.

Those interested in locomotion had paid little attention to median fins, but one previous study of zebrafish had determined that neural-crest cells — embryonic cells that give rise to cranial skeletal and connective tissues — are involved in median fin development. “That one experiment gave rise to the idea that neural-crest cells form the median fins,” says Cohn. To determine whether neural-crest cells and/or somitic cells — embryonic cells that develop into muscles and vertebrae — have a role, the researchers first attempted to microinject shark eggs with a fluorescent label.

But working with unusual animal models often introduces time-delaying technical hurdles. Sharks are seasonal breeders, which limits the availability of embryos. And, as Cohn and his colleagues found, embryos could not survive the saltwater intrusion that occurred during introduction of the fluorescent label.

So they pursued a molecular approach, cloning various genes that mark distinct cell types to determine which give rise to median fins. They found that although neural-crest cells do make a contribution, this is minor; somitic cells form the bulk of the fin. Cohn was most surprised to find that the genetic programme for fin development was first assembled in the somitic cells, then later in evolution redeployed to the tissue that gives rise to paired fins and limbs.

The scientists went on to document gene expression in lampreys, eel-like organisms with median fins but no paired fins. The shark and lamprey straddle the origin of paired-fin nodes along the evolutionary time scale. The lamprey's median fins also proved to originate from somitic cells.

The next question for Cohn's group is 'how old is fin development?' To find out, they'll study median fin development in amphioxous, the closest living invertebrate relative of vertebrates. “What's been most exciting for us is the flow of information between developmental biology and palaeobiology,” says Cohn, adding that the strength of evolution and development research is the integration of a wide range of disciplines in an effort to address evolutionary questions.