Even in their quietest and stillest moments, forests are places of fierce competition. Sunlight is the one of the most precious commodities here, and plants jostle, circumvent, and kill each other for prime positions beneath the incoming rays. Ferns are masters of this game; they excel at growing in the shade. Fay-Wei Li from Duke University thinks that their success depends on a single moment that happened around 180 million years ago, when an ancient fern stole a gene from another plant.

Ferns are sometimes portrayed as relics of an earlier phase of plant evolution, which were outcompeted by flowering plants and relegated to the bottoms of forests. But that’s not the case. Ferns may be an ancient group (they first arose 360 million years ago), but the vast majority of living members arose much later, during the Cretaceous period. By that time, flowering plants were already dominating the world. Ferns weren’t plucky holdouts consigned to some scrapheap of existence; they diversified in the shadows of other plants.

These lineages had a tool that almost all other plants (and indeed, other ferns) lack—a light sensor called neochrome. Most plants move towards sources of light using molecules that are sensitive to blue light, although some use red-light sensors instead. But neochrome is incredibly sensitive to both blue and red. That gives ferns an advantage because blue light is largely filtered out by the upper layers of a forest, while red light penetrates more deeply. Using neochrome, ferns could ‘see’ better in a shady, flower-filled world.

But where did neochrome come from? Li decided to find out. “I love ferns and I want to know why there are so many of them,” he says. “Neochrome seemed to be a great starting point to me, so I just decided to figure out its evolutionary history.”

That was easier said than done. When Li started, there weren’t a lot of complete plant genomes on record, so he didn’t have a lot of data to work with… and what he had made no sense. “I remember walking into my advisor’s office one day and telling her my PhD is doomed because I couldn’t figure this out,” he says. Salvation came from the One Thousand Plant Project—a massive initiative to study how plants, from algae to flowers, use their genes. Suddenly, Li had data galore. He wrote a programme to analyse it, “and one night, my Macbook terminal told me that it found a neochrome-like sequence in hornworts.”

Hornworts are usually found in greasy, blue-green mats, growing in damp or humid places. They’re even older than ferns, and were among the first plants to colonise the land.

It’s possible that the common ancestor of hornworts and ferns already had neochrome, and many of its descendants—including all trees and flowers—then lost this molecule. Alternatively, both hornworts and ferns could have invented neochrome independently. But Li’s analysis showed that both of these scenarios are extremely unlikely.

The fern and hornwort versions of neochrome are clearly related and shared a common ancestor. By comparing these modern versions and working backwards, Li calculated that they diverged from each other around 179 million years ago. By contrast, the ferns and hornworts themselves split off at least 400 million years ago.

This pattern, which doesn’t apply to any other fern gene, strongly suggests that the ferns acquired the gene for neochrome directly from hornworts. After that, the gene seems to have repeatedly hopped between different fern lineages.

These “horizontal gene transfers” are everyday events for bacteria, which seem to trade DNA with the same ease that we exchange emails. They’re much rarer in plants and animals, and many reported examples have been met with scepticism. But Li’s study has certainly won over Jeffrey Palmer from Indiana University. “I’ve read their paper closely, and I think their evidence is very strong and convincing,” he says.

Palmer is more measured about the idea that gaining neochrome allowed ferns to diversify in the shadow of flowers—it’s a plausible idea, but not one that’s been proven yet. If it was, “it would be the most significant horizontal transfer yet discovered in plants,” he says. Most transferred genes, including some that Palmer has discovered, don’t do very much. But neochrome “had the potential to really shake up fern physiology in a big way, and be really adaptive.”

Li is now looking for other horizontally transferred fern genes, to see if neochrome is exceptional or just part of a general pattern. And he’s also studying neochrome in hornworts to see what the gene