Scientists have for the first time discovered complex organic molecules that have survived fossilization, in the form of 350-million-year-old sea creatures found in Ohio, Indiana and Iowa.
The spindly animals with feathery arms known as crinoids appear to have been buried alive in storms during the Carboniferous Period, when North America was covered with vast inland seas.
Buried quickly and isolated from the water above by layers of fine-grained sediment, their porous skeletons gradually filled with minerals - but some of the pores containing organic molecules were sealed intact.
"There are lots of fragmented biological molecules - we call them biomarkers - scattered in the rock everywhere. They're the remains of ancient plant and animal life, all broken up and mixed together," says William Ausich of Ohio State University.
"But this is the oldest example where anyone has found biomarkers inside a particular complete fossil. We can say with confidence that these organic molecules came from the individual animals whose remains we tested."
The molecules appear to be aromatic compounds called quinones, which are found in modern crinoids and other animals, and which sometimes function as pigments or as toxins to discourage predators.
The team isolated the molecules by grinding up small bits of fossil and dissolving them into a solution which was put through a gas chromatograph mass spectrometer. A magnet separated individual molecules based on electric charge and mass, and computer software identified thems as similar to quinones.
The reason the crinoids were so well preserved is partly to do with the structure of their skeletons. Their long bodies are made up of thousands of stacked calcite rings, each a single large calcite crystal that contains pores filled with living tissue. When a crinoid dies, the tissue will start to decay, but calcite will precipitate into the pores, sealing organic matter within the rock. The location also helped: in the flat American Midwest, the rocks weren't pushed up into mountain chains or heated by volcanism.
The next challenge is to identify the exact type of quinone molecules found, and determine how much information about individual species can be gleaned from them.
"These molecules are not DNA, and they'll never be as good as DNA as a means to define evolutionary relationships, but they could still be useful," says Ausich.
"We suspect that there's some kind of biological signal there - we just need to figure out how specific it is before we can use it as a means to track different species."