BAR HARBOR − Famous for their sluggishness¹（迟缓，萧条） , turtles have been slow to give up the secrets of their evolution and place on the evolutionary² tree. For decades, paleontologists（古生物学者） who study fossils and molecular³ biologists who study genetics have disagreed about whether turtles are more closely related to birds and crocodiles or to lizards⁶. Now, two scientists from the Mount Desert Island Biological Laboratory in Bar Harbor, Maine, and their colleagues from Dartmouth College and Harvard and Yale Universities have developed a new technique for classifying animals, and the secret is out. Turtles are closer kin⁷ to lizards than crocodiles. To reach their conclusion, published in Nature News and Biology Letters, the research team looked at a newly discovered class of molecules⁸ called microRNA. Most of the genetic⁴ material or DNA⁹ that scientists study provides the code for building proteins, large molecules that form an essential part of every organism. But microRNAs are much smaller molecules that can switch genes¹⁰ on and off and regulate protein production. They are also remarkably¹¹ similar within related animal groups and provide important clues for identification.

"Different microRNAs develop fairly rapidly in different animal species over time, but once developed, they then remain virtually unchanged," said Kevin Peterson, a paleobiologist at MDIBL and Dartmouth College. "They provide a kind of molecular map that allows us to trace a species' evolution."

Peterson worked with Ben King, a bioinformatician at MDIBL. "My role in the study was to enhance our software so we could find new and unique microRNAs in the lizard⁵ genome," King said. "We identified 77 new microRNA families, and four of these turned out to also be expressed in the painted turtle. So we had the evidence we needed to say that turtles are a sister group to lizards and not crocodiles."

Though few creatures have been as puzzling as the turtle, the research team plans to use its microRNA analysis on other animals to help determine their origins and relationships as well. It is also developing a web-based platform to share the software with other researchers around the world.