By Paul Rincon
BBC News Online science staff
A tiny fossil found by a road cutting in Pennsylvania, US, could be the earliest example of an arm bone.
The earliest limbed creatures may have navigated shallow rivers
The 360-370-million-year-old humerus, or upper arm bone, indicates that limbs may have evolved for use in water and not to get around on land.
It suggests the earliest limbed animals were fish navigating shallow rivers, but its place in the evolutionary tree is the subject of some controversy.
US biologists have published details of the discovery in the journal Science.
The evolutionary process that transformed fins into limbs is poorly understood, despite being a key transition in evolution.
It led to the emergence of tetrapods, four-legged vertebrates that mostly live on land. Before tetrapods, vertebrates were confined to water.
"It's like a Rosetta stone, it helps us translate between these two different forms; the fish fin and the tetrapod limb," Dr Ted Daeschler, of the Academy of Natural Sciences, Philadelphia, US, told BBC News Online.
Limb by limb
"This grade of animal was part of the evolutionary history of all limbed animals including ourselves," he explained.
In the first half of the 20th Century, researchers postulated that the evolution of limbs was driven by the need to support and move the body on land.
The humerus evolved from being a mobile bone, as it is in the fish, to one that was stationary and acted as prop to support the body. But this had nothing to do with walking on land, the authors of this paper contend.
It is unclear what the rest of the animal was like
"It clearly shows that the lower limb was very muscular and directed towards the bottom, and the elbow permanently bent," Dr Daeschler said.
The authors propose that the animal, still known only by the number ANSP 21350, was an ambush predator that used its limbs to hold itself steady in the current of a shallow stream until prey came along.
Alternatively, the limb may have been used to move the animal through the shallow water.
The specimen will be an important addition to our understanding of the processes that allowed tetrapods to use their limbs for moving about on land.
"It can tell us how some of the structures that became useful on land first developed. Evolution seems to work by finding new functions for old structures," Dr Daeschler commented.
"Those features became useful when the animals had to hold themselves against gravity on land; they were already in the toolkit."
The authors have constructed a scheme to show relationships between some key fossils in the transition between lobe-finned fish and tetrapods.
But they admit they have left some important specimens such as Ichthyostega and Elginerpeton out of the scheme.
The fossil came out of a road cutting
Jenny Clack, senior assistant curator at the University Museum of Zoology in Cambridge, UK, said there were still many unresolved questions with regards to what kind of creature ANSP 21350 was and what exactly it used its arms for.
"The humerus has got some very peculiar features all of its own and those help us with sequences for acquisition of certain characters. But how it will fit in the long run I'm not sure," said Dr Clack.
Professor Robert Carroll, of the Redpath Museum, Montreal, Canada, told BBC News Online: "This individual bone is clearly that of an animal that is a very primitive tetrapod or a primitive lobe-finned fish or perhaps something in-between.
"The problem is that it isn't like any of the later humeri that you encounter in the later Carboniferous."