SOME of the stones used to create Stonehenge in Wiltshire may well have been first gone up in Wales, a new study has suggested.

Archaeologists have discovered evidence of quarrying for the world heritage sites' bluestones in Pembrokeshire - 500 years before they were erected in Wiltshire.

They believe the most likely explanation is that the stones were first used in Wales before being transported 140 miles to Salisbury Plain where they formed part of the monument's inner horseshoe.

Geologists have known since the 1920s that the bluestones were brought to Stonehenge from somewhere in the Preseli Hills but only now has there been collaboration with archaeologists to locate and excavate the actual quarries from which they came.

Stonehenge was built during the Neolithic period, between 4,000 to 5,000 years ago.

Both of the quarries in Preseli were exploited in the Neolithic, and Craig Rhos-y-felin was also quarried in the Bronze Age, around 4,000 years ago.

Project director Professor Mike Parker Pearson, from UCL Institute of Archaeology, said: "We have dates of around 3,400BC for Craig Rhos-y-felin and 3,200BC for Carn Goedog, which is intriguing because the bluestones didn't get put up at Stonehenge until around 2,900BC.

"It could have taken those Neolithic stone-draggers nearly 500 years to get them to Stonehenge, but that's pretty improbable in my view.

"It's more likely that the stones were first used in a local monument, somewhere near the quarries, that was then dismantled and dragged off to Wiltshire."

The discovery has been made by a team of scientists from UCL, the universities of Manchester, Bournemouth and Southampton, Amgueddfa Cymru - National Museum Wales, and Dyfed Archaeological Trust.

Professor Kate Welham, of Bournemouth University, believes the ruins of any dismantled monument are likely to lie somewhere between the two megalith quarries.

"We've been conducting geophysical surveys, trial excavations and aerial photographic analysis throughout the area and we think we have the most likely spot," she said.

"The results are very promising - we may find something big in 2016."

The Stonehenge bluestones are of volcanic and igneous rocks, the most common of which are called dolerite and rhyolite.

The scientists identified the outcrop of Carn Goedog as the main source of Stonehenge's 'spotted dolerite' bluestones and the outcrop of Craig Rhos-y-felin as a source for one of the "rhyolite" bluestones.

The special formation of the rock, which forms natural pillars at these outcrops, allowed the prehistoric quarry-workers to detach each standing stone with a minimum of effort.

"They only had to insert wooden wedges into the cracks between the pillars and then let the Welsh rain do the rest by swelling the wood to ease each pillar off the rock face" said Dr Josh Pollard, of the University of Southampton.

"The quarry-workers then lowered the thin pillars onto platforms of earth and stone, a sort of 'loading bay' from where the huge stones could be dragged away along trackways leading out of each quarry."

Radiocarbon-dating of burnt hazelnuts and charcoal from the quarry-workers' camp fires reveals that there were several occurrences of megalith-quarrying at these outcrops.

The megalith quarries are on the north side of the Preseli hills, and this location undermines previous theories about how the bluestones were transported from Wales to Stonehenge.

Previous writers have often suggested that bluestones were taken southwards from the hills to Milford Haven and then floated on boats or rafts, but this now seems unlikely.

"The only logical direction for the bluestones to go was to the north then either by sea around St David's Head or eastwards overland through the valleys along the route that is now the A40," said Prof Parker Pearson.

"Personally I think that the overland route is more likely. Each of the 80 monoliths weighed less than two tonnes, so teams of people or oxen could have managed this.

"We know from examples in India and elsewhere in Asia that single stones this size can even be carried on wooden lattices by groups of 60 - they didn't even have to drag them if they didn't want to."

The new discoveries may also help to understand why Stonehenge was built.

Prof Parker Pearson and his team believe that the bluestones were erected at Stonehenge around 2,900BC, long before the giant sarsens were put up around 2,500BC.

"Stonehenge was a Welsh monument from its very beginning. If we can find the original monument in Wales from which it was built, we will finally be able to solve the mystery of why Stonehenge was built and why some of its stones were brought so far," he said.

Further excavations are planned for 2016.

The findings, Craig Rhos-y-felin: A Welsh Bluestone Megalith Quarry For Stonehenge is published in the journal Antiquity.