MYS0CRWP5JR3NED·Mystery · site

Stonehenge

Stonehenge is a circular earthwork and stone monument on the chalk of Salisbury Plain. The present chronology, built from a Bayesian model of its radiocarbon dates, runs in five stages: the ditch, bank and 56 Aubrey Holes about 3000–2920 BC; the sarsen circle of 30 uprights with lintels and the five trilithons, with bluestones set among them, about 2620–2480 BC; and rearrangements of the bluestones down to about 1600 BC⁠[1]. The Aubrey Holes held cremated remains of at least 27 men and women buried between about 3000 and 2500 BC, making the site the largest known cremation cemetery of its age in Britain⁠[2].

Three kinds of stone make up the monument. The sarsens, sandstone blocks of up to 30 tonnes, come almost all from West Woods near Marlborough, 25 km north⁠[3]. The bluestones, about 80 originally, of 2 to 3 tonnes each, are dolerites and rhyolites from outcrops in the Preseli Hills of Pembrokeshire, where quarrying pits, wedges and hammerstones dated to about 3400–3000 BC were excavated at Carn Goedog and Craig Rhos-y-felin⁠[4]⁠[5]⁠[6]. The Altar Stone, a six-tonne sandstone slab lying at the centre, was shown in 2024 to carry mineral grains whose ages match the Old Red Sandstone of the Orcadian Basin in north-east Scotland, at least 750 km away⁠[7].

The open questions follow from the stones. Whether people carried the bluestones from Wales or ice did is argued between archaeologists and a few geomorphologists⁠[6]⁠[8]. Whether a Welsh stone circle at Waun Mawn was dismantled and its stones re-erected at Stonehenge, as claimed in 2021, has been contested on the excavation evidence and on the geochemistry⁠[9]⁠[10]⁠[11]. And what the monument was for has readings from cemetery and ancestral shrine to solar temple, healing centre, calendar and political statement of a united Britain⁠[12]⁠[13]⁠[14]. Of the 7 questions on this site, 1 are answered. The most likely answer to “How did the bluestones get from the Preseli Hills to Salisbury Plain?”, not yet confirmed: people quarried the stones in Preseli and carried them to Wessex, overland or with a water stage, around 3000….

Key questionsset by the editors

How did the bluestones get from the Preseli Hills to Salisbury Plain?Most likely: people quarried the stones in Preseli and carried them to Wessex, overland or with a water stage, around 3000…

Was a stone circle at Waun Mawn dismantled and its stones taken to Stonehenge?Most likely: waun Mawn is a group of small local stone settings with no connection to Stonehenge

What was Stonehenge for?Most likely: the first monument was a cemetery and ancestral shrine for a select group

Where did the sarsens come from?Most likely: fifty of the 52 surviving sarsens share one chemistry, matched to West Woods near Marlborough; the other two…

Where exactly did the Altar Stone come from, and how did it travel?Most likely: the stone came from the mainland Orcadian Basin, in Caithness or around the Moray Firth

Who raised the stones?Answered: neolithic people of Britain built it in stages from about 3000 to 2500 BC, with stones they quarried and…

All 7 questions

The site

The monument sits in a landscape of earlier and contemporary works: the Cursus, long barrows, the Avenue to the River Avon, and the great henge and timber circles of Durrington Walls 3 km to the north-east, where the builders' settlement and feasting deposits have been excavated⁠[15]. The axis of the sarsen settings, at an azimuth of about 49° to the north-east and 229° to the south-west, points to midsummer sunrise and midwinter sunset, and the first straight stretch of the Avenue runs along it⁠[16]. Ruggles and Chadburn, after surveying the alignment afresh, judge it deliberate and accurate to within a degree or so⁠[13].

The Aubrey HolesRachelle Haun, 2008. CC BY 2.0Artifact

The stones stand on an enclosure 110 m across. The sarsens were dressed on site; a core drilled from Stone 58 in 1958 and returned in 2019 gave the sample whose chemistry fixed the source at West Woods⁠[3]. The bluestones were set first in the Aubrey Holes, on the excavators' reading, and later moved into the double arc and horseshoe of the later stages⁠[12]⁠[1].

Map of the landscapeOpen the full map

Stones Landscape Where the stones came from

+−

StonehengeStone circle and henge

Mapped

The Avenue to the River AvonEarthwork (banked avenue)

Mapped

Stonehenge CursusEarthwork (cursus)

Mapped

Durrington WallsHenge

Within 1 km

WoodhengeTimber circle and henge

Mapped

Bluestonehenge (West Amesbury henge)Stone circle and henge

Within 1 km

Bush BarrowBowl barrow

Mapped

Normanton Down barrowsRound barrow cemetery

Mapped

Cursus barrowsRound barrow cemetery

Mapped

Old King BarrowsRound barrow cemetery

Mapped

New King BarrowsRound barrow cemetery

Mapped

Winterbourne Stoke crossroads barrowsRound barrow cemetery

Mapped

Stonehenge Down barrowsRound barrow cemetery

Mapped

MappedWithin 1 kmBase map © OpenStreetMap contributors

Timeline

Timeline

Squeezed where little happened. Each mark is a published claim, by how it stands:AcceptedMost likelyStill possibleDisputedRuled out

What was found

The 2008 re-excavation of Aubrey Hole 7 recovered the cremated bone that Hawley had reburied in 1935: at least 27 individuals, men and women and some children, dated from about 3000 to 2500 BC⁠[2]. Strontium isotopes in 25 of them show that about 40 per cent had not lived on the chalk in their last years and had values consistent with west Wales, and the carbon in some bone indicates cremation on wood from outside the region⁠[17].

In Wales, excavation at Craig Rhos-y-felin found a recess from which a rhyolite pillar had been detached, a stone-built platform, hammerstones and wedges, with charcoal of the late fourth millennium BC⁠[5]; Carn Goedog gave the same tools and a date of about 3400–3000 BC for extraction of the spotted dolerite that matches most Stonehenge dolerites⁠[4]⁠[6]. At Waun Mawn, 5 km from the quarries, the same team found four standing or recumbent stones and a number of stone holes on an arc which they reconstructed as a 110 m circle, abandoned before completion, with one hole whose shape they matched to Stonehenge's Stone 62⁠[9]⁠[18].

The Altar Stone's fragments, from excavations of the twentieth century, were studied by automated mineralogy and by dating detrital zircon, apatite and rutile. The grain ages are Mesoproterozoic and Archaean with a mid-Ordovician overprint, a Laurentian signature found in Britain only in the Orcadian Basin⁠[19]⁠[20]⁠[7]. Comparison with Orkney sandstones then excluded the islands and pointed to the mainland basin around Caithness and the Moray Firth⁠[21].

The Altar Stone (Stone 80)The Stones of Stonehenge (stonesofstonehenge.org.uk). CC BY 4.0Artifact

A rhyolite cobble excavated by Newall in 1924 was re-examined in 2025: it shows no glacial striations or shaping and is read as a fragment of a worked Craig Rhos-y-felin monolith⁠[22].

What is known

The sequence and dates are secure to within a few decades at each stage⁠[1]. The sources of all three stone types are fixed by geochemistry⁠[3]⁠[4]⁠[7]. People and animals came to the monument's landscape from across Britain: pigs eaten at Durrington Walls were raised in Scotland, the north of England and Wales as well as Wessex⁠[15], and some of the people buried in the Aubrey Holes had lived in western Britain⁠[17].

Quarrying of bluestone in Preseli is dated to about 3400–3000 BC, up to several centuries before the first bluestone settings at Stonehenge, which the excavators read as a gap during which the stones stood somewhere in Wales⁠[6]. No glacial deposits are known on Salisbury Plain, and the Newall boulder shows no glacial features⁠[22].

The Altar Stone's route from Scotland is not known. The authors of the 2024 study favour sea transport along the coast, given the distance and terrain⁠[7].

The explanations

On transport, the archaeological reading is that people moved the bluestones from Preseli, overland or by a combination of land and water, around 3000 BC; the quarry features, the dates and the absence of glacial evidence on Salisbury Plain support it⁠[6]⁠[22]. The glacial reading holds that Irish Sea ice carried the stones to the Plain and that Neolithic people gathered erratics; it rests on the general extent of glaciation and on the Newall boulder, which the 2025 re-examination does not support⁠[8]⁠[22].

On Waun Mawn, the excavators' position is that a large circle was begun and abandoned and some stones went to Stonehenge⁠[18]. Darvill reads the site as several small settings and the Aubrey Holes as post holes⁠[10]. Geochemistry shows the Waun Mawn dolerites come from Cerrigmarchogion, not Carn Goedog, so the site cannot have supplied Stonehenge's dolerites⁠[11], and a geomorphological critique finds no demonstrated circle at all⁠[23]. The claim stands as possible but not supported.

On purpose, the readings are not exclusive. Parker Pearson's team read the first monument as a cemetery and ancestral shrine for a select group, and the sarsen rebuilding as a monument whose stones from Wales, Wessex and Scotland made it a statement of unity among communities across Britain, at a time when new people were arriving from the Continent⁠[12]⁠[14]. The solstitial alignment is accepted as deliberate; the fuller observatory of Hawkins, with eclipse prediction, is not⁠[24]⁠[13]. Darvill's 2022 calendar of 365.25 days encoded in the sarsen numbers was answered by Magli and Belmonte as numerology⁠[25]⁠[26].

Published claims

Published claims and their status

Year	Who	Claimed	Status	Reason
1136	Geoffrey of Monmouth⁠[27]	Merlin brought the Giants' Dance, stones that giants had carried from Africa to Ireland, to Salisbury Plain as a memorial for Aurelius Ambrosius.	Ruled out	The stones are four millennia older than the story and came from Wales, Wessex and Scotland, not Ireland⁠[1]⁠[3]⁠[7]
1963	Gerald Hawkins⁠[24]	Stonehenge's alignments make it a solar and lunar observatory that could predict eclipses.	Disputed	Reviews of the alignments against the excavated plans accept the solstitial axis but find no case for an observatory⁠[16]⁠[13]
1971	G. A. Kellaway⁠[29]	Ice carried the Welsh stones to or near Salisbury Plain, and the builders gathered erratics rather than hauling stones from Wales.	Disputed	Quarry workings at the Preseli outcrops, the re-examined Newall boulder and river sediments from the Plain all weigh against it, though its proponents still argue for it⁠[6]⁠[22]⁠[28]
2009	Mike Parker Pearson and colleagues⁠[12]	Stonehenge began as a cemetery for a select group, with cremations dated from about 3000 BC.	Accepted	The re-excavated Aubrey Hole 7 yielded at least 27 cremated individuals, the largest cremation cemetery of its time in Britain⁠[2]
2015	Mike Parker Pearson, Richard Bevins, Rob Ixer and colleagues⁠[5]	The rhyolite outcrop at Craig Rhos-y-felin was quarried for a Stonehenge pillar in the late fourth millennium BC.	Accepted	The recess, platform, tools and dates were matched to Stonehenge debris, and a second quarry at Carn Goedog showed the same working⁠[6]
2019	Mike Parker Pearson and colleagues⁠[6]	The bluestones were quarried by wedging along natural joints at Carn Goedog and Craig Rhos-y-felin about 3400 to 3000 BC and moved overland to Salisbury Plain.	Most likely	The quarrying is well evidenced and human transport is the field's view, but the route and method are not known⁠[22]
2020	David Nash and colleagues⁠[3]	Fifty of the 52 surviving sarsens share one chemistry, matched to West Woods, 25 km north.	Accepted	A reanalysis argued that several sources cannot be ruled out, and the authors replied; the exchange is unresolved in print but the field holds the match⁠[30]⁠[31]
2021	Mike Parker Pearson and colleagues⁠[9]	Waun Mawn was a 110 m stone circle, dismantled about 3000 BC, and some of its stones were re-erected at Stonehenge.	Disputed	The authors later conceded the circle was never finished, and the dolerites left at Waun Mawn come from a different outcrop from Stonehenge's⁠[18]⁠[11]
2022	Timothy Darvill⁠[10]	Waun Mawn is a series of small local stone settings, not a great circle, and the Aubrey Holes held posts, not bluestones.	Most likely	The geochemistry and a geomorphological critique also find no link between Waun Mawn and Stonehenge⁠[11]⁠[23]
2022	Timothy Darvill⁠[25]	The sarsen settings encode a solar calendar of 365.25 days.	Disputed	Two archaeoastronomers rejected it as numerology with astronomical errors and no supporting analogy⁠[26]
2022	Richard Bevins, Nick Pearce, Mike Parker Pearson and Rob Ixer⁠[11]	The dolerites at Waun Mawn come from Cerrigmarchogion, not Carn Goedog, so Waun Mawn cannot have supplied Stonehenge's dolerite bluestones.	Accepted	No later analysis disputes the geochemistry, and the circle's excavators are among the authors⁠[11]
2023	Giulio Magli and Juan Antonio Belmonte⁠[26]	Darvill's sarsen calendar is numerology, with astronomical errors and no analogy to support it.	Accepted	No archaeoastronomer has taken up the calendar since⁠[13]
2024	Anthony Clarke, Christopher Kirkland and colleagues⁠[7]	The Altar Stone came from the Old Red Sandstone of the Orcadian Basin in north-east Scotland, at least 750 km away.	Accepted	A follow-up study ruled out Orkney, narrowing the source to the mainland basin; the outcrop itself has not been found⁠[21]
2024	Brian John⁠[8]	The Newall boulder, excavated at Stonehenge in 1924, is a glacial erratic, so ice brought the bluestones to Salisbury Plain.	Disputed	A re-examination found no glacial surface features and read it as a piece of a worked Craig Rhos-y-felin monolith⁠[22]
2024	Ronald Hancock and colleagues⁠[30]	The geochemistry of Stone 58 does not show a single source, and three to six sources cannot be ruled out.	Disputed	The West Woods authors replied that the method suits silcrete poorly; the response added no new analyses⁠[31]⁠[32]
2024	Clive Ruggles and Amanda Chadburn⁠[13]	The solstitial axis is deliberate and precise, the lunar claims are weak, and Hawkins's observatory does not survive scrutiny.	Accepted	It agrees with the earlier phase-by-phase review of the alignments⁠[16]
2025	Richard Bevins, Nick Pearce, Rob Ixer and colleagues⁠[22]	The Newall boulder has no glacial surface features and is a fragment of a worked Craig Rhos-y-felin monolith.	Accepted	River sands draining Salisbury Plain also lack the Preseli zircon ages that glacial delivery would bring⁠[28]
2025	Mike Parker Pearson and colleagues⁠[14]	With stones from Scotland, Wales and Wessex, the rebuilding of about 2500 BC was a political act uniting communities across Britain.	Most likely	The far-flung sources are established, but what the builders meant by gathering them is interpretation⁠[7]⁠[3]
2026	Anthony Clarke and Christopher Kirkland⁠[28]	Mineral ages in river sands draining Salisbury Plain show local sediment only, so the Plain was not glaciated.	Most likely	It is a single recent study, but it agrees with the absence of glacial deposits on the Plain⁠[22]
2026	Anthony Clarke and colleagues⁠[33]	Ice could have carried the Altar Stone part of the way toward Dogger Bank, but not to Salisbury Plain, so people hauled it most or all of the way.	Still possible	The model allows a partial ice stage, but Dogger Bank was under the sea before the stone arrived⁠[33]

What would settle it

The transport question would be settled by finding the bluestones' route: Neolithic deposits of bluestone debris or a trackway between Preseli and Wessex, or, for the glacial reading, any glacial till or erratics on Salisbury Plain with Preseli lithologies⁠[6]⁠[8]. Waun Mawn would be settled by full excavation of its circumference with control trenches outside it, and by matching Stonehenge bluestones other than the dolerites to its remaining stones⁠[10]⁠[11].

The Altar Stone would be settled by a geochemical match to a specific outcrop in Caithness or Moray, and the route by finds of worked Orcadian sandstone along the coasts⁠[7]⁠[21]. Purpose will not be settled by one find; isotopes and DNA on more of the cremated dead, and the dating of the Welsh and Scottish connections, would say who came and when⁠[17].