Suggested area (pink) across which Late Devensian erratics from northern Scotland might have been distributed. The map also shows the "footprint" of the Anglian glaciation, during which ORS erratics might have been carried even further southwards.
Here is the press release from Curtin University in Australia:
https://www.curtin.edu.au/news/media-release/study-details-epic-transportation-of-stonehenge-stone-across-ancient-britain/
ABSTRACT:
The Altar Stone, the 6000 kg central sandstone megalith at Stonehenge in southern England, is suggested to have originated from the Orcadian Basin in northeast Scotland, some 700 km away. However, its source location within this large basin remains unresolved and its mode of transport uncertain. Proposed mechanisms include Neolithic human transport via marine or overland routes or southward movement as a glacial erratic. Here, we combine sandstone provenance analysis with ice flow modelling to constrain potential sources and transport pathways. Ice sheet reconstructions show that southward transport vectors from northeast Scotland were highly localised, making precise source determination critical to evaluating glacial erratic transport plausibility. Candidate source areas farther south within the Orcadian Basin are more compatible with southward glacial dispersal, but show weaker correspondence with the detrital zircon age spectra that characterise the Altar Stone. By contrast, sandstones from Caithness on the mainland of northeast Scotland provide the closest match in zircon age structure; yet, modelled glacial transport from this region is predominantly towards the north‐east, with a localised south‐eastward pathway directed towards Dogger Bank. Glacial dispersal to Dogger Bank would reduce the anthropogenic transport distance required to Stonehenge from 700 km to 400 km. However, such a model raises an additional temporal problem, as Dogger Bank was inundated by post‐glacial sea‐level rise before the Altar Stone likely arrived at Stonehenge. Glacial transport may have provided an intermediate stage in the stone's journey, but alone cannot account for the final emplacement on Salisbury Plain. Even under a glacially assisted scenario, substantial anthropogenic transport would have remained necessary.
The article:
Clarke, A. J. I., Veness, R. L. J., Kirkland, C. L., Clark, C. D., Gandy, N., Emery, A. et al. (2026) From Highlands to Henge: Refining the Provenance and Transport Pathways of Stonehenge's Altar Stone. Journal of Quaternary Science, 1–8. https://doi.org/10.1002/jqs.70080
https://onlinelibrary.wiley.com/doi/epdf/10.1002/jqs.70080
Clarke, A. J. I., Veness, R. L. J., Kirkland, C. L., Clark, C. D., Gandy, N., Emery, A. et al. (2026) From Highlands to Henge: Refining the Provenance and Transport Pathways of Stonehenge's Altar Stone. Journal of Quaternary Science, 1–8. https://doi.org/10.1002/jqs.70080
https://onlinelibrary.wiley.com/doi/epdf/10.1002/jqs.70080
ABSTRACT:
The Altar Stone, the 6000 kg central sandstone megalith at Stonehenge in southern England, is suggested to have originated from the Orcadian Basin in northeast Scotland, some 700 km away. However, its source location within this large basin remains unresolved and its mode of transport uncertain. Proposed mechanisms include Neolithic human transport via marine or overland routes or southward movement as a glacial erratic. Here, we combine sandstone provenance analysis with ice flow modelling to constrain potential sources and transport pathways. Ice sheet reconstructions show that southward transport vectors from northeast Scotland were highly localised, making precise source determination critical to evaluating glacial erratic transport plausibility. Candidate source areas farther south within the Orcadian Basin are more compatible with southward glacial dispersal, but show weaker correspondence with the detrital zircon age spectra that characterise the Altar Stone. By contrast, sandstones from Caithness on the mainland of northeast Scotland provide the closest match in zircon age structure; yet, modelled glacial transport from this region is predominantly towards the north‐east, with a localised south‐eastward pathway directed towards Dogger Bank. Glacial dispersal to Dogger Bank would reduce the anthropogenic transport distance required to Stonehenge from 700 km to 400 km. However, such a model raises an additional temporal problem, as Dogger Bank was inundated by post‐glacial sea‐level rise before the Altar Stone likely arrived at Stonehenge. Glacial transport may have provided an intermediate stage in the stone's journey, but alone cannot account for the final emplacement on Salisbury Plain. Even under a glacially assisted scenario, substantial anthropogenic transport would have remained necessary.
============
ALTAR STONE
1 National Museum of Wales ( assumed to have been excavated Hawley in the 1920s)
2 Salisbury Museum (assumed to have been collected from the underside of the Altar Stone in 1844)
ORCADIAN BASIN SAMPLES
Two Old Red Sandstone rock samples from the Orcadian Basin, bought from Natural Wonders Ltd, Whitby:
1 Cruaday, Orkney
2 near Spittal, Caithness
Right from the outset there is bias that should have been picked up by the referees. In their introductory paragraph the authors refer to "three lithologies" in the "building block" collection at Stonehenge, specifically creating a seperate category for the Altar Stone. Ixer and Bevins, for reasons that are not difficult to discern, decided not so long ago that the Altar Stone should no longer be referred to as a "bluestone" but should be dignified by the creation of a new rock category. There is no earth science logic to that, since the bluestones (of all shapes and sizes) found in the Stonehenge landscape do not all come from Mynydd Preseli, and the Altar Stone is not the only sandstone. If we look at the sarsens and the erratic bluestones around Stonehenge, there are scores of lithologies represented, and ORS sandstone is just one of them.
This labelling device, right at the beginning of the article simply flags up the authors' specific intention to demonstrate that the Altar Stone was not just special or sacred, but that it was even more sacred than the bluestones.
In mentioning that Parker Pearson's Waun Mawn "lost bluestone circle" has been criticised, they cite Darvill (2022) who refuted the hypothesis on archaeological grounds, but they fail to cite my much more comprehensive 2024 criticism and dismissal (in the "Holocene" journal) of the MPP narrative on earth science grounds. One might have expected an acknowledgement of this in an article published in JQS..............
In referring to monolith transport, the authors refer to the glacial theory, but say "recent detrital zircon–apatite fingerprinting of Salisbury Plain river sediment found no evidence that glacial processes delivered the bluestones to southern England (Clarke and Kirkland, 2026)". It would have been more honest for the lead authors to state that their own zircon study is deeply defective, since the findings are based on just four river sand samples (accepted as valid without any refence to sampling procedures) and a lack of any control studies. This and other defects were discussed in another post on this blog:
We have similar problems with the other key samples used by Clarke and others in their studies of the Altar Stone. All 4 of them -- from Stonehenge, Orkney and Caithness -- are surrogate samples, not collected by the authors during fieldwork but taken from second hand sources:
ALTAR STONE
1 National Museum of Wales ( assumed to have been excavated Hawley in the 1920s)
2 Salisbury Museum (assumed to have been collected from the underside of the Altar Stone in 1844)
ORCADIAN BASIN SAMPLES
Two Old Red Sandstone rock samples from the Orcadian Basin, bought from Natural Wonders Ltd, Whitby:
1 Cruaday, Orkney
2 near Spittal, Caithness
The Altar Stone samples are simply assumed to be truly representative of the Altar Stone itself -- but if that attribution is faulty, then this article, like several others, becomes nonsensical. Given this fundamental research defect, why on earth did not one or another of the authors actually pop up north for a few days with a geological hammer and collect some proper, fully authenticated, rock samples?
Anyway, on a more positive note I welcome and agree with the paragraph on erratic transport routes and with the point that glacial erratics take thousands of years to be transported within the ice, during which ice flow vectors can shift in direction.
Quote:
This motivated us to model trajectories of glacial erratics, accounting for such shifting flow vectors over time to yield predictions of transport routes and distance. A complicating factor is that the Orcadian Basin sat near the ice sheet's north–south ice drainage divide (>1 km thick at times). Small movements (tens of kilometres) of this ice divide, as the ice sheet fluctuated, could have transported boulders in a multitude of directions. An investigation of the possible time-integrated flow routes is thus warranted to explore the plausibility of southward transport as an erratic. The key here is that rather than considering a single ice transport vector, we now examine if the Altar Stone could have first been transported in one direction and later in another as ice flow directions shifted. Could such multistage ice transport have carried the Altar Stone closer to Salisbury Plain? Given the complexity of ice flow directions in the region, refining likely locations of the source area is crucial to exploring this time-integrated ice flow analysis of possible erratic pathways.This is fine, and could also, of course, be applied to the west coast of Scotland in our attempts to discover how erratics apparently from the Western Isles might have been transported to the coasts of the Bristol Channel. Indeed, I have already cited another article by Veness and others in a recent post about Scottish erratic trails.
In the discussion of "detrital zircon data sets" from the Orcadian basin and elsewhere, the authors present no new fieldwork and no new samples. They depend entirely on the published work of other authors, and their conclusions are therefore based on pre-selected data sets, with a vast expanse of ORS sandstone territory unsampled and effectively ignored. This represents yet another powerful bias in the work.
On the matter or erratic trains, the authors suggest -- in line with BRITICE and other modelling work -- that erratics derived from Caithness and Orkney are unlikely to have found their way to the Dogger Bank during the Late Devensian Glaciation (LGM), but that they might have travelled south if they had previously been emplaced by eastward-flowing ice in (for example) the Wolstonian or Anglian Glaciation and then later entrained by a southward flowing ice stream. On the other hand ORS erratics from New Aberdour or the Inverness area could well have been entrained on the south side of an ice shed and transported southwards to the Dogger Bank. In other words, the modelling work suggests that the Inverness and New Aberdour areas are more likely to have provided Dogger bank erratics than outcrops in Caithness or Orkney. I have no problem with any of that, except to say that other outcrops of ORS around the Midland Valley Basin were even more likely to have provided erratics for the southward-flowing ice stream in the North Sea.
In the section of the article entitled "Human and ice transport?" the authors seem to lose touch with reality. First, they accept that the Stonehenge landscape "lacks the detrital signature expected from direct glacial delivery of megalith-derived material from northeast Scotland". They cite Clarke and Kirkland 2026, but (as mentioned above) they fail to appreciate that their conclusion is highly contentious, having been based on just four sand samples about which we know virtually nothing. Clarke and Kirkland are joint authors of the article which we are scrutinizing.......... enough said.
Then the authors embark on a series of speculations on "human agency" in the movement of the Altar Stone from Dogger Bank to Stonehenge which are so bizarre as to be embarrassing. How this section got through the refereeing process, maybe we will never know........
They suggest that Dogger Bank might have been a "topographic high point in the landscape" which might have held some cultural significance for Palaeolithic or Mesolithic inhabitants. They then suggest that the Altar Stone, found as an erratic, must have been special or sacred enough to justify collection and transport towards Stonehenge via maritime and land-based pathways three thousand years before it was incorporated into the Stonehenge monument. Talk about forward planning.........!!!
The idea of Dogger Island and its sacred stone -- seized upon, of course, by the media -- is so absurd that even our old friend MPP might be worried about it! There is no evidence at all that Palaeolithic or Mesolithic tribesmen in the period 9,000 - 7,000 BP had either the desire or the capability to move a six-tonne slab of rock from one place to another in Doggerland or anywhere else. As far as I am aware, there are no Palaeolithic of Mesolithic standing stones anywhere in Europe. At the time of Dogger Island's final inundation, and indeed for millenia before that, this was a densely wooded landscape of rivers, lagoons, bogs, fens and marshes -- good hunting, fishing and gathering territory. But I think we can safely dismiss the idea that the inhabitants knew that their island world was going to disappear, and that they planned and executed a move westwards for a large sacred stone while the going was good. Of course the media love this sort of narrative, but it's completely bonkers.
The "Dogger Bank scenario" involves "prolonged cultural significance or multiple-phase activity, across an exceptionally large temporal gap" -- but the authors also suggest that the boulder might have been moved south-westwards from the North Sea coast during the Anglian Glaciation and then dumped somewhere near the Chilterns ice edge. This would have involved a relatively short "anthropogenic transport route" along the Berkshire Ridgeway towards Stonehenge. In yet another scenario the authors suggest that the Altar Stone might have been ice-rafted through the Strait of Dover following one or another of the envisaged mega-floods, and then dumped somewhere on the south coast of England. That's a remote possibility, but before taking it too seriously we need to know much more about pre-Devensian glacio-isostatic adjustments and sea-level changes.
In summary, this is a very strange article. While I welcome the fact that some authors are at least taking the idea of long-distance glacial erratic transport seriously, the scenarios examined in the article all involve the human transport of a large bluestone monolith -- the Altar Stone. I'm still highly sceptical about that, and I think that the authors here have accepted far too easily that the Altar Stone did indeed come from the far NE of Scotland. I do not think that has been proved.
Clarke and Kirkland have rushed into print far too early. Their sampling and the use of surrogate samples suggests substantial bias. If I am to be convinced by their provenancing, I would like to see the lead authors of this piece get out into the field, collecting samples from many ORS locations in various parts of Scotland, specifically for the testing of their hypothesis. Once they have done that, and published their test results, I might be prepared to abandon my scepticism about the quality of their work.
See also Mike Pitts, who compounds the confusion:
