The Altar Stone - Maybe Not The Orcadian Basin?

 

Refining the provenance of the Stonehenge Altar Stone: multi-criteria screening of the Orcadian Basin, the exclusion of its highest-priority target on measured thermal data, and first-pass results from the Midland Valley

Tim Daw July 2026

Abstract

A UK-wide stream-sediment geochemical screen identified an East Caithness coastal cluster (cluster 15) as the strongest onshore barium and Ba/Rb anomaly over continuous Old Red Sandstone. Application of the primary thermal-maturity dataset (Hillier & Marshall 1992) shows that the cluster core lies entirely within a region of measured vitrinite reflectance >3 % R₀ and spore colours 10/11 — maximum palaeotemperatures in excess of ~250 °C on the authors’ calibration. This is incompatible with the Altar Stone’s preserved clay assemblage — mixed-layer illite/smectite (26–33 % of the clay fraction), tosudite, dioctahedral chlorite and kaolinite (Bevins et al. 2024, Table 4) — which records low-grade, sub-greenschist conditions. The mismatch is irreversible on current published data.

Orkney requires systematic re-screening. Nairn/Elgin has a favourable low-maturity, thinned-margin structural setting but only a weak, basement-associated geochemical signal (Ba/Rb ~15–17 over <25 % genuine ORS); it is a structural and thermal possibility rather than a validated geochemical candidate, and is untested for detrital zircon. Detrital-mineral data (Clarke et al. 2024, 2026) give strong terrane-scale support for a Scottish source but do not, at present resolution, exclude the Midland Valley, for which comparative multi-proxy data have not yet been published.

After the original geochemical screening was refined with the independent thermal, clay and barium-type filters it anticipated, and the highest-priority Orcadian target was excluded on measured evidence, the same transparent multi-criteria framework has been applied to the Midland Valley. First-pass results identify strong Ba/Rb anomalies over near-continuous Lower ORS in the Strathmore belt (cluster 148: Ba/Rb 37, 94 % genuine ORS) that pass the identical multi-element consistency filter, together with suitable fluvial sandy facies and documented aluminous clay parageneses (tosudite, kaolinite). The principal open constraint is thermal: the strong-anomaly Lower ORS is itself relatively mature (~1.2 % R₀), so the decisive screening criterion is whether least-mature ORS (≤0.7–1.0 % R₀) can be identified within the same belt from published maturity data. These are the leading candidates the expanded screen identifies; they can be judged against clay-mineralogical and maturity data as and when those are available.

Click to enlarge

1. The original geochemical screening was sound

The UK-wide composite stream-sediment screen correctly identified a high-amplitude Ba/Rb anomaly over near-continuous genuine Old Red Sandstone bedrock in East Caithness cluster 15 (~170 cells, ~42.5 km² along the Lybster–Clyth–Ulbster–Sarclet coast). The anomaly passed all multi-element consistency filters and showed no mafic or base-metal overprint. It converged with independent detrital-mineral evidence for a source within the crystalline basement that supplied the Orcadian Basin ORS. The method performed exactly as designed. (Cluster numbers in this document follow the UK-wide screen; East Caithness cluster 15 was numbered cluster 18 in the earlier Orcadian-basin-only screen.)

2. The Thermal-Maturity Objection to an East Caithness source

The screening paper anticipated a thermal-maturity objection at second hand (pending verification against the primary map). That verification has now been made against Hillier & Marshall (1992). The objection is stronger than the anticipation and is stated plainly here.

Hillier & Marshall (1992) mapped organic maturity across the Orcadian Basin from 248 spore-colour determinations and 182 vitrinite-reflectance measurements. Their high-maturity region of southern and central Caithness — vitrinite reflectance above 3 % R₀ and spore colours 10/11 (completely black) — extends along the coast from near Niandt in the south to just north of Wick (their Figs 2 and 3). Cluster 18 lies entirely within that envelope (cells span 58.30–58.36 °N).

Their Fig. 4 assigns measured reflectance directly to the subgroups that underlie the cluster: Lybster 3.1–3.7 % R₀, Hillhead 2.4–3.7 % R₀, Clyth 4.3–4.7 % R₀. On the Barker & Pawlewicz (1986) calibration used by the authors, reflectance above 3 % records maximum palaeotemperatures in excess of ~250 °C — the threshold of true metamorphism.

Precision is high at the scale of the data: direct measurements at named localities and type sections, with 10–20 % reproducibility at multi-sample localities and a spatial density of a handful of coastal sites across the ~7 km cluster coast. No plausible reading of measurement error converts 3–5 % R₀ into the sub-1 % R₀ values the Altar Stone requires.

The Altar Stone records much gentler heating, and its clay mineralogy is diagnostic. In the <2 µm fraction (Bevins et al. 2024, Table 4) it carries mixed-layer illite/smectite (26–33 %), tosudite (15–21 %), kaolinite (16–25 %), dioctahedral chlorite (12–13 %) and illite (14–19 %) — an aluminous, incompletely illitised, low-grade assemblage. Bevins et al.’s own Caithness Flagstone Group samples, from the same >3 % R₀ terrain, show the opposite: essentially pure illite with trioctahedral chlorite and corrensite or smectite, and no mixed-layer I/S, no tosudite and no kaolinite — the fully illitised, trioctahedral clay mineralogy that temperatures above ~250 °C produce. The Altar Stone thus retains substantial expandable interstratified clay and delicate aluminous phases (tosudite, kaolinite) that cannot survive the temperatures measured at the cluster coast. Independently, its bulk mineralogy (Bevins et al. 2024, Table 3) shows K-feldspar not detected and baryte at 0.8–1.1 wt % — the exact combination (no Rb-bearing K-feldspar, discrete diagenetic baryte) that drives the anomalously high rock-level Ba/Rb the screen was built to detect. The assemblage reads as low-maturity and aluminous — Moray-Firth-like rather than Caithness-flagstone-like.

The mismatch is irreversible on present published evidence. The cluster core is therefore excluded as a direct source of the Altar Stone.

3. Status of remaining Orcadian targets

With the East Caithness cluster core excluded on measured thermal and clay-paragenetic data, three further areas inside the Orcadian Basin require assessment: Orkney, the Nairn/Elgin margin, and the fault-bounded Latheron–Buldoo coastal window.

Orkney Most of Orkney lies at ~1–1.5 % R₀ (Hillier & Marshall 1992) and is therefore thermally favourable for preservation of expandable illite/smectite and delicate phases such as tosudite. However, published stream-sediment barium data (BGS regional geochemical atlas, Map 4) show elevated values over parts of the Stromness, Rousay and Eday successions, with several higher anomalies — particularly in south-west Mainland — spatially associated with the East Scapa Fault system and known mineralised areas. These are more likely to represent hydrothermal or vein baryte (the wrong genetic type for the Altar Stone’s early disseminated cement + tosudite association) than low-temperature diagenetic baryte in a mature sandstone.

Existing published mineralogical sampling on Orkney (standing stones plus limited outcrops) is geographically and facies-restricted, so negative results from that material bear only on the sampled subset. Notably, Bevins et al.’s own Eday Group sample (5514) carries 49 % mixed-layer I/S and 41 % kaolinite (Table 4) — the closest Orcadian clay analogue to the Altar Stone in the published data, though it lacks tosudite and dioctahedral chlorite. At the screening level Orkney is therefore not excluded but under-characterised: the published stream-sediment barium and Hillier & Marshall (1992) maturity maps have not been jointly resolved to facies level, and on present data no locality-level statement about Orkney is warranted either way. It is carried as an open candidate.

Nairn/Elgin corridor  The Moray Firth margin is thermally attractive: Hillier & Marshall (1992) and Hillier & Clayton (1989) place it among the lowest-maturity ground in the basin, with preserved expandable illite/smectite — a genuine point in its favour. Its geochemistry, however, is weak. The stream-sediment clusters along this margin are dominated by Moine Supergroup and Grampian Group crystalline basement, with only 0–23 % genuine ORS in the anomaly cells and modest Ba/Rb ratios (~15–17; e.g. cluster 44 at 22.7 % genuine ORS, Ba/Rb 16.6) — below both East Caithness (cluster 15: 98 % ORS, Ba/Rb 18.2) and the Midland Valley anomalies below. Because these anomalies sit largely on basement rather than on ORS, the barium is as likely to be basement- or pegmatite-derived as diagenetic. Nairn/Elgin therefore does not pass the ‘elevated Ba/Rb over near-continuous genuine ORS’ test that distinguished cluster 15.

The sedimentary environment transitions into fluvial and aeolian sandy facies. These lower-pH, leached conditions are the setting associated with the authigenic kaolinite and early diagenetic baryte cement seen in the Altar Stone. The broader Elgin/Nairn region also contains well-known fish beds in lacustrine facies, but the relevant units for comparison are the sandy fluvial and aeolian facies, which are typically less fossiliferous and represent a different diagenetic environment; the fish-bearing horizons elsewhere in the corridor are a screening consideration rather than a disqualifier of the candidate itself.

On present evidence Nairn/Elgin meets the structural and thermal criteria but not the geochemical one, and is untested for detrital zircon and clay paragenesis. It is best treated as a low-maturity fallback that would re-enter as a geochemical candidate only if data were to show a genuine ORS-hosted (rather than basement-hosted) barium signal. It does not rank ahead of the Midland Valley anomalies.

Latheron–Buldoo coastal window  One Caithness possibility survives, rated low. Hillier & Marshall (1992) map a small fault-bounded window of lower maturity along the coast between Latheron and Niandt, in the Robbery Head and Latheron subgroups, at ~1.5 % R₀ and bounded by the Latheron fault — distinctly cooler than the >3 % R₀ cluster coast a few kilometres to the north-east. This window lies ~2 km south-west of, and outside, cluster 15, so it is not the geochemical anomaly itself; and at ~1.5 % R₀ (~180–190 °C on Barker & Pawlewicz 1986) it is probably still too hot to preserve the Altar Stone’s expandable I/S and tosudite. Two things nonetheless keep it in play at low probability. First, with the Middle-ORS assumption dropped, its Lower ORS stratigraphy is no longer disqualifying. Second, the maturity here is blocky and fault-controlled, so a further-downthrown, cooler sliver within the window cannot be excluded on present mapping. The Buldoo standing stone (OS ND 2000 3369; 3.87 m, the tallest in Caithness, a slab of local sandstone bedded into nearby outcrop) sits in this window and shows that large monoliths were locally available — suggestive of availability, though not evidence of source. Against a clay-XRD expandability measurement on a coastal Robbery Head/Latheron sample the window would be decisively confirmed or excluded; on the maturity data the expected result is marginal. It is carried as a low-probability open candidate.

4. The Clarke et al. (2024, 2026) constraint and its documented limits

Detrital-zircon, rutile and apatite data provide strong statistical support for an ultimate source terrane within the crystalline basement that supplied the Orcadian Basin ORS, and significantly reduce the likelihood of non-Scottish origins. The provenance-level conclusion (a Scottish, Laurentian-margin source) is robust and is not disputed here. At locality resolution, however, comparative multi-proxy data from Midland Valley ORS have not yet been published, and the concordant zircon population is modest (56 concordant grains from three thin sections). The distinction between an Orcadian and a Midland Valley source therefore rests more on the multi-proxy chemistry than on zircon ages alone, and is not resolved at the resolution required once the thermal and clay filters are applied. The Midland Valley is not excluded.

5. Decision to expand the search

The original geochemical screening was sound. The independent thermal, clay and barium-type filters it anticipated have now been applied. The highest-priority Orcadian target (East Caithness cluster 15 core) is excluded on measured primary data from Hillier & Marshall (1992). No other locality inside the Orcadian Basin currently satisfies the full set of criteria. Consistent with the original method, the same multi-criteria framework can be applied to the next major Scottish ORS basin, the Midland Valley, and the remainder of this document reports that first-pass screening.

6. Application of the original Ba/Rb + multi-element screen to the Midland Valley

The same country-scale Ba/Rb anomaly map and multi-element consistency filter used in the original screening paper has been applied to the Midland Valley by zooming into the relevant portion of the BGS G-BASE and regional geochemical atlas datasets.

Several zones over Old Red Sandstone show elevated Ba and high Ba/Rb ratios. When the full multi-element consistency filter is applied (quiet signature across control elements, no strong coincident mafic or base-metal overprint), the following genuine-ORS candidates emerge (cluster IDs from the UK screen):

• Strathmore / northern Midland Valley Lower ORS belt — the strongest and most continuous anomalies. Cluster 148 (Arbuthnott–Garvock Group, ~27 km², Brechin–Montrose area) returns Ba/Rb 37.0 over 94.5 % genuine ORS — a higher ratio than East Caithness cluster 15 (18.2) over comparably continuous ORS. Cluster 176 (Stratheden Group, Upper ORS, ~13 km²) returns Ba/Rb 32.3 over 81 % ORS. Smaller 100 %-ORS Strathmore Group clusters (156, 160) sit at Ba/Rb ~16–18. Multi-element signatures are relatively quiet.
• Selected Upper ORS sandstone catchments in the central and eastern Midland Valley that pass the quiet multi-element filter. Cornstone (calcrete)-bearing Upper ORS represents oxidising, alkaline, commonly red-bed conditions — the opposite of the leached, reduced, grey-green setting of the Altar Stone — so these units are retained only where reduced, non-red sandy facies can be identified, and at lower priority than the Strathmore belt.
• Weaker or partially noisy anomalies over ORS that are retained at lower priority.

Anomalies spatially associated with known hydrothermal vein systems or major fault corridors are deprioritised at this stage, exactly as analogous signals were treated in the Orcadian work.

7. Further screens applied to the surviving Midland Valley candidates

Thermal maturity — Midland Valley ORS maturity is highly variable (Marshall et al. 1994), and this is the constraint that must be applied honestly, because the same Hillier & Clayton (1989) standard used above to exclude East Caithness applies here: illite/smectite expandability collapses to 5–10 % expandable layers by ~1.0 % R₀. The deeply buried Lower ORS of the Strathmore syncline — which hosts the strongest anomalies — reaches ~1.2 % R₀ or higher (≈165 °C on Barker & Pawlewicz 1986), and is therefore already too mature to preserve the Altar Stone’s expandable mixed-layer illite/smectite, tosudite and kaolinite. It follows that the thermal target is not the strong-anomaly Lower ORS itself but the least-mature ORS in the basin (sub-1 %, ideally ≤0.7 % R₀), which tends to be the shallower or marginal sequences that escaped deep burial. This reproduces, within the Midland Valley, the anti-correlation already documented in the Orcadian Basin between strong barium anomalies (mature, deeply buried or mineralised ground) and the low-maturity diagenetic setting the Altar Stone requires. Locating Midland Valley ORS at ≤0.7–1.0 % R₀ is therefore a prerequisite, not an assumption — the first test any candidate must pass.

Facies suitability — the passing Strathmore candidates overlie fluvial (braided-river) and locally aeolian sandstones. These leached, lower-pH environments favour authigenic kaolinite and early diagenetic baryte cement, matching the Altar Stone’s paragenesis. Reduced, grey-green sandy units are the target; oxidised red cornstone facies are not. Clay-mineralogical studies of the Midland Valley Lower ORS independently record aluminous interstratified phases including tosudite and illite/smectite (Hillier, Wilson & Merriman 2006) — the closest documented facies analogue for the Altar Stone’s tosudite–kaolinite association, and a genuine advantage over the Caithness flagstone interior. Tosudite is not itself a low-temperature indicator, however, so the requirement remains tosudite together with preserved expandable mixed-layer I/S and kaolinite.

Barium type — Some anomalies in central eastern Scotland are interpreted in the broader geochemical literature as derived from barite within the Old Red Sandstone bedrock itself — consistent with a diagenetic/authigenic origin in sandstone. Anomalies directly linked to known hydrothermal veins remain lower priority pending petrographic confirmation of early disseminated cement texture.

Resulting priority shortlist After the identical Ba/Rb + multi-element filter used in the original screening, followed by the same thermal, facies and barium-type filters applied to the Orcadian targets, the highest-priority areas are:

1. Strathmore Lower ORS belt (clusters 148 and 176, plus the adjacent 100 %-ORS Strathmore Group clusters 156 and 160), conditional on locating sub-1 % R₀ ground within it.
2. Reduced (non-red) sandy Upper ORS catchments away from known vein mineralisation, retained at lower priority pending facies confirmation.

These are the areas the screen flags for comparison against clay-mineralogical, maturity and petrographic data. The relevant discriminators, should such data be available or published, are illite/smectite expandability and tosudite (clay XRD) and early baryte cement texture and paragenesis (petrography).

8. Candidate comparison table

Candidate Area	Ba/Rb + Multi-element Screen	Thermal Maturity	Facies Suitability	Barium Type	Zircon Status	Overall Priority	Key Notes
East Caithness cluster 15 core	Pass	Fail (>3 % R₀, illite-only)	Lacustrine flagstones	Likely hydrothermal/vein	Match	Excluded	Direct measurements (Hillier & Marshall 1992); irreversible mismatch with Altar Stone paragenesis
Orkney (systematic)	Partial	Generally favourable	Heterogeneous	Mixed (fault-linked deprioritised)	Partial	Requires re-screen	Needs BGS Ba + maturity overlay; target low-maturity sandy facies away from major faults
Nairn/Elgin corridor	Weak (basement-associated)	Favourable (low maturity)	Fluvial/aeolian sands	Basement/pegmatite? (unproven)	Untested	Low (structural fallback)	Clusters on Moine/Grampian basement, 0–23 % genuine ORS, Ba/Rb ~15–17; structural/thermal only
Strathmore Lower ORS belt (cl. 148, 176)	Pass (cl.148 Ba/Rb 37, 94 % ORS)	Marginal (~1.2 % R₀; target sub-1 %)	Fluvial/aeolian sands; tosudite recorded	Bedrock-linked in places (unconfirmed)	Untested	High (new priority; thermal test pending)	Strongest new anomaly; must locate ≤0.7–1.0 % R₀ ORS within belt
Central/eastern MV Upper ORS	Partial	Variable	Fluvial; red cornstone facies excluded	Mixed	Untested	Medium	Retain only reduced non-red sandy facies away from veins
Latheron–Buldoo window (Orcadian)	Outside cluster (SW edge)	~1.5 % R₀ — borderline too hot	Robbery Head/Latheron ORS; local slabs (Buldoo)	n/a (not the anomaly)	Untested	Low (retained; clay test)	Fault-bounded low-maturity window SW of cluster; Middle-tier objection now moot; single XRD decisive

9. Conclusions

The original geochemical screening was sound. Application of the independent thermal, clay and barium-type filters it anticipated has excluded the highest-priority Orcadian target (East Caithness cluster 15 core) on measured primary data from Hillier & Marshall (1992). No other locality inside the Orcadian Basin currently satisfies the full criteria.

The search has therefore been expanded using the identical multi-criteria framework. First-pass results identify strong, genuine-ORS Ba/Rb anomalies in the Strathmore belt (clusters 148, 176) that pass the same multi-element filter and overlie facies-suitable, aluminous-clay-bearing Lower ORS. The principal open constraint is thermal maturity: the strong-anomaly Lower ORS is itself relatively mature (~1.2 % R₀), so the decisive screening criterion is whether least-mature (≤0.7–1.0 % R₀) ORS can be identified within the belt from published maturity data. Subject to that test, Strathmore is the strongest new target.

This is a screening exercise. It identifies where, on the criteria set out above, a source is and is not consistent with the published data; it does not propose fieldwork, sampling or laboratory work, and it claims no match. The candidates it lists can be judged against clay-mineralogical, maturity and petrographic data as and when such data are available or published, and the candidate list can be refined by higher-resolution overlay of thermal and mineralisation data where those exist.

Acknowledgements

S. Hillier is thanked for providing his paper. British Geological Survey stream-sediment and regional geochemical atlas data are acknowledged. All interpretations remain those of the author.

References (selected)

Barker, C.E. & Pawlewicz, M.J. 1986. The correlation of vitrinite reflectance with maximum temperature in humic organic matter. Lecture Notes in Earth Sciences 5, 79–93.

Bevins, R.E. et al. 2024. Was the Stonehenge Altar Stone from Orkney? Journal of Archaeological Science: Reports 58, 104738.

British Geological Survey. Regional geochemical atlases (Southern Scotland; UK stream sediment). G-BASE data.

Clarke, A.J.I. et al. (2024). A Scottish provenance for the Altar Stone of Stonehenge. Nature.

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

Clarke, A.J.I. and Kirkland, C.L. (2026). Detrital zircon–apatite fingerprinting challenges glacial transport of Stonehenge’s megaliths. Communications Earth & Environment, 7, Article 54. https://doi.org/10.1038/s43247-025-03105-3

Hillier, S. & Clayton, T. 1989. Illite/smectite diagenesis in Devonian lacustrine mudrocks from northern Scotland… Clay Minerals 24, 181–196.

Hillier, S. & Marshall, J.E.A. 1992. Organic maturation, thermal history and hydrocarbon generation in the Orcadian Basin, Scotland. Journal of the Geological Society, London 149, 491–502.

Hillier, S., Wilson, M.J. & Merriman, R.J. 2006. Clay mineralogy of the Old Red Sandstone and Devonian sedimentary rocks of Wales, Scotland and England. Clay Minerals 41, 433–471.

Marshall, J.E.A., Haughton, P.D.W. & Hillier, S. 1994. Vitrinite reflectivity and the structure and burial history of the Old Red Sandstone of the Midland Valley of Scotland. Journal of the Geological Society, London 151, 425–438.