Back in August I said I’d make the case for a Clyde-region Altar Stone somewhere with more room in it. This is the room.
The honest bit first, in one breath. I’m an artist with a geology A level from fifty-five years ago. Nothing below is a finding. It’s a preferred solution, which is a polite name for a hunch with references. The people who have actually put the Altar Stone under a microscope are Rob Ixer, Richard Bevins and their colleagues, and if you want the rock argued properly, start there. Brian John has the ice. Simon Banton has the best photographs of the stones. I’ve got renders and an opinion.
What everyone agrees on
The sand in the Altar Stone came from rocks north of the Iapetus Suture. Laurentia, the old northern continent. That’s the 2024 result and it’s a good one. Zircons don’t lie about their age.
What the papers went on to say, and the newspapers went on to shout, is Orcadian Basin, far north-east, and then Caithness specifically. That’s where I get off the bus. Orkney itself has since been ruled out by Bevins and colleagues, who went there and sampled it. Wrong feldspar, no baryte, no tosudite. Which leaves a comparison set of a handful of samples, two of them bought from a shop, standing in for a basin the size of Wales. “Matches the samples we’ve got” is not the same sentence as “came from here”. The sampling problem page goes through it slowly.
Mine: Firth of Clyde
My money is on somewhere in the wider Firth of Clyde. Not Ailsa Craig. Ailsa Craig is a lump of microgranite that ends up as curling stones. But there is Old Red Sandstone all round that firth: Arran, Bute, the Cumbraes, the Ayrshire shore from Wemyss Bay down to Largs, the bottom end of Kintyre. All Devonian, all fed from Laurentian ground, and most of it coarse, red, river-laid and nothing like the Altar Stone on a typical bed.
That’s the cost of the idea, and I’ll pay it up front. The Altar Stone is an odd rock. Pale grey-green, fine, micaceous, cemented with baryte and calcite, almost no potassium feldspar. A quiet-water bed. A pond in river country. Rift basins do that: coarse at the faulted edges, fine towards the middle, the odd still interval in a fluvial pile. That’s an inference about how basins behave, not a bed anyone has mapped, and I’m labelling it as such.
Still. The Altar Stone doesn’t match the typical Orcadian facies either. Everybody is hunting an unusual bed. The outcrops we can walk on are scraps of the Devonian landscape. Old Red Sandstone carries on under the Firth of Clyde and the North Channel, and two and a half million years of ice have had ample time to scrub a small basin off the map altogether. The bed I want may be under the sea, under the drift, or gone.
Baryte, since it always comes up. Barium in the cement is bed-level plumbing, not a regional badge. Bevins’ team found it largely absent from the Orcadian samples they tested. And the west had barium to spare: working baryte mines at Muirshiel above Lochwinnoch and at Gasswater near Cumnock.
Peter’s: Kerrera
Peter Kokelaar is a retired volcanologist who spent years dating granites in the Grampian Highlands and now picks erratics off Horton beach on Gower. His Stonehenge page is worth an hour of anyone’s time. His candidate sits further up the coast than mine: the Old Red Sandstone of Kerrera and the Oban district, in the Firth of Lorn, flaggy brown sandstones and red siltstones stacked in with the Lorne lavas.
His argument runs through the zircons. The inherited crystal cores that Clarke and colleagues used to point at the Orcadian Basin come from the deep crust of the Grampian Terrane, and that same crust was being worn down into sediment on the west side too. Same fingerprint, different postcode. The Grampian igneous rocks are barium-rich, which he reads as a fit for the baryte. And on Horton beach he has Lewisian gneiss, Harris anorthosite, Dalradian schist and Eigg pitchstone, every one of them from the far north-west, every one of them lying on a Welsh beach. That is his point and mine. The ice that flowed past the Scottish west coast delivered to the Bristol Channel. We have the receipts.
Lorn or Clyde. Two shaded areas on one map, both draining into the same ice, both untested. I’d take either over a shop-bought Caithness.
The proven leg
The transport is the part nobody has to take on trust. Ailsa Craig microgranite is the textbook indicator erratic and it turns up on Gower and along the Bristol Channel. Delivery to the Somerset coast belongs to the Anglian, around 478,000 years ago, or to something older that the Anglian tidied away. Buried channels full of till at Court Hill and Kenn near Clevedon, found when they cut the M5 through the ridge. Ice, in north Somerset, no argument. The evidence file scores every find, friendly or not.
Then the contentious gap. Fifty miles from Kenn to Stonehenge with nothing secure in it. Green counted fifty thousand pebbles from the river terraces round the Plain and found not one glacial rock. That’s the best punch the other side has and it lands. I’m not going to pretend otherwise. What I will say is that a weathered lump of grey-green sandstone is a very anonymous thing on a beach. Ailsa Craig granite gets noticed. This wouldn’t. Nobody has looked for a dull grey-green pebble – who would?
What would sink it
Offshore cores under the Clyde and the North Channel, checked for sandstone of the right character. Desk work, cheap. Irish Sea erratic collections searched for pale green, micaceous, baryte-cemented blocks. A hit in either place changes everything. A proper search that finds nothing weakens me. Hafnium isotopes pointing firmly and only north-east would weaken me more. A hunch that can’t say what would kill it isn’t worth keeping.
The long version, with the caveats in the right order, is on the preferred solution page.
Two lost sandstones and one long line of ice. I’ll settle for that.
Be happy for ever.