19 August 2026 geologyresearch-notes

Why the granite is a red herring

Everyone starts a Dartmoor gold search by looking at the granite. Three days in the BGS archive suggest that is exactly the wrong place to begin.

If you ask most people where to look for gold on Dartmoor, they point at the middle of the map. The granite. The tors, the high moor, the source of all those tin lodes. It is an entirely reasonable intuition: granite intrusions drive hydrothermal systems, hydrothermal systems concentrate metals, Dartmoor is one enormous granite intrusion, and the moor was mined for tin for something like eight hundred years.

The intuition is wrong, and the evidence against it is unusually clean.

The assays nobody quotes

In the 1920s Alfred Brammall assayed Dartmoor granite for gold. The results, as collected in the local literature, are these:

A troy ounce is 480 grains. The best rock on Dartmoor therefore assays at roughly one-fiftieth of an ounce per ton, around 0.55 grams per tonne. For comparison, the Cononish deposit in Scotland that supports an actual working mine runs at 15.9 g/t. Dartmoor granite is about thirty times poorer than ore, and that is the best sample.

Where the gold actually is

Meanwhile, the British Geological Survey spent the late 1980s and early 1990s panning streams in the South Hams, well to the south of the granite, and kept finding gold. The headline number from their synthesis report is that 44 of 450 drainage sites returned more than 0.5 ppm gold in panned concentrate. Not spectacular grades, but consistent, mappable, and over a large area.

What made it interesting was the shape of the grains. Many were dendritic: branching, fern-like, structurally fragile. As the report puts it, they are

very intricate in shape with projections which could not survive if transported in the streams for anything more than trivial distances from source.

That is a strong statement. It means the source is local to each anomaly, not washed down from the moor.

The palladium tells you where it came from

The clincher is chemistry. The South Hams gold is not pure. Many grains carry palladium, often in intricate growth zones, with rims running up to 13% Pd, and the survey turned up grains of potarite, a palladium-mercury mineral rare enough that finding it in a Devon soil sample is a genuine event.

Granite-related hydrothermal systems do not do this. What does is a much stranger process: low-temperature saline brines, around 86 °C, descending out of Permian red beds and depositing gold where they hit reduced, pyritic Devonian slate below the unconformity. This is now a recognised deposit style, and south Devon is one of its type areas, sitting alongside comparable occurrences at Tilkerode in Germany and Lubin in Poland.

The gold near Dartmoor, in other words, has nothing to do with Dartmoor. It has to do with a desert that covered Devon 280 million years ago and the salty water that percolated out of it.

Why this changes the search

Once you accept the model, the map redraws itself:

• Look at the Permian. Not the granite. That means the Crediton Trough to the north and the concealed Permian influence across the South Hams to the south.
• Look for the unconformity. The best ground is where structures in reactive Devonian or Carboniferous rocks pass beneath Permian cover; the BGS say so explicitly.
• Look at small tributaries. Fragile grains near their source means the little combes, not the main rivers.
• Stop expecting nuggets. This is fine, flaky, often dendritic gold. It will not sound in a detector and it will not look like the pictures.

There is a lovely irony in it. The tinners who found flakes of gold in their streamworks on the high moor were, on this reading, mostly finding traces that had nothing to do with the tin they were chasing. And the ground the modern evidence actually favours is farmland you need to knock on a door to reach.

Next: the byelaw that decides where you may actually do anything about it.