I live in Cilgwyn, not far from the Bluestone Brewery which is much patronised by the archaeologists every September, and on my land there is a very beautiful river valley. The river cuts through two dolerite dykes which have a major influence on both valley shape and gradient. Elsewhere we find Abermawr shales and mudstones, and meta-mudstones adjacent to the dykes where there has been a moderate degree of metamorphism or alteration due to baking. Where the hard rock outcrops, the valley sides are very steep indeed, and the river tumbles down through a series of cataracts.Above and below the dykes, the valley is much more open, with grassy meadows on the floodplain, more gentle sides and a river which is much more placid in appearance. All good standard geomorphology -- and exactly what one one would expect.
The point of this pleasant little tale (apart from the fact that I am rather chuffed to be the proud owner of two dykes) is that the geology map is highly generalised, and misses out on multiple details of rock distributions and outcrops. I know of other dykes in the neighbourhood which are not shown at all on the geological maps, as well as other outcrops of volcanic ashes and rhyolites. And there is great variation within outcrops which are shown as being relatively uniform on the maps. I have flagged up mapping inaccuracies on Waun Mawn, on Brynberian Moor and in the Brynberian Valley as well. I am not criticising the BGS surveyors who have produced fantastic maps over the years -- just pointing out that if the maps are not to appear too complex, there has to be an "appearance" of uniformity if the map is to be of any practical use to farmers, planners and the general public.
Against this background, I have always been critical of the claim from Ixer and Bevins that they have provenanced some of the foliated rhyolite fragments found at Stonehenge to "within a few square metres" on the rock face at Rhosyfelin. For a start, I have pointed out that none of the thin sections featured in the publications provides an exact match. And secondly, we have never seen an adequate rebuttal of the point made by me and many others -- namely that if a particular foliated rhyolite petrography is seen at point 8, it must also be observable at multiple other locations where that particular foliated layer outcrops at the ground surface. Even if rock samples have been taken from 30 sampling points within the Brynberian Valley, we simply do not know enough about the real and complex local geological map to justify what is really rather a wild and exaggerated claim about provenancing accuracy.