Some of the ideas discussed in this blog are published in my new book called "The Stonehenge Bluestones" -- available by post and through good bookshops everywhere. Bad bookshops might not have it....
To order, click

Saturday 30 March 2019

Have the bluestone quarry hunters called it a day?

The 2019 edition of the National Park "tourist newspaper" called COAST TO COAST has now been published, with all the events planned for the coming holiday season -- and for the first time since 2011 there is no mention of a bluestone quarrying talk at Castell Henllys (or anywhere else) presented by Prof MPP towards the end of September.    Last year he gave two talks on successive evenings, at the same venue -- but really had very little to say.

So this probably means that he and the rest of the jolly quarry hunters have at last given up on their hunt for Proto-Stonehenge and for even more bluestone "quarries"--  and have accepted that they have been flogging a dead horse for the last few years.  Two seasons of fruitless searching at Waun Mawn might have tipped the balance.

Or maybe it's the funding organizations who have come to the dead horse conclusion, have finally got fed up of MPP and the others refusing to accept that their findings are disputed, and have at last pulled the funding.........   Not before time, one might say, given that we have complained endlessly about scientific malpractice at Rhosyfelin, Carngoedog and Waun Mawn.  We have also noted with interest that the archaeological establishment which should be enforcing good practice appears to have turned a blind eye to all the misdemeanours of the dozen or so archaeologists and geologists involved.

So -- a September in Pembs free of archaeologists?   What a pleasant thought.......

I suppose there is a chance that funding applications are still being processed, and that digging plans  are not yet finalised.  We shall see.  I'll report on any developments.

Bluestone 33

Thanks to Neil Wiseman for this 2017 pic of him standing next to bluestone 33 (one of the spotted dolerite pillars) (I think we can call this one a pillar!) in the  bluestone circle at Stonehenge.  This is a rather interesting stone -- one of those claimed (somewhat foolishly) by Mike Parker Pearson as having come from his famous Carngoedog "quarry".  It has a rather battered surface, and is one of around ten bluestones that have been shaped or tooled in some way.

The six or seven bluestone which have the most obvious traces of shaping are in the bluestone oval or horseshoe.  The "worked" stones most often cited in the literature are lintels 36 and 150, stones 67 and 70 (uprights with tenons) and the tongue and groove "pair" (stones 66 and 68).

Tuesday 26 March 2019

Quarried bluestone pillars? You must be joking...

This is a slide from one of the talks which I do about the "bluestone quarries".  These are supposed, by Parker Pearson and his associates,  to be "pillars" and to have come from the Carngoedog "Neolithic bluestone monolith quarry".  If they have ever been anywhere near a quarry, I will eat my hat.......

Sunday 17 March 2019

The Carn Goedog bluestone provenancing is unsupported by the evidence

The geology of the north flank of Mynydd Preseli.  Source:  BGS mapping.

Some quotes from Bevins, Ixer and Pearce, 2014

The principal conclusion presented here is that at least 55% of the Stonehenge monoliths and fragments analysed to date can be sourced to Carn Goedog. The PCA plots support the association of the Group 1 Stonehenge dolerites with Carn Goedog but also suggest that Group 3 dolerites might come from Carn Goedog.....

Group 3 samples might be from an as yet un-sampled part of the Carn Goedog outcrop, bearing in mind that Jones et al. (2005), on the basis of PXRF investigations, identified that a number of the eastern Preseli outcrops were geochemically heterogeneous. Further sampling of the Carn Goedog intrusion would serve to clarify if this is the case or not.

Bevins, Ixer and Pearce — Carn Goedog paper, 2014.
Journal of Archaeological Science 42 (2014) 179e193

As we know, these rather circumspect conclusions have been solidified, over the course of five years, on the basis of no additional evidence, info firm statements from Parker Pearson et al (2019) that "At least five bluestone pillars (stones 33, 37, 49, 65, 67) were taken from Carn Goedog, and probably many more."  Elsewhere they claim that a further 4 bluestone megaliths at Stonehenge have also probably come from Carngoedog -- making 9 in all:

Recent geochemical analysis has revealed two main groups of Stonehenge spotted dolerite, the larger of which (Stones 33, 37, 49, 65 & 67) can be matched most closely with Carn Goedog (Bevins et al. 2014). The second group (Stones 34, 42, 43 & 61) has not been provenanced to a specific Preseli outcrop but may derive from Carn Goedog or from nearby outcrops such as Carn Breseb or Carn Gyfrwy. 

I have already scrutinized this claim, and have pointed out that all 9 of these so-called quarried pillars are not freshly quarried pillars at all, but battered erratic boulders and slabs.

Dear Reader,

Do these look to you like freshly quarried pillars?  If they do, you might need to get your eyes checked........

 Some of the boulders and slabs claimed by Parker Pearson to have been quarried at Carngoedog.  Pics -- courtesy Simon Banton


To summarise:
The existence of the Carngoedog bluestone quarry is completely dependent upon the accurate provenancing of fragments and samples of spotted dolerite from Stonehenge being traced back unequivocally to the Carn Goedog tor.  As we know, Bevins, Ixer and Pearce (2014) claim to have done that, on the basis of just three samples from the tor, three from Carn Breseb and two from Carn Sian -- although we have no information at all as to where on the tors the sampling points were.  I have raised serious concerns about this provenancing, since none of the Stonehenge samples match precisely with the samples from Carngoedog.  

Now I have followed up some references, and have discovered the paper summarised below.  It's a paper about a dolerite sill from Tal y Fan in Snowdonia, around 110m thick and emplaced in rather similar circumstances to the sills of the Fishguard Volcanic Series in NE Preseli.  What the paper shows is that within sills of this scale, there is great differentiation in the chemical composition and petrological characteristics of a sill from its edges in towards the centre.  There are also differences attributable to post-emplacement metamorphism or alteration.  

Quote from Bevins, Ixer and Pearce 2014:
For example, in a study of the altered, 110 m thick dolerite sill of Tal y Fan (an intrusive sheet of igneous rock) Merriman et al. (1986) were able to model flow differentiation processes during magma emplacement involving the segregation of olivine towards the centre of the intrusion on the basis of concentrations of MgO and Ni, elements which are preferentially partitioned into olivine.

One of the interesting things about this 1986 paper is that Richard Bevins is one of the authors!  So a sample taken from near the edge of the sill is not typical of the centre of the sill, nor of other parts affected by metamorphism.  We have no idea whether the three samples from Carn Goedog came from the centre of the sill, or near one of the edges.  

There must have been similar differentiation in all of the other extensive sills on the north flank of Preseli -- the Carn Bica sill, the Carn Sian sill,  the Carn Ddy Fach sill, the Craig Talfynydd sill and all the others.  So how can the authors of the 2014 paper know that the characteristics of the fragments taken from Stonehenge do not have better matches in unsampled parts of one or more of the other sills in the area?  Their sampling density in the field is nowhere near being adequate to rule out that possibility.

In other words, there is no way that they can state with any degree of confidence that they have actually identified Carn Goedog as a source for certain Stonehenge bluestones.  They should go back to Preseli, take maybe 50 more samples across the full width of the sills already mapped, and then tell us what the analytical results are.  Until they have done that, we can dismiss their "provenancing work" as useless.

If that sounds all very technical, apologies.  It's actually a matter of very simple logic.  Elementary, my dear Doctor Watson.


Petrological and Geochemical Variations within the Tal y Fan Intrusion: a Study of Element Mobility During Low-Grade Metamorphism with Implications for Petrotectonic Modelling
Journal of Petrology, Volume 27, Issue 6, 1 December 1986, Pages 1409–1436,
Published: 01 December 1986


The Tal y Fan Intrusion is an altered olivine dolerite sheet emplaced into a coeval sequence of subaqueous volcanic rocks of Caradoc (Ordovician) age in NE Snowdonia, Wales. Primary mineral and chemical variations across the 110 m thick sheet suggest that the magma was drawn from a zoned magma chamber, although the intrusion consolidated predominantly as a single cooling unit. An horizon of ferrodolerite resulted from in situ fractionation. Secondary mineral assemblages are indicative of the prehnite-pumpellyite and prehnite-actinolite fades, suggesting metamorphic alteration conditions of approximately 310°C and 1-85 kb. Major elemental variation largely reflects primary mineral variations across the intrusion, although Ca, Al, and Na show limited mobility in the outermost 4-5 m, related to breakdown of plagioclase feldspar during metamorphism. The LIL elements Rb, Sr, K, and Ba were highly mobile, particularly in the marginal zones, whereas Th, in addition to the incompatible elements Zr, Y, Ti, P, Nb, Ta, Hf, and the REE, was immobile even in the marginal zones. Accordingly petrotectonic modelling based on discriminant diagrams using these immobile elements is considered most reliable. The Tal y Fan Intrusion has characteristics transitional between N-type and E-type MORB, similar to tholeiitic within plate basalts. In contrast with other Ordovician volcanic sequences of the Welsh Basin, no subduction component is identified in the Tal y Fan magma, the LIL element enrichment observed being related to alteration.

Merriman, R.J., Bevins, R.E., Ball, T.K., 1986. Geochemical variations within the Tal y Fan intrusion: implications for element mobility during low-grade meta- morphism. J. Petrol. 27, 1409-1436.



From a previous post, in 2015.
A gentle reminder of why the "spot provenancing" of foliated rhyolites to within a few sq metres at Rhosyfelin is also nonsense:

This is an extract from our new paper, to be published on Monday:
Parker Pearson has pointed out to many visitors the “exact location” from which an orthostat was taken from the rock face and hauled or carried off to Stonehenge. That assertion appears to be based on the statement from Ixer and Bevins (2011, 2014) that they had provenanced certain rhyolite flakes at Stonehenge to “within a few square metres” of their sampling point 8, near the tip of the Rhosyfelin spur. However, they have not adequately demonstrated that level of precision, either through published thin sections from Stonehenge and Rhosyfelin samples, or through analysis of a very dense pattern of sampling points. The Stonehenge rhyolite flakes could even have come from a section of the spur which has been removed by the processes of glacial entrainment. Also, the “shelf” from which the monolith is supposed to have been removed is heavily abraded by either meltwater or ice action, indicating that it cannot have been quarried during the Neolithic.

The thin section showing the petrography (the Jovian fabric) of the foliated rhyolite at Locality 8 as discussed by Bevins and Ixer in their papers. Thanks to them for the photo.

The geologists have never demonstrated in print that there is anything in the Stonehenge rhyolitic debitage that is identical to the rock exposed at Richard Bevins's original sampling point 8........  If they have thin sections that prove a match, here is an offer to publish them on this blog......

In any case, as far as I can see, the petrography of the foliated rhyolite debitage at Stonehenge is quite variable, when seen in thin sections, as is the petrography of the samples taken from the Pont Saeson - Rhosyfelin area. A confounding factor is that the rock face at Rhosyfelin coincides with a series of closely-spaced fracture planes, extending for about 50 south-westwards from the tip of the spur. Bits and pieces of the same fracture plane are exposed along the whole face, which means that widely separated samples will have an identical, or almost identical, petrography. I am sure that samples will have been taken, and that this will have been confirmed by the geologists. I would like to see that in print. So whatever the nature of the rock at point 8 (location 11) may be, the same rock will occur in many other locations along the rock face and in other locations along the valley side where rock exposures can no longer be seen.

In summary, the level of accuracy in the geological "spot provenancing" for this locality is another fantasy, cited repeatedly and deliberately by the archaeologists as part of their campaign to "prove" that this is a Neolithic quarry. No matter what they may claim, this is NOT scientific evidence.

The proto-Stonehenge checklist

Waun Mawn on the flank of Mynydd Preseli -- the site of the latest fruitless search for Proto-Stonehenge.   Maybe the intrepid diggers will find the Holy Grail instead?

Below is a list of the sites that the archaeologists have been looking at since 2011, when geologists Ixer and Bevins published their first paper referring to Craig Rhosyfelin as "the dominant source of the Stonehenge rhyolitic debitage".  (The title of that paper was a nonsense, because the authors were depending on just the KNOWN debitage, given that only around half of the area of the stone settings has ever been excavated.  Let that pass........ but it was a sign of things to come, with the tendency for "over-interpretation" becoming increasingly obvious with every year that passed from 2011 until today.)  But even before that date, the two of them had been showing increasing interest in the origins of the bluestones and the debitage at Stonehenge, with a number of papers on the dolerites, rhyolites and sandstones.
Some of the earlier references are listed here:

.... and it's interesting that not one of them is in a geological journal.  All of them are in archaeological journals and popular magazines -- some of them peer-reviewed, and others involving publication at the invitation -- or the discretion -- of the editor.   One wonders how many of these articles would have been accepted for publication in geological journals, given the extra level of scrutiny which would have been involved.

Without going back through twenty or more articles, I can't be sure of this -- but my impression is that originally the two geologists were rather circumspect about how the bluestone monoliths at Stonehenge were picked up and transported.  But as soon as the two tribes of archaeologists (the Darvill - Wainwright tribe and the Parker Pearson tribe) became involved in the research  the glacial entrainment / transport hypothesis was dumped for no particular reason -- and from that point on the research flipped from being good science to being bad science -- with the focus placed firmly upon the confirmation of two ruling hypotheses.  The first was that the stones were quarried from sacred of significant sites, and the second was that the stones were transported from West Wales to Stonehenge by human agency.

So the working assumption has been, from 2011 onwards, that there are at least two Neolithic bluestone quarries in the Preseli area, and that the research simply has to concentrate hard on confirming that.  We have referred to this before as "assumptive research" -- we can perhaps forgive that within a "storytelling" discipline like archaeology, but it is unforgivable in a truly scientific discipline like geology.  The reputations of Ixer and Bevins as serious research scientists will eventually  be decided not by the likes of me and the followers of this blog, but by academic geologists.

I have said on many occasions -- in print and on this blog -- that the "spot provenancing" of Stonehenge rhyolite fragments to Rhosyfelin and Stonehenge spotted dolerite fragments to Carngoedog is deeply flawed, for a number of reasons.   But Ixer and Bevins, like the Parker Pearson team of archaeologists, are in a state of denial about the fact that their evidence and conclusions are disputed -- and a refusal to acknowledge  the existence of  the two 2015 papers by Dyfed, John and myself is not just academically indefensible but also an admission that their own evidence is very shaky indeed.

These are the "invisible"papers:

Brian John, Dyfed Elis-Gruffydd and John Downes (2015a). "Quaternary Events at Craig Rhosyfelin, Pembrokeshire." Quaternary Newsletter, October 2015 (No 137), pp 16-32.

Brian John, Dyfed Elis-Gruffydd and John Downes. 2015. OBSERVATIONS ON THE SUPPOSED “NEOLITHIC BLUESTONE QUARRY” AT CRAIG RHOSYFELIN, PEMBROKESHIRE". Archaeology in Wales 54, pp 139-148. (Publication 14th December 2015)

The latter article has had 1,386 reads so far, and it's intriguing that Messrs Ixer, Bevins and Parker Pearson appear to have missed it.


This brings me to "Proto-Stonehenge" and the concept of EVIDENTIAL COVER.  If you are a researcher proposing the existence of bluestone quarries, and your supporting evidence from your putative quarrying sites is very thin,  you need supporting evidence or evidential cover which has the capacity of giving you backup.  Ixer and Bevins and the archaeologists desperately need to demonstrate that their Neolithic quarrymen were digging for a purpose.  Because the rocks were sacred or because the locations were sacred?  Because there were political or religious motives for the digging?   Because they needed the monoliths for small local stone settings?  Or for  megalithic burial sites like long barrows or passage graves?  Or -- best of all from a PR point of view --for a giant ceremonial stone circle which would demonstrate both the skills of the Neolithic inhabitants of the area and the "value" or symbolism of what they created?  It's a easy hop from there into all sorts of wonderful hypothesising and mythologising -- and in such a scenario the storytelling skills of MPP of course come into their own.

If we listen to the contents of MPP's 'annual lectures between 2012 and 2018 (and look at his articles) all sorts of wacky ideas have been thrown in -- and these are the very ideas which have featured in his annual requests for research funds.   The media love it, the students who turn up for the September digs love it, and  apparently the funding bodies love it too, which is why there seems to be a continuous stream of research grants into MPPs bluestones project.  It seems that the funders are just as incapable of independent scrutiny and the editors of learned journals.

As the years have passed, the search for evidential cover has become more and more desperate.  How many sites have been investigated in the hope that they will "validate" the quarrying and human transport hypotheses?  We know of about nine, but there may well be more.  And what has been turned up?  Nothing at all -- no reasons have been found for the supposed quarrying activity, no quarrymens' villages have been found, no evidence has come forward to show that spotted dolerite or foliated rhyolite were somehow preferred to other slabs, boulders and pillars, and no evidence has been found for the use of sharp-edges or freshly quarried monoliths, either at Stonehenge or in West Wales. The proposed episodes of quarrying at Rhosyfelin and Carngoedog do not match up with anything relevant at Stonehenge, and indeed they are falsified by the radiocarbon evidence obtained and cited.  Worst of all, from the point of view of Ixer and Bevins and the archaeologists, is the fact that no trace has been found of Proto-Stonehenge............

Why don't they just give up, admit that they have already wasted enough public money on this wild goose chase, and get on with something more worthwhile?  More to the point, why is it that the funding organizations continue to feed public money into this worthless project? Answers on a postcard please.........


The "Bluestone Quarries"

Craig Rhosyfelin -- a so-called "bluestone quarry" falsified by the researcher's own dating evidence and otherwise hotly disputed.

Carn Goedog -- ditto

Carn Meini / Carnmenyn -- thought by Darvill and Wainwright to be the key spotted dolerite Neolithic quarrying site. It's now doubted that any of the Stonehenge monoliths or debitage fragments came from here.

The "Proto-Stonehenges"

Castell Mawr -- proposed as Neolithic henge linked to the "quarries" and now shown to be a Bronze Age / Iron Age feature with no links to bluestones or Stonehenge.

Bayvil -- ditto

Felindre Farchog -- prehistoric (?) enclosure / earthwork? A Neolithic henge site or quarrying settlement? After a short dig, idea dropped. Probably a medieval site.

Carn Goedog (traces below the tor, on the edge of Brynberian Moor) -- Neolithic quarrymen's village? Shown to be medieval.

Pensarn -- A site connected to Rhosyfelin? Shown in 2016 to be a Bronze Age cist grave with nearby Iron Age site.

Parc y Gaer -- A site linked to the Rhosyfelin "quarry"? Shown in 2016 to be a Roman site -- probably a villa.

Waun Mawn -- still being referred to as the possible site of "proto-Stonehenge" -- full results awaited, but apparently no links with bluestone quarries or with Stonehenge.

Bedd yr Afanc -- suggested as another proto-Stonehenge site, but previous excavations have shown up no signs of a henge or stone ring.

Banc Du -- "Might the Banc Du enclosure, with its extensive views across south-west Wales as far as the isle of Lundy in the Bristol Channel, have been a stopping place for the bluestones? Might they even have been erected here as a large stone circle?" (Parker Pearson et al, 2019). It's strange that this should be flagged up again, since Darvill and Wainwright demonstrated that there was nothing here other than a causewayed enclosure.

Saturday 16 March 2019

Strontium isotope results may be deeply flawed

Thanks to Jon for alerting me to this new research. Quote:

Reference maps showing strontium isotope data dominated by the isotopic signature of modern agricultural lime, do not show the true strontium isotopic composition of the area during prehistoric times, when the individual being studied lived, or the object being studied was in, created.  This can result in erroneous interpretations of the origin and movement of these prehistoric people and artefacts.
As readers of this blog may have noticed, I have been very circumspect about the results of work using strontium isotope measurements which purports to show the "origins" of either people or domesticated animals found in archaeological contexts.  Just put "strontium" into the search box, and assorted entries will pop up.  Here are some:


First, the report from Denmark:

Agricultural lime disturbs natural strontium isotope variations: Implications for provenance and migration studies on Neolithic feasts and far-travelled pigs

DOI: 10.1126/sciadv.aav8083

Strontium isotope maps are disturbed by agricultural lime
March 13, 2019, Aarhus University

Strontium isotopes are frequently used in archaeological studies to establish the provenance and migration history of prehistoric people and artifacts. Many of these studies may be based on incorrect data. A Danish study shows that agricultural lime can alter the composition of strontium isotopes dramatically, so that the modern isotopic signature of an area may be very different from the prehistoric signature.

A study by researchers at Department of Geoscience, Aarhus University, Denmark now shows that strontium isotopes may often be used incorrectly in archaeological studies, as the widespread use of added (strontium-rich) agricultural lime in low- to non-calcareous soils can dramatically alter the strontium isotopic composition of the surface waters running through them and the plants growing within them.
This is of special significance for strontium-isotope based provenance studies, where the strontium isotopic values measured in a prehistoric person's remains or in a given artefact are compared to measured strontium isotopic values in the surrounding, modern environment. Reference maps showing strontium isotope data dominated by the isotopic signature of modern agricultural lime, do not show the true strontium isotopic composition of the area during prehistoric times, when the individual being studied lived., or the object being studied was in created.
This can result in erroneous interpretations of the origin and movement of these prehistoric people and artefacts.
In their study published in Science Advances, the geologists Erik Thomsen and Rasmus Andreasen from Aarhus University discuss two prominent examples of this:
The iconic Bronze Age women, the Egtved Girl and the Skrydstrup Woman, who were found in Denmark in 1921 and 1935 respectively, but were recently (2015 & 2017) interpreted to have originated far away from Denmark. Moreover, the Egtved Girl was interpreted by the study's authors to have traveled back and forth between Denmark and another place, likely her homeland that was believed to be southern Germany.

These conclusions became a part of a larger framework of ideas of extended European mobility, migration, and trade, during the Bronze Age.
Conversely, the strontium data presented in the new study show that these two women could have obtained their strontium isotopic signatures within 10 km of their burial mounds, and do not indicate any cause to suspect that the women came from afar or traveled great distances during their lifetimes.

It is noteworthy that the effects of agricultural lime on the strontium isotopic composition demonstrated here is not isolated to this study's field areas in western Denmark but is likely to occur worldwide in arable areas with non-calcareous soils. The use of agricultural lime is ubiquitous in farming on less fertile soils to provide calcium for the plants and adjust soil acidity. Thus, many studies using strontium isotopes for provenance and mobility studies in these farmed low-calcareous areas may well need revision, and researchers should use care when sampling in these areas for such studies, in the future.

Explore further: The Bronze Age Egtved Girl was not from Denmark
More information: E. Thomsen el al., "Agricultural lime disturbs natural strontium isotope variations: Implications for provenance and migration studies," Science Advances (2019).


As far s I can see, this research is very sound.  And it has major implications for the UK -- especially for the Palaeozoic parts of Britain where soil acidity is high and where "liming" of the land has been going on continuously for 500 years as a means of increasing agricultural productivity.  What the researchers found was that the strontium signature in water and in plants went down quite substantially with increasing lime use in farmed areas.  That means that samples taken from such areas (and used in the creation of strontium isotope base maps) give signatures substantially lower than they should be -- and lower than they were when Neolithic / Bronze age human and farm animals were alive.  The amount of error will depend on exactly how and where the "base point" sampling was done.

So if we assume (as we must now do) that the UK maps used in "migrant provenancing research" use background figures that are too low (for example, 0.7097 against a true figure of 0.7124; or 0.7099 as against a true figure of 0.7131), they will need to be substantially corrected upwards and redrawn.  While this does not invalidate the principle that the Palaeozoic geological provinces of Britain have higher strontium signatures (mostly over 0.7100) than the SE of England, and that the chalklands may have signatures as low as 0.7075, there must be multiple inaccuracies. In this context, there is no way that the claim of Snoeck et al that they have done "accurate provenancing" of certain Stonehenge cremated bones to one little piece of West Wales can be sustained.  The claim was dodgy enough as it was -- now it surely has to be binned.

There is a sense of deja vu in all of this.  When new scientific techniques are adopted and found to be novel and useful, there is often a rush of publications announcing wondrous new discoveries.  Then a little later anomalies and errors become apparent, and correction factors begin to be introduced -- since all techniques have their limitations.  This has happened with radiocarbon dating -- and as I have said often on this blog, Quaternary chronology is still screwed up in the UK because of the chaos associated with the over-application of amino acid dating in the period 1980 - 2000.

Thursday 14 March 2019

Neolithic feasts and far-travelled pigs

A reconstruction of Durrington Walls.  Illustration courtesy Peter Dunn.

There is an interesting new study which moves along our understanding of all those pig bones derived from BBQs and orgies at Durrington Walls and elsewhere in Wessex.  The bones of 131 pigs are analysed, this time using five different isotopes in the measurements.  As with previous work using just oxygen isotope measurements, it is possible in theory to take bone samples from the sites of assumed mighty feasts, and to match them with a degree of confidence to certain parts of the UK.  The assumption is that the animals would have moved on the hoof -- and the authors argue that since pigs are not all that easy to drive over long distances (the drovers of the late 1700s found that too!) there must have been a powerful reason for people to have travelled very slowly from A to B -- attending a great ceremony might have sufficed, but maybe there were political or religious motives for the journeys as well.  As expected (with MPP as one of the authors) there is much speculation about the significance of "pan-British connectivity".

On examining the results of the measurements,  the key point is that the animals appear to have come from all over the UK, with no strong clustering in any bedrock / environmental region.  Quote:
Overall, the analysis of comparative data provides clear evidence for wide-ranging variation in isotope values in the Late Neolithic sample, indicative of animals and humans converging on the site from numerous different regions, some from a substantial distance away. This contrasts with the evidence from the flint and ceramic assemblages from Durrington Walls, which are overwhelmingly consistent with a local origin. The sarsen stones are likely to derive from the Marlborough Downs, approximately 32 km to the north, providing evidence for intraregional movement of substantial materials, but the inner ring and horseshoe of bluestones, deriving from West Wales, provides the best evidence for long-distance movement (28).


It is likely that maritime and riverine transport played an important role in these networks. This mode of transporting pigs has been used from prehistory to the present in Indonesia and Papua New Guinea (7, 36). Driving pigs overland would have represented a formidable challenge. Whatever mode of movement was used, a vast investment of effort would have been required. The transport of bluestones to Stonehenge from the Preseli Hills in Wales (28) demonstrates the challenges that communities overcame at these monumental complexes. There has been little research on mobility within Late Neolithic Britain, and the work on the bluestones provides the best evidence for interregional links. Note that some pigs have isotope values consistent with deriving from this location, further supporting the links between the Stonehenge landscape and West Wales. However, equifinality remains an interpretative issue, and it is plausible that these animals could also come from other locations (potentially southern Scotland or southwest England).

Results demonstrate that the Late Neolithic was the first phase of pan-British connectivity, with the scale of population movement across Britain arguably not evidenced in any other phase in prehis- tory. These long-distance networks were sustained by the movement not only of people but also of livestock. The complexes represent lynchpins for these networks, and it is not only the famous mega- lithic centers of Stonehenge and Avebury that were major foci. All four sites show long-distance connectivity, and there is no indi- cation that they served different networks; all drew people and animals from across Britain. After more than a century of debate concerning the origins of people and animals in the Stonehenge landscape, these results provide clear evidence for a great volume and scale of intercommunity mobility in Late Neolithic Britain, demonstrating a level of interaction and social complexity not previously appreciated. 

Note this:

"........the work on the bluestones provides the best evidence for interregional links. Note that some pigs have isotope values consistent with deriving from this location, further supporting the links between the Stonehenge landscape and West Wales." 

Circular reasoning, not for the first time.  Let's just ignore it. By and large, this new paper provides no evidence at all that West Wales was a "preferred location" in some way, with special links to the Stonehenge area.  That must have been a profound disappointment to certain archaeologists and geologists!

Aso, I have one serious reservation about this work. It gives a series of sites (in the Stonehenge and Durrington Walls area and across wider Wessex) artificial significance -- simply because they are the sites they happened to look at. The authors have not shown that the spread of apparent sources for all the pig bones was in any way unusual. They assume everything was moving towards a Wessex centre or focal point for big hog roasts and celebrations — but in the real world there might have been just as much centrifugal movement, away from Wessex and out towards all points north, south, east and west.  It's just that the edges have not been sampled.......


Here is a report from "Science" magazine:

More than 4000 years ago, people erected monumental circles of stones or wood all over southern England. Stonehenge is the most famous, but many other so-called “henges” were also built between 2800 and 2400 B.C.E. Nearby trash pits full of pig bones suggest they once hosted enormous feasts. But who was coming to these gatherings?

The pigs have now helped solve that mystery. Researchers studied the bones of 131 pigs from four henge sites in southern England, notably Durrington Walls, a wooden henge (pictured) built a mere 3 kilometers from Stonehenge, and Marden Henge, the largest of these monuments yet discovered. When the pigs were alive, their bones absorbed chemicals from their food and water, preserving a unique signature of each pig’s local environment and diet. The researchers measured the isotope ratios of five of these chemicals: strontium, oxygen, sulfur, carbon, and nitrogen.
Nearly no two pigs had the same isotope signature, the team reports today in Science Advances. That suggests they were brought to the henges from many different places rather than being raised locally for feasts. In fact, their isotope signatures match the environments in every corner of England, Scotland, and even Ireland. And that means the people attending the feasts—and contributing their pigs—likely came from as far away as western Wales, northeastern England, and even Scotland. It seems the henges have been tourist hot spots for millennia. 


R. Madgwick, A. L. Lamb, H. Sloane, A. J. Nederbragt, U. Albarella, M. Parker Pearson, J. A. Evans,  2019.  Multi-isotope analysis reveals that feasts in the Stonehenge environs and across Wessex drew people and animals from throughout Britain.
Sci. Adv. 5(3), pp 1-12.

The great henge complexes of southern Britain are iconic monuments of the third millennium BCE, representing great feats of engineering and labor mobilization that hosted feasting events on a previously unparalleled scale. The scale of movement and the catchments that the complexes served, however, have thus far eluded understanding. Presenting the largest five-isotope system archeological dataset (87Sr/86Sr, 34S, 18O, 13C, and 15N) yet fully published, we analyze 131 pigs, the prime feasting animals, from four Late Neolithic (approximately 2800 to 2400 BCE) complexes to explore the networks that the feasts served. Because archeological evidence excludes continental contact, sources are considered only in the context of the British Isles. This analysis reveals wide- ranging origins across Britain, with few pigs raised locally. This finding demonstrates great investment of effort in transporting pigs raised elsewhere over vast distances to supply feasts and evidences the very first phase of pan-British connectivity.

Tuesday 12 March 2019

The other "Carngoedog bluestone monoliths" are glacial erratics too

Carn Goedog -- the "Neolithic bluestone quarrying site" which can only be seen by those wearing rose-tinted spectacles

The other day I put up a post in which I said that it was quite extraordinary that Mike Parker Pearson and his merry gang (all 12 of them) had claimed -- in a peer-reviewed article -- that five of the Stonehenge spotted dolerite bluestones had been quarried from Carn Goedog, apparently without ever bothering to look at the shapes and surface characteristics of the stones themselves.  The word "negligence" doesn't do justice.  "Complete and utter cockup" would be nearer the mark.  Or we might say, if we want to use less colourful language, that the quarrymen have comprehensively falsified their own hypothesis.

This is the post:

Let's remind ourselves of the words used in the article:
Recent geochemical analysis has revealed two main groups of Stonehenge spotted dolerite, the larger of which (Stones 33, 37, 49, 65 & 67) can be matched most closely with Carn Goedog (Bevins et al. 2013). The second group (Stones 34, 42, 43 & 61) has not been provenanced to a specific Preseli outcrop but may derive from Carn Goedog or from nearby outcrops such as Carn Breseb or Carn Gyfrwy. 

Mike Parker Pearson, Josh Pollard, Colin Richards, Kate Welham, Chris Casswell, Charles French, Duncan Schlee, Dave Shaw, Ellen Simmons, Adam Stanford, Richard Bevins and Rob Ixer.

OK -- at the insistence of the geologist the wording is a little more circumspect than the breathless certainty of the press releases would have us believe -- but there is no doubt at all that the central message being put out by MPP to the media -- and to the world of academic archaeology -- is that 5 of the Stonehenge bluestone monoliths are certainly quarried from Carn Goedog, and that 4 others are in all probability from the same tor.

I showed in the previous post, using images from Simon Banton's website, that only one of the 5 cited stones (number 33) can be referred to as a pillar, but that is so small and so heavily abraded that it cannot possibly be referred to as a freshly quarried monolith.  The others are boulders or slabs with all of the essential characteristics displayed by glacial erratics.

I have now had a look at the other four cited or "assigned" monoliths, and things don't get any better.  Here are the images:

Stone 34 -- a well-rounded, heavily-abraded elongated boulder

Stone 42 -- another boulder with an irregular shape, rounded off edges and a great 
deal of surface damage

Stone 43 -- a well-rounded small boulder with a heavily weathered surface

Stone 61 -- another strange slab with rounded edges and heavy surface weathering.  This one does look as if it might have come from Carn Goedog or one of the other tors  -- not because it looks more like a pillar, but because the curving fracture is characteristic of the dolerite sills exposed to the west of Carn Alw.  But again, it would defy logic to claim that this looks like a freshly quarried orthostat.

So if we look at these nine "bluestone orthostats" which are deemed to "prove" that the Carn Goedog quarry actually existed, the best that can be said is that their geochemical and petrological characteristics may indicate an origin somewhere near Carn Goedog, but that as a group they can only be interpreted as glacial erratics entrained and deposited by the ice of the Irish Sea Glacier.

Sunday 10 March 2019

Ice moulded slabs on Carningli

This is an interesting photo, taken from near the northern tip of the Carningli upland, looking towards the SW.  Cwm Gwaun is in the distance.  I'm not sure of the photo source -- somebody put it onto the web.

You can see the embankments from the fortified site, and the roughly rectangular defended area where the Iron Age village was located.  But look at the slopes beneath the embankment in the lower middle part of the photo.  They are remarkable in that very little detritus or rock breccia has accumulated on them -- and if you look carefully you can see ice-smoothed or moulded slabs outcropping in a number of locations.  Elsewhere, on the flanks of the crag, there are considerable accumulations of boulders and scree.

This is no coincidence -- every time there is a glaciation involving the Irish Sea Glacier, the ice comes in from the north (the foreground of the photo) and under that assault most of the superficial debris that might have accumulated is simply swept away.  The last time that happened was during the Late Devensian glaciation, around 20,000 years ago.  I still think that at the peak of this glacial episode, the whole of Carningli was overriden by ice.  There are other apparently ice-moulded features near the summit crags.

Wednesday 6 March 2019

This blog goes onto the UK web archive

A number of faithful blog followers have raised with me the possibility that the whole blog (ten years' worth of posts and jolly discussions, almost 1.5 million hits and almost 2,500 posts) would probably disappear without trace if I was to die tomorrow or to be done over by some shady archaeologists dressed in long raincoats, trilby hats and dark glasses.........

Anyway, I was advised by a colleague that web sites and blogs deemed to have high value as representing aspects of the life of the nation could be  placed onto the national "web archive" and after going through some sort of screening process I have now been invited to issue a license to the National Library of Wales for the blog to be copied and archived.  That means that its content will be available till the end of time -- not necessarily on the open access web, but via the copyright libraries:  the British Library; the National Library of Scotland; the National Library of Wales; Bodleian Libraries, Oxford; University Library, Cambridge; and the Library of Trinity College, Dublin.

The license has been processed, and so the blog in its entirety will be available via the libraries just as soon as the copying is done.  After that it will be updated at six-monthly intervals.

Further information can be obtained from the UK web archive site, here:

So there we are.  We can all sleep easily in our beds from now on, in the knowledge that all of the rude things I have said about archaeologists are available for everlasting scrutiny and approval.

Clast-supported and matrix-supported raised beach deposits

Clast-supported storm beach at Abermawr, Pembs.  Here some of the cobbles are made of Irish Sea till!  There are no sediments in the spaces between the cobbles -- they are washed away to lower levels during every high tide. In the lower part of the storm beach ridge, beach sand fills all the spaces.

Matrix-supported raised beach cobbles in layers, Poppit, Teifi Estuary, West Wales

Matrix-supported raised beach at Porth Killier, St Agnes, Isles of Scilly.  The spaces between the cobbles are filled with a reddish colluvium

Cemented clast-supported raised beach from the Gower.   There is some fine-grained sediment between the clasts, but the clasts are very tightly packed together.

There is some discussion in the literature about the significance of clast-supported raised beach deposits and those that are matrix-supported.

We are talking, of course, about conglomerates, or unconsolidated conglomerates in which the clasts are for the most part well-rounded, and normally of a consistent size -- that size being dependent, to some degree, on the precise dynamics relating to wave energy and the type and availability of materials.

Clast-supported raised beaches are those in which the cobbles, boulders or pebbles are in contact with each other, with little no finer material in the intervening spaces.    In other words, the clasts "support" each other.

In contrast, in matrix-supported raised beaches the cobbles, boulders or pebbles may not actually be in contact, and the bulk of the material may be made up of a fine-grained matrix.  In other words, the matrix supports the larger particles.

There is a suggestion that clast-supported raised beaches are somehow more "authentic" and maybe younger than raised beaches that are matrix-supported;  and there are some -- in the debates about the Scilly Isles in particular -- who have argued that the matrix-supported deposits are very old indeed, and are in secondary positions, having been moved downslope during cold-climate or even periglacial conditions.

I do not think this is a sustainable argument -- and I think it would be a grave mistake to label one type of beach as old and redeposited, and the other as young and in situ.  To some degree all open-work storm beaches are kept clear of finer sediments as long as they are being washed and constantly rearranged by storm waves -- but as soon as they are isolated from marine coastal processes, modification kicks in.  If a beach on a platform is beneath a steep slope, it is inevitable that fine-grained sediments (generally labelled as "colluvium" will be washed down over and into the beach, gradually filling all available spaces.  In more open environments, alluvial or river processes will fulfil the same function.  And in yet other situations blown sand will be carried onto and into the beach, with sand grains filling the spaces.  Then we might have combinations of all these processes.

The main message is this:  if a raised beach is in a particular position, resting on a rock platform, it is in all probability in its "right" stratigraphic relationship with the deposits resting on it.  If you want to argue for something different, then you have to come up with some pretty smart additional evidence to support your case..........

A cholera cemetery

Not standing stones from the Neolithic or Bronze Age........ but striking similarities.

Photo posted on Facebook by Jeremy Thomas. A Cholera cemetery connected to the Tredegar Ironworks. High on a bleak moorland above Abertwssyg, Rhymney valley. Cefn Golau Cemetry -- there is a location map on Coflein.

A fantastic photo, but something unutterably sad. Apparently there were several such in Wales, but the other are lost.  Up to 600 people may be buried here -- many, I dare say, in unmarked graves.

Here is a link:

The Coflein record:

Tuesday 5 March 2019

Holocene peat beds exposed again at Abermawr

A fine photo from Casey Cilshafe showing the exposed peat beds exposed as a result of the lowering and re-configuration of the pebble storm beach.  As we can see, the surface has been modified by rills -- partly because of sea water running downslope when the tide is high, and partly because impounded fresh water from behind the storm beach is infiltrating the beach and flowing down towards the sandy beach (which is submerged in this photo).  This exposure has been partly re-covered by pebbles in the past few days.

Stream water flowing across an exposure of the peat beds -- this is not pure peat but "peaty mud" containing abundant plant remains and occasional tree roots and fallen branches.

Remains of a tree stump and root system on the surface of the peat bed.

A sharp contact  between an overlying peat layer (up to 10cm think nowadays -- it may well have been much thicker prior to millennia of erosion beneath the storm beach) and light grey / bluish clay with sand and silt.  This does not appear to be classic Irish Sea Till -- but it is probably part of the glacial sequence.

Vertical section showing how sharp the contact is between the  dark brown peat layer and the underlying grey clay.

About 10 m away from the peat bed exposures, there is an exposure of Irish Sea till, also showing through the pebble beach.  This confirms how thin and ephemeral the storm beach actually is.  Near the trowel we see a disintegrated lump of lignite.  Shell fragments are also present, as in the main exposure of Irish Sea till in the northern drift cliff.

Some previous posts on the Abermawr submerged forest and peat beds:

The raised beach -- another Abermawr exposure

After the winter storms, and some debris slumps and cliff-falls, there is a new exposure of the raised beach on the old rock platform at Abermawr North.  As we can see here, it is overlain by brecciated slope materials in a sandy matrix -- this is the "lower head" as previously described many times.   I was not prepared to risk life and limb to climb up and examine it -- the exposure is very difficult to access.  I estimate it to be about 3 - 4 m above HWM.

Now we can see the raised beach in patches for about 15m laterally -- always in the same position on top of a low rock cliff, and always overlain by "the lower head".  

I think I can see traces here of a rather smoothed rock platform at the top of the little cliff.  Some day somebody younger and fitter than me will get up there and take a proper look.....

Monday 4 March 2019

The Devensian ice edge on the Isle of Samson

Now that David Mawer of the Isles of Scilly Wildlife Trust has recorded the presence of gravelly deposits apparently of glacial origin on the west side of Samson, and now that Prof James Scourse has accepted that they may well be authentic indicators of Late Devensian ice, let's ponder on the precise positioning of the ice edge.

The map above is a first shot -- it will inevitably be corrected.  But we can assume several things, since topographic controls will have come into play on this island which has two hills, each about a hundred feet high.  First, there were probably lobes of ice pushing into the strait between Bryher and Samson, and into St Mary's Road to the south.   On the island itself, the ice may have pressed against North Hill and South Hill without being able to override them.  And there may have been a small lobe in the col between the hills.

Let's see how this all works out when somebody gets round to doing some more detailed fieldwork by following the shoreline right around the island. 

Saturday 2 March 2019

Annet, Samson and Tean

The isle of Annet, west of St Agnes in the Scilly Archipelago.  It's a low island, washed by storm waves -- so there may or may not be coherent glacial and related sediments there which are worth examining.......

The isle of Samson, where, according to James Scourse, glacigenic sediments have now been discovered on the west side by David Mawer, senior conservation officer of the Isles of Scilly Wildlife Trust.  In this photo South Hill is in the distance and North Hill is in the foreground.  We also see White Island, off the west coast.

The isle of Tean, with St Helen's beyond.  This irregularly-shaped small island may have been affected -- at least in part -- by Late Devensian glacier ice.  Bottom right, we can see Lowertown on St Martin's.