c. 3000 BC – c. 1500 BC

A basic henge is simply a, roughly, circular space, enclosed by a ditch and bank earthwork, with an entrance, or entrances, leading to the centre. As a rule (there are, of course, exceptions), the bank is built up on the outside of the ditch – an arrangement usually regarded as being unsuitable for defensive purposes. Most henges contained various types of feature – for instance, arrangements of stones, timbers or pits.

Swinside Stone Circle, Cumbria.

It is generally thought that they were ceremonial centres, where people would gather together to take part in religious rituals and other communal activities. Stone circles (where a space is enclosed by standing stones), presumably fulfilling a similar role, were also constructed. Some henges combine a ditch and bank with a stone circle – including the two most famous (and magnificent), Stonehenge and Avebury, both in Wiltshire.

The archaeological term ‘henge’ is actually a back-formation from the name Stonehenge (originating in Anglo-Saxon times – having the meaning ‘hanging stones’), but Stonehenge, which represents the culmination of stone circle engineering, is not at all a typical henge.


There are four wonders which may be seen in England… The second is at Stonehenge [Stanenges], where stones of remarkable size are raised up like gates, in such a way that gates seem to be placed on top of gates. And no one can work out how the stones were so skilfully lifted up to such a height or why they were erected there.
Henry of Huntingdon, Historia Anglorum (I, 7), written c.1130,
translated by Diana Greenway[*]

Stonehenge’s first chronicler, Henry of Huntingdon, mused on ‘the how’ and ‘the why’ of its construction, and nine centuries later archaeologists are still trying to provide satisfactory answers to those two basic questions. Considerable progress has, however, been made in unravelling Stonehenge’s history, and, thanks to Radiocarbon dating, answering a third important question: ‘the when’.

STAGE 1, c.3000 BC.  Work on Stonehenge began in about 3000 BC, with the digging of the ditch and bank – enclosing an area roughly 92 metres in diameter.[*] Uncharacteristically for a henge, the bank was on the inside of the ditch. (There was also a low outer bank, now severely eroded by agriculture.) When new, the chalk bedrock of Salisbury Plain would have ensured the earthwork stood out boldly in the landscape. There were two entrance causeways – the main, larger, one to the north-east and a smaller one to the south. There were probably arrangements of timber posts within the henge and at the main entrance.[*] On the inner side of the earthwork circle there may have been a circuit of fifty-six timber posts or standing stones. These hypothetical posts or, more controversially, stones, would have stood in the filled-in pits known as Aubrey Holes.[*] The most popular theory in recent years has been that they held timber posts, but archaeologists of the Stonehenge Riverside Project (SRP) excavated one in August 2008, and the condition of its chalk base led them to believe that it had once supported a standing stone.[*] Researching archive material revealed that similar crushing of the chalk had been observed in other Aubrey Holes during previous excavations.


The most intensive period of investigation at Stonehenge was between 1919 and 1926, carried out by William Hawley. An English Heritage press-release of January 2001 notes that: “Hawley excavated a large proportion of the eastern part of the monument, in a programme of work that is considered over-vigorous by today’s standards.”
Hawley excavated thirty-two Aubrey Holes – it was apparently his assistant, Robert Newall, who gave the pits that name. They were numbered in a clockwise direction – Aubrey Hole 1 being at about 2 o'clock, adjacent to the south end of the main entrance. Numbers 1–30, 55 and 56 were excavated under Hawley. Richard Atkinson (in association with Stuart Piggott and J.F.S Stone) excavated numbers 31 and 32 in 1950.
Hawley initially believed that the pits had held stones, but later, apparently influenced by the excavations at nearby Woodhenge in 1926–7, changed his mind in favour of timber posts. Atkinson (who directed excavations, and reconstruction work, at Stonehenge from 1950 to 1964) considered the Aubrey Holes to be, simply, pits.
Mike Parker Pearson is at pains to point out that one of his colleagues at the 2008 excavation, Julian Richards, was not convinced there was unequivocal evidence that the Aubrey Hole under investigation, number 7, had once held a stone.[*]

The SRP have gone further, proposing that the Aubrey Holes originally contained ‘bluestones’, which have generally been seen as much later arrivals at Stonehenge.[*] Most of the thirty-four Aubrey Holes that have so far been excavated contained cremation burials.


Excavations have, so far, found – in the Aubrey Holes, but also in the ditch and in the bank sides – 63 cremation burials (according to Mike PP.[*]  Mike Pitts has estimated that there might be as many as 240 cremations – making Stonehenge the largest known prehistoric cremation cemetery in Britain.[*]
A sample of charcoal, associated with a cremation in an Aubrey Hole, was radiocarbon dated around 1952 – early in the history of the technique – and the resulting date-range (3020–1520 BC) is too wide to be very helpful. In 2007, however, samples of remains from three cremation burials, which had been kept by The Salisbury Museum, were radiocarbon dated. The oldest sample was from Aubrey Hole 32 (excavated 1950). Its date (3030–2880 BC) placed cremations at Stonehenge from its very beginning – they had previously been believed to be later additions – and confirmed the, already generally accepted, notion that the Aubrey Holes belonged to Stonehenge’s earliest stage. The other two samples were from the ditch (excavated 1954), and their dates (2930–2870 BC and 2570–2340 BC) showed that cremation burials were deposited at Stonehenge over a period of at least 500 years. So few burials over such a long period might tend to suggest that the dead individuals belonged to a ruling elite. (In addition, two fragments of non-cremated human skull recovered from the ditch have produced dates of 2890–2620 BC and 2880–2570 BC.)
In 1935, a number (59 according to Mike PP) of cremations, unearthed by William Hawley during the 1920s, were reburied (accompanied by a commemorative plaque) in Aubrey Hole 7. This Aubrey Hole was reopened in August 2008, and the remains, which were mixed together in a dense mass, removed for further examination. (An undisturbed cremation burial was also found on the edge of the Aubrey Hole.) Analysis of the bone fragments confirmed that cremations were buried at Stonehenge during the period c.3000 BC–c.2500 BC, and showed that they were of both men and women, with a few children.[*]

The site seems to have stayed in this form for about half a millennium. The ditch was neglected, and slowly filled with sediment. In about 2500 BC, however, Stonehenge entered a period of development during which stones were arranged and rearranged, to eventually produce the monument whose impressive wreckage remains today.

STAGE 2, c.2500 BC.  The stone settings built at the centre of Stonehenge fall into two categories: those made from bluestones, and those made from sarsens. Sarsens are naturally occurring sandstone slabs, and recent research suggests that the majority of those used at Stonehenge were probably sourced from the vicinity of West Woods, 15 or so miles to the north.[*] In about 2500 BC, a circle (30 metres in diameter) of thirty sarsen uprights (standing approximately 4 metres above ground level), topped off by a continuous circuit of sarsen lintels, was erected around a horseshoe arrangement (the open end pointing in a north-easterly direction) of five, massive, sarsen ‘trilithons’.[*] Construction of this megalithic masterpiece would have taken an enormous number of man-hours. Once the, not inconsiderable, task of dragging the rough sarsens to the site had been accomplished, these, extremely hard, rocks were shaped and smoothed, and had joints sculpted into them using the most rudimentary of stone tools.[*] Having levered the uprights vertical, the lintels then had to be raised (up to 7 metres above the ground), and dropped into position.[*]

The first appearance of bluestones at the centre of the henge – indeed, their first definite appearance anywhere on site – was in a concentric double arc setting. The evidence for this placement is provided by the, so-called, Q and R Holes, which lie between the sarsen circle and trilithon horseshoe settings.


Plan of the sarsen settings at the centre of Stonehenge. The uprights shown black are those standing today. The orbit of the Q and R Holes lies between the trilithon horseshoe and the circle.
Richard Atkinson, who gave the Q and R Holes their designation, thought the setting (which is known to have been of bluestones because of chippings found in some of the holes) formed a full circle. However, only an eastern arc is clearly evident, though there are also isolated stoneholes in a couple of places on the same orbit. Possibly the setting was abandoned before it was finished, or perhaps later soil disturbance has effectively erased some of the evidence. At any rate, the original form of this setting is not certain, but there does appear to have been some kind of entrance arrangement at the north-east, opposite which there may have stood a large upright stone. (William Hawley found a worn axe-head, from Cornwall, at the bottom of a large circular pit in which the stone may have stood.) It may be that the large stone is still at Stonehenge, having been relocated, now known as the Altar Stone (of which more later).

There are no reliable radiocarbon dates for the double arc, but Richard Atkinson’s excavations, in the mid-1950s, indicated that it was the earliest stone setting in the henge’s centre. As a result, it has usually been placed a short while before the arrival on the scene, in about 2500 BC, of the massive sarsens – the theory being that this rather modest bluestone arrangement was quickly abandoned and removed, to make way for a much more ambitious scheme.[*] The SRP has, however, questioned the supposed stratigraphic relationship between the bluestone arcs and the sarsens, and suggested that the settings were contemporaries – that the sarsen horseshoe, sarsen circle and bluestone arcs were all part of a single design.


In 1993–4, English Heritage funded a project to reassess the results of 20th century excavations at Stonehenge. As part of the work, an extensive programme of radiocarbon dating was carried out. The fruit of this labour was published as Stonehenge in its Landscape in 1995 (in effect, it is this work that the SRP has refined and developed). It was thought that there was one deer antler sample relevant to the construction of the circle, and three relating to the trilithons. The spread of dates derived from the samples, however, was very large. Not only that, it seemed as though the circle had been built before the trilithons were erected, which, on purely practical grounds, seems unlikely. More recent research, by the SRP, demonstrated that two of the samples, which had been thought to relate to the erection of the trilithons, actually belong to later work. Eliminating these two samples led the SRP to date the construction of both the trilithon horseshoe and the circle 2620–2480 BC. (If all the samples are taken into account, the estimated date for the trilithons shifts to 2440–2100 BC.)[*]
A small excavation was carried out in the spring of 2008, by Timothy Darvill and Geoffrey Wainwright (who were not part of the SRP), in an attempt to find material that would date the bluestones’ placement in the Q and R Holes. No meaningful dating evidence was produced,[*] but the excavation convinced Darvill and Wainwright that Atkinson had been wrong about the stratigraphic relationship between the Q and R Holes and the sarsens. As a result, they, and subsequently Mike Parker Pearson (director of the SRP), came to the conclusion that the bluestone setting represented by the Q and R Holes was part of the same phase of construction as the sarsen settings. As a result, the SRP scheme dates the erection of the trilithon horseshoe, the bluestone double arc and the sarsen circle, in that order, to 2620–2480 BC.[*]

The bluestones underwent subsequent rearrangement (of which more later), and it would seem that there were about eighty of them in total – the remains of about forty-three survive (some are just buried stumps). As already mentioned, the SRP has suggested that there were bluestones planted in the fifty-six Aubrey Holes centuries before their first appearance at the henge’s centre. In 2009 the SRP excavated a newly discovered feature, just over a mile to the south-east of Stonehenge, adjacent to the river Avon. It had been the site of a small stone circle – about 10 metres in diameter, with an estimated twenty-five standing stones – at about the same time as Stonehenge’s beginnings.[*] The SRP have proposed that these stones were also bluestones. They suggest that the stones from both the Aubrey Holes and this newly found monument (which has been dubbed Bluestonehenge) were recycled in the centre of Stonehenge.[*]

The bluestones are of a variety of rock-types, and most of them evidently originated in and around the Preseli Hills of south-west Wales.[*] To transport dozens of two-ton stones some 150, as the crow flies, miles, over land and water, using muscle power alone, is an extraordinary achievement – in fact too extraordinary for some authorities to accept as credible. Evidence for the alternative theory – that they were transported to the vicinity of Stonehenge by a glacier (which would account for the motley mix of rock-types) – is though, at the moment, not overwhelming, and the debate continues.


In 2010, geologists Rob Ixer and Richard Bevins matched chippings of a particular rhyolite found at Stonehenge (it is assumed that the chippings are from a bluestone or bluestones that are now lost or buried) with rock from an outcrop called Craig Rhos-y-felin, on the north side of the Preseli Hills. An archaeological team (favouring the idea that the bluestones were transported to Stonehenge by manpower), headed by Mike Parker Pearson, carried-out five seasons of excavations at the site (2011–15), and published a paper in December 2015 (Antiquity Vol. 89, Issue 348), in which they described the “Welsh bluestone megalith quarry for Stonehenge” that they had uncovered.[*] However, the very same month, geologist John Downes and geomorphologists Dyfed Elis-Gruffydd and Brian John (favouring the idea that the bluestones were transported by a glacier) published a paper (Archaeology in Wales Vol. 54) refuting the conclusions made by Parker Pearson’s team:
… detailed investigations of the features at the site by the three authors of this paper suggest that there is no sign of human quarrying activity – either related to the removal of monoliths intended for transport to Stonehenge, or for any other purpose…
It is suggested, on the basis of careful examinations of this site, that certain of the “man made features” described have been created by the archaeologists themselves through a process of selective sediment and clast removal. An expectation or conviction that “engineering features” would be found has perhaps led to the unconscious fashioning of archaeological artifices.[*]
In August 2018 a paper submitted by a group of researchers, including Mike Parker Pearson, was published in Scientific Reports. The Abstract states:
Cremated human remains from Stonehenge provide direct evidence on the life of those few select individuals buried at this iconic Neolithic monument… New developments in strontium isotopic analysis of cremated bone reveal that at least 10 of the 25 cremated individuals analysed did not spend their lives on the Wessex chalk on which the monument is found. Combined with the archaeological evidence, we suggest that their most plausible origin lies in west Wales, the source of the bluestones erected in the early stage of the monument’s construction. These results emphasise the importance of inter-regional connections involving the movement of both materials and people in the construction and use of Stonehenge.[*]
In February 2021, Mike Parker Pearson and his archaeological team published a paper (Antiquity Vol. 95, Issue 379) proposing that some of Stonehenge’s bluestones had stood in a stone circle at Waun Mawn, in the Preseli Hills, before being uprooted and moved to Salisbury Plain.[*] Such few stones that remain of the erstwhile monument at Waun Mawn are of ‘unspotted dolerite’, as is a flake from one of the stones that has been removed from the site. In the paper, the stone to which this flake belonged is said to have left an “unusual pentagonal imprint” on the bottom of its stonehole, and this is said to match the base cross-section of an unspotted dolerite which still stands at Stonehenge (Stone 62). The authors conclude:
… it seems that Stonehenge stage one was built – partly or wholly – by Neolithic migrants from Wales, who brought their monument or monuments as a physical manifestation of their ancestral identities to be re-created in similar form on Salisbury Plain …
The story presented by Parker Pearson et al. – which invoked a nonsensical fairy-tale (called a “legend” in the paper) told by Geoffrey of Monmouth, about Merlin building Stonehenge with the stones of a giants’ stone circle brought from Ireland: perhaps it “contains a grain of truth” – garnered much publicity. The BBC made a documentary, Stonehenge: The Lost Circle Revealed – it was said that the base of Stone 62 and the unusual stonehole imprint “fitted like a key in a lock”, and Mike Parker Pearson referred to Stonehenge as “a second-hand monument”.  Somewhat embarrassingly, subsequent geological analysis showed that the stone used at Waun Mawn did not match Stonehenge Stone 62. The stones at Waun Mawn were probably sourced locally, from Cerrig Lladron about 1½ miles to its southwest. The probable source of Stone 62 has been identified as Garn Ddu Fach, an outcrop about 4 miles to the east-southeast of Waun Mawn.[*]

Professors Timothy Darvill and Geoffrey Wainwright consider the bluestones to be Stonehenge’s raison d'être. They believe the stones were thought to have healing properties, and that Stonehenge was “a prehistoric Lourdes” – a place of pilgrimage, to where people would travel in the hope of being cured.[*]

There were four isolated sarsens, known as the Station Stones, erected, just inside of the bank, on the corner points of a remarkably accurate rectangle framing the central settings.


The Station Stones were placed, more or less, on the same circuit as the, by then filled in, Aubrey Holes. Two of them – they are undressed sarsens – remain. One (Stone 93), though still erect, is just a 1 metre high stump. The recumbent Stone 91 is 3 metres long. Stones 92 and 94 are gone, but originally they were partially buried by low earth mounds – referred to as the North and South Barrows. There are no radiocarbon dates available for the Station Stones. All that can be said for certain is that they post date the Aubrey Holes (shown yellow on the plan).
Based on the similarity between William Hawley’s description of a surface he had uncovered beneath the South Barrow, and their own observations of house floors at the nearby site of Durrington Walls, the SRP believe that the remains of D-shaped buildings underlie the North and South Barrows.
The SRP dates the Station Stones c.2500 BC, i.e. in the same time bracket as the construction of the sarsen settings (and, in their opinion, the bluestone arcs) at the henge’s centre. If, however, sighting across the diagonals of the rectangle was an important part of the function of the Station Stones, then they would certainly have to predate the sarsen settings (though not necessarily the bluestones), which would have blocked the view.
There are theories concerning the astronomical significance of every feature of Stonehenge – that the positioning of each pit, post, stone and entrance was intentional, creating alignments that were important to the beliefs of the builders. Aubrey Burl, however, makes a rather more prosaic suggestion regarding the Station Stones: “a plausible interpretation is that they created a template whose intersecting diagonals established a centre for the forthcoming sarsen ring that the world knows as Stonehenge. If so, it might seem no more than prehistoric idleness that failed to withdraw them when the circle was completed.”[*]

In about 2450/2400 BC, the so-called Beaker People arrived in Britain.[*]

STAGE 3, c.2400 BC.  Around 2400 BC, the Avenue, a processional route leading to Stonehenge’s main, i.e. north-eastern, entrance, was constructed.


The Avenue is almost 3 kilometres long. From the henge’s north-east entrance, the first 500 metres runs dead straight, in the direction of the midsummer sunrise. The walkway is roughly 12 metres wide, flanked on each side by a bank and, outside that, a ditch. Though none are now evident, there may have been standing stones, at intervals, along the banks – it seems that holes were discernible in the early 1720s, when the antiquarian William Stukeley investigated Stonehenge. The Avenue then turns abruptly to the east, and bends southwards towards the river Avon. Although the Avenue’s earthworks are visible from the ground in the vicinity of Stonehenge itself, ploughing has virtually erased them along much of its length. In 1973 its route had been traced to within 150 metres of the Avon. At the end of their 2008 season, archaeologists of the SRP were investigating alongside the river when they discovered, not the parallel ditches of the Avenue that they were hoping for, but an intriguing curved ditch. Excavations had to wait until the following year. It turns out that the Avenue terminates at the monument which has been dubbed Bluestonehenge. Bluestonehenge began life as a stone (presumed to be bluestone) circle. After the stones were removed (presumably for use at Stonehenge), a henge, some 30 metres in diameter, was built.
The SRP discovered that the north-east entrance to Stonehenge is at the end of two parallel natural ridges, about 200 metres long, between which run parallel ‘periglacial gullies’ – these sediment-filled gullies would have produced stripes in the vegetation during dry spells – that just happen to align with the midsummer sunrise/midwinter sunset axis, and that the ridges underlie the banks of the Avenue (prompting speculation that this natural phenomenon was the reason that Stonehenge was built where it was). Sarsen and bluestone chippings – the debris of stone-dressing – found beneath the banks indicate that the Avenue post dates the construction of the henge’s central settings, and this is confirmed by dates obtained from four samples (of antler and animal bone) recovered from the Avenue’s ditches. These dates have usually been interpreted to suggest that construction of the Avenue spanned several centuries (the straight section from the henge’s entrance being built first), but the SRP have concluded that the whole thing was built around 2400 BC (2500–2270 BC), and then the silted-up ditches were cleared-out around 2200 BC (2290–2120 BC).

The henge’s silted-up ditch was re-cut.[*]

The, now recumbent (and colourfully named), Slaughter Stone is the sole remnant of a megalithic portal into the henge. Standing in the Avenue, but not on the centre-line, about 78 metres distant from the circle centre, is a massive unworked sarsen, called the Heel Stone.

Photograph courtesy of Adam Stanford
© Aerial-Cam/SRP 2008.
The Avenue was, until recently, cut by the A344 road. In 2013 the road was closed, and the section adjacent to Stonehenge dug-up and grassed over.
The Heel Stone stands in the foreground. Beyond the tourist walkway is the Slaughter Stone, lying in the entrance to Stonehenge.
Photograph courtesy of Adam Stanford
© Aerial-Cam/SRP 2008.
The Avenue was, until recently, cut by the A344 road. In 2013, the road was closed, and the section adjacent to Stonehenge dug-up and grassed over. The Heel Stone stands in the foreground. Beyond the tourist walkway is the Slaughter Stone, lying in the entrance to Stonehenge.


The Avenue does not, in fact, align very well with the entrance causeway. A theory that this is a false impression, due to the state in which William Hawley left the henge ditch and bank following his excavations in the 1920s, is demonstrably wrong, since an aerial photograph taken in 1906 shows the south-east terminal of the henge ditch extending into the path of the Avenue.[*]
The Slaughter Stone is a large (about 6½ metres long), worked, sarsen slab. Next to it are holes where two other stones (D and E) once stood. Radiocarbon assays of two antler samples found in Stonehole E provided an estimated date of 2470–2275 BC for the erection of the Slaughter Stone and its two companions.[*] The SRP, though, suggests that these stones were raised prior to this, in Stage 2 (which they date 2620–2480 BC), and propose that Stones D and E were removed during Stage 3 (which they date 2480–2280 BC).[*] There is, however, some evidence to suggest that two of the portal stones were present and still standing during the 17th century (see below), but when William Stukeley investigated Stonehenge, in the early-1720s, only the fallen Slaughter Stone remained.
Until recently the A344 road cut across the Avenue, just avoiding the Heel Stone – the above-ground portion of which is estimated to weigh 29.92 tons. Stoneholes show that two other monoliths (B and C) once stood between it and the henge’s entrance causeway.
In 1979 a stonehole was found (97) which suggested that, at one time, the Heel Stone had a partner that was, at some stage, removed and a ditch dug around the, now solitary, Heel Stone. Rodney Castleden comments:
I can offer no explanation at all for the removal of stone 97: as far as I know, no one has yet offered a plausible explanation. The pair of Heel Stones made a perfect ritual doorway for the midsummer sun …[*]
There is, though, a plausible explanation (one favoured by the SRP). Evidently Stonehole 97 was cut into the infill of an earlier, larger, hole. It may well be that this was the Heel Stone’s natural resting place. It was first erected into Stonehole 97, but was later moved into its present position.
Aubrey Burl suggests that the Heel Stone was first erected long before the henge was dug. He argues that it was one in a series of stones marking the line of an ancient trackway, and was later integrated into the monument.[*]
In 1620, King James I instructed the architect Inigo Jones to investigate Stonehenge. The result was a book, Stone-Heng Restored, in which Jones presents his vision of how the monument originally looked (below). He gives the Slaughter Stone a partner – but he also shows three entrances and six trilithons.
It seems, however, that, though he believed there were three entrances, Jones’ notion of how they looked was based on the observation of only one. After his death, his protégé John Webb wrote (Vindication of Stone-Heng Restored, 1665) that: “he [Jones] hath described in his Draught Two Stones … these were the two parallel Stones that stood upon the inside of the Trench, at the Entrance from the North-East”.
In John Aubrey’s plan, The Ichonographie of Stoneheng as it remaines this present yeare 1666 (below), the Slaughter Stone and its colleague are both shown, seemingly still standing. Further, Jones envisaged an opposing pair of uprights on the outside of the ditch, and Aubrey records one of them, again seemingly still standing, on his plan.

The whole monument is aligned on the axis of the midsummer sunrise and midwinter sunset. The tendency is to assume that it was the midsummer sunrise, as seen looking out from the centre of the stone settings down the Avenue (at the time Stonehenge was built, the sun would have risen just to the west of the Heel Stone), that was the main point of this alignment. It may be, however, that the midwinter sunset (the sun would have set between the uprights of the Great Trilithon), at the opposite end of the axis, was, at least, equally significant.

An arc of four stoneholes may be all that remains of a bluestone circle, built within the horseshoe of sarsen trilithons.

STAGE 4, c.2200 BC.  Round about 2200 BC, bluestones were arranged in a circle, running inside the sarsen circle. According to the SRP’s proposed sequence, the bluestones of this new circle had, until this time, stood in the Q and R Holes (the previous notion was that they had been removed from the site when the sarsens were installed, only to be brought back at this time). The small bluestone circle inside the sarsen trilithon horseshoe, assuming there was such a structure, was apparently restyled into an oval.[*] No actual stones are now present at the north-eastern end of the oval (though there are stoneholes), and it has been supposed they were removed in prehistoric times in order to produce a bluestone horseshoe setting that echoed the sarsen trilithon setting. It is perhaps the most controversial proposal of the SRP that this bluestone horseshoe did not exist. Based on the observations of Timothy Darvill and Geoffrey Wainwright, made during their 2008 excavation, it is suggested that the missing stones were robbed in Roman, and later, times, and that the prehistoric setting was always an oval.

Placed within the oval/horseshoe, near the south-western end – at the heart of the monument – was the, so-called, Altar Stone: a five-metre-long, dressed, greenish sandstone block of Welsh origin.[*] The Altar Stone is recumbent today, and broken in two, pinned down by stones from the collapsed Great Trilithon. [Plan]


It isn’t clear how many uprights formed the bluestone circle – perhaps sixty. After the removal of a number of stones at the north-eastern end of the oval, a horseshoe arrangement of, it seems, nineteen bluestones remained. In Stonehenge in its Landscape (1995), Rosamund M.J. Cleal states:
The present arrangement of Horseshoe and Circle must have been the product of an intentional remodelling of the monument… it is clear that at some stage the Horseshoe and Circle were intended to be the major Bluestone setting of the interior …
Chapter 7 (p.231)
But in Stonehenge: Exploring the Greatest Stone Age Mystery (2012), Mike Parker Pearson declares:
Out goes the bluestone horseshoe – there is no archaeological justification for regarding it as a separate prehistoric entity: it is more probably simply the degraded remains of the bluestone oval.
Chapter 20 (p.311)
The distinguished archaeologist Francis Pryor (Stonehenge, The Story of an Ancient Landscape, 2016) is, however, sticking to the old interpretation:
Somewhat later in Stage 4 the bluestones at the northeastern end of the oval were removed, to leave a U-shaped setting that echoed the much larger Trilithon Horseshoe immediately outside and above it.
Chapter 7
At any rate, whilst the stones employed in the oval/horseshoe are shaped and polished dolerite, the stones in the circle are a motley collection of mostly unworked stones. Aubrey Burl opines:
The circle bluestones, like leftovers, were hastily and carelessly set up as though they had been no more than a method of using up the remaining stones.
A Brief History of Stonehenge (2007), Chapter 9 (p.245)
Whilst Rodney Castleden says they:
… skulk among the sarsens, like dwarfs among giants, in an irregular ring …
The Making of Stonehenge (1993), Chapter 7 (pp.176–7).
Some bluestones display evidence of their use in previous settings. For instance, Stone 68 (shown rightabove) has a grove down one edge which could well match up with a projecting tongue on Stone 66 (which is now just a ground-level stump). Both of those stones were reused in the oval. Two stones from the circle (Stone 36 and Stone 150, both of which are now fallen and partly buried) have mortice holes carved in them, so it seems that at one time they were the lintels of bluestone trilithons.
The Altar Stone is the largest bluestone at Stonehenge. The shaping at one end indicates that it did, at one time, probably stand erect, but not necessarily in its final placement. The Altar Stone may once have been part of the setting represented by the Q and R Holes. It may have stood in the large circular pit (at the bottom of which a worn Cornish axe-head was found), in the orbit of the Q and R Holes, at the south-west, facing a possible entrance arrangement at the north-east. It may be that the Altar Stone was horizontal in the final arrangement of stones at Stonehenge.

STAGE 5, c.1600 BC.  About nine hundred years after the sarsens had been raised, two, inaccurate, concentric circles of pits (the Y and Z Holes) were dug around the outside of the circle. They never received stones – they were left open, and gradually silted-up. This apparently half-hearted project, undertaken around 1600 BC, seems to mark the end of Stonehenge’s usefulness.[*]

Built to inspire awe, Stonehenge was clearly a sacred place where rituals took place. Beyond that generalization, however, its purpose and meaning have long been the subject of speculation. The SRP has found evidence supporting Mike Parker Pearson’s theory that Stonehenge was but one component of a larger ritual complex. Durrington Walls, Britain’s largest henge (around 480 metres in diameter), and its smaller (around 85 metres in diameter) companion, Woodhenge, lie in a little valley less than two miles to the north-east of Stonehenge. In 2005, a metalled roadway connecting the Southern Circle of Durrington Walls – a 40 metre diameter timber monument within the henge – to the river Avon was discovered.


Durrington Walls has been severely damaged by ploughing, and is crossed by the A345 road. Three entrances have been detected: at the north-west, at the south (facing Woodhenge) and at the south-east.
In 1967, ahead of improvements to the A345, excavations found the remains of two timber-post circles. The smaller, two ringed, Northern Circle and, adjacent to the south-eastern entrance, the 40 metre diameter, six ringed, Southern Circle.
Recent investigations have indicated that the henge bank of Durrington Walls was built over pits that had previously held a circuit of large wooden posts. See: ‘Rethinking Durrington Walls: a long-lost monument revealed’, article in Current Archaeology Issue 320 (November 2016), freely available online.
Woodhenge was discovered in 1925, by one Squadron Leader Insall (who spotted it from his aeroplane). Its bank and ditch surrounded an arrangement, 44 metres in diameter, of six, concentric, timber-post rings. (The timbers were later replaced by a smaller, rectilinear, setting of stones). The remains of a child with a cleft skull were found near the centre of the henge. At the time of the discovery (1926), it was thought the child (probably a girl about three years old) may well have been a dedicatory sacrifice, but the bones were destroyed during the Blitz, so a modern reassessment is not possible. Woodhenge had one entrance, at the north-east.
Google Satellite View of Durrington Walls and Woodhenge (opens in new window/tab).

Whereas Stonehenge and the Avenue were aligned on the axis of the midsummer sunrise and midwinter sunset, the Southern Circle and roadway (Durrington Avenue) were aligned on the axis of the midwinter sunrise and midsummer sunset. The timber structure of the Southern Circle was linked, via Durrington Avenue, the river Avon and Stonehenge Avenue, to the stone structure of Stonehenge. Professor Parker Pearson believes that the, short-lived, timber structure represented the land of the living, whilst the permanent stones of Stonehenge represented the realm of the ancestors. Symbolic journeys between to the worlds of the living and the dead could be made via the monuments’ ceremonial Avenues and the river.

Nobody ever lived at Stonehenge, but at Durrington Walls the remains of houses that were in use at the time the central sarsen settings were erected at Stonehenge, and that pre-date the construction of the henge of Durrington Walls, have been found.

The Durrington Avenue and Southern Circle were the main components of a ceremonial complex, later encircled by the henge bank and ditch, which formed the focus of a very large settlement whose house floors have been well preserved beneath colluvium.[*]… The discovery of houses within, underneath and outside Durrington Walls suggests that a large area of the valley in which the henge lies was probably covered in dwellings… It is likely that these dwellings were lived in by the builders of the Southern Circle and Stonehenge, and latterly the Durrington Walls and Woodhenge henges.
Stonehenge Riverside Project, Summary Interim Report on the 2006 Season

Furthermore, the nature of the debris associated with the settlement, indicates that great midwinter feasts took place there.[*]

Professor Parker Pearson is persuasive – he may well be right – but, unfortunately, we can never know with certainty. As Samuel Pepys, having found the stones of Stonehenge “as prodigious as any tales I ever heard of them”, confided to his diary on the 11th of June 1668: “God knows what their use was!


This old monument does as much exceed in greatness the so renowned Stoneheng [sic], as a Cathedral doeth a parish Church …
John Aubrey (1626–97), Monumenta Britannica[*]

Avebury is about 17 miles north of Stonehenge. Even today its earthwork, enclosing an area roughly 350 metres in diameter, is impressive. [Plan]  When built the bank would have been something like 27 metres wide at the base and 6 metres high. The ditch was 9 metres deep and some 14 metres wide at the top, tapering to less than half that at the bottom.[*] There were entrance causeways at the north, south, east and west. These breaks in the earthwork were, millennia later, adapted to carry the roads serving the village of Avebury, which inhabits the henge.


The henge is now known as Avebury because that is the name of the village, but this wasn’t always the case. It used to be called Waledich – probably from the Anglo-Saxon weala-dic, meaning ‘ditch of the Britons’. According to Aubrey Burl* it was still being called Wall-dyke in the 19th century.
* Prehistoric Avebury Second Edition (2002), Chapter 2 (p.27).
Though the village of Avebury has Anglo-Saxon roots, its spread into the henge itself is relatively recent. Andrew Reynolds, in Archaeological Research Agenda for the Avebury World Heritage Site, §2.8 (February 2001), writes:
From the evidence available, it can be argued that early medieval settlement began immediately to the southwest of the henge monument, probably during the 6th century, and most likely comprised a single farmstead. By the early 9th century the settlement had moved northwards and eastwards, up to the west entrance of the henge itself… With the exception of property boundaries, settlement lay largely without the henge until the post-medieval period …
Freely available online.
Google Satellite View of Avebury (opens in new window/tab).
The Swindon Stone, on the right, alongside the road at the northern entrance to Avebury.
The Barber Stone is foreground left. In the middle distance are stones of the Southern Inner Circle. The pair of stones furthest right mark the southern entrance.

Some of the massive (in the region of 60 ton) stones that acted as portals can still be seen. Both remain at the southern entrance. One, the Swindon Stone, stands at the northern entrance. According to, antiquarian, William Stukeley (1687–1765), the Swindon Stone’s partner (a stone “of a most enormous bulk”) fell down, and broke, in 1722: “we saw three wooden wedges driven into it, in order to break it in pieces.”  One fallen stone marks the eastern entrance. Nothing remains at the western entrance. The portal stones are components of a ring, originally comprising about a hundred standing stones, set around the perimeter of the enclosed area. Thirty stones of this Outer Circle can still be seen, of which three are fallen. There were two inner stone circles, each roughly 100 metres in diameter. All the stones at Avebury are sarsens – found locally, on the Marlborough Downs, a couple of miles to the east. They have been left in their natural state – not smoothed or worked to shape, as happened at Stonehenge. It seems, though, that, in the circles, the stones were set so that their, naturally, smoothest side faced inwards.[*] The Southern Inner Circle comprised about thirty stones, set around a central Obelisk. Today, five stones from this circle stand in an impressive arc. The Obelisk is no more. It was, though, sketched by William Stukeley in 1723.[*] To the west of the Obelisk (its position is now marked by a concrete plinth), but still within the Southern Inner Circle, can be seen six smallish stones, standing in a line running roughly north-south – remnants of the so-called Z Feature.[*] Outside the Southern Inner Circle, to the south, stood the Ring Stone. All that remains is a stump, but according to William Stukeley the stone (“not of great bulk”) had “a hole wrought in it”.  Only four stones from the Northern Inner Circle can still be seen, two of which are fallen. It is possible that this circle wasn’t actually a circle at all. It may have been a horseshoe – open on the south. Either way, at its centre was the Cove – three stones set in a U-shaped arrangement – which opened towards the north-east (in the general direction of midsummer sunrise).[*] Two of the three stones still exist.[*] In 2003, ahead of engineering work to prevent them collapsing, a limited excavation of the Cove stones’ footings was undertaken. The mighty back-stone, which projects almost 4½ metres above ground, was found to extend more than 2½ metres below ground (the base was not reached) – it could weigh 100 tons – and OSL dating indicated that it had been erected in 3120±350 BC.

Avebury’s chronology is not clear. Even if the Cove’s back-stone was erected c.3000 BC, it does not necessarily follow that the Cove as a whole was built at that early date. Radiocarbon assays of antler-picks tend to place the massive henge earthwork in the period 2600 BC–2500 BC.[*] The outer stone circle probably followed soon afterwards. The inner circles could predate the earthwork. Two avenues of standing stones – approaching the western (Beckhampton Avenue) and the southern (West Kennet Avenue) entrances of the henge – are evidently broadly contemporary with the earthwork.[*]

Although the Avebury sarsens were not worked to shape, they were selected for their natural shape. This is particularly apparent from the stones which can be seen on West Kennet Avenue (most of which were re-erected by Keiller). Sarsens naturally occur in two forms: columnar and broader, lozenge-type, shapes. On West Kennet Avenue these shapes alternate along the length of the rows, and, also, opposing stones are of contrasting shape. There seems to be general agreement that columnar stones are symbolically male and lozenges female.

Avebury’s stones seem to have remained unmolested until the early-14th century. Then, perhaps because they were offending Christian sensibilities, many were tipped into pits and buried. In 1934, Avebury was bought by a rich archaeologist, Alexander Keiller (of the ‘Dundee Marmalade’ family). His intention was to restore the site to something like its former glory. He cleared away trees, undergrowth and rubbish. He removed fences, old barns, and derelict cottages. Indeed, he wanted to rid the henge of all buildings, and used to have houses demolished as they became vacant. Keiller’s investigations along the western half of the Outer Circle brought to light several buried stones. He re-erected them in their original positions. Where he found a stonehole, but the stone itself was missing, he had the position marked with a concrete plinth. Trapped by one of the buried stones was a human skeleton. It was dated, by coins, to the early-14th century, and was identified as the remains of a barber-surgeon, by virtue of a pair of scissors and a medical-looking probe or lance. It was thought he had been accidentally crushed whilst assisting with the burial of the stone, but a more recent examination of the skeleton seems to indicate that this was not the case.[*] Maybe he was pinned to the ground and died of suffocation, or, rather more sinisterly, maybe he had been murdered and his body buried with the stone.[*] At any rate, the re-erected stone is known as the Barber Stone. Keiller had begun excavating the Southern Inner Circle when his work was interrupted, by the outbreak of the Second World War, in 1939. The National Trust bought Avebury in 1942. When the war ended, Keiller’s poor health prevented him from resuming his excavations, though he continued to live in Avebury Manor until his death, in 1955.

Before the war brought his investigations to a premature end, Alexander Keiller had discovered, in the western half of the Southern Inner Circle, the buried remains of what was called the Z Feature. The stones of this feature (Keiller identified twelve, six of which were intact and were re-erected) had been arranged in a line, running roughly north-south, about thirty metres long, at each end of which the setting turned eastwards.[*] The eastern side of the Southern Inner Circle has not been excavated, but a geophysical survey carried out in 2017 places the Z Feature at the western end of a square setting of stones, with the Obelisk at its centre. Further, Keiller interpreted some linear cuttings in the chalk adjacent to the Obelisk as medieval – the traces of a shed built against the fallen Obelisk. The archaeological team responsible for the 2017 geophysical survey (the Living with Monuments Project), however, believe there is enough evidence to propose that there was once an “early Neolithic house-like structure” at the heart of the Southern Inner Circle.


Dom Barker, Josh Pollard, Kris Strutt, Mark Gillings, Jeremy Taylor, ‘Geophysical Survey across part of the Southern Inner Circle of the Avebury henge (April 2017)’:
This report describes the results of fieldwork carried out as one component of the AHRC-funded Living with Monuments Project that seeks to investigate the earliest structural activity at Avebury through a combination of detailed archival work (both antiquarian; earlier survey; unpublished 1930s excavation records) and targeted geophysical prospection…
… there are grounds for believing that a series of chalk-dug linear features encountered during 1939, and interpreted by Keiller as an unconventional, open-sided medieval ‘lean-to’ constructed against the bulk of the fallen Obelisk megalith (a 7m long sarsen) are in fact of early Neolithic date. This is based on: the absence of medieval pottery from these features; the presence here of a concentration of early Neolithic material; their morphology, where in plan the closest match is with a growing body of excavated early Neolithic houses and mortuary enclosures. Furthermore, the axis of the structure is shared by that of the ‘Z-feature’ stone-holes (stones i-xxii) of the Southern Inner Circle. It is hypothesized that the latter represent a later lithic elaboration of this ‘foundational’ structure.
(This paper is freely available online.)
Above, computer model of the Southern Inner Circle released by the Project in 2017. (The Ring Stone is the outlier, bottom-left.)

Surveys, both ancient and modern, have located a number of buried stones in the eastern half of the Outer Circle. Indeed, there exists an almost continuous arc of fifteen stones in the south-eastern quadrant (thirteen buried, two fallen, and one is missing), but there are no current plans to follow Keiller’s example and re-erect them.

The stones which survived, out-of-sight, underground are the lucky ones. John Aubrey (d.1697) says he had heard from a local parson:

… that these mighty stones, as hard as marble, may be broken in what part of them you please without any great trouble: sc. Make a fire on that line of the stone, where you would have it crack; and, after the Stone is well heated, draw over a line with cold water, & immediately give a knock with a Smyth’s sledge, and it will break like the Collets at the Glass house.
Monumenta Britannica

By the time William Stukeley visited Avebury (1719–24), the work of stone-breaking was in full swing.

Just before I visited this place, to endeavour at preserving the memory of it, the inhabitants were fallen into the custom of demolishing the stones, chiefly out of covetousness of the little area of ground, each stood on… The custom of thus destroying them is so late, that I could easily trace the obit of every stone; who did it, for what purpose, and when, and by what method, what house or wall was built out of it, and the like.[*]
Abury, Temple of the British Druids, with some others, Described (1743),[*] Chapter 3 (pp.15–16)

West Kennet Avenue connected Avebury to a stone circle, almost 2 miles to the south, whimsically called the Sanctuary. William Stukeley witnessed the destruction of this latter monument in 1723–4.


This Overton-hill from time immemorial, the country-people have a high notion of. It was (alas, it was!) a very few years ago crown’d with a most beautiful temple of the Druids. They still call it the sanctuary… It had suffer’d a good deal when I took that prospect of it, with great fidelity, anno 1723, which I give the reader in plate XXI [below].
Abury, Temple of the British Druids, with some others, Described (1743), Chapter 7 (p.31)
By the end of 1724, all the stones had been removed, and the land ploughed over. The site was found, using Stukeley’s work as a guide, in 1930. When it was excavated, not only were there the holes from the two concentric stone circles sketched by Stukeley (the diameter of the outer circle is about 40 metres), but there were also holes for six concentric rings of timber posts. Interpreting the building sequence of this monument continues to keep archaeologists entertained. It has been suggested that the postholes are the remains of roofed buildings (the same has been said of Woodhenge’s posthole rings), but it is probably fair to say that most archaeologists today are inclined to the view that they held free-standing posts. Possibly the posts had timber lintels, or perhaps they were decorated in the fashion of totem-poles.

John Aubrey recorded West Kennet Avenue on the plan of Avebury he made in 1663, but that was the only avenue of standing stones he recorded.[*] It would seem that Beckhampton Avenue, which connected to Avebury’s western entrance, had been so thoroughly erased that Aubrey failed to notice its faint traces. William Stukeley, though, did; or at least he claimed he did. Stukeley (who was ordained in 1729, becoming a Church of England vicar) became convinced that the religion of the Druids (who had built Avebury) was the antecedent of the doctrine of the Church of England.


In the Preface to Stonehenge. A temple restor’d to the British Druids (1740), Stukeley writes:
My intent is (besides preserving the memory of these extraordinary monuments, so much to the honour of our country, now in great danger of ruin) to promote, as much as I am able, the knowledge and practice of ancient and true Religion; to revive in the minds of the learned the spirit of Christianity, nearly as old as the Creation, which is now languishing among us; to restore the first and great Idea of the Deity, who has carry’d on the same regular and golden chain of Religion from the beginning to this day; to warm our hearts into that true sense of Religion, which keeps the medium between ignorant superstition and learned free-thinking, between slovenly fanaticism and popish pageantry, between enthusiasm and the rational worship of God, which is no where upon earth done, in my judgment, better than in the Church of England. And seeing a spirit of Scepticism has of late become so fashionable and audacious as to strike at the fundamentals of all revelation, I have endeavoured to trace it back to the fountain of Divinity, whence it flows; and shew that Religion is one system as old as the world, and that is the Christian Religion; that God did not leave the rational part of his creation, like the colony of an ant-hill, with no other guide than instinct, but proportion’d his discoveries to the age of the world, to the learning, wisdom, and experience of it; as a wise parent does now to his children. I shall shew likewise, that our predecessors, the Druids of Britain, tho’ left in the extremest west to the improvement of their own thoughts, yet advanc’d their inquiries, under all disadvantages, to such heights, as should make our moderns asham’d, to wink in the sunshine of learning and religion. And we may with reason conclude, there was somewhat very extraordinary in those principles, which prompted them to such a noble spirit as produced these works, still visible with us, which for grandeur, simplicity and antiquity, exceed any of the European wonders.

He developed the idea that:

The whole temple of Abury may be considered as a picture, and it really is so… When I frequented this place, as I did for some years together, to take an exact account of it, staying a fortnight at a time, I found out the entire work by degrees. The second time I was here, an avenue was a new amusement. The third year another. So that at length I discover’d the mystery of it, properly speaking; which was, that the whole figure represented a snake transmitted thro’ a circle; this is an hieroglyphic or symbol of highest note and antiquity.
Abury, a Temple of the British Druids, with some others, Described, Chapter 4 (p.18)
William Stukeley’s:
‘A Scenographic view of the Druid temple of Abury in north Wiltshire, as in its original.’
William Stukeley’s: ‘A Scenographic view of the Druid temple of Abury in north Wiltshire, as in its original.’

The Sanctuary, at the end of West Kennet Avenue, formed the serpent’s head, and Beckhampton Avenue formed the tail. Beckhampton Avenue, however, was not clearly visible on the ground, and, since Stukeley did tamper with his earlier findings to make them better fit his theory, it is no wonder that its very existence was doubted.


Stukeley’s published plan of the Sanctuary. The caption reads: ‘The HAKPEN or snakes head temple on Overton hill, called the Sanctuary.’
Stuart Piggott:
… there is evidence of one grave and deliberate falsification which can be traced by means of the assembled Stukeley papers. His published plan of the stone circles at The Sanctuary … shows them as ovals, not true circles. But the 1930 excavations of the site showed them to be in fact accurately circular, and we find that Stukeley’s original field-surveys of the site, which survive, show true circles and minor features completely in accordance with the facts as recovered by excavation. An intermediate link is an original drawing of c.1740 in which the positions of the stones are arranged as ovals in grey wash, but are superimposed on a faint pencil outline showing them as circles, while in one of the original plans an oval outline has been roughly sketched in. It appears that by 1740 the mystic serpent had to be given a more naturalistic head than that afforded by true circles, and the original survey was altered accordingly.
William Stukeley: an Eighteenth-Century Antiquary (1950) Chapter 4 (p.129)

Stukeley was vindicated, to some extent anyway, when traces of Beckhampton Avenue were found by archaeologists working around the Longstones in 1999, 2000 and 2003. The Longstones are, today, two standing stones, called Adam and Eve, halfway down Stukeley’s plan of Beckhampton Avenue. However, whilst Stukeley continued Beckhampton Avenue southwards, making it about the same length as West Kennet Avenue, no archaeological evidence was found to indicate that it extended beyond the Longstones.


The stone known as Eve is evidently the last surviving upright stone of Beckhampton Avenue. Adam – a massive sarsen slab, estimated to weigh 62 tons, which was re-erected in 1912, having toppled the previous year – is the last surviving stone of the so-called Longstone Cove. John Aubrey described this feature as: “three huge upright stones … like those at Aubury [i.e. the Cove at the centre of the Northern Inner Circle] … they are called the Devill’s Coytes.”  William Stukeley writes:
This Longstone cove, vulgarly call’d long stones, is properly a cove, as the old Britons call’d ’em, compos’d of three stones, like that most magnificent one we described, in the center of the northern temple at Abury; behind the inn… Mr. Aubry in his manuscript observations publish’d with Mr. Camden’s Britannia, speaks of them by the name Devil’s coits. Three huge stones then standing. It was really a grand and noble work. The stone left standing [i.e. Adam], is 16 feet high, as many broad 3½ thick. The back stone is fallen flat on the ground, of like dimension… The other was carried off by that destroyer Richard Fowler, together with many more, but seven years ago (when I was there).
Abury, a Temple of the British Druids, with some others, Described, Chapter 8 (p.35)
The 1999–2003 excavations revealed that the assumed three-sided cove had actually been a four-sided rectilinear setting of massive sarsens. Anyway, William Stukeley reckoned that Beckhampton Avenue carried-on past Longstone Cove:
The avenue continu’d its journey by the corn fields… Then it descends by the road to Cherill, ’till it comes to the Bath-road, close by the Roman-road, and there in the low valley it terminates, near a fine group of barrows … This point facing that group of barrows and looking up the hill is a most solemn and awful place … Here I am sufficiently satisfied this avenue terminated, at the like distance from Abury-town, as Overton-hill [the Sanctuary] was, in the former avenue … I am equally satisfied there was no temple or circle of stones at this end of it … The head of the snake was aptly represented by that double circle on Overton-hill … I apprehend this end of the avenue drew narrower in imitation of the tail of a snake, and that one stone stood in the middle of the end, by way of a close.
Abury, a Temple of the British Druids, with some others, Described, Chapter 8 (p.36)
The 1999–2003 excavations indicated that Longstone Cove (though it was actually a four-sided monument) marked the end of Beckhampton Avenue.

In his diary entry for the 15th of June 1668, Samuel Pepys wrote:

In the afternoon come to Abebury, where, seeing great stones like those of Stonage [Stonehenge] standing up, I stopped, and took a countryman of that town, and he carried me and shewed me a place trenched in, like Old Sarum almost, with great stones pitched in it, some bigger than those at Stonage in figure, to my great admiration: and he told me that most people of learning, coming by, do come and view them, and that the King did so: and that the Mount cast hard by is called Selbury [Silbury Hill], from one King Seall buried there, as tradition says.[*] I did give this man 1s [one shilling].
Standing over 30 metres high, with a diameter of about 160 metres, Silbury Hill is widely claimed to be the largest man-made prehistoric mound in Europe.

There have been three major excavations of Silbury Hill, which is less than a mile to the south of Avebury, since Pepys’ time. In 1776 a shaft was driven vertically through the mound; in 1848 a horizontal tunnel was dug to its centre; and between 1968 and 1970 further excavations, along the line of the 1848 tunnel, were carried out. Despite the local tradition mentioned by Pepys, no burial has been found. In 2000, the backfilling of the 1776 shaft collapsed, leaving a large hole in the Hill’s summit. Further investigation showed that there were also voids in the other diggings’ backfill, and that there was a danger of the whole monument collapsing. Essential works to stabilize the Hill were carried out in 2007–8, giving archaeologists access to its interior. New datable material was found, and, additionally, archive material was reassessed.[*] A, perhaps surprising, picture of Silbury Hill’s chronology has emerged. It began in about 2450 BC, as an unimposing gravel mound, less than a metre high and 9–10 metres in diameter. The site then, seemingly, underwent virtually continuous development, evolving through many stages, for between 55 and 155 years, at which point work stopped, leaving the Silbury Hill we know today. (Well not quite – it appears that, in the early-11th century AD, a spiral pathway was dug into the hillside, the top was levelled-off, and a timber fortification was erected on the summit.[*]) The monument’s start-date and its relatively brief history suggest it was the swansong of the local pre-Beaker population – it had no meaning in ‘Beaker culture’, so it was simply abandoned.[*]

 © Anthony Johnson 2008
Henry of Huntingdon’s is the earliest extant description of Stonehenge, and the earliest record of the name. In a footnote to her edition/translation of the Historia Anglorum (published 1996), Diana Greenway comments: “Possibly the presence of fallen stones at the site led Henry, or his source [Henry may not have seen Stonehenge himself], to believe that originally there had been a second storey of lintelled circles.”
Incidentally, Henry’s first wonder: “is the wind which issues with such force from the caves in the mountain which is named ‘The Peak’, that it drives back any pieces of clothing thrown in and tosses them up to a great height... The third is at Cheddar Gorge, where there is an underground cavern which many people have often entered, but although they have travelled a long way over dry land and over rivers, they have never been able to come out at the other end. The fourth is that in certain places the rain seems to rise up from the mountains and immediately fall on the plains.”
Deer antlers, used as picks or rakes, that were left at the bottom of the ditch have provided radiocarbon dates. Bayesian modelling of Stonehenge’s radiocarbon dates suggests the ditch and bank earthwork was constructed between 3000 BC and 2920 BC.
See: ‘Stonehenge remodelled’ (Antiquity Vol. 86, Issue 334, December 2012), freely available online, and Mike Parker Pearson, Stonehenge: Exploring the Greatest Stone Age Mystery (2012).
There is a complex of postholes – pits that once held timber posts – within the henge, between the centre and the southern entrance, and also at the main entrance. Only one posthole has provided a datable sample (a piece of pig-rib), and that produced a date around 2500 BC (2580–2460 BC), which is, of course, centuries after Stonehenge’s beginnings, and is actually about the same time that construction of the surviving stone settings started. However, where excavations have revealed overlaps of postholes and stoneholes, the postholes have always been the earlier feature. There are hundreds of postholes, and it seems reasonable to suppose that they span the history of the site – perhaps, in some cases, predating the henge itself.
Aubrey Holes are named after the antiquarian John Aubrey, who, in 1666, appears to have detected five of them. The plan below shows the ditch (in yellow), bank and Aubrey Holes.
The Stonehenge Riverside Project was a collaborative effort by five university teams, under the direction of Mike Parker Pearson (University of Sheffield), with co-directors Josh Pollard (University of Bristol), Julian Thomas (University of Manchester), Kate Welham (University of Bournemouth), Colin Richards (University of Manchester) and Chris Tilley (University College London). Between 2003 and 2009 the SRP undertook extensive fieldwork in Stonehenge’s environs, and carried out a reappraisal of material produced by earlier archaeologists.
Non-destructive chemical analysis of the fifty-two surviving sarsens (of around eighty originally) at Stonehenge revealed that fifty of them shared a similar geochemistry – two have distinctly different chemistry, both from each other and from the rest. A core sample from one of the fifty, which had been taken when the stone (Stone 58) was re-erected in 1958, became available in 2018 (it had been thought lost), and was subjected to more rigorous, destructive, testing. Comparison of the results with samples of sarsens occurring between Norfolk, in the east, and Devon in the west, demonstrated that Stone 58, and hence the rest of the fifty, was “most likely sourced from the vicinity of West Woods.”
See: ‘Origins of the sarsen megaliths at Stonehenge’, Science Advances Vol. 6 Issue 31 (31st July 2020), freely available online.
The, so-called, ‘bluestones’ are a heterogeneous collection of stones – mostly dolerite, but also rhyolite, volcanic and calcareous ash, and sandstone.
‘The original Stonehenge? A dismantled stone circle in the Preseli Hills of west Wales’, freely available online.
Trilithon: three stones – two uprights topped by a lintel. Three of the five trilithons are standing today.
Trilithon 4, pictured above, was re-erected in 1958, having collapsed in 1797. In 2011 Stonehenge was surveyed using laser scanning equipment. The precise mapping of the stones’ exposed surfaces enabled the weight of the above-ground portions to be estimated. The left upright above (Stone 58) is estimated at 24.04 tons, the right upright (Stone 57) 27.93 tons, and the lintel (Stone 158) 16.56 tons.
The lintels were secured atop the uprights by ‘mortice and tenon’ joints.
Left: tenon on top of an upright.
Right: mortice in a fallen lintel.
The lintels in the circle were also locked together by a tongue and groove joint. Presumably, such sophisticated joinery was commonplace in timber structures.
Of the thirty uprights in the circle, seventeen are now standing, and six lintels are in place.
The prone upright would have been moved forward on rollers and tipped into its hole. Some holes appear to have been given a sloping entry-ramp to ease this process. The projecting end would then be levered up, and the stone hauled vertical with ropes (probably using an A-Frame to increase leverage). The cavity was then packed with rubble.
There is no evidence to show how the lintels were raised. If an earthen ramp had been used, traces would have been found. One popular idea is that the lintels were placed on a log platform next to the uprights. The stone was then levered up, and other logs pushed (at right angles to the previous logs) under it. This was repeated, until the stone had been raised to a sufficient hight that it could be pushed across and located on the uprights. As with every other aspect of the operation, easier said than done!
The, now ruined, Great Trilithon (Trilithon 3), at the apex of the horseshoe, once stood over 7 metres tall – more than 2 metres higher than the sarsen circle.
Nick Pearce, Richard Bevins, Rob Ixer: ‘Portable XRF investigation of Stonehenge Stone 62 and potential source dolerite outcrops in the Mynydd Preseli, west Wales’, Journal of Archaeological Science: Reports Vol. 44 (August 2022).
Richard Bevins, Nick Pearce, Mike Parker Pearson, Rob Ixer: ‘Identification of the source of dolerites used at the Waun Mawn stone circle in the Mynydd Preseli, west Wales and implications for the proposed link with Stonehenge’, Journal of Archaeological Science: Reports Vol. 45 (October 2022).
The bluestones are much smaller than the sarsens, as the photo below demonstrates.
The bluestone on the left (Stone 63) has an estimated above-ground weight of 1.2 tons. The sarsen on the right (Stone 53, Trilithon 2) has an estimated above-ground weight of 18.14 tons.
The two rejected samples were found in a pit that Richard Atkinson believed was the ramp used to erect one of the uprights (Stone 56) of the Great Trilithon. The SRP have, however, convincingly shown that, whatever its purpose actually was (that remains uncertain), this large pit (it has never been excavated to its fullest extent) was not used as an entry-ramp, but it was dug in 2440–2100 BC, i.e. after the trilithons had been raised (2620–2480 BC). Indeed, the pit is so deep and in such close proximity to the Great Trilithon, that it may have, eventually, been the cause of the latter’s collapse.
Six radiocarbon dates were derived from samples found in one Q Hole. They covered the swathe of time from 3370–3090 BC to AD 1480–1640. The samples had suffered from a process called ‘bioturbation’, meaning they had been moved from their proper position in the layers of soil by the action of worms and other burrowing creatures, so no reliable figure for the digging of the Q Hole could be arrived at. In fact, only one of the dates (2460–2200 BC) was in the right ballpark and, in the immediate aftermath of the excavation, this date, ‘around 2300 BC’, received much media coverage – see: ‘Stonehenge birthdate discovered by archaeologists’, article in The Telegraph, freely available online. Darvill and Wainwright later collaborated with Mike Parker Pearson to produce the SRP chronology for Stonehenge.
The first version of Stonehenge lasted from the digging of the ditch and bank in 3000–2920 BC, to the arrival of the sarsens, in 2620–2480 BC. There are no radiocarbon dates available for the newly discovered stone circle, but it can be loosely dated from two flint chisel arrowheads, recovered from the packing of two stoneholes, to around 3400–2600 BC.
There is no absolute evidence that the stones which stood at Bluestonehenge were actually bluestones. The shape and size of the stoneholes rule out the possibility that they were sarsen slabs. The identification of them as bluestones is based on the imprints that some of them have left behind.
The removal of the stones from Bluestonehenge is dated, by an antler pick discarded at the bottom of an emptied stonehole, to 2460–2270 BC. This is corroborated by the date obtained from another pick, 2469–2286 BC, which had also been left on site after the stones’ removal.
The bluestones are mostly dolerite, but also rhyolite, volcanic and calcareous ash, and sandstone. The source of all of them has not been ascertained, but since the early-1920s* it has been recognized that the majority originated in Preseli.
* H.H. Thomas, ‘The Source of the Stones of Stonehenge’ (Antiquaries Journal Vol. 3, 1923).
See: ‘Message in the Stones’, article in Current Archaeology Issue 212 (November 2007), freely available online.
See Beakers and Bronze.
The SRP dates the re-cut ditch (which has a shallower and wider profile than the original) 2560–2140 BC (obtained from an animal bone). See: ‘Who was buried at Stonehenge?’ (Antiquity Vol. 83, Issue 319, March 2009), freely available online.
There are datable samples associated with these later bluestone settings: one of antler and one of animal bone from the new circle, and one of antler from the oval. Additionally, the orbits of the Q and R Holes and the possible small bluestone circle were cut by a mysterious large pit (previously thought to have been an erection-ramp for an upright of the Great Trilithon) that the SRP place in the period 2470–2210 BC. A stone of the oval was subsequently set into the filling of this pit. As a result, the estimated dates of construction are: 2270–2020 BC for the large bluestone circle, and 2210–1930 BC for the bluestone oval.
The Altar Stone (Stone 80) is classed as a bluestone, but it apparently did not originate in the Preseli region. Its origins may lie some 50 miles east of Preseli, in the Brecon Beacons.
See: ‘Craig Rhos-y-felin: a Welsh bluestone megalith quarry for Stonehenge’ (Antiquity Vol. 89, Issue 348, December 2015), freely available online.
The Z Holes are roughly 3½ metres outside the sarsen circle, the Y Holes roughly 11 metres. Seemingly, the intention was to have thirty pits in each circuit, matching the sarsens, but the work was evidently abandoned – the last Z Hole was never dug. One antler sample from Z Hole 29 and three from Y Hole 30 (both of these holes are in the north-east of their circuits) have provided radiocarbon dates. Rather than simply being abandoned tools, these antlers would seem to be deliberately deposited offerings. The constructors of the henge’s ditch, back in c.3000 BC, had deposited animal bones, but not just any animal bones – they were venerable relics, having been collected many years before the henge’s construction. Similarly, one of the antler samples from Y Hole 30 and the sample from Z Hole 29 are significantly older than the other two from Y Hole 30. Taking this into account, the proposed date for the digging of both the Y Holes and the Z Holes is 1680–1520 BC.
The houses are squarish and have floorplans similar to the famous stone built houses at Skara Brae, in the Orkneys. At Durrington Walls, where the houses would have been, predominantly, made from timber, none of the fixtures have survived, but there are slots in the chalk floor where the wooden furniture was located.
John Aubrey’s interest in Avebury began in January 1649, when he came across the monument, by chance, whilst out hunting. In 1663 Aubrey became a Fellow of the Royal Society (founded 1660). His comparison of Avebury, a cathedral, with Stonehenge, a parish church, came to the attention of the Society’s patron, King Charles II, and resulted in Aubrey giving the king a guided tour of Avebury: “I shewed him that stupendious Antiquity, with the view whereof, He and his Royal Highness, the Duke of Yorke, were very well pleased.”  Afterwards, in September 1663, Aubrey carried out a plane-table survey of Avebury.
John Aubrey is best known today for a collection of biographical titbits known as Brief Lives, but, as his namesake Aubrey Burl* points out: “he was, without question, the first great English fieldworker and archaeologist.”  John Aubrey’s archaeological notes, which he had drawn together to make a book, Monumenta Britannica, were only fully published in 1980 (Vol. 1) and 1982 (Vol. 2).
* Prehistoric Avebury Second Edition (2002), Chapter 2 (p.38).
Harold St George Gray directed excavations at Avebury in 1908, 09, 11, 14 and 22. This photograph, showing his workers, vividly illustrates the ditch’s original depth.
Sarsens, in their natural state, are found scattered on the ground surface. Indeed, they were also known as Grey Wethers, because of their resemblance to a grazing flock of sheep. The tops of the stones are rougher than the undersides due to weathering.
Stukeley wrote that the, by-then prostrate, Obelisk was: “of a vast bulk, 21 feet long, and 8 feet 9 inches diameter; when standing, higher than the rest.”
The five remaining stones of the Southern Inner Circle, with the smaller stones of the Z Feature behind. Concrete plinths mark empty stoneholes.
William Stukeley’s sketch of the Cove. He says that the missing, northern, wing fell down in 1713 (no doubt it was broken up for building material): “They told me it was full seven yards long, of the same shape as its opposite, tall and narrow.”  Incidentally, Stukeley, mistakenly, though that, apart from their different central features, both the inner circles (“temples”) were identical – each having an outer ring of 30 stones and a concentric inner of 12.
The two remaining stones of the Cove, photographed in June 2007.
Optically Stimulated Luminescence dating: a technique in which the length of time that has elapsed since a mineral was exposed to daylight is determined.
It may be that the massive earthwork, the remains of which we see today, replaced an earlier, much smaller, ditch and bank. Excavations have revealed the profile of a smaller bank buried under the final bank, but Aubrey Burl* argues that this was merely a guideline for the main earthwork.
* Prehistoric Avebury Second Edition (2002), Chapter 3 (pp.65, 81–2).
The skeleton was thought to have been destroyed in an air-raid during World War II. In 1999, however, it was found in the Natural History Museum. Examination of the skeleton suggested it was of a male of slight build in his late-20s, and revealed that he had a healed wound to his head – probably a sword cut. There was, though, no compelling evidence that the skeleton had been crushed under a stone weighing some 15 tons.
See: ‘Stones murder theory revealed’, article in Wiltshire Gazette and Herald (18th April 2002), freely available online.
A group of techniques (magnetometry, resistivity and ground-penetrating radar are commonly used in archaeology) utilised to detect and identify features below the ground surface.
Workmen probing with iron bars, directed by the Reverends A.C. Smith and W.C. Lukis, in 1881.
Geophysical survey carried out by the National Trust in 2003.
Stukeley names those he knew to be responsible for destroying the stones. In a reference to the use of fire to smash them, he writes: “One Tom Robinson the Herostratos of Abury, is particularly eminent for this kind of execution, and he very much glories in it.”
Stukeley mocks Tom Robinson in the above drawing.
Both John Aubrey and William Stukeley thought that monuments such as Stonehenge and Avebury were Druid temples.
John Aubrey’s plan of Avebury.
“From the south entrance runnes a solemne Walke, sc. of Stones pitch’d on end about seven foot high”.
Predecessor of modern Salisbury, built, by the Normans, within the defensive earthwork of an Iron Age hillfort.
John Aubrey:
No history gives any account of this hill; the tradition only is, that King Sil or Zel, as the countrey folke pronounce, was buried here on horseback, and that the hill was raysed while a posset of milke was seething.
Monumenta Britannica
In a paper, freely available online, entitled ‘The world recreated: redating Silbury Hill in its monumental landscape’ (Antiquity Vol. 81, Issue 311, March 2007), Alex Bayliss, Fachtna McAvoy and Alasdair Whittle, state:
The new programme for dating Silbury Hill was conceived from the outset within a Bayesian statistical framework. This allows the chronology of the monument to be formally estimated, using an explicit statistical methodology, from both the radiocarbon dates and the stratigraphic sequence revealed by archaeological excavation. This approach was introduced for the construction of archaeological chronologies more than a decade ago, and the impact of its routine application is beginning to become apparent.
Dated by pot sherds and a coin, of King Æthelred (the Unready), of c.1010. It doesn’t seem unreasonable to suppose that the Hill’s summit was fortified as a result of the Viking activity that characterizes Æthelred’s reign (see The Vikings Return), indeed, the Anglo-Saxon Chronicle records a battle in the locality (“at Kennet”), where English forces were defeated by Vikings in 1006.
See: ‘The many faces of Silbury Hill’, article in Current Archaeology Issue 293 (August 2014), freely available online.
According to Aubrey Burl:
The Beckhampton and Kennet avenues must have been built at the time when the bank and ditch were almost finished because several of the stones were supported by blocks of Lower Chalk from Avebury’s ditch.
Prehistoric Avebury Second Edition (2002), Chapter 8 (p.209)
Aubrey Burl* says Keiller found that the Z Feature stones: “had been supported by blocks of chalk from the ditch, just like the pillars of the Kennet avenue so that they also had to be contemporary with the construction of the earthwork.”  The same applies to the Ring Stone, avers Dr Burl, which, judging by the size of the pit it sits in, had evidently replaced a larger stone.
* Prehistoric Avebury Second Edition (2002),
Chapter 3 (pp.78–9), 7 (p.208), 9 (pp.240–1).
Mike Parker Pearson has summarized the findings and conclusions of the SRP in Stonehenge: Exploring the Greatest Stone Age Mystery (2012).
Aubrey Burl, A Brief History of Stonehenge (2007), Chapter 8 (p.207).
Rodney Castleden, The Making of Stonehenge (1993), Chapter 5 (p.130).
Mike Pitts, Hengeworld (2001), Chapter 15.
See: ‘The dead of Stonehenge’ (Antiquity Vol. 90, Issue 350, April 2016), freely available online.
See: ‘Stonehenge remodelled’ (Antiquity Vol. 86, Issue 334, December 2012), freely available online.
See: Stonehenge World Heritage Site Landscape Project: Stonehenge, Amesbury, Wiltshire, Archaeological Survey Report (2010), freely available online.
Aubrey Burl, A Brief History of Stonehenge (2007), Chapter 4 (pp.126–7).
‘Craig Rhos-y-felin: a Welsh bluestone megalith quarry for Stonehenge’, freely available online.
‘Observations on the supposed “Neolithic bluestone quarry” at Craig Rhosyfelin, Pembrokeshire’, freely available online.
‘Strontium isotope analysis on cremated human remains from Stonehenge support links with west Wales’, freely available online
See: ‘Feeding Stonehenge: cuisine and consumption at the Late Neolithic site of Durrington Walls’ (Antiquity Vol. 89, Issue 347, October 2015), freely available online.