Antiquity 1939 Volume 13 Pages 223-233

Antiquity 1939 Volume 13 Pages 223-233 is in Antiquity 1939 Volume 13 Issue 13.

Avebury [Map]. Summary of Excavations, 1937 and 1938 by Alexander Keiller (age 49).

Note 1. See Plate I, facing p. 232. My thanks are expressed to Major Allen for permitting the use of this air-photograph.


The reasons for selecting the northwest sector for the beginning of the excavations within the circles of Avebury, scheduled to last for at least a decade, by the Morven Institute of Archaeological Research (with the permission and cordial co-operation of H.M. Office of Works) was twofold. First, the conditions of indescribable squalor and neglect prevailing over most of the area of this section surpassed, if comparison is possible, those existing in any other part of the circle; indeed the tangle of rusty pig-wire, the accumulations, to a depth of nearly three feet, of old tins and broken bottles, around two of the standing stones, to say nothing of the refuse-heaps which filled part of the ditch almost flush with its edges, contributed ungenerously towards rendering the once majestic site of Avebury what it has been for centuries, the outstanding archaeological disgrace of Britain. Secondly, the presence of a veritable jungle of trees and undergrowth not only perpetually menaced the safety of this part of the monument, but had already done dama e to the preservation of the all-important features below ground-levef. Protests lodged in the recent past by those who had the welfare of the monument at heart had proved unavailing, and it was not until the eastern part of the sector passed by purchase into the writer's ownership that the work of preservation, which must, at least in the case of Avebury, always be regarded as of so much more importance than that even of excavation itself, could be begun. The scrub and brushwood were removed, as were privet bushes deliberately planted against the stones. The trees, almost without exception rotten to the core and actually growing on the monument, were cut down. The roots in the vicinity of the stones were laboriously grubbed up by hand, while those on the bank and in the ditch were blown up, gelignite being employed in place of any other explosive in order to avoid damage to the subsoil. That the work was undertaken none too soon, indeed in some cases unavoidably too late, was well exemplified when Stone 3, one of the four largest remaining stones in the Outer Circle, was excavated to its base, and it was found that tree roots had already forced it 1.75 feet out of the perpendicular, while lesser roots had penetrated into cracks in the sarsen and had split large blocks off the parent stone. If the excavations had borne no other result they would at least have confirmed what every archaeologist knows, that the presence of trees is wholly incompatible with the preservation of an ancient monument. Unhappily it may be regarded as a certainty that the northwest sector will not represent, during the excavations at Avebury, the only example of the truth of this dictum.

So far as the northwest sector was concerned no work of an archaeological nature had been recorded, save for a single unproductive cutting into the bank near the northeastern extremity by W. Cunnington and A. C. Smith1 in 1865. For evidence concerning the number of stones existing at various dates in historical times recourse must be had to the surveys of Smith and Lukis2 (1882), Crocker on behalf of Sir Richard Colt Hoare3 (1812), Stukeley4 (1724), and Aubrey5 (1663).

Note 1. Revd. A. C Smith, British and Roman Antiquities of the North Wiltshire Downs, 1885, P. 143

Note 2. ibid. p. 140, plate v.

Note 3. Sir Richard Colt Hoare, History of Ancient Wiltshire, 1821, vol. 11, plate XIII.

Note 4. Dr William Stukeley, Abury Described, 1743, Tab. I, frontispiece.

Note 5. John Aubrey, Monumenta Britannica (Bodleian Library, Oxford); plan made 1663; facsimile of plan in Wilts. Arch. Mag., 1862, VII, 224.

At the beginning of the excavations four stones1 (1, 3, 14 and 15) were still standing, while portions of four more (6, 7, 12 and 13) could be distinguished covered by a medieval field-boundary. This consisted of a low turf-covered dry stone wall connecting many of the stones, whether standing or fallen, and forming, as it transpired, the southern edge of a cart-track running westwards and then southwards from the main Swindon-Avebury road.

Note 1. The system of numeration of stones and stone-holes (or sockets within which stones previously stood) adopted by the Morven Institute in the northwest sector was to allot numbers consecutively in an anti-clockwise direction, Stone no. I being the standing megalith immediately to the west of the Swindon-Avebury road.

The presence of this field-boundary had precluded Smith from carrying out his investigations in 1881, which elsewhere in the Circle had consisted of probing, and in some cases of subsequent partial excavation undertaken in an endeavour to identify the existence both of buried stones and also the position of stone-holes. His survey, however, although showing the same standing stones and the same quantity of fallen ones, indicates the latter as representing nos. 5, 6, 11 and 12. The survey by Crocker, carried out for Sir Richard Colt Hoare, shows the above standing stones and six fallen (5, 6, 7, 11, 12 and 13). The plan by Stukeley as usual gives in this regard the fullest historical information. In his time ten stones were standing (I, 2, 3, 4, 8, 9, 10, 14, 15 and IS), while seven were fallen (5, 6, 7, 12, 13, 20 and 21) the last named being immediately to the north of the village street dividing the northwest and southwest sectors. At a point equidistant (96 feet) from Stones 12 and 14 Stukeley marked upon his plan within the area a fallen stone. Excavations carried out during 1937 proved conclusively that no stone had ever stood here or in the immediate vicinity. It is possible that Stukeley was misled in this regard either by the presence of a large natural field-sarsen introduced for building or other purposes, or by part of a destroyed megalith which had been abandoned at this spot. Aubrey marked seven stones, all of which can be identified with a certainty not always possible where his plan of Avebury is concerned. These represent Stones I, 2, 3,4, 14, 15 and 18.

During the excavations of 1937 six stones (5, 6, 7, 11, 12 and 13) were recovered from under the field-boundary, to which reference has been made above, and these were re-erected in their original stone-holes. Stones 5 and 6 had been felled apparently at the time of the formation of the boundary bank and had been broken transversely. Between, and on either side of them, the wall was found to have been composed not of natural field-sarsens but of pieces of broken megaliths. In the present year (1939) a system, devised during the previous season and employed in the southwest sector, was utilized here, consisting of taking plaster casts of all fractured surfaces in an endeavour to reconstruct as much of the broken stones as possible. Many of the fragments in the wall were proved to belong to Stones 5 and 6, and certain of these could be, and were replaced; although other portions, including many which joined each other, certainly had formed part of one or other of these stones. In each case at least one course was missing, rendering complete reconstruction impossible. Stones 7, 11 and 12 had been damaged to a greater or less degree in order to clear the route for the cart-track. Stones 16 and 17 were discovered buried in pits of the unusual depth of 8 feet and 7.7 feet respectively. These likewise were re-erected in their original positions. The site of Stone 17 lay within a cow-byre, but on application being made to the owner, Mr J. Peak-Garland, of Manor Farm, he courteously agreed to the demolition of this structure in the interests of the monument. At this stage the writer desires to record his sincere gratitude to Mr Peak-Garland, not only in this regard but for his sympathetic interest in the work, no less than for his invaluable practical assistance. This included permission both to remove trees and to re-erect fences, and it is not too much to say that without his continued co-operation the excavations, at any rate of the southern part of the area available in the northwest sector, could not have been carried out. Finally, the stone-holes of Stones z,4,8,9 and 10, were identified, in each case with an adjacent burning-pit indicating all too clearly the fate which the stones had suffered since Stukeley’s day.

In all the stone-holes quantities of clay of riverine origin were found introduced as a packing medium, while a large number of packing stones had been utilized in the stone-holes which had not suffered from subsequent disturbance. Holes which had held small stakes to minimize the friction, which would otherwise have been occasioned by the pressure of the base of the stone upon the side of the stone-hole opposite to that from which it had been erected, were found in the majority of undamaged stone-holes, as well as other features in the chalk indicative of sockets for baulks and vertical posts for various purposes. The stone-hole of Stone 6 was curious in so far that it would appear to have been excavated to a greater depth than was subsequently found necessary, the stone having actually been erected upon a foundation of closely packed supporting stones, carefully laid upright upon their edges. Stones 1, 3, 14 and 15 were excavated to their bases. Stone 1 was found to have slipped in its socket seemingly during, or at any rate shortly after, erection, accounting for its present position, which had previously always puzzled the writer. The position of Stone 14, standing obliquely to the arc of the Circle, has in the past given rise to considerable conjecture, much of it of a fanciful and far-fetched nature. On examination, however, the explanation was found to be quite simple. Owing to the breaking of a horizontal supporting stake, 6 inches in diameter, the stone had imbedded itself up against the northwest corner of the stone-hole, at the same time sliding on the sloping base of the stone towards the north. That this accident had taken place during the original erection was further shown by the fact that the part of the stone-hole in front of the stone, and at the northeast corner, had been filled in with packing stones.

A discovery of considerable importance was made when, on examining material which had been tipped into the ditch to form the course of a medieval track-way, a causeway of solid chalk [Map] was found to exist. An original entrance through the bank and over the ditch from the north was therefore proved to have existed. The breadth of this causeway could not be ascertained owing to its northeastern edge being at present under the main road to Swindon.

The excavation of a disturbed area of ground a little distance within the Outer Circle and situated between Stone 1 and Stone-hole 2 disclosed the unsuspected existence of a definite stone-hole, to the west-southwest of which, upon a platform slightly below the untouched level of the chalk, lay a quantity of packing stones partly surrounded by a semicircle of closely set stake-holes, which had presumably held anti-friction stakes. So close was this stone-hole (PLATE 11, 1), referred to as 'A' on the plan, to that of Stone I that it would not have been possible for a stone to have stood in each at the same time, and since Stone I has never fallen, Stone A must have antedated it. A further feature of interest lies in the fact that no sign of destruction of this stone by fracture was evident on the site. At a distance of 36 feet to the northwest, during the process of the uncovering of the causeway to the untouched chalk, a second stone-hole ('B' on plan) was subsequently uncovered, although, owing to the scarping of the causeway on both sides to a gradual slope, possibly in historic but more probably in prehistoric times, only the bottom of this stone-hole still existed. A third stone-hole ('C' on plan) was discovered 36 feet to the north-northwest of B (PLATE 11, 2). In this case, however, the northeast limit of the ditch had actually cut through the socket, providing thereby conclusive evidence that this stone-hole, and it is to be presumed Stone-holes A and B as well, antedated the bank and ditch and Outer Circle, and that the Circles of Avebury consequently represented a structure of two separate periods. On the assumption that Stone-holes A, B and C formed the arc of a circle, it was shown that a third setting must have existed of practically the same diameter as those of the southern and central circles, while the centres of these three settings would have lain almost in a straight line. It seems impossible therefore not to conclude that what one may term 'Avebury I' consisted of three settings of stones (whether consisting of single circles or double concentric circles remains to be demonstrated by excavation), unaccompanied by banks or ditches. At a later date, although not necessarily after any great lapse of time, a monument of a different type consisting of a ditch, a bank outside it composed of the material from the excavation of the ditch, and a single attendant circle of megaliths on the inner side of the ditch, was constructed, the line of the bank and ditch being broken by entrances consisting of causeways of solid chalk. Whatever significance may have been attached to the lay-out of the earlier structure was evidently ignored, if indeed it had not already been forgotten, by the constructors of the second part of the monument, in view of the apparent removal, if not actual destruction, of the northern setting consequent upon the formation of bank, ditch, and accompanying stone-circle. From the evidence obtained during the excavations of the West Kennet Avenue by the Morven Institute in 1934 and 1935,1 as well as those of 'The Sanctuary'2 on Overton Hill, and of one of the Long Stones at Beckhampton3 by Mr and Mrs B. H. Cunnington, it may be assumed that 'Avebury I' is to be allocated to the BI Beaker-culture during the earliest phase of the Early Bronze Age in North Wiltshire. Mr Piggott (age 28) has suggested that 'Avebury II' may have been erected by A Beaker-folk at a date which would be contemporary with the later occupation of Windmill Hill. As yet, however, no definite datable evidence to confirm this plausible theory has been obtained. Owing to a certain amount of the bank having been thrown down to form a relatively recent cart-track over the northeast corner, it was necessary to carry out excavations to identify the original limits at this point. During this work sections of a discontinuous palisade trench were uncovered, the purpose of which had been to support the material of the bank in the front and partly round the end. It would appear that no such precautions had been taken beyond this point, nor at the back of the bank where natural silting had been allowed to proceed unchecked. A possible explanation for this feature is that the original erectors were not concerned with the appearance of the rampart unless seen from within the Circle, and the fact that the stones themselves have been dressed tonly in such a manner as to provide the required forms or shapes, when viewed from a similar position, may be cited as a parallel to this outlook. Alternatively, and more probably, the purpose of those palisades was simply to prevent material silting down from the bank over the causeway or into the ditch.

Note 1. Antiquity, December 1936, pp. 417-427.

Note 2. M. E. Cunnington, Wilts. Arch. Mag., 1931, XLV, 313-14.

Note 3. M. E. Cunnington, Wilts. Arch. Mag., 1913, XXXVIII, 3-6.

The entire course of the ditch as far south as the farm road, which represented the southern limits of the area in which either excavation or preservation is at present possible, was cleared of modern and relatively recent refuse to the depth of the top limit of natural silt. This entailed an immense amount of labour, and since no question of stratification was concerned, a mechanical excavator was employed for the purpose. The material obtained therefrom, much of which had been derived originally from that part of the bank which had been thrown down in the 17th century A.D. to permit the construction of the farm buildings which still cover the site, was replaced where the course could be accurately determined. It was found that in historical times not only had material been thrown into the ditch from the western side but the eastern side had likewise been quarried away to a considerable extent from behind Stone 11 to Stone 17. The silted limit of the ditch on the west was identified by the evidence of a buried turf-line, while the original side was shown, after excavation, by the untouched chalk. The original line of the ditch could consequently be restored and this work has been in progress until the present year.

A positive network of medieval and more recent ditches intersected the entire area under excavation. The relative chronology of these has now been satisfactorily arrived at, but this need not be considered as a subject into the details of which it is necessary to enter in this place.

At the conclusion of the excavations no endeavours were spared to improve the condition of this part of the monument prior to throwing it open to the public. Much of the area was turfed, obstructing hedges were removed, boundary fences re-adjusted, or re-erected as far from the Outer Circle as was permissible. Entrance gates were provided at suitable points and finally notice boards containing maps and cognate information were placed at suitable points. As evidence of the increased public interest in Avebury it may be remarked that the attendance of visitors rose, following the excavations under review, from IOO/ZOO a week to an average of between 1000/1500 during the summer months. These figures have not only been maintained but considerably exceeded since 1937.


Those who visited Avebury a year or more ago will remember the southwest sector as open pasture, flanked on the north side, except at the northeast corner, by a row of houses in varying states of dilapidation. The field contained one standing stone (no. 17 on plan) and one fallen (11), while two (13 and 18) had been buried in such shallow pits that small parts protruded above the surface. Smith1 had referred to the existence of only one other buried stone, which proved on excavation to be no. 9, although situated some distance from the position indicated on Lukis’ survey. In Colt Hoare’s1 time no further stones existed, but no. 11 was then standing. Stukeley2 however in addition shows nos. 1, 12 and 14 as fallen, the remains of the first of which were recovered during this season’s excavations, while both the others were found to have been destroyed by fire. Nine stones, but whether standing or fallen is not indicated, appear on Aubrey’s3 plan, the identity of some of which cannot however be fixed with certainty.

Note 1. British and Roman Antiquities, 1885.

Note 2. Abury, 1743.

Note 3. Monumenta Britannica.

As an essential preliminary to the work of excavation two derelict cottages, long condemned by the local authorities as unfit for human habitation, were demolished, together with attendant outbuildings and a modern stable. Beneath the foundations of these buildings, and between them and the village street, was found the site of a pond, finally filled in, according to local records, about the beginning of the nineteenth century. In the bed of this, and incorporated with the foundations of an old forge, lay the greater part of a megalith (I) which had been largely broken up by direct fracture. Pieces of this stone were identified in an adjacent buried wall, some of which it was found possible at a later stage to attach once more to the parent block after it had been erected.

To the southeast of the Stone I, five successive pits (2-6) were uncovered in what had formed the garden of the cottages. In each of these pits stones had been buried; but in every case these had later been rediscovered and destroyed by the fire-and-water method. Had it not been for the subsequent habitation of this part of the monument it is safe to say that all these stones would by now have been re-erected in their stone-holes, parts of which in all cases could be distinguished. Six more stone-holes (7, 8, 10, 12, 14, and 22) were identified with adjacent burning-pits. Besides Stones 9, 13 and 18, referred to above, five more stones were discovered buried, all at a considerable depth. These, together with the fallen stone (II) were re-erected in their original positions.

On the completion of the excavation of the western half of the Outer Circle, therefore, over half the original number of stones, not excluding the area which is still covered by the buildings at the south end of the northwest sector, may be seen standing erect, a much more satisfactory result in itself than could have been foreseen at the beginning of the work. (PLATE III).

During the excavation of the buried Stone 16 [Map], a complete skeleton was found within the narrow space between the stone and the only unfinished part of the side of the burial-pit. It is evident that the remains were those of an individual who had been accidentally killed while engaged in completing the pit for the burial of the stone, which had apparently slipped or fallen owing to a support giving way, fracturing the victim’s pelvis, and also breaking his neck. The right foot was wedged beneath the fallen stone and it had consequently been impossible at the time of death to remove the corpse. It had therefore been covered over and the pit containing stone and body filled in. The date of the occurrence, and the burial of the stone, could be accurately fixed to within a few years by the discovery near the man’s left hip of a discoloured patch of soil, doubtless representing the remains of a leather pouch, upon which lay thrce coins; two silver pennies of Edward I, minted at Canterbury in 1307, and a sterling of the City of Toul. Other finds beside the skeleton included a pair of pointed scissors, which were from their form definitely those rather of a barber than a tailor, and a small iron object, with the vestigial remains of a wooden handle, which had apparently been a lancet or probe. These objects were found beside the left thigh. The discovery of a pair of scissors in England accurately identifiable to so early a date as the first quarter of the fourteenth century A.D. is interesting.

In passing it may be remarked that the generally accepted explanation that is in order to facilitate agriculture-for the burial of so many of the stones at Avebury cannot, on other grounds than the employment of a barber for the task, be regarded as adequate, at any rate so far as the northwest and southeast sectors are concerned.

The skeleton has been accepted by the Museum of the Royal College of Surgeons, while the associated finds are housed in the Museum of the Morven Institute at Avebury.

The standing Stone no. 17 [Map] was excavated to its base, and was proved to have slipped accidentally during erection, much in the same way as had Stone no. 14 of the northwest sector, which accounted for its present position being oblique to the arc of the circle. In this case certain of the anti-friction stakes had been broken by the accident, remains of carbonized wood being found for the first time in the stake-holes.