Back to Archive

Project and Fieldwork 2008

Excavations at Pullyhour

The excavation of a small henge monument at Pullyhour, Halkirk, Caithness: interim report:

Richard Bradley and Hugo Lamdin-Whymark

In the north of Scotland there are a number of earthwork monuments that have been described as ‘henges’ or hengiforms’. They have a number of features in common. All are small circular enclosures, with an external bank, a wide internal ditch and a surprisingly restricted interior, but it is far from obvious whether they are more closely related to the ceremonial monuments of the Neolithic and Early Bronze Age, or to Irish ring barrows which resemble them in surface appearance.

Until recently, none of the Scottish sites had been excavated and published. One had been investigated by Time Team without any result, although it was located not far from the find spot of the Early Bronze Age Migdale Hoard. Another had been recorded during road building at Lairg Sutherland and had a terminus post quem of 1350 – 1050 BC. It also included a cremation burial with a date of 1180 – 930 BC. The first research excavation did not happen until Easter this year, with the investigation of a well preserved earthwork at Pullyhour in Caithness. The site was located on the side of a valley overlooking the Thurso River, in a position with evidence of Mesolithic activity. There had also been standing stones and a cist burial in the vicinity. The earthwork had a single entrance and a well preserved external bank (Plates 1 and 2). It was exactly aligned on their mains of large cairn on the opposite bank of the river. In Caithness, monuments of this kind normally date from the Neolithic; the closest comparison is with a passage tomb beside Loch Calder.

The enclosure faced south and was also directed towards the full moon at midsummer which rises above the far horizon. Excavation demonstrated that the earthwork was built in two phases (Fig.1). The first was a circular enclosure with a broad internal ditch which must have held water (Plate 2). It had been built in open country which was not used for farming. A date from the old land surface shows that it has a terminus post quem of 1620 – 1450 BC. In its second phase the enclosure was converted to an oval ground plan by widening the ditch on the inside. A small horseshoe-shaped bank was constructed on its inner lip, and the area inside it was cobbled. At the same time the outer bank was enlarged and its inner edge was probably revetted on by a rubble wall. This phase is dated between 1320 - 1120 BC. A small post was erected in the centre of the enclosure.

The entrance was very narrow and was flanked by a stone kerb. Just outside it was a pair of posts. One was much larger than the other, and the base of the upright still survived (Plate 4). To our surprise it was a substantial piece of pine, a species which became extinct in Caithness at the end of the Neolithic period. This is confirmed by a date of 2580 -2340 BC. Its socket had been cut through the secondary bank, meaning that it could not have been erected before the Middle or Late Bronze Age. Not only had the builders aligned the enclosure on an ancient monument, they raised a massive post which must have been dug out of a bog.

Its actual age could not have been known, but they would have been aware that it was a relic of some antiquity. The only artefacts associated with the Bronze Age structure were two flint flakes.

Finally, the monument was decommissioned. The rubble supporting the bank was pushed into the ditch and the entrance was blocked by a small cairn. The pine post was uprooted and the stones that had held it in place were smashed to pieces and used to cover its socket. There is no archaeological evidence of further activity on the site.


First, we must thank George McDonald for permission to excavate at Pullyhour.

We also wish to thank the following people: Paul Humphreys and Emma Sanderson for making many of the arrangements for the excavation; Dorothy Maxwell (Highland Archaeology) for advice and information;

Ronnie Scott for providing the tools and for his drawing skills; Strat Halliday for providing the unpublished Royal Commission survey of the site and discussing it with us; Alex Brown for pollen analysis; Amy Poole for her work on the soil samples; Alison Sheridan (National Museums of Scotland) for advice on the waterlogged post and for arranging for it to be dated; Richard Tipping for information on bog pine; Dougie Scott for advice on the orientation of the monument; Rod MacCullagh for information on the excavation at Lairg; and, last but not least, the hard working team, many of them from the Caithness Archaeological Trust and the North of Scotland Archaeological Society, who carried out the work, often in dreadful weather.

Plan of Pullyhour Henge

Figure 1

Plate 1

Plate 2

Plate 3

Plate 4

Excavation Chapel Hill, Dunbeath 2008

Second Interim Report 2008

This season concentrated on further investigating the features excavated in 2007 and opening up a series of new trenches within the scheduled area. This was largely to determine the nature of several geophysical anomalies identified from the resistivity survey (GUARD project 1002) and to investigate the origins of one of the large radial walls present on the site. A gradiometer survey was carried out in the scheduled area but no significant anomalies were identified. Also, a limited geo-radar survey was carried out by ORCA, over a transect across the proposed Trench 6 in order to evaluate whether the technique would be useful for future application.

Trench 2 was enlarged in order to gain a better understanding of features found in 2007. Excavation revealed that the natural stratigraphy of gravel, boulders and clay had been altered in order to enhance the effect of the terrace. This was indicated by scarping marks in the clay slope which, combined with the enhancement of the boulder layer to create a wall, would have acted as both a visual and functional boundary around the terrace. The wall was completely destroyed prior to the 17th century, dated by a Charles II bawbee discovered above the toppled stones. The extension towards the outer boundary wall demonstrated evidence for continued agricultural activity over a large period of time, interspersed with episodes of dumping down the slope which led to the accumulation of a considerable depth of soil. Finds consisting of animal bone, pottery and metal objects and ranging from the 12th century onwards, suggests that this area had a long association with agriculture. Natural mixed sand and gravel at the base of the slope extended towards the edge of the adjacent river, with no upward slope which might have indicated a ditch. A deposit of large stones was uncovered at the western end of the trench, though whether this is a dumped layer serving as possible banking for the slope is uncertain.

Trench 3 was extended towards Chapel Hill and Wall A in order to further examine the extent of the cobbles found in 2007, as well as to continue the investigation into a linear feature indicated by resistivity. Excavation only allowed for two clear edges of the cobbled level to be uncovered, later shown to be two distinct phases, approximately 9m wide and running at an angle of c. 45 degrees to the wall at a comparatively shallow depth of only c. 0.2m below the surface. This odd angle makes interpretation as a pathway or road rather dubious; however a function as a working surface is certainly plausible when the sizeable amounts of slag uncovered are taken into consideration. Sherds of Late Medieval pottery including imported German stoneware, date the cobbled surface to the late 14th or 15th centuries.

Trench 4 was originally situated to examine the origins and purpose of one of the large extant walls (Wall A) which radiate from Chapel Hill. The trench was placed across a perceived blocked portion of the wall in order to see if there was any path or roadway leading up to the ‘gap’. For safety reasons, and to make as little intervention as possible, it was decided to locate the trench 1.5 metres away from the wall. It became apparent that a great depth of topsoil had accumulated over the archaeological deposits. At 0.7 – 0.8m there was a dark layer which extended across the whole trench and contained much evidence for burning. This layer contained slag and burnt clay along with a variety of pottery preliminarily dated to the late Norse period. The purpose of this layer remains unclear but it may be related to large scale industrial activity or a widespread destruction event. Sealed beneath this layer were a number of archaeological deposits.

The earliest feature encountered was a possible bank made up of compacted stones and gravel. Although the full extent was not uncovered, it was not very substantial suggesting that some attempt had been made to level it. A section of walling was revealed running parallel to the bank and apparently post-dating it (although further excavation would need to confirm this). The wall was of substantial build (1m wide) and of very high quality. There seems to have been some attempt to level the area between the bank and the wall by infilling the gap between them with a mix of soil and clay material, from which came fragments of 6th-7th century pottery. This would place the construction of the wall at least as early as the 7th century and the bank somewhat earlier. Although no floor surfaces have so far been revealed it is likely that the wall formed part of a structure probably relating to the early Christian phase of occupation. At some point the wall seems to have been demolished and the lack of tumble suggests that it was deliberately demolished probably to re-use the stone elsewhere.

After the wall was demolished a layer of brown soil build up suggests a change of use for the area. The layer contained burnt sheep bone and charcoal which is suggestive of a dispersed surface midden probably adjacent to an area of domestic activity. Also, associated with this period was a build up of sandy material against one side of the wall which contained flecks of charcoal and burnt clay, again suggestive of activity with the vicinity. Cut into the brown build-up layer were a number of features associated with craft-working activity and relating to a post-hole which was also cut into the top of the wall. One of the features was identified as a hearth and contained small fragments of burnt clay, which could be the remains of clay moulds, and another feature was associated with extensive burning and fragments of kiln furniture. Pottery finds from within the post-hole and other features suggest that this phase of activity relates to the later Norse period.

An offshoot from the main Trench 4 was excavated to examine the base of the Wall A and the relationship of the apparent blocked gap to the rest of the wall. Through narrow exploratory excavation it is believed that the base of the wall was revealed but without further excavation this cannot be confirmed. It was established that the wall was standing in the late Norse period and was probably not constructed much earlier. What was also apparent was that the ‘break’ in the wall does not seem to carry on to the whole depth of the wall; in fact it seems that it does not go much below the present ground surface. Furthermore, no traces of any pathway or road leading up to the gap could be found. This raises several questions about the phasing of the construction of the wall. It seems that while the base of the wall was initially constructed in the late Norse period, the upper portion, with the gap, may have been added at some point later, possibly in the 15th or 16th centuries.

Trench 6 was located across a low resistance linear anomaly, which was later reinforced by the geo-radar survey results suggesting a deep ditch-like feature. Excavations again revealed considerable a depth of plough-soil, beneath which was a mixed midden-like deposit within a widespread layer that also contained evidence for extensive burning. This included late Norse pottery as well as large quantities of slag and possible iron ore or haematite. Sealed by this were the partial remains of what appeared to be a hearth or post-hole, although further excavation is needed to confirm this. The natural was mixed gravel, stones and boulders, into which was cut a ditch approximately 1.1m wide and 0.75m deep, and revetted on its east side. Primary deposition contained only charcoal however the upper fills, which also covered the surrounding natural, contained pottery, burnt bone and charcoal fragments which are suggestive of occupation nearby. It is notable that the soil in this area becomes shallower as it slopes up towards the base of Chapel Hill, suggesting a lack of extensive cultivation.

Trench 7 was located across what the GUARD resistivity survey had suggested was a structure. The trench was dug to natural without any archaeological feature being discovered.

Trench 9 was located across one of several low mounds. Excavations were incomplete but exposed stones were interpreted as, either a robbed-out Bronze Age cairn, or a stone dump, a common feature on the site for recycling stone as a raw material. The discovery of a, higher than usual, proportion of water-worn quartz pebbles adds weight to the former interpretation. Work will continue in 2009.


Edward Oakley, Anne Sassin, Imogen Tompsett

Geophysical survey, excavation

Department of Archaeology, University of Nottingham

For further details contact Meg Sinclair at Dunbeath Heritage Centre on 01593 731 233

Continuation of excavations at Whitegate Broch, Keiss, near Wick, Caithness - 2008

The community excavation of Whitegate Broch continued, further detail of structural features was exposed with finds including Iron Age pottery and food debris. Most notable was the excavation of a beautifully constructed feature thought to be a well. It is not clear if the ‘well’ was purely functional or wether it had a ritual purpose; there is evidence to suggest that Iron Age people worshiped Chthonic (subterranean) deities.

All content © Caithness Archaeological Trust 2004

unless otherwise stated