Ross and Cromarty Archaeological Services West Coast Archaeological Services

Comar Wood Dun Cannich, Strathglass

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Comar Wood Dun Archaeological Evaluation and Measured Survey

Cannich, Strathglass, Scotland

Final Report

Ross and Cromarty Archaeological Services West Coast Archaeological Services Ryefield, Tore, Ross-shire, IV6 7SB The Salmon Bothy, Shore St, Cromarty, IV11 8XL Tel: 01463 811310 Tel: 01381 600726 Mobile: 07891 578998/07776 027306 Mobile: 07867 651886 [email protected] [email protected] www.rossandcromarch.co.uk - 2 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Results of the Archaeological Evaluation at Comar Wood Dun

Client Forestry Commission Scotland

National Grid Reference NH 32509 31008

NMRS No. NH33SW 42

Canmore ID 314324

HHER ID MHG55867

RoCAS Report 2014-27/COM13

Oasis No. rosscrom1-170218

Date September 2014

Authors Mary Peteranna, Steven Birch and Lynn Fraser

Acknowledgements

We wish to thank Matthew Ritchie, Giles Drake-Brockman and Neil McAdams of Forestry Commission Scotland for their assistance on-site and Edward Martin for capturing aerial photographs during the fieldwork. Thanks are also due to George Geddes (RCAHMS) for archival information. Fieldwork was carried out by Steven Birch (WCAS), Lynn Fraser and Mary Peteranna (RoCAS), Cathy MacIver (CMS Archaeology) and Sam Williamson, with assistance from Brian Duff and Greer Jarrett. Ordnance Survey mapping is reproduced by permission of Landmark Information Group under RoCAS licence LIG1044. Mapping containing Ordnance Survey data is subject to Crown Copyright 2014. All rights reserved. Licence number 100022432. This is an unpublished report.

- 3 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Contents

Summary 7

Introduction 8

Survey 10

Excavation 13

Walling 15

Interior features and stratigraphy 19

Posthole alignments 23

Enclosure wall 26

Later structures 27

Tree disturbance 29

Analysis and Interpretation 29

Post-excavation results 29

Phasing 33

Interpretation 35

Discussion and Conclusions 37

References 41

Tables

Table 1 Radiocarbon Dates from Comar Wood Dun

Appendices

Appendix 1 Comar Wood Dun, Cannich, Strathglass Artefactual Analysis Beverley Ballin Smith

Appendix 2 Comar Wood Dun, Cannich, Strathglass: botanical remains Susan Ramsay

Appendix 3 Animal bone from Comar Wood Dun, Strathglass Catherine Smith

- 4 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

List of Illustrations

Illustration 1 Comar Wood Dun, at the time of discovery and prior to felling of 7 mature Douglas Fir

Illustration 2 Location of Comar Wood dun 9 Illustration 3 Landscape position of Comar Wood dun 11 Illustration 4 Topographic survey of Comar Wood dun 11 Illustration 5 Landscape position of Comar Wood dun and nearby dun / fort sites 12 Illustration 6 Aerial photograph of the site at the start of the excavation, looking 14 south over Strathglass

Illustration 7 Measured survey plan of Comar Wood dun and enclosure 14 Illustration 8 Comar Wood dun, prior to excavation, looking ESE 16 Illustration 9 Looking over Trench 1 after initial excavation; facing W from the 16 courtyard; facing E from outside the dun

Illustration 10 Looking over Trench 2 after initial excavation 16 Illustration 11 Secondary outer wall face on the north side of the entrance passage, 17 and outside face of the dun wall in Trench 2

Illustration 12 Looking over Trench 5, facing NNW, showing the blocked up 17 passage and possible outside walling and slabs/steps; inset: blocked up passage

Illustration 13 Dun wall in section (Trench 1), facing S as viewed from the north 18 side of the entrance passage

Illustration 14 Possible intramural space on the N side of the dun (Trench 2) 18 Illustration 15 Mid-excavation plan of Trench 1 19 Illustration 16 Post-excavation plan of Trench 2 19 Illustration 17 N-facing section through bank of material against inner wall and later 21 wall (110)

Illustration 18 N-facing section through courtyard deposits with SE edge of hearth 21 (118)

Illustration 19 Aerial image of Trench 1, with hearth (118) and dun wall; inset: 21 rotary quern (SF 1.03) in situ in the SW corner of the hearth

Illustration 20 N-facing section drawings from Trench 1 showing dun wall and 22 courtyard deposits

Illustration 21 E-facing section of Trench 2 showing dun wall, possible intramural 22 space and interior deposits

- 5 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illustration 22 Post-excavation plan of Trench 1, showing posthole locations 24 Illustration 23 Looking south over the entrance passage, showing postholes (186, 24 131, 188 and 122)

Illustration 24 Posthole (191) at the start of excavation (left) and pit (151), post- 25 Excavation

Illustration 25 Section drawings of posthole features from Comar Wood dun 25 Illustration 26 Plan of Trench 3, showing the enclosure wall with entrance structure 27 Illustration 27 Looking SSE over Trench 3, showing the enclosure wall and 27 entrance structure (310)

Illustration 28 Plan of Trench 4 and Structure 1 and Trench 6 and Structure 2 28 Illustration 29 Looking NE across the interior of Trench 4 28 Illustration 30 Quernstones from Comar Wood dun; SF 1.03, SF 1.02 31 Illustration 31 Central hearth setting (118) 31 Illustration 32 Post-excavation image of Comar Wood dun, during the 2013 32 evaluation, showing the location of walling and postholes

- 6 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Summary

An enclosed dun was discovered by Forest of which the site was rebuilt and reused for District staff in Comar Wood, Strathglass, several centuries before abandonment. 1km southwest of Cannich, during a pre- Posthole alignments in the entrance passage felling check on the woodland in the and interior courtyard showed evidence for Inverness, Ross and Skye Forest District in the construction of timber structures inside 2010. The site was initially interpreted as a both areas. Other structural evidence galleried dun with a well-defined defensive uncovered included later interior walling outwork together with later buildings used to constrict the courtyard space and two constructed on the outside of the site within successive central slab-built hearths, the last the main structure and outwork. The dun and of which contained broken quern fragments. the area around it were clear-felled by hand Opposite the dun entrance on the west side, in 2013, in a manner which avoided was an entrance recess through the outer structural damage to the site. enclosure wall, while on the east side of the dun evidence for an opposing entrance Following site clearance, an archaeological passage was uncovered. The poorly measured survey and evaluation were carried constructed walls revealed inconclusive out to record the site and establish both the evidence for intramural gallery spaces and nature and extent of any surviving only a small amount of artefactual material archaeological deposits and any damage was recovered. caused by afforestation. The main objectives of the fieldwork were to recover material that Although poorly built walling and significant could be used to securely date the site and to tree root disturbance caused difficulty for the interpret its form and function. The results excavators, the results of the evaluation and would enhance the historic environment post-excavation analysis have expanded the record and Forest Design Plan and understanding of this site type, a prominently contribute to the Scottish Archaeological sited roofed structure which may have been Research Framework (ScARF). visited intermittently for ceremonial or defensive purposes. The excavation results The excavations revealed that Comar Wood also included evidence for Mesolithic forest dun had been constructed during the second clearance on the site and later buildings of an half of the 1st millennium BC. Evidence for unknown date within the collapse. two burning events was uncovered, after both

Illus 1

Comar Wood Dun, at the time of discovery and prior to felling of mature Douglas Fir (© Forestry Commission Scotland)

- 7 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Introduction Otherwise, a name embedded in local folklore Comar Wood dun is situated on a rocky crag suggests that the site was known as Creag na on a slight terrace on a southeast facing Fannich (McAdams pers comm). hillside above the River Glass. Located 1km southwest of the village of Cannich, in At the time of discovery, the site was set Strathglass, Inverness-shire, its position, within thinned mature conifers but had not 42km southwest of Inverness, places it at the been ploughed. There was also visible wind watershed dividing the Scottish west coast blow occurring upon the wall of the outwork and the central and eastern Scottish Highlands and in the enclosure. In order to inform future (Illus 2; 3). The glen runs strategically from forest management a detailed measured the east at the Beauly Firth, which forms the survey of the dun was carried out using a southwestern extent of the Moray Firth and plane table and alidade. The survey of the site, feeds into the North Sea, and the west at Loch as carried out by Forestry Commission Duich, which feeds into the Inner Sound of Scotland, indicated that there was potential Skye, the Minch and the Atlantic Ocean for significant buried archaeological deposits beyond it. Comar Wood dun is located almost and structural elements to have survived. centrally along this glen and forms one Observations from this survey noted that a element of a complex prehistoric and historic structure 11m in diameter with a western archaeological landscape centred on entrance was enclosed by walling, which Strathglass, Glen Moriston and Glen utilised the natural break of a slope and cliffs Convinth. These fertile river valleys penetrate to the southeast. The central structure the mountainous hinterland from the coastal comprised a well-preserved drystone wall, plains around the Beauly Firth, while Glen measuring approximately 4.8m wide and Moriston provides a through route to the standing up to 1.8m in height and contained shores of Loch Ness. The relatively flat valley three depressions interpreted as possible bottoms provide good quality agricultural intramural galleries. Several breaks in the land that would also have been attractive for enclosing outwork were noted as were two settlement during prehistory. small buildings built into the spread tumble from the dun and its outwork. During 2010, in advance of felling operations in Comar Wood, Cannich, a previously In 2013, in order to protect the site from unrecorded prehistoric enclosed, galleried dun damage during felling operations, twenty- was re-discovered by Forest District staff. It three mature trees were removed from the dun was only after the site was discovered that using a soft-felling technique, which involved reference to the site was found in the Royal gradually lowering a felled tree to a suitable Commission on the Ancient and Historical height to be processed. Monuments of Scotland’s Emergency Surveys of 1943, where a written entry states: The appearance of the monument before excavations commenced in 2013, with its Fort, Cannich circular form, evidence for intramural There is said to be a fort near to the N. end of galleries or cells and significant amounts of Comar Wood, somewhere to the W. of the stone, suggested that the site at Comar Wood foot-path that rises from the Cannich- comprised the remains of a monumental Fasnakyle road to the Cannich-Glen Cannich complex roundhouse; possibly a broch or a road and near its junction with the latter. The dun. These stone-built structures have long fort could not be found, but when the been a dominant feature in the study of the inventory survey is made it should be looked Scottish Iron Age, and their classification and for again – if possible at a season when the development has provided much debate. bracken is down. Sept. 16th 1943.

- 8 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 2 Location of Comar Wood dun

Recent work of the Scottish Archaeological archaeological remains in order to better Research Framework panel (ScARF 2012) understand the structure and to recover has identified the central Highlands as a key securely datable material, whilst also ‘black hole’ in terms of understanding the evaluating the extent of tree root damage. The context of enclosed places, with a particular evaluation provided evidence that this Iron issue being the lack of dating evidence, which Age roundhouse had been occupied and is a severe constraint on understanding reused during multiple phases over a period of enclosed places. The fieldwork objectives up to 600 years. were to evaluate the extent and nature of the - 9 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Survey The enclosure wall, best preserved on the west side of the site, comprised a mix of large A detailed contour and measured survey of stones, boulders and some large slabs the site and landscape features (Illus 4; 7), standing 0.7-1.2m high and spread 2-3m including the location of tree stumps on the wide. A short section of the outside enclosure structures, was conducted prior to wall face was visible on the north east side of commencement of the archaeological the site. Although there was no clear excavation. The measured survey was indication as to the original width of the wall, conducted using a Leica 705 Total Station. it is estimated to be 1.5m wide based on the The contour survey was conducted using a spread of collapse. The entrance to the staff-mounted Trimble GeoXR Rover. All enclosure was located on the west side of the survey data was three-dimensional and dun, slightly offset from the dun entrance, referenced to the British National Grid and with access to the site running from west to Ordnance Survey datum. east.

The collapsed stone was mostly moss- Two later buildings, in a degraded condition, covered, although a considerable amount of had been built against the southeast side of stone was visible due to displacement by the the dun, utilising the outer wall face as part of felling and clearance of trees and vegetation the building construction. Structure 1, which from the site. The stone used on the site was a measured 3.5m NE-SW by 2m internally, was fairly homogenous type: a pale grey, hard a subrectangular building comprising rubble- schist-type comprising sub-angular and sub- built walls of large and medium-sized stones rounded stones. Tree brash and other debris and slabs standing up to 1.2m at the west end obscured the visibility of much of the outer with collapsed stone spread 1.5-2.5m to all enclosure walling and the surrounding sides. There were no visible faces noted in the landscape. walling, with the exception of the dun outer wall face which formed the northwest inner The circular roundhouse structure measured face of the structure. The entrance, most 22.8m N-S by 22m E-W externally, with the likely in the southeast wall, had been stone spread increasing the size to 27m N-S destroyed by tree planting and root action. by 26.6m E-W, and standing to at least 1.5m above the ground surface. The outer wall face Structure 2, measuring 4m NE-SW by 2m was best preserved in the arc from north to internally, was the better preserved of the two west with some large facing stones visible, buildings with wall faces visible on its while the position of it was projected in southwest side. Also subrectangular, the places, particularly in the south where it was building comprised double-faced, drystone mostly obscured by collapse. The entrance, walling 0.7m wide, which stood up to 1.5m located on the west side of the structure, high in the south corner where it joined with comprised a linear hollow 2m-3m wide that Structure 1. The entrance, partially obscured ascended gently to the east where it entered by collapse, was located in the centre of the the inner courtyard. No definite features were southeast wall and measured 0.7-0.8m wide. visible within the entrance or the inner A third building, Structure 3, was built against courtyard, which, as projected, measured the inside of the enclosure wall on the approximately 13m in diameter. At the time southwest side of the site. A subrectangular of the survey the width of the roundhouse building, it measured 5m NW-SE by 3m wall was projected to be from 3.5m to 4.2m overall and 3.8m NW-SE by 2m internally, wide, being widest in the north – although this comprising large stones and boulders with was found later to be incorrect. Several some slabs. The wall was double-faced with depressions noted in the top of the wall were the outer wall face visible in the northwest interpreted as either intramural spaces or corner and both faces visible on the southeast robbed out stone hollows. gable end.

- 10 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 3 Landscape situation of Comar Wood dun

Illus 4 Topographic survey plan of Comar Wood dun - 11 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 5 Landscape position of Comar Wood dun and nearby dun / fort sites - 12 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis The entrance, partially obscured by collapsed dun measuring 24m by 16m internally, with stone, was located in the centre of the walls 4.5-5.9m thick, and stone outworkings. northeast wall. A gap in the dun enclosure Castle Spynie (NH54SW 9), located 4.5km to walling to the northwest of Structure 3 could the southeast of Beauly, is a near-circular have been as a result of stone clearance structure, 19.5m in external diameter with conducted to improve access to the later walls around 4m wide, classified as a dun or buildings on the site. Structure 4 was built on broch with outworks that appear connected the northeast corner of the enclosure and with natural out-crop and cliff to form an consisted of a small, subcircular rubble wall encircling, outer defence. The appropriate built into enclosure wall on the northeast side classification for these monuments, which of the site. share many of the features found at Comar Wood, is not clear from survey alone. The presence of later structures built into the collapsed stone of prehistoric sites is not an Preliminary interpretation of the site at Comar uncommon phenomenon with Altbreck Broch Wood suggested that it could have been (SM 1829) and Carn Liath long cairn (SM similar in form and construction to any of the 4752) being two examples. Unfortunately, it above sites that are located in the wider is difficult to ascribe a date to these later landscape in and around Strathglass. structures, without documentary evidence. However, it is only through excavation that elements recorded in survey can be It is also difficult to accurately interpret the conclusively interpreted. features of enclosed dun structures from survey alone. Within the surrounding landscape of Strathglass, a wide range of Excavation potentially contemporary sites have been identified (Illus 5), including forts and duns (a Tree brash littered the area between the dun few of which display evidence of and enclosure wall and the area beyond this at vitrification) and brochs. Few of these sites the start of the fieldwork (Illus 6). There was have been excavated or investigated in any some evidence for stone displacement, despite detail, although most have been visited over the extremely careful attention given to the the years by the Royal Commission on the site by the foresters. Prior to setting out the Ancient and Historic Monuments of Scotland. trenches, the site was cleared as much as possible of loose debris. Located to the southwest of Comar Wood dun, 3.5km along Strath Glass, Knockfin Four trenches (Trenches 1, 2, 5 and 7) were (NH32NW 2) comprises a wider, placed over the dun and inside the courtyard subrectangular structure, measuring 30m by of the structure while three trenches (Trenches 21m with walls 2.5-3m wide, built on a rocky 3, 4 and 6) were placed to explore external outcrop and with possible enclosure walling structures (Illus 7). Prior to excavation, the built up to the outcrop. Less than 4km to the trench areas did not appear to have suffered northeast in Strathglass is a possible broch or extensive damage by recent felling operations. dun, Dun Coille Struy (NH33NE 1). Situated The overlying vegetation was dense, dry moss on top of a steep rocky spur, the structure and a layer of soil/forest debris approximately measures 18.5m in diameter externally with 3-5cm deep. This deposit and loose rubble walls up to 4.2m thick and possible evidence were removed during the first stage of for a well-preserved oval guard chamber and excavation. intramural galleries. A second circular stone- built structure is located nearby at Struy Trench 1 was laid-out through the entrance of Bridge (NH43NW 1) measuring around 20m the structure, measuring 15m long on an E-W in diameter externally, although little further axis, and extending from approximately 2m detail is noted. Dun Mor (NH44NE 8), near beyond dun wall to centre of the courtyard. Kilmorack, is another roughly circular fort or - 13 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 6 Aerial photograph of the site at the start of the excavation, looking south over Strathglass

Illus 7 Measured survey plan of Comar Wood dun and enclosure - 14 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis The trench was positioned to evaluate the and archaeological contexts were sampled in inner and outer wall faces of the dun, bulk and partial quantities for wet sieving and structural elements of the entrance passage flotation. and courtyard, courtyard features such as hearths and the type and depth of deposits Initial excavation conditions were similar below the stone collapse. A number of tree across the site, with a thin pine needle mat, stumps were distributed within the limits of felling debris and moss overlying loose stone the trench and some tumbled stone was (Illus 8). Small and large tree stumps and root visible within the entrance passage. systems were also encountered through to the base of stratified layers. A light brown sandy After the first level of excavation, Trench 1 loam mostly compacted at the base of the was extended to better evaluate the exterior stone marked the transition between primary and interior wall faces, while Trench 7, a silting over the abandoned structure and small test pit, was located at the eastern end archaeological horizons. With the exception of Trench 1 to look for an opposing posthole of the dun wall outer face, removal of these on the opposite side of the site. layers failed to reveal any standing walls or features within the structure (Illus 9-10). Trench 2 was aligned north-south across the north side of the dun in order to target what Walling appeared to be a gallery or chamber within the dun wall. It measured 10.7m N-S by 2m, with The dun wall faces did not easily present a small 1m-wide extension on the west side of themselves and excavation was not the trench. straightforward. Inside Trench 1, removal of the rubble layer was expected to reveal Trench 3 was located over the one side of the passage walling and inner/outer wall faces. enclosure wall entrance, to look for evidence The presence of tree roots over and above the of contemporaneity with the dun and passage had caused substantial destruction to structural features in the entrance. Trench 5 the site, leaving once sturdy walls collapsed was set against the dun outer wall face on the and rubble-like in appearance. The exterior east side of the structure in order to evaluate a face of the dun wall was most visible on the section of the external wall which appeared to north side of the entrance passage, while the curve inward and to look for potential dating inner face did not clearly reveal itself. material below the wall. On both sides of the entrance passage (Illus Trenches 4 and 6 were placed over two later 15), the exterior wall face had been structures attached to the southeast side of the augmented with a secondary outer skin of dun. Trench 4 was situated in Structure 1 on walling (101). The primary face (102), as the southwest side of the dun to assess the exposed on the south side of the entrance, date and function of the building. The comprised well-set boulders with smaller structure utilised the dun wall as its northwest stones forming the upper courses, up to three wall with its southeast wall abutting Structure courses high (0.75m) above the ground level. 2. Trench 6 was situated in Structure 2 on the The secondary external skin (101) comprised southwest side of the dun to assess its date, large basal boulders with 1-3 courses of function and relationship to Structure 1. It smaller stones overlain by flat slabs on top also utilised the dun wall as its northwest wall (Illus 11). Vertical pinning stones marked the and it appeared that this building had been separation between the two faces, although constructed after Structure 1. this was not as clearly visible on the north side of the entrance, where the secondary wall Excavation of the trenches was conducted face was better preserved, visibly sloping over a two-week period during variable towards the interior and concealing the weather conditions. All artefacts and ecofacts primary face. were retained and stabilised for later analysis - 15 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 8

Comar Wood dun, prior to excavation, looking ESE with the entrance in the centre of the photograph

Illus 9

Looking over Trench 1 after initial excavation; facing W from the courtyard (left); facing E from outside the dun (right)

Illus 10

Looking over Trench 2 after initial excavation; facing N from the courtyard (left); facing S from outside the dun (right)

- 16 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis There was evidence that the secondary Other exposed sections of the outside wall exterior wall face on the south side of the face were revealed in Trenches 4, 5 and 6. passage, which sat on a stony-silt layer, had Interestingly, in Trench 5, on the east side of been rebuilt. Extra layers of stone (119) set the dun, the exterior wall face curved inward around this outer stonework appeared to be an to a 1m-wide section of wall (506) comprising even later modification to constrict the medium-large stones – in contrast to the large entrance passage opening. The dun wall core boulders of the main outer wall face (504). (105) consisted of a loose fill of medium- The change in construction was interpreted as large stones and smaller clasts with voids. a blocked-up passage and not a wall repair, as the change in wall alignment was located The outside face of the structure was easily directly opposite the main entrance, which identified on the north side of the site in had also been augmented with a secondary Trench 2. A loose layer of stone collapse from external skin of walling that elongated the the structure covered the face (201), part of entrance. Also in Trench 5, the blocked up which had been visible during the survey. The passage and the remains of a possible exterior wall survived as two-three courses of wall (505) adjoined possible step slabs (513) subrectangular boulders, the lowest of which and a compact layer of stones (503) possibly was packed in behind redeposited subsoil set as slabbing (Illus 12). These sections of (203). The wall core comprised large, loose stonework were set down after the passage stones (206) on the surface packed at the base was blocked up. with smaller stones in a silty matrix.

Illus 11 Secondary outer wall face on the north side of the entrance passage, facing E (left); outside face of the dun wall in Trench 2 (right)

Illus 12

Looking over Trench 5, facing NNW, showing the blocked up passage and possible outside walling and slabs/steps; inset: blocked up passage

- 17 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis On the inside, identification of the inner wall formed part of a paved surface associated face was hampered by large tree stumps and with the intramural cell – although this was root systems and poor preservation of the not picked up at the time of excavation. The structure. In Trench 1, on the south side of the excavators concluded that the interpretation in entrance passage, an extension to the trench Trench 2 remains unclear due to the poor edge eventually revealed an interior wall face quality of construction that resulted in poor comprising two courses with the larger basal survival of structural elements. Although stones set into the subsoil (Illus 13). On the further fieldwork would be required to north side of the entrance passage, the inner confirm the presence of intramural spaces wall face was not discernible due to the within this structure, it did appear that a presence of a large tree stump and its compact stone layer (223) represented interior damaging root system. walling abutting the upright, possible passage, stones. If the interpretation is correct, overall In Trench 2 (Illus 16; 21), which the wall width would have been approximately excavators had hoped would reveal the 4.4m of the north side of the dun. presence of an intramural cell or gallery on the north side of the structure, excavation Further inside the dun, a single-coursed through stone tumble and sediment eventually rubble-built wall was identified in both uncovered an interior face (213), which Trench 1 and Trench 2. In Trench 1, the wall formed a dun wall measuring 1.9-2.0m wide. (110) comprised medium subangular stones Unfortunately, most of the inner facing standing 3-4 courses high, built over a silt stones, some of which were upright slabs, had layer with small stones and cobbles and collapsed out. This had revealed the soil-stone frequent charcoal fragments and carbonised fill (208) at the base of the stone wall core roundwood timbers (168). The wall was (206). On the inside, a layer of stone tumble enclosed on both sides by compact stone and covered a charcoal-rich ashy silt (207) silt layers (109 and 182). In Trench 2, overlying the subsoil and was interpreted as a remnants of a single-course wall (215) in a spread of material from a primary burning similar position were identified in the eastern event. Opposite the inner dun face, remnants trench section. In the western trench section, of possible courses of stonework flanked by it appeared to have survived as basal slabs upright stones (214) were interpreted as under rubble with two courses of upper slabs, possible walling and a passage, which may enclosed by stone and silt layers (209/218) have formed part of an intramural cell (Illus and 210) on either side. In both trenches, the 14; 21). Several flat recumbent stones around walls were interpreted as a later demarcation the base of the upright stones may have of the courtyard space.

Illus 14 Illus 13 Possible intramural space on the N side of the dun Dun wall in section (Trench 1), facing S as viewed from the north side of the entrance passage ( Trench 2), with the inner wall to right and upright passage stones to left - 18 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 15 Mid-excavation plan of Trench 1

Illus 16 Post-excavation plan of Trench 2

Interior features and stratigraphy Finally, along either side of the entrance passage (Trench 1), two single-coursed stone The interior of the dun was relatively free alignments (166) on the south and (167) on from collapsed stone. However, banked up the north, were all that survived of stonework against the dun inner wall faces, under stone inside the entrance, which had been disturbed collapse, was a complex sequence of by tree root action. The stonework partially charcoal-rich lenses and mixed deposits covered postholes relating to an entrance containing small cobbles and stone clasts. In structure, suggesting that posts had been Trench 1, spread across the centre of the inserted prior to completing the walling inside levelled courtyard, shallow deposits the passage. During excavation, an alignment represented ash and charcoal layers over and of an angled stack of eight stones (103) was around the central hearth (118), which had uncovered in front of the south passage wall spread down the entrance passage as thin (166). One interpretation is that the stones had patchy lenses. The layers of material banked been placed there during a re-building phase against the inner dun wall have been but were never used. interpreted as material cleared out of the

- 19 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis courtyard during several phases of reuse (Illus tree root movement. The slabs were 17; 20). The lowest of these layers comprised surrounded by a kerb (113) on the north, east mixed stone and charcoal-rich sediment and south sides with the open end of the (107), most of which was packed in a bank hearth facing the entrance to the dun (Illus 19; against the interior dun face. This layer was 31). Two alignments of edge-set stones (114), sealed by a black charcoal-rich layer (168) which also contained a rotary quern fragment that contained a substantial amount of (SF 1.02) and may have been an earlier hearth carbonised hazel roundwood. Collapsed stone setting, continued off the west side of the kerb (108), later courtyard clearance (109) and and aligned with the entrance passage wall. eventual silting formed the remainder of the sequence on the west side of the later wall A lens of charcoal-rich silt (117) directly (110), which likely served to retain the above the hearth slabs and below the upper cleared out stone and debris and demarcate ash layer (115) appeared to be directly the later courtyard interior. On the inside of associated with the hearth. Below this hearth wall (110), which was built on top of the were the remains of an earlier and smaller burnt layer (168), there was a further bank of slab-built hearth (176), which covered an stone (182) that appeared to have been cleared earlier pit (183) cut into the subsoil. The pit up against the later wall. At the very base of was filled with a pale orange ash deposit the depositional sequence were patchy lenses (185) interpreted as a hearth material. This of charcoal-rich silt (153), which overlay the marked the separation between the hearth pit subsoil (116), an extensive area of which had and the hearth settings. During sectioning of been heat-affected (178). the features, it was also shown that the upper hearth slabs (118) had cut through the On the opposite side of the courtyard, intermediate ash layer (175) that sat on top of multiple tree root systems had disturbed the the intermediate hearth slabs (176) and that upper deposits. An extension to the trench the uppermost hearth kerb (113) cut through north of uncovered a later wall (193) aligned the lowest burning layer (153). This provided with the north side of the passage and which evidence for three sequential hearth structures connected to the continuation of (110) on the in the centre of the dun courtyard. A number north side of the courtyard. Below the wall, of small stakeholes, (140), (142) and (146), under a layer of silting was a black, charcoal- around the outside of the hearth probably rich layer containing collapsed, burnt timbers. related to cooking or screen structures around This was interpreted as part of the burnt layer it. (168) from the south side of the passage, and represented a burning event in which internal Below the spread of hearth material (111) was partition structures had collapsed. a mixed silty ash deposit (121) with charcoal, peat ash and small stone clasts, some of which Across the level area inside the dun courtyard, had been affected by heat. Larger slabs (148), a 30-35cm deep, fairly continuous sequence measuring 0.5m long by 0.5m wide and up to of deposits was excavated (Illus 18; 20). 0.15m thick, at the base of this layer were Below the upper silting layer (112), a mixed interpreted as the remains of paving between ash and silt deposit (115) overlay a charcoal- the entrance and the courtyard. The interior rich lens of material (111). Both layers surface of the dun was the orange sandy clay appeared to have generated from a large slab- subsoil (116) which contained extensive built hearth (118) and had spread across the vitrified patches (Contexts 178 and 184), courtyard to either side, with patches of (111) particularly in the area between the entrance also spreading down the entrance passage. and the courtyard. The charcoal-silt lenses The hearth, a square setting, measured 2.2m (153) on top of the subsoil were interpreted as long (E-W) by 0.8m wide. Small subangular the spread of debris associated with a primary hearth slabs included part of a lower rotary burning event – material that was also quern (SF 1.03), which along with the west uncovered in Trench 2 (207, 212) and Trench end of the hearth, had been fragmented by 5 (507). - 20 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 17

N-facing section through bank of material against inner wall (right, next to vertical scale pole), with later wall (110) in centre

Illus 18

N-facing section through courtyard deposits with SE edge of hearth (118) in front right

Illus 19

Aerial image of Trench 1, with hearth (118) at left end and dun wall on right side; inset: rotary quern (SF 1.03) in situ in the SW corner of the hearth

- 21 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 20 N-facing section drawings from Trench 1 showing dun wall and courtyard deposits; the postholes shown hashed lines in the top image are depicting their location in the trench, not the stratigraphic relationship

Illus 21 E-facing section of Trench 2 showing dun wall, possible intramural space and interior deposits - 22 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Posthole alignments the north side of the entrance. To the east side of the entrance passage, posthole (149) In Trench 1, a number of postholes and a contained two fills with mixed charcoal beam-slot were revealed at the lowest fragments and a small amount of burnt bone horizons inside the entrance passage and flecks. The opposing posthole (188) contained inside the dun interior (Illus 22; 23). One, some packing stones, mixed charcoal possibly two, postholes were uncovered fragments and a larger amount of burnt bone within the interior of the dun in Trench 2. fragments. Although some of the postholes contained in situ material, many of them showed signs of The two sets of opposing postholes in the disturbance from either removal of disused entrance were similar in terms of their posts, reinsertion of new posts and re-cutting position within the passage and their for new postholes (Illus 25). dimensions, varying from 0.60-0.65m wide and 0.65-0.75m deep. Each one was partially The entrance passage would have contained a covered by the entrance passage walls (166) timber structure, as represented by opposing and (167) and each one contained some postholes (122)/(124) and (149)/(188). On the fragments of burnt bone. This is in contrast to north side of the west end of the passage, postholes (127) and (131), which measured posthole (122) contained packing stones, 0.20-0.25m across and contained no burnt some of which were still in situ, a significant bone. The interpretation is that the larger amount of oak charcoal fragments and a small postholes (122)/(124) and (149)/(188) amount of burnt bone flecks. The fill of represented a timber structure during a opposing posthole (124) was nearly identical; secondary phase of construction, as the however the west side of it had cut through northwest posthole (124) had cut through posthole (127), a smaller slot which contained small posthole (127). Analysis of the burnt oak timbers, one of which ran from the environmental samples and radiocarbon dates west side of the posthole into the beam-slot appear to support these results. (135). The slot, which measured 1.7m long and 0.5m wide and 0.12m deep connected to On the inside of the dun interior face, posthole (122) on the south side of the excavation below the later wall (193) revealed entrance. a wide pit (151), containing postholes (174) and (191), which had cut through the later The smaller posthole (127) was similar in size destruction layer (168). A second, opposing to posthole (131), which contained small pit (186) was uncovered at a similar horizon amounts of charcoal on the south side of the on the south side of the trench. Opposite the entrance’s eastern end that measured 0.2m pits, to the east inside the courtyard, two across and 0.3m deep. The two post slots postholes, (160) to north and (164) to south, seemed to be associated, based on their size were uncovered below the ash layer (121). and placement, and possibly represented an Disturbance from tree root systems had earlier timber structure inside the dun caused difficulty in the excavation of the entrance. In addition, it was recorded that on features. Pit (151), measuring 1.2m long N-S the opposite side of the entrance from by 0.95m wide and up to 0.7m deep overall, posthole (127), below the later stonework contained a complex of fills and evidence for (119), that a dark, charcoal-flecked patch was two postholes (Illus 24). The primary posthole visible, but was not investigated. It is possible (174) lay directly below the later wall (193) that this material filled a small posthole linked and contained a setting of packing stones at to (127) and (131). It is not known if there the top and mixed charcoal fragments with had been another posthole opposing (131) on some burnt bone in the fill. The secondary the north side of the passage due to the posthole (191), which cut through (174) on presence of a large tree stump and root system the north side, contained a few packing that was not removed but caused substantial stones, mixed charcoal and some burnt bone difficulty for excavation at the eastern end of fragments. - 23 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 22 Post-excavation plan of Trench 1, showing posthole locations

Illus 23

Looking south over the entrance passage, showing (from left to right) postholes (186, 131, 188 and 122) - 24 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 24 Posthole (191) at the start of excavation (left) and pit (151), post-excavation (right)

Illus 25 Section drawings of posthole features from Comar Wood dun

Due to the sequence of excavation of (151) pit comprised a lower deposit with some overall, the section drawing does not fully charcoal flecks and at least two packing depict the settings of packing stones described stones (154) and an upper deposit (187) above. The later posthole (191) respected the contained packing stones, mixed charcoal and later wall (193), having been inserted through burnt bone fragments. As tree roots had the burning layer (168) on the north side of caused disturbance to the feature, it appeared the wall. that stonework from collapsed wall (110) had become embedded in the top, after post Posthole (186) on the south side of the removal. The pit was interpreted as the entrance, measured 1.2m long N-S by 0.75m remains of a posthole that had been re-cut by wide and 0.68m deep. The mixed fill of the a secondary post slot. However, due to the

25

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis disturbance, there were no cut boundaries dates obtained from two charcoal samples identified. The size of the feature, together selected from postholes (127) and (188) with its position opposing pit (151), does concur with the above interpretations. support this interpretation. The two features therefore appear to represent four individual Enclosure wall postholes from two different phases of a post structure inside the dun. The enclosure wall surrounding the dun terminated at two points against the edge of To the east of pits (151) and (186), two the steep outcrop on the southeast side of the opposing postholes, (160) and (164), were site. Trench 3 was positioned to assess the located. The two postholes were comparable, entrance through the enclosure wall, located measuring 0.45m across and up to 0.6m deep on the western side of the site, opposite and and with fills containing displaced packing slightly offset to the south from the dun stones and mostly oak charcoal fragments, entrance. The enclosure wall measured 2.5m while only posthole (160) contained some across and survived to a maximum height of burnt bone fragments. Considering the 0.75m high. The extent of the rubble suggests location of and size of postholes (160) and it may only have stood 1.5m high. Tree root (164), they could relate to a first phase disturbance was prevalent as with elsewhere interior structure, and might be associated on the site and, unfortunately, a large tree had with posthole (174) next to the dun inner wall. disturbed the north terminal of the entrance. It is possible that the two sets of post slots Despite this, the base layer of the external represent two interior rings of post settings, face of the enclosure wall survived along with but there is no evidence other than the circular a section of wall that may have formed a door shape of the dun, to support this. check or entrance cell (Illus 26; 27).

Finally, in Trench 2 at the base of the The wall construction was similar to the dun. archaeological layers inside the dun a small The battered interior face (306) comprised posthole (219) was uncovered. It measured large cobbles and stones packed against the 0.35m wide and 0.35m deep while a second wall core, while the exterior face (305) possible posthole (221) next to it measured comprised mostly slab base boulders set into 0.25m wide and 0.3m deep. the subsoil. Excavation against the wall faces revealed that the enclosure had been built Although there were some similarities in over a layer of smaller stones within the content of the posthole fills (discussed further subsoil which contained a small amount of below), it is the dimensions and locations of charcoal flecks (309). This layer appeared the features that was most informative in similar to the charcoal-flecked subsoil below working out relationships. The context of the the dun wall in Trench 5. posthole fills cannot be taken as material directly connected with the structures Although a large tree stump and its root represented, due to the insertion and removal system had disturbed the south side of the that would have taken place, particularly enclosure entrance, one course of surviving when considering later pit re-cuts. As Ramsay stonework (310) on the northwest side of the concludes (Appendix 2), all of the posthole wall provided evidence for a wall feature fills contained mixed remains of various associated with an entrance. The alignment of wood species, although in some cases oak boulders, set in from the inner/outer wall appeared in higher quantities. In addition, as faces formed a 1m-wide section of wall, discussed by Ballin Smith (Appendix 1), which may have been part of an internal cell scattered artefacts were recovered within the or door check for the entrance. The similarity fills of various postholes, suggesting as in construction technique and location of the Ramsay does, that the materials are residual enclosure entrance opposite the dun entrance and could have come from any of a number of suggests outwork is contemporary with phases. Despite these caveats, the radiocarbon construction of the dun. - 26 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 26

Plan of Trench 3, showing the enclosure wall with entrance structure (310) to right

Illus 27

Looking SSE over Trench 3, showing the enclosure wall and entrance structure (310) in centre front

Later structures be built. Within Trench 4, a small post setting Trenches 4 and 6 were placed over Structures (405) was uncovered in the east corner of the 1 and 2 on the southeast side of the dun to building and the remains of small paving evaluate the date and function of what stones (402), some of which had been appeared to be later buildings constructed disturbed by tree root movement, were within the collapsed stone from the dun. uncovered on the floor. The floor surface deposits revealed evidence for burning (403), Both buildings utilised the dun wall as their including reddened and charcoal-rich patches northwest wall (Illus 28; 29), while the with some fire-cracked stone. southeast wall of Structure 1 was abutted by Structure 2, which was the later of the two to

27

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis The depth of the deposit varied across the the construction of the buildings. The location trench, being deepest in the centre and to the of post setting (405) seems to suggest that it east with a subcircular, darker patch in the formed part of Structure 1, although it was centre of the trench. While there were no filled with a charcoal-rich deposit interpreted paving slabs uncovered in Trench 6, the as (403). However, it could be that this was amount of collapsed stone, indicated that the redeposited material after the post removal. outer wall of the dun must have survived to a There was no artefactual material recovered to greater height at the time of the building’s suggest a date for the buildings. It could be construction. surmised that the southwestern building served a different function due to the presence The walls of both buildings sat on layer (403), of slabbed flooring, in contrast to the bare which suggests that, whatever activity floor of the northeast structure. produced the burning evidence, it pre-dates

Illus 28

Plan of Trench 4 and Structure 1 (left) and Trench 6 and Structure 2 (right)

Illus 29

Looking NE across the interior of Trench 4, with the outer dun wall face visible on the left side of the image

28

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Tree disturbance root systems of Douglas Fir are relatively small compared to the standing elements of During the measured survey, the location of the tree; they generally form shallow root 210 tree stumps, mostly Douglas Fir with plates with little in the way of tapping roots. some Silver Birch, were recorded on the site. Therefore, much of the disturbance inflicted Although it appeared no ploughing had taken on the underlying archaeology has been place, some of the tree root systems were confined to the upper horizons. Certainly, the extensive, although relatively small when soft-felling technique employed by the compared to the height of the harvested trees. foresters was key to the preservation of the A significant number of trees had been quality of the monument. It was also fortunate planted within the inner courtyard of the dun, that no significant ‘wind-throw’ events had within the entrance passage, over the walls taken place in the past. Under such and enclosure wall and the later buildings. circumstances, the damage to any underlying According to Forestry Commission personnel, archaeology would most likely have been thinning of the Douglas Fir crop would have catastrophic. As well as the tree planting and taken place on several occasions, requiring felling, the poor construction quality and mechanical machinery, prior to the site being subsequent collapse were major contributing soft-felled in 2013. This later methodology factors in the difficulty encountered in was used in order to minimise any further unravelling the site during excavation. disturbance to the archaeological structures and deposits. Analysis and Interpretation Significant sections of standing wall had been reduced to the foundation stones and in some Post-excavation results instances even these large boulders had been displaced. It was obvious during excavation The excavation recovered only a small that one the reasons for the failures of the dun amount of artefacts, burnt bone fragments and outer wall were due to poor preparation of the charcoal samples, while flotation of sediment ground, poor stone quality, and low quality of samples recovered small quantities of construction methods. However, movement of environmental material and a few further mechanical vehicles over the site would also artefacts. Fortunately, the analysis of the have caused damage, evidence for which was materials, which came mostly from layers noted over the mostly-flattened enclosure spread across the dun courtyard and entrance wall. passage, and had infiltrated postholes and pits, has provided important information to Further destruction was caused by invasive aid the interpretation of the excavation results. tree root systems. Over time, as the roots The results of both the faunal and grew and expanded, they infiltrated walls and palaeoenvironmental assessments revealed the soils, probably encouraged by rich deposits in use of local woodland resources and the certain areas. Root movement had caused the presence of domesticated livestock on the site. collapse of structural elements and the mixing Analysis of the small finds concluded that of stratified layers, and this was particularly although many of the items could be noticeable in Trench 1 at the east end of the associated with manufacturing activities, the entrance passage where the stumps of five scarcity of material recovered could reflect trees had been recorded. In this area, the roots limited occupation of the site, possibly had disturbed what may have been an area of suggesting it had a specific non-domestic use. paving within the entrance and walling on the north side of the passage. Ramsay’s assessment (Appendix 2) provides a detailed analysis of the carbonised remains Considering the tree root damage identified in from bulk samples of the pits and postholes some areas of the site, much of the excavated on the site along with selected archaeology survived remarkably well. The samples of all stratified layers. While the - 29 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis evidence points to oak being the main timber The other posthole fills and surface deposits used for internal structural support posts, at contained mixed species with no single wood least during the first two phases of building, type dominating. This is not surprising given birch and hazel appear in significant the interpretation of two burning events on the quantities and were probably also utilised for site, associated with multiple periods of structural purposes. posthole re-use and restructuring of the dun. Carbonised remains of the local woodland Opposing postholes (122) and (124) at the species included oak, birch, hazel, alder, west end of the entrance contained identical willow and Scots pine. Other carbonised charcoal assemblages, dominated by a material present in the samples included very considerable amount of oak charcoal, while small amounts of mammal bone fragments, some birch was present too. Also inside the hazelnut shell, barley and indeterminate cereal entrance, the smaller postholes (127) and grains. The presence of burnt cereal grains (131) contained large amounts of oak charcoal (with at least barley present) and hazelnut with some birch present. Within the shells has provided a hint at food courtyard, opposing postholes (160) and (164) consumption. As does the small quantities of also contained large quantities of oak burnt animal bone fragments, which were charcoal, with significant amounts of birch identified by Smith (Appendix 3) as and hazel charcoal also recovered from the fill containing at least domesticated cattle and of posthole (164). The increased oak charcoal sheep or goat, with evidence for butchering content in the above postholes was recovered shown by the presence of knife cutmarks on a only from postholes that had not been rib bone (Sample 29) from the hearth ash disturbed by later recuts – with the exception layer (117). Further bone fragments were of small posthole (127), although it had only recovered from the context (169) between the been narrowly truncated by the secondary upper and lower hearth settings. The degraded posthole (124). state of the burnt grain and bone also indicates that the material was subject to Evidence for wattle or hurdle fencing or several periods of burning, not just associated partitions was recorded in the intermediate with hearth use or cooking. This is not destruction layer (168/179) where intensely unexpected given the evidence for at least two burnt deposits contained carbonised hazel substantial conflagration events on the site. roundwood with hazel wood pegs or trenails possibly also present (SF 1.12). In the same It is important to note that in Trench 1, a layer where birch was also significantly pattern there was in the recovery of burnt present, it was noted by Ramsay that birch bone fragments, which appeared in a much timbers might have been associated with a higher quantity from the fills of secondary roof structure. Although the type of wood phase postholes and the upper floor deposits; used for hearth fuel was inconclusive, the ash the highest quantity of burnt bone came out of layer (117) directly overlying the upper hearth the upper horizons (i.e. contexts 111 and 168 slabs contained charcoal dominated by alder, as opposed to contexts 153 and 507). and could suggest that this wood was used However, the spread of material across the primarily for fuel. floor deposits and posthole fills was mixed and conclusions are limited on this basis. A further possible pattern may be present with regards to the lack of willow and Scots pine The small assemblage of stone tools from the charcoal in all posthole fills and deposits site is typical of those from other Iron Age interpreted as part of secondary phases. sites (Ballin Smith, Appendix 1). The Willow charcoal dominates in the small fragments from two different querns (SF 1.02 posthole from trench 2, and is also found in and 1.03; Illus 30) are evidence for food small quantities in the first phase deposits processing on site, and this is supported by (107), (153) and (212), and could suggest that the presence of stone pounders/hammerstones willow was used during phase 1 only. (SF 1.14 and 7.01) found close to the hearth. - 30 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

0 20cm

Illus 30

Quernstones from Comar Wood dun; SF 1.03 (left and centre); SF 1.02 (right)

0 20cm

Illus 31

Central hearth setting (118); left: facing ESE, the re-aligned kerbstone (possible whetstone) in front; right, facing NE, quernstone (SF 1.03) in front - 31 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Illus 32

Post-excavation image of Comar Wood dun, during the 2013 evaluation, showing the location of walling and postholes

- 32 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis The presence of five other stone tools from postholes had been re-cut, with one secondary inside the dun (including pumice, SF 1.06. cut (191) going through an intermediate layer that was brought to the site from elsewhere) of burnt destruction debris (168). These provides some evidence of activities that may factors provided evidence that the dun was have taken place on the site. Pottery occupied and re-used during multiple periods. manufacture, animal skin processing, wood tool working and possibly metal-working The primary dun structure comprised a could be supported by the presence of fired subcircular, double-faced stone-built wall clay waste (SF 1.08), iron slag (SF 1.05 and with rubble core measuring approximately SF 1.11), wrought iron waste (SF 1.05), small 3.2-3.3m wide around the entrance, as defined fragments of bronze sheeting (SF 1.09), and by wall faces (102) and (106), enclosing an shaped hazel roundwood (SF 1.12). However, area approximately 18m N-S by 15m E-W. the material was found in such small amounts On the north side of the dun, the wall that this assessment is not conclusive. The measured 2.1-.2.2m wide with some evidence lack of further finds, coupled with the small for an intramural cell, represented by (201), amount of food plant remains and bone waste, (213) and (214). The entrance passage, could suggest that the site was occupied located on the west side of the structure, sporadically or that it had a specialised measured 1.6m across and may have purpose. While the presence of metal objects contained a slabbed surface (148). A timber and metal-working debris is interesting, there structure, possibly a porch and door frame, is no evidence to prove that this was taking was represented by postholes (127) and (131), place on-site – the material could have been and beam slot (135), which was linked to brought to the dun from another place or it posthole (127) by burnt timber fragments could have been manufactured in an area overlying the reddened, vitrified subsoil outside of the dun, probably during primary between them. There may also have been an occupation as shown by the recovery of the opposing entrance (506) on the east side of items from residual deposits. Of even greater the building. On the inside, a timber structure interest is the lack of ceramic artefacts on the may have supported a roof or other structure, site. as represented by opposing postholes (174) and the primary posthole in (186), and Phasing possibly a further setting of opposing postholes (160) and (164). There would have Due to the poor condition of the monument, been a hearth at the centre of the courtyard, interpretation of the phases remains uncertain. represented by pit (183) or slab setting (176). However, there were many consistent findings uncovered below the chaotic rubble The first phase of the building appeared to and debris. have ended after a fire, represented by a wide- spread burning layer (153)/(212)/(507) and Prior to excavation it was apparent that the the vitrified subsoil (178)/(184) at the same courtyard area was remarkably clear of stone horizon. Associated with this event was a rubble, despite the fact that it had been bank of charcoal-rich material (107), subjected to deep disturbance by tree planting containing heat-affected stone clasts and and root movement. In both Trench 1 and vitrified stone (SF 1.07), piled against the Trench 2, layers of material had been banked inner wall face above layer (153) and the up against the interior wall face while redeposited subsoil packed into the base of structural evidence for later interior walls the wall. Hazel roundwood charcoal from the (110) and (215) suggested the demarcation of burning layer, the lowest archaeological a secondary, constricted courtyard space. On horizon (153) in Trench 1, produced an AMS the outside of the dun entrance, it was clear date of 376-203 cal BC (SUERC-54243) that the outer wall face had been widened while the lowest archaeological layer (212) (1.01) and that an opposing entrance passage overlying the subsoil in Trench 2 produced an (506) had been blocked up. Inside the dun, almost identical AMS date of 371-201 cal BC - 33 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis (SUERC-54232) for hazel roundwood. In say if these changes were all contemporary, addition, these dates are almost identical to they all appear to be secondary to the earlier the AMS date, 382-204 cal BC (SUERC- elements of the site discussed previously. 54242), obtained from a hazel roundwood charcoal sample from the bank of material Birch charcoal from the fill of posthole (188) (107) against the inner wall face. Two further produced an AMS date of cal AD 70-224 AMS dates from charcoal taken from (SUERC-54240) and provides a suggested postholes interpreted as part of the phase 1 date range for the secondary occupation. This structure fell within the same period: willow date range is very similar to one obtained charcoal fragments from posthole (219) in the from material associated with a second dun interior in trench 2 produced a date of burning event. Represented by the 365-184 cal BC (SUERC-54238) and birch intermediate charcoal-rich layer (168/179), charcoal fragments from small posthole (127) this deposit contained carbonised timbers that in the entrance passage produced a date of lay below a later interior wall (110) and over 371-207 cal BC (SUERC-54241). Although the earlier clearance layer (107) and floor posthole fills can contain residual material deposit (121). Carbonised timbers from the after post removal or degradation, the material fire debris (SF 1.13) produced an AMS date from these two particular features was of cal AD 27-212 (SUERC-54247). Posthole homogenous and did not appear to have been (191) may have cut through this layer, which subjected to later disturbance. would suggest that a secondary interior courtyard structure was built after this event; A later phase of the structure included the otherwise this layer may have been disturbed addition of a secondary exterior face to the by the removal of the burnt post from its slot dun, represented by wall face (101), which after the fire, which would relate it to a period increased the wall width to approximately before the event. As with the first phase 3.8m wide, while further stone (119) was destruction layer, the destruction debris added around the outside of the entrance appeared to have been cleared out of the structure constricting the entrance to about courtyard and banked up over the earlier 1.1m wide on the outside of the dun. A layers against the dun inner wall face. second timber structure inside the entrance passage that probably supported wider posts At least two slabbed hearths fit into the is represented by opposing postholes (122) phasing discussed above, with the basal and (124) and opposing postholes (149) and hearth relating to the vitrified surface and (188). On the opposite side of the dun, a burning horizon from the first phase. Birch passage through the wall (506) was blocked roundwood charcoal from the ash layer (117) up, with the blocking stone overlying the first overlying the upper hearth slabs produced an phase burning horizon (507). On the interior, AMS date of 45 cal BC – cal AD 80 a second structure may have been built or (SUERC-54233). This date is very similar to repairs may have been done to an existing the date range from the overlying ash layer one, represented by opposing postholes (191) (115), which produced an AMS date of 41 cal and the secondary re-cut of (186). The dun BC – cal AD 120 (SUERC-54237) from hazel was initially interpreted as a galleried roundwood charcoal. The dates indicate that structure. Although it is difficult to fit into the earlier slabbed hearth (176) and possible phasing, in Trench 2, there was some hearth pit (183) fit into earlier periods, evidence for structural walling inside the dun. probably phase 1, as supported by the There did appear to be a hollow inside the stratigraphic relationship between these primary inner face (213) with possible earlier hearths and the vitrified subsoil (178). entrance flanking stones (214) and further The upper hearth (118) also seems to have stonework (223) next to it. On the basis of the been in use before the second burning event. rough construction of the walling, the possible Therefore the uppermost spread of material gallery space could be related to a secondary (111) and the ash layer (115) present in the phase of reuse. Although it is not possible to courtyard and extending from this hearth - 34 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis probably represent a spread of material during successive periods over the space of present on site during later occupation, as approximately 300-500 years. they are well sealed below later silting and collapse. It may have been at this time that the The archaeological evidence for two burning upper hearth setting was realigned. A large events within which various structural stone, which showed use-wear as a possible changes occurred has shown at least three- whetstone, located at the western end of the four different phases of re-use of the site. The setting appears to have been removed from its fairly clear condition that the courtyard was original position, where it formed the western left in at the time of abandonment may also kerb of the hearth. It was re-aligned on an represent further use after the construction of east-west orientation, in effect opening up the the later walls. hearth to the entrance. This stone was set into the upper ash later (115) and may have been Probably at a much later time, the collapsed moved at the same time as the broken quern structure was re-occupied, as shown by the was incorporated into the hearth. No presence of Structures 1-4. However, there interpretation was made about the alignment were no artefactual or stratigraphic of small upright stones (114) to the south side relationships uncovered to suggest anything of the final hearth, which also incorporated a other than the structural elements were built broken quern. They might have been the against the base of the outside dun wall face. remains the earlier, robbed out hearth setting It is not known if this took place during the (176) or another setting. post-medieval or medieval period, or perhaps earlier, although the shape and size of the The later occupation of the courtyard space buildings could suggest the former. At least, it was indicated by the construction rubble wall seems that the structures, which all appear of (110), which constricted the width of the similar build, pre-date the 1872 Ordnance interior space. Given its rough construction, it Survey, which showed no structures on or was probably built rather quickly as a front anywhere near the site. edge to the earlier debris piles, but still respecting the original entrance passage, as One final intriguing date derived from Scots marked by wall (193) which connected to it. pine charcoal fragments recovered from In Trench 2, a continuation of this later redeposited natural subsoil (508) below the interior wall appears to be represented by wall base of the dun outer wall face where a (215), in the same position as wall (110), second entrance passage was identified. This constricting the space to an area measuring charcoal produced an AMS date of 5740-5645 approximately 11m in diameter. After this cal BC (SUERC-54239), giving a hint of wall was built, another inner bank of material early forest clearance during the Mesolithic was formed on the inside of the walls (110) period, well before the period discussed in and (215) and may represent an even later this paper. period of courtyard clearance. The evidence may suggest continued infrequent and Interpretation informal occupation after the 1st-2nd centuries AD. The interpretation of Comar Wood dun is not without uncertainties. The condition of the The fact that the hearth deposit and overlying site, after at least two dramatic burning ash layer dates fall within the same range, but events, along with the mixing of layers from possibly earlier than the phase 2 destruction re-occupation and considerable tree root event, could indicate that the last period of disturbance, offers good reason to exercise continuous use of the dun took place between caution in the interpretation of certain the mid-1st century BC to the end of the 1st features. Detailed analysis of the fieldwork century AD, while abandonment took place and post-excavation results has shown that after the mid-2nd century AD. Thus, there is Comar Wood dun was occupied throughout evidence to show that the site was in use various structural phases, the relationships - 35 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis between many of which are hard to establish. and would have replaced the earlier one at this Radiocarbon dating, however, suggests that stage. These secondary phases came to an end occupation may have taken place during two during the latter part of the 2nd century AD or distinct periods in the Iron Age. 3rd century AD. Hurdle or wattle panels also appear to have been used inside the courtyard, The primary dun was built prior to the first re-using the post slots from the first structure. burning event, represented by the lowest layer Following this period, another re-occupation of burnt material spread through the entrance is represented by the construction of interior and interior of the courtyard and abutting the rubble-built walling while even later inner wall. Five consistent AMS charcoal clearance of interior debris was banked dates, places this time around the 4th-2nd against this wall face. The hearth may have centuries BC. Prior to this event, the first continued to be used at this stage until phase dun comprised an entrance passage abandonment. containing a timber structure on the west side of the site with possibly an opposing passage The chronology for the dun use falls within a on the east side of it, probably with a timber period of approximately 600 years, and structure inside the courtyard. The structure evidence for various phases is demonstrated was probably built no higher than 1.5m, with by the sequence of radiocarbon dates (Table double-skinned walls partially revetted into 1) falling in the middle to late Iron Age. the subsoil. An interior cell may have been Interestingly, the only evidence for earlier built into or against the northern dun wall, occupation or activity is 5740-5645cal BC possibly during this period but possibly later. (SUERC-54239), a date derived from pine The defensive enclosure wall, measuring charcoal, probably the result of forest about 2.5m wide and probably standing to clearance during the Mesolithic Period. 1.5m high, ended abruptly at the edge of a steep knoll and to the east and was entered The period of later occupation of Comar through a passage opposite the dun entrance. Wood dun, in the form of attached structures This outwork may have contained a door or to the monument, is unknown. There were no guard cell. secure deposits to provide a date as to when the two later buildings attached to the During the secondary phases of use, the fire southeast side of the dun, and the third built and occupation debris was raked out and piled into the enclosure wall southwest were against the interior wall face of the dun. The constructed. A small cell built into the entrance passage was elongated and a new enclosure wall on the northeast side, is timber structure erected inside it. This may be probably contemporary with the three the same time during which the opposing buildings. The subrectangular shape of the eastern passage was blocked up. A new three structures and the interior post-slot in timber structure was built during this phase or one, could indicate a medieval or post- the next in approximately the same location medieval use.

Table 1 Radiocarbon Dates from Comar Wood Dun

Calibrated Sample Context Calibrated 1- Delta13 Lab Code Description Uncal (BP) 2-sigma No. No. sigma (68.2%) C% (95.4%) SUERC- Charcoal : Corylus cf avellana roundwood 54232 002 212 12 rings; lowest occupation horizon in 2213 ± 30 359-208 BC 371-201 BC -25.9% (GU34494) Trench 2 SUERC- Charcoal : Betula sp roundwood; sample 54233 003 117 from directly above the hearth setting (118); 1976 ± 30 19 BC-80 AD 45 BC-80 AD -27.9% (GU34495) relating to hearth use

- 36 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Calibrated Sample Context Calibrated 1- Delta13 Lab Code Description Uncal (BP) 2-sigma No. No. sigma (68.2%) C% (95.4%) SUERC- Charcoal : Corylus cf avellana roundwood 54237 007 115 15-20 rings; sample from upper ash layer, 1962 ± 30 5-71 AD 41-120 AD -26.3% (GU34496) over Context 117; relating to hearth use SUERC- Charcoal: Salix sp; fill of post-hole [219] in 54238 024 220 Trench 2 - possibly part of inner post-ring in 2197 ± 30 357-184 BC 365-184 BC -26.4% (GU34497) dun, under Context 212 SUERC- Charcoal: Pinus sylvestris type; subsoil 5740-5645 54239 026 508 layer below the blocked-up passage (506) 6815 ± 30 5724-5672 BC -27.0% BC (GU34498) and burnt layer 507 SUERC- Charcoal: Betual sp; sample from fill of 54240 036 189 1875 ± 30 78-209 AD 70-224 AD -26.5% post-hole (188) (GU34499) SUERC- Charcoal: Betual sp; fill of posthole (127), 54241 045 128 2213 ± 30 359-208 BC 371-201 BC -25.6% support post on N side of entrance porch (GU34500)

Charcoal : Corylus cf avellana roundwood 3 SUERC- rings; relating to burning event at end of 54242 059 107 2228 ± 30 365-210 BC 382-204 BC -26.6% Phase 1; sample from sealed layer below (GU34501) carbonised structural timbers (168)

Charcoal : Corylus cf avellana roundwood 5 SUERC- rings; lowest occupation horizon in Trench 54243 060 153 2221 ± 30 361-210 BC 376-203 BC -25.8% 1; sample from sealed layer inside the (GU34502) courtyard against the inner wall face

SUERC- Charcoal : Corylus cf avellana roundwood; 54247 SF1.13 168 burnt timber withy/hurdle fragments related 1903 ± 30 69-129 AD 27-212 AD -28.7% (GU34503) to second phase collapse of dun

Discussion and conclusions The site at Comar Wood lies close to the watershed dividing the Atlantic Scottish west The positioning of Comar Wood dun at the coast from the Central and Eastern Highlands edge of a steep-sided knoll overlooking the and there are many similar site-types in the strath leaves no doubt as to the strategic glen and wider landscape. Given its central positioning of the site. The structure would location, and the interpretation of the site, it have been a permanent site in the landscape poses the question of whether Comar Wood for centuries, possibly marking guardianship dun was actually a gathering place, a over the land. The permanence of the monument located centrally to various tribal monument was proven during the evaluation, groups that visited it on occasion for showing that clear phases of occupation and ceremonial reasons. The fact that so few re-occupation took place within the same artefacts were recovered and such little structure, a substantially-built stone building domestic debris was encountered may support with a substantial wall enclosing and this hypothesis. Also, such an interpretation defending access to it. It was built to be seen, might offer an explanation as to why the fire but not to be accessed freely. The overall debris was banked up against the inside of the landscape position (Illus 5) places the site structure rather than being cleared out centrally in the glen that connects the east completely as would be expected in a coast at the Beauly Firth and west coast at permanent settlement structure. Loch Duich. As such, its position could have placed it prominently within territorial or Although portrayed as simple in form, the landscape boundaries in prehistory. There is duns are a very diverse class of monument, still much to be learned about this site-type possessing a variety of ground plans which and other so-called duns or forts, of which include the possible roofed ‘dun-houses’ as there are many that overlook the same glen. well as the much larger ‘dun enclosures’ which were almost certainly unroofed and - 37 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis were more like the much later Irish ring forts with walls averaging 5m thick, contained (Harding 1984). A good interpretation for this evidence for being roofed, contained a timber site would be that it was a dun-house – a entrance structure and was refurbished and re- substantial monument that is too small to be a occupied after a primary burning and fort and could have been roofed – with a vitrification event. The surviving walls had defensive enclosure. The location was been well-built, when compared to those deliberately selected as was the adoption of an recorded at Comar, but the layout of the enclosure wall – strongly supporting the case entrance passage at Langwell, including the for a defensive structure with a display of locations of the postholes, is remarkably prominence. The longevity of use is also similar to that seen at Comar Wood. Similarly clear, with groups repeatedly re-visiting and to Comar Wood phase 2, the substantial re-using this monument, a culture reinforcing postholes were recessed into the stonework in their presence on a landscape in some way, be the entrance, and may have provided the it for defence or display of power, both of structural support for horizontal timbers to which must certainly be linked. Comar Wood, support the stonework, whilst also being along with most of the other prehistoric inserted prior to the entrance stonework being settlement sites in the area, overlooks the laid (Nisbet 1996). The dun at Langwell did fertile river plains where the most reliable contain a well-built guard cell running off the agricultural ground was located. Firmly entrance passage, but no other intramural located within these agricultural landscapes, it spaces were identified or excavated. is possible to see the enclosed dun at Comar Interestingly, the vitrification of the subsoil Wood as overlooking and being embedded in was described as reddened to a depth of more these resources. than 15cm (Nisbet 1996, 55). Although this was not as deep as at Comar, the vitrification It has been suggested that the adoption of did infiltrate the subsoil to a depth of 2-5cm enclosure was a deliberate choice. Armit and and supports the burning of some collapsed Ralston (2003, 193), for example, suggest that timber structure. Similarly to Langwell dun, a perceived trend to enclose a space could the occupation deposits surviving inside were have been associated with factors such as an shallow, containing very small fragments of increasing emphasis on pastoral farming bone and no pottery. Bone from a midden brought about by climatic deterioration, or a spread below the outer wall also contained result of social change. The construction of sheep and cattle, the only two species of enclosing works could also be associated with which were identified at Comar Wood. In the a wide potential range of practical and secondary phase of Langwell, they did not symbolic meanings (Collis 1996; Ralston repair the wall prior to re-occupation, but 2006, 10-11), such as defining communities, constricted the space after clearing the as social defences, displaying status or entrance completely, also banking up deposits isolation, and as expressions of power through around the foundation stones of the wall but the mobilisation of labour (Hingley 1992, 39). keeping the central area clean. Radiocarbon Perhaps, it is possible that the occupants of dates taken on samples from the site produced these enclosed sites, including any associated results ranging from 500 cal BC to cal AD monumental buildings, were displaying 140, although one sample did return a identity, prestige and independence (Hingley surprising date of cal AD 910 to cal AD 964. 1992, 14-17; Armit 1997c, 27), although this remains an assumption based on current Another site in the West Highlands of models of Iron Age society. Scotland showing some similarities to Comar Wood is that of Rahoy in Morvern, Argyll, A close parallel for the type of site evaluated excavated by Childe and Thorneycroft (1938). at Comar Wood is a vitrified dun at Langwell This roughly circular stone-built dun had also in Strath Oykel, Sutherland, excavated in been heavily vitrified, which presented 1973 and 1974 (Nisbet 1996). Langwell dun, difficulties for the excavators in finding measuring between 15 and 15.5m internally surviving elements of the inner and outer - 38 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis walls. The structure measured approximately similar evidence for construction and re-use 21m externally and 13m internally, giving a during the same periods as Comar Wood dun, wall thickness averaging 4m. The structure particularly Tarlogie roundhouse, constructed included a large rectangular central slab-built c.360-170 BC and re-used c.AD 25-130 and hearth, but as with Langwell and Comar again c. AD 240-330 (Hatherley pers comm). Wood, few small finds were recovered. The Comar dun certainly shares many of the At Loch Glashan, Argyll, in 2003, a single characteristics of the dun at Langwell, and to evaluation trench was taken over the wall of some extent that at Rahoy. However, the the Atlantic roundhouse or dun courtyard in walls of the structure at Comar Wood display an effort to assess the damage caused by a relatively poor construction technique vegetation and to obtain dating evidence. The compared to Langwell and did not contain Atlantic roundhouse measured 28m in any timber lacing. Langwell was also external diameter, with the main wall preceded by a small, elongated, multivallate approximately 3.5m wide (Gilmour and fort, of which at least the inner circuit was Henderson 2011). As at Comar Wood, Loch timber-laced; Comar Wood produced no Glashan contained a secondary rubble-built evidence for earlier structures or settlement, internal wall parallel with the main wall and other than Mesolithic forest clearance. post-dating the main phase of occupation. Also at Loch Glashan, between the later It has been argued that sequences of structures rubble wall and the main wall inner face there such as those uncovered at Langwell, support was a deposit containing burnt material that MacKie’s (1976, 223-4) interpretations for abutted the main wall face overlying the the ‘miniaturisation’ of hill forts, which subsoil. This context produced two probably pre-date both plain and galleried radiocarbon dates of 393-205 cal BC duns (Nisbet 1996, 67-8). However the (SUERC-5478) from charred grain and 363- overall relationship between forts, simple 108 cal BC (SUERC-5478) from a hazelnut duns, galleried duns and brochs remains to be shell. A layer below the later wall, produced considered. similar radiocarbon dates from hazelnut shells, while the context itself yielded a The re-use and re-building of structures is yellow opaque glass bead typologically dated consistent with other Atlantic roundhouse to between the 2nd century BC and 2nd century sites, Langwell, Loch Glashan and Tarlogie AD (ibid). Similar to Loch Glashan, at Comar being just a few examples of this. Another Wood dun there was a bank of burnt material site, the complex roundhouse of Applecross against the primary inner wall face at a similar Broch, is also a prime example (Peteranna horizon associated with primary occupation. 2012). Comar Wood dun contains other As at Loch Glashan, this layer was located similar features to Applecross Broch, such as between the inner wall and a later, concentric quernstone re-use in slabbed features, unusual interior wall. Hazel charcoal from the layer at angled stacks of stone and rectangular hearth Comar Wood produced a radiocarbon date of settings with small upright stone kerbs. 382-204 cal BC (SUERC-54242), similar to However, in stark contrast to Applecross those produced from the Loch Glashan Broch, the excavations at Comar Wood failed deposit. to produce any dateable small finds. This is also reflected at sites such as Langwell, In 2013, a structure of similar proportions and Rahoy, Loch Glashan and Easter Rarichie. In morphology to Comar Wood dun was particular, Loch Glashan and Comar Wood excavated at Easter Rarichie on the Tarbat were both noted as aceramic, although fired Peninsula in the northeast Highlands. Located clay from Comar Wood may represent a by- on a heavily fortified hill, it was built and in product of ceramic manufacture. use from c.400-200 BC (Hatherley pers comm). Further excavation by Hatherley in Did people clean out their belongings in a 2014 in the same area has produced some methodical manner after abandoning such - 39 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis sites, or is the lack of objects relating to their for structural development and regional material culture providing evidence to suggest models. However, it has shown that such that these sites were only used temporarily; ‘keyhole’ investigations, designed to assess for social gatherings or as places to retire to in archaeological potential and recover times of strife? Such circumstances make it information to a pre-determined plan, can difficult to assign any particular function to provide valuable information and usefully these structures, whether domestic or increase the current corpus of knowledge on defensive in nature. At the same time, two such structures. A carefully constructed quernstones were found reused within the research design and effective trench hearth setting at Comar Wood, incorporating placement gained the optimum amount of disused domestic tools into a practical living information from a small area. area. This is a practical and simple reuse, commonly seen in floor surfaces and paved The evaluation also proved that it is difficult areas of Atlantic roundhouse sites (Mackie, to assign a monument type based on field 1971-2). The reuse of these sites and lack of survey alone; such monuments should be considerable artefactual material suggests that approached with an open-mind and can only these sites were nothing to do with elite be truly categorised through excavation. It settlements, but probably related to more was shown that what was thought to be a autonomous farming communities well-preserved, galleried dun was, in fact, establishing a presence and control over more like a monumental roundhouse of poor territory. The reuse indicates successive preservation. groups returning to the same site, although, for what purpose, it is unclear. It does seem apparent that the second phase of rebuilding involved more careful reconstruction, while the third phase was a rather rapid constriction of the space, and may not have even contained a built structure. Perhaps the later reuse, which appears to have taken place after the start of the 1st millenium AD, reflects changes going on within the culture of the period, social change in progress.

Like most of the fort, dun and broch sites distributed along the length of Strathglass, the monumental roundhouse at Comar Wood was built in a location to take full advantage of an extensive view shed; whether to work on established lines of sight between contemporary structures, to provide a defensive location, or to represent a symbol of status within the wider landscape. With its impressive dimensions, including a possible towering conical roof, contemporary outworks and its location set on the edge of rocky knoll whose slopes fall quickly towards the valley floor, the monument would certainly have displayed identity and prestige.

Further excavation of this site and other similar sites would certainly reveal much more information to establish a chronology - 40 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis References

Armit, I., 1997. Celtic Scotland. London: Batsford.

Armit, I., 2003. Towers in the North: the Brochs of Scotland. Stroud: Tempus.

Armit, I., 2005. ‘Land-Holding and Inheritance in the Atlantic Scottish Iron Age’ in Turner, V.E., Nicholson, R.A., Dockrill, S.J. and Bond, J.M. (eds) Tall stories? 2 millennia of brochs. Lerwick: Shetland Amenity Trust. 2005, 129–143.

Armit, I. and Ralston, I. B. M., 2003. ‘The Iron Age’ in Edwards, K.J. and Ralston, I.B.M. (eds) Scotland after the Ice Age. Environment, Archaeology and History, 8000BC – AD1000. 2nd ed. Edinburgh: Edinburgh University Press. 2003, 169-193.

Birch, S, Fraser, L & Peteranna, M 2014 ‘Comar Wood Dun, Cannich, Strathglass: Data Structure Report’. Unpublished Report

Childe, V.G., and Thorneycroft, W., 1938. The vitrified fort at Rahoy, Morvern, Argyll. Proceedings of the Society of Antiquaries of Scotland, 72, 23-43.

Collis, J., 1996. 'Hill-forts, enclosures and boundaries', in Champion, T.C and Collis, J.R., (eds) The Iron Age in Britain and Ireland: Recent Trends. Sheffield: Department of Archaeology and Prehistory, University of Sheffield. 1996, 87-94.

Gilmour, S and Henderson J 2011. “A 1st millennium BC Atlantic Roundhouse in Argyll: survey and excavation at Loch Glashan” in Proceedings of the Society of Antiquaries of Scotland, 141: 75- 102.

Harding, D.W., 1984. ‘The function and classification of brochs and duns’, in Miket, R. and Burgess, C. (eds) Between and Beyond the Walls. Edinburgh: John Donald. 1984, 206 -220.

Hatherley, C., 2014. Research post-graduate at Aberdeen University.

Hingley, R., 1996. Ancestors and identity in the later prehistory of Atlantic Scotland: the reuse and reinvention of Neolithic monuments and material culture. World Archaeology, 28(2), 231-243.

Mackie, E 1971-2. “Some new quernstones from brochs and duns” in Proceedings of the Society of Antiquaries of Scotland, 104: 137-146

MacKie, E.W., 2007. The Roundhouses, Brochs and Wheelhouses of Atlantic Scotland c.700 BC- AD 500: architecture and material culture, the Northern and Southern Mainland and the Western Islands. BAR British series 444(II), 444(1), 2 V. Oxford: Archaeopress.

McAdams, N., 2013. Forestry Commission employee.

Nisbet, H., 1996. Excavation of a vitrified dun at Langwell, Strath Oykel, Sutherland. Glasgow Archaeological Journal, 19 (1994-95), 51-73.

Peteranna, M 2012 Applecross Broch Community Archaeology Project 2006-2010. Unpublished data structure report.

- 41 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis RCAHMS 1942-3 Emergency Survey 2v, Typescripts, Page(s): INV 53 Held at RCAHMS A.1.1.SUR

ScARF 2012. Iron Age Scotland: ScARF Panel Report. Edinburgh: Scottish Archaeological Research Framework. Accessed 23/01/14 at http://www.scottishheritagehub.com/sites/default/files/u12/ScARF%20Iron%20Age%20Sept%2020 12.pdf

- 42 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Appendix 1

Comar Wood Dun, Cannich, Strathglass Artefactual Analysis Beverley Ballin Smith

INTRODUCTION The 2013 archaeological work at the dun in Comar Wood produced a small range of artefacts: stone tools and samples, fired clay, metal and metal working debris and organic objects (see Appendix 1). The stone finds included rotary querns, pounder/hammerstones, polishers/rubbers, a possible hone and pumice pieces. Most of these artefacts demonstrate the utilisation of the local geology - mica schists, quartz and sandstone. The querns are the most sophisticated pieces of the assemblage, requiring skill in their manufacture and also maintenance during their use. The other tools are largely expedient use of cobbles and pebbles. The only anomaly in this assemblage is the presence of pumice, which has been brought from the coast to the site. Samples of vitrified material formed during the firing of the dun, indicate the temperature that must have been generated during the process. Other finds include one wrought iron object, a little metal slag and two very small pieces of copper-alloy. There are in addition rare organic pieces of carbonised hazel roundwood, which complete this assemblage. The artefact types are typical of the Iron Age period and are considered contemporary with the occupation of the dun during phases 2 and 3, dated to approximately the middle of the first century AD.

METHODOLOGY All measurements, weights and descriptions of each find and sample were entered into a prepared database in spreadsheet format. Each artefact was described, with identification of its geology or material where possible. For stone, the worked surfaces, ends, edges, and pecked or hammered areas were examined and recorded according to accepted procedures; two metal objects were X- rayed, and the organic objects were examined microscopically. The Standard and Guidance for the collection, documentation, conservation and research of archaeological materials (IfA, 2008) was followed.

WORKED STONE The 12 stone artefacts located during the excavation fall into five categories (see table below). There are in addition three unworked stones. The artefacts exhibit the use of a variety of stone resources from the complex geology of the area. These including the local metamorphic rocks: micaceous schists of various types and hardness, quartz, quartzite and sandstone. Another identified rock is granite possibly from a boulder from glacial or alluvial deposits along the River Glass. The only piece that is imported is pumice, possibly a single large cobble. This was likely to have been brought to the site from either the west or the east coasts as the site is situated approximately half way between the Sound of Sleat and Loch Alsh to the west, and the Moray and Cromarty Firths to the east. No other non-local stones were found during the excavation.

SF Nrs Type Number 1.02 and 1.03 Querns 2 (total of 9 fragments) 1.14 and 7.01 Pounder/hammerstones 2 0.01, 0.02, 1.10 and 2.04 Polishers/rubbers 4 7.02 Possible hone 1 1.06 Pumice 4 fragments 1.04, 2.03 and 2.05 Unworked 3

Table 1: Stone artefact types

- 43 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Querns Two querns were found in close association with each other within a setting of hearth slabs (context 118) and within a stone setting (context 114) that extended the south side of the hearth. Both stones are fragments of upper stones of rotary querns and neither of the completed stones was particularly regular in shape.

The largest, SF 1.03 (Figure 1) comprises eight fragments of a relatively soft but coarse mica-schist that is prone to lamination, which together accounts for about 75% of the completed stone. Six of the eight fragments fit together, and the remaining two also join, but not to the rest. When found it had a substantial tree root growing through it that may have caused its fragmentation (Birch et al 2014, Plate 12). The quern probably broke during use and was discarded to be reused in the kerbing by the hearth. Part of edge of the stone is missing and the edges of the largest fragment have also broken away. The largest piece contains the central splayed perforation, which in the reconstructed quern is not central. Given the type of rock the stone is made from it is quite likely that it was reshaped during use when its edges chipped or broke away. It is estimated that the stone was c. 540 mm in diameter.

The upper surface of the stone is convex with rounded edges, where they survive, that vary in thickness from 31 to 54 mm. The lower, working surface retains evidence of pecking that roughened it not long before it was discarded. Several pieces of the reconstructed quern indicate that near its edges, the generally flattened worked surface became slightly dished or concave through use. Close to the perforation the working surface was, however, slightly domed. The perforation is splayed on both surfaces measuring c. 50 mm in diameter. A handle slot or hole was not present in the surviving pieces. The upper surface is slightly pink in colour in areas, indicating that it might have been burnt prior to its reuse in the hearth kerbing/paving. When found its lower surface was uppermost.

Figure 1: Reconstructed quern SF 1.03

The other quern SF 1.02 is tooled from granite and broken across the spayed central perforation. This slightly dome-shaped stone reduces in thickness towards its edges from a maximum of from 50 mm at its centre. The edges of the stone are rounded and the upper surface is worn through use. The central perforation is splayed and would have measured c. 60 mm in diameter. In addition, it has a small but intact splayed handle perforation, 30 by 32 mm in diameter that is positioned close to one edge. The base of the stone is mostly flat, but it is slightly hollowed around its central perforation and worn towards its surviving edge. Like the previous stone, the base was roughened by pecking not long before it broke and was discarded. The overall diameter of the quern was c. 380 mm.

- 44 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Pounders/hammerstones Two stones were used as pounders or pestles. SF 1.14 is a large quartzite cobble, 1130 g in weight, which has both ends facetted through use by pecking or pounding. One end also has a flaking scar though its use as a hammer. It was found in a bank of dumped material against the inside wall of the dun.

SF 7.01 is a quartz chunk or short bar, 365 g in weight, which was trimmed to fit the hand. The shaft has smoothed edges, even though the faces of the stone are rough. Both ends of the tool have facetted wear caused by the stone being used for pecking or pounding. Both ends have two worn facets. The tool was found in the top of a disturbed occupation context just below the topsoil in the centre of the dun.

Polishers/rubbers Four rounded water-worn stones, two of quartz and two of quartzite, have been identified as rubbers or polishers. SF 0.01 is a pebble, which is domed on one face and flat on the other. The middle of the flatter face has a smoothed area c. 45 by 38 mm where it has possibly been used as a polisher or rubber. Both this piece and SF 0.02 are unstratified. SF 0.02 is a small pebble, with one domed surface. The other is flatter with a distinct area of polish in its centre measuring 36 by 36 mm. SF 1.10 has a narrow band of slight facetted wear around parts of its circumference. It also has soot or other discolouration around its edges, suggesting it may have been used by a smith. It was found within a deposit of ash from hearth context 113. SF 2.04 is possibly unworked but some discolouration on one surface suggests use as a smoother/polisher. It was found on the surface of the occupation layer (context 212) within the dun.

Hone A single, elongated pebble, SF 7.02 is identified as a possible hone. Its edges are rounded and one surface is flattened and smoothed through slight use. This piece was found in the top of a disturbed occupation context just below the topsoil in the centre of the dun.

Pumice fragments SF 1.06 comprises four pieces of pumice. One is a rough bar, two are small fragments probably broken off the latter, and one has two smooth surfaces, possibly the outer skin of the pumice pebble. This piece is shaped like the rim or base edge of a piece of pottery, but the surface and shape are natural. All four pieces have impurities and probably derive from the same pebble. The pieces were found in the fill (context 123) of a posthole (context 122) inside the dun entrance passage.

Discussion of stone tools This small assemblage of stone tools is typical of those found at other Iron Age sites, where the manufacture of iron tools and implements was rare, excepting the occasional blade or knife (Nisbet 1996). The reliance, therefore, on stone for most domestic and industrial purposes remained a necessity for survival during this period. On Iron Age sites where the geology permitted, certain stones were used for specific tasks (see Ballin Smith 1994): fine-grained sandstones for hones and whetstones, quartz and other dense but fine-grained rocks for smoothing and polishing, heavy cobbles, usually fine-grained for manual hammering, pounding and grinding, and the occasionally pumice piece for sanding and smoothing. The assemblage from Comar Wood is not exceptional, in that the best available stones were chosen for the purpose to which they were intended, but from locally available resources such as rock outcrops, and most likely the bed of the River Glass (as the cobble and pebble tools indicate). Quartzite, quartz and schist were commonly used at Langwell Dun, Strath Oykel, Sutherland (Nisbet 1996, 64), but that site produced many artefacts but mostly of the types identified at Comar Wood.

- 45 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Two querns, apparently contemporary and found in close association with a hearth were made from different stones - granite (hard) and mica-schist (soft). As noted above, a granite boulder may have been used for one, and a block of mica-schist quarried for the other. Mica-schist querns are commonly found across the highland region of Scotland and further afield. However, as in the example of SF 1.03, the rock is prone to lamination and fragmentation. Flour ground from grain with a stone of this poor quality would have been extremely gritty and unpleasant to eat.

The occurrence of probably one cobble or pebble of pumice, later broken, is relatively rare on an inland Iron Age site, and suggests that the piece was deliberately brought to the dun. Pumice is more typically found on coastal sites, one example being Applecross Broch on the west coast of Wester Ross, where several pebbles of this material were found during the excavations (Peteranna 2012).

The stone assemblage indicates that food processing took place (querns and pounders), that iron knives were sharpened (the hone), and that the manufacture of wooden tools, leather goods and pottery required finishing (smoothers/polishers, pumice), and occasionally a large cobble was used as a hammer, possibly in metalworking.

VITRIFIED STONE Two samples of vitrified stone SF 1.07 and 1.11 were located in context 107 a dump of material formed from the clearance of the courtyard, which lay against the inner wall face of the dun. Their combined weight is a little over 2 kg and these irregular lumps of fused material include mainly sandstone as well as micaceous schist and other local rocks. Some of the rock has melted into an icing of flowing pale slag which has fused with other hot rocks during the vitrification process. Recent scientific research by Friend et al 2007 on two duns in north-west Scotland, suggest that vitrification of their stonework probably took place at temperatures of c. 850°C, because of the presence of mica and quartz. Earlier research suggested temperatures of 1000°C would have been needed for the process (MacKie 1969). Although further comparative analysis is need when referring to the Comar Wood dun, the geology of the Cannich area (mica schists) is not too dissimilar to that of the Morar and Moidart areas - both metamorphic rocks (British Geological Survey- igeology 2014), and therefore the processes of vitrification and temperatures attained may have been similar. The causes of vitrification have not been explored in this analysis.

METAL AND METALWORKING DEBRIS Iron Metal objects were rare on the site as only three iron objects and one of copper-alloy were found during the excavations (Table 2). The most recognisable iron object was SF 1.01, a fragment of modern chain link, which was found in the topsoil. This piece was not analysed further.

SF Nrs Type Number 1.01 Iron chain link 3+ (modern) 1.05 Iron object 1 1.09 Copper-alloy fragments 2 1.11 Slag 1 053s Slag 1

Table 2: Metal artefact types

A badly corroded lump of iron SF 1.05, was X-rayed and partly cleaned to aid identification. It was noted that large pustules had formed on its surface and it was identified as a small, irregularly- shaped lump of probably wrought iron. Its irregularity suggests it is most likely an off-cut from a

- 46 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis manufacturing process and not a functional designed item (pers. comm. Will Murray, Scottish Conservation Studio). This piece was found in the fill of a posthole (context 122) inside the dun entrance passage.

An X-ray of SF 1.11 confirmed that it was a rounded and irregular lump of slag. It was found within a bank of material dumped against the dun wall in the courtyard. A small piece of slag was found during the processing of soil sample 053 from the fill (context 165) of a posthole (context 164) within dun courtyard, lying to the east of the south side of the entrance passage. The slag was not analysed further.

Copper-alloy

SF 1.09, two small copper-alloy fragments were found in the primary fill of the re-cut (context 191) of a posthole (context 151) which may have been a roof support in the entrance passage to the dun. The fragments appear to be small off-cuts of thin bronze sheeting or plate, which have been cut or perforated (Figure 2). Their presence suggests some copper-alloy metalworking on the site.

Figure 2: SF 1.09 copper-alloy pieces (scale in mm)

Discussion of metal and metalworking debris The presence of a single iron object, two pieces of slag and a very small amount of copper-alloy within the dun does not indicate prolific activity associated with metal working. The low numbers of metal and slag, and its occurrence within the building can be due to a variety of reasons. Metalworking in the form of smithing of wrought iron could have taken place within a small furnace confined, for example, to a hearth such as context 118 with its extended kerbing. The occurrence of hammerscale in soil samples would be firm evidence of this, but the single iron object is not. The small fragments of slag, may also be evidence of iron working rather than iron production, as the latter could have been carried out some distance away from the building. If the copper-alloy fragments are the result of local manufacture, they too could have been the result of metalworking, possibly on the same furnace/hearth where smithing was taking place. The present evidence however, does not allow firm interpretation of the occurrence or use of iron or copper- alloy on the site, and it is worth considering that these finds may have been brought to the dun from other areas. One interesting fact is the presence of metal and metalworking debris in the fill of postholes associated with both the entrance passage and the inner post-ring of the building. It suggests that both metal and slag are residual items, possibly derived from a phase prior to that of the placing or replacing of timber in the postholes. Small fragments of slag and metal could easily have been dropped on the floor of the dun to later become incorporated in subsequent alterations of the building.

- 47 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

FIRED CLAY A single irregular piece of fired clay SF 1.06, weighing 8.1 g was found within the fill of a posthole (context 122) inside the dun entrance passage. This may be a fortuitous piece of clay burnt during the vitrification of the dun, as its shape is irregular. The lack of other pieces of burnt clay mitigates this find being a piece of daub.

ORGANIC OBJECTS - identifications by Susan Ramsay Two organic samples SF 1.12 and 1.13 were found within burnt deposits (contexts 168) that lay across the inner courtyard of the dun and through its entrance passage. The composition of the deposits included hearth and destruction debris, possibly from a second phase of occupation of the structure.

The samples were charcoal of hazel (Corylus cf avellana) roundwood with another hazel piece that was not roundwood. SF 1.12 has 13 growth rings and intact bark, and is a cut and shaped piece, which is broken. Its maximum diameter is 18 mm and is length 25 mm. Three pieces of hazel wood comprise SF 1.13. One piece is roundwood and is 40 mm in length with some bark present. It has 12 growth rings with a maximum diameter of 13-14 mm. The second fragment of roundwood has five growth rings and a maximum diameter of 12-14 mm and is 48 mm in length, again with some bark present. The third piece of hazel charcoal was irregular in shape and it was not from small roundwood. Tool marks were not identified on any of the pieces.

Figure 2a: SF 1.12 shaped piece, b: SF1.13 two pieces of roundwood

Discussion of organic objects These small pieces of carbonised hazel wood, mostly roundwood are a rare survival on an Iron Age (non-waterlogged) site, where their preservation is most likely the result of the firing and vitrification of the dun. Wooden objects of this period are most often found on crannog sites, such as that dug in 1960 at Loch Glashan (Crone and Campbell 2005). At that site a large variety of artefacts was recovered including large turned pieces, stave-built containers and smaller pegs, tools, spoons etc. Roundwood of c. 15-20 mm in diameter was used for trenails (structural pegs) mostly often made from ash, but hazel, alder and oak wooden pieces were also found on the site (ibid 42- 43), suggesting the use of local woodland resources, as most likely happened at the Comar Wood dun.

Another Iron Age crannog in Ayrshire produced a similar range of artefacts to that at Loch Glashan, where roundwood was used for pegs, spatulate tools and pins. It was noted that 'small objects such

- 48 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis as the awl and spatulate tools were whittled out of small twigs and branches of hazel and holly' (Crone 2000, 128).

At the later site of the Biggings, Papa Stour Shetland, wooden pegs were used for structural purposes associated with its wooden Norse building. All pegs were headless (Morris 1999, 190 and Illus 107) but some had shaped ends. The cut end of SF 1.12 is similar to these and this may indicate that it was a wooden peg or trenail. The smaller pieces of roundwood SF 1.13 from Comar Wood could be pins, but other possibilities are that they were intended to be used as wattle or as charcoal in a hearth or small furnace.

GENERAL DISCUSSION OF ARTEFACTS The finds are distributed across the dun and from the lowest to the latest phases of the site. The potentially earliest group of finds, a quartz rubber SF 1.10 and the small copper-alloy cut pieces SF 1.09, are those found in the fill of posthole 151. Although the posthole is thought to date to Phase 1, the earliest of the site, it is likely that the artefacts were incorporated when it was re-cut in Phase 3, and therefore they could be any date from phases 1 through to 3.

Phase 2 finds, from the burning event, include the burnt wood SF 1.12 and SF 1.13, a piece of iron slag SF 1.11 and a double ended quartz pounder/hammerstone SF 1.14. This group of finds suggest activity around the hearth that could have been industrial in nature such as smithing, with the hammerstone being used in metalworking activities, and the slag derived from those activities. The wood charcoal could have been structural (pins, trenails, even wattle) or possibly charcoal to stoke a small metalworking hearth.

The two broken and worn querns SF 1.02 and SF 1.03 are associated with the hearth 118 and its extension. Their reuse in the kerbing indicates they are probably of earlier date than the hearth, most likely Phase 1. SF 1.05, the piece of wrought iron, which was found in the fill of posthole 122 inside the entrance passage in Phase 2 could have derived from the hearth and possibly adds weight to the interpretation of the use of this feature for metalworking.

The fills (contexts 123 and 165) of post holes (contexts 122 and 164) are considered to be Phase 2 in date, and are associated with the entrance passage or the supporting inner ring of wooden posts in the centre of the dun. The artefactual evidence suggests this was debris from the floor of the dun: iron and slag pieces (SF 1.05 and 053 sample), and an irregular lump of burnt clay SF 1.06 The fragments of a pumice pebble SF 1.06 could have been discarded and included in the posthole filling as packing.

Two stone tools SF 7.01 and SF 7.02 (a double-ended pounder and a possible hone) were located in disturbed occupation deposits associated with a hearth in Trench 7. SF 2.04 was associated with an occupation layer in the possible cell excavated in Trench 2. The majority of evidence is associated with activities in the centre of the dun and in its entrance passage, suggesting that waste materials were regularly taken out of the building for disposal, with occasional smaller bits accidentally dropped onto the floor. Some of these became incorporated in the fills of postholes when the wooden supports of the dun were replaced.

There is nothing outstanding in either the distribution of finds or the range of finds from the dun. The numbers are small, reflecting the limited archaeological intervention at the site, but also perhaps the limited activities that took place there. The cultural evidence mirrors to some extent other Iron Age and later sites such as the crannogs (Crone 2000, Crone and Campbell 2005), brochs, as well as other dun sites, such as Langwell (Nisbet 1996). In contrast to the range and number of finds of the much larger cultural assemblage at Dun Ardtreck, Skye, excavated in the 1960s (MacKie 2000), the assemblage from Comar Wood is very narrow. What is surprising though

- 49 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis is the total lack of pottery from the site, which contrasts remarkably with, for example, Dun Ardtreck. However, this might not be unusual when compared to the finds assemblages from Langwell, Rahoy and Dun Lagaidh (Nisbet 1996, 68), where pottery was not recovered from any of these sites. It might be worth considering whether vitrified duns had specialised functions, and the limited range and number of finds is a result of this.

DATING Two of the radiocarbon dates from the excavations are most relevant to the artefacts recovered (highlighted in Table 3, below). The hearth (context 118) and its extension (context 114), which included both querns, produced calibrated dates in the latter half of the first century BC and the first century AD. This suggests that the querns were likely to have been in use prior to, or at the beginning of that date, but were discarded and then most likely reused during the first half of the first century AD. Other dates in the radiocarbon sequence for the site indicate that the dun's primary activity lay between 376 and 184 cal BC, and therefore the dating of contexts 115 and 117, along with the querns, are associated with a secondary or later activity.

The date range for the destruction deposits, contexts 168, found across the courtyard, partly overlap that of the hearth and kerbing but are also younger, to c. 176 cal AD. This indicates continued use of the site to that time and also to as late as c. 276 cal AD when it was abandoned (see radiocarbon date sequence for the whole site). However, it is possible to suggest that some of the destruction deposits and its artefacts are contemporary with the use of the hearth.

Radiocarbon dates from Langwell Dun (Nisbet 1996, 65-66), indicate that its original construction is older than that of Comar Wood Dun, but that its occupation and use continues through the second half of the first millennium BC and throughout the first century cal AD. However, the long time span indicated by the radiocarbon date is unsatisfactory from the point of view of comparison of activities between the duns.

Dates from Applecross Broch (Peteranna 2012) suggest some contemporary use associated with a hearth and deposits across its courtyard, with that of Comar Wood, but that occupation and use continued at later the site, into the middle of the 6th century cal AD.

These few examples show that the middle to late Iron Age of the western highland region is complicated, with early beginnings, secondary phases of use and in the case of Applecross, activity carrying on into the Pictish period. The picture of use at Comar Wood to some extent mirrors that at other sites, but the radiocarbon dates indicate that its use could have spanned c. 600 years, compared to c. 750 for Langwell and Applecross. However, more interpretation is needed on the origins, function and reuse of Comar Wood in order to understand its status in Iron Age settlement of the period.

Site Radiocarbon Context 2 sigma calibrated dating lab & no. dates Comar Wood Dun SUERC-54237, Context 115, ash layer 41 cal BC - 87 cal GU-34496 over context 177, which AD (92.6% overlies hearth slabs probability) context 118 Comar Wood Dun SUERC-54247, Contexts 168, destruction 48 - 176 cal AD GU-34503 layer considered (89% probability) secondary Langwell Dun GaK-4862 Timber on the dun base, 620 - 30 cal BC possibly predating the dun construction

- 50 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Langwell Dun GaK-4860 Charcoal from posthole at 640 cal BC - AD cal entrance to guard chamber 110 Applecross Broch SUERC-35373, Secondary organic deposit 80 - 240 cal AD GU-24486 abutting broch wall in courtyard Applecross Broch SUERC-35369, Dark red-brown granular 340 - 540 cal AD GU- 24485 sticky soil overlying small hearth slabs, Feature 2

Applecross Broch SUERC-35368, Charcoal-rich metal- 210 - 40 cal BC GU- 24484 working deposit within Feature 4 hearth, considered a primary(?) deposit Table 3: Comparative radiocarbon dates

BIBLIOGRAPHY Ballin Smith, B (ed.) 1994 Howe: four millennia of Orkney prehistory. Edinburgh: Society of Antiquaries of Scotland. Monograph Series Number 9, 185-213.

Birch, S; Fraser, L and Peteranna, M 2014 Comar Wood Dun, Cannich, Strathglass. Archaeological evaluation and measured survey. Unpublished report by Ross and Cromarty Archaeological Services and West Coast Archaeological Services.

British Geological Survey- igeology 2014 [Accessed on line 29-04-2014].

Crone, A 2000 The History of a Scottish Lowland Crannog: Excavations at Buiston, Ayrshire 1989- 90. Edinburgh: STAR.

Crone, A and Campbell, E 2005 A Crannog of the 1st Millennium AD: Excavations by Jack Scott at Loch Glashan, Argyll, 1960. Edinburgh: Society of Antiquaries of Scotland.

Friend, C R L; Dye, J and Fowler, M B 2007 New field and geochemical evidence from vitrified forts in South Morar and Moidart, NW Scotland: further insight into melting and the process of vitrification, Journal of Archaeological Science 34 (2007), 1685-1701.

MacKie, E W 1969 Timber-laced and vitrified walls in Iron Age forts: causes of vitrification. Glasgow Archaeological Journal 1, 69-71.

MacKie, E W 2000 Excavations at Dun Ardtreck, Skye in 1964 and 1965, Proc Soc Antiq Scot 130, 301-411.

Morris, C 1999 Late norse and medieval woodworking, wooden artefacts and products, in Crawford, B E and Ballin Smith, B The Biggings, Papa Stour, Shetland: the history and archaeology of a royal Norwegian farm. Edinburgh: Society of Antiquaries of Scotland.

Nisbet, H 1996 Excavation of a vitrified dun at Langwell, Strath Oykel, Sutherland. Glasgow Archaeological Journal 19 (1994-95), 51-73.

Peteranna, M 2012 Applecross Broch Community Archaeology Project 2006-2010. Unpublished data structure report.

- 51 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Appendix 2

Comar Wood Dun, Cannich, Strathglass: botanical remains Susan Ramsay

Summary Excavation of a prehistoric enclosed dun in Comar Wood, Strathglass, has provided evidence for the use of oak as the main structural timber. Charcoal from hearth deposits indicates that the inhabitants made full use of the locally available woodland resources but that there may have been some degree of selection of oak for fuel as well as for construction. There was very little evidence for food plant remains, with only small quantities of carbonised barley grain and hazel nutshell recorded. Radiocarbon dating suggests that the site was occupied between the 4th century BC and the 2nd century AD, although it is not possible to determine whether this was a continuous period of habitation.

Introduction The following archaeobotanical report details the processing, analysis and interpretation of carbonised botanical remains recovered from samples taken during excavations at Comar Wood Dun, Cannich, Strathglass (NGR NH 32509 31008). The archaeological investigation was carried out by Ross and Cromarty Archaeological Services and West Coast Archaeological Services on behalf of Forestry Commission Scotland between 29th August and 6th November 2013. The excavation revealed a prehistoric enclosed, galleried dun with a defensive outwork (Birch, Fraser & Peteranna, 2014).

Methodology

SAMPLE PROCESSING A programme of bulk sampling, plus spot finds, was undertaken in order to examine the carbonised archaeobotanical remains from Comar Wood. In total, 42 samples, representing 30 separate contexts, were analysed for botanical remains. The bulk samples were processed by flotation or wet sieving for the recovery of carbonised remains, using standard methods and sieves of mesh diameter 1mm and 500µm for flots (and wet sieved samples) and 2mm and 4 mm for retents from flotation. There was no indication that any of the samples had the potential for the preservation of uncarbonised plant remains through waterlogging.

MACROFOSSIL ANALYSIS Dried flots and sorted retents were examined using a binocular microscope at variable magnifications of x4 - x45. For each sample, estimation of the total volume of carbonised material >2mm and >4mm was made and modern contaminants were scored using a scale of 1-3 ‘plus’ marks. For each sample, all the charcoal >4mm was identified unless this proved to be too large an amount, in which case a known percentage of the total charcoal >4mm was identified. All carbonised seeds were also identified and any other plant macrofossil remains were noted.

The testa characteristics of small seeds and the internal anatomical features of all charcoal fragments were further identified at x200 magnification using the reflected light of a metallurgical microscope. Reference was made to Schweingruber (1990) and Cappers et al (2006) to aid identifications and vascular plant nomenclature follows Stace (1997).

Results Results will be discussed by trench and by feature type. Numbers in brackets indicate individual context numbers; round brackets relate to deposits or fills and square brackets indicate cuts or

- 52 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis features. The results are shown in Table 1 (Trench 1 – hearth deposits), Table 2 (Trench 1 – posthole fills and structural debris) and Table 3 (Trenches 2, 3, 4 and 5) at the end of this report.

TRENCH 1 Trench 1 was positioned through the entrance of the structure, from outside the outer dun wall, along the entrance passage and into the dun inner courtyard.

Hearth deposits (Table 1) Excavations within the entrance passage revealed several areas of charcoal-rich deposits (111) and (168) that were thought to be spreads from hearth rake-out deposits from a large slab-built hearth [118] or from a burning event. The charcoal assemblages included large quantities of alder, birch, hazel and oak, together with a small quantity of willow (in context (168) only) and a few poorly preserved carbonised barley grains. A wedge-shaped deposit of sandy sediment with stone clasts (107) overlay one of the patches of the charcoal-rich deposit (111). Context (107) produced large amounts of birch and hazel charcoal, with some oak also present, and is thought to represent the first phase of courtyard clearance. Hazel roundwood charcoal from (107) produced an AMS date of 382 – 204 cal BC (SUERC-54242). However, a spot find of hazel roundwood charcoal from (168) produced a much later AMS date of 27 - 212 cal AD (SUERC-54247).

The hearth [118] was surrounded by a stone kerb [113] on three sides, with an open side towards the entrance passage. Ash deposit (115), covering the central hearth [113], produced a similar charcoal assemblage to those discussed above, with alder dominating and lesser quantities of birch, hazel and oak charcoal also present. Hazel roundwood charcoal from ash deposit (115) produced an AMS date of 41 cal BC – cal AD 120 (SUERC-54237).

Below ash layer (115) but above the hearth slabs (118) was a further charcoal rich layer (117), which probably represents the primary use of the hearth. The charcoal assemblage from (117) was dominated by alder charcoal with a small amount of birch and traces of oak also present. In addition, a few carbonised cereal grains were also recorded but these were too poorly preserved to be further identified to species. This might suggest that they had been subjected to multiple episodes of burning within the hearth. Birch roundwood charcoal from (117) produced an AMS date of 45 cal BC – cal AD 80 (SUERC-54233).

The above carbonised assemblages are in accordance with what would be expected from hearth deposits, with the carbonised cereal grains giving some indication of domestic occupation. The presence of significant amounts of oak charcoal in (111) may suggest that there is some structural component to these deposits since oak is less commonly used on general domestic hearths, although it is often preferred for metalworking hearths.

Below layer (111) was a mixed deposit (121) with charcoal, ash and rounded stones, and below this was a further charcoal rich layer (153) overlying the sub-soil, which showed evidence of vitrification (169). The charcoal assemblage from (153) was dominated by birch, hazel and oak charcoal, with traces of alder and willow also present. In addition, a single carbonised barley grain and one fragment of hazel nutshell were the only evidence for food plant remains. Hazel roundwood charcoal from (153) produced an AMS date of 376 – 203 cal BC (SUERC-54243). The charcoal assemblage from the vitrified sub-soil (169) was also dominated by birch and oak charcoal, but in this context only a trace of hazel charcoal was present.

A section through the hearth revealed evidence for another, earlier, slab-built hearth [176] and it became clear that the vitrified sub-soil (169) and the charcoal rich deposit (153) were contemporaneous with this earlier hearth. A second ash context (175) overlay the hearth slabs [176] and the charcoal rich deposit (153). However this ash layer (175) produced only a trace of birch

- 53 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis charcoal and no other identifiable carbonised remains. Within layer (169) was a charcoal-rich lens (177) that produced a charcoal assemblage dominated by oak, with only a small amount of birch also present. This lens must also be associated with hearth [176]. The radiocarbon dates from the hearth deposits span a period of around six centuries, which suggests that the site was occupied for a significant period of time and that a degree of mixing of deposits had occurred.

Posthole fills and structural debris (Table 2) Two postholes, [160] and [164], were located at the east end of the entrance passage, within the courtyard of the dun. The fill (161) of posthole (160) had a charcoal assemblage dominated by oak but with birch, hazel and Scots pine type also present. The fills (165) and (181) of posthole [164] also produced large amounts of oak charcoal but there were also significant quantities of birch and hazel, with smaller amounts of alder, willow and Scots pine also recorded. Fill (181), which comprised the packing stones within posthole [164] also produced a few carbonised cereal grains, possibly barley, but in poor condition. These results suggest that the main posts were probably made of oak but that significant quantities of other carbonised material became incorporated into the posthole fills at some point, possibly during one of the destruction phases.

Posthole [186], located below a possible wall [108], contained a primary fill (154) and a secondary fill (187). Only the secondary fill (187) was examined in the study but it produced only small amounts of birch and oak charcoal, which were not sufficient to suggest a post had been burnt in situ. However, immediately to the east of posthole [186] fragments of carbonised timbers were noted, which were thought to be either the remains of collapsed roof elements or hurdle fencing. In addition two small finds (SF1.12 and SF1.13) of carbonised wood showed evidence for having been trimmed and tapered at the ends. Large quantities of charcoal was recovered from (179) and identified as birch and hazel, in roughly equal proportions. The small finds of charcoal were all small hazel roundwood fragments. This assemblage seems more indicative of hurdle fencing than roofing elements.

Postholes [122] and [188] were located on the south side of the entrance passage, with postholes [124] and [149] on the north side. It is thought that these postholes are the remains of a porch over the entrance of the dun. Posthole [122] contained a single fill (123) that had a charcoal assemblage dominated by large amounts of oak, with a small amount of birch the only other type present. The main fill (125) of the opposing posthole [124] produced an identical charcoal assemblage as found in (123). Posthole [124] had been recut at some time and this posthole [127] also produced fills (128) and (130) comprising large quantities of oak charcoal with small amounts of birch also present. Birch charcoal from (128) produced an AMS date of 371 – 201 cal BC (SUERC-54241). Posthole [188] on the south side of the passageway had a fill (189) that also contained only oak and birch charcoal, but in more or less equal quantities. However, birch charcoal from (189) produced a much later AMS date of cal AD 70 - 224 (SUERC-54240). The opposing posthole [149] contained a primary fill (157), which was not examined, and a secondary fill (156) that only produced small amounts of alder, birch and oak charcoal.

A smaller posthole [131], fill (132), was located adjacent to wall [106] in the entrance passageway and produced a carbonised assemblage again dominated by oak but with very small quantities of alder and birch also present. This is further evidence for an oak post, burnt in situ.

The removal of one of the entrance passage walls [167] revealed a posthole [151], which had subsequently been recut [191]. The lower fill (170) contained significant amounts of oak and birch charcoal, with traces of alder also present. This is a similar assemblage to those from most of the other postholes associated with the entrance passage. The upper fill (159) contained only small amounts of oak and hazel charcoal and provides little further information.

- 54 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

As with the hearth deposits, the AMS radiocarbon dates are not consistent and suggest significant reworking of charcoal from earlier periods of occupation.

TRENCH 2 Trench 2 was located across the north side of the dun to investigate what appeared to be a gallery or chamber within the wall of the structure. A charcoal-rich surface (212) was located beneath rubble associated with the inner dun wall. This context produced a charcoal assemblage dominated by hazel, with lesser amounts of birch, oak and willow also present. Hazel roundwood charcoal from (212) produced an AMS date of 371 - 201 cal BC (SUERC-54232). This might suggest the remains of a burnt hurdle screen within this context, which may be related to the end of the primary phase of occupation.

Two small postholes [219] and [221] were located below the inner wall bank [215]. The fill (220) of posthole [219] produced a mixed charcoal assemblage dominated by willow, with small amounts of oak, hazel and alder also present. In addition, a single carbonised barley grain and some fragments of hazel nutshell were also recorded from this fill. These finds do not suggest the remains of a post burnt in situ but are more consistent with hearth waste. Willow charcoal from (220) produced an AMS date of 365 - 184 cal BC (SUERC-54238). The AMS radiocarbon dates from Trench 2 are much more consistent and indicate that the primary period of occupation was around the 3-4th centuries BC.

TRENCH 3 Trench 3 was located over the possible entrance through the enclosure wall on the west side of the dun. Only a single context (303), identified as a charcoal patch, was examined from this trench. This context produced only a couple of pieces of oak charcoal, which may suggest a structural origin.

TRENCH 4 Trench 4 was situated in Structure 1 on the southwest side of the dun. This structure used the dun wall as its western wall with its south-eastern wall abutting Structure 2; both structures were thought to date to the Post-Medieval period. An occupation layer (403) within Structure 1 produced large quantities of birch and oak charcoal, with traces of hazel nutshell. This assemblage could be hearth waste but might also be evidence for structural elements.

TRENCH 5 Trench 5 was located against the dun outer wall face on the east side of the structure. A section of wall [506] that formed a blocked-up passage was built on top of a charcoal-rich silt (507), although on examination this silt produced only small quantities of alder charcoal. Below (507) was a layer of redeposited natural (508) but this context produced only traces of Scots pine type and oak charcoal. Scots pine charcoal from (508) produced an AMS date of 5740 – 5645 cal BC (SUERC- 54239) indicating that this charcoal is redeposited from a much earlier period of possibly natural burning. Below wall [505] was an old ground surface (512) but again this produced only traces of birch and oak charcoal. Outside the dun wall there was evidence for an external wall, which had been constructed of large stones within a stone-silt matrix (503). This silt matrix contained only small amounts of alder and hazel charcoal, which may be the remains of redeposited hearth waste.

Discussion The charcoal finds from the dun hearth contexts and from the posthole fills of the entrance passage show a range of AMS radiocarbon dates, which suggests that the site had been in use for several centuries. It is not possible to determine whether this was a continuous period of habitation or

- 55 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis whether there were several discrete episodes of use. Earlier fragments of charcoal seem to have become incorporated into later posthole fills making it difficult to date the individual features and to determine which features are contemporaneous.

The charcoal assemblages from the hearths indicate that the inhabitants were utilising locally available woodland resources for fuel. The fuel types represented were oak, hazel, birch, willow and alder, although the abundance of oak in domestic hearth waste is not common on Scottish sites after the Bronze Age. The results may simply indicate that oak was still a common woodland resource in the area, unlike the situation in central and southern Scotland. It may also be that elements of structural timbers have become incorporated into these hearth deposits during the destruction episodes that took place within the dun. However, a further possibility is that oak was being deliberately collected for hearths where high temperatures were required e.g. for small scale metalworking (Gale & Cutler, 2000). However, this suggestion would require further evidence in the form of metalworking waste or a concentration of metal finds from the site.

Very few food plant remains were recovered from this site, with only very small quantities of carbonised barley grains and fragments of hazel nutshell recorded. The barley grain that was recovered was extremely poorly preserved and looked as if it might have been subjected to several episodes of burning. The lack of food plant remains on this site might suggest that it was either not occupied by significant numbers of people or that it was only occupied sporadically. The presence of barley does not help to narrow down the main period of occupation since barley has been the commonest cereal type in much of Scotland from the Neolithic to the Medieval period (Dickson & Dickson, 2000).

Several of the posthole fills, particularly those associated with the entrance passage, showed evidence for the presence of structural remains. Oak dominated, as would be expected from large posts that had to support a substantial structure. However, significant amounts of birch and hazel were also present, which might suggest the presence of wattle hurdle panels or dividers within the structure itself. However, the range of AMS dates obtained from these features makes it difficult to determine whether the oak is contemporaneous with the other charcoal types. The birch and hazel charcoal may be from an earlier period of occupation. It is also possible that the birch was, in some way, associated with the roof structure but it is difficult to be any more than tentative with this suggestion. There was no evidence for burnt turf, heather, reeds or straw which might have indicated roofing material within any of the mixed deposits.

REFERENCES

Birch, S, Fraser, L & Peteranna, M 2014 ‘Comar Wood Dun, Cannich, Strathglass: Data Structure Report’. Unpublished Report

Cappers, R T J, Bekker, R M & Jans, J E A 2006 Digital Seed Atlas of the Netherlands, Groningen Archaeological Studies 4, Barkhuis Publishing, Eelde, The Netherlands.

Dickson, C A & Dickson, J H 2000 Plants and People in Ancient Scotland. Tempus, Stroud. Gale, R & Cutler, D 2000. Plants in Archaeology. Westbury Publishing, Otley.

Schweingruber, F H 1990 Anatomy of European Woods. Haupt, Berne & Stuttgart.

Stace, C 1997 New Flora of the British Isles 2nd Ed. Cambridge University Press.

- 56 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Comar Wood'13 Context 107 111 115 117 153 168 169 177 175 Sample 059 008, 021, 004, 007, 003 030, 055, 057, 064 049 051 065 047 013, 014 060 Description Inner bank of Charcoal-rich Spread of Charcoal-rich Charcoal-rich Burning layer Partially Charcoal-rich Ash layer material spread of ash from layer over lower deposit related to vitrified layer to W of below hearth against the material from hearth [113] hearth slabs - within Trench wooden surface hearth [113] paving [118] dun inner around the primary use 1 structure below hearth wall face - hearth down of hearth collapse slabs [118] first phase of the entrance during phase courtyard passage 2 of dun 'clearance' occupation Volume of charcoal 2-4 mm 470ml 300ml - 150ml 300ml 240ml 50ml <<2.5ml <<2.5ml Volume of charcoal >4 mm 400ml 225ml 165ml 500ml 170ml 500ml 35ml 10ml <<2.5ml % charcoal >4mm identified 25% 60% 70% 20% 65% 30% 100% 100% 100% Charcoal Common name Alnus cf glutinosa alder - 8 (4.33g) 72 (12.09g) 70 (16.01g) 1 (0.07g) 88 (34.98g) - 4 (0.11g) - Betula spp birch 36 (12.53g) 47 (11.05g) 33 (3.21g) 9 (1.63g) 57 (10.91g) 35 (5.90g) 50 (3.74) - 1 (<0.01g) Corylus cf avellana hazel 45 (10.58g) 31 (8.25g) 8 (2.57g) - 28 (16.00g) 5 (1.16g) 1 (0.06g) - - Quercus spp oak 11 (2.35g) 25 (8.92g) 6 (0.19g) 1 (0.18g) 31 (2.93g) 23 (2.79g) 34 (2.10g) 13 (1.79g) - Salix spp willow - - - - 1 (0.07g) 11 (1.62g) - - - Cereals Hordeum vulgare sl barley - - - - 1 4 - - - cf Hordeum vulgare sl cf barley - 9 ------Cereal indet indet cereal - - - 8 - - - - - Carbonised seeds Corylus avellana nutshell frag hazel nutshell frag - - - - 1 (0.04g) - - - - Table 1: Botanical remains from Comar Wood, Trench 1 – hearth deposits.

57

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Comar Wood'13 Context 123 125 128 & 130 132 156 159 161 165 170 179 181 187 189 Sample 038 044 043, 045 023 039 048 058 053 052 062 022 033 036 Description Fill of Fill of Charcoal- Charcoal- Fill of large Upper post- Fill of Fill of Outer Burnt/ Packing Packing Fill of posthole posthole rich fill of rich fill of posthole pipe fill of posthole posthole (primary) fill collapsed stones stones and posthole [122] inside [124] inside post/stake post/stake [149] = posthole [160] [164] of posthole roof/structura within fill inside [188] inside dun entrance dun entrance hole [127] hole [131] (150) [191] [191] l timbers posthole posthole dun entrance passage passage within dun [164] [186] passage Volume of charcoal 2-4 mm - - 320ml 525ml - - 90ml 120ml 100ml 100ml 180ml - - Volume of charcoal >4 mm 30ml 20ml 135ml 320ml 15ml 10ml 50ml 180ml 30ml 1500ml 140ml 10ml 10ml % charcoal >4mm identified 100% 100% 100% 25% 100% 100% 100% 100% 100% 33% 50% 100% 100% Charcoal Common name Alnus cf glutinosa alder - - - 3 (0.40g) 1 (0.39g) - - 5 (0.57g) 2 (0.05g) - 3 (0.16g) - - Betula spp birch 3 (1.22g) 2 (1.78g) 16 (1.55g) 4 (0.57g) 1 (0.36g) - 26 (1.78g) 7 (1.43g) 36 (1.97g) 39 (85.16g) 57 (7.33g) 1 (0.53g) 7 (1.33g) Corylus cf avellana hazel - - - - - 1 (0.46g) 3 (0.16g) 39 (4.53g) - 24 (84.74g) - - - Pinus sylvestris type Scots pine type ------1 (0.06g) 1 (0.08g) - - - - - Quercus spp oak 52 (10.09g) 15 (3.13g) 663 (29.28g) 172 (18.92g) 5 (0.62g) 6 (0.38g) 247 (8.07g) 133 (8.85g) 100 (3.92g) - 72 (6.29g) 4 (0.84g) 4 (1.38g) Salix spp willow ------7 (1.03g) - - - - - Cereals cf Hordeum vulgare sl cf barley ------5 - - Table 2: Botanical remains from Comar Wood, Trench 1 – posthole fills and structural debris.

58

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis

Comar Wood'13 Trench 2 Trench 3 Trench 4 Trench 5 Context 212 220 303 403 503 507 508 512 Sample 001, 002, 015 024 063 009, 011 017 020 026 025

Description Charcoal-flecked Fill of posthole Charcoal-rich Occupation Dun rubble but Charcoal-rich Charcoal- Old ground occupation layer [219] patch at E end of deposit may be later deposit - surface flecked, surface below inside dun trench that wall core occupation redeposited Feature 505 extends to S deposit natural Volume of charcoal 2-4 mm 400ml 20ml - 170ml - - - - Volume of charcoal >4 mm 480ml 25ml 5ml 105ml 10ml 10ml 2.5ml 5ml % charcoal >4mm identified c. 60% 100% 100% c. 60% 100% 100% 100% 100% Charcoal Common name Alnus cf glutinosa alder - 2 (0.09g) - - 5 (1.25g) 5 (1.12g) - - Betula spp birch 94 (3.98g) - - 100 (7.82g) - - - 3 (0.56g) Corylus cf avellana hazel 158 (30.97g) 5 (0.19g) - - 2 (0.64g) - - - Pinus sylvestris type Scots pine type ------2 (0.05g) - Quercus spp oak 14 (0.67g) 2 (0.14g) 2 (1.28g) 57 (3.33g) - - 1 (0.09g) 1 (0.07g) Salix spp willow 9 (1.90g) 52 (3.18g) ------Cereals Hordeum vulgare sl barley - 1 ------Carbonised seeds Corylus avellana nutshell frag hazel nutshell frag - 7 (0.08g) - 4 (0.04g) - - - - Table 3: Botanical remains from Comar Wood, Trenches 2, 3, 4 & 5.

59

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis Appendix 3

Animal bone from Comar Wood Dun, Strathglass Catherine Smith

Introduction and methods

Animal bone fragments extracted from samples from the excavation at Comar Wood Dun, Strathglass were examined in order to determine whether any were identifiable to species or particular bone element. All of the bone samples seen came either from hearth, hearth-related or destruction layers relating to burning of the dun and posthole or beam slot fills.

The bone samples consisted of highly fragmented mammalian material, all of which had been subjected to high temperatures to the extent that the bones were calcined and had lost most of their organic component. Although a small number of the fragments were recognisable as belonging to particular species, most were not and were therefore described according to size. Larger, thicker- walled fragments were described as large mammal (LM in the accompanying catalogue) and probably came from large animals, most likely cattle. Medium mammal bones (abbreviated MM) were most likely from sheep/goats or pigs. Indeterminate mammal (IM) was the term used to describe any other mammalian fragments. It should be stated that this system has its limitations, notably that the high temperatures to which the remains were subjected has led to a high degree of shrinkage due to burning out of the organic components and it is therefore difficult to separate large from medium mammals. It is possible for calcined bones to shrink to about half their original size depending on the intensity of the heat to which they have been subjected.

As a guide to the retrieval rate, the lengths of the smallest and largest fragments in the sample are given as a range (in millimetres) in the catalogue. Lengths of most individual fragments are not stated unless the bone was identifiable to particular species or element. (These lengths are merely a measure of the size of the fragments along their longest axis and do not relate to any formal system of anatomical measurement such as von den Driesch (1976)).

Results

The accompanying catalogue details the fragments found in each sample, in context order. Most of the fragments were unidentifiable due to heat damage and are logged as indeterminate mammal (IM), but, as is common with this type of material, small elements such as sesamoid bones survived intact due to their small size and dense structure and were identifiable to species. Two samples contained cattle sesamoids: Sample 46, Context 111 (a spread of hearth material or a separate burning event) and Sample 57, Context 107. Other cattle remains identified were a fragment of mandible with tooth alveoli from which the incisor teeth were missing (Sample 40, Context 156, a fill of a large posthole) and a distal fragment from a cattle second phalange or toe bone (Sample 46, Context 111).

Some ungulate tooth enamel fragments, most likely from cattle, were also preserved, if poorly. Tooth fragments were present in two samples: Sample 60, Context 153; a lower spread of charcoal- rich material and Sample 52, Context 170; outer fill of posthole.

One bone from sheep/goat was also noted. This was a very shrunken fragment from the proximal end of a first phalange. The epiphysis had fused, but as this bone fuses fairly early in the animal’s life it could only be stated that the animal was immature or adult at the time of death.

- 60 -

Comar Wood Dun, Strathglass: Final report on the post-excavation results and analysis One possible rib shaft fragment (Sample 29, Context 115), which unfortunately could only be described as large/medium mammal, still bore traces of knife cuts on the surface of the bone. Two paired parallel sets of two short cuts were observed, running diagonally to the long axis of the bone. These cuts would have been inflicted during butchery, specifically filleting, of the meat. The identical pairing of the cuts implies the same implement made both sets of marks.

Discussion

The analysis of this material demonstrates that even the most unpromising of archaeological samples may contain identifiable fragments.

Burnt long bones typically break up into many unrecognisable shaft fragments and since only a few areas of the long bones, particularly of larger animals, retain diagnostic characteristics, the larger the animal, the greater the number of fragments the bone breaks into and the less likely it is that they can be identified with any certainty.

By contrast, sesamoid bones, often overlooked in hand-collection of excavated material because of their small size, are diagnostic at least to mammal family if not to species and as noted above are often the only recognisable fragments to survive where intense burning has occurred. Similarly, foot phalanges are also small and their articular ends dense and so may also survive.

Despite the low rate of identification it can be stated that the bones of domesticated livestock were present in the assemblage and that at least two species, cattle and sheep/goats, were being utilised by the occupants of the site. Much of the burning was probably related to the destruction of the dun by fire, for example in Context 111, from which probable cattle remains were retrieved. Where hearth material was present it is possible that burning of the bones may have been deliberate, as a method of disposal of the remains of meals, although the catastrophic destruction wreaked by fire over the whole site is a more likely explanation. Bone fragments may have been accidentally incorporated into the fills of postholes during clearing of the site at the end of Phase 2 and the fragments themselves provide no evidence that they were introduced deliberately.

Reference von den Driesch, A 1976 A Guide to the Measurement of Animal Bones from Archaeological Sites (= Peabody Museum Bulletin No 1). Harvard University. Harvard, USA.

- 61 -