Hallhole Farm, Meikleour

In January, 2017, Alder conducted a watching brief on soil stripping in advance of the construction of an agricultural storage building to the south of Blairgowrie, Perth & Kinross. The site lay within a rich archaeological landscape: the Coin Hill unenclosed settlement, 300m E of the site; Black Hill Roman signal station 150m to the southeast; the Cleaven Dyke, a 2.5km-long Neolithic earthwork of international significance- the southeast terminal of the bank lay 700m northeast of the site, although pit alignments visible on aerial photographs appear to extend it further southeast towards the River Isla.  A Neolithic mortuary enclosure excavated at Littleour lay approximately 700m north of the site, on the far side of the Cleaven Dyke (Barclay, Maxwell et al 1998). The Neolithic long barrow of Herald Hill stands at the opposite end of the Dyke, close to the river. Neolithic monuments have also been excavated at Inchtuthill (a mortuary enclosure, Barclay & Maxwell, 1991) and Beech Hill House, Coupar Angus (Grooved Ware recovered from beneath a Bronze Age burial mound, Stevenson, 1995), both within a few kilometres of Hallhole.


                                                       Cleaning the site by hoe 

On the last day of the watching brief, fragments of prehistoric pottery began to emerge from several areas of the site. Soil stripping was by then complete, but following a site meeting with the client and Perth & Kinross Heritage Trust, it was decided to fully excavate the site before proceeding with cutting foundation trenches. After a clean-up with help from labourers from the Meikleour Estate, over a hundred separate features were identified and work began on recording and excavating these between 19th January and 3rd March, 2017. As suspected, the majority proved to be natural features, either tree or boulder boles or rabbit burrows- burrowing was apparent across the site and had in fact thrown up some of the potsherds originally noted- but 49 anthropogenic features remained, 43 of prehistoric origin identified by potsherds and lithics found within their fills. 


Flint lithics from Hallhole (Laura Fyles) 

The prehistoric features were pits and postholes, comprising four main clusters in the northeast and northwest corners, centre and south centre of the site, with several outliers in addition. There appeared to be chronological divisions between the clusters, with the northeast and northwest clusters containing Early Neolithic pottery (apart from two pits containing Bronze Age vessels) and relatively few lithics, mostly of worked quartz, while the central and south-central groups contained Late Neolithic Grooved Ware and abundant lithics of flint (including a high proportion of imported "Yorkshire flint").  


Pit feature F29 

A total of 76 bulk soil samples were taken from deposits within the excavated features. Paleobotanical analysis of several of these revealed the presence of 50-100 grains of naked barley (hordeum var nudem) and emmer wheat (triticum dicoccum), together with chaff, in a large pit (F29) in the northeast cluster, which also contained charred seeds of cleaver (galium aparine ), which is commonly found on waste ground around the edges of a settlement. As well as seeds, F29 contained a setting of heat-fractured stones, sherds of severely abraded Early Neolithic pottery and quartz tools. It lay within a few metres of several possible storage pits and a tight cluster of postholes that may have represented domestic activity including food preparation. An apparent structured deposit was also present, in a small, scooped pit (F41)

The northeast corner of the site 

C14 dates were obtained from fills of F29 and F41. A barley grain from F29 gave an uncalibrated date of 4820 BP, +30, calibrated to 3656-3526 BC (95.4% probability, with a younger date range of 3598-3526 cal BC probable at 60.6%). Hazel wood from the upper fill of F41 gave an uncalibrated date of 4729 BP, +29, calibrated to 3635-3377 BC (95.4% probability, with an older date range of 3635-3552 cal BC probable at 43.9%). Both samples therefore appear to represent activity around 3600-3500 BC, during the Early Neolithic. This may have been broadly contemporary with the construction of the Cleaven Dyke, as well as the earliest phase of activity at the mortuary structure at Littleour (3650-3100 cal BC). The dates are also compatible with the lithic and ceramic evidence, both of which point to Early Neolithic activity in this part of the site.

Grooved Ware from the centre of the site (Laura Fyles)

Meanwhile, the central and south-central clusters suggested three separate phases of activity. The vast majority of lithics and much of the Grooved Ware pottery from the site came from just two features:  a shallow pit (F68), which had been partly cut by a smaller posthole, contained three bipolar cores, two short-end scrapers, a thumbnail scraper and a sandstone grinding slab broken in two (a fragment of a similar slab was found in the northeast of the site; comparison with slabs found in Norway, where they appear to be of Mesolithic origin, suggests these examples are likely to be Early Neolithic, with the one in F68 probably re-deposited). A sub-circular posthole (F65) contained flakes of worked flint and a large amount of debitage, also found in F68. The quantities of debitage and waste flakes from these features suggested the presence of a knapping floor, through which F65, F68 and another pit (F64) were later dug; four smaller postholes forming a rectangle contained no debitage, suggesting they may have been part of a light structure contemporary with the knapping floor, which may also have been the origin of the many small fragments of Grooved Ware (some charred and mostly in the Durrington Walls style, uncommon in Scotland) later swept into F65 and F68. Hazelnut shell from the upper fill of F65 gave an uncalibrated date of 4157 BP, +25, calibrated to 2877-2635 BC (95.4% probability, with a later range of 2819-2661 cal BC being probable at 73.2%).

Structured deposit F17

> The features to the south of the central cluster appeared to form an alignment of pits and/or large postholes. F93, F16 and F77 were all large pits measuring between 0.80m - 1.14m across and all contained post-shaped depressions and possible ramps that suggested the presence of a large post for at least part of their existence. While F77 may have stood open for long enough to permit windblown silt to accumulate, F16 and F93 both appeared to feature tiplines indicating deliberate backfilling and generous quantities of charcoal; in F93, this and discoloured subsoil suggested the in situ burning of the post. A small pit (F17) on the lip of F16 contained two upturned Grooved Ware pots, both flat-based and decorated with incised marks, in what was certainly a structured deposit, possibly dug against an upstanding post in F16.  F16 and F93 both contained flakes of Yorkshire flint in their upper fills, against the lip of the cut, in contexts likely to represent sealing deposits closing off the feature. Hazelnut shell from a middle fill of F16 gave an uncalibrated date of 3997 BP, +29, calibrated to 2575-2469 BC (95.4% probability). Hazel wood from a band of carbonised material within F93 gave an uncalibrated date of 3965 BP, +27, calibrated to 2571-2350 BC (95.4% probability, with mid-range dates of 2505-2450 cal BC being probable at 49.5%). The dates for the pit alignment place it at the very end of the Neolithic. 

Pit alignment, south centre of site