1997:506 - DER18 (INCLUDING DER302 AND DER305), Killoran, Tipperary

Middle Bronze Age
This site, located in the centre of the TMF development area of Derryville Bog, was established by the IAWU (labelled 95DER18) to extend east–west across the Middle Bronze Age bog to dry land. Archaeological excavation over two seasons (Excavations 1996, 104, 96E0203 and 96E0298) found that the site dated from before 1440+9 BC (Q9349), ran west to east for 555m and was exposed for a period of up to 40 years. Investigation took place between May and August 1997 in eight cuttings opened by machine across all or part of a Bord na Móna field, 13m wide. Excavation revealed three phases to its construction and considerable variation in overall building technique over its length.

Environmental context
The builders of DER18 started from the Derryfadda peninsula in the east and crossed through tall reed-fen growing over a broad glacial ridge using a stone causeway. At the western terminus a much wetter discharge channel, c. 60m wide, of slow-moving water draining the bog was encountered. Attempts were made to cross the channel using wood from the western margins. Worsening surface conditions brought about by the presence of this heavy wooden deposit effectively dammed the channel, resulting in the flooding of the entire causeway. The effects of the stream overflowing caused the rapid and unusually uniform growth of raised bog over the entire length of the track, rendering it impassable.

Stratigraphic context
PHASE 1
The primary phase of construction was defined as two parallel rows of stakes set out as marker-posts through an area of dense reed-fen (dry enough to walk on in the summer months) from a glacial ridge in the east. This phase was radiocarbon-dated to between 1745 and 1405 cal. BC (Beta-1027530), constructed in a single season. The double stake row, containing around 1400 individual sharpened stakes, was orientated east–west and measured 456–488m in length, with each row no more than 1.4m apart. The stakes in each row were set on average 0.4m apart and were 60mm in diameter, and appeared to span the Bronze Age width of the bog, terminating 10–20m from the eastern landfall and terminating at the main bog discharge channel 6–6.5m from the western landfall.

PHASE 2
Stone, clay, and gravel paving formed the majority of this phase, set onto an already well-worn surface of compressed reeds in fen peat. This phase of the causeway was dated to 1605–1285 cal. BC (Beta-102751). The stone causeway measured 484–486m in length and 1.2–1.4m in width; stone paving was laid within the defined limit of the stakes, although several of the stakes were no longer used as effective markers, either from rotting or from being damaged during construction. In several western trenches the causeway was revealed as a loose stepping-stone arrangement of small angular cobbles; elsewhere it formed a tightly packed paving of boulders and clay. The purpose of this stone seemed to have been to weigh down the reed to create a matting surface on which to walk.

Localised areas of wetter peat resulted in the need for intermittent use of brushwood and sand as a foundation for the stones. The wood was laid in short stretches, 1.1m long and 1.4m wide, set between the stake rows. The eastern terminus was very well defined as a metalled roadway 2.2m wide crossing onto the glacial till and veering north-east, over 7m in length, suggesting that this phase was constructed from the east.

The western section of the causeway was constructed almost entirely of wood owing to problems encountered crossing the very wet discharge-channel zone. The terminus was very poorly constructed—a deep deposit of longitudinal roundwood, occasional planks and brushwood, often utilising dry root systems to zigzag across the channel, measuring c. 8m in length, up to 0.8–3.3m in width and up to 0.55m in depth. Several elements of this construction branched off to the south-east or north-west and appeared to be aborted attempts to find another access route from the west. One substantial branch track was assigned a separate site number, DER302.

PHASE 3
This revised phase represented the intermittent repair to the structure where it had sunk into softer peat. Oak planks, some with cut holes or mortises, were laid over transverse roundwood and between piles of stone, or longitudinal planks and/or roundwood were laid directly over the sunken causeway. This phase was dated to before 1440+9 BC (Q9349).

Rath House, Ferndale Road, Rathmichael, Co. Dublin