Roughly L-shaped, measuring no more than 3.5 metres by 2 metres and less than 30 centimetres deep, it is easy to overlook as a feature of any particular significance. But the care taken to make it, prising out sections of limestone along the natural bedding planes, and the material left inside it, suggest a purposeful use of this small carved space long before the floor above it was ever laid.

The hollow, recorded as F107b, came to light during excavations in what is called Cutting 2. On its floor, a mixture of quarry grit and charcoal-rich clay had built up in distinct layers, and within that deposit were fragmented organic refuse and scraps of pottery. A sample of animal bone taken from the fill was sent for radiocarbon dating, and the result placed the deposit somewhere between 920 and 540 cal. BC, placing it firmly in the Later Bronze Age. At some point after that accumulation, the hollow was sealed when the paved floor of Structure 2.1 was laid directly over it. That sealing is, in an odd way, what preserved the evidence: whatever activity produced the charcoal, the pottery, and the animal bone was effectively locked in place by the construction above it, left undisturbed until modern excavation reached it. The research was published by Claire Cotter in 2012, drawing on fieldwork carried out as part of a wider investigation of Dún Aonghasa and its enclosures.