A gazetteer of possible Neolithic and Bronze Age megaliths west of Exmoor — menhirs, one estuary row, Lundy, and the stones that only might be
Coloured dots are standing-stone records (confidence). Pale diamonds are 1830s tithe field-names — Longstone, Greystone, Hoarstone and the like. Those names may mark a prehistoric stone, a medieval bond-stone, or only a name; they are a separate layer, not dated monuments. West Middleton is scheduled as post-medieval staddle stones, not a prehistoric row.
North West Devon is not Dartmoor, and it is not Exmoor either. It has no Grey Wethers, no Merrivale, no Spinster’s Rock, and it does not collect the minilith settings and peat-buried rows that take over once you cross into the National Park. What it has is a thin scatter of quartz and slate uprights on the coastal downs, one buried stone row in the Taw estuary, nine granite menhirs on Lundy, and — as the eastern hinge — the Mattocks Down quartz pair with a lost row and lost circles on the last downs west of the moor.
This gazetteer covers standing stones in North Devon and Torridge west of Exmoor: Hartland to the Taw–Torridge, the Ilfracombe–Lee–Mortehoe coast, the downs as far as Mattocks, Hangman and Kentisbury, plus Lundy. It excludes Dartmoor; Devon Exmoor (Chapman Longstone, White Ladder, Furzehill, Ilkerton Ridge, quincunxes, Brendon); and Somerset Exmoor. Those moorland sites are the same Late Neolithic / Early Bronze Age world, immediately east of this map, and belong in an Exmoor gazetteer.
The list was drawn in 2026 from Historic England scheduled-monument open data, the Devon Historic Environment Record, National Trust heritage records for Morte, Lundy and Arlington, Palmer’s 1937 Ilfracombe survey (table, TDA 69, pp. 494–495, Plate LXIII), Chanter and Worth on Mattocks Down, and nineteenth-century tithe field-names. Grades keep scheduled or accepted megaliths apart from possible, doubtful, lost, folklore and rejected stones. National Trust estate lists of “standing stones” are used as leads, not ingested whole — rubbing-stones, waymarkers and field furniture sit in those lists beside genuine uprights.
The diamond overlay is the tithe-name layer: plot names from the 1830s apportionments (Long Stone, Grey Stone, Hoarstone, White Stone, Broad Stone, Great Stone, Quoit). A name is not a date. Some will be prehistoric standing stones; many will be medieval markers or later agricultural stones. Field-precision pins are used where a grid is known; otherwise the name is shown at the parish centre.
Almost every quartz stone on the Ilfracombe–Lee–Mortehoe downs in this gazetteer was first written up by Mervyn G. Palmer of the Ilfracombe Field Club, in ‘Standing Stones of the Ilfracombe District’, Transactions of the Devonshire Association 69 (1937). He numbered the Lee stones, measured in feet and inches, and noticed quartz in the hedges. He was writing before the National Grid, so his positions are estimates of latitude and longitude; Fowler (1993) warned that the descriptions are generally good and the grids only approximate. The Ordnance Survey, walking the same ground in the early 1950s, was colder: a large part of his forty-odd sites were field-clearance, rubbing-stones or bound-stones. That scepticism is useful and is not the last word. Whitestone, the Damage Barton three, Mattocks Down and the Gospel Stone survive the cut. Wrafton does not. Unmatched Palmer map numbers are kept as their own pins, graded possible or doubtful, not promoted to certain.
Almost none of these stones have a scientific date. Historic England’s scheduling formula for standing stones is “Late Neolithic to the end of the Bronze Age, for the few excavated examples.” The only well-dated stone row in Devon is Cut Hill on Dartmoor, which is mid-Neolithic — earlier than the conventional Late Neolithic / Early Bronze Age bracket used for most Dartmoor and Exmoor monuments. Yelland produced Mesolithic to Early Bronze Age occupation debris beside the stones, not a construction date. Treat every “Neolithic” label here as a possibility, not a result. The honest statement is: possible Late Neolithic / Early Bronze Age ceremonial stones, in a landscape that also contains rubbing-stones, bound-stones, erratics and folklore boulders.
Certain menhirs of real size are few: Whitestone (2.74 m of veined quartz) above Lee; the Gospel Stone / Greystone (2.13 m vein quartz) above Woolacombe; the three quartz stones north of Damage Barton; the Mattocks Down quartz pair (fallen 2.8 m and standing companion, with a lost 1630 row and lost circles — the eastern hinge of this map); Pilton Longstone (moved, holed, listed); and the Lundy group of nine. Yelland is the only excavated lowland row, now under silt. Two alleged cromlechs at Mortehoe (Woolacombe Hill and the Warren on Morte Point) are in the HER as chambered tombs; the OS in 1953 and every survey since have called them natural. They stay in the list as doubtful. Palmer’s remaining records are kept, graded, and not mixed with the scheduled monuments.