Caution in Attributing the Fremington Clay Series to Irish Sea Glaciation

A Case for Predominantly Fluvial and Periglacial Origins of the Fremington Clays, North Devon

Tim Daw

Cannings Cross, Wiltshire, UK

tim.daw@gmail.com

© Tim Daw 2026. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

Note: This paper is a revised and expanded version of an earlier synthesis published in 2025 (Daw, 2025; DOI: 10.13140/RG.2.2.22035.34089). The original intention was to defer any substantial update until full petrographic results for Taylor’s No. 7 hyalopilitic andesite — rediscovered in 2025 at Combrew Farm — were available, since that analysis bears directly on the question of distant versus local provenance for the most potentially significant exotic clast in the Fremington inventory. However, the degree of critical interest the original paper has attracted, particularly regarding the ice-dam height argument and its implications for the Ramson Cliff erratic, suggested that an interim revision incorporating the composite dam model and the findings of Daw et al. (2026) was warranted rather than leaving those points unaddressed. The petrographic results for No. 7, when available, will be reported separately and may prompt further revision.

Fremington Area Superficial Geology — light blue: till which contains the Fremington Clays; the Bickington–Hele ridge lies between them and the estuary. From https://geologyviewer.bgs.ac.uk/

Abstract

The Fremington Clay Series of north Devon has been central to debates on the extent of Middle Pleistocene glaciation in south-west England, often interpreted as evidence of Irish Sea ice incursion during the Wolstonian Stage (MIS 6). Undisputed glacial-marginal and ice-rafted processes demonstrably reached the present north Devon coastline, with associated deposits and erratics at low elevations (~10–30 m OD) and glaciofluvial outwash influence extending to ~55 m OD on the pre-existing Bickington–Hele bedrock ridge. However, stratigraphic, sedimentological, petrological, geomorphological, and chronological evidence — now reinforced by Daw et al. (2026) on the Ramson Cliff erratic (~80 m OD, for which glacial emplacement is considered highly improbable) and the 2025 rediscovery and re-examination of key clasts — warrants caution against designating the series as direct till from a high-standing ice sheet.

The deposits primarily reflect fluvial and paraglacial sedimentation in an ice-marginal Taw–Torridge setting, with overwhelmingly local (Dartmoor/Cornubian) provenance for embedded erratics. Sparse “exotic” signals — small clasts with possible Irish Sea affinities (including Taylor’s No. 7 hyalopilitic andesite) and abraded Irish Sea-type foraminifera — are best explained as dropstones or reworked rafted debris delivered into a proglacial lake at ~30 m OD. A refined composite ice-sediment dam model, drawing on modern proglacial analogues from New Zealand, Svalbard, and Patagonia, demonstrates that lake impoundment at this level can be achieved by a modest marginal ice lobe augmented by high sediment flux from periglacial catchments and confined by pre-existing topography, without requiring extensive onshore overriding. The ephemeral nature of such estuarine barriers explains the absence of a prominent terminal moraine while preserving perched glaciofluvial witnesses such as the Hele gravels. This framework confines significant glacial influence to the coastal fringe and aligns with offshore Bristol Channel evidence.

Keywords: Fremington Clay, low-level glaciation, fluvial-periglacial deposition, erratic provenance, ice-rafting, composite ice-sediment dam, proglacial lake dynamics, north Devon Quaternary.

1. Introduction

The Quaternary glacial history of south-west England remains contentious, particularly regarding the southerly limits of Irish Sea ice. Offshore evidence from the Celtic Sea and Western Approaches — including ice-rafted detritus records and glacigenic sediment distributions documented by Scourse et al. (2009) and synthesised by Gibbard & Clark (2011) — indicates a long-lived ice margin near 51°N during multiple cold stages, with Wingfield (1995) modelling sea-level and ice-margin interaction in the Irish and Celtic Seas. Recent methodological reviews (Lee et al., 2025) further emphasise the fragmentary nature of onshore glacial evidence in south-west England and the importance of distinguishing “limits of preserved evidence” from actual ice-sheet margins, reinforcing the need for caution when interpreting deposits such as the Fremington Clay Series. In north Devon, glacial-marginal or ice-rafted processes clearly reached the present coastline, depositing erratics and glacigenic materials on platforms and in low-elevation sequences around Saunton, Croyde, and the Taw estuary (~10–30 m OD). The Fremington Clay Series (up to ~30 m thick) lies at these low levels, while capping sands and gravels on the Bickington–Hele ridge reach ~55 m OD. These features confirm ice or floating ice impinging on the coast.

However, the extent of inland overriding remains disputed. This revised synthesis argues for continued caution in attributing the Fremington Clay Series to direct Irish Sea till from a high-standing ice sheet. It integrates new data: the re-evaluation of the Ramson Cliff erratic at ~80 m OD on Baggy Point (Daw et al., 2026 — for which glacial emplacement is now considered highly improbable), the 2025 rediscovery of Taylor’s No. 7 hyalopilitic andesite (pending full petrography), and a fully developed model of composite ice-sediment damming that explains the reconstructed ~30 m lake level without requiring extensive onshore ice. The series is interpreted as predominantly fluvial and paraglacial, with local sources dominant and any exotic signals compatible with limited rafting into a proglacial lake.

2. Historical Interpretations

The Fremington Clay has attracted geological attention since Maw (1864) first described it as a “boulder-clay,” and it is important to recognise that fluvial-lacustrine interpretations of the deposit are not revisionist: they predate the glacial consensus. Prestwich (1892) provided a foundational fluvial-estuarine model, describing the clay as an overbank accumulation in a river-fed embayment of the Taw estuary, characterised by fining-upward sequences from subangular local gravels to stoneless silts. This view aligned the deposit with raised beaches at 15–20 m OD and emphasised its confinement to the valley floor without evidence of widespread ice override. Ussher (1878) interpreted the underlying gravels as Taw River alluvium, highlighting the absence of exotic clasts or shear fabrics that might imply glacial transport.

Mid-twentieth-century reappraisals refined this fluvial paradigm. Balchin (1952) reframed the clay as an alluvial infill in a periglacial floodplain, underscoring its red-brown matrix, homogeneous texture, and lateral pinch-out as signatures of terrestrial reworking rather than glaciomarine diamicton. Mitchell (1960) acknowledged hybrid elements but prioritised fluvial origins for the basal units. Edmonds (1972) advanced a non-glacial model for the pebbly drifts at Fremington Quay, viewing them as solifluction reworked by Ipswichian floods into river terraces. These interpretations challenged Zeuner’s (1959) bottom-moraine proposal by emphasising paraglacial drainage diversions in the Taw–Torridge basin.

The glacial paradigm was revived by Stephens (1966, 1970), who interpreted the clay as Wolstonian till, and consolidated by Kidson & Wood (1974). However, the revival was persistently muddied by confusion between the sparse erratics documented in situ within the clay and the more numerous, far-travelled ice-rafted boulders on adjacent beaches at Saunton and Croyde. Taylor’s (1956) catalogues exacerbated this by grouping “Saunton and Fremington erratics” indiscriminately, amplifying onshore ice narratives without distinguishing the coastal boulders’ subrounded, striated forms from the clay’s subangular, aureole-affine pebbles at depth. This lumping overlooked elevation mismatches and transport vectors.

Clarification emerged through targeted reappraisals that disentangled these suites. Everard et al. explicitly refuted glacial linkages, noting that since the Fremington boulder clay overlies the equivalent of the raised beach, it cannot have been responsible for the coastal erratics at Croyde and Saunton, attributing the latter to ice-floe rafting. Madgett & Inglis (1987) surveyed 37 Saunton–Croyde boulders, correcting Taylor’s misidentifications and differentiating them as sea-ice proxies from the clay’s solifluction terraces. Modern syntheses, including Harrison (1997) in the Geological Conservation Review and Bennett et al. (2024), reinforce this resolution, portraying the clay as a continuous 4 km fluvial body with pseudo-laminated fines, while coastal erratics reflect Celtic Sea calving.

3. Stratigraphy and Sedimentology

The Fremington Clay Series overlies bedrock or basal gravels at low elevations (~10–26 m OD in exposures). Croot et al. (1996) carried out the most detailed modern investigation of the sequence, including a purpose-dug excavation from surface to bedrock at the former Brannam’s Pottery Clay Pits and Higher Gorse Claypits (SS529317). They delineated five units (Table 1). The upper stoneless clay — sometimes called the potter’s clay, after the Brannam’s and Fishley potteries that quarried it commercially — exhibits horizontal to pseudo-laminated bedding, fining-upward trends, and weak clast fabrics inconsistent with subglacial lodgement till. They are, however, entirely compatible with low-energy fluvial or glaciolacustrine settling in a valley-confined proglacial lake at ~30 m OD — which is the depositional environment that the bulk of the sedimentological evidence supports, and which the present paper accepts as the most parsimonious interpretation for the upper clay units.

Table 1. Fremington Clay Series stratigraphy (adapted from Croot et al. 1996).

Unit	Description	Thickness	Key Features	Interpretation
E (Head)	Gravelly sand/clay; angular local clasts	1–1.5 m	Cryoturbated; gradational base	Periglacial solifluction (Devensian+)
D	Clast-rich weathered red clayey silt	0.5–1.0 m	Over-consolidated; CaCO₃ 10–20%	Weathered glaciolacustrine/fluvial
C	Irregular sand/silt lenses; reworked fossils	2–2.5 m	Sharp contacts; no strong grading; OSL >26 ka BP (minimum)	Ice-proximal fluvial sands
B	Dark brown clay; stoneless base to clast-rich top	8–9 m	Pseudo-laminae; weak fabrics; >1,500 clasts analysed	Low-energy lacustrine; episodic flood inputs
A (Basal)	Clast-supported subangular gravels	1.5–2.0 m	Weak imbrication; local clasts	High-energy fluvial/proglacial outwash

 

Croot et al. examined more than 1,500 clasts in the 16–256 mm size range, extracted from grab samples at various levels within Unit B. All but a single exception could be accounted for in the bedrock geology within a 10 km radius of the site. That exception — a cobble-sized clast at approximately 5 m depth displaying typical flat-iron subglacial morphology with exceptionally well-striated faces — represents the only unequivocally glacially-transported in-situ clast recovered from the clay pits. Croot et al. were explicit about the weight this single find carried in their interpretation: without it, they stated, they “would have been forced to consider a much wider range of possible origins for the Fremington Clay Series.” The glacial character of the deposit thus rests, in the primary investigators’ own assessment, on a single clast rather than on the bulk sedimentology.

The critical question is not whether a proglacial lake existed — the fining-upward sequence, dropstones, and laminated silts make that case persuasively — but what kind of ice margin was needed to dam it, and what that implies for the lateral extent of glaciation along the open coast.

4. Provenance of Erratics: Altitude and Source Constraints

A fundamental observation is the consistent low elevation of in-situ erratics within the Fremington Clay: ~10–26 m OD in clay pits and cuttings, as recorded by Taylor (1956) and Arber (1964). These elevation data predate modern differential GPS levelling, and the original Ordnance Datum benchmarks used by Taylor and Arber carry uncertainties of ±1–2 m. Updated fieldwork in 2025 confirmed the approximate range by reference to current OS mapping. Coastal erratics on Saunton–Croyde platforms and in head deposits occur at or near sea level to slightly elevated positions consistent with raised beaches or periglacial reworking. No verified erratics in the main clay body exceed ~30 m OD.

The Hele gravels capping the Bickington–Hele bedrock ridge to ~55 m OD contain erratic clasts deposited by meltwater streams linked to the same marginal system. These include igneous types (dolerites, possible andesites) alongside dominant local Devonian–Carboniferous and Dartmoor-derived material. No detailed petrographic inventory exists specifically for the ridge gravels, but the assemblage overlaps with the sparse, mixed-provenance clasts in the lower clay and reflects fluvial transport and abrasion rather than direct ice rafting or high-level till.

Coastal versus inland erratics

A critical distinction must be drawn between the sparse erratics documented in situ within the Fremington Clay itself and the more numerous, far-travelled ice-rafted boulders on adjacent north Devon beaches. The latter, often lumped together in glacial models, include unambiguous Scottish and Irish Sea lithologies (e.g., Ailsa Craig microgranite, Purbeck flint) deposited via sea-ice rafting or storm transport during lowstands. Madgett & Inglis (1987) surveyed 37 Saunton–Croyde boulders, correcting earlier misidentifications and demonstrating that these coastal boulders are subrounded, striated, and clustered in head or beach gravels — quite distinct from the clay’s subangular, aureole-affine pebbles found at depths of 2–22 ft within the clay body. Taylor’s (1956) indiscriminate grouping of “Saunton and Fremington erratics” has long obscured this distinction, inflating the apparent exotic component of the inland deposit by conflation with coastal material of entirely different transport history.

Petrological summary

Petrological inventories of clay-embedded erratics (Table 2) list igneous and metamorphic types (dolerite, granophyre, andesite) amid dominant local Devonian–Carboniferous clasts (>99%; Croot et al., 1996). Dewey (1910) and Taylor (1956) provide detailed thin-section analyses confirming igneous dominance (spilitic textures in No. 6, ophitic in No. 10) with local affinities — Cornish spilites, Devon dykes — while noting morphological resemblances to Scottish types without geochemical confirmation. Quartz-dolerites and olivine-dolerites match Meldon Chert Formation dykes; hypersthene-andesites and granophyres evoke Dartmoor elvans and aureole rocks, distinguishable from Irish Sea equivalents via mineralogy (e.g., titaniferous augite in alkali micro-dolerites; Gilbert, pers. comm. in Croot et al., 1996). No. 8, an overlooked altered quartz porphyry from Fishley, exemplifies potential aureole sourcing, with epidote and apatite evoking Variscan intrusions mobilised via Taw floods.

Other clasts (spilites, dolerites, quartz porphyry) align with Dartmoor aureole or Cornubian sources (Madgett & Inglis, 1987). Sparse possible Irish Sea-affinity lithologies (<1%) fit occasional dropstones or reworked rafted debris into the proglacial lake.

A note of caution: No. 7

Taylor’s No. 7 (Combrew Farm driveway wall) warrants particular caution. The clast is a hyalopilitic andesite, ~16 inches across, well-rounded, glassy and brittle, with no augite — closely matching the original Scottish (Dumfries/Argyll) description given by Dewey (1910). Rediscovered in 2025, with full petrography awaited. The mineralogy as described — zoned, twinned acid labradorite, pleochroic hypersthene prisms, and magnetite gridiron in a brown glass base — does appear genuinely consistent with a Scottish volcanic source rather than a Dartmoor elvan or West Devon dyke, and the alternative local attributions proposed for other Fremington erratics sit less comfortably with this particular lithology. Pending geochemical confirmation, No. 7 should be treated as a probable genuinely far-travelled exotic.

Even if the pending analysis confirms a distant Scottish origin, however, the clast would remain one of very few exotics among >1,500 locally derived stones — entirely consistent with delivery as a single dropstone from floating ice calved from a marginal lobe in Barnstaple Bay, rather than requiring direct till transport by an overriding ice sheet.

Erratic inventory

No.	Location	Lithology	Description & notes	Alternative local / regional source	Key references
6	Combrew Farm / Bickington	Spilite (vesicular granophyre)	40×30×25 in; dark grey, porphyritic albite feldspars, micropegmatite groundmass, chlorite-replaced ferromagnesian, secondary granophyric vesicles with calcite. No striae. Isolated in middle of clay-bed.	N. Cornwall spilites (Crinan pillow-lava type) or Dartmoor volcanics	Dewey (1910); Taylor (1956); Arber (1964)
7	Combrew Farm / Chilcotts	Hypersthene-andesite (hyalopilitic)	16 in across; dark grey-green, glassy porphyritic acid labradorite (zoned, twinned), hypersthene prisms (pleochroic), magnetite gridiron in brown glass base; no augite/olivine. ~22 ft below surface, c. 1870.	Dartmoor elvan intrusions or W. Devon dykes	Dewey (1910); Taylor (1956); Arber (1964)
8	Fishley Pottery, nr Combrew	Altered quartz porphyry	47×19×16 in; light grey, holocrystalline granitic texture, phenocrystic quartz/felspar (up to 5 mm); crushed plagioclase, apatite prisms, red amorphous matrix, epidote. From clay-pit.	Porphyritic dyke W. Devon / Cornwall; Dartmoor aureole	Taylor (1956)
9	Brannam’s pits	Quartz-dolerite	c. 300 lb, ellipsoidal; grey, fine-grained, kaolinised felspar laths, primary quartz, fresh reddish augite, apatite needles, magnetite/calcite. In middle of brown clay.	Dartmoor Permian–Triassic dykes (Meldon)	Taylor (1956); Arber (1964)
10	Brannam’s pits	Olivine-dolerite	c. 300 lb, angular; darker grey, micro-pegmatitic ophitic, yellow olivine, ilmenite prisms, plagioclase tabs, slight quartz orientation. In brown clay; common Devon type.	Dartmoor olivine-bearing intrusions	Taylor (1956); Arber (1964)
—	Brannam’s, 17 ft depth	Olivine-dolerite pebble & Carboniferous grit slab	2 in rounded pebble (as No. 10); 5×1.25 in slab with red ferric oxide skin along cracks (post-inclusion infiltration).	Local fluvial rework (pre-embedding waterworn)	Taylor (1956)
13	Brannam’s pits (1962)	Quartz-dolerite	10 ft from top of clay.	Dartmoor dykes	Taylor (1956); Vachell (1963); Arber (1964)
—	Higher Gorse, Plymouth (1994)	Alkali micro-dolerite	Small striated boulder in main clay unit; plagioclase phenocrysts, titaniferous augite, vesicles.	Dartmoor micro-dolerite variants	Croot et al. (1996)
—	Pen Hill, Taw Estuary	Trachy-andesite	Partially buried in beach/estuarine sand (not in situ in clay).	Regional andesitic flows; fluvial rework	Croot et al. (1996)
—	Arber (1964), post-1957	Dolerite and granodiorite	Removed boulders, originally inside clay; later identified.	Dartmoor aureole dolerite / granodiorite	Arber (1964); Wood (1973)
—	General Fremington area	Spilite, grey elvan, quartz/olivine dolerite	Multiple small pebbles (50+), embedded 5–11 ft above base or at top/base.	Dartmoor aureole (elvan, spilite-like volcanics)	Taylor (1956); Arber (1964); Croot et al. (1996)

Table 2. In situ erratics in the Fremington Clay series: lithologies, descriptions, and alternative provenances (updated with Dewey, 1910; Taylor, 1956 records; excludes coastal ice-rafted boulders). Row highlighted in amber = No. 7, treated here as a probable genuinely far-travelled Scottish exotic pending geochemical confirmation.

Interpretation

Sparse exotics (<1% of clasts >1.5 cm) occur as subangular pebbles or rare striated cobbles (e.g., single microdolerite at 4 m depth; Croot et al., 1996), embedded at low elevations (10–26 m OD). Granites match Dartmoor’s Carboniferous pluton, mobilisable via periglacial clitter slopes and Taw entrainment (Evans et al., 2012). Dolerites align with local intrusions, distinguishable from northern equivalents via U-Pb/Hf isotopes — untested on archives (e.g., >1,500 clasts at Plymouth University; Taylor’s thin-sections at Cambridge). Flints and schorlrocks suggest short-distance fluvial/marine reworking, not ice-sheet transport (Daw, 2024a).

Recent syntheses and petrological reappraisals continue to support a predominantly local or regional provenance, with the Dartmoor pluton and its aureole emerging as the most parsimonious source (Bennett et al., 2024). Even for enigmatic types like No. 6, Dartmoor affinities remain viable, with geochemical tracers recommended for confirmation (Daw, 2025a). No. 7 is the strongest candidate for a genuinely distant origin, but its isolation among >1,500 local clasts is itself telling: a single far-travelled dropstone from calved ice is a very different proposition from till transport by an overriding ice sheet.

This profile favours hybrid fluvial–periglacial input: Taw–Torridge floods exported Dartmoor debris alongside local slates, explaining weak NW–SE fabrics without Irish Sea signatures. Rarity of true exotics (no chalk, minimal Scottish gneiss) and lack of concentration gradients refute sheet glaciation (Bennett et al., 2024), particularly when coastal rafted erratics are excluded.

5. The “Exotic” Signal: Foraminifera and Far-Travelled Clasts

While the series is overwhelmingly local in provenance, a subtle “exotic” signal is present and requires careful interpretation. A small number of clasts show possible Irish Sea affinities, including certain quartz-dolerites and hypersthene-andesites (notably Taylor’s No. 7) that parallel material in the Saunton/Croyde coastal deposits. These clasts are typically small pebbles or cobbles rather than large boulders. They are best explained as dropstones delivered by floating ice from a modest ice lobe in Barnstaple Bay, or as minor reworked material from the adjacent sea floor, rather than primary till.

The clay also contains a derived microfauna, including damaged and abraded Irish Sea-type foraminifera. Kidson & Wood (1974) reported up to eleven species, including Ammonia beccarii and Nonion labradoricum, from samples in the Fremington exposures. The diagnostic feature is their physical condition: tests are abraded, broken, and frequently infilled with secondary minerals, indicating significant transport and reworking prior to deposition. This is consistent with incorporation via glacial meltwater or ice-rafting into a freshwater or brackish lake, rather than in-situ glaciomarine deposition in which tests would be expected to show better preservation. The foraminifera therefore record proximity to Irish Sea-derived material without requiring that the ice itself advanced over the site.

6. Chronological Constraints

OSL dating of Units B–C yields ages of >424 ka BP (Croot et al., 1996), placing the primary clay deposition in or before the Anglian (MIS 12) rather than during the Wolstonian (MIS 6) stage to which the Fremington Clay has traditionally been correlated (Stephens, 1970). This has significant implications. If the clay predates the Wolstonian, then the standard model — in which Irish Sea ice advanced into Barnstaple Bay during MIS 6 to deposit the Fremington sequence — requires either that the dating is unreliable or that the deposit records an earlier, Anglian glacial episode.

The stratigraphic position of the clay — overlying Hoxnian-age gravels and underlying Devensian head — is consistent with either an Anglian or Wolstonian attribution, but the OSL ages favour the older assignment. Variability in terrace grading (four levels; Edmonds, 1972) suggests multiple cold phases, with the Fremington Clay potentially representing an Anglian fluvial legacy subsequently reworked during the Wolstonian. This complicates synchronisation with the Scilly “tills” (Devensian at Scilly; Scourse, 1991) and the Trebetherick deposits (locally derived; Wood, 1973), and underlines the danger of assuming a single, unified glacial event across south-west England.

The chronological uncertainty reinforces the case for caution: if the deposit is Anglian rather than Wolstonian, the entire framework of MIS 6 Irish Sea ice reaching north Devon requires reconsideration.

7. The “Missing” Dam: Post-Glacial Erosion and the Hele Gravels

A common critique of the proglacial lake model is the absence of a prominent terminal moraine or dam remnant at the Taw estuary mouth. This absence is, in fact, precisely what modern analogues predict for composite ice-sediment dams in high-energy settings.

At Nordenskiöldbreen in Svalbard, Nehyba et al. (2017) documented a small ice-dammed lake that developed in the early 1990s along the glacier margin in Adolfbukta. The lake was progressively infilled by a Gilbert-type fan delta, with fluvio-deltaic terraces recording multiple lake-level falls, before being largely obliterated by deglaciation, erosion, and fluvial redistribution within approximately two decades. This case demonstrates how rapidly an ice-sediment barrier can form, aggrade, and then breach or disperse once the ice nucleus retreats and high-energy fluvial processes resume — a process highly relevant to the post-glacial fate of any composite dam in the high-energy Taw estuary.

The primary surviving evidence of the dam’s effective height is preserved not at the estuary mouth but on higher ground to the south: the Hele gravels, which cap the pre-existing Bickington–Hele bedrock ridge at ~55 m OD. These perched glaciofluvial deposits, laid down by meltwater streams linked to the same marginal system, were high enough above the valley floor to escape the main phase of post-glacial fluvial incision. They contain erratic clasts (including igneous types such as dolerites and possible andesites) alongside dominant local Devonian–Carboniferous and Dartmoor-derived material. The basal Unit A gravels of the Fremington sequence itself represent the “roots” of the dam: the initial high-energy outwash phase before the basin deepened into a more tranquil lake environment.

8. Composite Ice-Sediment Dams: The Modern Evidence

The standard mental model of an ice-dammed lake is a clean wall of glacier ice blocking a valley, with the lake surface pressing directly against the ice face. In this model, the dam height is simply the ice thickness, and the ice surface must exceed the lake surface to prevent overtopping. This is the model implicitly invoked when critics argue that a ~30 m lake at Fremington requires ~30 m or more of solid ice — and therefore an ice surface capable of reaching the ~80 m Ramson Cliff erratic on Baggy Point.

Modern proglacial geomorphology shows that this clean-ice model is the exception rather than the rule. In most settings where glaciers interact with rivers carrying significant sediment loads, the actual barrier is a composite structure: ice, outwash gravel, glaciofluvial sand, debris flows, and — in marine-terminating cases — subaqueous fans. The effective dam height is the combined elevation of all these components, not the ice thickness alone.

8.1 The New Zealand Last Glacial Maximum: Outwash Fan-Head Damming

The most comprehensive modern analogue comes from the Southern Alps of New Zealand, where Sutherland et al. (2019) reconstructed the ice-contact proglacial lake systems associated with the Last Glacial Maximum across the entire mountain range. Their central finding is that the major LGM lakes — Tekapo, Pukaki, Ohau, Wanaka, Hawea, and Wakatipu — were not dammed primarily by moraine ridges or by monolithic ice barriers, but by outwash fan-heads: massive aggradational gravel bodies built by high-sediment-load braided rivers against and around the ice margin.

Sutherland et al. (2022) developed this model in detail at Lake Tekapo, where the glacier terminus was entirely buried by the outwash fan-head, with no terminal moraine visible as a discrete landform. The dam that created the lake — which persists today, long after the ice vanished — was the sedimentary mass itself. Shulmeister et al. (2019) provide wider context, noting that this pattern is typical across the western South Island, where high sediment supply overwhelmed the capacity of terminal moraines to form as distinct features.

The key insight for the Fremington debate is that these New Zealand lakes stood at substantial depths behind barriers whose ice component was only a fraction of the total dam height. Once the ice melted, the lakes persisted — Tekapo, Pukaki, and Ohau still exist today — held in by the residual sediment mass. The permanence of the lakes derives from the sediment, not the ice.

8.2 High-Arctic Svalbard: Rapid Infilling and Stabilisation

The Svalbard example, already cited in Section 7 for the ephemeral nature of ice-sediment barriers, also illustrates the positive side of composite damming. At Nordenskiöldbreen, fluvial sediment rapidly aggraded against the ice barrier, building a composite dam whose crest exceeded the ice surface in places. The effective barrier height fluctuated as sediment accumulated, eroded, and redistributed. At no point was the dam a simple wall of ice with a measurable freeboard; it was a dynamic, composite structure in which ice and sediment alternately dominated.

8.3 Perito Moreno, Patagonia: Ice, Debris, and Dynamic Water Levels

The Perito Moreno Glacier in southern Patagonia provides perhaps the most dramatic modern example of composite damming. The glacier periodically advances across a narrow strait to dam Brazo Rico, a branch of Lago Argentino, raising its water level by up to 20 m above the main lake before catastrophic drainage ensues. Crucially, the lake level during damming episodes rises well above what the subaerial ice thickness at the narrow strait would predict, precisely because the underwater and debris-armoured components of the barrier contribute to the effective seal. Leakage occurs not by overtopping but by subglacial drainage when the hydraulic head exceeds the ice overburden pressure — a process described in the broader GLOF literature by Carrivick & Tweed (2013).

8.4 A General Principle

These three examples — from a temperate maritime mountain belt, a High-Arctic archipelago, and a Patagonian ice field — illustrate a general principle well established in the proglacial lake and GLOF literature but not previously applied to the Fremington debate: in any setting where rivers carry significant sediment loads to an ice margin, the resulting dam is composite rather than monolithic. Carrivick et al. (2022) synthesise this principle in their review of coincident glacier and lake evolution across the Southern Alps, showing that sediment flux is as important as ice dynamics in determining lake existence, extent, and longevity.

9. Application to the Taw Estuary

9.1 Sediment Supply from a Periglacial Hinterland

The Taw and Torridge drain catchments that include the northern margins of Dartmoor and the extensive periglacial plateau surfaces of Exmoor. Evans et al. (2012) describe Dartmoor during Pleistocene cold stages as an independent ice cap with extensive periglacial slopes generating clitter fields, solifluction mantles, and thick head deposits. The subtle moraines at Slipper Stones imply thin, cold-based ice (<50 m thick), enhancing tors and dry valleys via frost action rather than erosion. Edmonds (1972) documents the terrace stratigraphy of the Taw valley, recording multiple episodes of fluvial aggradation consistent with high-sediment-load braided river systems.

The rivers feeding the Fremington lake basin were not clear-water streams draining stable, vegetated catchments. They were periglacial braided rivers with extremely high bedload transport rates, carrying gravel, sand, and silt derived from frost-shattered bedrock, solifluction deposits, and reworked older drift. The analogy with New Zealand’s high-sediment-load glacial rivers is direct: these are exactly the conditions under which outwash fan-heads build against ice margins.

9.2 The Composite Dam Model for Fremington

Applying the composite dam model to the Taw estuary, the scenario runs as follows. An Irish Sea ice lobe advances into Barnstaple Bay, blocking the combined outflow of the Taw and Torridge. The ice margin need not be a towering cliff; a modest, debris-charged lobe grounding in the relatively shallow waters of the inner bay would suffice. As the rivers back up, sediment-laden flow deposits outwash fans against the upstream (southern) face of the ice barrier. Gravel and sand aggrade rapidly, building fan-head surfaces that supplement the ice and raise the effective dam height.

The Fremington Clay itself records this process. The basal gravels represent the initial outwash phase. The overlying laminated silts and sands record a transitional phase as the basin deepened and the dam became more effective. The uppermost stoneless clay represents full lacustrine conditions, with suspension settling of the finest fraction in a quiet, deep-water environment. Dropstones from floating ice derived from the margin punctuate this otherwise tranquil record.

In this model, the 30 m lake surface does not require 30 m of solid ice. It requires a composite barrier — ice plus aggraded sediment plus debris — whose combined crest reached or exceeded 30 m OD at the topographically constricted estuary mouth. The pre-existing Bickington–Hele ridge at ~55 m OD provided lateral confinement on the southern side. The ice component of the dam may have been substantially less than 30 m thick, supplemented by the very outwash and debris that the Fremington sequence itself records.

9.3 The Self-Reinforcing Dam: Sediment as Both Product and Barrier

This interpretation is not ad hoc. The fining-upward sequence of the Fremington Clay Series — basal gravels to laminated silts to stoneless clay — is precisely what the outwash fan-head model predicts. In the New Zealand analogues, Sutherland et al. (2022) describe ice-contact lakes with identical stratigraphic signatures. The Fremington Clay is the sedimentary product of the process that created its own dam.

This self-reinforcing dynamic is a standard feature of composite ice-sediment dams. The dam grows in effectiveness over time without requiring any increase in ice thickness. A modest ice lobe that initially created a shallow, gravel-floored pond could, through progressive outwash aggradation, give rise to a deep lacustrine basin over decades or centuries. The Hele gravels at ~55 m OD, perched on the bedrock ridge above the zone of later fluvial incision, are the surviving high-water witnesses of this process.

10. Implications for the Ramson Cliff Erratic

The composite dam model has a direct and important consequence for the debate over the Ramson Cliff erratic. If a 30 m lake can be explained by a modest ice lobe at the estuary mouth, supplemented by outwash aggradation and laterally confined by pre-existing topography, then the lake level tells us nothing about the height, thickness, or lateral extent of ice on the open coast to the west.

The Ramson Cliff erratic sits at approximately 80 m OD on Baggy Point, a fully exposed coastal headland several kilometres west of the Taw estuary. Its petrography — an altered epidiorite or greenstone of approximately 700 kg — has been re-examined by Daw, Ixer & Madgett (2026), who argue that it aligns with local Cornubian or Dartmoor sources rather than distant Irish Sea material. But even setting aside the provenance question, the erratic at 80 m OD is not explained by a composite dam at 30 m in the valley below.

The argument that “the ice must have been higher than the lake” is true in a trivial sense — the ice surface at the dam must exceed the water surface, or the lake drains. But the dam was at the estuary mouth, in a topographically constricted setting. The ice there could have been 35 m thick (to maintain a few metres of freeboard above a 30 m lake) without extending laterally at that thickness along 10 km of open coastline to reach Baggy Point at 80 m. A lobate ice margin blocking an estuary is not a uniform ice sheet overriding a headland.

The absence of intermediate-elevation erratics between the ~30 m clay body and the ~80 m Ramson Cliff boulder reinforces this point. If ice were thick enough and laterally extensive enough to reach 80 m on Baggy Point, a scatter of erratics at 40 m, 50 m, 60 m, and 70 m along the intervening coastline would be expected. No such scatter has been identified despite over a century of fieldwork.

11. Discussion: Separating Dam Height from Ice Extent

The confusion at the heart of the ice-dam height argument is a failure to distinguish between two very different physical situations: the height of a composite barrier at a topographically constricted point, and the regional extent and thickness of an ice mass across open terrain.

In every modern analogue considered here — New Zealand, Svalbard, Patagonia — the dam height at the constriction point exceeds the ice thickness by a substantial margin because of sediment aggradation. The ice does not need to be thick everywhere, or present everywhere, or even the dominant component of the dam. It needs to provide a nucleus around which sediment accumulates.

The Fremington case is a near-ideal candidate for composite damming. The Taw–Torridge system drains a periglacial hinterland with prodigious sediment supply. The estuary mouth is a topographic constriction. The sedimentary record preserves a classic fining-upward proglacial sequence consistent with progressive dam growth. And the absence of a discrete moraine at the dam site is exactly what the New Zealand model predicts: in high-sediment-load settings, moraines are subdued or buried, and the dam is the outwash mass itself.

None of this denies the presence of Irish Sea ice in Barnstaple Bay. A composite dam still requires an ice nucleus. What it denies is the extrapolation from local dam height to regional ice extent. A 30 m lake in the Taw valley is consistent with a modest, topographically controlled ice lobe, not with a thick onshore ice sheet capable of overriding headlands at 80 m.

The same logic applies to the question of inland penetration. If the ice dam sat at the estuary mouth rather than deep within the valley system, the Fremington Clay records the lake that formed behind the barrier, not the advancing front of a glacier. The deposits are confined to a low-level bench along the southern side of the Taw estuary between Barnstaple and Instow, with no continuation of till or glacigenic sediment southward up the Taw valley. No erratics have been reported at or south of Barnstaple itself. The Hele gravels at ~55 m OD represent glaciofluvial outwash — material deposited by meltwater — not direct till from an overriding ice mass.

This interpretation sits comfortably within the broader pattern recognised by Lee et al. (2025), who highlight preservation biases and the predominance of low-elevation, ice-marginal or ice-proximal signatures in the southwest peninsula rather than widespread overriding by a thick ice sheet.

Croot et al. found that the Fremington Clay is mildly over-consolidated, with pre-consolidation pressures of 250–350 kPa. They considered this consistent with loading by an overriding ice mass, but were careful to note that fast-ice or a floating ice shelf could produce similar effects on coastal estuarine deposits. The composite dam model offers a further possibility: the over-consolidation may reflect the weight of aggraded outwash and debris on the upstream face of the barrier.

The regional context reinforces this interpretation. Even on the most generous reconstruction of Irish Sea ice extent in the Bristol Channel, onshore glacial evidence on the English side is confined to low elevations at the coastal fringe. The GCR review records a broad consensus that Somerset itself was not glaciated, while accepting glaciation of the Avon coastlands in the Kenn lowlands and Vale of Gordano at low elevations. The glacial limit in Sedgemoor appears to lie between Greylake and Langport Railway Cutting 6 km further south, which is erratic-free. Even the enigmatic deposits on Bleadon Hill at 82 m OD on the Mendip flank lack demonstrably glacially transported erratic material. The Fremington deposits thus sit within a consistent regional pattern: Irish Sea ice impinged on low-lying coastal areas and estuaries but did not climb significantly above sea level, penetrate far inland, or override higher ground.

12. Conclusion

The Fremington Clay Series records a modest ice-marginal setting in which fluvial and paraglacial processes dominated. The sparse Irish Sea lithologies and abraded foraminifera indicate proximity of a limited ice lobe in Barnstaple Bay but do not require voluminous direct till or extensive onshore overriding. The reconstructed ~30 m proglacial lake is explained by a composite ice-sediment dam — a modest ice nucleus augmented by rapid outwash fan-head aggradation from periglacial rivers and laterally confined by the pre-existing Bickington–Hele ridge — rather than a monolithic wall of ice.

The ephemeral nature of such estuarine barriers, demonstrated by modern analogues from Svalbard, explains the absence of a prominent terminal moraine at the estuary mouth. The surviving Hele gravels at ~55 m OD preserve the perched meltwater component of the dam system. The Ramson Cliff erratic at ~80 m OD on Baggy Point — petrographically local, lacking any supporting suite of high-level deposits or erratics at intermediate elevations — remains an isolated outlier whose glacial emplacement is highly improbable. Chronological constraints (OSL ages >424 ka BP) further complicate the standard Wolstonian attribution, raising the possibility that the deposit records an Anglian rather than MIS 6 event.

Elevation data remain central to this interpretation: low-level erratics and deposits (~10–30 m OD) support coastal glacial influence; glaciofluvial outwash reaches ~55 m OD on higher ground; the 80 m outlier does not fit. Dam height in the valley and the lateral extent of onshore ice are separate questions with separate answers. Ongoing geochemical analysis of archival clasts — including Taylor’s No. 7 — and potential further study of the Hele gravels will continue to refine provenance and test these interpretations against the primary evidence.

This interpretation confines significant glaciation to the coastal fringe and has implications for broader claims of glacial transport across southern England, including the contested theory that Irish Sea ice carried the Stonehenge bluestones from Wales. If the ice did not climb above ~30 m OD even in the confined setting of the Taw estuary, a fortiori it did not reach Salisbury Plain.

The story advances through careful scrutiny of the rocks, the topography, and the modern analogues — not through assumptions about uniform ice dams.

 

References

Arber, M.A. (1964). Erratic boulders within the Fremington Clay of North Devon. Geological Magazine, 101(3), 282–283. https://doi.org/10.1017/S0016756800049517

Balchin, W.G.V. (1952). The erosion surfaces of Exmoor and adjacent areas. The Geographical Journal, 118(4), 453–476. https://doi.org/10.2307/1791282

Bennett, J.A., Cullingford, R.A., Gibbard, P.L., Hughes, P.D. & Murton, J.B. (2024). The Quaternary geology of Devon. Proceedings of the Ussher Society, 15, 84–130. https://ussher.org.uk/wp-content/uploads/benettetal1584130v2.pdf

Campbell, S., Hunt, C.O., Scourse, J.D., Keen, D.H. & Stephens, N. (eds) (1998). Quaternary of South-West England. Geological Conservation Review Series No. 14. Chapman & Hall, London (JNCC).

Carrivick, J.L. & Tweed, F.S. (2013). Proglacial lakes: character, behaviour and geological importance. Quaternary Science Reviews, 78, 34–52.

Carrivick, J.L., Sutherland, J.L., Huss, M., Purdie, H., Stringer, C.D., Grimes, M., James, W.H.M. & Lorrey, A.M. (2022). Coincident evolution of glaciers and ice-marginal proglacial lakes across the Southern Alps, New Zealand: Past, present and future. Global and Planetary Change, 211, 103792.

Croot, D.G., Gilbert, A., Griffiths, J. & van der Meer, J.J. (1996). The character, age and depositional environments of the Fremington Clay Series, North Devon. In: Charman, D.J., Newnham, R.M. & Croot, D.G. (eds), The Quaternary of Devon and East Cornwall: Field Guide. Quaternary Research Association, London, pp. 14–34. (Related note in Quaternary Newsletter 80.)

Daw, T. (2025). Caution in Attributing the Fremington Clay Series to Irish Sea Glaciation: A Case for Predominantly Fluvial and Periglacial Origins in North Devon. DOI: 10.13140/RG.2.2.22035.34089.

Daw, T., Ixer, R. & Madgett, P. (2026). A review of the Ramson Cliff erratic: evidence of high-level ice flow? Quaternary Newsletter, 167, 13–19. https://doi.org/10.64926/qn.20517

Dewey, H. (1910). Notes on some igneous rocks from North Devon. Proceedings of the Geologists’ Association, 21(4), 429–434.

Edmonds, E.A. (1972). The Pleistocene history of the Barnstaple area. Report of the Institute of Geological Sciences, 72/2. HMSO, London.

Evans, D.J.A., Harrison, S., Vieli, A. & Anderson, E. (2012). The glaciation of Dartmoor: the southernmost independent Pleistocene ice cap in the British Isles. Quaternary Science Reviews, 45, 31–53.

Gibbard, P.L. & Clark, C.D. (2011). Pleistocene glaciation limits in Great Britain. In: Ehlers, J., Gibbard, P.L. & Hughes, P.D. (eds), Quaternary Glaciations — Extent and Chronology: A Closer Look. Elsevier, pp. 75–93.

Kidson, C. & Wood, R. (1974). The Pleistocene stratigraphy of Barnstaple Bay. Proceedings of the Geologists’ Association, 85(2), 223–237.

Lee, J.R., et al. (2025). Refining the known extent of major onshore Quaternary glaciation in the UK — Types of evidence, nomenclature and uncertainty. Proceedings of the Geologists’ Association. https://doi.org/10.1016/j.pgeola.2024.12.002

Madgett, P.A. & Inglis, E.A. (1987). A re-appraisal of the erratic suite of the Saunton and Croyde areas. Transactions of the Devonshire Association, 119, 135–144.

Maw, G. (1864). On a supposed deposit of boulder-clay in North Devon. Quarterly Journal of the Geological Society of London, 20, 445–451.

Mitchell, G.F. (1960). The Pleistocene history of the Irish Sea. Advancement of Science, 17, 313–325.

Nehyba, S., Hanáček, M., Engel, Z. & Stachoň, Z. (2017). Rise and fall of a small ice-dammed lake — Role of deglaciation processes and morphology. Geomorphology, 295, 662–679. https://doi.org/10.1016/j.geomorph.2017.08.019

Prestwich, J. (1892). The raised beaches and ‘head’ or rubble drift of the south of England. Quarterly Journal of the Geological Society of London, 48, 263–343.

Scourse, J.D. (1991). Late Pleistocene stratigraphy and palaeobotany of the Isles of Scilly. Philosophical Transactions of the Royal Society of London Series B, 334, 405–448.

Scourse, J.D., Haapaniemi, A.I., Colmenero-Hidalgo, E., Peck, V.L., Hall, I.R., Austin, W.E.N., Knutz, P.C. & Zahn, R. (2009). Growth, dynamics and deglaciation of the last British–Irish ice sheet: the deep-sea ice-rafted detritus record. Quaternary Science Reviews, 28, 3066–3084.

Shulmeister, J., Thackray, G.D., Rittenour, T.M., Fink, D. & Evans, D.J.A. (2019). The Last Glacial Maximum (LGM) in western South Island, New Zealand: implications for the global LGM and MIS 2. Quaternary Science Reviews, 213, 44–66.

Stephens, N. (1966). Some Pleistocene deposits in North Devon. Biuletyn Peryglacjalny, 15, 103–114.

Stephens, N. (1970). The West Country and Southern Ireland. In: Lewis, C.A. (ed.), The glaciations of Wales and adjoining regions. Longman, pp. 267–314.

Sutherland, J.L., Carrivick, J.L., Shulmeister, J., Quincey, D.J. & James, W.H.M. (2019). Ice-contact proglacial lakes associated with the Last Glacial Maximum across the Southern Alps, New Zealand. Quaternary Science Reviews, 213, 67–92.

Sutherland, J.L., Evans, D.J.A., Carrivick, J.L., Shulmeister, J. & Rother, H. (2022). A model of ice-marginal sediment-landform development at Lake Tekapo, Southern Alps, New Zealand. Geografiska Annaler: Series A, Physical Geography, 104(3), 151–182.

Taylor, C.W. (1956). Erratics of the Saunton and Fremington areas. Report and Transactions of the Devonshire Association, 88, 52–64.

Ussher, W.A.E. (1878). The chronological value of the Pleistocene deposits of Devon. Quarterly Journal of the Geological Society of London, 34, 449–458. https://doi.org/10.1144/GSL.JGS.1878.034.01-04.33

Vachell, H.A. (1963). An erratic boulder from the Fremington Clay. Geological Magazine, 100, 94.

Wingfield, R.T.R. (1995). A model of sea-level in the Irish and Celtic Seas during the end-Pleistocene to Holocene transition. Geological Society, London, Special Publications, 96, 209–242.

Wood, T.R. (1974). Quaternary deposits around Fremington. In: Straw, A. (ed.), QRA Easter field meeting handbook. Quaternary Research Association, pp. 30–34.

Zeuner, F.E. (1959). The Pleistocene Period. Hutchinson, London.

 

Appendix 1 — Original Erratic Descriptions

Selected excerpts from Dewey (1910) and Taylor (1956), reproduced here for reference as the originals are not readily accessible online.

From Dewey, H. (1910)

Plate XXIX from Dewey (1910), showing thin-section microphotographs of the Fremington igneous erratics.

Boulder No. IV (Combrew Farm garden): “The rock is dark grey-green, with large pale olive-green felspars which are glassy and easily chipped from the rock. Microscopic examination reveals its glassy and porphyritic nature. It possesses felspars of two generations, but both are acid labradorite. The larger ones form a quarter of the rock, and the smaller occur in about equal quantities with the glassy base. The ferromagnesian constituent is a rhombic pyroxene which occurs to the entire exclusion of all other ferromagnesian minerals, for there is no augite, hornblende, or olivine. It occurs as small prisms, nearly the same size as the small felspars of the ground mass, with good cleavages, strong pleochroism, and straight extinction.”

“Magnetite is abundant, and occurs as rods and feathery masses, and also as fine thin lines in the glassy base forming a network or gridiron structure. The crystalline constituents are embedded in a brown glass which constitutes about half the rock.”

“The rock may be described as a hypersthene andesite. In many respects it resembles the Tholeite of Watt Carrick, Dumfries, and the hypersthene rocks of Curachan, Loch Craignish, Argyll, but all of these rocks contain considerable quantities of augite, whereas this rock is free from augite.”

From Taylor, C.W. (1956)

Page from Taylor (1956) showing locations and descriptions of the Fremington erratics.

Taylor (1956): photograph of erratics at Combrew Farm.

No. 7: “The next erratic of this group is the hyalopilitic andesite, also previously described with No. 6 above. It is now situated on the right of the gated portion of the driveway to Combrew Farm, and is a glassy, brittle andesite, quite different from any of the foregoing rocks. Well rounded and about sixteen inches across, it contains no augite, but otherwise resembles similar rocks of Dumfries and Loch Craignish, Argyllshire.”

No. 8: “A section of this specimen under the microscope shews it to be a much altered quartz porphyry, with crushed and irregular crystals of plagioclase, porphyritic quartz and long prisms of apatite, with a matrix nearly amorphous and red; epidote appears to have replaced part of the mosaic. This is regarded as a highly peculiar textural type, which may be derived from a fairly local source, such as the porphyritic dyke, west of the coasts of Devon and Cornwall.”

 

Appendix 2 — Rediscovery of the Fremington Clay Erratics, November 2025

The farm was visited on 2 November 2025 and the principal erratics photographed and measured. All key stones described by Taylor (1956) remain in situ or in their recorded relocated positions.

No. 7: Hyalopilitic andesite (Combrew Farm driveway wall)

This is the most important erratic in the Fremington inventory for the question of distant versus local provenance. It remains on the right of the gated driveway to Combrew Farm, exactly as Taylor described it in 1956.

No. 7 in its position on the Combrew Farm driveway wall, November 2025.

No. 7 viewed from the driveway approach.

No. 7 close-up showing the glassy, brittle texture and well-rounded form.

No. 7: detail of surface. The dark grey-green colour and glassy groundmass are clearly visible.

No. 6: Spilite / vesicular granophyre (Combrew Farm entrance)

A boulder by the farm entrance matches the quoted size (40 × 30 × 25 inches) and description (“a dark grey, finely crystalline rock”). If it is the same erratic, it has been rotated since Taylor’s photograph.

Possible No. 6 at the Combrew Farm entrance, November 2025.

Detail of the boulder beside the stone wall.

Further view of the boulder.

Close-up showing dark grey crystalline texture.

Other roadside stones

Two smaller roadside erratics are conglomerates, one resembling a sarsen-type puddingstone. They do not match any of the erratic descriptions in the literature.

Conglomerate boulder near the farm entrance.

Second conglomerate boulder; possible puddingstone type.