Southerndown’s Dunraven Bay brings together a sandy beach, striking Early Jurassic limestone cliffs and fossil-rich shore platforms. Look closely at the Blue Lias and you can see curved oysters, other bivalves, ammonites and traces of life on an ancient seabed. The changing rocks also reveal how a Jurassic sea spread across an older limestone landscape.
FIND FREQUENCY: ♦♦♦♦ – Bivalves, especially Gryphaea, are conspicuous in some beds. Ammonites and other fossils occur, but finding an identifiable loose specimen is never guaranteed.
CHILDREN: ♦♦♦♦♦ – The main bay offers a rewarding family visit with close supervision. Keep children away from cliffs, deep pools and the water’s edge; rocky platforms become slippery and disappear beneath the tide.
ACCESS: ♦♦♦♦♦ – The car park above Dunraven Bay gives a short approach to the shore. Parking, toilets and visitor facilities are listed by the council; check signs for current charges and opening arrangements.
TYPE: – Coastal cliffs and foreshore within Southerndown Coast SSSI. Plan a photography-first visit: do not hammer cliffs, excavate platforms or damage fossil-bearing beds. Public access does not establish permission to remove fossils.
DIRECTIONS
♦ Southerndown lies on the coast south of Bridgend. Approach on the B4524, then follow the road through the village towards Dunraven Bay.
♦ Use the signed car park above the bay and walk down the normal public approach. Follow current parking notices rather than relying on an old all-day tariff.
♦ This guide covers Dunraven Bay itself. Examine accessible rock on either side only while the tide and conditions allow; do not extend the walk around Witches Point into Temple Bay on the assumption that the route will remain open.
♦ Bay reference: 51.446338, -3.606273. Check tide times and weather before leaving; plan a falling-tide visit with ample time to regain the main beach and normal exit.
FOSSIL HUNTING
Begin by looking at the rock surfaces rather than trying to break them. Dunraven Bay’s limestone platforms and loose shore material show shells in several ways: complete valves, weathered moulds and sections cut through shells by erosion. A photograph with a scale and a note of the exact position can record a fossil while leaving the protected exposure intact. The alternating limestones and muddier beds host different assemblages, so a shell-rich patch is more informative than an assumption that every layer should contain the same fossils.
The curved oyster Gryphaea arcuata is one of the bay’s best-known fossils. Its thick, curved lower valve contrasts with the flatter opposing valve. Some beds near Seamouth contain exceptionally dense oyster concentrations, but these are particular horizons rather than a measure of how many loose fossils a visitor will find. Oysters are bivalves, not brachiopods. Their shape, attachment areas and growth lines make useful subjects for close-up photographs.
Other bivalves broaden the picture. The museum’s Dunraven identification sheet illustrates Plagiostoma and large Pinna. The latter can remain in their former life position, and erosion may expose wedge- or diamond-like shell sections rather than a complete upright shell. Observe these structures in place: removing a shell would destroy information about how the animal lived in the sediment. Do not assume that every large shell section belongs to the same genus.
Ammonites occur in both the marginal deposits and the more regularly bedded offshore succession. The museum sheet illustrates small Schlotheimia and larger arietid ammonites, including Arietites; the centres of large examples may be missing. Ribs, the shape of the coil and the preserved outer whorl matter more than size alone when identifying a specimen. Published records of Vermiceras, Coroniceras, Arnioceras and Euagassiceras are particularly valuable for dating the strata, but a weathered or incomplete beach ammonite should not be assigned to one of these names merely because it was found here.
A coral-rich limestone on the Dunraven foreshore was historically described as containing abundant Montlivaltia haimei, with fewer corals in adjacent limestones. The museum sheet also illustrates a coral as Stylophyllopsis. These are separate source identifications, not a demonstrated synonymy. Small five-rayed or star-shaped Isocrinus stem pieces are another useful find to recognise: they are fragments of crinoids, not complete starfish. Gastropods also occur, but the broader coastal literature’s long species lists should not be treated as a checklist for every bed in this bay.
Carbonised wood, including branch or trunk material, records plant debris carried into the Jurassic sea. Burrows preserve a different kind of evidence: the activity of animals moving through or dwelling in the ancient seabed. Trace their branching or repeated form on the rock surface without cutting into the bed. Calcite veins and replacement textures can resemble biological structures; a photograph of the surrounding rock helps distinguish a fossil from later mineral growth.
Fish and marine-reptile remains are reported from the wider offshore succession and the old locality guide, but they are uncommon and no particular reptile species can be promised from Dunraven Bay. The old guide favoured the southeastern side where muddy layers reach the foreshore; this is a useful historical observation, not a guarantee of vertebrate finds or a reason to approach unstable cliffs. If a potentially important bone is visible in bedrock, record its position and seek advice from the site manager or a museum instead of excavating it.
Significant discoveries, geological research and conservation milestones at Southerndown.
1866 – The Southerndown Beds are named
Tawney introduced the name Southerndown Beds in the early study of the coast’s marginal Jurassic deposits. The term remains useful historically, although the facies boundaries are not uniform time horizons.
1922 – Dunraven’s succession and fossils are documented
A.E. Trueman described the bay’s limestone and shale succession, its oyster-rich and coral-bearing horizons, and ammonites at several relative positions. His work also used local oysters in correlation and evolutionary interpretation; the old idea of a simple straight evolutionary progression is not treated as established fact today.
1960 – Marginal and offshore environments are compared
A. Hallam’s sedimentary and faunal study helped clarify the contrast between the coastal marginal deposits and the more regularly bedded offshore succession, providing an important framework for understanding the Southerndown coast.
1965–1968 – Sedimentology and fossil distribution are investigated
F.J. Wobber examined sedimentology, microfacies and ecological distribution along the coastal transect. The work helped explain local environmental variation, but its wider-coast fossil records cannot all be assigned to the main Dunraven Bay exposures.
1990 – The modern measured packages are published
The revised Bridgend geological memoir documented Dunraven’s separate limestone–mudstone packages, the Seamouth Limestone and their ammonite evidence. It explicitly recognised the limits to detailed correlation through the fault-bounded bay.
1998 – Historic oyster collections are traced
Michael G. Bassett documented the survival and transfer of Swansea research collections, including oysters collected at Southerndown by Trueman and Trevor Marchant. Some original catalogues and handwritten labels survived; the reported collection totals covered several localities rather than Southerndown alone.
1999 – Dunraven oysters contribute to growth research
Work by Jones and Gould on Gryphaea growth and age included Dunraven material, as documented in the later geological conservation review. The locality’s oysters continued to inform research beyond their use as field-guide fossils.
2002 – Dunraven mineralisation gains a revised age constraint
A study by Simms, Little and Rosen recorded driftwood-associated mineralisation high in Dunraven’s Porthkerry succession. This showed that the mineralisation could be no earlier than Sinemurian, while its precise later timing remained uncertain. The paper’s better-known coral reinterpretation concerned neighbouring Slade Trough.
2004 – The coastal geological record is synthesised
The Geological Conservation Review brought together the Pant y Slade–Witches Point succession, distinguishing Dunraven’s younger offshore strata from the marginal-facies sections farther west and setting out their scientific importance.
2009 – A preparation study reveals hidden shell damage
Christian Baars examined partly silicified Gryphaea arcuata from a Southerndown road cutting. Acid treatment removed calcite still present inside apparently silicified shells, demonstrating how preparation could leave an attractive exterior while destroying internal support and information.
GEOLOGY
Dunraven Bay exposes part of the Blue Lias Formation of the Lias Group, including the Porthkerry Member. The succession includes Hettangian and lower Sinemurian rocks. Marginal deposits change age laterally, while Dunraven’s offshore strata include a substantial lower Sinemurian record; describing the whole bay as Hettangian misses much of its geological history. The older literature used ‘Porthkerry Formation’, but the current BGS rank is Member. The change of rank does not create an additional rock unit.
The Jurassic sea advanced across an uneven landscape of much older Carboniferous limestone. This older substrate is visible at Trwyn-y-Witch, or Witches Point, where the contact is an unconformity. Near the former shoreline, limestone and chert fragments mixed with shell debris to form conglomeratic and sandy-textured limestones historically called Sutton Stone and Southerndown Beds. These marginal facies pass laterally and upwards into the more familiar offshore limestone–mudstone alternations. Their boundaries change age from place to place, so they are not simple time lines that can be traced unchanged around the bay.
Within the offshore sequence, thicker and thinner limestone beds alternate with calcareous mudstones. The rock changes reflect differences in the original sediment and seabed conditions, subsequently modified by burrowing, compaction and mineral growth during burial. The limestone-rich Seamouth Limestone forms an important local marker, while overlying packages vary in their proportion of limestone and mudstone. Ammonites establish Bucklandi and Semicostatum zonal evidence, but some individual correlations remain tentative.
The bay occupies the narrow Dunraven graben between the Slade and Dunraven faults. Movement and differential subsidence during sedimentation are among the explanations proposed for local thickness and facies changes. This is why the well-known measured beds at Nash Point cannot simply be copied onto Dunraven. Even a recognisable local limestone package need not match each individual bed of a neighbouring section.
The rocks also record a later mineral history. Barite and galena occur in the coastal succession, and a published observation links mineralisation with driftwood high in Dunraven’s Porthkerry Member. That occurrence can be no earlier than Sinemurian; a possible younger Jurassic origin has been discussed, but the timing is uncertain. It should not be labelled as a precisely dated event at 200 million years. The former claim of Triassic deposits extending 300 m offshore belonged to Ogmore-by-Sea, not to this Dunraven Bay succession.
Detailed geology and stratigraphy at Southerndown / Dunraven Bay.
Scope and order. This account concerns Dunraven Bay, with the older substrate at its southern headland, Trwyn-y-Witch. The Slade Trough to the west and the shore beyond Witches Point are separate exposures. The succession below runs from older substrate towards younger Jurassic rocks; local changes in facies and faulting prevent a single bed-by-bed log from representing every part of the bay.
Carboniferous substrate and the Jurassic unconformity. Older, deformed limestone is exposed in the lower part of Trwyn-y-Witch beneath the Jurassic marginal deposits. The erosional contact represents a substantial break in the record. Limestone and chert fragments incorporated into the younger sediments can therefore be derived from the older landscape, whereas shells living in the Jurassic sea belong to the younger deposit. The numbered Slade Trough Beds 1–9 and its distinctive geometry are not the Dunraven offshore section.
LIAS GROUP
Blue Lias Formation
Sutton Stone and Southerndown marginal facies. Sutton Stone is typically massive or rubbly white bioclastic limestone, with corals, chert and fragments of Carboniferous limestone. The Southerndown facies is generally blue-grey, thinner and more regularly bedded, including chip conglomerates and calcarenites with shaly partings. These historical facies interdigitate and pass laterally and upwards into offshore limestone and mudstone; their gradational boundaries are diachronous, not two uniform time planes. At Dancing Stones [SS 8828 7322], Schlotheimia supports the Hettangian Angulata Zone. The BGS account places the highest marginal Southerndown rocks more broadly in the Bucklandi Zone, and the conservation review discusses Bucklandi and possible Semicostatum continuation inland. These differing positions must not be compressed into a Hettangian-only age for every marginal exposure. South of Trwyn-y-Witch, the youngest marginal facies reaches the Portlocki Subzone and offshore facies the Laqueus Subzone; that separate side of the headland is not the main bay log.
Seamouth oolite and local boulder horizon. At Seamouth [SS 883 733], oolite in the upper marginal facies has an uneven upper surface. Shale above contains boulders of the same oolite encrusted by oysters. About 400 m to the west this boulder bed disappears and the transition into the offshore facies changes. It is a local sedimentary relationship, not a persistent marker assumed to occur everywhere around Dunraven Bay. About 400 m west of Seamouth, the transition also shows an increase in mean chert-clast size, a sharp upward decrease in lithoclast abundance and increasing pyrite. These changes help distinguish the local transition without imposing the Seamouth boulder bed on the whole shore.
Porthkerry Member
Terminology and measurement. The BGS memoir called this the Porthkerry Formation; the current Lexicon assigns it member rank within the Blue Lias Formation. The following packages are the memoir’s Dunraven succession, not the whole member’s regional thickness or the height of a single cliff. Its top is not exposed here. Regional minimum coastal and borehole totals cannot be substituted for the local measured intervals.
Lower package at Dancing Stones — 19 m. Planar-bedded wackestones alternate with calcareous mudstones in an approximate limestone:mudstone proportion of 60:40. The lowermost beds yield Vermiceras cf. solaroides, V. conybeari and V. caesar, supporting the Conybeari Subzone. In the upper 3 m, Coroniceras cf. longidomus suggests a late Conybeari or Rotiforme Subzone age, with that alternative retained. Faunal evidence suggests an equivalent of the memoir’s regional Unit B, but the amalgamated marker beds recognised in Unit B elsewhere are not identified here.
Seamouth Limestone — 4 m. Fine- to medium-grained wackestones and packstones are divided by thin mudstone partings. Coroniceras cf. rotiforme at the top supports the Rotiforme Subzone. Its lithology, fauna and thickness suggest correlation with regional Unit C, but this does not establish identity between every local bed and the Nash Point log.
First overlying package — 8.5 m. Nodular or wavy-bedded limestones alternate with mudstones in an approximate 45:55 ratio. Low beds contain Coroniceras cf. rotiforme. Arnioceras and Euagassiceras recorded 6 m above the Seamouth Limestone tentatively indicate the middle or upper Semicostatum Zone. This places the zonal transition low in the overlying succession, but does not identify a precisely fixed boundary bed. The lower succession above Seamouth probably corresponds to regional Unit D at Nash Point, but bed-for-bed correlation through Dunraven is not possible; the individual Nash marker beds are not imported.
Next package — 3.7 m. Planar-bedded limestones and mudstones return to an approximate 60:40 limestone:mudstone proportion. It is a separately described local package, not an additional named formation or member.
Next package — 15 m. Mudstones contain nodular limestones, with limestone and mudstone represented in an approximate 50:50 proportion. The nodular character contrasts with the more planar bedding of the packages below and above.
Highest recorded package — 10 m. Planar-bedded limestone and mudstone have an approximate 75:25 proportion, making this a distinctly limestone-rich part of the recorded sequence. No upper formation contact is exposed. The absence of that contact must not be replaced by a guessed complete thickness or an invented overlying unit.
Coral and oyster-rich horizons. A Dunraven foreshore limestone contains abundant corals recorded historically as Montlivaltia haimei, with smaller numbers in adjacent limestones. No secure modern bed number or thickness is assigned to that coral horizon here. Certain Seamouth mudstones contain very dense Gryphaea arcuata populations: published estimates of approximately 60% shell volume and 60–95% life-position shells describe selected beds, not the whole bay. Pinna in life position provides further evidence of seabed communities preserved where they lived.
Eastern coastal offshore facies. The conservation review describes individual limestone beds in the eastern part of the site ranging from 0.04 to 0.76 m thick, forming nodular or tabular bands within the Porthkerry succession. This is an individual-bed range, not the thickness of the six measured Dunraven packages. Fossils in muddy sediment are commonly compacted, and some beds contain fibrous calcite known as “beef”. Rare organic-rich laminated shales contain pyritised ammonites, fish debris and impoverished bivalve assemblages. These are local facies observations without an invented assignment to a particular numbered bed or measured package.
Structure, deposition and later change. The Slade and Dunraven faults bound the narrow Dunraven graben. Synsedimentary movement and differential subsidence have been proposed to explain local changes in facies and thickness and the difficulty of bed-for-bed correlation. Original depositional differences were modified by bioturbation and diagenesis; the alternation of limestone and mudstone should not be reduced to one settled universal mechanism. Thalassinoides-type burrowing is recorded through the Porthkerry succession, without requiring an unverified ichnospecies assignment to every trace on the foreshore.
Driftwood-associated mineralisation. Barite and galena were observed with driftwood high in Dunraven’s Porthkerry Member. Their formation can be no earlier than the Sinemurian host strata; a possible middle Jurassic age has been discussed, but remains uncertain. The coral illustrations in the same study mainly concern neighbouring Slade Trough, with comparative specimen BMNH R37234 explicitly from Brocastle. Neither the illustrated Brocastle specimen nor the Slade colonies should be relabelled as Dunraven Bay vouchers.
Trueman’s 1922 historical section: a separate framework. The early account described a prominent limestone about 8 ft thick, approximately 60 ft above the Southerndown Beds, made of ten or more nodular layers with thin contorted shale partings. Above it lay roughly 8 ft of limestone and thick shale, then about 70 ft of shale with sparse nodular limestone, followed by more calcareous beds. These approximate historical units and positions cannot simply be equated with the later 4 m Seamouth Limestone or substituted for the modern measured packages.
Historical lower reefs and ledges. A reef under Dunraven, 10 ft below the prominent band, yielded large slender arietids, small silicified ?Arnioceras cf. spirale, Coroniceras cf. rotiforme and C. cf. rotator. Ledges ENE of Dunraven Castle another 10 ft lower contained crushed Vermiceras cf. conybeari and V. cf. caesar among large arietids; the author stressed that accurate identification was usually impossible. Similar lower beds, including an exposure 30 ft lower west of Sea Mouth, yielded Gryphaea aff. obliquata and occasional Lima gigantea. A separate reef south of Sea Mouth, NE of the castle, contained abundant Montlivaltia haimei. These are historical identifications with their original uncertainty, not a newly revised modern species list.
Historical prominent band and overlying shales. In or immediately above the prominent band, Trueman recorded stout tuberculate arietids and a possible Arnioceras cf. spirale. Selected shales immediately above contained silicified Gryphaea aff. incurva, Pleuromya sp. and Pleurotomaria cf. anglica; the report of shells making up about half the rock refers to some beds only. Higher shales also yielded Gryphaea aff. incurva, with no ammonites recorded in that account. A single well-preserved Arnioceras bodleyi was reported from the highest limestone. Higher faulted beds were inaccessible and part of the interpretation relied on fallen material, so this is not a precise modern in-situ range chart.
Historical marginal-facies measurements and locality limits. Trueman separately estimated 45–50 ft of marginal Southerndown Beds in the cliff subsection, with Zygopleura sp., ammonite impressions and Gryphaea; lateral variation made older sections difficult to match. On the north side of Witches Point he estimated 25–40 ft of Sutton Stone and about 14 ft of conglomeratic Southerndown-type limestone. These are historical measurements at specified exposures, not a single uniform thickness beneath all Dunraven Bay. His longer mollusc list from the south side of the headland belongs beyond this guide’s main-bay scope. About 10 ft below the top of the Southerndown Beds, oyster shapes led Trueman tentatively to suggest the lowest part of his historical Vermiceras subzone, although no ammonites had been found there. Lower beds contained forms resembling Gryphaea dumortieri. This is an explicitly historical oyster-based correlation and resemblance, not a revised identification or a connection to the separately catalogued, broadly located BGS specimen.
References. Wilson, D., Davies, J.R., Fletcher, C.J.N. and Smith, M. (1990), Geology of the South Wales Coalfield, Part VI: the country around Bridgend, Chapter 6; BGS Lexicon, Porthkerry Member; Simms, M.J., Chidlaw, N., Morton, N. and Page, K.N. (2004), British Lower Jurassic Stratigraphy, Pant y Slade–Witches Point account, pp. 119–128; Trueman, A.E. (1922), The Liassic rocks of Glamorgan, especially pp. 262–264; Simms, M.J., Little, C.T.S. and Rosen, B.R. (2002), Lower Jurassic coral study; Amgueddfa Cymru, Southerndown fossil identification sheet.
SAFETY
Keep well clear of the cliffs and any fresh falls. Rock can fall at any time, including in fine weather, and a hard hat does not make the cliff foot safe. Do not shelter beneath overhangs or climb loose blocks.
Tides can cover the beach and cut off the ends of the bay. Witches Point has strong tidal currents, with rescues documented offshore; tides can also cut off shore routes. Check the forecast and tide times, start on a falling tide and return before the route narrows. There is no universal safe number of hours around low water; conditions, swell and the intended route all matter.
Limestone platforms, seaweed and wet shingle can be slippery. Wear footwear with good grip, supervise children closely and avoid wave-washed edges. Carry a charged phone and tell someone your plan. In a coastal emergency call 999 or 112 and ask for the Coastguard.
EQUIPMENT
Take a camera or phone, a small scale, a hand lens, a notebook and sturdy footwear. Tide information, suitable clothing and drinking water are more useful than heavy collecting tools for this protected shore. Do not bring a hammer for working the cliffs or solid-rock fossil layers.
If the landowner or site manager has confirmed that your intended collection of loose material is permitted, use small protective bags and labels so that a specimen’s exact locality and circumstances are not lost. Keep photographs of it before moving it.
CLEANING AND TREATING
For any legitimately collected loose specimen, begin with gentle dry brushing and examination under magnification. Retain supporting matrix around delicate shells and avoid forcing a split through a fossil-bearing limestone. Fragile, mineralised or potentially important material is best assessed by an experienced preparator before treatment.
A preparation study used Gryphaea arcuata from a Southerndown road cutting, rather than an unrestricted sample from this protected beach. Shells that looked silicified on the outside retained calcite internally; acid treatment dissolved that calcite and left hollow, fragile shells. This is a direct local reason to avoid acid cleaning and not to assume that an apparently resistant exterior protects the whole fossil. Beekite rings are replacement textures rather than original shell ornament.
Record the specimen and its matrix before cleaning. Do not apply a general acid recipe, varnish or consolidant simply because another Blue Lias fossil was treated that way; mineral composition and preservation vary.
IDENTIFY YOUR FINDS
Need help identifying a fossil? Share clear photos, where you found it and its size with the community.
FURTHER READING
Amgueddfa Cymru: Southerndown fossil identification sheet
BGS: Geology of the South Wales Coalfield, Part VI, the country around Bridgend
Geological Conservation Review: Pant y Slade to Witches Point
Geological Curator: Baars’s study of partly silicified Southerndown oysters
Simms, Little and Rosen: Lower Jurassic corals and mineralisation
Vale of Glamorgan Council: Dunraven Bay visitor information
ACCESS RIGHTS
Dunraven Bay lies within Southerndown Coast Site of Special Scientific Interest. The site-management statement discourages collection of in-situ fossils and minerals, and the museum’s local field sheet asks visitors not to hammer cliffs or solid-rock fossil layers. Leave bedrock specimens, mineral veins and the shore platforms undamaged; photography is the appropriate starting point.
The public visitor approach and parking arrangements do not themselves grant permission to remove loose fossils. Check with the landowner or site manager if you intend to collect, and follow current signs and any additional restrictions. Do not assume council or National Trust ownership, or extend permission for one part of the shore to adjoining land.
It is important to follow our ‘Code of Conduct’ when collecting fossils or visiting any site. Please also read our ‘Terms and Conditions‘
Updated 2017, Edited by Jon Trevelyan 2014
LINKS
♦ Fossil Discussions
♦ Fossil Articles
♦ Buy Fossils, Tools and Equipment
♦ Buy Crystals, Meteorites, and Artefacts
♦ Join Fossil Hunts
♦ UK Fossils Network
