Westbury on Severn Fossil Hunting

1822 – First published measured outline
William Daniel Conybeare and William Phillips described a measured succession through the coloured marls and overlying pale beds at Garden Cliff.

1842 – Insects and vertebrate remains reported
Brodie reported insect remains from thin limestone near the top of the section. Richard Owen also referred to reptile remains from Westbury, providing an early record of the locality’s vertebrate fossils.

1860 – Detailed sections and Triassic age discussed
Accounts by Sir William Vernon Guise and Thomas Wright drew on sections examined with J. Jones and W.C. Lucy. Wright used the fossils in assessing the Triassic age of the beds between the red marls and the Lias.

1865 – Etheridge’s bed-by-bed section published
Robert Etheridge described the Garden Cliff succession in detail, including the Pullastra sandstone, bone-bearing layers, Pecten beds and Estheria horizon. His bed numbers remain important for interpreting later studies.

1901–1903 – Richardson’s measured section developed
Linsdall Richardson published a detailed account of Garden Cliff and used its numbered beds in regional Rhaetic correlation. Published citations give differing dates, 1901 and 1903, for the volume’s paper.

1904 – A single synchronous bone bed challenged
Reynolds and Vaughan questioned the proposed correlation of Garden Cliff’s main bone bed with the basal bed at Aust. Wickes likewise emphasised that bone-bearing lenses occur at different levels, an issue revisited by modern research.

1911 – Westbury Beds terminology consolidated
Richardson proposed Westbury Beds as an alternative to the many overlapping names then used for the dark lower Rhaetic succession, using Garden Cliff in comparisons with other sections.

1935 – Pachystropheus named from locality material
Friedrich von Huene named Pachystropheus rhaeticus using material from Garden Cliff. The name became central to later studies of the locality’s small marine reptile.

1974 – Nick Large collection donated
The Nick Large collection entered Bristol Museum. Garden Cliff specimens from this collection were subsequently included in the 2022 reassessment.

1977 – Bone-bed sedimentology analysed
Sykes described the lower Garden Cliff succession and separately analysed finer and coarser parts of the Upper Pullastra sandstone, distinguishing different concentrations and grain sizes of phosphatic remains.

1980 – Westbury Formation adopted
The national Triassic correlation scheme formalised Westbury Formation within the Penarth Group. The BGS Lexicon identifies Westbury-on-Severn as the type area.

1992 – Material collected for renewed vertebrate study
Glenn Storrs collected the samples now registered as BRSUG 29953 in December. They later formed part of the quantitative microvertebrate study.

1993 – Reptile relationships and trace fossils investigated
Storrs and Gower proposed a choristodere relationship for Pachystropheus, while Wang studied the locality’s horseshoe-crab traces and sedimentary setting. The reptile interpretation was revised in later work.

1996 – Pachystropheus anatomy studied in detail
Well-preserved Garden Cliff remains contributed to the detailed investigation of Pachystropheus by Storrs and colleagues, highlighting the scientific value of the locality’s lightly abraded bones.

1997 – Geochemistry tested bone-bed origins
Clive Trueman and Michael Benton compared rare-earth-element signatures in bones and sediment at Garden Cliff and Aust. The Garden Cliff material showed closer correspondence with its matrix, consistent with less reworking.

2009 – Type-section status proposed
R.W. Gallois proposed Garden Cliff as the type section for the Westbury Formation in a reassessment of the Penarth Group. Garden Cliff’s exposed section helped focus discussion of the formation’s definition.

2010 – New trace fossil named
Allington-Jones, Braddy and Trueman named Radichnus allingtona from Garden Cliff, with holotype BRSMG Cd5331. Their ichnological and geochemical analysis supported a shallow, restricted environment periodically disturbed and oxygenated by storms.

2019 – Red mudstones interpreted in greater detail
Milroy, Wright and Simms analysed the Westbury succession as part of a regional study of saline mudflats and temporary lakes. The work recognised more varied depositional conditions within the red-bed succession.

2021 – New samples and measured section obtained
The team behind the subsequent 2022 paper collected BRSUG 29954 in June and recorded a new lower Westbury Formation section, including several bone-bearing levels.

2022 – Microvertebrate fauna and bone-bed correlation reassessed
Harriet Williams and colleagues published the first comprehensive study of Garden Cliff’s microvertebrate assemblage, distinguishing sample provenance and questioning both a supposed coarse basal bone bed and simple time-equivalence with Aust.

2024 – Pachystropheus reclassified
Jacob Quinn and colleagues reassessed Pachystropheus rhaeticus, including its original name-bearing material, and identified it as a thalattosaur. This revised the earlier proposed choristodere relationship of the Garden Cliff reptile.

The main bed-by-bed account runs from oldest to youngest and follows the GCR’s modified Richardson–Sykes column. Its historical measurements are retained as one source scheme. Etheridge’s earlier numbering, Sykes’ lower section and the fieldwork published in 2022 follow separately because their boundaries, thicknesses and bone-bed interpretations cannot safely be merged.

MERCIA MUDSTONE GROUP

Branscombe Mudstone Formation

Twyning Mudstone Formation (22.25 m). Red marl with grey/blue-grey zones, mainly angular fracture and thin gypsum veins. Twyning Mudstone Formation is the older name used in the GCR table; the current BGS name is Branscombe Mudstone Formation. The 22.25 m value belongs to that published column. The 2022 paper separately describes about 15 m visible at the north-western end and 20–25 m in its general red-marl account; these are different scopes of observation. Rare conchoidal and cuboidal fractures are also described in the GCR red-marl row. The GCR also records local lamination, ripple markings and vertical gypsum veins; the beds dip approximately 2° south-east.

Blue Anchor Formation

Blue Anchor Formation (5.49 m). Grey-green marl, weathering grey-blue, yellow-grey or white; conchoidal fracture.

PENARTH GROUP

Westbury Formation

Basal bone bed (0.05 m). Discontinuous sand of mixed grain size with derived Blue Anchor mudstone clasts. The published row records bones, coprolites. This is the basal bed reported in the older synthesis. Its position was questioned in the 2022 re-investigation, which suggested that loose material might have been assigned to the base incorrectly. It is not an established part of the newer field log.

Basal shale (c) (0.15 m). Unsandy black fissile shale with silt patches and Blue Anchor pellets/flakes in lowest 0.025 m.

Basal shale (b) (0.15 m). Black fissile shale with fine sand patches.

Basal shale (a) (0.1 m). Black shale containing sandy patches. The published row records bivalves, bone-bed particles.

Lower Pullastra Sandstone (0.13 m). Pale grey calcareous micaceous slightly pyritic siltstone, splitting in two laminae. The published row records vertebrate debris in lower lamina. Sykes’ 1977 lower-section account gives 0.4 m for the named lower siltstone rather than the GCR’s 0.13 m. The two source values are retained separately.

Shale above Lower Pullastra bed (0.61 m). Fissile calcareous black shale with some siltstone layers. The published row records bivalves. The GCR gives 0.61 m; Sykes’ abbreviated lower section gives 0.6 m.

Upper Pullastra Bed (0.3 m). Sandstone: lower medium-grained part richer in larger fossils; finer upper part has tiny phosphatic debris. The published row records vertebrate/phosphatic fragments. The finer upper and coarser lower parts were analysed separately by Sykes. This older named interval is not automatically equated with every later use of Pullastra Bed.

Shale above Upper Pullastra (0.45 m). Fissile black shale with thin pale calcareous siltstone and limestone.

Higher bone bed (0.025 m). One to four pyritic bedded laminae with crystalline upper surface and medium-grained sandy layers. The published row records vertebrate fossils. The GCR narrative places this level about 2 m above the formation base and about 0.4 m above the Upper Pullastra Bed, with an approximate thickness of 0.03 m. Its table gives 0.025 m. It is distinct from the disputed basal bed. The GCR narrative describes a matrix with approximately 70% disseminated pyrite, unabraded pyrite crystals on the basal surface, sporadic quartz grains and nodules among the vertebrate remains and coprolites. Ripples occur on both surfaces, with muddy infills of small channels above and some weak alignment of bones below.

Fissile shale above higher bone bed (0.5 m). Fissile black shale with pale silt layers.

Sandstone with shale seam (0.05 m). Sandstone containing black shale midway.

Imperfectly laminated shale (0.1 m). Black shale with weak lamination.

Thinly laminated shale (0.55 m). Black shale with fine lamination.

Laminated shale with sandstone layers (0.66 m). Black laminated shale with grey micaceous calcareous sandstone near base and 0.30 m below top.

Hard limestone (0.02 m). Slightly pyritic black-blue hard limestone.

Sandstone–shale interbeds (1.25 m). Calcareous and pyritic sandstone interbedded with laminated black shale.

Pecten-bed interval (b) (0.01 m). Limestone, locally sandy. The GCR groups this interval with Etheridge’s bed 12, including his Lower and Upper Pecten Beds; neither name is assigned to this thin subrow alone.

Pecten-bed interval (a) (0.86 m). Poorly laminated selenitic black shale; calcareous pyritic sandstone 0.12 m below top. This is the companion subrow of the Pecten-bed interval, retaining the GCR grouping rather than a new one-to-one correlation.

Lilstock Formation

Cotham Member

Lower Lilstock shale (1.68 m). Grey marly shale breaking conchoidally.

Euestheria-bed / Etheridge bed 14 Estheria Zone (0.3 m). Limestone. The two historical horizon names are preserved; their use here does not assert that every old fossil identification has been revised.

IIe (0.18 m). Grey marly shale with weak lamination.

IId (0.01 m). Rippled calcareous pyritic sandstone.

IIc (0.28 m). Laminated grey shale.

IIb (0.01 m). Pyrite-rich non-calcareous sandstone.

IIa (1.82 m). Laminated grey marly shale.

Ib (0.06 m). Blue-grey to brown limestone with conchoidal fracture.

Ia: Insect Limestone (0.06 m). Blue-grey to light-brown limestone; upper 0.02 m fissile. The Insect Limestone belongs to the interval from which rare fragmentary insect material was historically reported. The broader Cotham record also includes gastropods.

Upper calcareous shale (0.56 m). Finely laminated brown and grey calcareous shale. The GCR table labels these upper beds Lilstock Formation, while its locality narrative discusses the Cotham Member. That is the basis of the member framework here; no separate Langport thickness is supplied.

LIAS GROUP

Lias Group limestone (0.05 m). Hard grey to black-blue limestone. Only this basal Lias limestone is measured in the GCR column. It is not evidence for a full regional Blue Lias succession at Garden Cliff.

Wright’s 1860 section: youngest to oldest. The following sixteen beds retain his descending order, imperial measurements and historical fossil identifications. They are separate from Etheridge, Richardson, the GCR and the later field section.

Wright Bed 1 (4 in). Ostrea bed: hard dark-grey argillaceous Lias limestone, many surface shells. Historical fossils: Ostrea liassica; Modiola minima; Cardium n.sp. Wright placed this historically among the lowest Ammonites planorbis-series beds.

Wright Bed 2 (1–2 ft). Greyish clay.

Wright Bed 3 (4–8 in). Monotis bed: cream-coloured argillaceous fissile limestone. Historical fossils: Upper lamina: Monotis decussata abundant. Lower: Myacites musculoides (Schl. ?), Cardium Rhaeticum, Modiola minima, Monotis decussata, Ostrea liassica. The upper and lower faunal laminae have no separately supplied thicknesses.

Wright Bed 4 (1–2 ft). Greyish shaly clay.

Wright Bed 5 (1 ft). Estheria bed: light-grey nodular limestone, locally shelly, prominent band. Historical fossils: Estheria minuta in nests; Pecten Valoniensis in shelly portions.

Wright Bed 6 (8–10 ft). Dark friable shale with many fossiliferous seams. Historical fossils: Small compressed Conchifera, undetermined.

Wright Bed 7 (1 ft). Dark shaly clay with many compressed shells. Historical fossils: Pullastra.

Wright Bed 8 (4 ft). Dark shale with many seams of compressed shells. Historical fossils: Compressed shells.

Wright Bed 9 (2 in). Pecten bed: dark argillaceous limestone. Historical fossils: Pecten Valoniensis numerous and compressed.

Wright Bed 10 (6 ft). Black shale. Historical fossils: Rare fossils; bodies resembling coprolites.

Wright Bed 11 (1 in). Bone Bed: greyish calcareo-siliceous band with bone debris and much pyrite, true bone breccia. Historical fossils: Saurian/fish bones, reptile teeth, Saurichthys, Acrodus, Hybodus, Ceratodus, many coprolites. These are the broad historical identifications, rather than redetermined modern specimens.

Wright Bed 12 (2 ft). Black shale.

Wright Bed 13 (9 in–1 ft). Dark-grey micaceous sandstone, ripple-marked upper surface, prominent band and large shore slabs. Historical fossils: Avicula contorta, Cardium Rhaeticum, Pullastra arenicola, Modiola minuta. The source prints Modiola minuta here, differing from its use of minima elsewhere.

Wright Bed 14 (2 ft). Black shale.

Wright Bed 15 (4 in). Grit resembling Bed 13, much pyrite, fish scales/teeth. Historical fossils: Fish bones, scales, teeth; Pullastra arenicola.

Wright Bed 16 (2 ft). Hard black shale. Historical fossils: Bodies resembling coprolites. Grey Keuper marl lay below this numbered succession.

Wright separately described more than 80 ft of exposed Keuper marls and a 2–4° south-east dip. These observations are outside the sixteen numbered beds.

The Jones and Lucy section published by Guise in 1860: oldest to youngest. This independent sixteen-bed scheme ascends through the succession. Its numbering, original wording and conflicting measurements remain separate from Wright’s descending column.

Jones/Lucy–Guise Bed 1 (2 ft). Hard black shale. Historical fossils: Coprolites resembling those in higher bone bed.

Jones/Lucy–Guise Bed 2 (5 in in the later correction; original feet mark disputed). Grit and iron pyrite; fossiliferous weathered surface. Historical fossils: Teeth, coprolites, Pullastra arenicola. The original table repeats a feet mark beside 5, but Williams and colleagues explicitly correct this to 5 inches. The conflicting printed value is not adopted as a reliable five-foot bed.

Jones/Lucy–Guise Bed 3 (2 ft). Black shale like Bed 1.

Jones/Lucy–Guise Bed 4 (8 in). Band like Bed 2. Historical fossils: Pullastra arenicola.

Jones/Lucy–Guise Bed 5 (2 ft). Black shale as Beds 1 and 2.

Jones/Lucy–Guise Bed 6 (1 in). Bone Bed chiefly pyrite. Historical fossils: Teeth, coprolites, palates, bones.

Jones/Lucy–Guise Bed 7 (6 ft). Black shale as before. Historical fossils: Coprolites and other unspecified remains.

Jones/Lucy–Guise Bed 8 (2 in). Pecten bed.

Jones/Lucy–Guise Bed 9 (4 ft). Black shale.

Jones/Lucy–Guise Bed 10 (1 in as printed). Pullastra bed. Historical fossils: Pullastra. The printed one-inch value belongs to this scheme and differs from Wright’s account.

Jones/Lucy–Guise Bed 11 (9 ft). Shale.

Jones/Lucy–Guise Bed 12 (1 ft). Grey nodular limestone, uncertain fracture. Historical fossils: Estheria locally abundant.

Jones/Lucy–Guise Bed 13 (1 ft in table). Clayey shale. The table gives one foot, whereas the prose describes about four feet of clayey shale above the Estheria Bed.

Jones/Lucy–Guise Bed 14 (Unmeasured). Limestone band: lower and upper fossil-rich surfaces. Historical fossils: Lower Myacites, probably M. liassinus; upper Monotis decussata abundant. Brodie also reported Cypris, plants, Cyclas and fish scales. The prose adds occasional Modiola minima and repeats the tentative Myacites liassinus identification on the underside. This is table Bed 14 but the limestone is called 13 in the preceding prose.

Jones/Lucy–Guise Bed 15 (2 ft). Greyish-blue clay.

Jones/Lucy–Guise Bed 16 (Unmeasured). Bottom bed (uppermost item in this ascending list). Historical fossils: Printed Modiola and Ostrea minima. The original fossil wording is ambiguous. The later planorbis, Johnstoni and angulatus succession discussed nearby in the paper was not traced at this exact exposure.

The Jones/Lucy–Guise account separately describes 70–80 ft of green and red New Red Sandstone marls underneath the numbered succession, dipping approximately 2° south-south-east.

Etheridge’s 1865 numbering, as reproduced in 2022

This historical graphical scheme is described from its reproduced column, from oldest to youngest. Its Rhaetic boundaries and old fossil names are not current lithostratigraphic or taxonomic assignments. Some fine labels are combined or ambiguous in the reproduction, so no extra thickness or unnamed bed has been invented.

Bed 1 (not given). Six alternating red, grey and white marl beds.

Bed 2 (3 ft 3 in). Marl above red-line marker.

Bed 3 (14 ft). White/grey marl interval historically called Sully Beds.

Bed 4 (1 ft 3 in). Dark/rusty shale at lower fossil-bearing junction. Historical labels record coprolites, teeth, scales.

Bed 5 (5 in). Lower fossil-bearing sandy bed. Historical labels record Gyrolepis Alberti, Sargodon tomicus, Pullastra arenicola, Saurichthys apicalis, Avicula contorta.

Bed 6 (1 ft 10 in in Fig.3 a; 2 ft in Williams narrative). Dark shale.

Bed 7 (1 ft). Pullastra Bed. Historical labels record Cardium Rhæticum, Modiola minima, Avicula contorta, Pullastra arenicola.

Bed 8 (1 ft 6 in). Shale below main bone bed.

Bed 9 (1 in + 6 in + 1 in shown as three rows). Bone-bed interval with internal shale. Historical labels record Acrodus minimus, Gyrolepis Alberti, Nemacanthus, Saurichthys apicalis, Pullastra arenicola.

Bed 10 (7 ft). Cardium shales.

Bed 11 (2 ft 6 in). Contorta shales.

Bed 12 (2 ft). Lower and Upper Pecten Beds enclosing shale. Historical labels record Plesiosaurus (Upper Pecten Bed label).

Bed 13 (7 ft). Fossiliferous shale/marl interval. Historical labels record Pecten Valoniensis, Cardium Rhæticum, Axinus cloacinus, Anatina Suessi, Lima præcursor, Myacites musculoides, Pleurophorus angulatus, Modiola minima.

Bed 14 (1 ft plus adjacent 6 in label whose allocation needs original caption). Estheria Zone; subsidiary c marker below. The adjacent 6 in label and subsidiary c marker cannot be allocated confidently from this reproduction.

Bed 15 (3 ft). Shale/marl interval.

Bed 16 (3 ft 6 in). Shale interval. Historical labels record fish remains.

Bed 17–18 (5 in combined label in reproduced figure). Insect limestone interval. Historical labels record Monotis decussata, Modiola minima, Myacites musculoides. The two numbers remain combined, as shown; separate thicknesses are not supplied.

Bed 19 (1 ft 6 in). Shale interval.

Bed 20 (3 in). Upper limestone below Lower Lias. Historical labels record Modiola minima, Ostrea Liassica.

Sykes’ 1977 lower-section scheme

Sykes’ sequence below runs from youngest to oldest. It is retained separately because the Lower Pullastra value and the interpretation of the basal bone bed differ from later accounts.

Shale with white silt (0.5 m).

Pyritic bone layer (0.025 m). Its upper surface is bedded and crystalline, with interlayers of fossiliferous medium sand.

Shale/siltstone/limestone (0.45 m).

Upper Pullastra sandstone (0.3 m).

Bivalve-bearing shale with siltstone (0.6 m).

Lower Pullastra siltstone (0.4 m). This siltstone contains bivalves.

Three-part basal shale (0.4 m).

Discontinuous basal bone bed (0.05 m).

The subdivisions below belong within Sykes’ existing package totals. His 0.4 m Lower Pullastra siltstone remains a separate source value from the GCR’s 0.13 m.

Basal shale, lower subdivision c (0.15 m). Marl pellets and flakes occur in the lowest 25 mm, without sand.

Basal shale, middle subdivision b (0.15 m). Patches of fine sand interrupt the shale.

Basal shale, upper subdivision a (0.10 m). Bivalves and bone debris occur in this upper part. Together, the three subdivisions form the 0.4 m basal shale.

Upper Pullastra sandstone, finer upper portion. The finer sediment contains tiny phosphatic remains. Its Table 8 analysed residue comprised 86.6% quartz, 0.1% phosphatic material and 13.3% pyrite, with small amounts of mica excluded. No separate thickness was specified for this portion.

Upper Pullastra sandstone, coarser lower portion. Medium-grained sand contains larger fossils. Its separate Table 9 residue comprised 84.43% quartz, 2.58% phosphatic material and 12.99% pyrite, with small amounts of mica excluded. These are sample-residue proportions, not whole-bed composition or fossil abundance.

Historical basal bone-bed sample. Table 10 records 23.21% quartz, 20.14% phosphatic material, 1.85% Tea Green Marl and 54.80% mudstone in the analysed residue; minor pyrite and selenite were excluded. The later fieldwork questioned the basal placement of the older reported bone bed, so this analytical sample is not treated as proof of a securely located modern basal layer.

The 2021 field section published in 2022

Underlying Blue Anchor mudstones. Blocky blue-grey mudstone lies below the Westbury Formation. In the plotted section the formation boundary is at the 1 m mark; heights above the base therefore differ from the raw plotted scale.

Basal contact. The team found an abrupt contact with rusty, iron-rich papery shale, without the coarse basal conglomerate described in the older synthesis. The contact lay locally 90–120 cm below the prominent sandstone. Its varying separation suggests local downward cutting; the absence of a visible coarse lag does not establish an absence of erosion.

Thin sandstone, about 0.7 m above the base. A thin sandstone interrupts the basal black shales. It is not given a separate formal bed number.

Approximate graphic intervals below the main sandstone. Figure 3b places the Westbury Formation base at plotted 1 m, with shale up to a thin sandstone around plotted 1.7–1.8 m. More shale extends to the burrowed lower contact of the prominent Pullastra sandstone around plotted 2.2–2.3 m; that sandstone continues to roughly plotted 2.5 m. These are readings of an unnumbered graphic, rather than precise author-supplied bed thicknesses.

The lowest bone-marker datum differs within the source. The figure places its lowest bone symbol near plotted 1.4 m, about 0.4 m above the depicted formation base. The caption instead calls the lowest bone-bearing level 1.4 m above the base, while the prose associates an indistinct bone level with the Pullastra sandstone. These statements do not establish one consistent lowest horizon.

Rippled sandstone, about 1.4 m above the base. The prominent Pullastra sandstone shows wave ripples and bears Chondrites, Thalassinoides and Selenichnites on its lower surface. An indistinct bone-bearing layer occurs above. The loose blocks sampled as BRSUG 29954 were only tentatively linked with this horizon.

Principal pyritic bone bed, about 2 m above the base. The layer is about 0.025 m thick, with small lenses, ripple cross-lamination and mud drapes. The body text describes lenses 10–30 cm wide, whereas the abstract says up to 20 cm. The main text and Fig. 3 caption use about 2 m above the formation base; the Fig. 4d caption gives 2.8 m without resolving its datum. These descriptions are kept distinct.

Further bone-bearing level, about 3.8 m above the base. This position is given in the Fig. 3 caption. The figure establishes the level, but does not provide a full independent lithological description.

Shale and an unlabelled thin band. Above the prominent sandstone, shale is interrupted by the principal bone layer near plotted 3 m, consistent with about 2 m above the depicted formation base. It continues to a separate pale or dashed band around plotted 3.7 m. That band has no secure lithological or fossil name in the caption.

The two upper drawn sandstones. Shale continues to a sandstone at approximately plotted 4.8–5.0 m; burrow marks occur at its base beside a mollusc symbol. Another shale interval reaches about plotted 5.4 m, followed by the uppermost drawn sandstone to approximately 5.7 m. The higher bone marker at plotted 4.8 m agrees with the caption’s 3.8 m above the formation base. Neither drawn sandstone is confidently equated with the separate prose account of an inaccessible sandstone four metres above the principal bone bed. The graphic ends within the Westbury Formation and supplies no modern measured Cotham section.

Higher bivalve-bearing sandstone. Another passage places a thin sandstone 4 m above the main bone bed and records Selenichnites beneath it. It was inaccessible in situ. This statement is not equated automatically with the 3.8 m-above-base bone horizon because the positions and descriptions differ.

Fish, reptile and invertebrate evidence

The processed microvertebrate samples BRSUG 29953, collected by Glenn Storrs in December 1992, and BRSUG 29954, collected by the authors in June 2021, are distinct from the museum slabs and from in-situ trace observations. The 2022 authors used both a total-fragment count and a more restrictive standard counting method. Neither measures numbers of living animals or visitor finding probabilities; loose-block provenance also limits exact bed assignment.

2022 Table 1 counts (total / standard). Lissodus minimus: 224 / 23; Pseudodalatias barnstonensis: 1 / 0; Hybodus cloacinus: 14 / 0; Rhomphaiodon minor: 139 / 47; fin spines: 11 / 11; hemicentra: 4 / 4.

2022 Table 1 counts (total / standard). placoid scale morphotype 1: 47 / 47; placoid scale morphotype 2: 17 / 17; placoid scale morphotype 3: 34 / 34; placoid scale morphotype 4: 10 / 10; placoid scale morphotype 5: 4 / 4; hybodont scale morphotype 1: 3 / 3.

2022 Table 1 counts (total / standard). hybodont scale morphotype 2: 1 / 1; ctenacanthid scale: 3 / 3; Gyrolepis albertii: 63 / 32; Severnichthys Birgeria type: 118 / 34; Severnichthys longidens type: 76 / 34; osteichthyan scale morphotype 1: 23 / 23.

2022 Table 1 counts (total / standard). osteichthyan scale morphotype 2: 19 / 19; osteichthyan scale morphotype 3: 5 / 5; osteichthyan scale morphotype 4: 23 / 23; coprolite: 7 / 7.

Some source totals differ: the systematic text gives 219 rather than 224 for the most numerous small shark teeth, 12 rather than 11 fin spines, four rather than three ctenacanthid scales, and 24 rather than 23 for each of osteichthyan scale morphotypes 1 and 4. The text also mentions seven placoid morphotypes, but the descriptions and table enumerate five. The counts above follow Table 1, with those differences retained rather than filled in. For the “Birgeria” material, the systematic prose gives 107 + 10 = 117 specimens, while Table 1 gives 118.

The authors provisionally retained Severnichthys for tooth forms linked historically with Birgeria and Saurichthys, while acknowledging an alternative classification. The museum material adds a Nemacanthus monilifer spine on BRSMG Ce16900, a Sargodon tomicus tooth on Ce10813 and a Pseudodalatias barnstonensis tooth on Cb4876. Sargodon was absent from the two processed samples. Taxonomic labels on historical teeth should not be changed in bulk.

Fish jaws with teeth still attached. The 2022 account describes two unidentified jaw examples, including the figured BRSUG 29953-21. BRSMG Ce17771 separately contains a jaw with a “Birgeria”-type tooth attached. These records do not justify a more precise identification of the otherwise unassigned jaws.

The processed material includes four tooth morphotypes of Lissodus, five placoid-scale morphotypes, two hybodont-scale morphotypes and four bony-scale morphotypes in the enumerated descriptions. These are specimen categories, not additional species or beds. Rounded crushing teeth, pointed cusps, isolated scales and broken fragments preserve different kinds of anatomical evidence.

Museum limb material assigned to Pachystropheus rhaeticus includes BRSMG Ce17773–Ce17776 and Ce17778–Ce17779, with some individual elements only tentatively identified. Ce17771 is a mixed block with bones, spines and coprolites; Ce17770 includes Pachystropheus bones and a larger worn plesiosaur epipodial. Ichthyosaur teeth and fragments include Cg3025 and material in BRSUG 29953. A possible dinosaur neural spine on Ce16900 was too abraded for a secure determination. No Pachystropheus was recovered from the two processed sample collections.

Historical limb identifications and their uncertainty. The material attributed to Pachystropheus includes BRSMG Ce17774, a possible small femur; Ce17775, an uncertain limb; Ce17776, a distal femur; Ce17778, a possible humerus in the prose; Ce17779, a left humerus shown in posteroventral view; and Ce17773, an unidentified limb fragment. These are separately qualified museum elements, without new exact-bed assignments.

The ichthyosaur account reports six possible microtooth fragments, of which only two preserve whole tips, and illustrates the larger tooth BRSMG Cg3025. The historical heading Ichthyosaurus does not establish a secure genus-level identification for every fragment.

The 2024 reassessment places Pachystropheus rhaeticus among thalattosaurs, replacing its proposed choristodere relationship. This is relevant to Garden Cliff’s original name-bearing material; it does not make the later Somerset specimen part of this locality’s finds.

The body-fossil record also includes historical Ostrea, Protocardium and Pecten shells, often isolated and pyritised, and Lingula. Rare insects and gastropods are reported from the Cotham Member. The older aggregate vertebrate list includes Gyrolepis, Ceratodus, Acrodus, ichthyosaurs and plesiosaurs; the later account stresses that Ceratodus is nearly absent compared with Aust. These are source-specific records, not equally likely finds.

A separate shell-bearing horizon. BRSMG Cf8854 carries Protocardia rhaetica among randomly oriented shells from another fossiliferous horizon, explicitly outside the bone bed. The 2022 synthesis also cites Richardson’s historical ophiuroids with bivalves and other benthos in sandy horizons; it does not present those as new identifications from the 2021 samples.

The 2022 synthesis cites fish-tooth oxygen-isotope values of 17.8–18.8 per mille from Fischer’s 2012 work and discusses reduced salinity. This is that synthesis’s interpretation of an earlier study, rather than a direct measurement of present water conditions.

Trace fossils and environmental interpretation

The recorded traces include branching Chondrites, almond-shaped Lockeia, Thalassinoides and the feeding traces Selenichnites hundalensis and Crescentichnites antarcticus. The 2010 study adds Planolites beverlyensis, Oniscoidichnus, Rhizocorallium irregulare and Taenidium serpentium, together with a qualified unusual Walcottia form and Merostomichnites-like structures. Earlier work records Diplocraterion burrows in the lower Westbury Formation infilled by overlying black mud, as well as Skolithos. These observations are not all securely tied to the same sandstone bed.

On the BRSUG 29953 Chondrites surface, an underlying lamination bears unoriented, almond-shaped bivalve traces. BRSMG Cd5331 is the separately figured Lockeia slab. Preservation and relief affect interpretation; the different traces on a slab are not automatically the same structure as Radichnus.

Radichnus allingtona was established in 2010 from holotype BRSMG Cd5331, on the underside of a light-brown micaceous fine sandstone beneath a bone-bearing bed. The same slab number occurs in the later description of Lockeia, but the two names concern different trace structures. Body fossils, feeding marks, burrows, ripples and sediment chemistry together indicate changing shallow-water conditions, including restricted, poorly oxygenated intervals and storm disturbance followed by colonisation. They do not describe one fixed environment for every bed.

Milroy and colleagues’ Westbury succession illustrations distinguish weakly stratified red-brown silty mudstone from pale grey-blue laminated silty mudstone. Figure 13A, labelled source depth 8 m, shows millimetre-scale ripple lenses; Figure 12A, labelled depth 23 m, shows the weakly stratified and thinly laminated facies together. These plotted-source positions belong to the study’s wider succession and are not current visitor heights or independently matched numbered Garden Cliff beds. The regional interpretation links the contrasting fabrics with changing conditions on saline mudflats and in temporary lakes; the unreproduced full composite should not be substituted for Garden Cliff’s own measured schemes.

References
Benton, M.J., Cook, E. and Turner, P. (2002), Westbury Garden Cliff, in Geological Conservation Review 24, pp.237–241.
Sykes (1977), British Rhaetic bone-beds, Mercian Geologist 6, 197–239, especially pp.206–209.
Etheridge, R. (1865), historical section reproduced in Williams et al. (2022), Fig. 3a.
Williams, H. and colleagues (2022), Microvertebrates from the Rhaetian bone beds at Westbury Garden Cliff, Proceedings of the Geologists’ Association 133, 119–136, doi: 10.1016/j.pgeola.2022.01.002.
Allington-Jones, L., Braddy, S.J. and Trueman, C.N. (2010), Palaeoenvironmental implications of the ichnology and geochemistry of the Westbury Formation, Palaeontology 53, 491–506.
Milroy, P., Wright, V. P. and Simms, M. J. (2019), Dryland continental mudstones: Deciphering environmental changes in problematic mudstones from the Upper Triassic (Carnian to Norian) Mercia Mudstone Group, south-west Britain, Sedimentology 66, 2557–2589.
Quinn, J.G. and colleagues (2024), doi: 10.1080/02724634.2024.2350408.
BGS Lexicon: Westbury, Lilstock, Cotham, Blue Anchor and Branscombe units.
Wright, T. (1860), Garden Cliff and Wainlode sections, Quarterly Journal of the Geological Society 16, pp. 377–379.
Guise, W. V. (1860), Jones and Lucy’s Garden Cliff section, Proceedings of the Cotteswold Naturalists’ Field Club 2, pp. 188–190.