Chilcompton’s old New Rock colliery spoil contains shale and sandstone brought up from deep Carboniferous workings. Plant stems and fern-like impressions can be seen in suitable loose shale, but the steep, overgrown public-path route is primarily a place for careful observation unless the landowner has agreed to collecting.
FIND FREQUENCY: ♦♦♦ – Plant impressions occur in some shale pieces, but vegetation and weathering make visible material variable; rare Writhlington fossils should not be expected here.
CHILDREN: ♦♦♦ – Older, closely supervised children may manage the steep route; loose or muddy spoil can make it unsuitable for younger visitors.
ACCESS: ♦♦♦ – Follow the signed public path beside the former colliery and industrial area. Parking and collecting require their own permission.
TYPE – Mixed spoil from New Rock Colliery, rather than an in-situ sequence of beds.
DIRECTIONS
♦ From the A367 take the B3139 towards Chilcompton, then the B3356 south through the village. The old collecting locality is beside the industrial area south of Downside Abbey and school.
♦ The original approach follows a public footpath beside a farm near the industrial estate and crosses the old spoil. Follow current signs and check local route conditions rather than entering private yards or using an unsigned shortcut.
♦ The original locality reference is 51.25359°N, 2.50433°W, close to the former New Rock workings. The path climbs steeply across the spoil; choose a shorter or alternative walk if the slope is unsuitable for your party.
♦ Park only where lawful and permitted. Do not assume industrial-estate spaces are public parking, and keep farm access, business entrances, gateways and traffic routes clear.
♦ Obtain landowner permission before taking fossils or leaving the public path. Do not excavate the walking surface, slopes or tree roots to expose fresh rock.
FOSSIL HUNTING
The useful material is fine-grained shale carrying plant impressions, stems and occasional fern-like foliage. Begin with naturally weathered surfaces in loose pieces already visible beside the permitted route. Tilt a slab towards the light: a shallow impression may show its outline more clearly in low-angle illumination than when viewed straight on. Dark shale can preserve fine detail, while weathered red or brown pieces may show weaker impressions.
Look for repeated biological patterns rather than assuming that every dark mark is a fossil. Parallel ribs interrupted by joints can indicate a plant stem, and branching fronds may preserve rows of small leaflets. Vein patterns and leaflet attachment are useful for identification, but the site’s documented visitor record does not justify assigning every fragment to a named genus or importing a complete flora from another Somerset tip.
New Rock spoil is mixed mining waste. The position of a slab on the path, and its red or black colour, cannot establish a formation or coal-roof horizon. Keep the precise find place and rock description with any authorised specimen. Record whether it was naturally loose, and photograph the impression before handling it further.
The historical mine records include a few non-marine bivalves from mudstone above the Two Coal seam, tentatively identified by the Geological Survey as Anthraconauta phillipsii. This is an underground record, not evidence that shells are abundant in the present path debris. The scientifically important insects, arachnids and other rare animals at Writhlington belong to a different rock store and should not become Chilcompton collecting expectations.
If the landowner allows a little loose material to be taken, select representative pieces without disturbing the path. Thin shale can fracture across a fossil, so support it, retain both halves of any existing split and avoid repeatedly opening fresh slabs. Any permitted splitting should be done safely away from other path users; digging into the path or beneath roots is inappropriate.
Overgrowth and past collecting affect what can be seen, and the exposed material can change. A visit can still be worthwhile for understanding the mining landscape even when few impressions are visible. Leave large blocks, rooted ground and unstable slopes alone rather than trying to recreate the mine’s original exposures.
Mining, geological research and stratigraphic discoveries at New Rock, Chilcompton
1819 – New Rock Colliery sunk
The shaft initially reached Globe Coal at 166 m depth and was later extended to Garden Course at 360 m, bringing deep Carboniferous rocks into the surface mining landscape.
1824 – Early shaft section published
Buckland and Conybeare published a Chilcompton New Pit record later regarded by the Geological Survey as probably New Rock. Its seam names were subsequently reassessed.
1886 – Workings extended towards Great Course
Under Frederick Spencer, an incline was developed towards the deeper Great Course rather than extending the shaft further.
1965 – Pennant succession synthesised
The Wells and Cheddar memoir described the worked New Rock coals within the basal lithological Pennant succession.
1968–1969 – Closure reveals key marine markers
B. Kelk and R. J. Wyatt identified the Winterbourne and Crofts End marine bands during closure work, improving correlation of the southern Somerset coal succession with Bristol.
1969 – Linked mining operation dismantled
Following closure of the connected Mendip operation, equipment was sold for scrap and shafts were infilled with spoil. This historical account does not establish the present condition of every New Rock feature.
1993 – Mine sections and structure consolidated
The Bristol memoir assembled New Rock seam measurements, intervals and faulted underground geometry, including the New Rock–Old Moorewood section.
2009 – Modern formation and member terminology established
The BGS framework clarified Pennant Sandstone Formation and Downend Member usage, including the older lithological base in Somerset relative to Bristol.
GEOLOGY
New Rock Colliery worked a deep, folded and faulted Carboniferous succession. Its principal worked sequence belongs to the lower Pennant Sandstone Formation, represented in modern terminology by the Downend Member of the Warwickshire Group, with deeper workings extending into the underlying productive Coal Measures. Discarded rock from mining reached the surface as spoil, so the present heap mixes material rather than exposing one formation in order.
The sandstones record river-channel deposition, while mudstone and siltstone accumulated on floodplains and in poorly drained basins. Plant debris entered the finer sediment, and peat developed into coal. Seatearths beneath seams preserve the soils in which vegetation grew. These changing environments explain the alternation of hard sandstone, softer shale, coal and clay-rich partings in the mine records.
The Pennant boundary is defined by the arrival of characteristic sandstone, and its position varies across the coalfield. In Somerset the first Pennant-type sandstone lies between the Aegiranum, or Crofts End, and Cambriense, or Winterbourne, marine bands. It is therefore older than the equivalent lithological boundary in Bristol and should not simply be equated with a traditional Middle–Upper Coal Measures boundary. Winterbourne Marine Band is a marker horizon here; the separate Bristol-area Winterbourne Formation is not inserted into the New Rock succession.
The principal worked seams include Great Course, Garden Course, Warkey, Two Coal, Small Coal and Globe. Their coal leaves, shale partings and intervening sandstone intervals are known from historical underground records. Recognition of the two marine bands during closure work in 1968–1969 greatly improved correlation with other parts of the coalfield, although the modern visitor does not see those marker bands laid out along the spoil path.
Red-coloured rock does not establish the presence of the formal Barren Red Member, and the known New Rock workings do not support assigning this tip to the Farrington Member. Colour can be affected by weathering or alteration, while the original source of any individual spoil piece may be uncertain. The broad late Carboniferous age is appropriate here; a falsely precise date or a complete regional member-age range would imply more certainty than the mixed material provides.
Detailed geology and separately scoped historical New Rock mine intervals
This account concerns New Rock colliery spoil beside the Chilcompton industrial area, close to the former shaft at ST 6479 5057. It is not a measured outcrop through the footpath. Underground measurements below are retained in their historical units and are explicitly separated from the loose material visible today.
WARWICKSHIRE GROUP
Pennant Sandstone Formation
Downend Member
The principal worked New Rock coals lie within the lower Pennant succession. The modern Downend Member is defined lithologically, with Pennant-type sandstone beginning in Somerset between the Aegiranum/Crofts End and Cambriense/Winterbourne marine bands. The boundary therefore differs from Bristol. The member has a broad Bolsovian–Asturian regional range, but that whole age range is not demonstrated by the present mixed heap. Deeper mine material extends towards the underlying South Wales Coal Measures, and an individual loose block cannot always be assigned to the Downend Member.
Older sources used Downend Formation and New Rock Group or Series with changing boundaries. The Winterbourne Marine Band is a correlation marker here, not evidence for the separate Bristol-area Winterbourne Formation. Neither the Farrington Member nor Barren Red material is established as a component of this exact spoil by the recovered mine records.
Lower workings: Standing or New Hit seam. A branch towards Old Moorewood encountered inferior coal 46 yd from the Fern Rag drift. That figure locates the seam along a working; it is not coal thickness. The subsequent historical account places about 640 ft of measures without workable coal between Standing Coal and the then-used Pennant base.
Possible Dungy Drift. A seam cut 150 yd beyond Standing Coal was tentatively identified with Dungy Drift and comprised 48 in of coal mixed with rubbish, not 48 in of clean coal. The succeeding 300 ft of strata up to the historical Pennant base contained no known workable coal.
Crofts End or Aegiranum Marine Band. The band was identified about 230 ft above Dungy Drift in the local mine records. It marks a marine incursion within the predominantly non-marine succession; no local band thickness or borrowed Bristol fossil assemblage is assigned to it.
Little Course and Firestone comparisons. These coal horizons occur in regional descriptions and nearby sections, but the detailed mine account says they were not detected at New Rock or Strap. Their measurements from Mells or Pitcot are therefore not treated as New Rock beds.
Great Course. This worked seam lies in the basal Pennant succession below the Winterbourne Marine Band. The older regional synthesis describes a 4 ft Great Course about 90 ft above its Pennant base, but neither figure is an unambiguous modern measurement of a bed on the Chilcompton path or of the exact New Rock seam.
Strap Coal: neighbouring-workings comparison. The Survey gives 18–20 in of coal, approximately 140–180 ft above Great Course, particularly in Strap Pit workings. Those are regional comparison measurements rather than an exact New Rock seam section.
Winterbourne or Cambriense Marine Band. The marker lies below Garden Course, probably in clift above Strap. Its recognition, with Crofts End, during the 1968–1969 closure period allowed more secure correlation with Bristol. No local numerical band thickness is established.
Garden Course: Plan 2, roof downwards. An unmeasured clift or silty-mudstone roof overlies 2 in of stony coal. Below is a 16–24 in upper coal leaf containing shaly streaks, then 12–30 in of clod or mudstone parting, then a 24 in bottom coal leaf. The floor is unmeasured pan, or seatearth. Each coal leaf and parting belongs to this historical underground seam section.
Garden Course regional relationships. The digitised Survey text places Garden Course “about 150” above Strap but omits the unit, so no numerical spacing is adopted from that passage. A conglomeratic band higher than Garden Course contains large quartz pebbles and pink and green igneous pebbles in the broader Nettlebridge account; it is not identified as a separately measured New Rock bed.
Garden Course to Warkey. Warkey is approximately 140 ft above Garden Course in the New Rock shaft. Its top-to-bottom seam section comprises 11 in of coal with fusain, a 57 in shale parting and a 15 in lower coal leaf. Fusain is charcoal-like fossil plant material; the thick shale separates the two coal leaves.
Warkey to Two Coal. The interval in the Holcombe–New Rock area is about 80 ft, with sandstone, mudstone partings and false seams. It is an interval between worked horizons, not a single 80 ft sandstone bed.
Two Coal roof and seam. The roof mudstone yielded a few non-marine bivalves originally called juvenile Anthracomya, later presumed to be Anthraconauta phillipsii. The identification remains tentative. Below the roof, the New Rock record gives a 10 in top coal, 36 in of fireclay and a 12 in bottom coal. This shell occurrence is a historical underground observation, not evidence of abundant shells on today’s path.
Two Coal to Small Coal. About 50 ft of intervening measures separates these seams at New Rock and in the western area. A more detailed lithological subdivision is not supplied for this interval.
Small Coal: section reopened during the Second World War. Beneath an unmeasured clift roof lies 4–6 in of binching or parting material. The upper coal is 18–30 in thick and the lower coal 24–48 in, above a pan floor. These ranges record the historical section, not a constant coal thickness throughout the mine.
Small Coal to Globe, ascending sequence. The lower interval contains 115 ft of mudstone and sandy mudstone, followed by a thin coal whose thickness is not given. Above are 180 ft of hard sandstone and 35 ft of mudstone. The Survey states an overall Small Coal–Globe interval of 330 ft, which matches the sum of the three numbered clastic intervals; the unmeasured thin coal is not assigned an invented extra value. Different totals at Pitcot and Holcombe are separate observations.
Globe seam, roof downwards. Grey mudstone forms the roof. A 12 in upper coal leaf locally pinches out; below are 36 in of rubbish and a 26 in lower coal leaf, resting on pan. The changing upper leaf helps explain why one isolated coal fragment cannot reconstruct the complete seam.
Early shaft names and depths. Buckland and Conybeare’s 1824 Chilcompton New Pit section is considered probably New Rock by the Survey. Its seam called Warkey occurred at 533 ft 7 in depth but may actually be Globe. The seam called Small Coal at 438 ft 7 in may correspond to Daniel’s Holing, not the correctly named Small Coal below Globe. These are shaft depths and uncertain historical correlations, not seam thicknesses.
Daniel’s Holing and inferred higher coals. A New Rock shaft account places Daniel’s Holing 91 ft above the Old Globe Coal Hole at 541 ft shaft depth. Reinterpretation of the early section suggests three coals above Globe, respectively 36 in, 36 in and 25 in thick, but that reconstruction is tentative. It is not proof that each is represented by a recognisable bed in the present heap.
Faulted mine structure. The New Rock–Old Moorewood panel of Survey Figure 28 shows shafts and inclines, Mesozoic cover, faults, crushed coal and shale, and the named Globe, Small, Garden, Winterbourne, Great and Crofts End horizons. The other panels illustrate different mines and cannot be merged into a single New Rock log. No numerical thickness is inferred from the drawing.
Older regional Pennant synthesis. The 1965 memoir gives about 460 ft, predominantly sandstone, from Great Course to Garden Course. Above Garden it describes a broad package of about 300 ft of sandy mudstone followed by 200 ft of sandstone, with Two Coal and Small Coal in the lower part and Globe at the upper boundary. These generalised regional relationships are kept separate from the later, more specific New Rock measurements; they are not averaged into one artificial column.
Inferior coals in the Moorewood–New Rock level branch. Within the structurally difficult lower productive Coal Measures, the level branch cut several dirty, inferior seams, the thickest 42 in. The Survey discusses them within roughly 600 ft of measures whose correlation is complicated by folding and faulting. They may represent deteriorated equivalents of seams farther east, but the individual correlations are tentative and the 600 ft is not a single coal bed.
Great Course roof at New Rock. The local workings usually had grey mudstone above the coal, with clift feathering between the mudstone and seam. This is a New Rock roof-rock observation; the complete coal-leaf measurements from Mells, Pitcot or Moorewood are separate mine sections.
Garden Course summary and older New Rock section. The 1993 Survey’s broad description gives a Garden Course thickness of about 42 in before setting out the more detailed variable Plan 2 section. The 1965 memoir separately records at New Rock a top dirty coal 1–2 ft thick, an unbedded mudstone parting 1–2.5 ft thick and a bottom coal 2 ft thick. These source-specific records should be compared, rather than combined into extra coal leaves or substituted for the Plan 2 ranges.
Regional Globe estimate and shaft cover. The 1965 regional Pennant synthesis describes Globe as about 3 ft thick; the more detailed New Rock coal leaves remain those listed above. The same memoir records over 130 ft of Triassic Dolomitic Conglomerate in the New Rock shaft. This is younger cover above the Carboniferous workings, not an extra Pennant interval or a measured thickness exposed on the spoil path.
Additional labels in the underground cross-section. The New Rock–Old Moorewood panel also identifies White Axen, Perrink and Main towards Old Moorewood, the Edford Lag Fault, Carboniferous Limestone and the historically named Quartzitic Sandstone Group. These labels establish structural and stratigraphic context around the workings; they do not show that every labelled unit supplied the present collecting spoil.
References.
Kellaway and Welch (1993), Bristol district Geological Survey memoir, lower New Rock workings and Nettlebridge/Downend sections, Figures 20 and 28.
Waters and colleagues (2009), Carboniferous framework, pp. 136–137.
Green and Welch (1965), Wells and Cheddar memoir, historical Pennant account.
BGS: New Rock Colliery mining history.
SAFETY
The old path can be steep, uneven, overgrown, muddy and slippery. Wear footwear with a good grip and supervise children closely. Avoid loose edges, disturbed ground and unstable trees; the absence of sea cliffs does not make a spoil heap hazard-free.
Do not enter mine workings or investigate possible shafts. Historical records of infilling do not establish the condition of every feature today. Keep to the signed route, stay clear of farm and industrial traffic and do not obstruct entrances.
Never remove material supporting the path, a slope or tree roots. If a permitted loose slab is split, wear eye protection, keep others well away and stop if it cannot be handled safely.
EQUIPMENT
Bring sturdy boots, a hand lens, camera, notebook and small scale. These are sufficient for observing naturally exposed plant impressions without disturbing the public path.
If removal of small loose specimens is allowed, use soft wrapping and a rigid box, supporting thin shale flat and keeping counterparts together. Ask what tools are permitted; picks and excavation tools should not be used on the path or rooted slopes.
CLEANING AND TREATING
Photograph and label any authorised specimen before cleaning. Loose dirt can sometimes be removed with a very soft dry brush, but fragile carbon films and impressions are easily damaged. Stop if the surface flakes or the fossil loses definition.
Do not routinely soak, coat or chemically treat plant-bearing shale. Support thin pieces and let damp material dry slowly. Keep matrix and counterpart halves until the fossil has been identified, and seek advice before consolidating a crumbly or unusual specimen.
IDENTIFY YOUR FINDS
Need help identifying a fossil? Share clear photos, where you found it and its size with the community.
FURTHER READING
BGS: Mendip coal mining, including New Rock Colliery
BGS Bristol district memoir: New Rock seams and marine bands
BGS Wells and Cheddar memoir: the historical Pennant succession
BGS Lexicon: Downend Member
BGS Lexicon: Pennant Sandstone Formation
BGS Carboniferous stratigraphical framework
GOV.UK: access to private land
ACCESS RIGHTS
The described route follows a public footpath across the old mining landscape. A walking right does not give permission to leave the route, excavate the spoil or remove fossils. Ask the landowner before collecting and follow any current signs or access conditions.
Keep the path safe and undamaged for other users, leave crops, roots and vegetation alone, and use only parking that is lawful and permitted. No current blanket collecting permission or dedicated public parking arrangement is established for the tip.
It is important to follow our ‘Code of Conduct’ when collecting fossils or visiting any site. Please also read our ‘Terms and Conditions‘
LINKS
♦ Fossil Discussions
♦ Fossil Articles
♦ Buy Fossils, Tools and Equipment
♦ Buy Crystals, Meteorites, and Artefacts
♦ Join Fossil Hunts
♦ UK Fossils Network
